Gilbert Murray

이윤진 2011. 1. 24. 13:24











Oxford University Press

London Edinburgh Glasgoiv Copenhagen

New Tori Toronto Melbourne Cape Town

Bombay Calcutta Madras Shanghai

Humphrey Milford Publisher to the University




This edition contains, besides a good deal of revision in detail,
some new material on the historic background of Homeric myths
and new illustration of the traditional book by the analogy of
stage plays, as well as a small additional appendix. My general
attitude towards Homeric problems remains much the same
as I have explained in the Preface to the Second Edition, though
I hope that, in spite of the War and the League of Nations, I
have learnt something more about Homer in the last ten years.

In that period the analysis of the poems has been treated by
two acknowledged masters, Wilamowitz and Bethe.^ It is
worth noting that while agreeing in general view and method
they differ greatly in their particular conclusions. That result
indicates, in my opinion, not that the method is faulty but that
the available evidence is insufficient. It can show us, as I have
said elsewhere, the kind of thing that must have happened ; it
can seldom tell us the exact thing that did happen. Dr. Leaf's
Troy (191 2) and Homer and History (191 5), though much in
both of them fails to convince me, have added greatly to our
knowledge of Homeric geography, and show all the lucidity and
grip of real hfe which specially distinguish their author among
learned men. From Professor Chad wick I have learnt at last to
understand what a ' Heroic Age ' really is, an invaluable lesson
for Greek scholars. From Professor J. A. K. Thomson I have
had real help in the imaginative understanding of Homeric
poetry, and in particular of the relations of the bard to the chorus.
The ninth and tenth chapters of his Sttcdies in the Odyssey
appear to me to have brought light into the very heart of what
is specially called ' the Homeric Question '. From him also I
have adopted the conception of the Achaioi as a north-western
tribe, forerunners of the Dorians, and quite distinct from the

' Die Ilias und Homer, by U. von Wilamowitz-Moellendorff (1916, 1920) ;
Homer, Dichtiing ufid Saga, by Erich Bethe. Ilias 1914 : Odyssee, &c., 1923.



older Greek civilization, which was the parent of Aeolian and
Ionian. This view, based chiefly on the evidence of the three
independent groups of Achaean inscriptions, was also suggested
by Chadwick and has been adopted by many later writers. It is
not in any way essential to my general position, but it seems
to me to account for the facts better than any other current

My belief in the importance of the Attic element in the poems
is in general, though not in detailed, agreement with the position
taken by such writers as Cauer, Bethe, and Mulder, and has lately
been confirmed by the striking Sprachliche Untersuchtmgen of
Wackernagel (191 6).

Among critics who take a radically different view from mine,
I have learned most from Drerup {Das Homerprobleni in der
Gegenwart^ 1921, being vol. i of a large work on Homerische
Poeiik). The masses of interesting material which he has
collected and the acuteness of his historical arguments console
one for the over-polemical tone of what he openly calls a
' Kriegsbuch '. In this connexion I would fain pay a tribute to
the Grtmdfragen of the lamented Paul Cauer, of which the third
edition lies before me, a work which, apart from its learning,
seems to me to acquire real beauty by its patient candour and

I rejoice to find myself more in agreement with Mr. T. W.
Allen's learned and ingenious book, The Homeric Catalogue
of Ships (1921), than I had expected beforehand. Mr. Allen at
least regards the Catalogue as an ancient non-Homeric document,
originally composed for another context and, in spite of additions
and omissions, still bearing clear marks of its origin, and I see
with sympathy that he has brought down upon himself the wrath
of a more perfect ' unitarian ', Drerup, by these dangerous
concessions to human reason. But there is one point of principle
on which I think that he has overstated his case. He treats the
Catalogue as a legal document, and argues that any alteration in
it which may have occurred in the course of ages must be no
mere corruption, such as is common in literary texts, but a
' forgery made with a material and quasi-legal purpose '. Such
a view appears to me to rest on a misunderstanding, and I have
discussed it in Appendix J.


The last few years have produced several new and interesting
studies of the peculiarities of Homeric style in narrative, in similes,
and in the almost impossible task of describing battles.^ It
would be quite easy, of course, to attribute them all to the
personal tastes of a hypothetical individual. one critic, for
instance, postulates a poet who knew several dialects and was
proud of mixing- them, and who also enjoyed making fun of the
gods. But, as Jebb emphasized long ago, the thing to be
explained is not a personal taste but a tradition. An eccentric
Homer is an impossibility. A tradition is a social fact, based on
the unspoken agreement of poets and audiences, from which
neither can vary widely or abruptly. And the problem to be
understood is how and why through many generations the
normal Greek public expected its epic poets to speak in a
particular artificial dialect, to use a particular type of simile and
description, and to obey certain subtle and probably unconscious
rules of symmetry. Certainly the artistic instinct implied in both
poet and audience by these facts is very remarkable, but perhaps
not more so than the exquisite conventions of symmetry and
proportion which were traditional in other forms of Greek art,
for instance in architecture and the carving of bas-reliefs.

G. M.

Christ Church, Oxford,
July 1924.

' See the authors quoted in Cauer, Gri(ndfragen, iii. 5, to whom may be
added Shorey in Class. Phil., July 1922 ; Fraenkel, Honcrische Gleichnisse,
1 92 1 (which however I have not seen), and J. T. Sheppard, The Pattern of
tlie Iliad, 1923. See also the bibliography by Miilder in Korte's Jahres-
bericht iiber die Fortschrittc dcr kl. Altcrtuinswissenschaft, Vol. 182 (1920,
1), with his trenchant but over-subjective criticisms.


In preparing this second edition I should like to thank several
friends for notes and suggestions, among them Mr. E. E. Genner,
Professor Cruickshank, Mr. J. A. K. Thomson, and Mr. Andrew
Lang. T have also derived profit from some of my reviews, both
EngHsh and foreign.

It was vain, I suppose, to hope that even the most pacific and
wary walking would take one far into Homeric territory without
rousing the old lions that lie wakeful behind most of the larger
stones. I have listened with mixed feelings to their threatening
voices. The sportsman within me would like to go gun in hand
and bag a few of the most dangerous ; the philosopher is resolved
to do them no injury, but merely try, gradually and indirectly, to
make them friends to man. While still avoiding controversy,
therefore, I have tried in this edition to state more clearly or
correctly or patiently a number of arguments which seem to
have given trouble in the first; I have in many places added
or altered a word or two in view of fresh evidence ; especially
I have added a new chapter on the known history of the Homeric
text. The rest of my book proceeds mainly in historical order,
and deals largely with regions in which there is no record ; this
new chapter reverses that order and reaches back, step by step,
from the known to the unknown.

I am anxious to find common ground with my unitarian critics.
I only differ irreconcilably from those who reject all analysis ab
imtio ; who assume as an unquestioned starting-point that, towards
the end of the second millennium B. C, when to the best of our
knowledge there was no Greek literature, a single miraculously
gifted man, of whose life we know nothing, living in the heart of
a rich, widespread, and romantic civilization, which no history
mentions and all excavation has signally failed to discover, com-
posed for an audience unable to read two poems much too long
to be listened to ; and then managed by miraculous but un-


specified means to secure that his poems should be preserved
practically unaltered while flying viva per ora virtini through
some six extraordinarily changeful centuries. These stalwarts
do not wish to be persuaded or argued with. But for the rest
of us a meeting-ground is possibly within sight. If the Iliad is
a traditional book, in which old material has been reshaped by
later bards — whether we suppose a gradual development of a
Trojan story or an Achilles story, or a fictional reshaping of old
poetry which had originally nothing to do with Achilles nor yet
with Troy, or all these together— the difference between Wolfians
and unitarians is really one of degree.

Of course the Iliad is a unity. Every successful version of
a traditional poem is that. Every new poet who recited and
thereby modified the Iliad produced or meant to produce a
unity. Nay, the very arguments which are used to prove
a complex growth in the past will serve to prove a unity in
the present. For almost every discrepancy or awkwardness is
deliberately smoothed out and reconciled. There are no naked
impossibilities, there are no crude and unpalliated contradictions.
The poets who worked upon the Iliad were too good artists for
that. Wherever we can discern the tracks of the ' Diaskeuasles '
we can nearly always discern also the pains he has taken to
conceal his tracks.

The original substratum is a vei^a caiLsa : the poem as a whole
cannot be conceived without it. The reshaping by later poets
or editors is a vera caztsa : it is demonstrated by the history of
the text. The task of the unitarian, then, is, somewhere between
the first sources of the Ilt'ad and the last additions, to find some
one poet whose work utterly surpasses that of all who came
before or after him. For my own part, I leave that quest to
scholars of more confident temper ; the little I have to say about
it will be found on pp. 238 ff.

The subject of Homeric language needs a few words to itself,
both because of the good recent work done upon it (see p. 168 f )
and because the questions at issue are often misunderstood. The
cardinal fact about the language is the extraordinary mixture in
it of old and new, in forms, in constructions, in manners of thought.
This mixture has, of course, been explained in various ways ; to
me it is merely the natural mark and stamp of a Traditional


Book, preserved, renewed, conventionalized, and unconsciously
modified — always within the limits of the convention — by many
successive generations of reciters.

But the critics of thirty years ago were apt sometimes to go
wrong by not recognizing the complexity of the problem before
them, and trying by means of the language to determine the
comparative date of particular books as wholes, or of the two
poems as wholes. It is true that there are differences of style ;
slight but decided differences, which every good scholar, however
he may explain them, feels. But it is impossible to cut out any
large section of the poems clean and say : ' Every line of this is
written in language of a particular date.' on the hypothesis
which I follow, of course, any such expectation would be un-
scientific. The mixture of old and new is all-pervasive. The
oldest parts have passed through the lips of scores of later poets ;
the latest parts — even the most confessedly apocryphal additions
of the ' wild papyri ' — are largely made up of old lines and
phrases, and are always composed in the old convention.

Any satisfactory examination of the language statistics must
bear these considerations in mind and realize the diflSculty of its
task. It must never be satisfied with merely counting unanalysed
phenomena. It must always dig below the ' surface corruption '.
It must never use repeated or inorganic lines as if they were
necessarily original or organic in the place where they happen
now to stand. It must take full account of differences of subject
as naturally producing differences of vocabulary. And it must
of necessity, if it is to do much good, practise an extreme deli-
cacy of sensitiveness to language. When the whole poem has
been uniformly clothed in conventional epic diction, when each
rhapsode has deliberately written to the best of his powers in
' Homeric ' language, it is only by a delicacy of observation
surpassing his that we can hope to detect his deviations from
standard. This sounds very difficult ; but it has often been done.
After all, we scholars have unlimited time : and the rhapsodes,
though skilful, were unsuspicious.

As to my own particular views, I am conscious of a slight
change, or advance, of opinion on one important question, and it
is a pleasure to acknowledge here a debt of gratitude to my


famous and inveterate foeman, Mr. Andrew Lang-. I only wish
the change were one which Mr. Lang were more likely to accept
as typical of true repentance.

I speak with diffidence on points of pure archaeology, but in
his book on The World of Houter ^ Mr. Lang seems to me to
have shown that phase after phase of that world, where it is not
Mycenaean, agrees with nothing that we know on solid earth
before the sixth and fifth centuries. That is, the common opinion
which places ' Homer's world ' on solid earth in ' post-Mycenaean '
times, from the tenth to the eighth century, is confronted with
greater difficulties than ever. Our archaeological evidence is
now fairly abundant, and no such world has been discovered.
Of course there are old Mycenaean or ' Achaean ' elements.
But, apart from these, Mr. Lang argues in detail that the men's
dress, the women's dress, the corslets and armour, are markedly
different from those of the earliest vases, and just like those of
the sixth and fifth centuries. The dress is that worn by the
' older men of the wealthy classes ' a little before the time of
Thucydides (Thuc. i. 6). The same is true, as I rejoice to find
Mr. Lang saying, of the Homeric gods. They are, apart from
traces of a wilder background, the gods of Pheidias. All our
study of Greek religion has long been telling us so. The same,
I would say, is true of the moral tone of Homer. Allowing for
certain data in the saga, Homeric morals and ' religion ' in the
higher sense (see Mr. Lang's excellent remarks on p. 120) are
those of pre-sophistic Athens at her best. The expurgations of
which I make so much use point on the whole in the same direc-
tion. We have no reason to think that the cruelties and in-
decencies which I believe to have been expurgated were specially
objected to in the time, say, of the dipylon vases. The tone
of Xenophanes, Thales, and Heraclitus is, I think, enough to
show that they would pretty certainly be condemned in Ionia as
soon as the great age of Ionia was well established. It is at any
rate perfectlj' easy to show that they were all condemned in fifth -
century Athens (see pp. 263 f.).

Of course Mr. Lang and I interpret these facts differently.
I take them as confirmino- the evidence for the Pisistratean recen-
sion and the fluid condition of the poems in the fifth and fourth

' Chapters viii and ix.


centuries. All this is developed in my new chapter on the
history of the text. Mr. Lang supposes that about the year
looo there was a pure ' Achaean' age uncontaminated by Ionia,
very brief and therefore unrecorded, very local and therefore
undiscovered, which happened in all the above respects to be
surprisingly like the age of Pisistratus, 450 years later, though
different from all ages between.

If the corslets are work of the sixth century or later, a much
greater part of the elaboration of the Iliad than I formerly
ventured to suggest must belong to the time of Pisistratus or even
of Aeschylus. And I do not shrink from this conclusion. We
know for certain of only one great creative age in Greek literature,
that which extends, roughly speaking, from Aeschylus to Plato.
But doubtless there lived strong men before Aeschylus ; the
begitming of the great age may confidently be extended to Solon
or to Thales. All through this age we know that something
called Homer was constantly recited : we have strong evidence
to show that, even at the end of it, the text was still fluid and
liable to be re-written. Of course we must not forget the old,
the very old, substratum. But if we find upon that substratum
work of a peculiar architectonic greatness, a peculiar humanity
and eloquence and smoothness of diction, a peculiar dramatic
form and tragic intensity, is it not reasonable to suppose that it
acquired those qualities during the only age in which we know
that Greece had them, or something like them, to give ?

Mycenae and Cnossus in their prime may conceivably have
had such qualities. But the poems are not Mycenaean, much
less Minoan. The great age of Greece certainly had them ;
and during the great age the poems were certainly still being
recited and had not yet reached a final form. Between those two
ages Greek civilization has little to show that rises above the
level of respectable barbarism. one cannot indeed quite suppose
that masses of old epic poetry lay completely dead and buried
till some sixth-century Kynaethus dug them up. The epic con-
vention is too fixed, the whole style is too intelligible, for that.
And our miserable remains of the Rejected Epics illustrate sug-
gestively what the substratum, or the sources, of the Iliad may
have been like, before they were glorified. There is a separate
inquiry there. But it looks as if we must face the probability


that a far larger amount of real creative work than we ever
suspected was done upon both Ihad and Odyssey by poets not
far removed either in date or in spirit from Pindar and the great
Athenians : that the history of Greek literature is after all a
great and intelligible contimmnt^ not one shining prehistoric
island, then centuries of darkness, and then all the rest.

There has been a great output of books on Homer in the last
three years ; I mention here only a few that may be useful to my
readers.^ Dealing with the general question, we have to welcome
a second edition of Paul Cauer's lucid and fascinating Grund-
fragen der Homerkritik (Leipzig, 1910), to which in my twenties
I owed a large debt of gratitude, and an Italian translation of
Drerup's well-known and copiously illustrated Horner^ enlarged
and improved. A new book, Georg Finsler's Homer (Berlin,
1908), gives an extraordinarily comprehensive and compressed
account of almost all sides of Homeric criticism ; Professor
Seymour's useful Life in the Homeric Age (Macmillan, 1908) is
full of minute and sober observation ; the short Probe eines
wissenschaftlichen Komvteniars su Homer^ by E. Hermann
(Hansaschule-Festschrift, 1908), is particularly promising. I hope
it will be carried further.

Dealing with the actual analysis of the poems, Wilamowitz, in
a paper on Q {Sitzzmgsber . d.k. Preussische^i Akademie^ 1910,
xxi), has argued very persuasively that most of that book was
probably composed to make room in a connected Iliad for two
existing but independent lays, I and K. Another excellent
article is Hektor's Abschied, by Erich Bethe [Abhandlungen der
k. Sdchsischen Ges. d. Wissenschaften, xxvii, No. xii), arguing
that in the main the author of Z was also the author of /2,
and, though a late poet, perhaps deserves the name of Homer.
There are certainly marked similarities between the two books.
Dr. K. Rothe's Ilias als Dichitmg (Paderborn, 1910) is a very
erudite and pleasing restatement of the conservative position.
He considers that Homer (i) used old epic material freely, but
turned it all to his own artistic ends ; (2) that when he had finished
the poem he sometimes turned back to it and added pieces;
(3) that he lived in a charming court in Ionia, founded by the

' A very complete bibliography is in Rothe's articles, first in BursiarHs
Jahresberichi, and afterwards in Zeitschr.f. d, Gymnasiahvesen .


last king of Mycenae, who had fled thither from the Achaeans,
and betrayed other personal weaknesses which are reflected in
the figure of Agamemnon. This, of course, seems to me like
a fairy-story, but much of Rothe's criticism is good. Mr. T. W.
Allen's articles on the Homeridae, the Epic Cycle, and the
Catalogue are also written from a severely unitarian standpoint
{Classical Quarterly I (1907), 11 (two articles); J. H. S. xxx.
pp. 292-323). Mr. Andrew Lang's World of Hotter (Longmans,
1910) restates his old views with some interesting modifications
in the light of recent literature. Mr. Shewan's Doloneia
(Macmillan, 191 1) is an industrious and gallant attack upon all
critics who have either spoken disrespectfully of K or thought
its style in any way peculiar. Van Gennep's litde Question
d'Homcre has a useful bibliography by A. J. Reinach (Paris, 1910).
Dr. Verrall's volume of essays, T/ie Bacchants of Euripides
(Cambridge, 19 10), contains two valuable papers on Homeric
subjects : The First Horner^ showing that in the fifth century
' Homer ' meant much more than ' The Iliad and the Odyssey '
and suggesting that the first Epic Cycle dates from Pisistratus ; and
The Mutiny of Idonieneus, arguing a harmonization of sources
in Iliad K-N. Among new attempts at analysis of the poems
we have Pick's Eittstehung der Odyssee (Gottingen, 1910), terse
and masterly, like all that Pick writes, though involved with
improbable speculations ; and Miss Stawell's striking work,
Homer and the Iliad, a book full of fine observation and poetical
understanding. She attempts to reconstruct an ' original Iliad '
(omitting most of B, all H, 0, I, X, IV, H, half O, and much of
the later books, but keeping at all costs Z and fl), and fortifies
her results by a further study of the language ; this ' original
Iliad ', however, probably made free use of older poems.

A somewhat new form of ' unitarianism ' is put forth in
Dr. Mulder's vigorous and valuable book. Die Ilias tmd ihre
Quellen (Weidmann, 1910). 'An abundance of unassimilated
material in spite of a constant effort after uniformity ' is his descrip-
tion of the problem, and he finds its solution in the hypothesis
of a single gifted and artificial poet who, by processes of daring
fiction, wrought a new poem out of numbers of old ones — the old
Thebais, a Meleager epic, a Heracleia, a Pylian epic, an Achilleis,
and others. The Iliad was thus produced in Ionia about the


year 625, the Odyssey somewhat later at the Court of Pisistratus.
His poet does much the same work as the ' Bearbeiter ' or
' Diaskeuast ' of earlier scholars, only more of it.

The ' surface corruption ', already ably treated of late years by
such editors as \'^an Leeuwen and Professor Piatt, is the subject
of many clever and interesting conjectures in Mr. Agar's
Homerica. Perhaps I may be allowed to urge every student
who wishes either to study the language or to enjoy the music
of Homer to accustom himself to ' thinking away ' this destructive
and often unmetrical surface-corruption. For English readers
the best method is a constant reference to such texts as I have
mentioned above, together with an occasional reading of Fick,
The outline of this problem, as of most others affecting Homer,
will be found in Father Browne's Homeric Study (Macmillan,


G. M.

Christ Church, Oxford,
May 191 1.


These lectures were written in response to an invitation from
Harvard University to deliver the Gardiner Lane Course for 1907.
Only some half of them were actually so delivered. The subject
had been so long forming itself in my mind, and I was also so
anxious not to allow any mere lack of pains to prove me un-
worthy of the honour thus offered me, that I soon found my
material completely outrunning the bounds of the proposed
course. I print the whole book ; but I must confess that those
parts of it which were spoken at Harvard have, if it is not
egotistical to say so, a special place in my affections, through
their association with the constant and most considerate kindness
of Mr. and Mrs. Lane and of many others who became in varying
degrees my xenoi in America.

The book touches on some subjects where, feeling more than
usually conscious of the insecurity of my own knowledge, I have
not scrupled to take advantage of the learning of my friends.
On several points of archaeology and primitive history I have
sought counsel from Professor J. L. Myres ; on points of Old
French from Miss Pope of Somerville College; on Semitic
matters from my colleague Professor D. S. Margoliouth, whose
vast stores have stood always most generously open to me.
In a more general way I am conscious of help received from
Mr. J. W. Mackail and Mr. T. C. Snow, and above all from
Miss J. E. Harrison, who read the Lectures in MS. and called my
attention to much recent foreign literature which I should other-
wise have neglected. The debt which I owe to her Prolegomena,
also, will be visible on many of the ensuing pages.

In subjects such as these the conclusions reached by any writer
can often be neither certain nor precise. Yet they may none the
less be interesting and even valuable. If our evidence is incom-
plete, that is no reason for not using it as far as it goes, I have


tried throughout the book never to think about making a debating
case, or taking up the positions most easy to defend ; but always
to set out honestly and with much reflection what really seems to
me to be most like the truth. I feel, indeed, that 1 ought perhaps
to have stated my evidence much more fully and systematically.
My excuse is that the lectures were originally written almost
without books of reference, and that when I went over them to
verify my statements and cite my authorities I hesitated to load
the book with references which might be unnecessary, and which
in any case were rather in the nature of afterthoughts.

As regards the Homeric Question, which forms in one way or
another an important element in my subject, I have long felt that
the recent reaction against advanced views has been largely due,
not indeed to lack of knowledge, but to inadequate understanding
of what the ' advanced ' critics really mean. A good part of my
present work has therefore lain in thinking out with rather more
imaginative effort many of the common phrases and hypotheses
of Homeric criticism. My own views are not, of course, identical
with those of any other writer. Among English scholars I agree
most closely with Dr. Leaf, and may almost say that I accept his
work as a basis. For the rest, I follow generally in the main
tradition of Wolf, Lachmann, KirchhofF, Wilamowitz. But
the more I read, the more conscious I am of good work being
done on all sides in the investigation of Greek religion and early
history, and of the astonishing advance which those subjects have
made within my own memory. The advance still continues.
Archaeologists are throwing shafts of light even across that Dark
Age of which I speak so much in Lectures II and III. My own
little book, heaven knows ! indulges in no dream of making
a final statement of the truth on any part of its field. It is only
an attempt to puzzle out a little more of the meaning of a certain
remote age of the world, whose beauty and whose power of
inspiration seem to shine the more wonderful the more resolutely
we set ourselves to understand it.


New College, Oxford,
Sept. 1907.






1. Greek Poetry and Thought conceived as forces in the service

of man 1-3

Other modem views of the Greeks :

a. The Greek as ' Classic ' ; true, if rightly understood . . 3-7

<^. The Greek as ' Pagan '; directly false 8-21

Hellenism always in an uphill war against ' Paganism ',
e. g. in the questions of:

1. Human sacrifice II-IS

2. Slavery 15-18

3. The subjection of Women 19-20

4. Impurity and Cruelty 20-21

c. The Greek as Levantine 21

2. Poetry and Progress. Can Poetry advance ? Can interest in the

bettering of life be a poetical emotion ? . . . . 22-24

3. The Puritan and the Artist. The bettering of life implies their

union, and this is found in Hellenism .... 24-25

Sdphrosyne 26

-Greek (as in



1. The prehistoric cities followed by a Dark Age

The stones

The traditions

2. A commercial civilization, not Greek but pre

stanced by Mycenae, Troy, &c.) .

3. Ruined by invasions of peoples from the North

The earliest mixture ....

Aegean peoples, Pelasgi, Hittites
Northerners, Achaeans, &c. ; complications of

4. Processes of invasion

Condition about the thirteenth century B. c.

Last stages ; by land and by sea ; seizing an island








This picture confirmed by the foundation-legends of the

Ionian cities 55~57

5. The Dark Age; the Walled City 57-58

Religion of the Polis 58-60



The wreck of institutions ....

1. Agricultural sanctions; the ox .

2. Tribal gods : the breaking up of tribes

3. Heroes, oracles, the dead .

4. The local Korai ; the Kotiroi .

5. The family, patriarchal and pre-patriarchal
Hesiod's Fifth age and the survival oi Aides and Ne7nesis
Aidos and Netnesis

The meaning of the words

Hesiod's five deadly sins ; the sanctity of the helpless
Small importance of Aidds in later Greek philosophy


Effect of this early anarchy on Greek civilization as a whole







1. Nature of a book in early times 93-99

^ LogoV QXidi^ Logioi Andres' 93-96

The Muses. The ' Grammata ' 96-99

Long life and growth of a book 99-100

2. Examples 100-107

(The Pilot series, Hamlet, Lear, Richard III, 100-105 ; the

Pseudo-Callisthenes, 105-106; The Song of Roland, 107).

3. Especially the Hebrew scriptures 107-119

Deuteronomy 107-108

J, E, andP; their sources and growth 108-113

Disturbing influences in the history told by the Pentateuch . 114-119

a. Unconscious: myth, tribal spirit, inadequate language . . I14-115

b. Conscious: archaism and expurgation 115-119


A. Expurgations: The Homeric Spirit

>■ The Poems represent an Idealized Past 120-122

Expurgation in the Alexandrian Critics 122-124

Traces of earlier expurgations 125-140



a. Impurity 125-126

b. Cruel and barbarous practices (torture, stripping the dead,

decapitation, poisoned arrows, &c.) 126-130

c. Human Sacrifice 130-140

1. Sacrifice of Virgins. The Taboo of the War Path .131-135

2. Sacrifice of Divine Kings. Division between Man and

God 135-140

Expurgation baffled :

Hector; cf. Uriah ; how Homer rises to the occasion . . 140-143
The Iliad in Education ; the Homeric Spirit .... 143-145

B. Evidences

Unconscious changes of Custom . . . . . . . 146-168

1. Armour and tactics ; the Mycenaeans and the ' Bronzen Men ' 147^159
(General armour, 147-149; tactics, 150-154; the thorax,

154-157; superpositions, unrealized epithets, 157-160.)

2. Bronze and iron 160-161

3. Burial and burning 161-162

4. Altars and Temples . . 162-164

5. Hedna and dowries 164-165

6. Houses ; Megaron and Palace, &c 165-168

C. Peculiarities

1. Tradition and Artistic Fiction; the latter prominent in Greek

Epos 169-170

Contrast with the Book of Judges; how the material oi Judges

might appear in a Greek epic 170-172

— Tradition and Fiction : the basis of the Iliad \s Tradition . . 173-174
The Iliad using older sources; Bellerophon and Lycurgus

from the Corinthiaca 174-177

Explanation : Reciprocal quotation in traditional books . . 177-178
Some sources of the 7/m</: the Catalogue ; Mulder . . 178-182

2. What is the //zW/? Not a Chronicle poem 183-184

Not exactly a Lay 184-185

But a Lay incrusted with additions and made to outgrow all

natural limits 1 86-1 87

(Fictional use of the Wrath-motive, 186.)

3. For what purpose can such a poem have been destined ? . . 187-194
The /'(m(i:/'^,?«<a:^a-tradition and its meaning .... 188-190

Before the /"awrt/z^^^rt^a ? 191-193

Epideixis 194


D. Historical Content of the Iliad

Is Homer more historic than the non-Homeric Tradition ? . .195-198

Theory of ' historical nucleus ' 199-201

History in Fiction 201-202

The Catalogue 202-204

Fiction and Myth 204-205

Helen 205-206

Achilles 206-208

Agamemnon 208-210

Odysseus and the Calendar 210-212

Thersites 212-215

Aias, Diomedes 215-217

E. Historical Content of the Iliad, continued

The historical fact always transmuted 218

Echoes of pre- Homeric warfare 219-226

Sarpedon and Tlepolemus 219-220

1\iQ Androktasiaz : Phaestus, Adrastus 221-223

Original geography : Homes of Andromache, Hector, Helen . . 224-227

Greek and Northern epics 227-230

Conjectural results 230-232

The successive ' Births ' of Homer 232-237


How scientific analysis affects our appreciation of the Iliad, and

where its greatness lies 238-260

We must consider the real Poem, not the hypothetical Poet . 239-242
Difficulties 242-249

A. Apparent vices of the Iliad as a poem :

1. Subject second-rate 242

2. Lack of finish in hidden parts ; contradictions, &c. . . 242-243

3. Loose use of linguistic forms 244-245

4. ' Ready-made ' descriptions and similes 245-249

(Examples from M : the baffled lion, 245-247 ; the two
oaks, 247-249.)
These seem to show a lack of originality or even of sincerity.

B. Answer. The nature and the standards of a Traditional Book . 249-260

' Originality' and intensity of imagination 249-251

Intensity of imagination in the Iliad and the Old Testament.

(Helen and Jehu.) 252-255



This intensity not connected with self-assertion, but with self-
devotion 255-256

Position of a late ' Son of Homer ' 257-258

Hence comes the peculiar appeal of most Traditional Books . 258-260


1. Homer is never born in Miletus or Ephesus: why? . . . 261

Spirit of Ionia in its bloom ; Thales, &c 262-263

The ultimate failure of Expurgation to meet the demands of this

new age 263-264

Homeric theology satisfied neither local superstition nor

advanced thought 265-267

2. The Milesian Spirit and its treatment of the Homeric Gods . 268

Examples :

Titanomachia in Hesiod and Theomachia in Homer . 269-271
Hieros Gavios and Dios A pate 271-273

3. The Epos dies in Ionia ; but the Heroic Saga finds a new vehicle

in Attic Tragedy, which goes back to a severer and more

antique spirit 273-275

Examples :

lo in Aeschylus : the Suppliant IFomen, the Prometheus . 276-279
Conclusion 279-281


1. The verbal text becomes uncertain when traced back beyond

150 B.c 282-295

The Alexandrian critics 283-284

What were the ' pre-Alexandrian ' texts ? 285-295

Evidence of Papyri 286-289

Evidence of quotations 289-294

Conclusion : special case of Plato 295

2. The incidents of the story become uncertain when traced beyond

about 400 B. c. (The Wall, the Catalogue, the Doloneia,

the end of the Odyssey) 295-297

3. The meaning of ' Homer' becomes uncertain when traced beyond

about 450 B. c 298-299

4. The Panathenaic Recitation included only the //tad and the

Odyssey 299-301

5. The traditional text comes from Athens ; no non-Attic text seems

to have survived 301-302

What text was used at the Panathenaea ? 302-304

Pisistratus : the last glimpse of land 304-306

The data and the problem 307-316



A. The Pharmakoi and Human Sacrifice

B. Torture of Slave Witnesses .

C. The Thalassocrats ....

D. Hubris, Dike, Horkos . . .

E. The Pseudo-Callisthenes

F. Stages of Old French Poems : Roland and

G. Expurgation in the Hymn to Demeter
H. The Epic Cycle

I. Transliteration from the 'Old Alphabet '
J. The Catalogue as a ' Legal Document '


St. Alexis







These lectures form the first part of an attempt to study the
growth of Greek poetry from a particular point of view, namely,
as the embodiment of a force making for the progress of the
human race. By progress I understand some gradual ennobling
and enriching of the content of life ; or, to adopt the magnificent
language of the document known in Scotland as ' the Shorter
Catechism ', some movement towards the attainment of that ' chief
end of man ', which is ' to glorify God and enjoy him for ever '.

This conception of all the arts, even poetry, as being so many
forms or parts of the service of man, may strike a hearer at first
as somewhat modern and removed from ancient habits of thought.
But I think the truth is just the opposite. The idea of service to
the community was more deeply rooted in the Greeks than in us.
And as soon as they began to reflect about literature at all— which
they did very early — the main question they asked about each
writer was almost always upon these lines : ' Does he help to
make better men ? ' ' Does he make life a better thing ? ' We all
know with what rigid and passionate Puritanism this view is
asserted by Plato. But Plato can never be taken as representing
the average man. There is better evidence of ordinary feeling
in the Frogs of Aristophanes.^ ' on what grounds should a poet
be admired ? ' says Aeschylus, and Euripides answers — ' For his
skill, his good counsel, and because we make men better in their
cities '. Amid all the many cross-currents of criticism illustrated
in the Frogs, there is no protest against this judging of poetry
by its fruits. The principle is accepted by all parties.

Among later writers the idea of the service of man, or the
bettering of human life, has become habitual and familiar.
Diodorus begins his history by a reference to the long chain of

' V. 1008, 1035, and the whole scene : cf. also Isocr. iv. § 159, and elsewhere.

27C0 3


historians who ' have aspired by their own labours to benefit our
common life '.^ Polybius speaks of history as the most obvious
help towards ' the correcting of life '.

Thucydides, as we all remember, will be content if his work,
whether interesting or uninteresting to an audience, is judged to
be useful. Denys of Halicarnassus sums up the praises of the
Athenians by saying, in the very language of an old Delphian
decree, that they * made gentle the life of the world '.

Theologians and philosophers, especially those of the more
ratlonahst schools, carry the conception further. The traditional
Gods are explained as being so many great men of past ages
who have in their various ways served humanity. ' That which
benefits human life is God,' said Prodicus in the fifth century B.C.
' Deus est mortali iuvare mortalem,' says Pliny from a Stoic source
in the first A. D. And in later times the view is always widely
current, a common meeting ground for Euhemerist, Stoic, and
Epicurean. The history taught in schools largely consisted, if
we may generalize from our extant Scholiasts, in lists of these
benefactors of mankind :

Inventas aut qui vitam excoluere per artis,
Quique sui memores alios fecere merendo.'^

' Diod. i. I Tols Iblois nouois o}c[)e\ri(Tai tov koivov ^'lov i(\)CkoTini]6rjaav. Cf.
Polyb. i. I {fir]8efilav . . . fToifiorepav Tols avdpMnon diopdaxriv) ; Thuc. i. 22 ; Dlon.
Hal. de Thncyd. p. 919 'Adrjvaloi , . . oi tov kolvov jiiov f^i]p.epo)aai>Tts ; idem,
iv. 25 (p. 701 R) on Servius TuUius. Herodotus, as one might expect, has
more of the mere artist about him : he writes, ws pTjTe to. yevofxtva e^ avdpa>7rcov
Tw XP°^V e^iTrjXa yivrjrai, prjTf fpya ptydXa re /cat datpncTTa, to. fifv "EXXijcrt, ra Se
^ap^upoia-i anohexOivTa, ciKkm yeprjrai (i. i). Compare also the remarkable
language of a Delphic Inscription of the second century B.C., in Bulletin de
Corr. Hellenique, 1900, p. 96, conferring honours on certain Athenians :
"ESo^f Toii 'Afi(j)iKTvoaiv ' eneiSrj yeyopevai Kal (rvueiXex^ai rep^fiTwi' crvvobov nap'
'Adqvaiois (Tvp^e^rjKe npQiTov, hv 6 8ripos inravTutv rmv iv (ivdpanois ayadav apxrjyos
KaraaTadeis, iy pev tov Oqpimbovs ^iov pfTi]yny(v tovs avdpdaiTovs els fjpepoTrjra,
Trapairios 6' eyevfidi] Trjs npos aWrjXovs Koivcovlns, eianyayoav Trjv tcov pvarrjpLoifi
7rapd8oaiv Kal 8ia tovtcov napayyeiXas tois anacnv ort peyiaTov ayadov eariv tv
avdpoinots fj wpos eavTOvs XPW'-^ '''^ '^'^'' TicrTiy, eVi 8f tu>v hoQevratv vno tS)v Becov
TTfpX tu)v dvdpanroov v6pa>v Kal Tfjs naiSdas ... ' Decreed by the Amphictyons
of Delphi : "Whereas it was in Athens that a union of the craftsmen of
Dionysus (i.e. tragic actors and poets) first arose and was gathered together ;
and whereas the People of Athens, the established leader in all human
advance, first won mankind from the life of wild beasts to gentleness ; and, by
introducing the Mysteries and thereby proclaiming to the world that the
greatest good for mankind is a spirit of help and trust toward one another,
hath been part maker of the co-operation of men with men, and of the laws
given by the gods for the treatment of men and of education ... *

- Plin. H/sL Nal. ii. 7. 18. Vergil, Ac?i. vi. 663. Cf. Lucr. v, latter part. I


It is the very language and spirit of that service of humanity
which lies at the heart of the practical religion of the present
day. The modern artist or admirer of art is apt to be offended
by it. Not, I think, justly. In a Greek society the artist was
treated frankly as a friend and fellow worker. He helped to
make life beautiful, which is at least one large and obvious way
of making it good. In a modern society he is a distinguished
alien, approached with a mixture of adulation and mistrust. We
suspect that what he calls beautiful may be really wicked.

I must take for granted many fundamental theses. That man
has progressed, for one thing, and that the direction in which
Western civilization has moved is on the whole a good one. I
think that few of us seriously deny these propositions ; and those
who do would not be moved by my arguments.

Now we find it generally admitted that the seeds of Western
civilization are mostly to be found in Greece and not elsewhere.
Yet it is curious how seldom Greek Literature is regarded from
this point of view, as an embodiment of the progressive spirit, an
expression of the struggle of the human soul towards freedom
and ennoblement.

We have had in abundance the classical point of view. The
Greeks have been the Classics, the masters in art and letters,
models of a finished and more or less unapproachable perfection
in form. Or rather, to put it more accurately, the Greeks round
about the fifth century B. c, and the Romans of the centuries just
before and after the Christian era, have been peculiarly the
Classics, and other writers have been admitted to various degrees
of classic dignity in proportion as they approached to the two
great periods.

Now I should like, if time permitted, to trace this conception
to its origin. Unreal as it sometimes sounds, it has its base in
mere fact. The Greeks and Romans of those two periods did,
for some reason or other, produce in most departments of thought
better work than any of the generations that succeeded them for
some thousand years or so ; and what is more, the generations of

suspect that this view of human history was largely inspired by the great work
of Dicaearchus, Bios 'EXXaSo?. He was an immediate disciple of Aristotle ; the
Lt/e of Hellas was a history of Greek civilization. Fragments in F. H. G. ii.

B 2


the decadence had the extreme good sense to see it. As regards
literature, the point is too obvious to need illustration. Let us
take a quite different field, the science of medicine. If a man
wished to learn medicine in the later ages of the Roman or
Byzantine empires, and right on to the Renaissance, to whom did
he go for his knowledge ? He went, as far as I can make out, to
various handbooks and epitomes of the works of two ancient
doctors ; of Galen, a Greek who practised in Rome in the year
1 60 A. D., and of Hippocrates, a Greek who practised in Cos and
Athens in the fifth century B. C. And Galen's own work largely
takes the form of a commentary on Hippocrates.

There is an interesting MS. extant of a treatise on Dislocations
by one Apollonius of Citium in Cyprus. The MS. was written in
Constantinople about the year 950 A. d., and it begins with a paean
of joy over the discovery of the works of this ancient surgeon,
with his accurate drawings to show how the various dislocations
should be set. The text was written out. The illustrations
were carefully copied. Where the old drawings were blurred or
damaged the copies were left incomplete, lest some mistake
should be made.^ Why } Because this ancient surgeon, living
about 150 B.C., knew how to set dislocated limbs a great deal
better than people who lived a thousand years after him. It was
a piece of good fortune to them to rediscover his work. And
his writing, again, takes the form of a commentary on the fifth-
century Hippocrates. Hippocrates' own writing does not look
back. It is consciously progressive and original.

That is what the Classics once were. I will not attempt to
trace the stages through which their empire has waned and their
power to help us dwindled away. What they now possess is a
limited but a most interesting domain. I will express it in this
way. There seems to be in human effort a part that is progressive
and transient, and another which is stationary or eternal. In some
things we find that a very third-rate person who happens to have
been born in 1 860 can teach us far more than a great genius or
a great reformer who was born in 1 760. About electricity, for
instance, or steamships. In the other sphere it is the quality of

^ See Schone's introduction to his large edition (Teubner, 1896), where
this point is proved. See also Greek Medicitte t'tt Rome, by Sir Clifford
Allbutt. 1 92 1.


the man or his work which tells. And it tells almost unaffected
by distance : what was once beautiful is still beautiful ; what was
once great of soul is still great. And if Shakespeare was born
nearly 400 years ago, and St. Paul 1900 and Aeschylus over 2000,
those facts do not seem to make any noteworthy difference in the
value of their work. This distinction is, I think, implied in the
current phrase which says that the ancient Greeks are still classics
in point of style. "

Now, in the narrow sense of style, any such view as this would
be almost grotesque. No modern historian could possibly model
his style on the strange contorted language of Thucydides ; no
playwright co^d copy Aeschylus. Aeschylus and Thucydides
were men of extraordinary genius who irresistibly bent the Greek
language to their will. They are not, in any literal sense, models
of normal style. If, however, we understand 'style' broadly
enough, so that style means the same as ' form ', and ' form ' in-
cludes ' spirit ', then, I think, the principle is true. The classical
books are in general the books which have possessed for mankind
such vitality of interest that they are still read and enjoyed at a
time when all the other books written within ten centuries of
them have long since been dead. There must be something
peculiar about a book of which the world feels after two thousand
years that it has not yet had enough. one would like to know
what it is that produces this permanent and not transient quality
of interest. And it is partly for this purpose that we study the
Classics. In some few ways one can know. Form or spirit in
some sense lives longer than matter ; austerity perhaps lives
longer than sweetness ; what is simple and serious lives longer
than what is merely clever. Much more remains unanalysable,
or can only be found by study of the books themselves. But
there are qualities that make things live ; and that which lives
becomes classical.

Yet I think that this kernel of truth is involved in much error.
It is probable that these models of style, as they were read both
in the Middle Ages and the Renaissance, were often bad models
rather than good. The accident was imitated, not the essence.
And the influence of the most living and original of all literatures
produced the corruptness of Classicism, a style almost certainly
very vicious, and that for two reasons. First, because it attempted



to reproduce In an uninflected language all kinds of exquisite
effects, largely connected with the order of words and the building
of periods, which are only possible and natural in a highly inflected
language. Secondly, because, in its appreciation of the immense
imaginative value of tradition and allusion, it groped round for
a tradition and found only one that was foreign and exotic and
therefore could not truly serve its purpose. There Is a great
grandeur in the prose of Milton and Hooker ; there Is at least
quaintness in the poetical style, largely inspired by Ovid, which
ran riot during the Renaissance, a style in which people called the
sun * Phoebus ' or the moon ' chaste DIan ', and were proud of
knowing stories of a complicated mythology which was not
accessible to ' the vulgar '. There are traces of something like
classicism in Greek poetry, I admit. They are the first signs of
its decay. The classicist spirit is just so far related to the living
spirit of Greek poetry, that It is a ranker form of the same poison
by which Greek poetry died.

That sort of eighteenth-century or Renaissance classicism is
perhaps dead, or no longer an active danger to the understanding
of Greek. But there are other classicisms which threaten us still.
Scholars in talking of the classics have allowed the object of their
study to become confused with the medium through which they
approach It. It is as though a man could not think of the stars
except in terms of telescopes, or of mountains and sea except in
terms of railway journeys and hotels. Nearly all of us approach
the classics through an atmosphere of education, with its con-
comitants of dictionary and grammar, its unnatural calm, its
extreme emphasis upon dutifulness and industry, and the subtle
degradation of spirit produced by its system of examinations.

vSome indeed take another path. From Winckelmann onwards
there have been many critics who felt, for obvious reasons, that
they could understand a Greek statue more easily than a Greek
poem. Hence comes another sort of classicism, a tendency to
explain the poems by the statues. A false road ; partly because
the Immense majority of extant statues are not Classical Greek,
but Graeco-Roman, and marked with the taint of the decadence :
but far more because, in the essence of things, poems are made of
quick words, and statues of stone, things that are not alike and
never have been.


The fact seems to be that the understanding of Greek poetry
needs first a good deal of hard linguistic study, and then, since
every one who likes poetry must have in himself some germs of
a poet, a poet's readiness of imaginative sympathy. As things
are, the poetical minds are often repelled by the grammatical
drudgery : and the grammarians at the end of their labours are
apt to find that their little spring of poetry has dried up.

The wise want love, and those who love want wisdom :
And all best things are thus confused to ill.

As to all these dogmas about what is Classical, I think we
should be on our guard. Classical and modern ; classical and
romantic ; classical and Christian ; there are no doubt some real
differences corresponding to these phrases, but I would urge
respectfully upon any student who loves poetry, that he should
approach his ancient poets quite simply and take what they have
to give him, not start off by expecting them to be ' classical ' or
' statuesque ' or ' pre-Christian ' or anything else. The more you
understand them, the less of these differences you will feel. And
for a simple reason : that the differences lie largely in the accident
of our own remoteness. We stand very far off, and have to
strain our eyes. For us the comparison of ancient and modern
is largely a comparison of something half-seen at a distance with
something which we know intimately. We are apt to see only
the bold outlines ; we are apt to miss the little lights and shades,
the quick vibrations of emotion that existed to a Greek in some
particular word or phrase, and therefore we think they are not
there. We mentally translate the words into a sort of dictionary
language, never very apt indeed, but, we hope, at least dignified ;
removed alike from subtlety and from littleness because it is
emptied of most of its meaning ; serene and unemotional because
we have not the knowledge or the sympathy to catch, across this
gulf of years, the peculiar thrill of what was once a ' winged
word ' flying from soul to soul. It is perhaps in this department
that the most pressing work of pure scholarship remains to
be done.

That conception of the Greeks as Classic, then, has a basis of
truth. It is only apt to be misstated, and so to darken counsel.



There is, however, a peculiar modification of it — which is almost
the direct opposite of the truth ; a conception of Hellenism as
representing- some easy-going half-animal form of life, untroubled
by conscience or ideals or duties, and the Greeks as a gay uncon-
scious hedonistic race, possessing the somewhat superficial merits
of extreme good looks and a mythically fine climate. There is
no reason to suppose the ancient Greeks miraculously handsome,
any more than to suppose that there is no dirty weather in the
Aegean. This view has so little of the semblance of truth about
it, that one wonders how it can have arisen. There are of
course the causes mentioned above, the presence of the Graeco-
Roman statues and the special diflSculties of understanding the
finer sides of the Greek language. But this particular conception
of the Greeks as ' Pagans ' comes, I think, largely from the mere
need of an antithesis to Christianity on its ascetic side. Christian
apologists, anxious to associate all the highest things in the world
with their own religion, have proceeded to make the Greek a
sort of type of what the natural man would be without Christia-
nity. And they have been met half-way by the rebels of their
own flock, intellectual people of an artistic, a revolutionary, or
a pleasure-loving temperament, who have turned against the
narrowness or conventionality of their Christian surroundings,
and then accepted, as a rough embodiment of their own rebellious
ideals, some imaginary Pagan Greek.

That would explain why this odd ideal of the Pagan Man
should be abroad at all. But why should the Greeks be chosen
as representing him ? Partly for their mere eminence. They
are the chief representatives of high civilization outside modern
Christendom. Partly, I think, from a disproportionate attention
sometimes bestowed on particular parts of Greek literature. But
largely for a reason peculiar to their own case, which I believe
to be very influential. We shall meet with it often during these
lectures. It is that we, living in an age when certain great strides
in human progress seem to be securely made and to need no
more thinking about, look back upon these early pioneers of
progress with some lack of historical sympathy, and attribute to
the Greek spirit itself a number of primitive habits which it was
not quite strong enough to conquer or else had not the leisure to
grapple with.


Anthropologists have shown us what this Pagan Man really is.
From the West Coast of Africa to the Pacific Isles in many vary-
ing shapes he meets us, still with the old gaiety, the old crowns
of flowers, the night-long dances, the phallus-bearing processions,
the untroubled vices. We feel, no doubt, a charm in his simple
and instinctive life, in the quick laughter and equally quick tears,
the directness of action, the unhesitating response of sympathy.
We must all of us have wished from time to time that our friends
were more like Polynesians ; especially those of us who live in
University towns. And I think, in a certain limited sense, the
Greeks probably were so. But in the main, as all classical
literature shows, the Greek and the Pagan are direct opposites.
That instinctive Pagan has a strangely weak hold on life. He is
all beset with terror and blind cruelty and helplessness. The
Pagan Man is really the un regenerate human animal, and
Hellenism is a collective name for the very forces which, at the
time under discussion, strove for his regeneration. Yet, histori-
cally, one of the most characteristic things about Hellenism is
that, though itself the opposite of savagery, it had savagery
always near it. The peculiar and essential value of Greek civili-
zation lies not so much in the great height which it ultimately
attained, as in the wonderful spiritual effort by which it reached
and sustained that height. The pre- Hellenic civilization of the
Aegean area was in some ways very high. Minoan Crete, for
example, produced larger buildings, better drainage, and in some
respects a livelier art than classical Athens: it certainly con-
trolled greater masses of concentrated capital. It does not how-
ever seem to have possessed much of the special Hellenic inspira-
tion. And the village communities of the mainland, whether of
Northern or Southern origin, cannot have been much above the
level of savagery. But the rise of Greece began from something
a little worse than the average level of barbaric Aegean societies.
It began, as I hope to show in the second of these lectures, in the
dark age which resulted when even these societies, such as they
were, fell into chaos.

Allowing for indefinite differences of detail, there seems to be
a certain primitive effortless level of human life, much the same
all the world over, below which society would cease to be; a
kind of world-wide swamp above which a few nations have built


what seem like permanent and well- weathered dwellings. Others
make transient refuges which sink back into the slough. La
nostalgie de la boue — 'home-sickness for the mud' — is a strong
emotion in the human race. one sees it often in individual life.
One can think of many instances in history: Hellenic king-
doms like that of the Seleucidae in Syria ; many provinces in the
decline of the Roman Empire ; the west of Asia under the rule
of the Turks ; the rush of reaction in ancient Egypt after the
religious reform of Amen-Hotep ; or, again, the many efforts
after higher religion in India, and the regular falling back of
each reformation into the same primitive slough.

Now, as Greek civilization rose from the swampy level of the
neighbouring peoples, especially the various pre-Semitic races
just behind the Aegean coasts, it could not shake itself clean all
at once. Remnants of savagery lingered on in obscure parts of
life, expurgated as a rule and made comparatively innocent, but
still bearing the mark of their origin. Such remnants, as a
matter of fact, tend to receive undue attention. The Greeks
themselves are puzzled at a strange practice. Herodotus says
that the explanation of it is sacred, and better not mentioned.
Pausanias describes it with an antiquarian's zest. Plutarch has a
comforting theory of its real allegorical meaning. Our own
friends the anthropologists, to whom all true Hellenists owe so
much, naturally revel in such things. They search antiquity
eagerly for traces of primitive man, for totems, cannibalism,
human sacrifice, and the like. The traces which they discover
are of the greatest value. But I think they have often mistaken
the reverberation of an extinct barbarity for the actual barbarity

What strikes one most in Greek society is not so much any
bad things that were actually done. Of course there were bad
things, and always have been in all societies. It is rather the
frightful proximity of worse things still. Practices that to us
seem like the scarce credible stories of a remote past were to
the fifth-century Athenian possibilities and even dangers. The
jungle grew thick and close all around them, and the barrier
between seemed very weak, very impalpable.

You will notice in the ordinary language of ancient writers a
characteristic which throws light on this aspect of Greek life.


Non-Hellenic nations are nearly always spoken of by their tribes
or races— ' Ethne '—Pelasgians, Macedonians, Phoenicians; the
Greeks are spoken of by their cities, or, what comes to the
same thing, by their islands— Milesians, Phocaeans, Eretrians,
Athenians. on the mainland it is the Polls or circuit wall that
forms the essential boundary of the nation ; in the case of the
islands, Samos, Naxos, Aegina, it is the equivalent wall of sea.
Every Greek community is like a garrison of civilization amid
wide hordes of barbarians ; a picked body of men, of whom each
individual has in some sense to live up to a higher standard than
can be expected of the common human animal. As the shield
is the typical weapon of the Greek warrior, so the wall is the
typical mark of Greek civilization. It is one of the facts that
most need remembering in order to understand the greatnesses
and the flaws of Hellenism, that it was represented everywhere
by a handful of men holding an outpost, men who wrought their
wonderful day's work in political and moral wisdom, in specula-
tion, in beauty of outward form and inward imagining, with an
ear ever open to the sternest of life's calls, and the hated spear
and shield never far out of reach. No wonder that the task was
too hard for them ! As a matter of fact, Greek civilization itself
was never for a long enough time well policed and organized,
its remoter villages were never thoroughly enough educated, to
make it secure, even in its central places, against some sudden
blind resurgence of the savage.

Take, for instance, the case of Human Sacrifice. The memory
of a time when human beings had been deliberately slaughtered
as a way of pleasing God runs through the literature of the fifth
century as of something far-off, romantic, horrible. We may
compare it to our own memories of the burning of heretics and
witches, deeds which we know to have been done quite lately, by
men very like ourselves, and yet deeds which we can scarcely
conceive as psychologically possible to any sane being. In just
the same way, to the earliest of the great Athenians, Aeschylus,
the sacrifice of Iphigenia is something monstrous, beyond under-
standing.^ The man who did it must have been mad. To
Euripides such acts are generally connected with a study of the

^ AtVxpo/iiT^Tiy rdXaiva irapaKona TTparoTirjfxwv, Aesch. A^". 222. But the whole

passage should be read.


worst possibilities of a savage mob, or of scheming kings led by
malignant and half-insane priests. In an interesting fourth-
century document, the dialogue called 'Minos', which is attri-
buted to Plato, human sacrifice is treated as the extreme of
what is ' to us unlawful ', and yet, the speaker insists, it was at
one time and among certain people ' the law ' ; and there are
rumours still, he adds, of strange sacrifices in the secret places of
Arcadian hills ! ^ It is the tone in which we might remind our-
selves, for instance, that even in the last decade or so women
have been tortured as witches in the Abruzzi or in Ireland. The
writer himself, and the society which he addresses, feel them-
selves entirely remote from such practices.

And yet how close to them on all sides this abomination
pressed, closer indeed than they knew! It is not only that it
continued throughout all antiquity to be practised in times of
great crises by all the barbarians of the Mediterranean coasts.
It is not only that we find Hipponax describing the ritual execution
of the phannakoi at Ephesus, a grotesque and possibly a some-
what cruel business which clearly was a sort of mock human
sacrifice. Hipponax was a satirist of the sixth century B. C,
with a liking for horrors, and Ephesus was a partially barbarian
town. But we find the thing creeping closer than that. In a
well-known passage of the Frogs Aristophanes ends up a
passage of comic abuse of certain persons much admired by his
opponents by saying that, ' in the old days, people would have
thought twice before using them as phannakoi'' — ' Scarecrows,'
shall we say .'' or ' Guy-Fawkeses ' ? The word means literally
' human medicines ', or ' scapegoats '. Late and careless writers
speak as if these pharniakoi were actually sacrificed. But for-
tunately we happen to have a fragment of an ancient third-
century historian, Ister, who explains what this odd business
really amounted to. Two persons, one for the men of the city,
one for the women, were led out as though to execution. They

* p. 315 b. He refers also to the descendants of Athamas as practising a
similar sacrifice. But there he is misinformed or, more likely, straining his
point in the argument. In the Athamas ritual the victim escaped. See texts
in Roscher's Lexicon. Mayer (ib. ii. p. 1509) compares a Pelops-Oenomaus
ritual in Rhodes, in which the sacrificing priest pursued the victim with a
spear, but was first blindfolded and had to run hand in hand with two small


wore necklaces, one of white figs, the other of black. They seem
to have been solemnly presented with cake and figs, and then
scourged and pelted out of the city — treated, in fact, very like
the Lion and the Unicorn, I hasten to add that the scourging
was done with little twigs and bulbs of the skilla^ or wild squill,
and the pelting with similar ineffective objects. The victims are
said to have been volunteers, and chosen for their ugliness ; and
various smaller details in the ceremony are meant to be grotesque
and absurd. At the end, the phannakoi were supposed to be
dead and their ashes were thrown into the sea. The ceremony
was an ' imitation ', says Ister, of a stoning to death.^

When did it become an imitation ? When was it, as it must
originally have been, a real stoning to death ? We cannot say.
The Human Medicine is the relic of a very ancient, very wide-
spread, pre- Hellenic barbarity, which the Greeks have not swept
altogether away, but have allowed to live on with its teeth

But the abomination creeps closer still. There Is a story
about Themistocles told by Plutarch on the authority of one
Phanias of Lesbos. Phanias wrote some 200 years after the
alleged incident, and some of the other stories he tells do not
command credence : for instance, the statement that once in the
Chersonese fish came down in the rain.^ Still the story, as he
tells it, is not incredible. And it exactly illustrates the points
which I wish to convey. 'When Themistocles as admiral was
making the chief sacrifice beside his flag-ship ' — this was in the
last crisis of the Persian invasion, just before the battle of Salamis
— ' there were brought up to him three prisoners, men of great
beauty, gorgeously arrayed and adorned with gold. When
Euphrantides the prophet' — there is sure to be a prophet in
such a business ! — ' saw them, since the holy fire at that moment
burst into a great and brilliant flame, and there was a significant
sneeze on the right, the prophet clutched Themistocles by the

^ See Appendix A, on the Phannakoi. The ritual was probably a charm
for ripening figs; see Paton in Rev. Arche'ologiqne, 1907, p. 51. He argues
that Adam and Eve were (PapixaKoi, The word seemed in Greek to be the
masc. of ipdpfiaKov, 'medicine' ; but it was probably a foreign word. Hence
the d in Ionic, as in Ad/jeTo? and other foreign words. In Attic the a is short
by analogy from <pdppaKov.

* I find that I was wrong to doubt Phanias's word here. There had been
a waterspout at sea.


right hand and commanded him to dedicate the young men and
sacrifice them all, crying on the name of Dionysus Omestes (the
Raw-Devourer). " Do this," he said, " and there is deliverance
and victory for Hellas," Themistocles was horrified at the
prophet's strange and monstrous demand. But, as so often
happens in great crises and times of suffering, the multitude,
putting all their hopes in something irrational rather than in
reason, shrieked to the god with one voice, dragged the
prisoners to the altar, and, as the prophet commanded, com-
pelled the whole sacrifice to take place.' It is not said that
Themistocles performed the act. (Plut. Them, xiii.)

Now the evidence for the story is weak. Themistocles is both
the shadiest and the most maligned of great Greek statesmen.
The whole story may be an outrageous slander invented by his
enemies after his ostracism. But that scarcely alters its historical
significance. It was, apparently, a story actually told. It must
have been, if not true, at least possible —not beyond the bounds
of credibility to excited persons.

As a matter of fact, it is just on occasions like this that
human sacrifices have most tended to occur : in a disorganized
army or a rabble full of fear, egged on by some fanatical priest or
prophet. There were bloody doings in Rome when the fear of
Hannibal was strong, judicial murders of vestal virgins, buryings
alive of ' Gallus et Galla, Graecus et Graeca ' in the Forum
Boarium. (Livy, xxii. 57.) There was a great burning of Jews,
we may remember, after the earthquake of 1 755 at Lisbon.

Perhaps the most tragic case, however, was the outbreak of
human sacrifice at Jerusalem in the seventh century, inspired by
the Imminent terror of Assyria. Jews who had been taught to
believe that Yahweh was their only refuge saw, or seemed to
see, with despair that their sacrifices were availing nothing.
They must give Him more : give Him anything in the world, if
only He will avert the horror of an Assyrian conquest, with its
pyramids of heads and Its prisoners flayed alive. Looking about
them, these unhappy devotees saw the human sacrifices of Tyre
and Sidon, and knew that there was still one thing which they
might offer. No wonder Yahweh did not hear them, when they
were giving less than the heathen gave ! So began the burnings
of children at the tophet in the vale of Hinnom. Of course the


practice was denounced by the prophets, and comparatively soon
ceased. The point to observe is that in Greece, and it would
seem in Greece alone throughout classical times, we find no
parallel to this kind of thing. A desperate attempt was made
by the superstitious party to force a crime of the sort upon
Pelopidas, in the terrible moments before the battle of Leuctra.^
But it failed. Human sacrifice was barbaric, not Greek. If the
Themistocles story is true, that one bloody outburst of super-
stitious fear stands alone. There were other occasions on which
all the conditions for such a deed seem to have been present.
Think of Xenophon's Ten Thousand after Cunaxa: think of
Nicias's army after the last battle before Syracuse. All the con-
ditions for the thing are there; but not the thing. The very
idea is incongruous to one's conceptions of Nicias or Xenophon.
— That is Hellenism.

Human sacrifice, then, is one of the barbarities which Hellenism
successfully overcame. It was either abolished entirely or else,
as in the case of the pharmakoi at Athens, reduced to some
harmless ceremonial which satisfied religious conservatism with-
out inflicting much harm on human beings.

But there were other strongholds of the primitive beast in man
which even Athens was not powerful enough to conquer. To
take three points : we find among the Greeks the institution of
slavery, fixed and unshaken ; women in a markedly subject con-
dition as compared with our own times, though far removed
again from the seclusion of the East ; and lastly, proceeding
partly from the institution of slavery, partly from certain forms
of military organization, some startling phenomena of what we
should call unchastity in the relations of the sexes. And then
we imagine that these things are characteristically Greek ! They
are just the reverse. They are the remnants of that primaeval
slime from which Hellenism was trying to make mankind clean.

The Greeks are not characteristically slave-holders. All the

* See Appendix A. The case in Philostratus, Vit, Apol. iv. 10, where the
thaumaturge ApoUonius of Tyana, being at Ephesus during a plague,
recognized a certain deformed beggar as being a demon of pestilence, and
set the crowd to stone him to death, was a horrid act on the part of an
unauthorized mob, not a dehberate human sacrifice approved by the law.
But the Asiatic cities were terribly infected with barbarism by the tjme
of Nero. The incident has elements of the pharniakos rite in it.


world held slaves, and had always done so. The Greeks are
characteristically the first human beings who felt a doubt or
scruple about slavery ; who were troubled in mind by it, who
thought, wrote, schemed, in the face — as far as we can judge — of
absolutely overmastering social needs, to be rid of it, some two
thousand years before it was abolished in Christian Europe. I
do not refer specially to the efforts of isolated reformers. The
Cynics, we know, condemned slavery root and branch. The
Stoics and certain religious organizations from the fourth century
onward refused to recognize its existence, and professed to count
all men free. Euripides was troubled by it, and can scarcely get
the subject off his mind. The sophist Alcidamas seems to have
made a preaching tour round the Peloponnese to induce all states
to combine in a general emancipation ; and, curiously enough,
was not murdered. But the tone of the non-reforming writers is
equally interesting as evidence. Homer, though of course no
thought of doing without slaves ever crosses his horizon, speaks
always of slaves with a half-puzzled tenderness. Slavery is to
him a terrible thing that may happen to any man, and will ' take
away half of his manhood '. The heroes are as courteous to the
slaves, Eumaeus and Eurycleia, as to one another. Plato, bred
in a far from democratic circle and generally in protest against the
ideals of the great sophists of the fifth century, does not care to
denounce slavery. In his ideal Republic he abolishes it silently
by merely constructing a state without slaves. In the Laws^
written in his old age, when the cloud of reaction had settled
darkly upon his mind, he accepts it as an existing fact and makes
elaborate regulations for the protection both of slave and of
master.^ The attitude of his opponents, the sentimental demo-

* See Laws, pp. 777-8. ' If slavery " takes away half a man's manhood "
how is one to deal with slaves ? Some masters utterly mistrust their slaves,
treat them like wild beasts, with whips and scourges, till they make them
many times worse than before. . . . No doubt the human animal is ill-
tempered, and not at all easy to manage, when you introduce " the necessary
distinction of servant and master ". It is a bad business ; the only rules
perhaps are not to have slaves of the same country or the same language,
and then to be scrupulously just in dealing with them, more so tha7iivitlt your
eqtials. The only test of true justice is the way a man behaves to those whom
he can wrong with impunity. only the unstained can sow seeds for virtue" .. .
that is a rule to be remembered by every master and prince and strong man
in dealing with those weaker than himself.' It will be noticed that Plato
does not draw much distinction between * servant ' and * slave '. He is


crats, can perhaps be deduced from the beginning of his dialogue,
Euihyphro, or on Piety. The man who gives his name to that
dialogue is satirized as a type of the pious and ultra-superstitious
Athenian democrat. When Socrates meets him, Euthyphro is
going to Athens to prosecute his own father for homicide,
because the said father has caused, though not intentionally, the
death of a slave who had killed another. Euthyphro has been
apparently on the best of terms with his father ; he admits that
he had great provocation, and that the slave probably deserved
to die. But he will not allow a slave to be murdered any more
than another man : and, what is more, though he expects to be
laughed at and thought ' mad ', he is confident, if he can once
get a hearing, of winning his case.^ The father, I should remark
in passing, would not be put to death.

It is unfortunate, perhaps, that our principal representative of
ancient Greece upon this question should be Aristotle. Aristotle
is, like Plato, somewhat anti-democratic ; and, unlike Plato,
devoted to common sense. It is his common sense, perhaps,
that obscures his vision most. He saw that in the existing state
of society slavery was a necessary institution. Its abolition
would have meant anarchy, perhaps famine. And Aristotle does
his best to show that the necessary institution is also just and
' according to nature '. It is the same line that was adopted by
the fathers of the early Christian Church.- Some men are born
to obey, others to rule. Put down a dozen Greeks in a bar-
barous country : in a few months you will find the Greeks giving
orders and the natives obeying them. But his arguments do not
matter so much. The important thing is that he found it
necessary to argue. Slavery could not, to a thoughtful Greek,
simply rank as an accepted thing. No doubt Aristotle had
a solid majority behind him : a majority composed of plain men
who had no intention of seeing their business hampered by
philosophers, and doubtless of those same obscurantists who

perhaps more troubled than most moderns by the existence at all of servants
and masters, though far less troubled by the existence of slavery proper
as a form of service.

^ Observe how Euthyphro extracts a high moral lesson from the most
revolting myths of Hesiod : 'wrong-doing must be punished, however high
the offender. Zeus did not spare even his own father.'

"^ Cf. Susemihl and Hicks, Ar. Politics, p. 24, n. 4.

2760 Q,


afterwards prosecuted him for impiety : not a majority of
philosophers nor idealist democrats. The two most influential
schools, Cynics and Stoics, stood on the other side. The
popular writers of the New Comedy ^ appealed to the public
with sentimental denunciations of the unnatural thing.

I do not in the least wish to deny that the slave-trade assumed
enormous importance in Greece. The slave-trade in later
antiquity was largely in the hands of the maritime Greek cities,
just as in the seventeenth and eighteenth centuries it was in the
hands of England, and for the same reason : because the slave-
trade went with the general carrying trade. Polybius counts
among the first necessaries of life for a large town ' cattle and
slaves'.^ Wheat is mentioned as secondary. And it stands to
reason that, wherever one set of men have had absolute power
over another, there must have been cases of extreme cruelty.
One should remember, however, that Athens, the most Greek
part of Greece, was remarkable for her gentleness to the slave
population. It was part of her democratic ideal. Her friends
praise her, her critics and enemies ridicule her, for making her
slaves indistinguishable from free men.^ That is something.
But I think the main point which distinguishes Greece from other
ancient communities, here as elsewhere, is not something actually
achieved, but something seen and sought for. In Greece alone
men's consciences were troubled by slavery, and right down
through the centuries of the decadence, when the industrial slave-
system ruled everywhere, her philosophers never entirely ceased
protesting against what must have seemed an accepted and
inevitable wrong.

' Cf. Anaxandrides, fr, 4, Philemon, fr. 94 (Kock) : especially how God

eXevdipovs cttojjwc TrdvTas rfj (fyvaeiy
dovXovs 8e fifTfiTorjaeu t) irXfove^ia.

(' covetousness transformed them into slaves ')•

^ iv. 38 Ilfjos fiiv Tcis dvayKaias rod /3tou ;^petnf to t( dpennnra Koi to tS>v tis ras
dovXfias dyoiJtvav aapaTcou nXrjdos — odious language, certainly.

* For instance, [Xen.] Respub. Athen. i. 10 ff. (hostile) ; Dem. Phil. iii. 3
(friendly) ; Plato, Rep. 563 B (satirical on the licence and self-confidence
of slaves, male and female, in a democratic state). on the torture of slave
witnesses, see Appendix B. The best recent discussion of Greek slavery is
in A. E. Zimmern's Greek Co7)Wionwealth ; see also his articles in the
Sociological Review, Jan. and April, 1909. He distinguishes 'apprentice
slavery ' and 'chattel slavery' ; in Greece we have chiefly the former. [See
also Heitland Agricol i, 1921, esp. final chapter.]


The Greeks were not characteristically subjectors of women.
They are the first nation that realized and protested against the
subjection of women. I speak, of course, of nations in some
state of social complexity. For in primitive agricultural com-
munities the women who worked in the fields were in most ways
as free as men. on this question, again, I should not lay stress
on the evidence of the isolated reformer. We all know how
Plato in the Repiiblic preached the complete emancipation of
women from all artificial restrictions whatever. But some time
before Plato other philosophers,^ and well-known philosophers,
must have advocated the same ideas, because we find all the
regular ' Woman's Right ' conceptions ridiculed in Aristophanes
considerably before the Republic can have been published.
And there is this to observe, unless my impressions deceive me :
Aristophanes, a strong conservative writing broad comedy for the
public, seems quite to understand the ideas that he is handling.
He treats them as funny, as offering material for scurrilous jokes,
but not in the least as things unheard of or incomprehensible.
He understands his opponents better than, for instance, Mary
WoUstonecraft was understood by the writers of the Anti-Jacobin.
Before Aristophanes, again, there was Euripides, studying the
woman's case with persistent insight and eloquence. Euripides
was a genius too extraordinary to be useful as evidence of what
his average contemporaries thought ; except, indeed, of what
they must have thought after he had spoken. But consider for
a moment the whole magnificent file of heroines in Greek tragedy,
both for good and for evil, Clytemnestra, Antigone, Alcestis,
Polyxena, Jocasta, even Phaedra and Medea: think of the
amazing beauty of the Daughters of Ocean in the Promethezis^
and of the Trojan Women in the play that bears their name.
They are all of them free women, free in thought and in spirit,
treated with as much respect as any of the male characters, and
with far greater minuteness and sympathy. I doubt if there has

^ I strongly suspect, Protagoras. In Diog. Laert iii. 37 and 57 a statement
is quoted from Aristoxenus and Favorinus (no doubt using Aristoxenus)
that 'almost the whole of the Republic'' was taken from Protagoras's
Antilogica. Aristoxenus is a good authority, though spiteful. If this is at all
true, the Lysistrata (b. c. 41 1), and perhaps the Ecclesiazusae (b. C. 392 or 389 ?),
must have been aimed at ideas of Protagoras, as the later Gyfiaecocratiae
of Amphis and Alexis were aimed at those of the Reptiblic. Cf. Plato,
Rep. V. p. 457 b.

C 2


ever, in the history of the world, been a period, not even
excepting the Eh'zabethan age and the nineteenth century, when
such a gallery of heroic women has been represented in drama.
And such characters cannot surely have sprung out of a society
in which no free women existed.^

The third point is hard to discuss fully, but the explanation of
it is very similar. A great deal of ancient unchastity comes
directly from the institution of slavery : for female slavery was,
in large part, another — and perhaps on the whole a worse — form
of the custom of prostitution. Much, again, was a mere relic from
the religious ritual of pre-Hellenic peoples, and much was
a survival from the times when Greece was invaded and con-
quered by Northern tribes inadequately provided with women.
As for the myths, their immorality arises mostly from some very
simple misunderstandings. Every little valley community was
apt to count Its descent from some local ancestress and the tribal
god, a being who was often imagined In shapes not human, as an
eagle, a swan, or a river-bull. A time came when these various
local gods were gradually merged in the great Achaean master-
god, Zeus. The process was a thoroughly good and progressive
one ; but it had an unexpected result upon Zeus's reputation. It

^ Attic Law, in many respects primitive, is markedly so with regard to
women. A woman was always under the tutelage of the head of her family,
who would as a rule be her father, or, on his death, her eldest brother. She
thus had a constant protector against any maltreatment by her husband.
The guardian could annul the marriage and take her home. She also had
her own property. on the other hand, a bad guardian could torment a
woman almost as much as a bad husband can now : e.g. he could get money
from the husband by threatening to annul the marriage. The father could
transfer his right of guardianship to the husband, then the wife was under
her husband's ' coverture ', as now. When he died, the wife either fell under
the coverture of the next head of her husband's family, or could be left by
will to some person of her husband's — and in practice no doubt her own —
choice. A great deal of the Attic treatment of women strikes one as
exaggeratedly romantic. They were to be ' rulers of the hearth '. They
blushed at the sight of a strange male. To lose his wife's esteem was the
greatest blow that could befall an honourable man. (The man in question
risked losing it by being caught hiding under a bed to escape the tax-
gatherer. — Dem. Androt. 53.) Epicharmus the poet was actually fined, in
Syracuse, for making a broad joke in the presence of his wife. one is
reminded of the Attic vases in which men are freely caricatured or treated
realistically, but women nearly always idealized. Family life must have been
extremely correct, to judge by the rarity of cases or mentions of adultery in
our rather plentiful law-court literature.— on this subject I can now refer to
Prof. Vinogradoft"'s Histo7-ical Jurisprudence (Oxford, 1922), especially vol. 2.


provided him with a collection of human consorts, and of strange
disguises, which caused much veil-drawing on the part of the
religiously-minded and much open laughter among the profane.
The same sort of explanation applies to those few elements in
Greek myths or ritual which strike one as cruel. They are
nearly all of them little hard deposits of ancient barbarity left in
the outer strata of Hellenism. Take the Marsyas story. The
Greeks, when they penetrated to the town of Celaenae, deep in
the heart of Further Phrygia, found a local tradition how a native
god had flayed alive the native hero or king, Marsyas. The
origin of the myth is not certain. Dr. Frazer takes Marsyas for
one of his primitive vegetation-kings, who were slain periodically
as the harvest is slain, and their skins or some similar relic some-
times preserved till the next year.^ It may, again, be a remem-
brance of some Assyrian conquest ; for the Assyrians when they
conquered a place often expressed their satisfaction by flaying
their prisoners alive. However that may be, the guides who
showed the Greeks round Celaenae, wishing to call their god by
some name which would be intelligible, had called him Apollo.
Most barbarian gods were either Apollo or Heracles. So the
hideous story takes its place on the remote outskirts of Greek
myth, a thing that was perhaps never believed, and would no
doubt have been forgotten had not the academic sculptors of the
fourth century made use of the mythical ' flayed man ' to illus-
trate the distribution of the human muscles. It is the same with
a dozen other cases. At Apamea, quite close to Celaenae, the
Asiatic population kept up a very ancient rite of sacrificing divers
beasts by burning them alive. The Syro-Greek Lucian describes
the business as something curiously barbarous and uncanny.'^
These things are in no sense characteristically Greek. They are
remnants of the state of things which the highest Greek civili-
zation up to the end of the fifth century B.C., a small white-hot
centre of spiritual life in a world of effortless barbarism, tried to
transform and perished in the attempt.^

' AttiSy Adonis, and Osiris, chap. v.

^ De Dea Syria, 49. Something similar, however, occurred at Patrae in
Achaia. Cf. Paus. vii. 18, 11.

' I will not discuss a third view, the Greek as a Levantine. Many very
good writers make use of this conception, but I think that, if pressed, it is
misleading. The much-abused modern Levantine owes his general bad name


It is then from this pomt of view that I wish to discuss certain
parts of Greek poetry : as a manifestation of the spirit of upward
striving in man, which we roughly describe as Progress. But
here a further question suggests itself. I feel that many among
my hearers, especially perhaps among those who care most for
art and for poetry, will protest against regarding poetry from
this point of view at all. Science, they will say, progresses : but
poetry does not. When we call a poem immortal, we mean that
it is never superseded : and that implies that poetry itself does
not progress.

This doctrine, when rigidly held, is apt, I think, to neglect the
very complex nature of most of the concrete works of poetry.
One may gladly admit that the essential and undefinable quality
that we call poetry, the quality of being poetical, is one of the
eternal things in life. There is something in Homer and the
Book of Job which cannot be superseded, any more than the
beauty of a spring morning or the sea or a mother's love for
a child can be superseded. But, after all, this essential spirit has
always to clothe itself in a body of some sort, and that body is
made up of elements which admit of progress and decay. All
the intellectual elements of poetry are progressive. Wider fields
of knowledge may constantly be thrown open to the poet.
Beauty may be discovered in fresh places. There may be
increased delicacy, or at least increased minuteness, of observa-
tion. There is, most important of all, a possibility of change
in the emotions which form the raw material of poetry.
Wordsworth was not, perhaps, so great a poet as the Post-
exilian Isaiah, yet Wordsworth would not have howled for joy
that ' The mountains should be molten with the blood of Edom '.

to habits which come chiefly from historical causes. He is shifty, servile,
cowardly, because for centuries he has been held in subjection by somewhat
ferocious and markedly unintellectual aliens. He has had to live by dodging,
and has the typical qualities of a subject race. The ancient Greek was him-
self a ruler, and had on the whole the virtues and vices of rulers. The race
elements are not the same either. The Levantine, mixed as he is, is not
largely influenced by fair-haired conquering Northerners. Even the geo-
graphical conditions, though physically not much changed, are psychologi-
cally different. The Greeks are still the sailors and traders of the Levant.
But what is now petty huckstering in obsolete sailing-boats was then the
work of great adventurers and leaders of men. So that its moral effect on
the sea-folk was different. (I should add that, as far as my personal know-
ledge goes, I do not agree with the ordinary wholesale condemnation of the


And, still more certainly, the writers of Isaiah would have been
utterly incapable of taking any interest in the subjects of most
of Wordsworth's poems. Poetry, in this way, can be taken both
as evidence of the progress attained by a society, and as a
force in its further progress. Indeed, the best poetry provides
sometimes the strongest, because the most subtle and unsuspected,
force ; and the most delicate, because the most living and
unconscious, evidence. The conscious moralist often seems
rather stupid and arbitrary — he is certainly an unpopular
character — and the conscious legislator perhaps worse. The
poet has over both of them the immense advantage that he is
not trying to say what he believes to be good for other people,
or what he believes that they believe to be good for them, but is
simply expressing what he himself loves most.

But what I am most concerned with now is a rather different
point. I want to suggest, first, that the mere interest in human
progress in general is a possible source of poetical inspiration,
a source quite as real and quite as poetical as any other. And
secondly, that this particular source of inspiration is rather
unusually strong in Greek poetry.

Many critics speak as if for a poet to be interested in progress
was a sort of disgrace or a confession of prosiness. I disagree ;
I think human progress may be just as much a true inspiration to
a poet as the lust of the eye or the pride of life. Of course it is
not so to all poets : there is very little of it in the final stages of
Homer, little in Pindar and Catullus, just as there is little in
Shakespeare or Chaucer. on the other hand, it is the very
breath of life to Aeschylus, Euripides, and Plato, as it is to
Shelley or Tolstoy.

Let us take as an example the last work of Condorcet, written
by him in hiding when condemned to the guillotine. He first
intended to write an answer to his false accusers and a justification
of his political career. And then, in the face of death, that dis-
cussion somehow seemed to him less important ; and he preferred
to work upon the subject which he felt to be the greatest in the
whole world, Le Progyes de r Esprit Htmiain^ The Progress of
the Human Spirit. It is much the same subject, ultimately, as
that of the enormous work projected by the late Lord Acton — a
history of Human Freedom. An interest in this subject implies,


I think, at the outset an intense feeling of the value, for good and
ill, of being alive. Here we are, you and I and the millions of
men and animals about us, the innumerable atoms that make our
bodies blown, as it were, by mysterious processes somehow to-
gether, so that there has happened just now for every one of us
the wonder of wonders, a thing the like of which never has been
nor shall be : we have come to life ; and here we stand with our
senses, our keen intellects, our infinite desires, our nerves quivering
to the touch of joy and pain, beacons of brief fire, it would seem,
burning between two unexplored eternities : what are we to make
of the wonder while it is still ours ?

There is here, first, an interest in human life as a whole, and
secondly, a desire to make it a better thing than it is. That is,
we shall find two main marks of this spirit : First, what is properly
called realism ; though the word is so constantly misused that we
had better avoid it. I mean, a permanent interest in life itself,
and an aversion to unreality or make-believe. (This is not in-
consistent with an appreciation of the artistic value of convention.
We shall have opportunities of considering that point in detail.)
Secondly, a keen feeling of the values of things, that some things |

are good and others bad, some delightful, others horrible ; and a
power of appreciating, like a sensitive instrument, the various
degrees of attraction and repulsion, joy and pain.

Here we run upon one of the great antitheses of life, and one
which, it seems to me, is largely solved by the progressive, or I
may say, by the Hellenic spirit ; the antitheses between asceticism
or Puritanism on the one hand, and the full artistic appreciation
of life on the other. In real life and in literature these two spirits
fight a good deal. But both, of course, are parts of one truth
If life is to be enriched and ennobled, you must first of all have an
appreciation of life. A man who refuses to feel and enjoy life
destroys it at its very heart. on the other hand, any strict Puritan
can always point to an immense amount of wreckage produced
by great appreciation of the joys of life, and also to a large
amount of good safe living produced by the principles of avoiding
pleasure, dulling the desires, and habitually pouring cold water
into your own and other people's soup, ' to take the Devil out of
it.' There is plenty of opportunity for dispute here in real life.
In speculation there seems to me to be none. The truth simply


is that in order to get at one desirable end you have to sacrifice
another. The artistic side of man insists upon the need of under-
standing and appreciating all good and desirable things: the
ascetic side insists on the need of a power to resist, a power even
to despise and ignore, every one of them, lest they should hinder
the attainment of something better.

The combination of these two, the appreciation of good things
and the power to refuse them, is characteristic of the spirit of
progress. I think most scholars will admit that it is also eminently
characteristic of Greek civilization. The enjoyment and appreci-
ation of life is too deeply writ on all Greek poetry to need any
illustration, though one might refer to the curious power and
importance in Greek life of two words, KdWos and Hocpia, Beauty
and Wisdom ; to the intensity of feeling which makes 'EXiris,
Hope, or ToXfia, the Love of Daring, into powers of temptation
and terror rather than joy : to the constant allegorizing and trans-
figuration of those two gods of passion, Dionysus and Eros.^ But
the principle of asceticism was at least equally strong. Whether
we look to precept or to practice, the impression is the same.
In practice a respectable ancient Greek allowed himself some
indulgences which a respectable modern would refuse : but for
the most part his life was, by our standards, extraordinarily
severe and frugal. To take one instance. Hippocrates, the great
fifth-century physician,- says in one passage that many doctors
object to their patients having more than one meal in the twenty-
four hours : but for his own part, he thinks that, though to most
healthy people it makes no difference whether they have two
meals or one, still some slow digesters cannot stand more than
one, while other delicate persons are positively the better for two !
Our healthy persons have four ; and our invalids fall not far short
of a dozen. All the great schools of philosophy, again, were in
various degrees ascetic. The general admiration felt by the
ancients for every form of frugality and hardihood strikes one as
altogether extreme. The praises of Sparta show us how severity
of life, coupled with courage, sufficed in the popular judgement
to cover a multitude of sins. Yet Greek asceticism is never like

^ These points are excellently brought out in Cornford's Thucydides
Mythistoricus, chap, ix, xii, xiii.
^ De Vet. Med. io=p. 593, fAopoa-Lredv and apta-Trjv are the alternatives.


Eastern asceticism. The East took its asceticism in orgies, as it
were ; in horrors of self-mutilation, bodily and mental, which are
as repellent in their way as the corresponding tempests of rage
or of sensuality. Greek asceticism, though sometimes mystical,
was never insane. It was nearly always related to some reason-
able end, and sought the strengthening of body and mind, not
their mortification.

One cannot but think, in this connexion, of that special virtue
which the early Greeks are always praising, and failure in which
is so regretfully condemned, the elusive word which we feebly
translate by ' Temperance ', Sophrosyne. The meaning oisophro-
syne can only be seen by observation of its usage — a point we
cannot go into here. It is closely related to that old Greek rule
of MrjSep dyav, Nothing too inuch^ which seems to us now rather
commonplace, but has in its time stayed so many blind lusts and
triumphant vengeances. It is something like Temperance, Gentle-
ness, Mercy ; sometimes Innocence, never mere Caution : a temper-
ing of dominant emotions by gentler thought. But its derivation
is interesting. The adjective a-axppcov or cra6(l)pa)u is the correla-
tive of 6Xo6(f)pcov, a word applied in early poetry to wizards and
dangerous people. 'OXoocppoiv means ' with destructive thoughts ',
aoxppcov means ' with saving thoughts '. Plutarch,^ writing when
the force of the word was dead, actually used this paraphrase to
express the same idea. There is a way of thinking which destroys
and a way which saves. The man or woman who is sqpkron
walks among the beauties and perils of the world, feeling the
love, joy, anger, and the rest ; and through all he has that in his
mind which saves. — Whom does it save "i Not him only, but, as
we should say, the whole situation. It saves the imminent evil
from coming to be.

It is then in this light that I wish to consider certain parts of
Greek poetry : as embodying the spirit of progress, that is, of
both feeling the value and wonder of life and being desirous to
make it a better thing : and further, with that purpose in view,
as combining a spirit of intense enjoyment with a tempering
wisdom, going into seas of experience steered by Sophrosyne.

1 Z)e Trafiquillitate, 470 D vovv aatTrjpia <f>povovvTa. 'oXoocppap is used of
Minos, Aietes, Atlas— also of a hydra, lion, and boar.



If we regard Greece as the cradle of European civilization, we
cannot help some feeling of surprise at its comparative lack of
antiquity. True, we have evidence of a civilization existing in
Crete and the Islands of the Aegean as far back as the end of the
Stone Age. But, for one thing, our knowledge of this civilization,
though based on abundant and skilfully sifted material, remains
enigmatic and conjectural, inasmuch as it depends upon our
interpretation of the stones, not upon literature : and, what is
more important, it is emphatically not the civilization that we call
Greek. I do not mean only or especially that the builders of
the earliest Cretan palaces were, as far as we can judge, of differ-
ent race and language from the Greeks. I mean that this civiliza-
tion, so far as we know it, has few or none of the special marks
that we associate with Hellenism. But of that hereafter. In any
case there lies between the prehistoric palaces of Crete, Troy, or
Mycenae, and the civilization which we know as Greek, a Dark
Age covering at least several centuries. It is in this Dark Age
that we must really look for the beginnings of Greece.

In literature and in archaeology alike we are met with the
same gap. There is a far-off island of knowledge, or apparent
knowledge ; then darkness ; then the beginnings of continuous
history. At Troy there are the remains of no less than six cities
one above the other. There was a great city there before
2000 B.C., the second of the series. It used red ware for its
pottery, wherever the first city had used black, and it seems to
have traded the newly discovered copper of the South-East
against amber from the Baltic. In this second city there was dis-
covered a fragment of white nephrite, a stone not hitherto found
anywhere nearer than China, and testifying to the distances
which trade could travel by slow and unconscious routes in early


times. That city was destroyed by war and fire ; and others
followed. The greatest of all was the sixth city, which we may
roughly identify with the Troy of Greek legend. Of this city
we can see the wide circuit, the well-built stone walls, the terraces,
the gates, and the flanking towers. We have opened the treasure
houses and tombs, and have seen the great golden ornaments
and imports from the East. Then we see the marks of flame on
the walls : and afterwards what ? one struggling attempt at a
seventh city ; a few potsherds to mark the passage of some genera-
tions of miserable villages ; and eventually the signs of the Greek
town of New Ilion, many hundreds of years later and well within
the scope of continuous history.

It is the same in Crete. City upon city from prehistoric times
onward flourishing and destroyed ; palace upon palace, beginning
with the first building of Cnossos, in a peculiar non-Hellenic
architecture ; proceeding to those vast and intricate foundations
in which Sir Arthur Evans finds a palace, a citadel, and a royal
city round about, the growth and accumulation of many hundreds,
perhaps thousands, of years. The ornamentation of the walls is
there, telling of the rise and decay of a whole system of decorative
art : fragments of early religion, the Bull-God or Minotaur
seated upon his throne ; the ' horns of consecration ' bristling
everywhere ; the goddess UoTvia 6r)p5iv, Queen of Wild Beasts,
now bearing a dove upon her head, now twined with serpents ;
sometimes in human shape, sometimes a mere stone pillar erect
between her rampant lions : sometimes a monstrous fetish.
There is the Divine Battle-Axe, that Labrandeus from whose
name the fable of the labyrinth seems to have arisen ^ : a being
who has not yet reached human shape or separate existence as a
' God ', but exists simply in the ancient bronze axes, scores of
which remain driven into the rock of the Dictaean cave, over-
crusted with a stalactite growth of stone, testifying to a worship
forgotten and uncomprehended. There are porcelains reminding
one of Babylon, ornaments from Egypt, marks of a luxurious
king's court, a gaming-table inlaid with gold and coloured

^ See, however, on the Labyrinth, Lecture V below, p. 138, note^, and
especially Burrows, The Discoveries in Crete, pp. 107-32. He connects
\nj3vpiv6os with \avpa and Aavpfiov. (So, I believe, did Wiedemann.) The
catastrophe which I am specially considering is, of course, that of ' Late
Minoan III'.


marbles, women acrobats, bull-fights, or perhaps, if we look
close, something more barbaric than bull-fights — boys and girls
thrown for the ' Bull of Minos ' to gore : then shattered gates,
flame-blackened walls and evidences of calamity, a feeble pulsing
of life outside the ruined palaces, and afterwards silence.
Centuries later a new Crete emerges, a Dorian island, rigid, self-
centred, uninfluential, in the full light of Greek history.

It is the same with the cities of the Argive plain, Mycenae and
Tiryns. They possessed less importance, and were inhabited
for a less vast stretch of history, than the cities of Cnossos and
Troy.^ But the treasures yielded to the excavator, especially in
Mycenae, are very great in proportion to the importance of the
town, and the historical problem is simpler. We all know the
Mycenaean remains : the Lion Gates, the earlier shaft graves,
and the later vaulted graves ; the remains of mummified kings ;
the skeletons in masks of gold, with their weapons, their drinking-
bowls, and sometimes the ashes of burnt sacrifice lying beside
them. And in the end, as in Troy and Cnossos, the marks of
flame upon the walls, traces of a dwindling population still
hovering about the old town, and quickly degenerating in the
arts of civilized life ; and then a long silence.

Such is the evidence of the stones. And that of literature
corresponds with it. There is an extraordinary wealth of
tradition about what we may call the Heroic Age. Agamemnon
king of Mycenae and Argos, Priam king of Troy, and the kings
surrounding them, Achilles, Aias, Odysseus, Hector, Paris, these
are all familiar household words throughout later history. They
are among the best-known names of the world. But how
suddenly that full tradition lapses into silence ! The Epic Saga
— I mean the whole body of tradition which is represented in
Epic poetry — the Epic Saga can tell us about the deaths of
Hector, of Paris, of Priam ; in its later forms it can give us all
the details of the last destruction of Troy. Then no more ;
except a few dim hints, for instance, about the descendants of

It is more strange in the case of Mycenae and Sparta.

* Under Tiryns an earlier city has recently been discovered. See W.
Dorpfeld, Athen. Mitth. 1907.


Agamemnon goes home in the full blaze of legend : he is
murdered by Aegisthus and Clytemnestra, and avenged by his
son Orestes : so far we have witnesses by the score. But then ?
What happened to Mycenae after the death of Aegisthus ? No
one seems to know. There seems to be no Mycenae any more.
What happened in Sparta after Menelaus and Helen had taken
their departure to the islands of the blest ? There is no record,
no memory.

In Crete there is less tradition altogether. one great name,
Minos, forms the centre of all Cretan legends. Minos is never
quite flesh and blood, like the Homeric heroes, Agamemnon or
Achilles. He is almost like that more than shadowy personage,
Creon, whose name means ' ruler ' and who appears in so many
myths of the mainland whenever a mere ' ruler ', and nothing
more, is wanted. We meet Minos in many different generations,
in many different characters, most of them probably possessing
some historical significance. He is the son of Zeus, or, still more
august, not the son but the ' gossip ' or familiar friend of Zeus,
a suitable position for the hereditary Priest-King, himself an
incarnation of the Cretan Bull-God.^ He is the Just Judge of
the Underworld, as befits the great legislator who was the
fountain of law to many islands and dependencies. Again, he is
the bloody tyrant of the Theseus myth, who gives the seven
youths and seven maidens to his man-slaying Bull. The reason
for that is painted on the walls of his labyrinthine palace, showing
— apparently — Minoan ladies looking on while the Bull is played
with and maddened by skilled performers till he is let loose on
his victims. Then again we have the boasting Minos of the
Bacchylides poem ; the mere royal father of the lost Cretan
prince, Glaucus ; the father or husband of many guilty and
romantic Cretan heroines, Phaedra, Ariadne, Pasiphae.

After Minos, what is there .'' Idomeneus in the Iliad^ a hero of

^ Atof /xfyaXov oapiarr^s, t 179: cf. Plato, Minos 319 D. See below,
Lecture V, p. 136 and note there. I suspect that Minos was a name like
* Pharaoh ' or * Caesar ', given to all Cretan kings of a certain type, and, further,
that the king was held to be the personification or incarnation of the Bull-God.
As to the evidence for a Minos existing at different dates, Prof. Burrows
remarks that the Parian Marble puts Minos in the fifteenth century B.C. and
also in the thirteenth, and that Diodorus (iv. 60) and Plutarch ( Vit. Thes. 20)
tell a similar story. See, however, Ridgeway, ' Minos the Destroyer, &c.',
Proceedi7igs of Brit. Acad. IV.


the second rank, a man grey on the temples, too old to run but
not too old to deal terrific blows, and withal a litde grim and cruel.
There is also another significant feature. Nearly all the false
stories told by Odysseus in the Odyssey are tales of Crete, as is
also the true romance of the life of Eumaeus the swineherd.
The later proverb associated Crete with ' romancing '. Was it
that Crete was full of stories of a past greatness which to the
ordinary forgetful world seemed merely incredible, as Juvenal
afterwards sneered at ' Graecia mendax ' because Herodotus had
preserved more ancient history than he could believe ? So near
had Minoan Crete come to complete oblivion.^

In Thebes, as in Troy, the tradition is more intelligible because
it explicitly leads up to a catastrophe. Many problems require
to be cleared up about the Theban traditions, even after Bethe's
work upon the subject. The prehistoric remains, as we said
above, are not prominent or remarkable, chiefly, no doubt,
because the place was never left for a long time deserted. It is
with Thebes as with Argos, with Athens, with the many sites of
towns on the coast of Asia Minor and the Italian Riviera.
Continuous occupation has destroyed gradually and surely the
remains of every successive period. But the Theban traditions,
as preserved in literature, are particularly rich, and they lead up
clearly to our Dark Age or Period of Ignorance. There is first
a strange race, Cadmeans, the people of Cadmus, ' the Eastern
Man,' '^ in possession of the city. The tradition is clearly not of
their making, for they are credited with all the crimes and
pollutions in the calendar : especially sexual crimes, which people
always impute to their enemies. Three generations of the
Cadmeans, Laius, Oedipus, and the sons of Oedipus, between
them commit pretty well all the crimes that can be committed in-
side a family. Unnatural affections, child murder, father murder,

* Cf. Hdt. vii. 171. Crete had formerly been 'emptied' by an expedition
of Minos to Sicily. Then 'in the third generation after the death of Minos
came the Trojan wars. . . . After the return from Troy there came famine and
pest slaying both man and beast, and Crete was made empty a second time.
Then came the present Cretans'— i. e. the Dorian tribes — 'and inhabited it,
together with the sui-vivors.'

^ Heb. D"Ip qedein, the east. Greek tradition calls them ' Phoenicians ', but
it is not clear what that term exactly denotes. Ridgeway thinks they were
* red ' Thracians {Early Age, p. 629). Cf. his ' Who were the Dorians ? ' in
Anthr. Essays to E. B. Tylor (1907). See, after Eeloch and Berard, Burrows,
op. cit.^ p. 141 f.


incest, a great deal of hereditary cursing, a double fratricide, and
a violation of the sanctity of dead bodies — when one reads such
a list of charges brought against any tribe or people, either in
ancient or in modern times, one can hardly help concluding that
somebody wanted to annex their land.^ And this was doubtless
the case. The saga gives us full details up to the quarrel of
Eteocles and Polynices and the Expedition of the Seven Greek
Champions, the chariot-fighting on the plain and the assault on
the Gates. The seven were defeated : so far we hear all at
length. Then much more briefly, with much less reality, we
are told that their sons made another expedition and took
Thebes. That is, the citadel of the Cadmeans eventually fell,
and nothing more is said or known.

It is the same wherever we turn our eyes in the vast field of
Greek legend. The ' heroes ' who fought at Thebes and Troy
are known ; their sons are just known by name or perhaps a little
more : Diomedes, Aias, Odysseus, Calchas, Nestor, how fully the
tradition describes their doings, and how silent it becomes after
their deaths !

Let us consider these destroyed cities a little closer. We can
perhaps make out the kind of civilization on which their
greatness rested, and the causes of their fall. For observe
this : though we can see in some cases from the evidence of the
stones that these cities came at last to a violent end, it is by no
means clear that it was any definite shock of war which really
destroyed the Aegean civilization. There Is no tradition at all
that the realm of Minos was sacked In war ^ : no real tradition
of the sack of Mycenae. And even in the cases of Troy and

* There is also extant a simpler version, before the self-defensive slanders
had been developed, in which the heroes are slain at Thebes simply iiijXov
ivfic Ol8nr68no (Hes. Er£a, 162), in an honest cattle-raid.

" Mr. J. L. Myres reminds me of Plutarch's story of 'Tauros the sea-
captain ', who was the real lover of Pasiphae, and his sea-fight off Cnossos.
This is possibly a very faint echo of a real tradition ( Vi'L Thes. xix and
preceding capp.). There would be no great siege in any case, since Cnossos
and Phaestus were open unfortified cities ; their fall would follow quickly on
the destruction of the Minoan fleet. Sir A. Evans however doubts whether the
sack of Cnossos and Phaestus was the work of a foreign army at all {B. S. A.
xi. p. 14). He suggests that both Palaces may have been destroyed by
earthquake at the time of the great volcanic eruption which destroyed


Thebes the testimony is suspicious. The Epos must say that
Troy eventually was taken, but the Epos knows that Achilles did
not take it, but failed and was slain. A son of Achilles, a mere
replica of Achilles, has been invented to come afterwards and
take it. Of course the Iliad as it now stands implies the future
fall of the city, but it need not have done so in an earlier form.
Nor need the Odyssey. The disastrous returns of the Greek
heroes and the fall of the house of Agamemnon point rather to
an unsuccessful expedition than to a great conquest. And how
does it happen, one may ask, that so many Greek lays were based
on the subject of ' Wraths ', or quarrels between leading chiefs,
between Agamemnon and Achilles, Odysseus and Agamemnon,
Odysseus and Aias, Achilles and Odysseus .'' Does it not look —
I take the suggestion from Prof. Bury — as if there was need of an
excuse for some great failure ? At any rate the actual tale of the
Sack of Troy, though immensely influential in later literature,
does not seem to be recorded in any very early form of the saga.
And even incidents which have a special air of verisimilitude
about them, like the stratagem of the Wooden Horse,^ may
represent only a brilliant afterthought of what ought to have
been done. I lay no stress on this point, except to suggest that
it is curious, if the war really ended in success, that the great
national saga in its early forms should not tell of the success,
but only of disastrous ' Returns ', together with a quarrel, or
several quarrels, between the chiefs — incidents well calculated to
excuse failure.

Exactly the same thing is the case with the Theban tradition.
A great expedition against Thebes is well known to the Epos,
that of the Seven Chieftains, led by the far-famed Adrastus.
That expedition, we are told, was defeated and all the seven slain.
» only,' the story adds, ' Thebes did fall in the end. Some people
who cauie afterzoards took it.' The names of these later comers
are not very certain. They are only the ' Ekgonoi ' or ' Epigonoi ',^
the ' men-born -after ', more shadowy even than Pyrrhus-Neopto-
lemus, son of Achilles. The general result seems to me to suggest
that, in the first place, the Epic tradition of the Greeks knew of

^ I suggest that it may refer to a siege tower of the Assyrian type. My
translation of The Trojati Woinen, p. 86.
■'' "EKyovoi, Eur. Suppliatits, 1224. 'ETTiyoyot is of course the usual name.

8760 J)


certain Iieroic expeditions against Thebes and Troy, but knew
also of their defeat ; and secondly, this tradition had much later
to be combined with the fact that in reality Troy and Cadmean
Thebes had ceased to be. Can we see anything in the historical
conditions which makes such a hypothesis probable ?

I suggest, to put it briefly, that these great fortress-cities
depended for their greatness upon industry and commerce, and
that during the period of persistent barbarian invasions industry
and commerce were destroyed. They resisted successfully the
direct shock of war ; but were gradually undermined by poverty.
All of them, as a matter of fact, are situated at the junctions
of important trade routes. Crete, for instance, a rough and
mountainous island, credited by Strabo with ' some fruitful glens',
is geographically, in Sir A. Evans's phrase (/. H. S. xiv), 'the
stepping-stone of continents,' lying in the mid-route between
west and east,^ between south and north. The Cnossian treasury
records seem to speak of a large trade in oil and silphium, as well
as horses and cattle and precious metals. They had already
invented, it would seem, the talent and the drachma. The lines
from Phoenicia and the great Babylonian hinterland, from Egypt,
from Libya, all tended to join at Crete on the way to the West,
the Northern Aegean, or the Black Sea,^ Some centralizing
power then must have arisen in the island, and the maritime trade
of such harbours as Kydonia and Hierapytna — the east of the
island seems to have remained isolated — served to support the
great central city of Cnossos. Thebes, again, as Strabo explains,
commanded the roads between three seas, the Northern Aegean,
the Southern Aegean, and the Corinthian Gulf.

But let us consider the point more in detail in two cases where
it is not so easily seen.

Mycenae, as M. Victor Berard has well explained, is what is
called in Turkish a Dervendji\ that is, a castle built at a juncture
of mountain passes for the purpose of levying taxes on all traffic
that goes through. There is the rich plain of Argos opening
southward to the sea. At the north of it are mountains ; beyond

^ See also Hogarth's address to the Royal Geographical Society, 1906. A
road running north and south has since been discovered.

"^ p. 400, from Ephorus. See also Berard, Les PMniciens et VOdyssie,
i. 225 f. Compare, for what follows, pp. II f. (Mycenae) and 79 f. (Troy).


them the plain of Corinth and Sikyon opening on the Corinthian
Gulf. Among these mountains, at the north-east corner of the
Argive valley, with no sea near, and no arable land anywhere
about it, stands this isolated castle of Mycenae, thickly walled
and armed to the teeth. It is hard to see how such a place could
live, and why it needed such military preparations, until we observe
that it forms the meeting-point of a very ancient system of artificial
roads, cut and built of stone, and leading from the Argive plain
to the Corinthian, from the southern sea to the northern. If
Mycenae stood alone, she formed a sort of robber stronghold,
which lived by levying blackmail on all the trade that passed.
But almost certainly she did not stand alone, and M. Berard's
explanation is only a part of the truth. Agamemnon was master
of a large realm, including ' all Argos and many islands ', and
Mycenae stood at the centre of it, able to keep open the trade
routes between the northern and southern seas, and ready to strike
with horse and foot in any direction where defence was needed.
Paved roads, as Eduard Meyer has pointed out, are meant for the
passage of chariots, not merely of caravans ; and whether or no
the commerce along these roads was anything considerable — a
point which depends largely on unsettled questions about Corinth
— Dr. Leaf is clearly right in regarding Mycenae as built more
for war than for commerce.^

M. Berard's explanation of Troy is even more instructive, though
it also looks rather different on further consideration. It has to
be modified by the observed fact that Troy does not show great
affinities with the islands, and does with Thrace and its own
hinterland. But there is more in it than that.

Six cities were built on that particular site, and six destroyed.
There must have been some rare attraction about the place, and
some special reason for destroying the cities built there, Greek
legend, in speaking of the destruction of Troy by Agamemnon,
always remembered that it had been destroyed before, though it
ran all the previous expeditions into one — when old Telamon
rose from his rest in Salamis, and gave himself to Heracles

^ See Leaf, Hojner and History, 220-8: E, Meyer, Gesch. d. Alt. ii. 170,
180. Leaf thinks there would be very little commerce on these roads, at any
rate if Corinth was not in existence in Mycenaean times, a point on which
the excavations are not conclusive. (Allen, Catalogue, p. 64.)

D 2


For the wrecking of one land only,
Of Ilion, Ilion only,
Most hated of lands.^

Tradition tells us that Heracles went ' because of Laomedon's
horses'^ and M. Felix Sartiaux has pointed out in his book on
the present domain of Troy that ' sa principale industrie est
1 elevage de chevaux '. The Trojans in the Iliad have, like Hector
himself, the special epithet of ' horse-taming ', and have almost a
monopoly of names compounded with -ittttos? But horses were
not the whole of Priam's wealth. Strabo (p. 680) speaks of ' gold
mines in Astyra near Abydos, of which little is left now, though
great slag-heaps and excavations are evidence of ancient working ',
But a larger stream of riches, and a far stronger cause of unpopu-
larity, lay well within reach of Priam's walls.

We know that in later times there was a vast body of trade
passing up the Hellespont, joining Mediterranean civilization
with that of the Black Sea. Obviously a city commanding this
trade would grow rich : but Troy does not seem at first sight
to be in the right position for commanding it. The older
city, Dardania, had lain higher up on Mount Ida, the Iliad tells
us (T 218), in safe retirement. But as the Trojans grew stronger,
or as they discovered a more tempting source of wealth, they
ventured nearer the sea. Yet even so Troy lies some miles inland
on the slopes of a hill commanding only a narrow swampy plain
with sea at each end of it. In modern times such a position is
not of much worth. But in the conditions of ancient seafaring it
was priceless.

Down the Bosphorus and the Hellespont there blows an almost
incessant wind and there flows an extraordinarily strong current.
If you bathe in the sultry heat down below Tenedos, near My tilene,
you may find yourself suddenly in swift and almost icy water

^ Eur. Troadt's, 806 (sentiment of the whole passage, rather than any definite
words) : cf. I24I Tpoia re noXfov (KKpirov fiKTOVfiivr).

^ E. 640 evex "'■irTruu Aaofiedouros, Eur. Tro. 8o6 ff. drv^ofifvos TrtoXtui'.

* See Sartiaux, Troie (1915), p. 130 : Prof. Grace Macuidy, C. Q. Jan. 1923.
'ItTTTcdaiioi in pi. occurs twenty-three times, always of the Trojans ; their allies
are QprjiKes 'nriTOTroXoi, naioits iniTOKopva-Tai, <tp{y€S iTnropcixoi. The horse is
said to have been introduced by the ' tumulus-people ' (Mongol nomads) from
the steppes of Asia in the third millennium B. c, when they first learnt to ride
and drive horses as well as to live on the milk of mares. See Myres in Grw-
bndge Ancient History, \, pp. 106 ff. on Trojan chariots see note p. 152.


sweeping straight from Russia. This current is at its strongest just
off Cape Sigeum, the promontory in front of Troy. At the
present time small steamers have some trouble in passing there,
and sailing ships can be seen waiting by the score under the lee
of Tenedos, till by utilizing stray puffs of favourable wind they
can tack round that difficult cape, and^ proceed by hugging the
eastern shore. In ancient times, when boats were small and
voyages short, their difficulties were much greater. The greatest
of all was the absence of galvanized iron tanks to hold drinking-
water. An ancient ship carried its water in heavy earthenware
jars, and if it was weatherbound for a few days on a waterless
coast the results were disastrous.

Now, M. Berard's view was that, though it does not look so
on the map, the plain of Troy is really an isthmus, or at least a
formation to which he can apply his famous ' Law of the Isthmus '.
Ancient cargoes were so light, and ancient ships so little accustomed
to long voyages, that in numerous cases where we should make a
longer voyage in order to avoid trans-shipment and 'breaking
bulk ' the ancients made straight for the nearest land, unpacked
their bales on to the backs of mules or men, and reshipped them
again at the nearest sea. on many isthmuses they even dragged
the ships overland rather than make a further round by water-
Thus, M. Berard suggested that merchants wishing to trade with
the Black Sea ports disembarked their cargo at the southern
end of the narrow swampy plain, carried it across on mules and
re-embarked it on the further side, paying heavy taxes to Priam
or Laomedon as they passed ; and that this is the origin of the
wealth of Troy. In reality, however, as Dr. Leaf has shown,^
this does not seem to have been the ancient practice. Traffic
from the south, if it wished to avoid the Etesian winds and the
Black Sea currents, took to the land much earlier, at Assos. By
St. Paul's day this old track had developed into a Roman road
from Troas to Assos, along which he travelled. He was then
going from north to south, so most of the party preferred to go
by sea, with the current (Acts xx, 13).

The real clue to the importance of Troy in Mycenaean times is
the fact that it ceased to be important as soon as there were
Greek settlements up the Hellespont and on the Black Sea.

^ Leaf, Troy, pp. 254-85.



When once there were friends along the coast, Greek ships passed
freely up to the Euxine, and allowed no blackmailing power to
grow up again at the mouth of the Hellespont or on the Black
Sea. Before that time, with the currents and wind against them,
and no fresh water except in the Scamander, Greek ships had not
been able to pass into the Hellespont except by permission of
Priam. In all probability they did not pass at all. The plain of
Tro}'^ was the actual meeting-place where the trade of the Euxine
touched that of the Aegean. Certainly the allies of Troy described
in Iliad B seem to live on a network of trade routes, and in
pondering on the great quantities of silver found in the Second
City one cannot but remember the allies or vassals who came ' from
far-off Alybe, where silver is born '} Priam's misfortunes were
so great that tradition is kind to him. But the perjuries and
extortions of Laomedon ring loud in legend. Was it simply
because the toll at the Hellespont was too oppressive to be
tolerated, that all maritime Greece felt involved in the oppression,
and volunteered to destroy the blackmailing citadel again and
again .'' Or was it, more simply still, that the position was so
valuable that one band after another of northern warriors,
Thracians, Dardans, Troes, Teukri, Phrygians, Achaeans, fought
for the possession ?

There are many problems still waiting solution about these
fortified centres of exchange, if I may so call them. How far did
they form a uniform empire or federation ? Was Mycenae normally
an outpost of Crete or an enemy of Crete, or when did it change
from the one to the other ? What relation did either of them
bear towards Troy, or towards the prevailing powers in Asia ?
Of what race or races were their kings ? How far was there a
conscious difference between the ' Minoan ' or Island race with its
sea-coast settlements and the less advanced masses of Anatolian
or ' Hittite ' peoples of the hinterlands ? In any case it is, I think,
perfectly clear that this Aegean civilization was not what we call
Greek. Its language was, as far as we can judge, not Greek.
Its art, though we can recognize in it many of the elements that
went to the making of Greek Art, was in itself not Greek. As
a matter of fact there were no Greeks in the world in those days,
any more than there were, let us say. Englishmen before the

^ B, 857 Tr]K66iv e^ ^A\v^r]s odt t dpyvpov earl yevidXt],


Angles came into Britain, or Frenchmen before the Franks
invaded Gaul. The Greek people was a compound of which
the necessary constituents had not yet come together.

We must recognize, however, that the existence of such rich
and important centres, dependent entirely upon sea-borne
commerce, argues both a wide trade and a considerably high
and stable civilization. We must not forget that piece of white
nephrite which, so archaeologists daringly assure us, came to
Troy all the way from China. And we must by no means regard
the masters of these cities as mere robber chieftains or levyers
of blackmail. Commerce dies if it is too badly treated ; and
Aegean commerce lived and flourished for an extremely long

These empires, if we may call them by so large a name, were
broken up by migrations or invasions from the north. In early
times, so Thucydides tells us, all Hellas was in a state of
migration.^ We hear of all sorts of migrant tribes ; of Hellenes,
Achaioi, and Pelasgoi ; of Carians and Leleges ; of Minyae ; of
the sons of Deucalion, Ion, Pelops, Danaus, and the rest. Most
of all we hear of the great migration of the Dorians,^ somewhere
about 1000 B. C. It is the habit of Greek tradition to remember
chiefly the last of a series of events. It remembers the last
migration, as it collected the last of the lyric poets, the last
tragedies, the last form of the Epos. And modern research shows
us that there were many successive waves of migration from the
north and north-west.

If we go back to the Stone Age, it seems likely that there was

* Thuc. i. 2 (fialveTcii Tj vvv 'EWus KaXovjiivrj ov miXnt /3f/3aiQ)S oiKOVfiept], dWa
fxeravaiTTdaeLs re ovaai ra Trporepa Ka\ paSicof eKacrroi ti)v eavrav anoXeinovTes,
/3info^cyoi vTTo Tivcov del nXetovaiv — a wonderful description.

'^ Aoipuv = 'hand', as in Hcsiod's deKubapos, Homer's eKKnibeKddcopos. The
Lambda (X) which served as the sign on the Spartan shields is not likely to
have been originally a letter of the alphabet ; perhaps it was a picture of a
hand in profile pointing downwards with the thumb sticking out. .Some of
the pictograms for ' hand' are like that. I suspect that the Dorians were the
' Tribe of the Hand ', and that dcopo}', ' gift ', is a thing ' handed ' or a buona
matto, and Ubu>ni the physical act of ' handing ' or ' moving the hand ', rather
than the moral act of 'granting', a use which survives in many poetical
phrases : e. g. Eur, Her. 1402 blbov 8epr] crr)v xf'P«- — Boisacq connects
Awpievs with dopv, comparing the names Awplfiaxos, Acopicpdui^s and a(r;^e'Swpos
*a wild boar' (spear-resister).


in Greece a very primitive' Hittoid' or non-Aryan population,
profoundly Influenced at certain centres by the advanced material
civilization of Crete. on these conditions, probably well before
2000 B. c, came an immigration of the peoples who afterwards
became the main Greek stock and who spoke a language that is
the parent of Greek, They were an Iranian people, akin to the
ancient Persians : Greek and Old Persian are unique in having
a Middle Voice and a dual number and in turning initial s into an
aspirate. They worshipped Zeus, the Aryan Sky-God. They
came, following their flocks in bullock carts, perhaps by way of
the Russian steppes, down through Thrace, Macedonia and Thes-
saly. They taught the inhabitants to speak Greek, and were the
ancestors of those Aeolians and lonians who were afterwards
driven before the various floods of later invasion. For later on
there came waves of peoples not entirely nameless : a great
movement of Thraco-Phrygian tribes with eastern linguistic
aflinities, saying satem instead of centum or kKarbv — who
perhaps founded the main civilization of Troy : "then the
Achaeans proper, not very numerous, consisting mainly of
chieftains and condottieri with their bands of followers from the
North-West ; and lastly, closely akin to the Achaeans, the North-
Western race of Dorians in their full mass.

The above attempt to make some statement on a subject which
is extremely uncertain may be inexact in almost every detail, but
it will serve as a hypothesis to work on, and is based on the
views of considerable authorities, chiefly Professors Burnet ^ and
J. A. K. Thomson. By the time we approach the borders ot
Homeric story, we can hazard some few more definite statements
about these North-Western or ' Danubian ' immigrants. They
were of Aryan speech : their language, as we see in the remains
of the Doric or North-Western dialects, was gradually assimi-
lated to the Greek that was being already spoken in Middle and
Southern Greece. It is commonly said, indeed, that the
Achaeans actually brought the Greek language with them and
imposed both it and the Epic form upon the natives ; but apart

^ See Burnet's articles on Pythagoras and Socrates in E. R. E. The
augment and some points in declension could also be cited as links between
Greek and Old Persian. The Achaioi perhaps spoke a ' Celtic ' language. See
especially Archaeology in Greece 1919-1921 by A. J. B. Wace in J.H.S, xli.
This first mixture of Hittoid //?^j Proto-Hellenic peoples makes a ' Helladic'
Age, on which supervenes a Bronze Age invasion from Crete and the south.


from other objections to this theory, the Achaeans had nothing
like enough time for such an achievement. Both the language
and to some extent the form of Epos must have been in Greece
already. The invaders seem to have been, to a preponderant
extent, tall and fair, warlike, uncivilized. Authorities differ about
the shape of their heads. They too worshipped the typical
Aryan patriarchal Sky -God. ^ They used, in the later streams
of invasion at any rate, iron weapons, including the terrible leaf-
shaped iron sword which was to cut its triumphant way through
all Europe ; they carried round metal shields, and fastened their
cloaks with ' fibulae ' or safety pins. The description of the
Thracians given by Herodotus in his fifth book would probably
have been true some six centuries earlier of all these invading
Northerners. Professor Ridgeway, who has helped so greatly
our understanding of the two elements in early Greek life, has
unfortunately over-simplified his statement of the case by
speaking as if there were one homogeneous invading race,
and one homogeneous race of aboriginals. He operates with
'Achaeans' from the north, and aboriginal ' Pelasgians '. The
terminology is convenient, but perhaps dangerously convenient
since neither part of the antithesis is really simple.

First the Pelasgians. In antiquity the name Pelasgian stood
broadly for pre- Hellenic. There were two main views, not
always consciously distinguished. To Herodotus the Pelasgi
were the aboriginal inhabitants of Greece, pre-Hellenic and
therefore barbaric, though in course of time they became for the
most part Hellenized. According to him the Lacedaemonians

Recent explorations show that Mycenae was unimportant in ' Helladic ' times ;
then came Minoan influences and greatness; then Helladic influence re-
emerged. The Lion Gate, Cyclopean Walls, and ' Treasury of Atreus ' are,
from this point of view, late. Cf. Wace and Thompson, Prehistoric 'J'kessaly,
on movements from the south ; also Ridgeway, Early Age, p. 645 ff. (which
however fails to account for the place-names).

^ See Mr. A. B. Cook's great book on Ze^is, Cambridge 1914. The name
Zeus at any rate is Aryan, not Hittite. However, the evidence is pretty
clear that there was a patriarchal Sky-god and Thunder-god in Greece
before the Achaeans came. Of course every Greek god is an immense
complex ; it is impossible to call one Achaean and another Pelasgian. Zeus
in classical times has usually dark hair, whatever that is worth as evidence
(A 128); he is called 'Pelasgian' (n 233; cf. Strabo, p. 329); is identified
with the Cretan Bull-god and Kouros-god, and has many strange non-Achaean


were of Hellenic race and immigrant, the Athenians were
Pelasgians, who had never left their home : though he introduces
this division by the statement that these two cities were ' the
most powerful of the Hellenes '. Thucydides, on the other hand,
seems to think of Pelasgians as part of the same general stock
over which Hellen and his sons gradually acquired such power
that they all adopted the name Hellenes.^ If we attempt to
reach the historical fact that lies beneath this confused language,
Professor Myres would suggest that the Pelasgians were a real
set of tribes, with northern rather than Aegean affinities, whom
we find first in places like Dodona, the Hellespont, and Pelas-
giotis, then, as they move under pressure from above, in various
parts of Greece ; in Crete, in Argos, in Attica, especially and
permanently in the islands of Lemnos and Imbros, where two
inscriptions in a non-Greek language have been discovered, and,
under the influence of Herodotus, accepted as Pelasgian. As
a matter of fact it seems clear that one at least of these inscrip-
tions is Etruscan ; it is even included in the Corpus Inscriptio-
muii EtrtiscartLiJi. Shall we then conclude with some of the
ancients that Pelasgians and Etruscans are the same race ?
Sophocles speaks of ' Tyrrhenian Pelasgi ', and Herodotus
remarks that the people of Plakie and Skylake in the Hellespont
spoke the same language as those of Cortona in Etruria.^ This
seems to be a mere confusion. The strange maritime people
who were called by themselves Rasna, by the Egyptian
monuments Turscha, by the Greeks Tyrseni, and by the Romans
Tusci or Etrusci, seem certainly to have reached Western Italy
from the East of the Aegean, and may well have left several
settlements on the way. The name Tyrseni is believed to tje
simply ' tower men ' (from ' tyrsis ') and was doubtless applied to
any band of sea-rovers who built themselves fortified posts in
Greek waters. And any strange language occurring in that area
was likely without much inquiry to be called ' Pelasgian '.

^ Hdt. i. 56 : Thuc. i. 3. Compare Strabo, v. p. 220 d/3;^atoV ti (^v\ov Kara
rrjv 'EXXfiSa Tracrnv enirroXnanu. He regards them as ^cipjiapoi {crXf^ov t] avfi7r<i(Ta
'EWiis KaToiKin ^npjSdpaiv inrrjp^f to naXmuv vii. p. 32 1), but he seems there to
be thinking of degrees of civilization, not of racial affinities. I doubt if he
seriously disagreed with the statement of Dionys. Halic. Ajitiqq. Ronwi.
i. IT r)v Kni TO tS)V neXaaywv yevos 'EWtjvikov, (k Hi\oTroPVTjaov to apx^iov.

^ Hdt. i. 57, reading KpoTcova for KprjaTUfa.


It may well be that the Pelasgi were a particular tribe specially
prominent in Greece before the Hellenic period, and it may be
that historians can discover some of its habitations and wanderings.
But the great dispersion of the name and the apparent lack of
connexion between the different peoples called by it lend great
persuasiveness to a suggestion of Dr. Leaf's. He compares the
name ' Welsh ', which means ' march -men ' and which was applied
by the Teutonic tribes to all their neighbours to the west and
south. Gauls, Britons, Italians, Dacians, were all ' Welsh '.
The ' Walloons ' of Belgium and the ' Vlachs ' of the Balkan
peninsula are the same. They are the strange peoples just
beyond the Teutonic border. If Pelasgoi is connected with
77-eAay, ' near ', the word would mean ' neighbour ' and would
denote the nearest strange people to the invading Greeks, as
their shifting border moved down from the north-west.^

Thus it would seem on analysis that either ' Pelasgoi ' is a name
like ' Welsh ', meaning the next people beyond the border, or
else it originally denoted a particular tribe, which for some reason
ended by giving its name as a general term to denote the whole
population of pre-Achaean Greece.

This is a perfectly normal phenomenon in the history of race

names. The tribe whose name spreads need not even be the

most important : it need only be the most conspicuous as seen

from over the border. Both Wessex and Mercia were bigger

than the kingdom of the Angles, and England was unified under

the headship of Wessex ; but Europe generally called the whole

country after the Angles, because the Angles were nearest to

Europe. All Europeans to the Saracens used to be ' Franks ' ;

all Greeks to the Asiatics were ' sons of Yawan ' ; just as in Italy

they were ' Graeci ' from the name of a certain Epirot tribe which

was much in touch with South Italy ; in Greece itself they were

' Hellenes ' from the name of a dominant tribe in South Thessaly.

It is safe to use Pelasgian in the two senses if we carefully avoid

confusing them.

1 See Leaf, Troy, chapter vii, esp. pp. 332 ff. Also Skutsch in Pauly-
Wissowa on Etrusker. Soph. fr. 270 (Pearson). — on the Pelasgians as a
historical tribe, see Myres in /. H. S. xxvii, who traces the ancient ' Pelasgian
theory' to Ephorus. Pelasgians are mentioned at Dodona, n 233 (apparent-
ly), Hes. fr. 225 (K), and Hdt. : Pelasgiotis, B 681 fif. (apparently), and later
writers: Hellespont, see Myres on B 84off. : Hdt. i. 57, ii. 51. Lemnos in
Homer (A 594, Q 294) is occupied by Hephaistos' people, the Sinties.


The little that we can make out about the race affinities of the
real aborigines is based chiefly on the names of the places which
they inhabited. All over Greece we find the towns, mountains,
rivers, and, curiously enough, the flowers, called by non-Greek
names. Names like Larisa, Corinthos, Zakynthos, Hyakinthos,
Olympos, Arisbe, Narkissos, Parnassos, Halicarnassos, are no
more Greek than Connecticut and Poughkeepsie or Alabama are
English, or Morbihan and Landes are French. And an examina-
tion of these non -Greek place-names, as carried out with great
ability by Kretschmer and Fick, leads to a result which is on
general grounds satisfactory. There is a great system of place-
names in a language still unknown to us, which reaches across
the mainland of Greece, the islands of the Aegean, and practically
the whole immense peninsula of Asia Minor : a language which
is clearly not Semitic, and in the opinion of most scholars not
Aryan either, and which must therefore have belonged to that
pre-Semitic population of Asia Minor of which the most
distinguished group is the Hittite.^ Anthropologists and

^ Especially Fick, Vorgriechische Orisnamen (1905) and Hattiden und
Daniibier in Griechenland (1908), illuminating books: also Kretschmer,
Geschichte der Griechischen Sprache (1896). Conway, however, argues that
this language— quite distinct from Etruscan-Pelasgian— was Indo-European,
though of course not Greek. {B. S. A., viii. pp. 125 fif., x. pp. 115 ff.) He
starts from the three short inscriptions found at Praesus, a town said to be
* Eteocretan ', in the east of the island. They are comparatively late, saec.
vi to iv, in Greek letters, but in an unknown language which bears affinities
to Venetic and Osco-Umbrian. Conway takes this language as = Eteocretan
and Eteocretan as = Minoan. For an historical criticism of this view see
Burrows, Crete, pp. 151 ff.

It is rash to decide till we know more of the Hitlite language, which may
now soon occur. H. Winckler's excavations during 1906 and after at
Boghaz-Koi in Cappadocia have resulted in: (i) a proof that Boghaz-Koi
was the capital of the Hittite kingdom ; (2) the discovery of the state archives,
consisting of many large complete tablets and over 2,000 fragmentary ones —
correspondence from Hittite vassals and from Egypt. The earliest are of the
same date as the Tel-el-Amarna letters, and contain notes for the Assyro-
Babylonian version of the treaty between Rameses H and the Hittite king,
Chetaser. The Hittite language is written both in hieroglyphics and in
cuneiform. The hieroglyphs are different from those of Egypt and have not
proved easy to decipher, though they were first noticed in 1736 and a corpus
of them was published in Wright's Empire of the Hittites in 1886. (See
Campbell-Thompson, A New Decipherment of the Hittite Hieroglyphics,
Oxford, 1913.) The cuneiform documents consist chiefly of some 20,000
tablets found by Winckler at Boghaz-Koi written in various languages, includ-
ing Semitic Babylonian, Sumerian, and Hittite proper. See Die Sprache
der Hittiter by Friedrich Hrozny, Leipzig, 1917. (Hrozny carried on
Winckler's work after the latter's lamented death.) Also Caractere Indo-
Europien de la Langice Hittite, by Carl Marstrander, Christiania, 1918, with


measurers of skulls tell us that there were in the Aegean lands
before any Northerners arrived on the scene two distinct races —
a dark long-headed Aegean race with littoral habits, never
going far from the sea ; and another dark short-skulled
Armenoid race, inhabiting the highlands on both sides. How
far these races were conscious of their respective unities, how
far the ruling Minoans were racially distinct from the surround-
ing peoples, are questions which we need not at present face.
The Aegean world was certainly divided into many little tribes
and communities, which no doubt fought and hated one another
as gladly as so many Celtic clans. But the remains show that,
generally speaking, they were homogeneous in culture though
by no means all at the same level. And we shall, with this
apology, speak of them in future under one name as pre-Hellenic
or Aegean.^

And opposed to these aboriginal or quasi-aboriginal races
stand the invaders from the north, Professor Ridgeway's
' Achaeans '. The case is exactly similar. The Achaeans
formed one of the many immigrant tribes ; but the name spread
beyond the bounds of the tribe and was used by the Aegean
peoples to denote the northern races in general. In Homer it
seems to include all the warriors, of whatever blood, who have
fallen under the lead of the northern chieftains. But we should
not forget that there were many branches of the invasion. From

review by Campbell-Thompson in the Times Literary Sttpplement, October 23,
1919. It appears that Hittite has (i) noun case endings, (2) possessive
pronouns, and (3) some verbal inflections of markedly Indo-European
character and indeed very like Greek. (The suffixed pronouns were : Sing.
1. -mi, -mu, -m\ 3. -/; 3. -s. PI. i, -na; 2, -7ii; 3. -u. The verbs showed
an augmented past tense, e.g. a-da-i, 'he gave'; a-k-t, 'he came'. The
nouns have accus. sing, in -«, and, as Prof. Sayce pointed out, nom. in -s.
Prepositions are ab{a), ta, mit, -kan. Times, Nov. 22, 1912, reporting a
lecture by Campbell-Thompson.) The vocabulary, however, is difficult to
equate with Indo-European, and Dr. Cowley and Mr. CamplDell-Thompson
consider the affinity of Hittite not yet proven. See Hogarth's article
HITTITES in Ejicycl. Brit, xi; also the note in O. Weber, Die Literatur der
Babylonier und Assyrier (Erganzungsband ii of Der Alte Orient), p. 275,
and Garstang's Land of the Hittites (1910); and especially A. E. Cowley,
The Hittites, 1920.

^ The question of Semitic and Egyptian influence or settlement among
these aborigines can be left aside. There was much interplay between Egypt
and Minoan Crete, but no whole nation came in from the south or east as
there did from the north.


the forests of Central Europe, guided by the valleys and moun-
tain passes towards Dodona and towards Thessaly, came divers
Achaeans and Hellenes ; more to the east came tribes of the
same blood, afterwards called Macedonian and Thracian.^ one
of these Thracian tribes, the Bhryges, crossed into Asia, like the
Cimmerii and the Gauls after them, and drove a wedge of
northern population into the midst of the native ' Hittoids ', If
any one is inclined to over-simplify his conception of these racial
movements, he might find a useful warning in a study of Phrygia,
or of one part of Phrygia, the Troad. If we take the various in-
vaders of the Troad in early Greek times, we find first the
' Phryges ' or ' Bryges ' : their name seems to have kept the old
Indo-Germanic bh which the Greeks could not pronounce. Also
the Troes or Trojans ; also a branch of the Paiones, who gave
their name to a part of Northern Macedonia; further, some
northern neighbours of the Paiones, the Dardanoi, led by a royal
tribe called Aeneadae ; some of their southern neighbours, the
Mygdones ; a tribe which disappeared early, called Phorkyntes
or Berekyntes ; some Thracians, not further specified, from the
Chersonese ; and lastly the Trares. Those are the northern
invaders only. The races already settled in the land seem to
have included a main body of Leleges, a race generally known
as aboriginal further south, in Caria; some Pelasgi, who had
probably come from Thrace ; Gerglthes and Teucri, the latter
being perhaps a royal tribe ; and, if we are to believe the Iliad^
important settlements of Lycians and Kilikes. And how many
other tribes may there have been, whose names are not preserved
to us ? That is the sort of complex of races which existed in one
small piece of territory.

And meantime, further to the west of Greece, came the pres-
sure of other and more barbarous peoples, called by the general
name of lUyrians, who eventually occupied the regions of Albania
and Epirus, and resisted Greek civilization till long after classical

But, to get rid of these names and come closer to reality, what

^ O. Hoffmann, Die Makedonen, ihre Sprache und ihr Volksthum (1906),
confirms Kretschmer's results. The language is a dialect of Greek, akin to
Thessalian, but influenced by 'non-Greek' Phrygo-Thracian and Illyrian.
The chief mark is, of course, Mac. ^ y 6 for Greek (j) x ^' The eastern wing
of the Migrations seems to have been the earlier.


are we to conceive these invasions to have been like ? Very
different, I think, in different circumstances. It is almost a rule
in history, that before any definite invasion of a new territory
there is a long period of peaceful penetration. The whole pro-
cess of the northern migrations must cover a period of many
centuries. In the beginnings it is not an army that comes to
invade. It is some adventurers or traders who come and settle :
some mercenaries who are invited in. Or again, it is a few
families who move a little further up a mountain, or a little on
the other side of a pass, breaking up new land where it happens
to be unoccupied. For a great part of the process, on the main-
land at least, these may have been the normal modes of advance :
on the one hand, a gradual increase of northern soldiers and
northern oflScers in the armies of the Aegean powers ; on the
other, a slight change in the possession of farms and pasture
grounds, in which the stronger race steadily got more and the
weaker less. But violence certainly came in, and in the later
stages the very extreme of violence. While there was room for
both races there was perhaps little or no fighting. But a time
always came when there was no room. Of that later.

One thing seems clear. While the great masses of the various
northern peoples were steadily pushing downwards on the main-
land, small bodies of chiefs or adventurers seem to have gone
forth into the Aegean region to carve out for themselves little
empires or lives of romance. They were ' invited in ', as Thucy-
dides puts it (i. 3), as allies or mercenaries or condotiieri in the
various cities. And, like other condotiieri^ they had a way of
marrying native princesses and occupying vacant thrones.^ It is
just what the Normans did in their time. About the year 1035
Robert Guiscard set out from Normandy, so Gibbon tells us, as
a pilgrim, with only one companion. He went south, and ended
by becoming King of Calabria. ' Under his command the
peasants of Calabria assumed the character and the name of
Normans.' Just so Agamemnon's followers assumed the charac-

^ As we shall see later, there is ground for suspecting that descent in these
communities went by the female side, so that to marry the queen or princess
was the normal way of becoming a king. So Xuthus = Creusa, Oedipus =
Jocasta, Pelops = Hippodameia, Menelaus and Agamemnon = the daughters
of the native king Tyndareus, &c. Cf. the numerous instances in Frazer,
Kingship, chap. viii. (Also Leaf, Homer and History, cap. ii.)


ter and the name of Achaeans.^ In the eleventh and twelfth
centuries A. D. you could find little bands of the Northmen
established at various points of the Mediterranean, as kings and
nobles among an inferior population. ' The gradual association,
incorporation, or alliance of the Scandinavians with the nations
they came to plunder or destroy is perhaps the most decisive
fact in the story of the Christian Middle Ages, and affords a
basis or starting-point for every subsequent development.' So
writes Professor Beazley of the mediaeval Normans.^ And just
the same might be said of these other invading Northmen in
Greece in the thirteenth and fourteenth centuries before the
Christian era.

The great citadel of Troy had a northern king, a Phrygian,
Similarly in all the other centres of Aegean power we seem to
find Northmen ruling. Minos indeed was aboriginal, and even
divine : but the tradition makes him first into a ' friend ', then
into a son, of the Achaean Zeus ^ ; and Idomeneus, the Cretan
chief of the Iliad^ is clearly counted among the Achaeans. The
ships of the ' peoples of the sea ' are under the Achaean Aga-
memnon. He is the very type and king of the Achaeans : but
it is interesting to notice that his family tree is derived from
Phrygia.* If this is right, Agamemnon belonged to those same
Northmen who had come eastward by way of Thrace to occupy

^ Gibbon, cap. Ivi. There is a good account of these sons of Tancred in
Demolins, Comment la Route cree le Type Social, ii. pp. 313 ff. Just so with
the Dorians: Halicarnassus was founded by ' lonians from Trozen' with
Dorian leaders. It counts as Dorian. Hdt. vii. 99 : Strabo, p. 653, &c.
So, too, Tarentum : Tdpai'Ta Se dnmKicrav (J.ev Aa/ceSat/xo^/ioi (Perioikoi and
Parthenioi)' otKia-rrjs 6e fyevfro STraprtarny ^dXavdos, Paus. x. lo. 3. The
dialect is Achaean = Perioikian, but the colony is called ' Dorian *. So
the 'Spartan' army at Thermopylae, 300 Spartans in 5,000 odd, besides
Helots. Meister, Dover imd Achdej\ p. 22 ff.

^ Dawn of Modern Geography, pt. ii, chap. i. ' See above, p. 30.

* Pelops is nearly always a Phrygian (Soph. Ajax, 1292; Hdt. vii. 8 and
II ; BacchyHdes, vii. 53, &c.). Pindar says a Lydian \0L i. 24, ix. 9). After-
wards the ideas are confused, and he is merely Asiatic. (So Thuc. i. 9.)
Observe that his alleged ancestor, Tantalus, was not originally a son of Zeus,
but'°an 6api(TTrjs like Minos ; i. e. not an Achaean, but a native prince, and
Agamemnon's descent from him a fiction (Eur. Or. 11 ; Pind. 01. i. 43, &c.).
Tantalus also appears as the first husband of Clytemnestra, slain by
Agamemnon (Eur. /p/t. Aul. 1150). His being non-Achaean may perhaps
explain why Zeus sends him to Tartaros with Sisyphus, Salmoneus, Tityos,
Ixion. (See J. E. Harrison, Prolegomena, pp. 336 ff.) Pelops' ally and
victim, the charioteer Myrtilus, bears the name of the Hittite king, Mursil,
who was also a charioteer. Strange !


Troy : and when he led an army against Priam he fought in a
specially close sense against his own kindred.

The later Greek imagination liked to think of Troy as an
Asiatic city, and to make the Trojan War a type of the age-long
struggle of West and East, Aryan and Semite. There are abun-
dant symptoms of this tendency in the Iliad (e.g. $ 88, X 48).
But it seems likely that in the earliest records the Trojan chiefs
were of the same race as the Achaeans. There is no difference
of language. The difference of language comes in between the
Trojans and their own allies, the ' Carians with barbarous tongue '
and the various peoples in whom ' there was not one language
nor one voice '} Their mode of fighting is exactly the same as
that of the Achaeans. Their gods are the same. Nay, if we
examine carefully into that question the result is rather curious.
According to Homer, a typical Achaean oath is by the trinity,
Zeus, Apollo, and Athena.^ And this trinity in the Homeric
poems must have been originally on the side of Troy ! Apollo
fights openly for the Trojans. Zeus is constantly protecting
them, putting off their evil day, and rebuking their enemies.
Athena indeed appears in our present Iliad as the enemy of
Troy. Yet it is to her that the Trojans especially pray. She
is the patroness of their city, she the regular Achaean ' City-
holder ' : and it is when the Palladion, or image of the protect-
ing Athena, is stolen away, that Troy eventually can be taken.
In Euripides' Trojan Woinen^ one may add, the treachery of
Athena in turning against her own city is one of the main notes
of the drama.

» B 867, A 437.

* The trinity does not occur outside Homer. ('Originally a trinity of
Father-Consort-Son, as usual in Aegean religion? Cf. births of Hephaistos,
Erechtheus &c.' JAKT.) on Zeus see note on p. 41. Athena is on one side
merely the Athenaia kore (see p. 97), on another, as Pallas, she is the
palladion, or divine thunder-shield which falls from heaven ; as such she is
a 'daughter', almost a mere attribute, of the Thunderer. Apollo has some
aboriginal characteristics, e. g. he is a stranger to the other Olympians, who
fly before him, in the Homeric hymn ; but as Sun-god (I have lived to see
this old view, which is based on firm ancient authority, re-emerge from the
depths of unfashionableness) he is closely associated with the Sky-god, Zeus.
See J. E. Harrison, Prolegomena, p. 461 f. ; Wilamowitz, 'Apollo' (Oxford,
1908) ; and in Hermes, 1903, p. 575. Also Th. Reinach, Itanos et Cinventio
scuti, Rev. de Vhistoire des religions, 1909, p. 331. The parallel between the
patriarchal Zeus in Greece and Othin in Scandinavia is very striking : invading
gods accepted as supreme by the native populations and imperfectly assimi-
lated to the old system of gods. See Chadwick, Odt of Oilmi.

2760 E


One great city, as we saw above, did not accept Achaean
rulers. In Thebes the Cadmeans, whoever they may have been,
held out to the end. The war of The Seven has a different look
from the ordinary wars of one Achaean band against another.
The Minyai in Orchomenos were destroyed more easily. Thebes
seems to have remained Hke an island in the flood of Achaean
invaders. She had them to the north of her in Thessaly and
Phthia, to the west in Phocis and Aetolia, to the east (probably)
in Euboea, to the south-west in Argos. And, if we are to
believe tradition, it was from this farthest southern point that
they turned, determined to tolerate no more the great fortress of
the alien race.

But in the main, if we try to conceive the Aegean in, say, the
thirteenth century B. c, we must think of the ancient seats of
power as generally standing, but at each palace a northern chief
established as king with a band of northern followers about him.
Their power was based pardy on sheer plunder, partly on the
taxes yielded by a constantly decreasing trade. It was an un-
stable condition. Some northern Agamemnon might sit at
Mycenae, a northern Idomeneus at Cnossos. They might have
imbibed a fair amount of civihzation. They were perhaps good
rulers. No one could doubt their valour. But too many of their
own kinsmen were prowling the adjacent world. It was only by
memory that they knew the

Riches that Ilion held, the walled and beautiful city.

Of old in the passing of peace, ere came the sons of Achaia.

Fewer and fewer caravans of laden mules plodded up the stone
ways of the Argive mountains. Fewer and fewer fleets of trad-
ing boats came to pay toll in the harbours of Southern Crete.

In this state of weak equilibrium there came further shocks
from the north-west. Other tribes pressed down on the main-
land, through Thessaly down to AetoHa, over from Aetolia to
Boeotia, to the north of the Peloponnese, to Elis : by sea came
the most dangerous of all enemies, hordes of dispossessed men,
who must plunder and slay, or else die. It was possibly with
some view of saving his dynasty and consolidating the various
bodies of chiefs who would otherwise be troubling him that


the Agamemnon of the time gathered his expedition of 'all
Achaeans ' against Troy, and won — if he did win it — his more
than Pyrrhic victory. Troy indeed fell, but all Achaean Greece
fell with it. A storm, says the tradition, scattered the returning
kings over the face of the deep. Some came home to die, some
were lost, some settled in strange lands. But for certain their
glory was gone, their palaces shaken, and the names of their
sons are blotted out from the page of history. Those old sea-
rovers had among them a special title of honour, TrToXi-rropOos,
' Sacker of Cities '. It represented the height of romantic glory
as well as the most profitable of financial co7^ps. And the same
idea, seen from the other side, haunted also the imagination of
the civilized city-dwellers. A beleaguered city, with the men
fighting in front and the women and old men watching from the
wall, forms a regular subject of epic description. It is depicted,
for example, on the shields of Achilles in Homer and of Heracles
in Hesiod. The same theme appears on a silver vessel from
Mycenae and on reliefs and mosaics from Minoan Cnossus.^
The theme took epic shape afterwards in the tales of Thebes
and Troy. It was in these ages the crowning triumph or the
ultimate horror of human life.

At first when the rovers sacked a city, they could in a way
rebuild it or have it rebuilt. They assimilated enough of
Aegean civilization at least to live in the castles of those whom
they conquered. But the same thing occurred here as in Rome
afterwards. As the ruder hordes and the vaster numbers pressed
down ; as the pre- Achaean races had sunk in numbers and in
confidence ; there came at last tribes who could destroy but
not build nor even keep, ' sackers-of-cities ' who burned and
shattered, and then could make no more of their conquest than
to live huddled in war-parties among the ruins.

One must probably conceive two different processes of migra-
tion, by land and by sea respectively. By land, a whole tribe
or nation tended to push on, carrying with it its women, its
normal possessions, its flocks and herds. Though even on land
there were many varieties in the intensity of the struggle. In
Boeotia, for instance, the conquering race, pushing over from the

^ Cf. I//ad 2 509, Asj^is 239 ff. Evans, O'etan Palaces, i. p. 314. The
mosaic is Middle Minoan II.

E 2


west, seems to have settled without much massacre and without
any formal enslavement of the resident population. one result
of this comparative clemency was a subsequent harshness. The
oligarchies in Boeotia continued through several centuries
peculiarly severe and illiberal. The subject race had been
admitted to something so nearly approaching equality, that it
needed — in the judgement of its masters — continual thrashing.
In most of Thessaly, in Argos, Corinth, Sparta, the natives were
reduced to varying degrees of slavery. They became, like the
Gibeonites, hewers of wood and drawers of water: like the
Messenians,^ they ' walked as asses walk, weighed down with
heavy burdens '. In Attica the invaders seem to have been few
and weak. They merely merged with the old population. one
cannot even discern a definite ruling class. It is a fact worth
noting by those who study questions of race, that among both
the Greeks and the Hebrews the most prominent and charac-
teristic part of the nation was also the part most largely mixed
with the race of the despised aborigines. The tribe of Judah
had the largest Canaanite element.^ As for the Athenians, they
always claim to be children of the soil, and Herodotus actually
goes so far as to describe them as ' not Greek but Pelasgian '.

But what of the migrations by sea ? The centre of Greece is

really not Athens nor Sparta nor any state of the mainland. The

real centre is the Aegean ; and the migrations by sea are both

more characteristic and for after-history, I venture to suggest,

more important. When a tribe moved by land it took most of its

belongings with it. When it had to cross the sea a possession

must needs be very precious indeed before it could be allowed

room in those small boats. Of course there are cases where a

deliberate invasion is planned, as the Saxons, for instance,

planned their invasion of Britain. The fighting men go first and

secure a foothold ; the rest of the nation can follow when things

are safe. In historical times, when the Athenians left Attica

before the advance of the Persian army, they took their wives and

even their herds across the narrow waters to Salamis and Aegina.

When the Phocaeans deserted their city and fled to the west, they

^ Tyrtaeus 6.

^ See e.g. Driver on Gen. xxxviii ; Cheyne also remarks on Edoniite and
North Arabian elements in Judah, £';/6-. Bid/, s. v.


seem to have begun by taking- their womankind at least as far as
Chios, where they might hope to find a breathing-place.^ But
these were more organized or at least less helpless peoples ; the
movement was well thought out beforehand, and there was friendly
land near. In the earlier migrations of the Dark Age a tribe, or
mass of people, seldom took to the sea till driven by the fear of
death . That was no time to think of taking women or herds. You
might desire greatly to take your young wife — or your old wife, for
that matter ; but you would scarcely dare to make such a proposal
to the hungry fighters about you. You might wish to take your
little boy. But would the rest of us, think you, choose to be en-
cumbered with another consumer of bread who could never help
in a fight, who might delay us in charging or flying, might cry
from the pain of hunger or fatigue and betray us all } No, leave
him on the beach, and come ! Put some mark on him. Probably
some one will make him a slave, and then, with good luck, you
may some day knock up against him and pay his ransom.

When we are off on the sea^ what is the prospect before us ?
We have some provisions, though no water. Instead, we take
guides who know where there are springs near the sea-shore in
divers islands and unfrequented promontories. We can move by
night and hide in caves during the day. The guide probably
knows places where cattle may, with some risk, be raided. Better
still, he knows of some villages that have been lately attacked by
other pirates, where the men are still weak with their wounds.
Not all their flocks have been killed. We might well take the
rest. If we stay at sea, we die of thirst. If we are seen landing,
we are for certain massacred by any human beings who find us.
Piracy on the high seas will not keep us alive. In the good old
days, when the Northmen first came, pirates could live like
fighting-cocks and be buried like princes. But the business has
been spoiled. There are too many men like ourselves, and too
few ships with anything on them to steal. If we go back to our
old home, the invaders have by this time got our women as slaves,
and will either kill us or sell us in foreign countries. Is there
anywhere an island to seize ? There are many little desert rocks
all studded over the Aegean, where doubtless we have rested

* Hdt. i. 165. Cf. the career of Dionysius of Phocaea as a pirate, vi. 17 ;
of the Samians, vi. 22 ff. And in general Thuc. I, 5.


often enough when the constrained position of sitting everlastingly
at the oars has been too much for us ; rested and starved, and some
of us gone mad with thirst under that hot sun. A waterless rock
will be no use. Can we seize some inhabited island } Alone we
are too weak ; but what if we combined with some other outlaws ?
There are some outcast Carians in like plight with ourselves in
one of the desert caves near. In our normal life we would not
touch a Carian. Their weapons are no gentleman's weapons.
Their voices make one sick. And their hair . . . ! But what does
it matter now .''... And with them are some Leleges, who worship
birds ; some unknown savages from the eastern side, dark-bearded
hook-nosed creatures answering to babyish names like ' Atta ' and
' Babba ' and ' Duda' ; and — good omen ! — some of our old enemies
from near home, the tribe that we were always fighting with and
had learned to hate in our cradles. A pleasure to meet them
again ! one can understand their speech. We swear an oath
that makes us brothers. We cut one another's arms, pour the
blood into a bowl and drink some all round. We swear by our
gods : to make things pleasanter, we swear by one another's gods,
so far as we can make out their outlandish names. And then
forth to attack our island.

After due fighting it is ours. The men who held it yesterday
are slain. Some few have got away in boats, and may some day
come back to worry us ; but not just yet, not for a good long
time. There is water to drink ; there is bread and curded milk
and onions. There is flesh of sheep or goats. There is wine,
or, at the worst, some coarser liquor of honey or grain, which
will at least intoxicate. one needs that, after such a day. . . . No
more thirst, no more hunger, no more of the cramped galley
benches, no more terror of the changes of wind and sea. The
dead men are lying all about us. We will fling them into the
sea to-morrow. The women are suitably tied up and guarded.
The old one who kept shrieking curses has been spiked with a
lance and tossed over the cliff. The wailing and sobbing of the
rest will stop in a day or two : if it torments you, you can easily
move a few paces away out of the sound. If it still rings in your
ears, drink two more cups and you will not mind it. The stars
are above us, and the protecting sea round us, we have got water
and food and roofs over our heads. And we wrought it all by


our own wisdom and courage and the manifest help of Zeus and
Apollo. What good men we are, and valiant and pious ; and
our gods — what short work they make of other men's gods !

There Is no trait in the above suggestion that is not drawn from
a real case. I have been imagining the case of a quite small island.
More often not a whole island was at stake, but only a promontory
or a foothold. Nor do we, of course, ever hear the whole compli-
cations of a conquest. It is always simplified in the tradition.

In Chios, for instance, we hear that there were first Carians, to
whom a settlement of Abantes from Euboea had joined themselves.
Then came an invasion of refugees from Crete — surely not of
pure Cretan blood — who gradually grew and mostly drove out
the Carians and Abantes. From Strabo ^ we hear, significantly
enough, of a quite different founder of Chios, a man called
Egertios, who brought with him ' a mixed multitude ' {avufieLKTov
TrXfjdos). It afterwards counted as one of the chief Ionian
cities. In Erythrae there are Cretans, Lycians, and that mixed
Graeco-barbaric race called Pamphylians, whose dialect seems to
show that they spring from some settlement of old prae- Achaean
vikings.^ Later came new immigrations from all the Ionian cities.
It was rather different at Colophon and Ephesus on the mainland.
In both cases there was an ancient pre- Hellenic oracle or temple
in the neighbourhood. In Colophon there were Greeks from
Crete, from Boeotia, from the west of the Peloponnese : if we
may believe the epic tradition, there were fragments of many
other tribes as well. They forced a settlement somehow on the
land ; living perhaps, as Wilamowitz suggests, in ' Blockhuts ' on
the shore, fighting for a permanent foothold in the barbarian city.
In Colophon they are accepted as a ruling caste, and get possession
even of the oracle. In Ephesus they are weaker ; they have a
position rather as clients of the great temple, and ' Diana of the
Ephesians ' remains at heart barbaric till she can break out into
confessed monstrosity in the Roman period. Round another
sanctuary, the little rock of Delos, there grows up a peculiar

* xiv. p. 633. The main sources for these colonization traditions, outside
the epos, are Strabo and Paus. vii.

' The half-barbarous Pampliylian dialect has enough points in common
with Cypro-Arcadiun to show this. See Thumb, Handbuch d. Gr. Dialekte,
p. 298, and Meillet in Rev. d. Etudes Grecqttes ior 1909, pp. 413 ff.


federation of people from divers parts of the Aegean, a league
whose business it is to meet at Delos for certain festivals, to pay-
proper dues to the holy place and to keep it sacred. They w^ere
called ' lawones ', lones, and the name spread gradually to a large
part of the Greek people.^

Nearly everywhere on the mainland and in the isles there are,
as we have said, old place-names in a language not Greek, but
earlier than Greek. But there are exceptions. In Cos we know
of an invasion from Crete. And there all the place-names are
Greek. What does that mean ? Is it that in this particular
island, large and fertile as it is, if the Greek invaders wanted to
ask the name of a mountain or a river, there was no single native
voice — not even a woman spared for a concubine — to answer
them, so that they had to name all the places anew ? I see no
other plausible explanation. Different was the end in Lemnos.
If tradition is to be believed — and, in the one large point where
it can be tested, the tradition is confirmed by history — there was
once done in Lemnos that act of vengeance for which one's
unregenerate instinct thirsts in thinking over the bloody and
relentless tale of these conquests. The men of Lemnos were
duly slain. The women were duly enslaved as concubines.
But they were trusted too soon : either they nursed the memory
of their wrongs longer than other women, or in some way they
had an opportunity denied to others. At any rate the native
women rose and murdered their invaders, and the island was
never completely possessed by the Greeks during all the classical
period. It was a hard task for an island in that position to keep
itself un-Hellenized. But somehow Pelasgians gathered there.
Later on, when a part of the population showed some tincture
of Greek manners and claimed descent from the Argonauts, it
was expelled. When the children born of some captured Greek
women began to show their Greek blood, they were murdered
and their mothers with them. The ' deeds of Lemnos ' ring with

* For all this paragraph see Wilamowitz's illuminating lecture Die lonische
Wanderung {Sitzungsber. Berlin. Akad. 1906, iv). As to Ephesus, the
* multimammia' form of Artemis is of course barbaric, and belongs to the
regular Anatolian mother-goddess. It is most remarkable that the recent
excavations at Ephesus have unearthed nearly fifty figurines of the goddess,
pranging from the eighth to the fourth centuries B.C.,' in none of which
is there ' any'approximation to the " multimammia " type rendered familiar
by statuettes of the Roman period'. — Hogarth, in the Times of Nov. 2, 1906.


an ominous sound in early Greek proverb, the extreme of horror,
no other deed like them.^

This is the sort of picture that we can recover of the so-called
Dark Age. It is just the stage of society which Professor
Chadwick has shown to be in so many regions of the world
the Mother of Heroic Poetry .^ It was the climax of a long
process of ' barbarian invasion ', in which a young half-savage
race gradually entered into the stored riches of an ancient
civilization, becoming successively awe-struck visitors, servants,
enemies, mercenary chieftains, adventurers, plunderers, and
masters. It was a time, as Diodorus says, of ' constant war-
paths and uprootings of peoples ' ^ ; a chaos in which an old
civilization is shattered into fragments, its laws set at naught, and
that intricate web of normal expectation which forms the very
essence of human society torn so often and so utterly by continued
disappointment that at last there ceases to be any normal
expectation at all. For the fugitive settlers on the shores that
were afterwards Ionia, and for parts too of Doris and Aeolis,
there were no tribal gods or tribal obligations left, because there
were no tribes. There were no old laws, because there was no
one to administer or even to remember them : only such com-
pulsions as the strongest power of the moment chose to enforce.
Household and family life had disappeared, and all its innumer-
able ties with it. A man was now not living with a wife of his
own race, but with a dangerous strange woman, of alien language

' Aesch. Cho. 631, Hdt. vi. 138. The story fits in with known historical
facts ; yet perhaps it is not safe to trust it. It has too much the look of a
myth built upon a religious cult of some kind. First the women of Lemnos
kill the men ; then the men kill the women (and children) ; thirdly, when
the Minyans of Lemnos are in prison in Sparta, their wives change clothes
with them and save them (Hdt. iv. 146).

^ The Heroic Age, Cambridge, 1912. Professor Chadwick discusses six
societies which have produced Heroic Poetry, and decides that in five of
them at least (p. 459) ' The Heroic Age can be traced back to a similar series
of causes. Firstly, we find a long period of "education", in which a semi-
civilized people has been profoundly affected from without by the influence ot
a civilized people. Then a time has come in which the semi-civilized people
has attained a dominant position and possessed itself, at least to some extent,
of its neighbour's property.' Such an age is a ' crowded hour of glorious life ',
when tribal and religious sanctions are broken down, and the individual tries
his fortune with his sword and his «/jeT7}, a time of 'Mars and the Muses'
See Preface, p. v.

^ YivKvai (TTpaTfiai Kai iJ.(TapaaTa(rns. Cf. of course all through this discus-
sion the ' Archaeologia ' of Thucydides i. Also see Appendix C, on the List
of Thalassocrats.


and alien gods, a woman whose husband or father he had perhaps
murdered — or, at best, whom he had bought as a slave from the
murderer. The old Aryan husbandman, as we shall see hereafter,
had lived with his herds in a sort of familiar connexion. He
slew ' his brother the ox ' only under special stress or for definite
religious reasons, and he expected his women to shriek when the
slaying was performed. But now he had left his own herds far
away. They had been devoured by enemies. And he lived on
the beasts of strangers whom he robbed or held in servitude.
He had left the graves of his fathers, the kindly ghosts of his own
blood, who took food from his hand and loved him. He was
surrounded by the graves of alien dead, strange ghosts whose
names he knew not and who were beyond his power to control,
whom with fear and aversion he tried his best to placate. one
only concrete thing existed for him to make henceforth the centre
of his allegiance, to supply the place of his old family hearth, his
gods, his tribal customs and sanctities. It was a circuit wall of
stones, a Polis ^ ; the wall which he and his fellows, men of
diverse tongues and worships united by a tremendous need, had
built up to be the one barrier between themselves and a world of
enemies. Inside the wall he could take breath. He could become
for a time a man again, instead of a terrified beast. The wall
was built, Aristotle tells us, that men might live, but its inner
cause was that men might ' live well '. It was a ship in a great sea,
says a character in Sophocles {Aiit. 191), whose straight sailing
is the first condition of all faith or friendship between man and
man. The old Kore or earth-maiden changes her type, and
appears on coins wearing a crown made of a city-wall. The
Polis had become itself the Mother-Goddess, binding together all
who lived within its circuit and superseding all more personal
worships. When this begins we have the germ of historical

This religion of the Polis was, I think, in the later ages of
Greece, the best, and is to us the most helpful, of ancient religions.
It has this in common with the others, that it implies in each

. ^ This is the use in Homer, preserved later in the words TroXt'^o), TroXtcr/ia.
Of course in safer times the cities spread and far outgrew the old Polis, which
was then apt to be called Acropolis — as at Athens. And some warlike tribes
went on living without a wall, /cura /ccu/xas, like the Spartans and the Northern
invaders of Italy.


citizen the willing sacrifice of himself to something greater than
himself. It has also to the full their passionate narrowness.
But it differs from all the others in many things. It has its roots
in knowledge and real human need, not in ignorance and terror.
Its rules of conduct are based not on obedience to imaginary
beings, but on serving mankind ; not on observance of taboos,
but on doing good.

'Apera noXv/xoxOe yevei ^poreio), says Aristotle in the first line
of his one curious outbreak into lyrics, ' Arete much laboured
for by the race of man.' It is one of the common burdens of
early Greek poetry, of Pindar, Hesiod, Phokylides, Simonides,
this thirst of men for Arete^ the word that we translate ' Virtue '.
It is more, of course, than our Virtue ; more even than the
Roman Virtus. It is 'goodness' in all the senses in which
objects can be called good, the quality of a good sword, a good
horse, a good servant, or a good ruler. The religion of the
Polls did essentially make men strive to be more of worth, to be
' good men '. Think for a moment of the judgements passed upon
his characters by the Deuteronomic compiler of the Book of Kings.
A sweeping judgement is passed for good or evil on almost
every king ; and on what is it based 'i First, on the question
whether the king followed exactly the precepts and taboos
ascribed to the deity worshipped by the writer ; and secondly,
whether he duly prevented even that deity being worshipped
anywhere except at the writer's own temple. Great rulers like
Jeroboam II or even like Omri, who is treated by the Assyrians
as the very founder of Israel, are passed over with scarcely more
than the mere statement that they ' did evil in the sight of
Yahweh '.

Now the Jews who wrote under the influence of Deuteronomy
represent a religion extraordinarily noble and enlightened.
Compared with the immense majority of ancient religions it
stands upon a mountain top. Yet contrast with these distorted
judgements of the Deuteronomist those passed by Plato in the
Gorgias on the great democratic statesmen of Athens. Plato
was perhaps the most theologically-minded of the great Greek
writers ; he writes in the Gorgias with great bitterness ; and I
think his judgements extremely prejudiced. Yet from beginning
to end he bases his indictments of the various statesmen on one


question only, their service to their fellow men. Have they
made Athens better and happier ? It looks as if they had ; but he
denies it. ' They have filled the city with docks and arsenals and
tributes and such trash, instead of Sophrosyne and righteousness.'
It is the difference between a soul in bondage and a free soul.
But to reach that freedom the Greeks had to pass first through
fire and then through a great darkness. That is the subject
which we will consider in detail in the next of these lectures.



I WISH in the present lecture to consider in detail some ot
those sanctions of tribal custom and religion which were exposed
to change or destruction in the anarchy of the great Migrations :
and then, in the apparent wreck of all, to study the seed of
regeneration which seems to have been left.

I do not know that we can begin better than by following
a curious by-path of the decline of tribal religion, the history of
*our brother the ox'. Not that it is specially characteristic of
Greece. It occurred over most of Europe and Asia. But it is
one of a multitude of changes that must have befallen with some
intensity and sharpness of outline in the Dark Age of Greece.

Professor Robertson Smith has shown with great skill the
position of the domestic animals in the early agricultural tribes,
both Aryan and Semitic. The tribe or kindred was the whole
moral world to its members. Things outside the tribe were
things with which no reasonable man concerned himself. So far
as they forced themselves on the tribesmen's attention, they were
bad, unclean, hostile. And the tribe consisted of what.^ Of
certain human beings, certain gods — one or more — and certain
flocks of animals. The thing that made them one was, according
to Dr. Robertson Smith's most suggestive explanation, that
sacred thing in which Life itself is, the common blood running in
the veins of all. This statement is no doubt a little too explicit.
The oneness of the tribe was a thing taken as obvious, not
a thing reasoned about. But as far as there is any conscious
analysis the blood seems to be taken as the ground of unity. It
was in the flocks as much as in the men. Nay, sometimes
rather more ; since the god himself was often in some sense an
ox, a sheep, or a camel. If we are, say, the Sons of Moab, then
our God Chemosh is the god of Moab and our cattle are the


flocks of Moab. They have shared our food and we have drunk
of their milk. The common blood runs in us all.

It would actually seem, from the evidence, that certain early
agricultural folk never used their domestic animals for ordinary
food. They would not so shed the tribal blood. They killed
wild animals, or, if chance offered, the cattle of strangers. Their
own animals were not killed except for the definite purpose of

Now, if anything went wrong with the tribe for any unknown
cause, if the harvest was bad, the cattle sick, the water scarce, the
neighbouring tribes overbearing, the cause was usually sought in
the attitude of mind of the god. ' The world was against them ' ;
in other words Chemosh was angry with his people, or had
forgotten them. His feeling for his kindred was becoming faint.
It must be renewed. And the regular and almost universal
method of renewing it was to take some of the living blood of
the tribe, take it especially while warm and living and full of its
miraculous force, and share it between the god and the people-
You went where the god lived, or you called him to come to
a particular pit or stone or heap of stones — an altar — and there,
after due solemnities, you shed the sacred blood for him to drink.
Feeding the god caused no great difficulty. It was easy to pour
the blood into the pit or upon the altar : and that rite always
remained. There was more awkwardness, and consequently
more variety of usage, about providing for the tribesmen them-
selves. For men began early to shrink from consuming raw
flesh and blood, and devised other ways of appropriating the
virtues of the miraculous liquid.

There is only one criticism to pass on this. It is that Robert-
son Smith's discovery was a little greater than he realized. For
he assumes a period in which there already exists some definite
personal god with whom to share the sacrifice, and we know now
that there was a previous period in which there was not yet
a personal god. There was the tribal blood there was also the
live animal that bore in it the life of the tribe, set apart and
consecrated, till it became full of magical vitality. The personal
god seems to have been made by abstraction and ' projection ' out
of this magical mana^ out of the ritual dances, the desires and
fears of the tribe. The bull was not holy because the god had


touched him ; the god himself only existed because the bull was
so charged with holiness and creative power.^ Now, as you
spared the ox in ordinary life because he was your brother and
fellow labourer, so you slaughtered him on a great occasion for
the same reason. Had he not been your brother, the sacrifice
would have lacked half its power. If we consult the collections
of anthropologists, we shall find many various ways in which
this feeling of brotherhood with the domestic animal is expressed.
The Todas of South India, for instance — that tribe to whom
anthropologists owe so much — sacrifice a buffalo once a year
only. When the victim falls, men, women, and children group
themselves round its head, and fondle, caress, and kiss its face,
and then give way to wailing and lamentation. In other cases
you beg the animal's forgiveness before slaying it, and explain to
it the dire necessity of the case, or the high honour you are
really conferring upon it. Or you arrange that it shall seem to
desire to die. You make an elaborate apparatus for self-decep-
tion, so that the beast may seem to ask you to let it die for the
tribe.2 You even arrange that it shall kill itself. I do not think
any clear distinction can be drawn here between the practices of
different races. The early Aryan peoples seem to have had
this conception, and therefore probably the Achaeans had it.
Whereas, on the other hand, the clearest instances surviving in
Greece in historical times seem to belong to the strata of more
primitive peoples. The word applied to this slaughter of the
domestic, the familiar and friendly, animal, is regularly 06j/ oy, the
legal word for ' murder '. And the locus classictts on the subject
is Theophrastus' ^ description of the Athenian festival called
Bouphonia, or Ox-murder, which contained an elaborate ritual
for ridding the various actors in the ceremony from the guilt of
the murder of their friend. The slayer flies for his life. Every
one concerned in the ceremony is tried for murder. Those who
drew water for the sharpening of the weapons are tried first : but
they only drew the water, they did not sharpen the axe and
knife. The sharpeners are next accused, and produce the men

^ See note on p. 275. * R. Smith, Religion of the Seiijites, p. 309.

* Cf, J. E. Harrison, Prolegomena, p. ill, note i, and authorities there
cited. There is ,1 similar (/joro? of a bear practised to-day in Saghalien. Cf.
the sacral Bear-slaying in the Kalevvala.


to whom they gave the weapons after they were sharpened.
These produce another man, who struck the victim down with
the axe : he another, who cut its throat. This last man accuses
the knife, which is solemnly pronounced guilty and thrown into
the sea. And besides all this, it has been arranged that the ox
shall have gone up to the altar of his own free will and eaten of the
sacrificial grains, thereby showing that he wished to be slain.
Further still, the dead ox is quickly stuffed, set on his feet, and
yoked to a plough as if he had never been killed at all ; it had
all been a bad dream.

Now what, in its ultimate element of human feeling, does this
mean ? When you have stripped off the hocus-pocus, the
superstitious make-believe of getting rid of pollution by a number
of dodges which can deceive no one, there remains at the back
a seed of simple human feeling that the act of slaying your old
kinsman and fellow worker is rather horrible : the feeling that
any honest man has about the killing of a pet lamb for food. It
was a thing, so Greek tradition tells us, that man In the golden
age did not do.^

The Bouphonia took place in Attica, where there was, practi-
cally speaking, no violent migration, and where a large element
of the ;^old population mingled gradually and peacefully with
a small element of the new. one finds traces of the same spirit
in the epics of the mainland. Heslod, in this respect represent-
ing a stationary society which had either recovered from the
violence of the Migrations or had preserved throughout them
much of the peaceful agricultural tradition, always speaks of the
ox as a sort of kinsman and partner. ' A house, an ox, and
a woman ' {Erga, 405) are what man needs for the facing of the
world. Heslod {Erga, 436 ff.) likes his ox to be nine years old :
his ploughman to be forty, and not stinted of his due dinner of
bread. You know one another's ways by that time, and feel
comfortable together. Clearly a nine-year-old ox Is not kept for
eating. Notice again how Heslod speaks of keeping the oxen
indoors and well fed in the cold weather {Erga, 452 ; contrast

^ R. Smith, Religion of Semites, p. 304, and Plat. Laws, 782 C 'Opcf^LKoi
Tivfs /3ioi. ' Plutarch ' in his brilliant essays nepl SapKo^ayia? takes just the
opposite view : the savage can be excused for flesh-eating, the civilized man


559) ; of the east winds {Erga^ 504) in the month of Lenaion,
' evil days, they just skin the ox, all of them ' ; of the cold dawn,
how 'it puts yokes on many oxen' [Erga, 580). During the
winter storms, too, you and your little girls can sit inside by the
fire and keep warm, but the wind blows through the ox's hide,
it cannot be kept out, and through the fell of the shaggiest goat.
But not the sheep. Their wool is too thick, and they do not
mind {Erga^ 5 1 2 fF.). Do you observe the sentiment of it all ?
How the ox is a friend, a member of the family.

The name they kept for him tells the same story. You will
remember the regular phrase in the older poetry efAnro^ay
eA«/ca? ^ov'i, the two epithets of rather dim and unrealized
meaning that are habitually applied to cattle. ElXiTroSe?,
' rolling the feet ', is an antithesis to the word applied to horses,
depcri7ro8e9, ' lifting the feet '. A horse steps high, a cow's foot
makes a more horizontal curve. And what of the other word
e'Ai/cey ? The Greeks understood it as ' curly-horned ', the
opposite of /3o(5i/ opOoKpaipdcou, ' straight-horned cattle '. There
were the two breeds in early Greece. But one should notice
this about the two adjectives : that they both belong to the class
of familiar names or nicknames applied to well-known animals —
names like ' puss ' and ' bunny '. Hesiod, our earliest farm-yard
poet, is full of such names : he has a nickname even for the ant
and the snail and the octopus, ' wise-wit ' (778) and ' house-carrier '
and ' no-bones ' respectively. — The hare is tttco^, ' trembler,' and
the goat jirjKciSy ' bleater,' the hog, rather less politely, is aiaXo?,

* grease.' ^ And this explains a little difficulty. "EXiKes means
' curly ', or ' crumpled ' ; and Dr. Leaf, in his invaluable commen-
tary to the Iliady objects that it is scarcely possible language to
speak of a ' crumpled cow ' when you mean a cow with crumpled
horns. True, if the word were still a simple adjective with no
special connotations. But it is not : it is a name, almost a pet-
name. When Hesiod's forty-year-old ploughman came down as

^ Unless indeed aiaXos merely meant (i) hog, (2) hog's grease. Sheep
seem to have no nickname. — In general cf. 530 fif., where 'the horned and
hornless wood-sleepers ' in a snowstorm go with their tails between their
legs, like a lame man bent over his stick. It is the same spirit. There is
intimacy with animals in general, even the snake in the new fragments is

• No-hair,' arpixos {BerL Klassikerfexfe, V. i, p. 36) ; a wild boar is ' spear-
stopper', dcrxeSwpof. But with the ox there is much more.

2760 P


usual rather before dawn and met his nine -year-old cow, I
suppose he addressed her as Helix ; he said, ' Good morning,

And when for some grave reason this cow or ox had to be —
what shall we say ?— ' murdered ' is the old Greek word — it was
a solemn occasion. Take a case where the feeling is already less
keen, the sacrifice at Nestor's house at Pylos in the third book of
the Odyssey (421-63). Nestor is, of course, a Homeric hero, but
he is now back at home, under the normal influences of home life.
The occasion is a special one. There has been a visible appear-
ance of Pallas Athena, and it is necessary to honour her, perhaps
to renew the tribal bond with her, in an extraordinary way.
' Let some one go to the field,' says Nestor, ' for a cow ; and the
ox-herd is to come with him. And bring also the goldsmith
Laerkes, to put gold on the horns of the cow. And everybody
wait here.' Then follows a solemn description of all the
apparatus and the details: the goldsmith's tools and work : the
purification of every person present to receive what may be called
the sacrament of the kindred blood : the suitable sacrificial
vessels placed so that it may not be spilt upon the ground —
where it might pollute the earth or even cry for vengeance : the
man appointed to strike, and the man appointed to cut the throat.
Then, as the cow is struck, ' the daughters and the daughters-in-
law and the august wife of Nestor all shouted aloud.' It was not
a mere cry of sorrow^ it was an o/oiuge, a special religious cry for
frightening away evil influences from the stream of ' our brother's '
sacred life.^ one would like to know if there was originally
something of that in the wail of the Todas.

^ You uttered an olo/ugc when any one had a fit, to frighten away the bad
kervih.\c\\ had seized him; in the case of Jason's princess {Medea, 1 170-7)
it proves to be something much worse than a fit, and the ololuge turns into
a wail of horror. For brotherly feeling toward the ox cf. Aelian,' V. H. 5. 14 ;
an old law at Athens says, ' Slay not the ox accustomed to plough or waggon,
for this animal shares the labours of man.' Also Plut. Solon, 21 : Solon
forbade sacrifice of oxen at funerals. (Probably for religious rather than
sumptuary reasons.) Cf. Hollis, The Nandi, p. 20: 'They [the Nandi
tribes] love their beasts, as they say themselves, more than anything in the
world : they talk to, pet, and coax them, and their grief is great when a
favourite sickens and dies.' I owe this reference to Mr. W. R. Halliday.—
I know of two Papuans who committed a motiveless murder from pure rage
and grief at the death of a favourite pig. There is also a ritual in Papua in
which the men kill a boar with clubs : the women wail, and then chase the
slayers into a river.


Contrast with this timid, religious, almost tender slaying of the
ox, the habitual sacrifices of the litad — and of those parts of the
Odyssey where the sacrificer is not in his own land. Compared
with Nestor's sacrifice, they seem like the massacres of a slaughter-
house, followed by the gorging of pirates. The heroes make
merry, ' Slaughtering sheep beyond number and crook-horned
swing-footed oxen.' They ' sit all day long even to the setting
sun feasting on measureless ox-flesh and sweet strong wine '.
The sacrificial terms are there, but are somehow shortened and
made brutal. The only people in the Odyssey who behave like
that are, first, the wicked suitors, who devour Odysseus' flocks ;
and secondly, Odysseus' own men when they are acting as pirates,
and slaughtering the herds of the Cicones. These exceptions
give us the clue. The heroes of the Iliad have crossed the sea,
and are no longer dealing with their own kindred. The oxen
they slaughter in droves are only strangers' oxen, not their own
familiar herds. They kill them as light-heartedly as they would
kill the strangers themselves. They think no more of the ox as
a member of their tribe. The distinction of their hecatombs lies
only in the general largeness and expensiveness of the whole

It may be objected to my method here, that the difference in
question is merely that between peace and war, and is not
specially connected with the Migrations. My whole answer to
that will come gradually. But it is at least the difference between
peace and a prolonged and disorganized state of war in which
ordinary wont and use has been forgotten. And that was just
the state produced by the Migrations. Of course Homer's
picture is in a dozen ways idealized and removed from history.
Yet, in the main, the chiefs of the Iliad^ adventurers who have
forced a landing on a foreign shore and live in huts on the beach,
year out, year in, supporting themselves by plunder and decimated
by pestilences, never quite strong enough to capture the native
city, nor weak enough to be finally driven into the sea, are exactly
in the normal position of these outcasts of the Migrations. In
their minds, as Achilles expresses it, X-ql'crTol [ilv yap re y3oey —
' cattle can be got in raids '. But let us consider the other
influences that held these men before the Migrations, and see
what became of them afterwards.

F 2


First, then, their definite gods.^ The Achaeans at one time
must have been organized in tribes, or federations of tribes, and
a tribe must naturally have a tribal god. The conception may-
seem somewhat abstract, because in the Greek pantheon as we
know it every personal god has behind him a long and tangled
history. Each has been made up out of very many elements, of
diverse origin, attached by historical processes to some one name
— or perhaps two names, since so many Greek deities have at
least two — and welded eventually into a sort of fictional unity by
the devices of the poets. Even Zeus, the northern patriarchal
Sky -god, suffered many modifications: for example, when for
purposes of theological harmony he was transformed into the
long-lost son of his conquered enemy, ' Pelasgian ' Kronos. Let
us think away these historical complications, and consider what
would probably happen to a pre-Hellenic migrant Achaean with
regard to his tribal god. The^business of that god was, of course,
to fight for and protect his tribe. His character, and his attributes,
so far as he had any, were, for the most part, simply the character
and attributes of the tribe. That is, to the tribesmen themselves
he had no noticeable character : he was just what a reasonable
god naturally would be. If they used bows, so presumably did
he: but they did not think the matter worth mentioning. If
they were characteristically bards, smiths, seafarers, spearmen,
mine-workers, naturally their god presided over all they did.
Thus to a stranger coming across the tribe the god would
produce a definite impression : he would be a smith, a ruler
of the sea, a spearman, a god of mines, a singer. That is
perhaps how, when a federation of tribes was made, there arose
departmental gods, with special attributes and almost always
special geographical homes : a Lemnian Hephaistos, an Athenian
Pallas, an Argive Hera, a Cyprian or Cytherean Aphrodite.

Now, as long as the tribe remained whole, the god of course
was with it. He had his definite dweUing-places : the Pytho
or Patara, the Bethel or Mamre, where he could be counted upon
to appear. Even when the tribe moved, he, in a slow and
reluctant way, moved with it. He was present wherever the

^ Some types of pastoral and agricultural gods and divine kings might
be treated here, but the same argument can easily be applied to them.
See pp. 135 and 205 ff. below, Lectures V and VIII, Also above, note on p. 49.


tribe was, though on great occasions it might be safer that the
chiefs should send embassies back to him, to make sacrifice at
some Dodona, some Sinai, some Carmel, where he had for certain
been present to their fathers.

But in these sea-migrations the tribe was never whole. The
chieftains can still call on their Achaean Zeus, and he hears or
rejects their call : but there is a feeling that he is not present as
he once was. He has to be called by his old names, with a
feeling of the distance that lies between : ' Zeus,' prays Achilles
at Troy, ' I^ord, thou of Dodona,^ thou Pelasgian, dwelling far
away.' The titles— whatever ' Pelasgian ' may mean— serve the
purpose of showing that you really know w^ho he is and belong
to him. Our old Thessalian Zeus seated on his throne at Dodona,
why should he listen to the crying of strange men in Asia ?
' There be very many things between, shadowy mountains and
ever-sounding sea.' But each of these words will attract his
attention. It is as if Achilles said, ' Zeus, thou who art my own
lord, who hast known my fathers at Dodona.'

Zeus did, in a way, move from mountain to mountain, just as
the Muses did. The Muses were first at home in Pieria and
Olympus, and then moved south to Helicon and Parnassus,
doubtless accompanying their worshippers. Zeus was actually
established on Mount Ida in front of Troy when Achilles prayed to
him as Dodonaean. He had come there with his Phrygians long
since. But the Zeus of Mount Ida was the god of Troy, and
surely could not accept the prayer of Troy's enemies. There
is a painful embarrassment. Zeus of Dodona is opposed to Zeus
of Ida. The tribe is divided against itself.^

Even in the Iliad^ amid all its poetical refurbishment of life,
there remain these unconscious marks of the breaking up of the
Achaeans. But it is clear from those cases which we considered of
the various Ionian colonies that the real Greek settlements of
the migration consisted of the most miscellaneous gatherings
from various tribes, together, I should imagine, with a leaven of
broken men, whose tribal belongings were forgotten. Now
among such a (jvyni^iKTov ttXtjOos — such a 'mixed multitude' as

Z(v ava, AaBcavdh , JJeXaa-yiKe, U 233. ZenodotUS, *^;yco^'a^f : evidently a

good and ancient variant : ' thou of the Oak Tree.'

^ Cf. above, Lecture II, p. 49, about Zeus, Athena and Apollo.


Strabo phrases it — the influence of the definite tribal gods would
be reduced almost to nothing. The common ' Wall ' has to
supersede them. Partly perhaps from some innate tendency of
the mind, but largely also from the force of circumstances, there
is a diametric opposition in this matter between Greeks and Jews.
The Jews seem to have found their kinsmen in Moab worshipping
a tribal god, Chemosh, according to rites practically identical
with their own. They, or at least the sacerdotal party which
prevailed among them, immediately regarded Chemosh as an
enemy and a devil, and where they observed some small differ-
ence in the ritual they magnified it and regarded it with loathing.
The ordinary Greeks would have said : ' The Moabites call Zeus
Chemosh, though some say he is Heracles rather than Zeus.'

Now, when gods are fused or renamed like this, they must
needs become less living and definite. For one thing, the taboos
or sacred practices change. In Greece itself some people who
would have died rather than eat a mouse seem to have mingled
with others who felt in the same way about lizards. Their gods
were both identified with Apollo.^ When an avoider of mice
found his friend eating mice freely near Apollo's temple and
meeting with no condign punishment, he must naturally have
been filled with religious anger. For a generation or so the
anger may have remained, latent or visible. But eventually, it
would seem, a time came when both parties ate what they liked,
and both, on the other hand, paid an easy toll to their gods
by joining in solemn sacrifices of the taboo animals on suitable
days. The religion had come into conflict with the common
conveniencies of life, and been beaten.

A tribal god, as we have seen, could move. As long as any
fair number of his tribe could keep together, he was present
among them. But other objects of worship were not movable.
Among the pre- Greek populations the most prevailing and
important worship was that of the dead. All Asia Minor is still
strewn with the graves of innumerable worthies, whom the course
of history has turned into Mahometan Walis or Christian Saints.
The old races called them ' Heroes '. They were much the same

^ Apollo Smintheus (A 39) : cf. Isaiah Ixvi. 1 7, and the original form of
Sauroctonos. on Smintheus see Lang, Otstoin and Myth^ pp. 103-20.


as the Roman Lares, ghosts of dead friends and ancestors, duly-
laid in the earth and worshipped with a few simple ceremonies
and small regular offerings of food and drink.^ Good scholars
have written of this worship as if it consisted entirely in the fear
and placation of dangerous ghosts. In later writers, like Plutarch,
there is evidence that points in this direction. But originally and
normally it is clear that this was not the spirit of ancestor-worship.
The ghost of the friend who loved you loves you still, unless you
in some way starve or injure him. The dangerous ghost is the
ghost of a strange kin. This conception certainly affected the
whole of Greece, and was one of the strongest religious bonds
regulating"private life. The gigantic tombs of the great kings of
legend, alien and steeped in crime, dominated the imagination of
the mainland right on into the classical period. Both Aegean
and Northerner were bound to their tombs by a thousand
delicate and powerful ties.

But the men of the Migrations had left their fathers' graves
behind them. The ghosts whom they ought to have fed and
cared for were waiting in the old lands helpless, with parched
lips, staring through the dark earth that lay above them.^ And
in the new lands where now they trod they were surrounded by
strange graves where lay not their own fathers, but the fathers of
the men they had wronged and slain, ghosts who hated them.
All later Greece was full of these unknown graves. They
devised many ceremonies to appease the ghosts. For one thing,
they were honestly frightened. For another, they knew that
their own dead were lying in the same condition, and they
vaguely trusted that perhaps at home also the strangers were
doing well by them. But it is a timid, uncertain honour that they
give. They may at any time be bearing some particular pollution
which specially kindles the dead man's rage. They know not

^ Babrius (second century A. D. ?) says definitely (fab. 63) that the gods
are the cause of good, the heroes of evil. Similarly, the still later Sallustius
says that god causes good, and the daemon evil. This becomes the normal
sense of daifiav in post-Christian writing. But contrast Hesiod, Erga,
123 fF., where the Heroes are blessed guardian angels, Saifjiovfs iaOXoi. The
account in Paus. vi. 9. 8 of the mad Cleomedes of Astypalaea illustrates the
sinister kind of hero. For the whole conception compare the Choephoroe
or the Oedipus Coloneus : the hero lies in the grave charged full of curses
and blessings. Cf. Harrison, Prolegomena^ p. 9, pp. 326 fif.

^ Cf., for instance, Eur. Tro. 1083 o-u /xev (ftdinevos dXaiveiS, ndauTOS awBpos,


his name, and cannot call him. He is only the Hero, one of the
sainted dead, the evcppove?, the XPI^"^^'^'^

One thing- indeed they could contrive, in rare cases, by the help
of their best areteres^ the medicine -men and makers of charms.
They could call the soul of their own dead hero from his grave
and keep it following their ships to the new settlement, there to
enter into an empty tomb which they had made for it. In this
way Phrixus, who had died in Colchis at the farther end of the
Black Sea, was brought back to Thessaly. In this way
Melanippus was brought from his ancient grave in Thebes to
Sikyon,in the hope that his presence would cause his old enemy
Adrastus to move to a new grave further away.^ Achilles seems
to have changed his grave several times, from Phthia to Skyros,
from Skyros to Troy, from Troy to the happy island of Leuce.
But there were difficulties in this process. A people flying from
a conquering foe could never carry it out. And perhaps the
practice itself was not very old. It seems to have needed the
help of a doctrine about the soul rather less concrete and material
than that of the old Aegean races. And one doubts whether,
when all was done, the ritual always carried conviction.

Very often the tomb of the dead hero had oracular powers.
His children in their perplexities could draw upon the wisdom of
their great ancestor, as the Persians in Aeschylus' tragedy seek
counsel from their dead Darius.^ Probably these oracles formed
the greatest engine of divine authority in most of the pre-Hellenic
tribes. And, as far as one can make out, an oracle never moved.
When a change of population took place, either it was forgotten,
as happened often and often ; or else it was for some reason
spared or partly annexed by the new possessors of the land.
Priests of the old race were often left in charge, and the old

^ As to these nameless or unknown ' heroes ' the clearest evidence is
Diog. Laert. i. lo, 3, ' one finds even now Kajh. roiis Brj^iovs roov'Adi^vaicov jSco^xovy
dvcovvfjiovs.' Perhaps also the frequent anonymous inscriptions — 'iKf'aios
fjpaiL avedrjice, ijpcai 6 opos, &c. For particular cases cf. Pans. iii. 13. 7
(ijpms Ttf), X. 33. 9 {bvvd(TTr]s dvrjp) : in X. 4. lo the unknown person has
become ' either Xanthippus or Phocus '. So i. 35. 7, ' he is not really Geryon,
but only Hyllus ! ' vi. 6. 7 ff. he is evil, hostile, and nameless, and is at last
driven out. Cf. also i. 43. 3 (Aisymnion at Megara), i. 34. 3, v. 15. 12
(generalizing the dead).

^ Pind. Pff/i. iv. 159, cf. Eustath. Odyssey, p. 1614 on avaKXrjo-is: Melanippus,
Hdt. v, 67.

^ on Oracles, cf. my Fotir Sfages, p. 51 f-


worshippers, when a time of safety came, could make pilgrimages
back to it. Nearly all the oracles of Greece were taken over on
terms by the incoming Northmen. The holy place eV AeX(f)oi9i
among the Delphians, which had once belonged in joint owner-
ship to an Earth- Mother and an underworld serpent, typical of
some departed hero, passed over, with or without battle, to the
Olympian prophet, Apollo. Apollo took the oracle of the
Abantes at Abac and that of the Carian clan of the Branchidae
among the barbarians in the neighbourhood of Miletus. on the
other hand, for some reason or other he left the Lebadean hero,
Trophonius, in peace, and the dead man continued to give
oracular dreams in the old cave according to the old rites. But
our present concern is with the men of the Migrations. What-
ever happened, they were cut off from their dead. To those
fugitive Abantes, for instance, who helped to settle Chios, it
mattered little whether their deserted oracle at Abae still spoke
or was silent for ever. They at any rate had no guidance
from it.

Nay : there was something worse. At times like these of the
Migrations it was best not to bury your dead, unless indeed you
could be sure of defending their graves. For you have all of you
now done, and are doing, things which must make men hate you
as your fathers and grandfathers were never hated in their
ordinary intertribal wars. You are taking from men everything
that they live by, their land, homes, wives, cattle, gods, and the
graves of their fathers. And the beaten remnant of those you
have wronged, unable to requite in due kind your many murders,
are skulking round by night, as you well know, homeless and
mad with rage, to do you any chance harm they can. They may
catch some wounded men, some w^omen, or children. They may
sometimes carry off some dead from the field of battle. At the
worst they can dig up some of your fallen comrades from their
graves. And then will be repeated the well-known orgy ot
helpless, pitiful revenge, the lust of unhappy hate trying in
a hundred ways to find its peace. For however magnificent you
may be, you conquering races, you cannot make men broken-
hearted with entire impunity.

There is hardly anything in Greek antiquity which is so


surrounded with intense feeling as this matter of the mutilation
or dishonouring of the dead. Throughout all poetry, through
the Epos, tragedy, and the historians, it rings, a hushed and
vibrating note, telling of something scarce to be spoken, a thing
which to see makes men mad. Scholars are apt to apologize for
this earnestness as a peculiarity of ancient feeling which we have
a difficulty in understanding. But I fancy that every one who
has come across the reality feels much the same as a Greek did ;
English soldiers who find their dead comrades mutilated in wars
with savages, or the combatants on both sides in the sempiternal
strife in the south-east of Europe, where Christian and Moslem
still are apt to dishonour infidel corpses.

There was one perfect way of saving your dead from all out-
rage. You could burn them into their ultimate dust.^ The
practice was the less painful to the feelings of the survivors,
inasmuch as the Northerners, who were now influential among
them, had used it in their old homes, in the forest country from
which they came. For cremation, like the other Homeric
custom of roasting meat, is a practice which demands abundance
of wood. But in Greece the other system seems generally to
have held its own. Even at Mycenae, where there were Nor-
therners in possession, the dead are buried, not burned. And
Greek language about the other life is on the whole far more
affected by the conceptions dependent on burial. The dead are
always \66vlol, 'people of the earth'; their realm is below.
The ghosts are not thought of as so much Kvia-q, or vapour of
burnt flesh. And the practice of cremation might well have been
forgotten entirely had not this special time of unrest revived it.
The grave was no longer safe. And men burned their comrades

^ Cf. I Sam. xxxi. 12, where the men of Jabesh-gilead burn the bodies of
Saul and his son, to save them from further outrage by the Philistines.
Burning seems to have been strongly against Israelite feeling ; many com-
mentators emend the text. Andrew Lang suggested to me to compare
Amos vi. 10 (obscure), Jeremiah xxxiv. 5 (Zedekiah : ' with the burnings
of thy fathers'), 2 Chronicles xvi. 14 and xxi. 19 (Asa and Jehoram). In
Scandinavia there is some evidence to show that cremation came in with
the cult of Othin. Othin's dead were burned and their souls went off to
Valhall. In the older belief they were buried or 'howe-laid', and stayed,
souls and all, in the howe, and ' exercised a beneficent influence on the for-
tunes of the family ', or defended the grave when it was broken into. This
is curiously similar to the condition in Greece. See Chadwick, Cult of Othin,
p. 58.


to save them from dogs, birds, and enemies. Sometimes we find
that, instead of burning, they buried them in peculiarly sacred
places, or in unknown and secret graves, for the same reason :

Lest angry men
Should find their bones and cast them out again
To evil.^

There was another form of worship which might have been
expected to persist, or at least quickly to recover itself.
Throughout the region that we are concerned with, from Wes-
tern Greece to the heart of Asia Minor, it seems as if every
little community in pre-Hellenic times had worshipped a certain
almost uniform type of goddess.^ An Earth-Mother or Moun-
tain-Mother in the eastern and the pre-Hellenic communities,
Mother of fruits and trees and of wild beasts, she is apt to be a
Maiden and a Bride as well, and in Greek lands is perhaps best
imagined as a Kore (Maiden). She is really the Earth as
Woman, passing at different times through all the different
normal phases of woman's life. Sometimes the intoxication of
the east is strong upon her, and, like Babylonian Ishtar, she is in
her own person all that woman can be. She is at once virgin,
mother, sister, wife and harlot, and her virginity is ever renewed.
She rears and nurses the Koiiroi^ or men-children, of the tribe.
The tribe itself is her Kotiros^ her child and in the end her
husband or lover. When Greek Sophrosyne prevails there is
an end, generally, to these dangerous confusions ; Mother-Earth
is distinguished from her Maiden daughter, and both protect the
purity of the home.

As we meet the full-blown deities of classical Greece we find
this original Earth-Kore embodied in various types. There is
of course the KorejzJ^r excellence^ daughter of Demeter, whose
more terrible name Persephone is seldom spoken. But there
are many others. The ' Athenaia Kore ' has been transmuted
into the Virgin Pallas ; her Kouros is an Odysseus or a Perseus,
whom she guides towards virtue and wisdom. The Argive

^ Eur. Med. 1 380.

^ See J. E. Harrison, Pivlegomena, pp. 257-322, ' The making of a
Goddess ' : W. M. Ramsay, Cities and Bishoprics of Phrygia, i. 87 fif. :
Hastings, Dictio7iary of the Bible, extra vol., p. 135 f. : Frazer, Attis, &-x.,
chap, iii, and Golden Bough : and A. Evans iny. H. S. xxi. pp. 170-80, and
B. S. A. ix. p. 85 f.


Kore has become Hera, the wife of Zeus : she has now no kouros,
though she watches over Jason and others, and perhaps she once
had Heracles to train in hard ordeals. Others are merged in
virgin Artemis, with her saintly votary Hippolytus ; or in the
opposite figure of the Cyprian Kore, Aphrodite, with her beloved
Adonis. At the back of such figures, in the dim distance, is an
old tribal organization, still alive in many parts of the world,
called by the general name of Initiations, or sometimes Mysteries.^
The boys on reaching maturity are taken to the Man's House,
learn their dances, pass through their ordeals or ' completions '
{r^X^Tai), and become, by initiation into the Mysteries, full men of
the tribe, fit to beget the tribe's children, to plough the tribe's
earth, and to stand in war against the tribe's enemies.

The Kouros is as prominent among Greek gods as the Kore
among goddesses. Apollo, Ares, Hermes are typical kouroi ;
in the recently discovered Cretan Hymn of the Kozivetes the
Megistos Kouros, Greatest of Kouroi, who leads the mystic dance
of initiation, is Zeus himself. And of course the same conception
has left its mark on most of the heroes. Doubtless In settled
conditions, where order and Themis held their own, the ordeals
and the teaching that prepared for them had a great influence
on character. The Australian and Red Indian evidence proves
this, without drawing on the enthusiastic testimony borne by
ancient writers to the moral influence of the Mysteries. But
when a tribe was broken, flying or fighting for its life, the
august rules of Themis must have gone by the board and the
moral training become chiefly an apprenticeship in brigandage.

Now, one set of agricultural people driven over seas and taking
refuge in the land of another would, as far as one can guess,
generally find themselves in the midst of the worship of another
Kore so close to their own that they could at once accept her.
Yet one must remember, first, that the fugitives were as a rule cut
off for some time from agriculture : and secondly, that every
Kore was certain to have secret rites and perhaps a secret name
to which the strangers would not be admitted. As a matter of
fact, there is something to be deduced from the geographical

* on these Rites de Passage see J. E. Harrison, Theinis, chap. I, 2 : my
FoKr Stages, chap, i : Webster, Primitive Secret Societies : Schurtz,
Altersklassen und Miimierbunde.


names which remained in vogue for the various Korai. If names
like Paphia, Cypris, Cytherea, Erycina, &c. persist throughout
antiquity, it clearly means that, even when a certain set of Korai
were definitely merged under the name of Aphrodite, still Our
Lady of Paphos was felt to be different from Our Lady of
Cythera or of Eryx. It is worth while remembering that even
at the present day in Spain the people of two neighbouring
villages will throw stones at one another's Madonna.^ Frazer^
and others have shown how much of the taboos and moral ideals
of primitive communities were bound up with the Corn Maiden
and her Mother. But all must have been rudely broken and
destroyed for the generations of the flight by sea.

In one respect especially this antique worship of the Kore was
bound up, if we may believe some of the ablest of modern investi-
gators, with the influences of daily domestic life. We must
distinguish two forms of the family in early Greece, which
corresponded roughly, though not exactly, with a division of
races.^ The Northerners had, as had been abundantly proved,
the regular Aryan organization of the family under the headship
of the male. The women and the children were in mami eitis^
under his hand. He had accepted the duty of defending them
against danger, and they correspondingly had to obey his will.
Relationship was counted through the male side, and the son suc-
ceeded to his father's estate. Indeed it is remarked by Isaeus that
a mother, though by nature the closest relation a man can have,
is not mentioned in the list of degrees of kinship in Attic Law.*

^ Cf. the following extract from an account of the eruption of Etna in
June 1923 :

' At Piedimonti the population brought out from the church the statue of
St. Anthony, their patron saint, and placed it in the central square of the

' At Linguaglossa on Tuesday morning the inhabitants took the pastoral
staff from the statue of St. Egidius, the patron saint of Linguaglossa, and
carried it in a supplicatory procession along the front of the lava. The popular
belief is that on a previous occasion the staff of the saint had miraculously
stayed the descent of the lava. But the population of Castiglione, when it
heard of this procession, became excited and alarmed. The miracle, they
thought, might have resulted in diverting the course of the lava from the
direction of Linguaglossa to that of Castiglione. Therefore they marched to
Linguaglossa, intending to stop the procession, A fight ensued.'

^ Cf. especially Psyche's Task.

^ Cf. the Auge (Heracles) and Aithra (Theseus) stories, and above, p. 47, note.
Isaeus xi. 17 {fJ.T]Tp6s opofia) 6 crvyytvearaTOV fiev t]P rfj <pvaet TTUvroiPf eV de
Tois ay)(i(TTeiais oixoXoyovfxevas ovk ((Ttiv.


Monogamy was the rule in Greece, though not among the

But this patriarchal system, however defective it may seem to
modern critics, and however much it was corrected by the
humanity of later Greek Law, was in its day an innovation and
a radical reform. Among the pre-Hellenic tribes the mother
was the natural centre of the family. She stayed and ruled the
household, while men fought and hunted and wandered. one
family might easily have different fathers or even uncertain
fathers ; and when the mother wanted a male protector she
turned to her brother, not to one of her mates. The children
were directed by their maternal uncle. The property descended
from mother to daughter. The sons, we have reason to believe,
generally provided them with dowries.

It is curious to read the conflicting enthusiasms felt for and
against this ' matriarchal ' or ' matrilinear ' system by feminists
and anti- feminists at the present day. Clearly one of the two
systems was suited to one type of society and one to another.
But both were families.^ Both must have possessed that power
of trying the temper and training the character In which the
family stands unrivalled. An institution In which life becomes
unbearable unless people treat each other with pretty constant
consideration, and habitually suppress their own more lunatic
claims, is clearly of enormous educational value. History seems,
up to the present, to have decided In favour of putting the
family under the more intellectual and muscular though less
sympathetic and affectionate head. And at times such as we are
considering, when life was rough and hard, the weaker part of
the human race very likely made a good bargain in exchanging
freedom for protection. But it is important to remember, when
we glibly speak of the higher conception of morals and the
purer family life of the patriarchal Aryans, that after all the
relation of mother to child is probably, even to our ideas, the
deepest, most influential, and. If I may use such a word, the
most holy of human relationships. And this relation was not

^ See especially Tylor in The Ninetee7ith Century, July, 1896 : A. B. Cook,
Classical Review, xx. 7 (' Who was the wife of Zeus?') : Farnell, in Archiv

ftir Religionsivisse7ischaft, 1904, vii (severely critical) : Frazer, Kingship,
Lecture VIII. The subject is admirably treated in Vinogradoff's Historical

Jurisprudence, vol. i. pt. i, The Elements of the Family.


only preserved by the older system, but was preserved in a
clearer and more authoritative form. The influence of the
patriarchate on religion is, of course, overpoweringly great.
Protestant and Mahometan countries are entirely dominated by
it. Yet if one tries to think for a moment of the vast volume of
prayer that is steaming to heaven at any one hour from all the
corners of the world, or, shall we say, of Christendom, I wonder
if he will find any more intense, more human, more likely to
achieve its end, than the supplication which rises from all parts
of Southern and Eastern Europe to that most ancient and many-
named Madonna, who has sat throned upon her rocks and been
a mother of many erring children from thousands of years before
the coming of Christianity. And further, if a man, who believes
somehow in the reality and ultimate worth of some religion of
gentleness or unselfishness, looks through the waste of nature to
find support for his faith, it is probably in the phenomena of
motherhood that he will find it first and most strikingly. Every
living animal preys upon every other : true : yet a mother
partridge will fight a dog to save her chickens, and a tigress die
in defence of her cubs. The religious system connected with the
matriarchal household, based on the relation of mother to child
and no other, must be counted, I think, among the great civiliz-
ing and elevating influences of mankind.

And, though this point is perhaps taking us too long, ought
we not also to consider the extreme beauty of those fragments or
elements of the Greek saga in which the young hero is befriended
and counselled by a mother or a guardian goddess ? Think of
Heracles and Athena, Odysseus and Athena, Perseus and Athena,
Jason and Hera, Achilles and Thetis. Achilles, we are duly
informed, was the son of Peleus, Peleus in himself is a great
saga-figure ; and it is a fine story how he caught and won his
sea-goddess, how she bore his son, and how, being divine, in the
end she could not dwell with him, but went back to her blue salt
caverns. Yet how little, as a rule, Peleus matters to his son !
When Achilles is in grief it is to his mother Thetis that he prays,
his mother Thetis that helps him. And few beings even in the
Iliad have the magic of that sea-spirit, so unearthly and yet
so tender.

No. Do not let us condemn too carelessly the home of the


pre-Hellenic peoples which knew of mothers and children, but
not much of husbands. Both forms of home must have acted as
powerful moral influences in man's life before the time of the
migrations by sea, and both equally were destroyed at that time,
and their divers ties and tendernesses battered out of existence.
' As for this trouble about Briseis,' says Agamemnon to the
envoys, ' tell Achilles that I will giv'^e him seven Lesbian women
down, and I promise him that, when we take Troy, he can pick
out twenty Trojan women — any twenty excluding Helen.' And
Briseis herself has not a proper name. The word Briseis is only
an adjective derived from the town of Brisa or Bresa in Lesbos.
She is ' the girl from Bresa '.

So much for the respect for woman which forms a part of the
tradition of both forms of home. And what of the father ? It
is interesting, though not strange, how keenly this question of
the treatment of fathers is felt. It was the same in the early
Aryan household, and throughout historical Greece. It is the
same, I should imagine, in all societies except those in which
people, like the rich at the present day, live on incomes derived
from accumulated stores of wealth and are consequently far
removed from the groundwork of human needs. In all poor or
precarious societies there is an assumption that the children owe
the parents a definite debt for their food and rearing. The
parents fed and protected the child when he was helpless. Now
that the old man cannot fight, the son must fight for him : when
he cannot work, the son must support him. Yet when men are
flying or fighting for their lives, when every weak hand or slow
foot brings danger to the whole party, there must have been
many old men left by their sons to save themselves as best they
might. The conscience of the Greek Saga was stirred on the
point. Not without purpose does it tell us how Aeneas in the
very flames of Troy, when every delay might mean death, would
not move without ' father Anchises ', and, when Anchises' strength
failed, faced all the dangers of flight amid armed enemies with
the old man upon his back. That is what the saga calls ' piety ' !
It is the other side of Hesiod's complaint, how the men of those
days, the generations that came just after the Trojan War, cursed
and deserted their old parents.

For there is a passage in Hesiod which reads almost as if it


were a direct description of this period of the Migrations, the
time when all the old sanctions which guided life have been
broken by the stress of a too great trouble. The passage comes
with an effect of interruption in the midst of the story of the Four
Ages of Man, the Golden, Silver, Bronzen, and Iron. Four they
must of course have been : but, as the poem now stands, there
comes a curious break after the Bronzen Men. They are
followed by the Heroes who fought at Thebes and Troy, and
they by the Iron race. This looks as if the Heroes were a mere
interpolation, and with the Iron Men we returned to the original
story. But the description of the Iron Men is in a style different
from that of the two earlier races. The Iron Men are not
creatures of mere idyllic badness. Through the dimness of the
half-childish story, through the formahty of the stiffly poeticized
language, one feels something of the grit of real life. And it is
a life very like that which we have just been analysing: the
homeless, godless struggle of the last migration. And it is as-
cribed to just the same point of history, the Dark Age which
followed p.eTa ra TpcoiKd, after the fall of Thebes, Troy, and
Mycenae {Erga., 156 ff,).^

But when the Earth had covered away this race also, then
Zeus son of Cronos made yet a fourth upon the land, more
righteous and valiant : the divine generation of the Heroes,
which are called half-gods of early times over the boundless
world. Bad war and awful battle slew them all ; some at
Seven-Gated Thebes, the land of the Cadmeans, died
battling about the flocks of the son of Oedipus : and some
War took in ships over the great gulf of the sea to Troy- land
for the sake of fair-haired Helen. Where verily the end of
death clouded them round.

And father Zeus, son of Cronos, gave them a life and
familiar places far away from men, settling them at the ends

^ It is almost impossible to date the subject-matter of a given part of the
Erga. As we have them, they represent early material, Boeotian, Phocian,
and other, in a late Ionized form. See on this point Lectures IV and V
below. The story of the Four Ages is probably of dateless antiquity ; the
addition of the Heroes and the re-shaping of the Iron Men may possibly have
been originally made in Ionia and afterwards taken over into the poetry
of the mainland. But the passing of the Arnaioi, Minyai, Lapithai, fragments
of Thracians and Phrygians, &c. through Boeotia, would produce equally
well the condition here described ; and it is simplest to suppose that the
whole passage, re-shaping and all, is Boeotian or Phocian. The Dark Age
affected the whole of Greece.

27i;o G


of the world, far from the immortals, and Cronos is king
among- them. And there they live with hearts untormented,
in the Islands of the Blessed, beside deep eddying ocean,
happy Heroes, and the mother of corn bears to them thrice
in the year her honey-sweet harvests.

Then the Fifth Men — would that I had never been among
them, but either had died before or been born after ! For
now is a race of iron. And never by day shall they have
rest from labour and anguish, nor by night from the spoiler.
The gods shall fill them with hard cares. . . . The father no
more kind to his children, nor the children to their father,
nor the guest true to the host that shelters him, nor comrade
to comrade : the brother no more dear to his brother, as in
the old days. Parents shall grow old quickly and be
despised, and will turn on their children with a noise of
bitter words. Woe upon them : and they hear no more the
voice of their gods ! They will pay not back to their
parents in old age the guerdon of their feeding in childhood.
Their righteousness in their fists! And a man shall sack
his brother's walled city.

There shall no more joy be taken in the faithful man nor
the righteous nor the good : they shall honour rather the
doer of evils and violence. . . .There shall be a spirit of
striving among miserable men, a spirit ugly- voiced, glad of
evil, with hateful eyes.

A spirit of striving, I have called it : the Greek is ^fjXo?, envy,
competition, the struggle for life. But observe that the end is not
yet ; though all normal sanctions have failed, the men of the Fifth
Age have still something to lose :

Then at the last, up to Olympus from the wide-wayed
earth, the beautiful faces hidden in white veils, away to the
tribe of the immortals, forsaking man, shall depart Atdos
and Nemesis}

How shall we attempt to translate the beautiful words ? ' Ruth

' There are interesting imitations of this passage in Eur. Medea, 439 ff. :
^(^aKt 8' opKCxiv X''P'f» oiiS' iT AiScos 'EX\(iSt to ixfyiiKa fihti, aWepta S' ayenra.
Also in the new (1911) papyrus of the Cynic poet Kerkidas : 'Afjuu 8e Umav
Ka\ fxer Atdas dyada /jieXfrW 6{os yap nvra Km ^ijifdis Kara yav. (' Paian '
here is the 'Healer' of sick humanity.) — Nemesis appears elsewhere as a
form of Artemis and as the Kore of Rhamnus, and Mr. A. B. Cook
makes the very interesting suggestion that Ne/xeo-ty is to vifios as Kaxeo-n to
\dxoi. Nemesis is thus like the Celtic Nefnctona, from nefneton, a sacred
wood, or the Latin Diana Nefnorensis. Her statue at Rhamnus had stags
in its crown and an apple-branch in its left hand. (Paus. i. 33. 3.) See
Appendix D.

Ill AIDOS and nemesis 83

and Wrath ' might serve. But let that pass for the moment.
The time which the prophet feared never came. Those two
goddesses stayed with man in his lonehest and worst hour, and
provided, if I read the history aright, the most vital force in the
shaping of later Greek ethics and poetry. A full understanding
of the word At'dSs would take one very far towards the under-
standing of all the hopes and creations of the Greek poets.

AlSco? is usually translated ' Shame ' or ' Sense of Honour ',
and Ne/zeo-iy, by an awkward though correct phrase, ' Righteous
Indignation.' The great characteristic of both these principles,
as of Honour generally, is that they only come into operation
when a man is free : when there is no compulsion. If you take
people such as these of the Fifth Age, who have broken away
from all their old sanctions, and select among them some strong
and turbulent chief who fears no one, you will first think that
such a man is free to do whatever enters his head. And then,
as a matter of fact, you find that amid his lawlessness there will
crop up some possible action which somehow makes him feel
uncomfortable. If he has done it, he ' rues ' the deed and is
haunted by it. If he has not done it, he ' shrinks ' from doing it.
And this, not because any one forces him, nor yet because any
particular result will accrue to him afterwards. But simply
because he feels aidos. No one can tell where the exact point of
honour will arise. When Achilles fought against Eetion's city,
' he sacked all the happy city of the Cilician men, high-gated
Thebe, and slew Eetion : but he spoiled him not of his armour.
He had atcfos in his heart for that ; but he burned him there as
he lay in his rich-wrought armour, and heaped a mound above
him. And all around him there grew elm-trees, planted by the
Mountain Spirits, daughters of Aegis-bearing Zeus.' ^ That is
aidos pure and clean, and the latter lines ring with the peculiar
tenderness of it. Achilles had nothing to gain, nothing to lose.
Nobody would have said a word if he had taken Eetion's richly-
wrought armour. It would have been quite the natural thing to
do. But he happened to feel aidos about it.

Aidos is what you feel about an act of your own : Nemesis is
what you feel for the act of another. Or, most often, it is what

^ Z 417. The word used is cre/3«f, not aldots : but in the connexion it comes
to the same.

G 2


you imagine that others will feel about you. If you feel disposed
to run away in battle, think of the pe/xia-i^ dvOpooiroiv ! People
will put that act to your account. When the elders of Troy look
upon Helen, ' Well,' they say, ' if men fight and die for such
a woman as that, ov vefxea-cs: none can blame them' (F 156).
Helen herself when she is expected — of course by a goddess :
no human being would be so shameless— to go to Paris and let
him make love to her immediately after he has emerged with
doubtful honour from his battle with Menelaus, refuses roundly :
' I will not go : veix^cra-qrbi' Si Kev ht) — it would be a thing to
feel nemesis at ' (P 410). When Achilles is justly angered with
Agamemnon, at first none can blame him (I 523) : but if he
persists after Agamemnon has sued for forgiveness, then there
will be nemesis : people will be indignant. He will know he is
doing wrong. (Observe, of course, that Nemesis does not mean

Let us follow this spirit of Nemesis for a moment, and then
return after wards to her still more interesting companion. In the
above instances the nemesis, the ' wrath ' or righteous indigna-
tion, has been that of definite witnesses or associates. There are
people who have seen your act, and know. But suppose no
one sees. The act, as you know well, remains veiiea-qrov —
a thing to feel nemesis about : only there is no one there to feel
it. Yet, if you yourself dislike what you have done and feel
aidos for it, you inevitably are conscious that somebody or some-
thing dislikes or disapproves of you. You do not look at the
sun and the earth with peace and friendliness. Now, to an early
Greek, the earth, water, and air were full of living eyes : otitheoi^
of dawtones^ of keres. To Hesiod and Homer they are ' myriads,
from whom there is no escape nor hiding'. one early poet^
says emphatically that the air is so crowded full of them that
there is no room to put in the spike of an ear of corn without
touching one. And it is they who have seen you and are
wroth with you for the thing which you have done !

The word Nemesis very soon passes away from the sphere of
definite human blame. Coarser and more concrete words are

* a\i<^\ he KTJpfs \ elXevvrai, Kevei) b' eicrdva-is ov8' aWepi, Bergk, F'r. adesp. 2 B,
reading aQipi, as is shown to be right by the quotation in Aeneas of Gaza
(p, 399 E). — See J. E. Harrison, Prolegomena, p. 170, note.


used for that: oveiSea, y\r6yoL. Nemesis is the haunting impal-
pable wrath of the Earth and Sun, the Air, the Gods, the Dead.
Observe, it is not the direct anger of the injured person : it is
the blame of the third person who saw.

Now let us be clear about one point. You will sometimes find
writers who ought to know better expressing themselves about
these matters in a misleading way. They say, or imply, that
when a Greek spared an enemy he did not do it from merciful-
ness or honour as we understand the words, but because it was
a part of his religion that Zeus would have a grudge against him
and punish him if he did otherwise. This may be true of a given
superstitious individual. But as regards the race it is putting the
effect for the cause. It was the emotion of the race that first
created the religious belief If the early Greeks believed that
Zeus hated the man who wronged a suppliant, that belief was
not based on any observed behaviour on the part of Zeus. It
was merely that they themselves hated the man who did so, and
felt that their god must hate him.

There are, then, certain actions which cause the feelings of
aidos and nemesis, of shame or ruth, when a man thinks of doing
them himself, of righteous Indignation when he sees them done
by others. Let us notice more closely what these actions
generally are. How far, for instance, do they coincide with the
objects of our own, or the mediaeval, feeling of ' honour ' } First
and most obvious, there are the actions that Imply cowardice :
they bring the simplest and crudest shame : ' Aidos, O ye Arglves,
will ye not stand ? ' ' Put In your hearts aidos and nemesis, . . .
I would not rail against one that was a weakling, for holding
back in battle : but you are chieftains ! . . . I have nemesis against
you in my heart ' (IV 95 ff.).

Secondly, actions that Imply falseness : lying and perjury. I
doubt If the word ever occurs In this sense In Homer, but that Is
because questions of false swearing never arise among Homeric
heroes. The false stories told by Odysseus In the Odyssey are
merely ruses of war. The treason of Pandaros Is something
which that unfortunate person might have felt shame for had he
lived. The poet himself seems a little ashamed of mentioning
such behaviour on the part of a hero, even a hostile hero, and
arranges as usual to lay the real g^ilt upon a god. Homeric


heroes do not need the aidos which prevents or ' rues ' falseness.
But it is common enough in Hesiod and Theognis and in tragedy.

Thirdly, actions that imply what we may loosely term im-
pudence or lack of reverence. The cases are few : Helen's words
above quoted are in point. So no doubt would be the boldness
of Niobe in boasting herself against the goddess Leto (/2 602 ff.),
or the impudence of Thersites in the second book of the Iliad.

All these might be included as objects of any current concep-
tion of ' Honour ' : but there is a fourth sense, by far the most
widespread and significant, which reaches a good deal beyond
the ordinary mediaeval ideal. It is the horror of cruelty or
treachery towards the helpless. Any sympathetic reader of
early Greek poetry will have noticed the importance, or indeed
the sanctity, attached to three classes of human beings : strangers,
suppliants, and old people. What is there in common between
the three.-* Nothing, I think, but their helplessness. Realize
what a stranger is, in a primitive society. He is a man with no
home, no friends, no one to protect him from injury, no one to
avenge him afterwards. He has not even his own sanctuaries to
shelter him, or his own tribal god. And again, a supphant :
a suppliant is any man or woman who formally casts away all
means of self-defence and throws himself upon your mercy.
That is the essential thing ; though of course, when he could, the
helpless man tried to influence your feelings in divers other
ways. He associated himself with something that you held
sacred. He sat on the steps of an altar : he touched some sacred
object : he lay on your door-step and threatened to starve unless
you took him in ; he contrived with his hand to touch your face
or your beard. But those are all accessories. The essential is
confessed helplessness. And all their literature shows what
horror the early Greeks felt at the notion of definitely and
formally rejecting a prayer made by the helpless, a horror some-
times amounting to what we should call moral weakness. They
expressed this generally in theological language. ' The stranger
and the suppliant come from Zeus.' ' Zeus is the watcher of
stranger and suppliant ' [l 270) ; ' The very Thunderer follows
the alSoios //C6r?;y ' (?; 165, 181); his own titles are 'iK^TtjcrLos
and Uitvios and even ' A(l)LKTOi>p}

^ on Zei/s 'A^iKTcop, see Lecture XI, p. 275.


And thirdly, old people. Here there enters in, no doubt,
some element of the patriarchal sanctity of a father ; but I think
that the helplessness of age is again the main reason for an old
man or woman being ai8oTo9. That explains why they are, like
beggars, strangers, suppliants, especially under the guardianship
of the gods, and in particular of Zeus. It explains why the older
they are the more is their claim on Aidos : why the blind are
classed with them.^ It may . be objected that, if helplessness is
the criterion, children also would be aiSoioi. The answer is
interesting. Ordinary children are not specially aiSoioi, or
charged with sanctity, because they have their grown-up relations
to protect them. But orphan children are.

There are some five deadly sins, says Hesiod in the Brga, of
which you cannot say that one is worse than another. They
are all beyond 1;he pale {Erga^ 327 ff.) :

It is all as one thing — the man who does evil to
a suppliant and to a stranger ; the man who goeth into his
brother's bed; the man who in heartlessness sins against
orphan children ; the man who reviles his old father on the
bitter threshold of age, laying hold of him with hurting
words : with that man Zeus himself is wroth.

These sins consist of four offences against the helpless and one
breach of a fundamental family taboo. All adultery was a most
grave offence. But if this particular form of it is chosen as the
worst, that is the doing of Aidos. Your brother trusts you, and
is often at your mercy. That is what makes him sacred.

For apart from any question of wrong acts done to them, there
are certain classes of people more aiSoioL, objects of aidos, than
others. There are people in whose presence a man feels shame,
self-consciousness, awe, a sense keener than usual of the im-
portance of behaving well. And what sort of people chiefly
excite this aidos ? Of course there are kings, elders and sages,
princes and ambassadors: alSoloi ^a(nXf]e9, yipovTes, and the
like : all of them people for whom you naturally feel reverence,
and whose good or bad opinion is important in the world. Yet,

^ Cf. Soph. O. T. Z7A-7i where commentators, from not seeing this point,
have altered the text. Oed. ' Thou art a child of unbroken night, so that
neither I nor any other who sees the light would {av) ever harm thee.' Ttr.
• It is not my doom to fall by thy hand,' &c. So MSS., and cf. 448 below,
where Tiresias repeats the same statement.


if you notice the language of early Greek poetry, you will find
that it is not these people, but quite others, who are most deeply
charged, as it were, with aidos ; before whom you feel still more
keenly conscious of your unworthiness, and whose good or ill
opinion weighs somehow inexplicably more in the last account.
The disinherited of the earth, the injured, the helpless, and
among them the most utterly helpless of all, the dead.^ All
these, the dead, the stranger, the beggar, the orphan, the merely
unhappy, are from the outset alSoToi^ 'charged with aiSm.'
Wrong them, and they become, ipso facto and without any word
of their own, dpaiot or Trpoa-rpoTraLOL, incarnate curses, things
charged with the wrath of God.^

The feeling seems to have been very strong. one must bring
it into connexion with the various stories of gods who were dis-
guised as beggars, and went through the world ill or well entreated
by different men according to their different natures. It is the
counterpart of what we, in our modern and scientific prose, call
' a sense of social responsibility ' or the like ; the feeling roused
more or less in most people by the existence of great misery in
our wealthy societies. To the Greek poet it was not scientific,
and it was not prose. It was an emotion, the keener because it
was merely instinctive and was felt by a peculiarly sensitive people ;
an emotion of shame and awe, and perhaps something like guilt,
in meeting the eyes of the oppressed of the earth ; a feeling that

^ * Do you feel aidos for the dead body of one that hated you ? ' the wise
Odysseus is asked in the Ajax; ' His goodness is more to me than his hate'
is the answer, an answer full of aidos {Ajax, 1357). ' The stranger and the
beggar are charged with aidos,' says Eumaeus in the Odyssey, and the
adjective nlbolo': is a regular epithet of a stranger. But mere unhappiness
is enough : ' A miserable man must needs rouse aidos in you,' says Oedipus
{O. C. 247).

* UpooTponnios is not *• turning 07ieself towards', as L. and S. say: it is
the adjective from npoarponf] which is the opposite of diroTponr], ' aversion.'
As you can by sacrifice, &c., try to 'avert' the dniidova? so you can 'bring
them upon ' somebody. Thus an injured suppliant has a power of npoa-Tpoirr] :
he brings down the gods upon his injurer. A criminal brings them down on
himself and those who are infected by his cfyoy. These words are very often
misunderstood ; e.g. the cfydoyyov apnlov o'Ikois of Iphigenia {Ag. 237) was not
a spoken curse — which would make the passage hideous — but the mere
crying of a murdered daughter, which necessarily involves an dpd. So when
Philoctetes charges Neoptolemus to look him in the face : t6i> irpoarponainv,
Tov iKiTrjv, S) crx^'^^^^ 5 he means : ' Me, charged with the wrath of God ; me,
who kneel before thee, O hard heart ' (P/nt. 930). — I do not mean to deny
that you can say in Greek rpaneadai eVl or Trpos ea-Hav. Aesch. C/to. 1038,
Eum. 205.


a wrong done to these men is like no other wrong ; that what
these men report of you ukimately in the ear of Zeus will outweigh
all the acute comments of the world and the gratifying reports of
your official superiors.^

If you look into the history of later Greek Ethics, it is rather
a surprise to find how small a place is occupied by Aidos.
Even to Plato and Aristotle it has become little more than an
amiable quality, the absence of which is particularly repulsive.
It has quite ceased to be the guiding force of men's moral life.
These two philosophers, of course, belong to a particular school :
they are aristocratic and intellectual ; both perhaps too much
inclined to despise those emotions which appeal to man's simplest
instincts and have a touch of the animal in them. If we possessed
any complete books by the more democratic and less authoritarian
philosophers, by Protagoras especially and Democritus, our im-
pression might be different. Among the philosophers of the
Roman period Aidos has quite faded away. It plays no part in
Epictetus. It is barely mentioned by Sextus Empiricus. only
Kerkidas the Cynic, rejector of all organization and system and
convention, falls back to primitive feelings and asks that life shall
be guided by Paian — the Healer— and Aidos. one can see the
reason for this ; indeed, the many reasons.

For one thing, Aidos is a mere emotion, and therefore incalcul-
able, arbitrary, devoid of principle. A man may happen not to
feel the emotion, and then you have nothing to appeal to. Or
again, if he has the emotion, there is no way ofjudging its strength.
An emotion which is made the whole moving principle of conduct
grows with what it feeds upon : it is never sated : it moves towards

* I have sometimes wondered how it happens that slaves are never spoken
of as charged with aidos. A particular slave may be treated with aidos.
He may be protected and helped because he is a stranger or a beggar. But
the word is not regularly applied to a slave. I think the reason is, as
Euripides says, ' Why speak of ruth where ruthlessness is the law ? ' The
whole institution was a negation of Aidos ; a refusal to listen to the emotion
m question. If you made a man your slave, that showed you did not regard
him with aidos. So the less said about it the better. As the Ocean Spirits
in the Prometheus tell us— with a different meaning— the clank of the
riveting of a prisoner's fetters frightens Aidos away {Prom. 134). Of course
a wrong done to a slave was hated by the gods and, one might hope, duly
avenged. But that was the same with animals. EiVl ki\ kwShj e'pcvCfs— there
is vengeance m heaven for an injured dog. on the ramifications and possible
origin of Aldas, AiKi;, OpKos, Scc, see Appendix D.


the infinite. That way madness lies, as the lives of so many of
the saints have show^n us. Besides, behind any morality based
upon emotion there is the question whether you ought or ought
not in a particular case to feel the emotion : and if not, why not .-*
It is there that the real principle of Ethics comes in. The later
philosophers wanted to understand, not merely to feel. They had
to build up conduct into a consistent rational system. It would
help them little if men said, ' Follow the leading of Aidos,' ' Love
your neighbour,' ' Pity humanity.' Such rules will help tjie con-
duct of men. But they do not provide an answer to a speculative
problem. Perhaps the main thing which the philosophers got
from Aidos was Aristotle's doctrine of the Mean : the observation
that in any emotion or any movement there is a possible best
point, which you should strive to attain and shrink from passing.
A great liberating doctrine, no doubt ; but one with the emotion
all gone from it. That was what served Aristotle's purpose best.

Again, there is an historical reason for the decline in the im-
portance of Aidos. Aidos, like Honour, is essentially the virtue
of a wild and ill-governed society, where there is not much
effective regulation of men's actions by the law. It is essentially
the thing that is left when all other sanctions fail ; the last of the
immortals to leave a distracted world. In an ordered society
there are all the more concrete sanctions to appeal to — the police,
the law, organized public opinion.

In a well-organized society the great majority of men are under
compulsion to behave better than they naturally would, if left to
themselves. It often strikes me, in certain parts of early Greek
poetry, that one gets a glimpse of a society in which, by the
breaking up of ordered life, men were compelled to be worse than
nature intended ; where good and merciful men had to do things
which they hated afterwards to remember. one may recall the
character in Herodotus,^ who wished to be the most righteous
man in the world, but was not permitted by circumstances. As
a rule in fiction (where motives of flattery cannot come into play)
rich men are wicked. It is obviously more interesting, as well as
more gratifying to the reader's feelings, to make them so. But
in Homer the rich men are apt to be specially virtuous ; d(f>P€ib9
dfivficov, 'rich and blameless' (E 9). one is reminded of the

' Herodotus, iii. 142.


naive desire of the old poet Phokylides, first to acquire a com-
petence and then to practise virtue. The project is amusing to
us, as it was to Plato. We know so much of the result of that
scheme of life. Yet think of that son of Teuthras in the Iliad^ who
' dwelt behind the strong walls of Arisbe, rich in all livelihood,
and was beloved of men. For he built his dwelling by the road-
side and showed love to all who passed.' ^ one might almost
think that, like Phokylides, he had made some resolution to
' practise virtue ' in this way, feeding the hungry and washing
the feet of strangers. But, in any case, it is easy to imagine how,
in a time like that of the Migrations, a decent man who had passed
through the horrid necessities of the struggle for bare life, and
was at last safe and prosperous with a strong wall around him,
would become just like these rich men in Homer, thankful to live
at last blameless and gentle towards gods and men.

The suggestion is little more than a fancy. But it occurs to
me in connexion with another. When we compare the civilization
and character of Greece and of Rome, we are struck, among
many other differences, with some broad general divergence.
The Roman seems to have all the faults and the virtues of
successful men. He is severe, strong, well-disciplined, trust-
worthy, self-confident, self-righteous, unimaginative, a heavy
feeder, a lover of gladiatorial games. The Greek, less gregarious,
less to be relied upon, more swept by impulse ; now dying
heroically for lost causes; now, at the very edge of heroism,
swept by panic and escaping with disgrace ; capable of bitter
hatreds and massacres in hot blood, of passionate desires and
occasional orgies ; but instinctively hating cruelty, revolting from
the Roman shows, frugal, simple and hardy to a degree which we
can with difficulty realize : above all, possessed of an unusual power
of seeing beyond himself and of understanding his enemies ; caring
for intellect, imagination, freedom, beauty, more than for force
and organization, crying aloud for orderliness and symmetry, in
part from mere artistic sense, in part because he knew his own
needs and dangers ; much as Plato prayed to be delivered from
poetry because poetry was to him a seducing fire. The causes
of such a difference are innumerable. There may have been a
greater proportion of pre-Aryan elements in Greek civilization.

' z 15.


There were important geographical differences. But one cause,
I think, is the early experience of the Greek race during the
great sea-migrations. The Romans had an almost steady history
of stern discipline, of conquest and well-earned success : the
Greeks at the beginning of their history passed through the very
fires of hell. They knew, what Rome as a whole did not know,
the inward meaning and the reverse side of glory. They knew
the bitterness of lost battles, the sting of the master's lash ; they
knew self-judgement and self- contempt, amazement and despair.
They had their triumphs and conquests. They must, I suppose,
be counted, even politically, among the most successful races of
mankind. But in their highest successes, in the times both of
Pericles and of Alexander, there is always something daring and
precarious. Their armies are always fighting against odds ; their
little cities trying by sheer energy and intellect to stem the strength
of great military empires. It is a wondrous fabric held together
for an hour by some splendid grasp of human genius, not one
based on strong material foundations by the gregarious and half-
conscious efforts of average men. They began their life as a
people, it would seem, in a world where palaces and temples
were shattered, armies overthrown, laws and familiar gods brought
to oblivion. Perforce they questioned and they wondered. It
is not for nothing that Greece produced Heraclitus and Plato,
thinking that all things pass like a stream and that the idea is
the only abiding reality. Thus, like the prophet in Callimachus'
great poem, they saw early the world that is behind the ordinary
world of human strivings, more real and more intangible ; and
throughout their history somehow this ideal haunted the race, a
vision perturbing their sight, unfitting them for continued empire,
yet shedding strangely over their defeat a splendour denied to
their conquerors.'

^ Call. Lavacra Palladis 87 reKvov aXacrre, \ ft'Ser ^ kBavaias (rradfa kqi
Xayovas, \ dXX' oi>k deXiov naXiv oyj/tai.



So far we have been considering the people: I wish now
to turn to the literature. For one of the clearest facts that we
know about these driven fragments of society who form the soil
from which Hellenism sprang is that they must have had a
literature. The vast store of prehistoric tradition preserved in
the Greek heroic saga is evidence enough. The Northerners
can scarcely have known the art of writing before some few of
them learned it in Greece. But it is quite conceivable that in
very early times they possessed epic lays, and that these lays
were in dactylic verse. So much we can conclude from various
formulae imbedded in the Homeric language. on the other side,
the Cretan script, coming on the top of other evidence which
was already sufficient, shows that long before the Migrations
there were scribes and ' wise men ' in the Aegean who had the
power of writing.

I am not proposing to discuss the Homeric Question, but rather
to put forward some general considerations preliminary to the
Homeric Question. If the men of the Migrations possessed a
literature, that literature was not in the least what we mean by
' Homer ', viz. the Iliad and the Odyssey. It was not even what
the Greeks of the sixth and early fifth centuries meant by ' Homer ',
viz. the whole body of heroic tradition as embodied in hexameter
verse.^ It must really have been something far more primitive
and less differentiated, of which the didactic epos, the lists of
ancestors, the Stesichorean lyrics, the local chronicles, the theo-
logical, magical, and philosophical writings, as well as the heroic
poems, are so many specialized developments. It has long been
clear to students of early Greece that the Iliad and Odyssey are
not primitive poems. Not only their art and construction, but

^ See Wilamowitz, Homerische Uniersuchtmgen, pp. 329-80.


their whole outlook on the world and the gods is far removed
from that of the most primitive Greeks known to us. Both
poems, indeed, contain a great deal of extremely ancient matter :
but both, as they stand, are the products of a civilized age and a
long process of development. It is the beginnings of literature
that we are now considering.

Let us begin by trying to imagine the position and practice in
an early society, say in Ionia before Thales, or on the mainland
before Solon, of what the Greeks generally described as a Xoyioy
dj/rjp, or ' man of words '.^ I say ' words ' because I despair of
an adequate translation oiLogoi. The conception Logos ^ ' word '
or 'speech', had, as we all know, a peculiarly distinguished
history among the Greeks. It was the word spoken : it was the
power of language ; it was the word which implies reason,
persuasion, interpretation, and which settles differences instead
of the armed hand ; it was thus the word which mediates between
the soul of man and man, or, in theological language, between
man and God ; to the philosopher it was the silent but eternal
word upon the lips (5f Nature, the speech by which the Cosmos
expressed its inborn reason. But for our present purpose it is
another aspect of the Logos that comes into play. The Logics
Aney^ or Man of Words, was the man who possessed the Things
Said, or traditions, which made up the main sum of man's
knowledge. He knew what Logoi really existed, and what were
mere inventions or mistakes. He could say Xoyos" kcrriv^ much
as a Hebrew could say ' It is written '. This implies, what is of
course the case, that Greek saga was mainly preserved by oral
tradition.^ Yet it would be rash to assume that there were no
writings. The extant Cretan records are far earlier than any
possible Homer. The ancients themselves tell stories of the
' books ' of the early minstrels. The use of MSS. by the
composers of our Iliad and Odyssey seems almost as certain as

^ See also Prof. Butcher's Lecture on ' The Written and the Spoken Word '
in Some Aspects of the Greek Gejiitis.
' Ar. Frogs, 1052 : ' What I said about Phaedra, was it not an tiv Xo-yoy ? '
' I see that Drerup has mistaken my meaning, Omero, p. 68, note. I
discuss the books because they are there to discuss ; the oral tradition in each
case was more important, as I explain, but it has vanished. The MSS. of
the Rohifid still exist, but no one living can hear ' Thorold ' or Taillefer
improvising. See the excellent remarks of F. Bolte on Rhapsodische Vor-
tragskunst in Netie Jakrouche?; 1907, I. Abt. xix. 8.


such things can be, and, though those composers themselves
belong to a much later date, the fragments of minute and, we
may add, uninteresting history preserved in the epic suggest the
use of some surer and more positive method than mere oral
tradition. The Man of Words, we may assume, would in many
cases not trust entirely to his memory, but would make a
permanent Logos of his own in the shape of a book.^

A book in those days was not what it is now. It was not
a thing to be given to the public, not a thing to be read
for pleasure." one can find parallels in the East or in the
Middle Ages. There was the great book of Michael Scott, the
magician, which was read by no man but one, and was
buried in its master's grave. There was the book of Thoth,
carried off by Nefrekepta ; the Book of Catyllus, reported by
the Spanish Mandeville.^ There is the great list of Arabic
chronicles, the rule of which is that each chronicle was the
property of the author or of his heir, and could not be read by
others without his permission. There are the innumerable and

^ Mr. Lang suggested to me the comparison of the Gaelic ' sennachie ' :
' sean ' = old ; * seanachas ' = story, tradition ; ' seanachaidh ' (pronounced
' shen-ach-ay ') = a man of tales, historian. It seems quite clear that the
sennachies could not read or write.

^ All through antiquity a book remained a thing to be recited from, or to
be read aloud to an audience by a skilled person. It is partly due to facts
like this that the oral repetition of stories continued so extremely late in
human history to be the normal way of keeping alive the records of the past,
even if the past was vitally important. In the case of the Gospels, for
instance, where a modern would have considered it of absolutely overwhelm-
ing importance to have a written record as soon as possible of the exact
deeds and sayings of the Master, we find, as a matter of fact, that it was left
for a considerable time to oral tradition. Compare the well-known phrase of
Papias (died c. a.d. 135), deliberately preferring a third-hand oral report to
the written word : —

'Whenever any person came my way, who had been a follower of the
Elders, I would inquire about the discourses of those elders, what was said
by Andrew or by Peter or by Philip or by Thomas or James, or by John or
Matthew or any other of the Lord's disciples, or what Aristion and the Elder
John, disciples of the Lord, said. For I did not think I could get so much
profit from the contents of books as from the utterings of a living and abiding
voice.' (I cite from Estlin Carpenter's First Three Gospels, p. 4.)

In the time of Papias there were libraries with books by the hundred
thousand, yet a book is still to him a dead and troublesome, mode of com-
munication. He is said to have been rather a stupid man, -navv afUKpos top
voiv. But a thousand years earlier than Papias this attitude of mind was the
normal one.

' Gr\^i\\'% Stories of the High Priests of Me7nphis\ Spanish Mandeville,
fol. 1376. (I owe this reference to Mr. W. R. Halliday.)


constantly varying MSS. of stories like the Arabian Nights^
each copy originally meant to be the private stock-in-trade of a
professional story-teller. In all these cases the man lived by his
book. It must be kept from the public ; above all, it must be
kept from the eyes of professional rivals. It can be given or
bequeathed to a son or a favourite disciple, as in the Greek story
one of Homer's scrolls, the ' Cypria ', served as his daughter's
dowry, another, the 'Taking of Oechalia', was left to his heir,
Creophylus.^ For the ancient Man of Words was not exactly a
story-teller, not exactly a chronicler, not exactly a magician.
He was all three, and something more also. His Logos con-
tained, with no distinction of subject, all that he specially wanted
not to forget, or, at least, all that was worth the immense trouble
of writing down, letter by letter.

There was an ancient Greek tradition, superseded in general
by the Cadmus story, which somehow connected the invention
of writing with Orpheus and the Muses. Orpheus' voice seems
to have recorded itself in books in some mysterious way.^ And
the Greek bards always owe, not only what we should call their
inspiration, but their actual knowledge of facts to the Muses.
The Muses 'are present and know all things'. They are, to
Hesiod at least, ' the daughters of Memory.' " Hesiod professes,
roughly speaking, to be able to sing about everything ; but he
always explains that he is dependent on the Muses for his know-
ledge. Other sources ofknowledge are indeed recognized. When
giving the names of all the rivers in the world, Hesiod stops at a
certain point and says that for the names of the rest you had better
consult the people who live on their banks, and they will be able
to tell you {Theog. 370). But most often he consults the Muses
{Tkeog. I ff., 105 ff., 966, 1022, Catalogues). So does Homer for
such subjects as the Catalogue of the Greek army (cf. a 'j^B 486,
761, cf. M 176). one suspects that that consultation was often

^ Cf. the case of Jendeus de Brie, author of the Bataille Loquifer, cent,
xii : he * wrote the poem, kept it carefully, taught it to no man, and made
much gain out of it in Sicily where he sojourned, and left it to his son when
he died'. Similar statements are made about Huon de Villeneuve, who
would not part from his poem for horses or furs or for any price, and about
other poets. — Gautier's Epopees Fjafigaises, vol. i, p. 215, note i, cited in
Lang, Homer and his Age.

* Qp/j(T<rnis iv aaviaiv, ra? | 'Op(f)(ia KnTeypn\l/fii | ytjpvs, Kur. A/c, ^67~9'

^ T/ieogony, 54, 916: for subjects, 100-15.


carried out by the bard retiring to some lonely place, or maybe
barricading the door of his hut, bringing forth a precious roll, and
laboriously spelling out the difficult letter-marks. rpd/x/iaTa,
the Greeks called them, or ' scratches '. And right on in mid-
classical and later times the name for a scholar was ' grammatikos '.
He was a ' man of grammaia ',^ one who could deal with these
strange 'scratches' and read them aloud, knowing where one
word ended and another began, and when to make big pauses
and little pauses. For things like that were not indicated in the

You will have noticed that a wise man in antiquity— and the
same is true of the Middle Ages — generally has a boy or disciple
attached to him. And the first thing which that disciple learns
when he begins to be ' wise ' himself is to read in his master's
book. Not in any book, mark you. They did not learn reading
in that way. You were not expected to understand the grain-
mata unless they were first read aloud to you. The case is
clearest with Semitic books, where the vowels are not written at
all, and in some cases the meaning cannot possibly be made out
for certain without help from the writer of the book. But it was
the same in the Middle Ages: with Michael Scott's book, for
instance. It was the same with various of the old Sanskrit books,
the meaning of which has in some places been absolutely lost
because there was a breach in the series of disciples to whom the
meaning was orally explained by the master. The thing that
most tangibly constituted a disciple was the power to handle, or
to read in, his master's book. Of course a very clever man would,
if you gave him time, be able eventually to make out other books
too. But that would be a special undertaking.

This limitation. If you think of it, is inevitable. In the first place
there will probably be no other books In the neighbourhood on
which to practise. Then further, it must be remembered that, as
the man's book is a private thing, so also is his method of making
signs. Handwritings always differ; and the handwriting of
a man who practically never saw any other person's handwriting
and who used his own merely to make notes for his own private
use, not to be read by others, would be sure — even apart from

^ See Rutherford's Scholia Aristophanica vol. iii. chap, i, and my Religio




the writer's own conscious wish for secrecy — to grow in a
hundred little ways specialized and abnormal. I have seen an
Arabic book which professes to give the special alphabets ^ used
by the ancient sages, Cleomenes, Plato, Pythagoras, Scalinus,
Socrates, and Aristotle, all of them different, ' in order that none
should know them but the sons of wisdom.'

Consider, then, the position of a man who possesses such
a book, and also can make gravtinaia himself. Suppose he
hears news of stranee events which he would like to record
accurately. Suppose he is lucky enough to hear another wise
man expounding new lore, or giving details on a subject
where his own book is vague. Suppose he finds, or borrows, or
inherits from a wise relation — wisdom runs in families — another
book containing valuable information. In all these cases he will
want to make additions and changes in his own book. Let us
consider how he is likely to set about it.

It is a difficult process to conflate two or more accounts of a
transaction into one, difficult even for a modern writer, with
all the battery of modern appliances at his command ; clear print,
numbered pages, indices to show you just where and how often
a subject is mentioned, paragraphs and chapters, divisions of
words and sentences, and abundance of cheap paper for making
notes and rough copies. Our ancient sage had his book written
on very expensive material, usually the skins of beasts carefully
prepared. He could not lightly throw away a scroll and write
it again. He had no facilities for finding references ; no index,
no pages, no chapters, no stops between sentences, no divisions
of any sort between one word and another ; only one long un-
divided mass of graminata^ not by their nature well calculated
to be legible. on the other hand, he probably knew his own
book by heart. It was an advantage which sometimes betrayed

What he generally did was to add the new matter crudely at
the end of the old. He could write on the margin or between the

^ Ancient Alphabets, by Ahmad bin Abubekr bin Wahshih, translated by
Joseph Hammer, London, 1806. 'Every one of these kings invented,
according to his own genius and understanding, a particular alphabet in
order that none should know them but the sons of wisdom ' (p. 14). Are the
'sons of wisdom' the disciples of the wise? The book is said to have been
written An. H eg. 241. It is concerned with alchemy.


lines. At a pinch, he might cut the hide with a knife and sew
in a new strip at a particular place. He had only to make the
roll intelligible to himself. And any one who has had experience
of the difference between a MS. fit to be sent to the printer and
a MS. that will do to lecture from will appreciate what that means.
No book has come down to us from antiquity exactly in this
state. All the books that we possess have at some time been
published, and therefore prepared in some sense to be intelligible
to the reader. But many Greek books retain clear marks of the
time when they were not meant to be read by strangers, but
only to serve the professional needs of the writer. The later
Homeric hymns, containing merely a number of suitable openings
and closes for recitations, point pretty clearly to the handbook of
the professional reciter. The voluminous writings of the Peri-
patetic school which come to us under the name of Aristotle bear
innumerable traces of their composition for private use in the
school. So do the remains of Hesiod ; so do, as far as I know
them, most of the late magical writings. In oriental literatures
the instances are, I believe, even clearer.^

In imagining the proceedings of this old sage we have taken
one particular crisis, as it were, in the history of his book. But
all the ancient traditional books which have come down to us
have, without exception, passed through many such crises. The
book which contained the whole Logos of the wise man was apt
to be long-lived. It was precious ; it had been very difficult to
write ; it was made of expensive and durable materials. It

^ Peculiarly instructive is the record of the first writing down of the text of
the Koran. Islam, being historically a late religion, has its origins exception-
ally well attested. Zaid Ibn Thabit was entrusted by Abu-Bakr, the first
Caliph, with the task of collecting the Prophet's revelations — Surahs —
preserved in part only in the breasts of the faithful. He made one official
copy for the Caliph, being guided by his general knowledge of the credibility
of his witnesses. As the informants naturally varied in dialect, a question
arose as to the genuine dialect in which the revelations came; this was
determined to be Korashite. The third Caliph made a public edition, thus
really establishing the Koran. Unofficial copies proved to be in existence.
These were all destroyed, and official copies sent out to the chief towns.
The same editor was employed for this work of publishing, perhaps because
only he could read the first edition with any certainty. The official copies
were, after all, not much more than 7nemoriae technicae. one who had read
the text with a teacher could afterwards recall what he had read thereby ;
one who saw the text for the first time would be confronted by an enigma. —
I take this from Prof. Margoliouth's AIohammedanisi?i, chap. ii.

H 2


became an heirloom : and with each successive owner, with each
successive great event in the history of the tribe or the com-
munity, the book was changed, expanded, and expurgated. For
the most jealously guarded book had, of course, its relation to
the public. It was not meant to read ; it was meant to recite
from. The Logos only came into full existence when spoken to
an audience, and obviously it had to suit that audience. It w^as
the source of stones and lays which must needs be interesting ;
of oracles and charms and moral injunctions which must not seem
ridiculous or immoral ; of statements in history and geography
which had better not be demonstrably false. The Logos must
needs grow as its people grew. As it became a part of the
people's tradition, a thing handed down from antiquity and half
sacred, it had a great normal claim on each new generation of
hearers. They were ready to accept it with admiration, with rever-
ence, with enjoyment, provided only that it continued to make
some sort of tolerable terms with their tastes, under which general
head we must include their consciences and their common sense.

I am tempted to take Instances from our own times to illustrate
w^hat I mean by a traditional book. The most obvious of our
traditional books are collections of mere information like
Whitaker's Almanack and the Statesman's Yearbook, or those
strange prophetic Almanacs and magic Herbals which continue,
I suppose, to enjoy a flourishing though subterranean existence
in all European nations. Or we might take the various Guides
to Navigation published by various countries. The Pilot series,
issued by the British Admiralty, seems now to hold the field ;
but M. Victor Berard ^ has traced its origin step by step from a
remote past, through French, Italian, Dutch, Spanish, Latin,
Greek, and perhaps Phoenician sources. An historical lawyer,
again, could show the same process of traditional growth in
various legal codes. Such literature reminds one much of the
Catalogues of Hesiod.

But the best modern instance of a traditional Xoyos is provided
by stage plays. A play which lives beyond a century or so
regularly consists of (i) a slowly changing book to guarantee
its potential existence and (2) a series of frequently changing

^ Les Phinicie7is et VOdyssie, i. p. 52.


performances before an audience to bring it to fullness of life.
The best parallels to an ancient lay, of course, would be those
plays which are not printed, but exist only in the prompt-books
and the actors' memories. Unfortunately, under modern condi-
tions, it is almost impossible for a successful play to live long
without getting printed. And the printed book with its hundreds
or thousands of identical copies produces a terrible fixity of text.
The traditional Mummers' Play, having had no authoritative text,
has almost perished of degradation. But one would like to have
records of the variations of George Barnwell or The Stranger
between their first production and the times when Thackeray
saw them ; or even of plays like East Lynne and The Private
Secretary^ which are acted all over the English-speaking world
at the present day. But let us take at once the most august

Hainlety as every one knows, long before it was an English
play was an ancient Scandinavian story, not invented by any one,
but just living, and, doubtless, from time to time growing and
decaying, in oral tradition. The first mention of it is in a song
composed about the year 980 ; the first complete version is in
Books III and IV of Saxo Graiumaticus about the year 11 85.
There is a later form in the Icelandic ' Ambales Saga ', and,
of course, there are innumerable variants in all parts of the

There was a play called Hamlet extant in England about
the year 1587, apparently not by Shakespeare and doubtfully
attributed to Kyd. Many difficulties and peculiarities in Shake-
speare's Hamlet are presumably due to the unconscious discord
between the old matter which he took from Kyd and the new
matter which he invented.^ The first version of Shakespeare's
Hamlet is in the Quarto of 1603 (perhaps really printed in 1602),
a version widely different from that which we generally read.
It has only 2,143 verses as against 3,891 in the Globe, the order
of the scenes is occasionally different, the names of the characters
are not all the same : for Polonius we have Corambis, and for

' See Corpus Hamleticicm, by Professor Josef Schick, Munich, vol. i, 1912,

^ See especially Hamlet, an Historical and Comparative Study, by Prof.

Elgar Stoll, Univ. of Minnesota, 1919 : The Genesis of Hamlet, by Prof. Lewis

of Yale (Holt, 1907) : Hamlet once More, by the Rt. Hon. J. M. Robertson,



Reynaldo, Montano ; the Queen, too, is definitely innocent of her
husband's murder.

The Second Quarto is dated 1604 and describes Itself as
' enlarged to almoste as much againe as it was, according to the
true and perfecte coppie '. The First Folio was published after
Shakespeare's death in 1623. It omits a good deal that was in
the Second Quarto and contains some passages which are not in
the Second Quarto but were in the First. In nearly all cases,
however, it has rewritten and altered them. There are many
critical questions about these three versions which we can here
pass by, but surely one thing is clear to any one who uses his
imagination. The three versions which happened to be printed
cannot possibly represent all the variations of the play which
were spoken on the stage. Shakespeare took an old play and
rewrote it. He rewrote his own play again and again for
many years. He added a great deal to it, but he sometimes
cast out again his own additions. Who can tell what additions
or what cuts were made at various performances where Shake-
speare himself was producing ; what additions or what cuts by
other producers, when Shakespeare was not there ? And after
his death, though as far as we know strange hands did not make
further additions to the play, as they certainly did to the
Homeric poems and to some Greek tragedies, there are great
variations of text all through the eighteenth and nineteenth

There is a very instructive book in twenty-five volumes called
The British Theaire^hy Mrs. Inchbald, published at the beginning
of the nineteenth century. It consists of a collection of the plays
acted at Drury Lane, Covent Garden, and the Haymarket,
printed from the prompt-books then in use under the authority
of the managers. one can there see what shape Richard III
had assumed after Colley Gibber and other distinguished people
had produced it. one can see Restoration comedies expurgated
in accordance with later eighteenth-century taste ; Congreve
making references to the Coronation of George IV; and some
old wine in Farquhar's ' Beaux' Stratagem ' altering its date from
1706 to 1792. The process is normal, inevitable, and, to my
judgement, entirely right, if the plays are to live and hold an
audience. It differs from the similar process in antiquity chiefly


because the ancients in pre-classical times had no sense of
obh'gation to the authentic words of an author, and, in any case,
were not, like the adapters of the eighteenth century, chained by
the leg to a printed text. The eighteenth-century producers
could not treat Hamlet as Hamlet himself proposed to treat The
Murder of Gonzago : ' Set down and insert in't ' on the spur of
the moment ' a new speech of some dozen or sixteen lines.' They
left out more than they added ; in particular, they expurgated
profanities. They said ' For Heaven's love ' instead of ' For
God's love' and the like. They expurgated what seemed
grotesque or violent, or what the Alexandrians would have
called ttTTpeTrey, ' unseemly '. They did not let Marcellus hit at
the ghost with his partisan (Act I, Sc. ii). They cut out all the
' Old mole ' and ' Truepenny ' business in Act I, Sc. v. They
removed the indecencies of Hamlet's conversation with Ophelia ;
and, while they left in the whole of ' To be or not to be ' without
a word changed, they cut out the whole of ' O, what a rogue and
peasant slave am I ! ' And this, roughly speaking, was the form
in which Hamlet gradually conquered the world throughout the
eighteenth and early nineteenth centuries. At least, it was a
form very much nearer to this than to either of the texts that
appeared in Shakespeare's life. It is instructive to imagine how
much greater the changes would have been if, with the same
amount of enthusiasm for Shakespeare and the same skill and
creative power in dramatic composition which existed in the
eighteenth century, there had been no printed books for the
public to read, while Dryden and Pope and Garrick and
Johnson had had a fairly free hand in retouching the prompt-
copies !

But having realized from Hamlet what might have happened,
let us notice what actually did happen in the full light of the
eighteenth century, in the midst of printed text and professional
critics, to two of the most famous plays of Shakespeare. I quote
from the fascinating book of Prof. George Odell, Shakespeare
from Bettertoii to Irving (Scribner's, 1920) :

In 1 68 1 Nahum Tate's King Lear was shown at Dorset
Garden. Tate's dedication to his ' esteem'd Friend, Thomas
Boteler, Esq.' says that in Shakespeare's Lear he had found
' a Heap of Jewels unstrung and unpolish'd ; yet so dazzling


in their disorder that I soon perceiv'd I had seiz'd a Treasure.'
The three most striking alterations were ( i ) an ' Expedient
to rectify what was wanting in the Regularity and Proba-
bility of the Tale ', by making Edgar and Cordelia (who
never meet in the original play) lovers from the start ; (2)
the omission of the Fool ; and (3) ' making the Tale conclude
in a Success to the innocent distrest Persons ', Lear restored
to his throne, the wicked sisters dying of poison, and Edgar
and Cordelia married. ' Yet,' says Tate, ' was I wrack'd
with no small fear for so bold a change, till I found it was
well receiv'd by my Audience.'

The first change was due to a desire for love-interest of
the stilted 'heroic' kind then popular on the stage. It
involved doing away with Cordelia's suitor, the King ot
France, and an implication that her cold answer to Lear
was due to hatred of Burgundy and love for Edgar ; it
necessitated keeping her in England and compelling her to
wander about on the heath in the fearful storm, accom-
panied by an interpolated confidant, Arante, useful for
sending on errands. Cordelia comes to beg Gloster to help
her father in the storm ; Edmund's passion is excited by her
beauty, and he sends two villains to the heath to capture her
and carry her to him. She fortunately and fortuitously stops
in front of the cave of Edgar, who drives off the rogues and has
an opportunity to behold his long- lost love at every lightning
flash. The well-known portrait of Mrs. Cibber as Cordelia
gives a picture of this interpolated scene. The omission of
the Fool removed from the play one of the most fascinating,
unearthly characters in Shakespeare ; he was not restored
to the English stage till 1838. The third alteration took
from the sufferings of Lear all their bleak, elemental tragedy,
and reduced the play to melodramatic limits. Finally,
the Edmund-Goneril-Regan episode was unpleasantly am-

Tate's mangling persisted for a century and a half;
Shakespeare's Lear was never once acted in all that time.

Again :

We have seen how great was the vogue of Tate's Kmg
Lear ; another play to live on even longer in equally
mangled form was Colley Gibber's Richard III^ played
first by the author at Drury Lane, in 1700. This version
has really never been driven from the stage ; it is probably
a more effective acting vehicle than Shakespeare's. It
simply strings together bits oi Henry VI ^ Part III, Richard
II and Richard III, interpolating even much of the best
part of the first scene from Henry IV, Part II, where North-


umberland learns of the death of Hotspur, Many lines of
this are given to King Henry in his first scene, where he is
informed of the death of his son. Henry V also contributes.
Into Richard's sohloquy on the eve of battle, fourteen lines are
interpolated from the fourth chovwsoi Henry V\ lines more
out of character it would be hard to imagine. A few other
lines are included from Henry F, about the host of ' mounted
scarecrows ', over which the greedy crows fly, ' impatient for
their lean inheritance.' This play, then, is a thing of shreds
and patches. It omits many passages of Shakespeare's
Richard III, Clarence's dream and Margaret's curse, for
instance, and it interpolates one by Gibber himself, that in
which Richard informs his wife— Lady Anne — that he is
weary of her, and means to marry her successor. The
pathetic scene of Elizabeth's farewell to her sons is also
Gibber's ; it contains but little Shakespearian material. The
aim is to make the leading character, as Hazlitt says, more
villanous and disgusting ; hence, the play opens with several
scenes from the end of Henry VI, Part III, showing the
murder of the King by Gloster. It has always been a
thriller, and, as Shakespeare's play is not among his best,
perhaps no great harm is done.

Few people know that the line so much admired by
critics, the fearfully succinct line assigned to the tyrant,

Off with his head ; so much for Buckingham !

is Gibber's ; also the frequently-quoted ' Richard's himselt
again ! ' ^

These instances show how, even in modern times, and even
among the greatest works of imaginative art, this process of
traditional reshaping to suit the needs of new generations has
gone on under the nose of critics who deny its possibility or
regard it as a wicked outrage on poetry. In antiquity the
process was both more habitual and more drastic, but of course
the evidence is far less full.

Perhaps the most curious instance in Greek literature is the
work which comes to us under the name of Callisthenes' Life of
Alexander. It is the source of all the mediaeval romances of
Alexander, and old translations of it are extant in Latin— one
made in the fourth century and one in the tenth— Syriac,
Armenian, Goptic, Ethiopic, Persian, Turkish, Malay, Siamese,

' Odell, vol. i. pp. 53 ff. and 75 ff.


and doubtless other languages. The basis in each case is a word-
for-word translation, but in every language the substance varies ;
for it was told in each country \yy Jongleurs and story-tellers who
added, omitted, and altered with a view to their audience. For
instance, Alexander is usually — in accordance with mediaeval
taste — made the child of a secret amour between his mother,
Olympias, and the exiled wizard king of . . . Of what ? Of
whatever country is most likely to please the audience. The
earliest version was written by an Egyptian Greek. Consequently
Alexander begins as a son of Nectanebos, king of Egypt. Then
he is a Persian, and so on. one version, in Ethiopic, leaves him
the son of his proper father, Philip, but makes Philip a Christian
martyr, who committed suicide on hearing from a prophet that
some day the Creator of the world would be crucified.

But it is not only the different translations that vary. Every
copy of the book differs from every other. As one editor,
Meusel, puts it : ' Like the MSS. of the Nibeluugenlied^ every
MS. represents a different recension.' ' The writers,' says Karl
Miiller, 'combined the offices of scribe and author.' That exactly
expresses it. Each scribe who earned his living by it made it as
good, as edifying, as entertaining a history as he could. The
book became a thing of tradition, and grew with the ages.^

The oldest version seems to have been written in Greek, in
Egypt, in the time of the Ptolemies. So much can be made out.
It professes to be the work of the philosopher Callisthenes, a real
person, who accompanied Alexander on his campaigns, and
whose real works have perished.^ We can also trace with some
probability an earlier stage of the same story : viz. a series of
imaginary letters, between Alexander and his friends, composed
by some sophist in Egypt not long after Alexander's death.

I will not speak of the mediaeval epics, the NibelungeiiHed^
the Arthur Legends, or the great French epics centring in the
Chanson de Roland. Each one of these subjects has its own
peculiarities and special difficulties ; but each one would illustrate
our main thesis equally well. Let me merely quote some words
of Gaston Paris to illustrate the nature of a traditional book.

^ See Appendix E. The Pseudo-Callisthenes.

"^ An interesting fragment of Callisthenes has lately been discovered, cited
by Didymus ou Demosthenes. (Teubner, I907. A papyrus.)


He is speaking of the controversy whether ' the author of the
Song of Roland' had ever seen the valley of Roncesvaux, where
the scene of his battle is placed. The great savant answers : —

The Song of Roland is not a work composed in one effort
at a given moment. It comprises in itself elements of very
different date and origin. Some go back to the immediate
impression of the event which it celebrates ; others have
been introduced in the course of centuries by professional
poets, who invented wholesale episodes calculated to increase
the interest of the poem and develop its power of heroic
and national inspiration. . . . The name of the author of the
Song of Roland \s Legion. And among those who, from the
seventh to the eleventh century, would have the right to rise
and answer any appeal addressed to that author, it would
be very rash to affirm that not one had ever passed by
Roncesvaux, at a period when so many people used that road.

How many controversies about Homer might be answered in
similar words ! ^

The most instructive example of the growth and change of
a traditional book under ancient conditions is to be found, I think,
in the Hebrew scriptures. I often wonder that the comparison
has not been more widely used by Greek scholars. The
scientific study of the Old Testament has been carried out with
remarkable candour and ability by many Semitic scholars of the
last two generations. The results of their researches are easily
accessible ; the main results may be said, in a sense, to be practi-
cally certain. You cannot, indeed, often say with certainty in
any particular place of difficulty, ' This is what happened ' ; but
you can very often say with certainty, ' This is the sort of thing
that must have happened.'

The subject is one of great interest. I fear, however, that
interest largely depends on details ; and I am compelled here to
content myself with the merest outline of the main facts about
the growth of the Pentateuch.

The central voice and the informing spirit of the Old
Testament is the Book of Deuteronomy. We all know its main
characteristics : an insistence on a rigid and highly spiritual

' Gaston Paris, Legendes du Moyen Age, p. 46 ff. See also Appendix F
on the Roland and the Vie de St. Alexis.


monotheism, and an avoidance of all remains of idolatry : a great
system of law, governing in a theocratic spirit all the details of
life, and resulting in an ideal too strict, and in some ways too
high, to have ever been carried out in practice: lastly, for the
sake of this purity of religion and morals, which was associated
with the conception of the Jews as Yahweh's peculiar people, and
the Temple at Jerusalem as the one seat of correct ritual and
doctrine, an Intolerant condemnation of all other places of worship,
however sacred, and a ferocious dread of all foreign elements
which might corrupt the orthodoxy of the chosen race.

Deuteronomy was found in the Temple by certain sacred
persons — we are not told who had put it there— in the eighteenth
year of King Joslah (B.C. 621 : 2 Kings xxli. 8 ff.). It was
accepted at once as the standard of a great religious reformation.
Joslah supported the Deuteronomists, and the reformation was
successfully carried through. Now among the other tasks which
the reformers had before them was the re-editing of the ancient
traditional books of the people. They needed reform in count-
less ways. Both of them, indeed, must have been originally
pagan and polytheistic. I say ' both ' rather than ' all ', because
In the main we can distinguish two great documents, which have
been welded by the Deuteronomists into the narrative of the
Pentateuch. one of the most obvious differences between them
Is that In one God Is called ' E:iohim '—the word translated ' God '
in our version, though It is really a plural ; In the other he is
called Yahweh, or Jehovah, the special unspeakable name of the
Hebrew God, translated In our version 'The Lord'. The
documents are called ' Jahvlst ' and ' Elohist ', or J and E

J seems to have been composed— that Is, put together out of
more ancient material— in Judah In the ninth century ; E in Israel
In the eighth. They were very similar in general contents.
Each was an almost undifferentiated tribal Logos, a sort of
history of the world and all the things in It that were worth
writing down.

A copy of J or E before the Deuteronomists altered It would
be, for Semitic historians, the most valuable book in the world.
The strange thing is that the reformers were able to carry their
project through. It was necessary for them not only to alter

IV J, E, AND P 109

their own versions at Jerusalem, but to suppress all old copies
that differed from their own. Had the kingdom of Israel still
been standing, the task would scarcely have been possible.
There must have been, one would imagine, copies of the old
books unexpurgated in the sanctuaries ^ of the Northern Kingdom.
But Israel was now in captivity, and most of the extant copies of
his old half-pagan books had doubtless gone with him. There
was little danger of their idolatrous voices being heard from
Halah and Habor and the river of Gozan and the cities of the
Medes. Yet even so there were difficulties in Judah itself
There seems to have been a regular military expedition against
the remnants of Paganism, a formal destruction of the old High
Places, and a massacre of the priests at Bethel. At last Jerusalem
stood alone as the only sanctuary, and the reformers had
undisturbed control of the Book. one is reminded of Greek
stories about the interpolation of Homer, how Solon or
Pisistratus or another bolstered his city's claim to the island of
Salamis by citing a passage in the Iliad^ which the opponents of
Athens thought spurious but were not, apparently, able to convict
by producing an authoritative text with different wording.

So far, then, we have found in the Pentateuch a document
compiled from three sources, the earliest written in the ninth
century, the latest about the year 621. But that is to leave out
of account, at any rate as regards Genesis, the greatest, or at
least the most formative and omnipresent, of all the sources.
The whole book was revised again, increased by large stretches
of narrative, and, roughly speaking, brought into its present
shape after the return from exile, between the years 440 and 400
B. C. This reviser, known to critics as P, was a member of the
priestly caste. He wrote, among other things, nearly the whole
of Leviticus. That is to say, in an average chapter of Genesis
we may read a verse written in the ninth century followed by one
written in the fifth, a gap of four hundred years. And some-
times the gap will occur in the middle of a verse. Sometimes
other sources, of unknown date, will intervene.^

' * But was there any connexion in Ancient Israel between the priestly
caste and literature ? The later Sophcr was the literate man.' D. S. M.

"^ e. g. Gen. ii. 4 is partly J and partly P. (See Driver's Commentary.) So
is xiii. II, while xiv is from an unknown source. (Abraham, Lot, and


Of course, even apart from the wholesale excision of paga-
nism from the most ancient books, the peculiar qualities of these
versions must have been much clearer when the books existed as
separate wholes. We know them only in fragments : and those
fragments have all passed under the hands both of revisers and
of religious reformers, who must both consciously and uncon-
sciously have modified the more striking discrepancies of style or
statement between their various sources. Still, the differences
are even now pretty clear : I take a few points from Canon
Driver's Introduction to Genesis}

J, or the Jahvist document, is a Logos of the most broadly
human interest. It is full of poetry and drama. It delights in
explaining the origin of human institutions — why men wear
clothes, why snakes crawl, why child-birth is painful : who
invented agriculture, pastoral life, music, metallurgy, the drink-
ing of wine : how men came to have different languages : why
Moabites, Ammonites, Canaanites, Edomites, are what they are,
the cause being generally some significant first action, or some
oracle spoken by a patriarch.

The writer is full of interest in the sacred sites of Palestine,
the altars, pillars, trees, and high places, and the reasons why each
one of them is sacred. He has no idea of condemning any of
them. They had not yet come into competition with the Temple
at Jerusalem. He calls God by the name ' Yahweh ' from the
beginning, and supposes that the true religion naturally belonged
to the primaeval patriarchs. In this, of course, the other
prophetic book, E, differs from him. In E the ancestors of Israel
'beyond the river' were idolaters (Joshua xxiv. 2, 14, 15), and
the name Yahweh is not revealed to man till Exod. iii. 14.
Again the Yahweh of J is frankly and naively anthropomorphic.
He not only feels human emotions, but he performs sensible acts ;
he 7notilds man out of earth, he plants a tree, he shuts iLp Noah
in the ark, he smells burnt meat, ivrestles with Jacob, and takes
<2^the wheels of the Egyptian chariots.

Now let us contrast with this the work of the latest writer of
all- P takes no interest in the origin of human institutions, only
in ritual : no interest in sacred sites, only in the Temple at
Jerusalem ; his God is, practically speaking, never anthropo-

^ Differences of J, E, P.


morphic. His history of the world has been mapped out in
a scheme of genealogies and dates, and especially of covenants
between Yahweh and his chosen people, Israel. There are three
stages of history marked by a gradually diminishing length of
human life, and by the revelation of God under three distinct
names: Elohivi^ El Shaddat—tho. obscure name revealed to
Abraham in Gen. xvii — and finally Yalnvch. The Patriarchs
raise no altars, perform no sacrifices. ' No act of worship seems
to be thought of till the appropriate place has been constructed
and the right persons appointed for its performance. The first
sacrifice recorded is that of Aaron and his sons in Lev. vii.'
The promises of God are strictly limited to Israel itself, and the
abiding presence of Yahweh with his people is dependent on the
directions for the exact construction of the tabernacle (Exod. xxix).
It is all sacerdotal through and through.

Thus there is a period of four hundred years between the
earliest and latest of the large integral documents constituting the
Book of Genesis. But the period of growth was much longer
than that. In the case of Genesis the argument does not come
out quite so clearly ; we can take our illustration more easily
from the Books of Samuel. As the earliest source in Samuel we
have the so-called ' Court narrative ' of David, attributed to the
tenth century B. c. At the other end there are considerable
slices of narrative which are found in the ordinary Hebrew text,
but not in the Septuagint translation, which was made about the
year 200 B. C. Of this fact two explanations are possible.
Either, and this seems the simpler hypothesis, the narratives in
question were not in the Hebrew text from which the Septuagint
was translated ; or else they were in the Hebrew text, and were
deliberately left out by the translators. on either hypothesis it
is clear that the authorized text was not definitely established.
A traditional book of which large parts can be left out or put
in at discretion is still in the stage of growth. The Book of
Samuel, then, was in process of growth for considerably more
than seven hundred years. And that is without reckoning the
small corruptions or verbal changes which seem to have occurred
much later. In some books, for instance, there are changes
directed against the claims of Christianity.

But, returning to the Pentateuch : when J or E was first com-


posed, it was not composed out of nothing. Each of them was
really put together in the same way as the whole composite
Pentateuch of the Priest, by taking an older existing book,
copying it out, adding, omitting, and sometimes altering. Many
of these earlier sources are quoted by name, as the Iliad quotes
the older Argonautica. There is the Book of Jasher. From it
come the standing still of the Sun and Moon (Joshua x. 12), David's
lament over Saul and Jonathan (2 Sam. i. 17), and perhaps some
verses spoken by Solomon when the Ark was brought to the
Temple (i Kings viii. 12). The song in Num. xxi. 14, again, ' is
it not written in the Book of the Wars of the Lord ? ' In these
cases the name of the older book is explicitly given. Much more
often it is omitted. Sometimes a quotation betrays itself by
being in verse, like the Sword-Song of Lamech, and the oracles
spoken over their respective children by Noah, Isaac, and Jacob.
But an insertion from a prose work would be hard to detect :
and even the verse was apt to be worked back into prose (see
commentators on i Kings viii. 12).

Among other sources would be the mere tribal traditions, such
as we have in the Book of Judges. Sometimes they are full and
clear, and seem to depend on written documents. Sometimes a
tradition consists merely of a name and a burial-place. ' After
him Elon the Zebulonite judged Israel : and he judged Israel ten
years. And Elon the Zebulonite died and was buried in Aijalon
in the land of Zebulon.' Aijalon Is probably the same word as
Elon. The chronology will not work. And the story seems
merely to mean that there was at Elon or Aijalon an unknown
grave which was regarded with reverence.

There was more detailed tradition at the various ancient sanctu-
aries, Hebron, Bethel, Gilgal, and the like, a source particularly
prominent In J and E, but discountenanced by the priestly editors.
There were fragments of history or learning adopted by hearsay
or otherwise from more advanced nations. This is a regular
process in primitive races, and is admirably illustrated in Professor
Margohouth's short Life of Mohammed} That prophet was
constantly picking up scraps of Christian and Jewish lore, and
incorporating them, with inevitable mistakes, in his Koran. In

^ Especially pp. 106 ff. Mohammed got Goliath's name as Galut ; the
name of Saul, David's other enemy, he had forgotten, so he made him Talut.


the Hebrew scriptures there seems to be an especially large debt
to Babylon, such as the stories of the Creation and the Flood ;
certain fragments about Abraham, who perhaps had the honour
of meeting the great law-giver Hammurabi or Amraphel ; and
many elements in the Hebrew laws themselves.

Now I realize that all this description must remain rather
ineffective when unaccompanied by detailed illustrations. But
the detailed illustrations would clearly take us quite beyond the
limits of our present subject. And it is, of course, not any part
of my business to prove the truth of the analysis of the Pentateuch.
I merely take the results reached by a consensus of the best
Semitic scholars, in order to show the sort of process which was
normal in the formation of one type of ancient Traditional Book,
and the qualities which naturally resulted therefrom. To produce
such a composite work as one of these books in its later stages
without inconsistencies and awkward joints would be difficult, as
we said above, even for a modern editor with all his mechanical
accessories and his opportunities of revision. To the ancient
editor the difficulties were insuperable. And, as a matter of fact,
most ancient compilations betray themselves. I will not dwell on
the various doublets and inconsistencies which careful reading
discovers in the Pentateuch ; the two divergent accounts of the
Creation, and of the Flood, with traces of a third in which there
was no Flood ; the inaccuracies of the chronology so laboriously
inserted by the Priestly writer — ancient numbers, when at all
complicated, seldom come out quite right ; much less on the
many small confusions, like that of the two wives of Esau who
are mentioned three times, each time with different names ; nor
yet on such curious formal points as the case of the Twelve Tribes
of Israel, which are mentioned again and again as twelve, yet
always add up as thirteen. Such weaknesses as these are normal
things among primitive historians. If they serv^e to illustrate the
writer's lack of critical control over his complex material, they
also are often evidence of his good faith.^

^ The Jahvist, very simple and anthropomorphic, narrated how Yahweh
' moulded ' a clay man and breathed life into him, and planted a garden and
put the man to keep it. Then as the man was lonely, Yahweh made all
sorts of beasts as companions for him, but none was quite satisfactory till he
made a woman out of one of the man's ribs, and then the man was content.
The Priestly Document, more, advanced and scientific, gives the other story

2760 I


I hope that by now I have succeeded in illustrating two points
about these ancient authorless books ; first, the immense periods
of time during which they remain fluid and growing ; and second,
the difficulties which they have in combining their multiplex
sources. The object which I have in view is, of course, Homer.
And I wish now to notice briefly some two or three more of the
phenomena characteristic of this kind of writing, in order that
we may know their faces again when they meet us in the Iliad.

First, there are the various disturbing influences that are apt
to affect the primitive historian. I will not lay stress on mytho-
logy and folk-lore, such as we find in the story of Samson, the
Sun -man, ^ or in the Babylonish part of the Creation : nor on
what I may call Romance, or the story-teller's instinct, such as
we find in the narratives of David and of Joseph. These factors
are enormously powerful in Greek legend ; Semitic scholars
differ as to their influence in Hebrew. I will not lay stress on
the tribal spirit, with its ramifications of patriotic devotion, party

of the six days of creation, with a gradual process of development, as it were,
from the lowest forms of life up to the highest, culminating eventually in man.
We cannot be sure about the account of the Israelitish Elohist ; for the
Reviser, while combining the other two, omitted it altogether. Similarly in
the Flood, the Jahvist tells how Noah took seven of each clean animal and
two of each unclean ; how the flood lasted some ninety-four days ; and how
Noah came out at the end of the time and offered sacrifice. The Priest tells
how Noah took two of every animal, with no distinction of clean or unclean ;
that distinction, he apparently argues, cannot have been known to Noah,
because it was first revealed to Moses in Lev. xi and Deut. xiv. He tells how
the flood lasted a year and ten days, and how at the end God made a covenant
with Noah and set his bow in the heavens for a sign thereof. There seems
also to be a trace of a version in which the first Man was not called Adam,
but Enosh — the other Hebrew word for man. As to the chronology so care-
fully introduced by the Priestly writer, Canon Driver shows that Judah
'marries, has three children, and after the third of them has grown up
becomes a father again, and through the child thus born becomes a grand-
father, all in the space of 22 years'. (Thirty-five would seem to be about
the minimum possible.) The age of Ishmael at the time of his casting out
varies between babyhood and adolescence. So does Benjamin's. The wives
of Esau are given in Gen. xxvi as Judith, daughter of Beeri, and Bashemath
d. Elon : but in Gen. xxxvi they are Adah d. Elon and Bashemath d. Ishmael.
And in chapter xxviii the daughter of Ishmael is Mahalath. one can see
what sort of process this implies. The compiler of the two, or the three,
narratives did not keep constantly looking forward and backward. He had
no index to show him all the places where he had mentioned Esau's wives,
and help him to reduce them to order. In the case of the more important
matters his memory no doubt served him, and he arranged his story consis-
tently. But in smaller things, which were not of real gravity to him, he
copied his authorities faithfully without noticing the occasional contradictions.
• lir^V' from K'lOK' 'Sun '.


feeling-, and odiiuu iheologicinn^ forces at times responsible for
the wildest misreadings and misrenderings of history. We must
remember that as a rule an ancient writer only recorded what he
wished to have remembered : that his book was only read within
his own tribe or circle, and that his only business with his tribe's
enemies was to injure them. He thought tribally. He used his
book as he would use his sword. But consider, as one significant
point, the helplessness of language which generally dogs these
early writers as soon as they have anything complicated to express.
The writer of Gen. x. 15, for instance, wishing to express the
relation of the Canaanites of the interior to the Phoenician city of
Sidon, can only say : ' And Canaan begat Zidon his first-born.'
The relation of the Canaanites to the Hittites, a great foreign
nation which seems to have had some settlers in Canaan, was
certainly different. But it is expressed in the same way : ' Canaan
begat Heth.' The tribe, the alien city, the foreign nation, are all
treated as individuals, and their complicated relations reduced to
that of father and son.^ Similarly Bethuel is mentioned as a
person, the father of Rebekah, but his brothers Huz and Buz are
tribes. Machir in Gen. 1. 23 is a person : in Num. xxxii. 40 he
is a clan : in Num. xxvi. 29 he ' begets ' Gilead, which is a district.
That district again ' begets ' the judge Jephthah — perhaps rather
a special case, since Jephthah had no legitimate father.

The disturbing influences hitherto considered are all, In the

main, unconscious. Let us consider for a few moments two

conscious influences. Then we can make an end of these

Semitic analogies and return to Greece. In the first place, is

there in such a book as Genesis, for example, any conscious

archaism } The answer is clear. The latest of all the writers

of the Pentateuch, P, is the one who is most particular to give

an archaic and primaeval colour to his narrative. He has used

his historical imagination, and constructed a remarkable picture

of the age of the patriarchs, quite unlike his own age or even

that of his immediate authorities. According to him, the

Patriarchs knew not the name of Yahweh, knew no altars, no

^ The statement in x. 6, ' Ham begat Canaan ', is different. It is definitely
untrue, and comes from tribal animosity. It suited the Israelites' self-respect
to think as ill as possible of their not very distant kinsmen, the Canaanites.
Consequently these undoubtedly Semitic tribes are assigned to Ham, the

I 2


sacrifices, no difference between clean and unclean meats. All
these things were specially revealed to them at later and definitely
mentioned periods. The earlier writers, J and E, are much less
particular. Their writing was centuries older, but the picture
which they draw is actually more modern. They allow Abram
to come to ' Bethel ', or pursue his enemies to ' Dan ', without
being troubled by the reflection that those names were only
the later representatives of ' Luz ' and ' Laish '. The Jahvist
tells us that in Seth's time ' men began to call upon the name
of Yahweh', without thinking it necessary to revise his earlier
narrative in which both the name and the person of Yahweh
seem to be known to all. Probably, if we only knew it, they
also archaized after their fashion, but, if they did, it was nothing
to the archaizing of the Priest. It so happens that the Hebrew
priestly writers were not interested in such things as the
comparative antiquity of bronze and iron or the date of the
Dorian migration. But, if they had been, you may be sure
that they would never have allowed a mention of iron nor a
hint of the existence of Dorians to defile their pages. These
things are of importance for Homer.

The practice of archaism is closely related to something far
deeper and more wide-reaching, the practice of expurgation.
In the case of these ancient and traditional books, which carry
on the Logos of one age to grow into the Logos of the next,
there must always emerge points of belief or feeling or conduct
where the new age differs from the old. In advanced states
of society, where the books exist in large numbers and the text
cannot be tampered with, the usual resort is allegory. All that
is objectionable is interpreted as meaning something else. But
while the books are still growing, two courses are open to each
new set of revisers. The simplest is tacitly to alter the document
and cut out from the venerable book all that seems unworthy
of it. This is expurgation. The other, more complex and
more dependent on an advanced historical sense, is to recognize
the difference in manners, and to try even in the new writings
to maintain the colour of the older age. That is archaism. one
may say that on the whole archaism is the normal practice, in
style, in vocabulary, and in the selection of facts to relate. But
when the writer is brought face to face with something which


he honestly hates or disapproves, then his archaism breaks down
and he resorts to expurgation.

Now the whole of the Pentateuch is permeated by a conscious
didactic purpose, and therefore by the spirit of expurgation.
For one of the processes which have formed the Pentateuch is
the gradual conversion of the sagas of primitive Semitic pagans
into the great book of Jewish monotheism. At what date the
early sources ceased to be pagan is open to doubt ; but that
they were once pagan is practically certain ; and probably the
work of the Deuteronomists and the Priests consisted almost as
largely in their unseen excisions of objectionable matter as in
the composition of their great codes, Deuteronomy and Leviticus,
and the innumerable small additions by which we now trace
them. Of course, as a rule, we have no means of knowing what
expurgations or omissions have been made. The thing is cut
out, and there is an end to it. But sometimes the excision has
not been complete, or has in some way left traces. Let us take
some instances.

There is the curious set of cases in which the word Bosheth^
' Shame ' or ' Shameful Thing ', has taken the place, or distorted
the form, of some genuine but objectionable word. For instance,
the title Melekh^ King, was applied to Yahweh as to other
deities : and at one time in the seventh century human sacrifices
were offered to him under that name. This was an abomination
to the purer Jewish feeling. Wherever the word Melekh
occurred in descriptions of these rites, the practice in the
Synagogue was to avoid pronouncing it and say instead Bosheth.
To indicate this, though the consonants of MLKH were not
altered in the text, the vowels of Bosheth were written under
them. Hence arose an imaginary word ' Molekh ' — afterwards
corrupted to ' Moloch ' — which was then taken for the name
of some unknown god of the Gentiles,

Again, the word ' Ba'al ' : this word, meaning Lord, or Master,
was originally a perfectly innocent title, applied to Yahweh as
well as to the gods of Canaan, Consequently many Hebrew names
in early times were formed from Ba'al. But to a later age they
sounded idolatrous, and they have nearly all been altered.
Saul's son Ishba'al (' Man of the Lord ') is turned into Ishbosheth,
* man of shame.' Jonathan's son Meriba'al becomes Mephibosheth.


In the case of Jerubba'al or Gideon a different line was taken.
The name must really have meant ' Ba'al founds or strengthens ' ;
but it is carefully interpreted as a sort of calembour or play on
the sound of the words, so as to mean ' Let Ba'al plead '. This
explanation then gives rise to one of the usual stories of the
confounding of the false God, Gideon defies Ba'al, and Ba'al
cannot plead, but remains dumb^ (Judges vi. 32).

To take a different kind of expurgation, there seems to be
some omission in the story of Cain s sacrifice (Gen, iv. 5). No
reason is given for its rejection. Probably the point of the
story lay in the ritual which Cain followed. There must have
been — so at least many authorities believe— some description of
the two rituals, Cain performed his sacrifice in some way that
was considered unholy or savouring of the gentiles. The older
story mentioned Cain's ritual in order to condemn it, the later
editors declined to speak of it at all. There is almost certainly
a great omission just before the story of the Flood, in the passage
(Gen, vi, i ff,) which tells how ' the sons of God saw the daughters
of men that they were fair, and took them wives of all that they
chose '. The next two verses are confused and unintelligible,
and the subject is promptly changed.

These instances, few as they are, will perhaps suffice to
establish the mere fact that expurgations have occurred. They
may also incidentally show how vitally the study of the expur-
gations in an ancient book helps towards the understanding of
its whole spirit. The expurgations and the interpolations ; all
that a man rejects from his traditional teaching and all that he
puts in its place ; a knowledge of those two together will surely
contain the main secrets of all that is most alive in the man's
own character. And the same is true of an age. The interpola-
tions and expurgations, if we followed the subject up, would

^ Exactly the same process has given rise to the mysterious 'Abomination
of Desolation' set up by Antiochus Epiphanes, in the well-known passages
of Daniel (ix. 27, xii. 11). The word for abomination, Heb, f'^ptJ', is used
exactly like ri'^Q to supply the place of the unmentionable name Ba'al,
What Antiochus really 'set up' was Baal Shamdwi, the Lord of Heaven ;
an altar, that is, to Zeus Ouranios. In place of Ba'al we say Shiqquf,
abomination : and in place of Sliamdim, heaven, which is here equally
unclean, inasmuch as it is part of the name of a heathen god, we put the
almost identical word Shomcin, from a word meaning to destroy or lay waste.
Baal Shamdim becomes shigqtig Shomem ; the Lord of Heaven becomes
the ' pollution ' of ' desolation '.


teach us much about the age of the Deuteronomists and the later
age of the Priests.^ And I wish now to apply this method, at
least in one of its aspects, to Homer. I shall not attempt to face
the question of interpolation. It is too complicated a subject.
But the traces of expurgation in Homer have been very little
studied, and seem capable of yielding some interesting results.
We will consider them in the next lecture.

* I have not attempted to analyse the expurgations of the Deuteronomists
(D), or to find out what sort of thing they most objected to. The above cases
are nearly all expurgations of idolatry or paganism, and that is evidently and
by far the greatest preoccupation of the revisers. There are also some
expurgations of immorality. As regards cruelty, they were much less
particular than Homer, provided that the cruelty was directed against
suitable objects. They approve of the ferocity of Samuel (i Sam. xv) and
the Herevi generally: i.e. the extermination of all living things, beast and
human alike, in heathen countries. (See Ban in Enc, Bibl., and compare
the Scandinavian custom of dedicating hostile armies to Othin by throwing a
spear over them.) They allow even such a sympathetic hero as Gideon to
' thresh ' the elders of Succoth ' with thorns of the wilderness ', without
comment ; the same may be said of David and others. In this particular
one may note that the very late book, Chronicles, expurgates its sources :
e. g. 2 Sam. viii. 2 : ' And he smote Moab (and measured them with the line,
making them to lie down on the ground : and he measured two lines to put
to death, and one full line to keep alive). And the Moabites became
servants to David and brought gifts.' This is repeated in i Chron. xviii. 2,
except that the Chronicler oj/iits the 7vords in h-ackeis.

Similarly the account of the taking of Rabbah, where David 'brought forth
the people that were therein and put them under saws and under harrows of
iron and under axes of iron, and he made them pass through the brick-kiln'
(2 Sam. xii. 31), is omitted altogether in Chronicles. (Driver and others,
however, think that torture is not intended here, but only slavery.) on the
other hand, when religious motives come in, the latest writers can be very
savage. See l Kings xiii. 2 and 2 Kings xxiii. 20, where Josiah's wholesale
sacrifice of the priests of Baal is described with exultation. (The end of
chap, xxiii is ascribed to a very late source, but the tone is really much the
same in the rest of the chapter, which is by J.)

Not perhaps actual expurgation, but something very similar, seems to
have been at work in those cases where we find that certain very old parts
of our extant composite narrative were not included in the Deuteronomic
revision. For instance, in the Book of Judges, D is not responsible for chap,
ix (Abimelech : a story possessing historical interest, but no religious value),
nor for xvi-xxi. He ended Samson at xv. 20, after the jaw-bone victory, at
the words: 'And he judged Israel in the days of the Philistines twenty
years.' The part omitted consisted of Dalilah and the end of Samson ; the
stories of Micah, the Danites, the sin of the Benjamites, &c. — all somewhat
unedifying. Similarly in Samuel, D has no hand in i Sam. xxviii. 3 to end
(Witch of Endor), which breaks the continuity of his narrative ; nor in 2 Sam.
ix-xx, which contains all the intimate Court stories, Bath-sheba, Rabbah,
Taniar, &c. D ended his narrative of David with the fdsume in 2 Sam. viii.
iSfif., 'And David reigned over all Israel, &c.' These stories are not later
inventions. They come from the oldest material, and must have lain before
D, who deliberately rejected them. They were, however, preserved and
eventually inserted into the composite narrative which we now possess in an
age which was more open than that of D to historical, antiquarian, or merely
human interests.


' As for these passages and all others of the sort, we will beg Homer and
the other poets not to be angry if we draw our pen through them.' — Plato,
Rep. iii. 387 b.

In considering the subject of Homeric expurgations I will take
my instances chiefly from the Iliad., because I believe the Iliad
to be, in the ancient phrase, ' more Homeric ' than the Odyssey.,
that is, both to have more of the definite Homeric spirit, and
to have undergone a more thorough process of revision and

But before studying what things the poets reject we should
be clear what sort of world they wish to represent. And we
can. They are with conscious art depicting a past age, but
an idealized past; a past, as Grote says, which never was a
present. It is the normal method of high romance ; the method
of Scott in depicting the ages of chivalry ; of Tennyson or
Morris in writing about the knights of King Arthur. There is
a conscious use of tradition and of what may almost be called
archaeology; a genuine attempt to be true to the manners of
a simpler society. Yet, with all its simplicity, it must be an
age of chivalry, of heroism, Nausicaa, the king's daughter,
may go out with the dirty clothes in a mule -cart to wash them
by the sea shore : so simple was life in those days ! But her
beauty and stature, and her language, and the language of others
in speaking to her, have a majesty and a graciousness which,
I suspect, the poet did not find or expect to find among con-
temporary young women. Those were the days when a hero
was really a hero and a princess a princess !

The picture of the past is constructed with wonderful con-
sistency. Where there is a slip or a doubtful point we find that


it has nearly always been observed and discussed by the
Alexandrian scholars. Greek taste, at the time when our text
was finally fixed, was extremely careful about anachronisms.
Thus it is likely that a text of the time of Pisistratus or earlier,
if we could find one, would have many more anachronisms than
our present text, which was punctiliously watched and re-edited
by generations of ancient scholars. But from quite early times
the Homeric poets must have been imaginatively interested
in the Heroic Age, and have tried to represent it vividly.
Though all the higher Greek poetry is about the same ideal
past, no other poems treat it with quite such scrupulous tender-
ness as the Iliad and Odyssey. No other poems have been
so fully studied and commented and re-edited.

Attempts have been made, and are still far too common, to
suggest that the Homeric picture of the Heroic Age is actually
the work of a naive and primitive poet who lived in it, and simply
described his own surroundings. But, apart from other difficulties
in this theory, it is definitely disproved by a curious difference
between the narrative itself and the poet's comments upon it.
In the narrative we have the complete heroic world. The
heroes of the Iliad consume only heroic food, consisting chiefly
of ' unspeakable flesh and sweet strong wine '. They eat
enormous slices of roast ox or sheep or boar, and that three
times a day. They do not condescend to boiled meat, much
less to fish, fowl or vegetable, milk or cheese. When the
companions of Odysseus are twice reduced to fish and wild-
fowl, it is a case of extreme need (/t 331 ff , 5 368). In the
similes, however, there is quite a lot about fishing, alike with rod
and net and spear ; about diving for oysters and the advantages
of a sea rich in fish. There are similes taken from the catching
of larks and pigeons, and perhaps from hawking. There is
much about milk and cheese, and one mention of boiled pork.
That is the poet's own work-a-day world, where people had
at most two meals a day and meat was a scarcity, not the world
of the great Zeus-born heroes.

Similarly in metaphors we hear of the trumpet (^ 219, $ 3^8),
which, as the ancient scholars had observed, is unknown in the
narrative. So riding is mentioned twice in similes (O 679, € 371),
but in the body of the narrative inn^e? are charioteers. (The


riding in K is, of course, exceptional.) Crowns and garlands
were a commonplace of Greek social life in classical times, and
are referred to in metaphors freely {N 736 and Schol., E 739,
A 36, 0^153, S 485, K 195); but the ancients have noted that
they are never used by the Heroes. It is very significant, as
will be seen further on, that, though the Heroes habitually fight
with bronze, when the poet wants a hard metal for a metaphor
he mostly speaks of iron.^

Thus it is certain that the poet is representing with conscious
art a past age of heroism and chivalry. Naturally, therefore,
he excludes from it not only what is low and modern, but also
a good deal that was really ancient, but indicated the squalor
and brutality of the past, rather than its chivalry.

But first a word as to method. If only we had still two or
three versions of the I/md belonging to different times, such as
we have of the Roland, the Alexis, or the Nibelungenlied, our
task would be plain. If we had any remnant of a pre-Attic
text, or even if one of our fragmentary pre-Aristarchean MSS.

^ See Drerup I, chap, iv ; and cf. the following passages: — Fish: n 747
TroXXoi/y av Kopeanev avtjp o8( Trjdia 8i.(pS>v, see note below. 5 368 alel yap n^pi
viifTov a\a>nevoi l^^dvaafTKOv yvapnTTOis dyKicrTpoLaiv, erfipe 8e yucrrtpa Xi/jlos. n 406
0)5 ore Tis (f)a>s . , . iepov i)(dvv tK ttovtoio 6vpa(€ XiVw Kol fjnoni )(a\K<3 (e'X/cei).
/x 251 0)9 S or eVi TrpojSdXo) aXievs 7rtpip.TjKe'i pd^8a> l)(dijai tois oXiyoiai SoXoi/ Kara
fiSara ^dWcov . . , Nets E 487 M"? ""Wf) ws dyj/icn XiVoi' aXovre Tramypov , . . J2 80
Iris plunges into the sea p.oXv^8aivr] iKeXrj ts ^va-aov opovaev fj Te . , , epxerai
wiirjaTTJa-ip in IxOvai Krjpa (ftepovaa, x 3^4 ^^^^ dead suitors lying in a heap ws
t' Ixdvas ovs 6' dXirjes . . . fit/cruo) e^epvaav noXvaiTco. Spearing : k I24 Ixdvs S' ws
TTfipovTfs drepnea baira (f)€povTo, Good fish as an economic asset is among the
blessings of the good king, t 113 tjktt; 8' i'p,7rf8a fi^Xa, ddXaa-aa 8t napexu Ix^vs.
Perhaps also I 360 'EXXria-novTov in IxOvdevra.

Birds caught in a net : x 4^8 00s 6' St dv fj KixXai ravvalnTepot, fje n-eXeiai tpKei
ivinXrj^wai, to 6" ((TT^Kt] eVt ddpvca . . . Hawking : X 302 oi 8' as ulyvnioi , . . in
opuldeaai dopcuai . . . X'^'-P^^^'- ^^ t' duepes nyprj.

Milk and cheese : B 471 'HiJTe /jlvuimp d8ivd(ov edvea TToXXfi, at re KaTa cnnQfihv
noip.VT]iov . . . ore re yXdyos dyyea 8eii€i. Cf. II 643. A 433 as t o'Us . . , fivpiai
(arijKaaiv dixeXyopevai ydXa XevKov. E 902 w? S' or' dnos ydXa XiVKov ineLyopevos
<Tvv€nr]^ev . . . (rennet curdling milk). Boiled pork : <I> 362 ws 8e Xe/Sz/y feZ 'iv8ov
ineiyupfvos nvp\ ttoXXw, KvitTrjv peX86pfvos dnaXoTp€(peos (ridXoio. on which
Schol. T observes ' The poet himself knows the use of boiled meat, but does
not represent the heroes using it'. So on the oyster-diver in n 747 the
Schol. T says ' This does not agree with the general life of the heroic age.
Not even the luxurious Phaeacians or the Suitors are represented as using
such things (as shellfish), . , . He represents them using neither fish nor
birds, tho' Odysseus' companions do try them under stress of need (8 368).
In general he avoids such usage because of its meanness (t6 piKponpcnis) and
makes them use roast flesh, so that he can say of Achilles * Automedon held
the flesh and Achilles cut ' ; imagine what the effect would be of making the
Son of Thetis clean a fish or boil soup ! '


were complete! As it is we are forced for the most part to
search in our present text for small things that look suspicious
and lead us to probabilities, not facts. Yet there is some
positive and definite evidence also. Our knowledge of the text
of the Iliad does, after all, just reach back to the time when
it was not yet absolutely fixed, when it was still possible for
a reader who greatly disliked something in his MS. of the Iliad
to ' obehze ' it or cut it out as unworthy of Homer. If we study
the passages deleted or condemned by the earliest critics known
to us, it is impossible not to see that, though the text was by
then almost fixed, the process of expurgation was still active.
Passage after passage is condemned or criticized as a7r/oe7rey,
' unseemly '. The only cases that are perfectly demonstrable,
of course, are those in which two versions are preserv^ed ; where
either our text contains lines which some other authority con-
demns, or some other authority preserves lines which have been
dropped out of our text.

For instance, there are four lines in the Phoenix story (I, 458-
61) describing that hero's wish to murder his father, of which
Plutarch tells us that Aristarchus cut them out ' in fear ', because
of their bad morals. There is a line just above (453) "where
Phoenix, speaking of his mother's infamous suggestion, says
' Her I obeyed aiid did' (rfj 7n66/xrji' koL ep^ia) : certain ancient
critics read the line, 'Her I disobeyed and did not' (rfj ov
nidofxrju ovS' ep^a). In the Story of Ares and Aphrodite in the
Odyssey, the scholia tell us of ten lines (6 333-43) which were
absent from some copies ' because of the unseemly suggestion ',
while we know that some ancient critics rejected the whole
episode. There is an interesting deletion on quite other grounds
in H 195-9. Aias is going forth to single combat with Hector,
and bids his companions to pray to Zeus ' silently within your-
selves , so that the Trojans at any rate may not hear' — and
so use counter-prayers. A little mean, that, especially for Aias ; a
little like mere witchcraft, such as the Norse heroes so vehemently
denounce and repudiate. It is followed by a healing line: 'Or
pray openly if you will ; I am not afraid.' But the best critics,
Zenodotus, Aristophanes, Aristarchus, all unite in rejecting the
whole passage. It was safest away.

It is instructive to look through the whole list of passages


rejected from the Iliad by the great critics. Our testimony is
miserably deficient, especially for Zenodotus, who matters most
of the critics because he was the earliest and the most drastic.
But there is enough to show that they rejected a great quantity
of lines on pure grounds of expurgation : ^ passages where the
Gods misbehave more grossly than usual, passages which
attribute to the heroes coarse language or unworthy motives,
above all, passages where the suspicious eye of a moralist saw
traces of the work of those infamous persons who misinterpreted
the relations of Achilles and Patroclus.'-^

* Whether they had or had not MS. authority for their deletions does not
affect the argument. If they had, then the pre-Alexandrian confusion was
even greater than our existing evidence proves. And we should then have to
assume (i) that the attempt to expurgate these passages had been made
before Zenodotus (which is pretty certain) ; (2) that in the turmoil of texts
the critics were largely decided by expurgatory motives. I append a rough
list of some typical ' unseemly ' passages which were condemned by ancient
critics :

Unseemliness in the Gods : A 396-406 (gods frightened by Briareos) ;
B 111-18 (Agamemnon's blame of Zeus); B 157-68 (Hera to Athena);
r 396-418 and 423-6 (Helen and Aphrodite) ; 35-40 (violent speech of Zeus
followed by apology) ; 385-7 (Athena puts on her father's tunic) ; 420-4
(Iris repeats message and adds insults of her own); A 78-83 (Gods angry
with Zeus; perhaps other reasons for excision); S 317-27 (the Leporello-
catalogue of Zeus's amours) ; O 5 a (decidedly dTrpenes, and dropped from our
texts); O 18-31 (threats of Zeus to Hera); O 212-17 (threats of Poseidon);
n 432-58 and 666-83 (discussion of gods about Sarpedon and its sequel;
probably some religious objection at work to reinforce critical reasons) ;
2 356-67 (Zeus to Hera); * 471, 475-7 (Artemis reviling Apollo); O 20 f.
(Apollo and the dead Hector; religious expurgation); Q, 23-30 and 71-3
(proposal that Hermes should steal the corpse, and statement that he could
not); 423 (cf. 20 f.).

Unseemliness in Heroes : A 225-33 (' Drunkard with the eyes of a dog,'
&c.) ; B 193-7 (treachery imputed to Agamemnon); 164-6 (abusive
language?); 284 ('reared you in his own house, bastard as you were') ;
I 458-61 (Phoenix and his father) ; A 794 f. (suggestion of cowardice in
Achilles) ; n 89-90 (mean motive in Achilles) ; T 77 (Agamemnon not rising :
ajrpfnes); Y 180-6, 195-8, 205-9,251-5 (all in the discourteous scene between
Achilles and Aeneas); ^ 804-6, 810, 824 f, (the barbaric gladiatorial
combat); fi 556-7 and 594 f. (unworthy motives). one may add 189
(giving horses wine ; barbaric) ; and the abusive language of Thersites in
B 227 f. and 231-4.

^ The primitive character of these practices is proved by the archaic
inscriptions of Thera and convincingly explained by Eethe in Rhein. Mus.
N.F. Ixii. 438-75. By this rite the full warrior adopted the squire and
imparted his ina7ta to him. Cf. Preuss in Archiv fur Reltg. 1910. That the
'silence of Homer' is intentional is proved by E 266 and Y 231 f. (on Gany-
medes). The clearest text about expurgation is Schol. T on n 97 : TravTfXm
fK^XrjTfov Tous S' (TTLXovs, says Aristarchus with unusual emphasis ; 'Zenodotus
was right in suspecting that they were inserted vtto tcop dp(T€vt.Kovs tparas thai
\€y6vT<ov nap' Ou.tip(o koX vnofoovvTav nai^iKa elvai Ap^iWea UnrpoKKov.'


This is the only part of the subject which is difficult to discuss.
It is too important to omit altogether. The evidence is clear
that there existed in early times, among both Aryans and
Semites, and notably among the Dorians, who are generally
reckoned among the more primitive races of Greece, certain
forms of sexual irregularity which were in the end totally con-
demned by the Jewish and the Athenian law, but were tolerated
in various parts of the Aegean and even in such well-conducted
communities as Crete. Sodom and Gomorrah, according to the
tradition, were consumed by fire from heaven. The tribe of
Benjamin was almost blotted out. Laius, king of Thebes, was
involved in a fearful curse, together with his whole race. But
early Greek traditions testify both to the existence and the
toleration of these practices. Now Homer has swept this whole
business, root and branch, out of his conception of life. Exactly
the same spirit is seen at work when we compare the rude
ithyphallic Hermae of ancient Greek cults with the idealized
messenger of the Gods in the Odyssey. But that is merely one
instance : for this kind of expurgation really pervades the whole
of our Homer.

Closely akin to this is the spirit in which our present text of
the Odyssey treats the marriage of Alcinoiis and Arete, the king
and queen of the Phaeacians. ' Her name was Arete, and she
was born of the self-same parents that begat king Alcinoiis '
{r\ 54 ff.). Exactly ; Hesiod too, the scholia tell us, made the
royal pair brother and sister. There are abundant instances of
that sort of marriage in the houses of the ancient divine kings.
The royal blood was too superhuman to make it desirable for
the king to wed any one lower than his own sister. Hera her-
self was sister and spouse of Zeus. The Pharaohs and the
Ptolemies after them made a practice of having their sisters for
queens. In the first of Griffith's Stories of the Priests of
Memphis the doctrine that the only fit bride for Nefrekepta is
his sister is explained and insisted upon. Hesione was sister
and wife to Prometheus, though Aeschylus, gently expurgating,

Schol. A agrees. They ought also, while they were about it, to have
expurgated the word Trcp in O 130. The passage of Aesch. Myr7nidones,
fr. 135, Nauck, has been grossly misinterpreted by Athenaeus. Blass
has pointed out that the words are addressed to Ares ; his ' kisses ' are the


makes her only half-sister by the father {Prom. 559). Such a
queen was doubly august. Arete, we are told, ' was honoured
as no mortal woman is honoured in these days, of all who hold
their houses under a husband's rule.' She was hailed like a god
when she went abroad (j; 66 ff.). This is the genuine language
of the Saga, and we know how to understand it. But in classical
Greece there had arisen a spirit to which such a union was
' unholy ', incestitm. And as we read on in the Odyssey we find
a genealogy inserted which in somewhat confused language
explains that when the Saga said ' parents ' (to/ctjg)!') it only
meant 'ancestors', and when it said that Alcinous' brother,
Rhexenor, died ' childless ' {ccKovpoi') It only meant ' without
male child ' ! Arete was really the daughter of the said brother.
It was only a marriage between uncle and niece.

Next, there has been a very careful expurgation of divers cruel
or barbarous practices, especially, I think, of those which seemed
characteristic of inferior races. The Iliad is full of battles, and
of battles fought with extraordinary fire. Yet the spirit of them
is not savage. It is chivalrous. No enemy is ever tortured.
No prisoners — with one exception to be noticed later— are ever
maltreated. Let us take two special cases where signs of ex-
purgation are visible.

We know that the dead body of Hector was dragged by
Achilles round the walls of Troy. That seems bad enough.
It seemed so to the poet : and the repentance of Achilles
is the main theme of the last two books of the Iliad. But
a far worse story was really handed down by the tradition.
There are fragments of the rude unexpurgated saga still extant,
according to which Hector was still alive when his enemy tied
him to the chariot rail and proceeded to drag him to death.
Sophocles, always archaic in such matters, explicitly follows this
legend (^j'h';*:, 1 031). So does Euripides (^;/t/?'(9?//. 399). Even
so late a writer as Vergil seems to adopt it.^ In fact, it may be
said on the whole to dominate the tradition. But Homer will
have none of it (X 361-95). Hector was dead — we are told so
not only in explicit language, but with rather peculiar repetition

* Aeneid ii. 273 'perque pedes traiectus lora tumeniis! Vergil was
probably copying the llin Persis in this passage.


— before Achilles began the deiKea epya, 'the shameful deeds.'
' And a dust cloud rose about him as he was dragged, and the
long dark hair spread wide, and all the head lay in the dust,
which before was beautiful ; but now Zeus gave him up to them
that hated him, to be foully wronged in his own fatherland.'

Again, there is, as we have said, no torture in the Iliad. But
there is a passage where a particularly dreadful wound is de-
scribed with, possibly, a certain gusto. The writhing man is
compared to a bull struggling in a net, and his pain is dwelt
upon. So far, perhaps, some older poet. But immediately a
saving line is added— a line of the sort that is technically called
* inorganic ', that is, which can be added or left out with no effect
upon the grammar or continuity. It runs, 'So he struggled
quite a little while^ not at all long'' {\iivvvBd irep, ov tl fxaXa Srju^
N 573). Now In the Odyssey, which, as I have said, is less
rigorously cleaned up than the Iliad, there is one scene of
torture. It is where the treacherous handmaids and the goat-
herd are to be killed. It has been decreed that the handmaids
shall not ' die by a clean death '. They are then hung up in a
row with nooses round their necks, ' so that they should die in
grievous pain *. So far, I think, the older poet. There follows
instantly the same saving verse : ' Their feet struggled for quite
a little while, not at all long ! ' (^ 473). The torture of women
was unpleasant even to an audience which approved the cruelty
to the goatherd.

Take another case, equally clear. The ordinary practice of
Homeric war allowed a warrior to take his dead enemy's armour.
This has, I suppose, been the case in all ages. But there was a
way of stripping the slain which added a sting of outrage to the
spoiling. The victor tore the dead man's tunic and left him
naked. This practice has been for the most part expurgated
out of the poems. Heroes are allowed to speak of it as a
possibility, or even to threaten it.^ But they are not allowed
actually to practise it. There are two instructive passages. In
iV 439 Idomeneus has pierced a man through the breast, and
then ' rends his tunic about him '. That is not pleasant : so the
line is added, ' even the tunic of bronze, which aforetime pro-

' B 416, n 841, just as they speak ot olKia to the dead as a possibility,
n 545, 559> and often.


tected his body from death.' The tunic becomes a tunic of
bronze. It was only the man's breastplate that Idomeneus
' rent ' ! In another passage, too (A lOo), there are signs of a
confused effort to escape from this barbarity. Agamemnon has
slain some men and taken their armour ; then he leaves them
' with their breasts gleaming, when he had stripped off their
tunics'. So it must originally have run. But in our present
texts, instead of ' tore ' or ' stripped ', there is a word [TreptSva-e)
which occurs nowhere else in Homer, but which must by all
analogy mean ' drew round ' or ' put on '. Agamemnon has
decently drawn the dead men's tunics over them ! ^ There are
many struggles on the part of commentators. There is a variant
reading which settles the matter by saying nothing about tunics
at all. Perhaps the most curious thing, linguistically, is that the
force of the context was too strong for the natural meaning of
the word TrepLSvco, and in later Greek it was normally taken, on
the strength of this passage, as meaning ' to strip '. Of course,
this sort of thing breeds confusion, and the corrector is no doubt
prepared to face it. The audience may be puzzled for a second.
But that will pass. If you told them that Agamemnon, their
great king, did on the battle-field one of those revolting things
that barbarians delight in and all decent Greeks utterly abjure,
the awkwardness would not pass so easily.

Another very interesting instance has been pointed out to
me by Professor J. A. K. Thomson. All through the poem
the heroes threaten at times to cut off one another's heads, and
sometimes in hot blood actually do so (e.g. A 147, N 202 ff.).
In P 39 Euphorbus threatens to carry off the ' armour and head '
of Menelaus; at 125 Hector is dragging Patroclus in order to
'cut the head off his shoulders with sharp bronze'. In ^ 177
Hector's heart urges him to cut off Patroclus' head and fix it up
on a post, like an African chief And in the same book, 334,
Achilles, addressing the dead Patroclus, says, ' I will not bury

^ CTTijdecri naiJLcf)nivopTaf, eVfl Trfplbvcre ;(iTcoj/tTy, A loo. See various interpre-
tations and strange constructions in Ameis, Anhang. Povelsen, the first of
modern scholars to point out the proper meaning ofnepiSvw, actually thought
that Agamemnon put on the shirts himself. Van Leeuvven and others call
the lines spurious or corrupt. The ancient v. 1. referred to is iirel kXvto. rev^^e^
aiTr)vpa. Aristarchus himself made nafjLipaiyoyTns agree with x"""^'"fj ^n
obvious makeshift.


thee till I bring to thee here Hector's armour and head.' Com-
pare X 348.

' Now I think,' writes Mr. Thomson, * that in the original story
Achilles carried out his threat. Look at the passage where
Achilles' dealings with the body of Hector are described,
W 2^ ff. "So spake he, and devised upon godlike Hector
hideous deeds (deLKea epya) : having stretched him prone by the
bier of Patroclus . . ." He did what ? Presumably deeds that
deserved to be called ' hideous ', but all that follows is : " in the
dust ; and the Myrmidons began to put off their armour and
loosed their steeds." I cannot get away from the impression
that something objectionable has been left out after rauvcraa^,
and the threat beforehand enables us to guess what that some-
thing was.'

It is interesting in this connexion to remember the story in
Herodotus ix. 78, how Lampon, son of Pytheas, proposed to
King Pausanias, after the battle of Plataea, that he should cut off
the Persian Mardonios' head and fix it up on a pole, and the rage
with which Pausanias rejected such barbarity; or the horror
with which Aeschylus speaks of ' lands where men's heads are
cut off and their eyes put out by process of law, boys castrated
as eunuchs and men mutilated, stoned and impaled' {Euin. 186).
Such deeds belonged to the " Beastly Devices of the Heathen,"
and were not likely to be tolerated in Homer.

Again, there is the matter of poisoned arrows. There is no

doubt whatever that the primitive inhabitants of Greece poisoned

their arrow-heads. The very word for poison,^ ro^iKoy, means

' belonging to an arrow '. And many myths tell of the incurable

and burning pains caused by arrows. The arrows of Heracles

in Hesiod (Aspi's, 132) 'had death on the front of them and

trickled with drops ' (cf. Scholia). Think of the hydra-dipped

shafts of Philoctetes. Think of the arrows of Apollo, bringing

pestilence. Think also of the peculiar word, so often applied to

arrows and arrow wounds, d^vKTo?, 'from which there is no

escape'. Does it not mean 'incurable' much more than

> This has been questioned, but cf. Strabo, p. 165 d 'l^rjpiKov 8e kqI t6 eV
€0(1 naparldfadai, to^ikov, o (rvvTideacnu (k jSoTcivrjs (Te\iva> Trpoa-opoias. Also
Dioscorides, vi. 20 to^ikou, en tov to. rd^a rau ^ap^upcov m avrov xp^fo^nt-
This puts the point exactly: poison was barbarous. Cf. also Luc. Nigrin.
yj and Paul. Aegin. 5. 53, where to|ikoV is a special poison.

2760 J^


' unerring ' ? The same thought explains why Eros is generally
armed with arrows, not with a great spear. He makes a wound
which looks slight, which perhaps hardly shows: but there is
in it a burning poison from which the stricken man does not

Now in the Iliad this poison has been completely cleaned
off from the arrow-heads. Poison is treacherous, ungentle-
manly ; a weapon for low barbarians, not for heroes. Yet you
can see from a number of lines what the arrows originally were.
Old phrases have been left unchanged: when Pandaros shoots
Diomedes in the shoulder he shouts in triumph that he cannot long
' support the strong arrow ', that is, that he cannot long survive
(E 104). In J 139 the arrow only just grazed Menelaus's skin ;
but Agamemnon immediately thought he would die.^ In v. 218
Machaon the leech attends to this wound, and the first thing he
does is to suck out the blood. Why, unless it was poisoned ?
In £1 394 the story is told how Heracles once wounded Hera with
an arrow, and ' the incurable pain laid hold of her '. Archers in
Homer chose out an arrow ' unshot before ', whose poison had
not been rubbed off {A 117, &c.). An arrow is habitually
described by epithets which gain point as soon as we remember
that arrows once were poisoned. They are ' bitter ', ' charged
with groans ', ' a foundation of black anguish '.^ The Odyssey\
as before, being less expurgated, is more explicit. In a 261 we
are told how Odysseus once went to Ephyra, to Ilos, son
of Mermeros — an ominous name — to seek a man-slaying drug
to anoint his arrows withal. But Ilos would not give it him.
He feared the nemesis of the eternal gods. ' But my father,' the
speaker continues, ' gave him some. For he loved him greatly.'
The Odysseus of the earliest legends must of course have used

We come next to a more complicated subject. With one excep-
tion, to be considered later, both Iliad and Odyssey are com-
pletely expurgated of the abomination of Human Sacrifice.

^ Of course, in the present course of the story, Agamemnon is reassured by
finding the wound slight.

^ niKpbs oldTos, ^ikea (TToPoevTa, fitXaivecop epfi odvvdav (whatever epfia may

^ Cf. Laws of Matm, vii. 90. * In war no poisoned weapons are to be
used, and no insults are to be addressed to a fallen enemy.' I take this note
from Mr. Romaine Paterson's eloquent book, The Nemesis of Nations.


The Homeric spirit would liave no dealings with such things.
It had too much humanity : it had too httle intensity of supersti-
tion. It did not denounce human sacrifice as Jeremiah, for
instance, denounced the rites of the Tophet outside Jerusalem.^
It is not Homer's way to denounce a thing that he objects to.
He merely sweeps it out of existence.

The early Greek myths are full of human sacrifices. one can
think at once of Menoikeus, Athamas, Phrixus and Helle, the
children of Heracles, Macaria, Iphigenia, Polyxena, and the
numerous virgin-martyrs of tragedy. If these stories were mere
fiction, it would be possible — though still difiicult — to hold that
they were unknown to ' Homer ' : that they were the horrid in-
ventions of later poets, trying to outbid their predecessors. But
they are not fiction. Nearly all of them come straight from some
ancient and disused religious rite, or some relic of very primitive
tradition. Iphigenia, for instance, is a form of an ancient an-
thropoctonous goddess, identified with Artemis.^ Polyxena is
a queen of the Underworld, ' Poly-xeina,' ' She of the many
Guests,' the wife of ' Polydector ' or ' Polydegmon '. Some of
these bloody traditions are doubtless Phoenician, and therefore
later.^ But others are pre-Hellenic. And even those due to
Phoenician influence were more than early enough for those
middle and later generations of the Homeric poets which were
mainly responsible for the work of expurgation. In the case of
Iphigenia, indeed, one can almost see the marks of the excision.'*
Now Homer has cut out these stories for their revoltingness, just
as he cuts out the cannibalism of Lycaon and Pelops, or the
mutilations of the Hesiodic gods. That is a sufficient reason,

^ Jer. vii. 31, xix. 5 ff., xxxii. 35; Ezek. xvi. 20 f., 36, xx. 26, 31, xxiii. 37,
39. Cf. Mic. vi. 6-8, &c., and laws in Dent. xii. 31, xviii. 10, &c.

^ Artemis-Iphigenia worshipped in Hermione, Paus. ii. 35. i. Cf. Hesych.
'l(})iy€veia' rj "Apre/ni? (Farnell, C7(//s of Greek States, vol. ii, chap, xiii, note 34),
and ^coiMos deas /loi ^vrifxa rrjs Aios K6pt]s (Eur. /. A. 1444), i.e. Artemis's altar
was Iphigenia's tomb !

^ on the date of the main period of Phoenician influence in Greece see
Myres in C. R. x. pp. 350 ff., and my article 'Odysseus' in the Quarterly
Review for April, 1905.

* (B 303-29 ; cf. Aesch. Ag. I15 ff., and the Cypria.) In Aeschylus and
the Cypria, when the bad omen occurs, Calchas declares that Artemis is
wroth with Agamemnon and demands the sacrifice of Iphigenia. In B, when
the hiiva TreXoipa invade the hecatombs and the Greeks are silent with horror,
Calchas rises and declares — merely that they will take Troy in the tenth
year ! one cannot but suspect that originally there was a price demanded
for that victory.

K 2


and, as regards the Odyssey^ it may be the only one that operates.
But if we look closer into the old stories of human sacrifice we
shall see that the subject has ramifications, and that there were
other causes contributing to this cleansing of the Homeric atmo-
sphere. With most of them we shall sympathize, with one
possibly not.

To take the latter first. The stories of human sacrifice that
have come down to us in myth are nearly all, for some reason or
other, sacrifices of virgins. one cannot be quite sure whether
this is due to history or to romance. The stories generall}^ occur
in the climax of a tragedy or some similar place, where they are
intended to produce an effect of romantic horror. So that
naturally young virgins are chosen as the victims, rather than,
let us say, middle-aged merchants. Yet, on the other hand, it is
likely enough that when such deeds were done it was more the
practice to slay a young girl than a man. The girl was more
likely to be ceremonially perfect : she was of less value to the
tribe ; she would be, at the best, more ready to die willingly,
and, at the worst, easier to kill.

Now the Odyssey stands on a different footing ; but I suspect
that these stories would have been rejected from the Iliad^ not
only because human sacrifice was a barbarity, but also because
the stories involved too intense an interest in women.

The Achaioi of the Iliad are habitually described by a rather
curious phrase, Kapt] Kojiocovre^^ not so much ' long-haired ' as
' letting the hair on the head grow long '. We may remember
the long hair of the Spartans at Thermopylae, and the Acoptee^
Tpix-diKe^. As to the original meaning of this phrase, I cannot
help suspecting that we may follow up a hint thrown out long
since by Robertson Smith. It means that the men were votaries,^
They had made a wow - vTroax^eais is the Homeric word^ — to
take Troy, and this implied a vow not to do certain specified
things until they had taken Troy. Like the warriors of the Old
Testament, they were consecrated.^ In modern language they

* Analogous cases in Religion of the Semites^ p. 333, and Additional Note i,
Taboos incident to Pilgrimages and Vows,

^ B 286 fF. vnodx^fTii of the Greeks. In B 349 it is Aio? vTroVxeo-ir ; in B 339
it is opKia. In Y 84 Aeneas viriaxixo (had made a vow) to fight Achilles,
The Franks had similar practices,

^ Cf. 2 Sam. xi. 1 1 (Uriah), i Sam, xxi. 4 f., and War in Enc, Bibl. Cf.
also Paus. i. yj. 3, viii. 41. 3 (hair kept for river worship).


were taboo while on the war-path, and the duty of never cutting,
combing, or washing the hair was the visible sign of various
other abstinences. The most important among these was
abstinence from sexual relations with women. I think that the
Iliad is quite consistent throughout in the recognition of this
taboo, a somewhat surprising fact. For the Poems seldom care
to be consistent about anything that does not occupy the front
plane of a hearer's attention. The nearest approach to a breach
of it is perhaps the situation in A. It seems odd that men under
a vow of this sort should quarrel about women-captives. But it
only seems odd because we think of the siege of Troy as a long
period. The Greeks had some hopes of taking Troy that very
day (J3 29, 66, 413), and then the vow would be ' off '. Agamem-
non's language is strictly correct (vv. 31, 113). He always
associates his love of Chryseis with ' home ' and ' returning to
Argos '. True, Achilles and Patroclus do not observe the taboo
in I, but that is because they have definitely renounced it, as they
have renounced their part in the war (I 665 ff.).i Agamemnon
seems to have observed it (I 133, 275). Nestor is too old to be
bound by it, and is waited upon by a handmaid, Hecamede
{A 624). I suspect that the peculiar woman-ignoring atmo-
sphere of the Iliad may have been due originally to this ancient
taboo of warriors on the war-path ; and that later, when the
actual religious ground had been forgotten, there remained
a womanless atmosphere and a feeling that any female interest
was out of place in a high story of war. That is why there is no
Brunhild or Guinevere among the motive forces of the Iliad'.
only a Patroclus. Love for a friend and fellow soldier is the only
love austere enough for this strife of heroes.

The exceptions to this ignoring of women are to be found
among the women of Troy, chiefly Helen and Andromache.

^ Cf. "^ 144, where Achilles renounces, for specific reasons, the vow not to
cut his hair. This perhaps explains the breach of the taboo in S2 676.
There seems to be a dim recognition of some such custom as I suggest in
Schol. AD on B 11, explaining the words Kapr\ koixoooptss. ' The Greeks of old
used to let their hair grow long dpfrrys kqi dvdi)eias x<^P'^-' Where was the
'courage and virtue' unless it was in some vow of the war-path ?— Of course
it is not suggested that everybody who was not keeping his vow had his hair
short; e.g. Hector in X 402, Euphorbos, P 51 f., and of course Paris, whose
motives are obvious. There may be a reference to this custom in the
complaint of Ajax's sailors, e'/jcoroj/ d' epcbrwj' aTreTrava-fv o'Iimi (6 TroAe/xos), Soph.
A J. 1205.


The Trojans were not under any such vow as the Achaeans.
They vAOuld have been only too glad for the war to stop any day.
They w ere not growing their hair long. In a Trojan atmosphere
women can be described and made interesting. It is in a Trojan
atmosphere, in the close neighbourhood of the great parting
of Hector and Andromache, that we have the one mention in the
Iliad of tragic or guilty love, the story of Anteia's passion for
Bellerophon. And how sternly it is cut down to a bare rhumd
of facts ! That whole subject, which has formed the most fruitful
spring of modern drama and romance, occupies in the whole
Iliad six lines out of some fifteen thousand ! (Z 160-5). These
Trojan princesses in the Iliad and many beautiful passages in
the Odyssey show how the Homeric poets could write about
women if they would. But in the case of the Trojan women
themselves we may notice two points. In the first place, splendid
as their pictures are, there is no love interest about them. The
whole of that subject is steadily ignored. Secondly, the great
passages all occur in what are generally considered as late parts
of the Iliad : and, as we shall often have occasion to notice, the
later parts of Homer show in many ways a growth of the spirit of
drama or tragedy. To the mind of a poet who had begun to
move toward that great conception, the position of the women in
a besieged and doomed city must have been in itself a subject of
such compelling interest that he might well venture to the very
verge of his traditional field in order to treat of it. Andromache,
the loving and noble wife of the great enemy, is a being made
for tragedy.

But outside these two or perhaps, if we add Hecuba, three
Trojan women there is a steady suppression of female interest in
the Iliad. There is no sacrifice of Iphigenia; no sacrifice of
Polyxena.^ The Amazons, firmly seated as they are in early
Epic legend, are only mentioned in late and so-called spurious
passages (T 189, Z 186). The crimes of the great wicked
heroines, Clytemnestra, Epicaste, Eriphyle, Procne, Althaia,
Skylla, and the like, are kept carefully away from the Iliad, and

^ Cf. Paus. i. 22. 6 of Polyxena : 'Homer did well to omit so savage a
deed ; and he did well, I think, to represent Skyros as captured by Achilles,
therein differing from those who say that Achilles lived in the company
of the maidens at Skyros.' The case of Clytemnestra in the Odyssey is
peculiar, and needs separate treatment.


allowed only a scanty mention in the Odyssey. There is nothing
about Creusa, Aeneas's wife, though she was an important
character in saga and received worship as a goddess. There is
nothing about the prophetess Cassandra. The prophesying of
Troy is done by a man, Helenus. Through nearly all the Ilmd
there reigns that austere and unsympathetic spirit which breathes
in the words attributed to Pericles, ' that a woman's fame is to be
as seldom as possible mentioned by men, either for praise or
blame ' (Thuc. ii. 45). This Thucydidean spirit is curiously
different from that of Aeschylus and Euripides. It is quite
different even from that of the Odyssey. It is a spirit so mon-
strously arrogant that we are apt to overlook a certain grandeur
which it possesses. When one thinks of the part sometimes
played by women in history — for instance, in French history —
one must feel, to put it at the lowest, a certain perverted spiritual
dignity in the fact remarked upon by Wilamowitz, that in the
whole political history of Athens there is only one woman, but
she pervades everything: the mail-clad Virgin of the Acropolis.

The victims, then, in these stories of human sacrifice are in
most cases virgins. But they have another characteristic. They
are all, without exception, persons of royal blood. That is to
say, they all owe their original creation to that dark and wide-
reaching tract of early religion which has lately been illuminated
to us by the work of Dr. Frazer. At the back of them stands
that to us almost incomprehensible being, which somehow com-
mended itself to the mind of primitive man, the divine king
who embodies the life of his tribe, and who must be born
anew at fixed periods lest that life should grow weak. He is
generally called a vegetation spirit, since the welfare of the trees
and crops is the first need of an agricultural tribe. But he affects
not only the fruits of the soil^ but also the flocks and the human
beings. So it is better to consider him as embodying the life,
or the vital force, of the community. As such he is the seed and
origin of the tribal god. If the tribal god is a beast or totem, as he
may be, it is because at a pre-theistic stage such a beast was the
chosen vehicle of the tribal life.

I will not spend more words in explaining this worship of the
divine king ; is it not written in the Golden Bozig/i, in the History
of the Early Kingships and the lectures on A Ills, Adonis, and


Osiris ? In their origin the slaughtered king, the god-king, and
the beast-king belong to the same region.^ They were largely
identical beings. In Greek mythology as we know it, these
beings, like other barbarisms, have been in divers ways trans-
formed ; but we can see their traces.

In Phthiotis, in Thebes, and in Athens we meet well-known
stories of the usual type : the city is doomed to destruction unless
one of the royal blood shall die for the people. In Athens
the last king, Codrus, sacrifices himself. In Thebes the one
remaining male of the royal line, Menoikeus, sheds his life-blood
into the dragon's den. In Phthiotis the stories are more con-
fused. Phrixus and Helle fly away, though Helle ultimately
dies ; the king Athamas is condemned to die, but always escapes
at the last moment. In some cases, it would seem, the divine
king was ipvicopo?. He was allowed to live for ' nine seasons','^
and then was removed before the sacred force had time to abate.

* on the original Greek ^acriXevs or 6(6s as medicine-man, and the KpaTos
KOI /3/n, or 7nana, that filled him, see Atithropology aiid the Classics, p. 75 f.
The history of this divine inana would well repay a monograph. It is always,
I think, associated with the power of the thunder. In Hesiod, Nike, Kratos
and Bie are always at the hand of Zeus ; in Call. H. Jov. 67 it is they who
made him king ; in the P7-omethetts, of course, they are his ministers. The
divine kings of the Ptolemaic period regularly possess v\.Kr]v km KpiWos els t6v
cinavra xpovov, implied in their /SatriXeia (Dittenb. Orient. Gr. 90, 35, and note
102) ; or KpuTos alone, or (xioTtjpia Km Kpciros, or aooTijpla koI viKT}, or the like.
The same with Roman emperors : Ditt. Or. Gr. 614 init. ; 625, 5 ; 678 init.
Id. Sylloge 757. 932, 5. Our own liturgy has familiarized us with a develop-
ment of this, T] ^naiXela /cat ?; dvvapis Knl fj 86^a els rovs nlavas. In earlier
times Tyrtaeus (4) says it is right that the Kings, honoured of God, should
lead {apx^iv), 8r]pov 8e nXTjdei viKrjv Kal Kcipros 'iireadai, the real divine power
should belong to the demos ! Solon (5) claims that he has given the Kpiiros
to the Demos, *as much as is sufficient,'

'^ As to evvecopos, the first thing to notice is that the word m.eans ' of nine
seasons ', and leaves us to find out what the ' season ' is. And as a matter of
fact it varied in successive ages. First, in the time of the primitive Moon
Calendar it was a month or a quarter (Eustath. k 390) ; at another stage
it was a half-year, a summer or a winter, a mode of reckoning which has left
its traces even in Thucydides. Last, when the Solar Year was well established,
it was a year. We shall find traces of all three uses ; for the present the
second is the most important. What, then, is the meaning and the special
relevance of nine half-years ? In the first place, let us realize that when the
Greeks said ' every nine half-years ' they did not mean ' every four-and-a-
half years' as we should; they meant every four years. Just as, when
reckoning in whole years, they called the same period a Penteteris, *a
five-yearly period.' 'Evvf copor means the same as * penteteric '. The special
importance of the four-year period is, of course, that by a little adjustment,
and giving the Olympiad 50 and 49 months alternately, it enabled the Solar
and Lunar years to coincide. Hence the great four-yearly games and festivals.

Minos, we learn, iweapos jBaalXeve, Aios fiiydXou oapiaTrjs (t 1 79). I cannot


Nine seasons comprised the life of the two vegetation-heroes, the
Sons of the Threshing-floor, Otus and Ephialtes, who tried
to scale heaven and were slain (A 311). Nine seasons also,
strangely enough, formed the limit of each incarnation of the
divine Minos, the perpetual king of Crete (r 1 79). Mr. A. B.

help suspecting that Minos was a divine king, periodically subjected to
some ordeal or deposed or murdered ; i. e. the Bull-King was regularly every
nine horai driven into the Bull-God's cave and there, really or ostensibly,
sacrificed. Compare a coin of Magnesia, a great centre of Bull-worship,
in which the Bull is kneeling at the entrance to a cave, which it seems about
to enter. It kneels, of course, as a sign of willingness. {Brit. Mus., Ionia,
xix. 9; I owe this reference to Miss Harrison.)

The evidence is: (i) He ruled for nine horai, therefore presumably he
somehow ceased to rule at the end of that period. (2) We have the definite
tradition that he went up into the Cave 'every nine years' to converse with
Zeus, to receive new commandments {Trpo(TTdyyi.aTa or vofiovs) and give an
account of his stewardship (Plato, Mifws, 319 d. Laws, 624 b, 630 d, 632 d ;
Strabo, pp. 476, 482, 762, citing Ephorus and Plato). 'Zeus' is merely the
Greek way of naming the Cretan Bull-God. The word 'years' has crept in
with the change of custom in reckoning. (3) This going into the cave of the
Bull-God can hardly be separated from going into the Labyrinth to be slain
by Mino-tauros. And the bloody tribute of seven youths and maidens was,
according to Plutarch, sent to the Minotaur 'every nine years' {Vit.
Theseus, xv). Did they conceivably at some stage die with the king or for
him ? It is noteworthy that the said divine Bull was originally ' made angry'
{f$rjypia)dr]) against Minos by the special wrath of Poseidon {Apld. iii. I. i, 3),
which looks as if originally it was JVIinos himself who was supposed to be
killed by it. (4) It bears out these suspicions that we have no saga-tradition
of Minos's death. (The first is Hdt. vii. 170, how a Minos was killed in
Sicily and his tomb worshipped.) That is, perhaps, he did not die, or his
death was a secret. He went into the holy cave and came out rejuvenated
after his converse with God. — There is, or was a few years ago, an ordeal in
Lower Nigeria, by which people go up a sacred road to the cave of the ' Long
Juju', and, if condemned, never come out again. Minos's mother, Europa,
who, as a young girl (I cannot find if she was nine years old), was carried off
by the Bull-Zeus, was also the wife of Asterios, which was the name of the
Minotaur. Minos himself pursued Dictynna-Britomartis 'for nine months' ;
at the end of which time she threw herself into the sea (Schol., Eur. Hip.
1 130). Has the proverbial 'nine-year-old ox' of Hesiod [Erga, 436) any
bearing on this subject? Aristotle, Hist. An. 575 b, says that an ox is at his
prime when nevTfTrjs or evveapos : 'which is the same thing'. In view of the
connexions between Crete and Sparta, it is interesting to find that the Ephors
' every nine years' watched for falling stars and then sent to Delphi to ask if
the kings should continue to reign or not (Plut. Ag-is, li). Cf. also Aetia
Graeca, 12 (Charila sacrifice), and Paus. viii. 2. 6 (the were-wolves resume
human shape after nine years). Also Pind. ap. PI. Meno 81 b (p. 133 Chr.)
eVflTO) ereif avbibo'i y\rvxa9 ttoKiv. The way in which these rituals stuck to the
letter of nine horai while freely varying the meaning of hore is instructive to
a student of human nature.

I subjoin the other passages where the word ivviwpof occurs in Homer : in
K 19 the mystic bag given by the King of the Winds is da-Kos /3o6f iweapoio :
il>. 39°) Kirke's enchanted victims are auikoiaiv ioLKores evveupoia-tv : cf.
Eustath. ad loc: in 2 351 Patroclus' wounds are filled dXeicfiaTos eupeupoio,
which had some magic power, us (f)app.aKa)8q rqv bxiva^iiv e^o^roy says Schol. A.


Cook has shown how Minos was a bull-god as well as a king,^ and
established his connexion with other periodic kings, such as the
Olympian victors. It is pretty clear from various evidence — the
Minotaur itself would be enough — that Minos on certain occasions
wore the bull-mask which asserted his divine nature. It was the
same with that other perpetual king, Pharaoh. At the periodical
feast of the royal marriage Pharaoh was masked as Osiris and
Pharaoh's wile as Isis, the deities whose incarnation they were.
I will not multiply instances from the daemonic masks of tragedy,
the apotropaic masks of comedy, the totem masks of Red Indian
tribes, the bull-headed and snake-headed maidens and youths in
the Mithras ritual. I will not dwell upon ^oa>7rL9 iroTVLa'Hpr] and
the yXavKcoTTiSa Kovp-qv. There can be no doubt that these names
reach back ultimately to a cow-goddess and an owl-goddess.^ And
we shall see in a later lecture how real is the historical connexion
between such saga-figures as Agamemnon, Diomedes, Achilles,
and these part-human, part-animal, part-divine tribal kings. But
it is just this sort of barbaric bestial haziness that Homer will
least of all things tolerate. For Homer there are no cow-
goddesses nor yet cow-headed goddesses, no owl-goddesses nor
yet owl-headed goddesses ; only a goddess in supremely
beautiful form who takes a blameless interest in cows or is
attended by a faithful owl.

And in just the same spirit Homer has drawn sharp and clear
the dividing line between men and gods. There are no persons

* See Mr. Cook's remarkable articles in Class. Rev., 1903, and Zeus vol. i.
pp. 491 f. (Minotaur the Cretan crown prince masquerading on a Bull), 662 n,
527 n. See also Bethe on Minos as the Bull-god of the Kefti (Egyptian for
' Cretans') in Rh. Mus. N.F. Ixv. The saga reflects the fights of the Kefti in

"^ See also Cook on 'Animal Worship in the Mycenaean kgQ,^ J.H.S., 1894.
' The custom of wearing a mask of the deity worshipped is common in the
religions of animal worship, in Egypt, Mexico, the South Seas, and elsewhere.
Lang, Myth, Ritual, and Religion, ii. 284 ; ib. 130. Cf. also Moret, Caractere
religietix de la Mojtarc/iie igyftientie ; Dieterich, Mithrasliiurgie. The
main Greek text for Pharaoh is Diod. i. 62. The fabled metamorphoses of
Proteus into various animals or a tree or fire are explained by the priests :
ev eOei yap elvai Tols kut' AiyvnTOv Swdarais TrfpiTidicrdai irtpl rrjv Ke(f)n\rjV
XeovTcov KOi Tnvpcov Koi SpaKovrcov npornfias, aqpeia ttjs apx^s' Kai nore fiev 8ev8pa
TTOTe Se TTvp, £(TTL 8 OT€ KOI BvpiapuTcov eucoScof exf'i' fVi Trjs Ke(f)aXf)s oiiK oXiya,
Koi 8ia TovT(ov ap.n fxeu envrovs di evnpiTTeiav Koap.e'iv, afia 8e tovs aXXovs els

Kaji'nrXrj^Lv liyeiv Ka\ 8iicn8aip.ova 8iu6e(Tiv. The trees and the fire are perhaps
invented for the sake of the Proteus story in the Odyssey, but the rest of the
account seems to be true.


in the Iliad or Odyssey^ as there are in the rest of Greek tradition,
who appear now as one and now as the other. There is a de-
finite avoidance of the makeshift bridge which satisfied Hesiod :
' the divine race of heroes, who are called demi-gods.' (See Leaf
on M 2'^, and Schol. BL, ibid.) Kings may be descended from
gods, and specially favoured by particular gods. But that is all.
The peasants of the Peloponnese continued long after Homer's
time to worship at the altars of a being called Zeus-Agamemnon.^
They may have been far from clear as to the distinction between
the God Cronos and his son Pelops at Olympia.^ But in the
Iliad Zeus, son of Cronos, is quite definitely a king of gods ;
Agamemnon, son of Pelops, definitely a king of men. There is
no shade of confusion between them.

It was a remarkable achievement of the Hellenic intellect, this
clear realization that a man was not a god, and that it was no
use calling him so. It needed such clearness of sight, such
daring, such humanity.^ We can see how hard the step was
when we reflect how small a part of the human race rose to the
height of following it. Think of the divine honours paid ages
after this to the Roman emperors. Think of the senate agreeing
to Caligula's claim of such honours for himself and his horse.*
No doubt there were mitigating circumstances in Caesar-worship.
The divine horse was an admitted eccentricity. Sensible men

* I see that Dr. Farnell doubts this ; in deference to so high an authority
I cite my grounds for the statement at greater length: Lycophron, Il23ff.
(where Cassandra prophesies f'/xo? 5' tiKotVf;? Zev? . . . STrapnuTau . . . KXtjOrja-eTai),
also 335, 1359 ff., and Scholia. Also Clem, Al. Protrept. pp. II, 18, cites
Staphylus for the worship of ' Kya\>.k\).vova riva Ala iv ^TrdpTj]. Usener has
pointed out what looks like an early trace of the same worship in Aesch.
Choeph. 255 ''"i ToO dvrripos Kai crt TifxcovTos fiija (cf. also ibid. 358, TrpowoXos re
ra>v n€yi<TTCtiv | xl^oviav (Ke'i Tvpavvcov). This may be a case of the well-known
sort, where two gods clash until one is made the priest or Trpon-oXo? or
Kkjibovx^s of the other, as e. g. Iphigenia was k\t)8ovxo^ of Artemis. Agamem-
non was King of Sparta (Stesich. 39, Simon. 20), and died at Amyclae (Find.
P. xi. 32), where Pausanias saw his tomb.

"^ See Mayer in Roscher's Lexicon, 'Kronos': especially ii. 1507 ff.
Observe that Pelops is Kronios, and that he also conquered Kronios. Paus.
vi. 21. II.

' Of course the making of the god in the first instance may have involved
a confusion of thought; the god may be only a projection of the 'mana'
of the medicine-king or the medicine-beast, or even simply 'le ddsir collectif
personnifie '. See below, p. 275, note. But the advance remarked in the
text was nevertheless enormous.

* Caligula also was an oapio-T?)? of Jupiter Capitolinus, exactly like Minos.
Suet. Calig. 22.


were conscious that the worship was in some sense metaphorical.
Politicians found it useful for testing and impressing the loyalty
of a distant oriental population. But the fundamental fact of
the matter is that such deification of kings did not seem to
educated Romans a thing unfamiliar or absurd. The old Roman
kings themselves, as Dr. Frazer has shown, had been in their
time personifications of gods. The various kings whom they
had conquered were all gods, the kings of Egypt, of Syria, ot
Parthia. The old Hellenic spirit was not then alive to testify.
The half-Greek Alexander and his generals had walked up and
down in barbaric places, where the old unpurified swamp was
still lying in the sun, and had caught the contagion of savage
ideas. La nosialgie de la bone laid hold upon them. Alexander,
who destroyed classical Greece, insisted that he was a god, and
the son of a divine snake. Demetrius received a semblance ot
divine honour even in Athens. That is just the atmosphere
which Homer and the spirit of early Hellenism had cleared
away — one might have hoped, for ever.

Like other morbid growths of the primitive human mind, these
deifications of living kings have had some particular develop-
ments that were beneficent and even splendid. But the verdict
of sane thought is against them. It is not only that their history
is written in blood. It is that they are in their very essence
degrading to humanity. And their abolition during the few
centuries in which the Hellenic power stood unbroken might of
itself be taken as a fair measure of the importance of Greece to
human progress.

So far, then, the cases which we have taken are instances ot
successful expurgation. The reforming Hellenic spirit has
ultimately, with what difiiculties and against what opposition we
know not, executed its will. Let us now consider a place
where it was baflled. Such passages were sure to occur in
a traditional book. For the first business of all these ancient
poets was to record history : and at times it happened that
objectionable facts were clearly and ineradicably fixed in the
history. The panegyrist of David who compiled our Book of
Samuel could not ignore David's treatment of Uriah. The poet
of Achilles cannot ignore the savagery of his hero's triumph.


The origin of the Uriah story in the midst of a tradition so
greatly modified for the glorification of David is in many ways
difficult to explain.^ But in the case of Achilles we may take it
as certain that in some early form of the saga, and even of the
poem, the ferocity of his revenge was part of his glory. Hector
did, it is true, by miserable treachery, contrive to kill Achilles'
dearest friend. But what a revenge our great Achilles took !
He tied Hector by the heels to his chariot, and dragged him to
death : all his friends looked on and dared not interfere. Then
he had fun with the body in all sorts of ingenious ways day by
day, till there was nothing left of it. Much the Trojans could do
to stop him ! And as for Patroclus, a round dozen of Trojan
nobles were slaughtered over his grave. That was how Achilles
treated his enemies. That kept the dogs in their place.

Now what was to be done with such an incident as this ? To
Homer — if we may use that name to denote the authors of the
prevailing tone of the Iliad— it was all odious and ugly. But it
was too firmly fixed in the tradition to be denied. A part of the
story, indeed, could be modified. Hector was saved from torture.
As we saw earlier, he was killed first, and dragged behind the
chariot afterwards. But what of the sacrifice of the twelve
Trojans ? Any sacrifice was an important and lengthy act. The
ordinary sacrifice of a bull in the Iliad has five lines allotted
to it, or ten, if we count in the roasting operations {A 458-67,
B 421-30). You would expect this sacrifice to have at the very
least twenty. As a matter of fact it is crowded into a shame-
faced line and a half! (^ 175). And that line and a half is
merely part of another sentence : it has not a whole verb to
itself. And it is followed by what certainly looks like one of the
extremely rare phrases of moral condemnation in the Poems :
' Yea, his heart devised evil deeds.' You could scarcely have a
clearer case of a poet recording a fact against his will. It is in a
very different tone that the Book of Kings records the human
sacrifices of the pious Josiah, when ' he slew all the priests of the
High Places that were there, upon the altars, and burned men's
bones upon them ' (2 Kings xxiii. 20 ; cf. i Kings xiii. 2, where
the word used is ' sacrifice ').

Even so, however, the fact stands recorded, and so does the
^ Though see note at end of Lecture IV. The Deuteronomists did omit it.


maltreating of Hector's corpse. No other corpse is so treated in
the Iliad. It is a difficulty like this that brings out the real
greatness of Homer. The whole of the last two books of the
Iliad is occupied with the psychological tragedy of this foul
action of Achilles.

In the first place there is not the faintest doubt as to the general
sympathy of the narrative. The gods, the reader, the poet, are
all at one. There is no exultation in the barbarity : there is only
shame and regret. I will go further. Of all the thousands ot
ferocious young soldiers, Greek, Roman, mediaeval, and modern,
who in their various days have read the Iliad and been ordered
by their teachers to admire it, it is hard to imagine a single one
rising from these last two books with a feeling that it was a fine
feat to do as Achilles did, and mutilate your dead enemies. But
the wonderful thing that Homer does is to make you understand
Achilles' state of mind. The cruelties which he practises are
those of a man mad with grief, a man starving and sleepless, who,
when he yields at last, yields in a burst of helpless tears. And
it makes some difference, also, that Achilles is deliberately
giving up his own life. He has the special supernatural know-
ledge that his revenge will be followed immediately by his death.
He heaps all that he has, as it were, upon the pyre of the friend
whom his own petulance and pride have caused to die,^

Homer, with his vibrating sympathy, his amazing language,
and that fiery splendour of narrative which seems almost to have
died out of the world when the Iliad was complete, can carry off
these deeds of horror, and leave Achilles a hero. Yet, even so,
Achilles as a subject for poetry, like the actual Achilles of
legend, paid for these savageries with an early death. It is
curious how little the Greek poets cared for him. He was the
uncontested hero of their greatest epic ; yet Greek literature as a
whole tends to pass him by. There is one lost Achillean trilogy
by Aeschylus, of which it would be rash to speak : there is one

* Starving and sleepless for twelve days, €i 31 ; tears, X2 510 ff. His own
death, 2 96ff. ; cf. his wonderful words to Lycaon, * 106-13: 'Nay, friend,
die like another! What wouldst thou vainly weeping? Patroclus died, who
was far better than thou. Look upon me ! Am I not beautiful and tall, and
sprung of a good father, and a goddess the mother that bare me? Yet, lo.
Death is over me and the mighty hand of Doom. There cometh a dawn of
day, a noon or an evening, and a hand that I know not shall lay me dead,' &c.


poignant and clever study of Achilles in Euripides' Iphigenia
in Aztlis. Late philosophers and pedagogues idealized or
allegorized him at their pleasure. But he inspired little great
poetry, and roused little imaginative interest compared with
lowlier heroes. He was associated with some of the faults that
Greece most hated, and he had not enough depth and variety of
character to make him fascinating in spite of them. Even the
man of many wiles, whose record in so many ways was far from
stainless — for instance, in that little matter of the arrows — speaks
much more in accordance with normal Greek feelinof. When his
great victory is accomplished and his wife and house delivered
from outrage, and the old Nurse is about to shriek for joy, he
bids her keep her joy in her heart, and refrain and make no cry :

Unholy is the voice
Of loud thanksgiving over slaughtered men (x 412).

One cannot help remembering in this connexion that the Iliad
in the fifth century occupied a central place in Greek education.
All well-born youths were trained upon it. And later Attic
writers speak with enthusiasm of the moral superiority of Homer
— and when they say ' Homer ' they chiefly mean the Iliad —
over the other ancient poets. Such is the common way of human
idealism. You first imagine ideal heroes and reshape your old
traditions till they yield the patterns that you desire : then, for-
getting they are your own creation, you contemplate them as
real historical beings and are kindled into a burning desire to be
like them. Whether this educational use of the Iliad began in
Ionia as early as the seventh century, which is likely enough, or
whether it only began in Attica in the sixth and fifth, we can
hardly help supposing that it had some share in these processes
of purification with which we have been dealing. The hand of
the schoolmaster certainly seems to have been at work — though
of course by different methods — in the case of another poet
much used in education, Theognis. Such parts of his poetry as
are obviously unedifying are relegated to a sort of appendix at
the end of the book, and in many MSS. are omitted altogether.^
But our evidence fails us. The use of the Iliad and Odyssey in

' Edifying passages from the old Ionic hortatory writers seem to have
been introduced into Homer. See Mulder, as cited below, Lecture VII,
p. 186. Also Brdal, Pour tnieux connaitre Homere, pp. 14 f.


education in classical times is a known fact, and a fact which
must have operated in the way required. It is a vera catisa.
Yet it is quite likely that the educational use itself is also a result
of some original moral superiority in the traditions of the con-
quering Northmen.^

Further consideration of this subject would lead us too far
afield. I am content for the present moment if I have shown
the mere fact that there was in the formation of the Iliad^ and to
a less extent in that of the Odyssey^ a strong element of reform and
expurgation. The tradition of early Greece, vast and tangled in
its wealth of varied beauty and ugliness as some South American
forest, was left by the Homeric poets a much cleaner and colder
thing than they found it. In this result two influences chiefly
were at work. First, a general humanizing of the imagination,
the progress of a spirit which, as it loved beauty, hated cruelty
and uncleanness. Secondly, the remnants of a race prejudice.
The relations of the Northern and the aboriginal elements in the
Homeric poems are involved, when you come to details, in
inextricable unity. But in its origin the ' Homeric ' convention
seems to represent some far-off idealized image of the Achaean
or northern spirit : the spirit of those scattered strong men who
in their various settlements were leading and shaping the Aegean
world. The special myths, beliefs, and rites that were character-
istic of the conquered races are pruned away or ignored, the
hero-worship, the oracles, the magic and witchcraft, the hocus-
pocus of purification : all that savours of ' the monstrous
regiment of women ', the uncanny powers of dead men, and the
baleful confusion between man and god.

Yet race prejudice is not quite the word. It is a race ideal,
and more than a race ideal. For it finds its main impulse not in
any maintenance of actual Northern tribes, past or existing, but
in the building up of something yet unborn. The earlier bards
had perhaps no name for this thing ; it was only a quality which
one felt in true Achaioi, Danaoi, or Argeioi. The later poets
knew it as Hellenism. True, the great division between Hellenes
and barbaroi is never in so many words expressed in the con-
ventional language of the Epos. The words are, no doubt, too

* See note on p. 263 f.


modern. They would break the convention, and are deliberately
excluded. But the feeling is there so strongly that eventually
the name cannot be kept out, and it enters, when it does enter, in
a strengthened and more un-epic form : ' Pan- Hellenes ' or,
rather more disguised, ' Pan-Achaioi '.^

Hellenism, as has often been remarked, denotes really not a unity
of race, but a unity of culture. Through all antiquity the sons
of Hellen were reckoned according to the spirit, not the flesh.
And the word ' Pan-Hellenes ' expresses just this. It implies
a readiness to extend the great name to all who are willing to bear
its burden, all who will live as Hellenes and take sides with Hellas.

Students of early Greek tradition are constantly brought up
against a certain broad contrast, between what is Homeric and
what is local. The local religion, the local legend, the local
feuds between Greek and Greek — these are things for which
Homer has in general no place. The Pan-Hellenism of Homer
strikes a reader even at first sio-ht : but it strikes him much more
keenly when he reflects in what a network of feuds and fears
and mutual abhorrences the life of primitive communities is
involved. ' Thou shalt not abhor an Edomite ; thou shalt not
abhor an Egyptian,' says the Deuteronomist, breaking down the
wall of hatred at particular points by definite injunctions. The
Homeric bards issue no such commands. They strike unnoticed at
the root of the whole system. They draw into the great orbit of
the Epos the ancestral heroes of all the Achaioi, Argeioi, Danaoi-
They show ' all Greeks ' labouring together, all of them suitably
idealized, all good men and true. They ignore everything that
is really tribal and exclusive, all the peculiar local rites, the taboo
tombs and secret names, which formed the very core of each
little village worship. They will deal only with such gods as can
stand publicly in the eyes of all Greece. It was a great attempt,
and it involved a considerable imaginative sacrifice. But mean-
time the new nation came into being. It worshipped Zeus
Hellanios, it attended the great Pan-Hellenic festivals, and there
in the four-yearly Homeric recitations Hellas found its book.^

^ UaviWtjvcs B 530, Ylavaxaioi at least 12 times.

^ For an instance of the extension of this spirit to the ' Homeric' Hymng
see Appendix G.




But let us turn to a question of evidence. I have been arguing
on general grounds that what we should expect to find in the
Homeric poems is some form of Traditional Book, which, like
the Song of Roland, or the Nibehmgenlied, or the Pentateuch,
or even Hamlet or the Covent Garden text of Richard III, has
reached its present form by a process of gradual growth and
constant rehandling. That is what we should expect. And our
study of the expurgations confirms our expectation. But is
there in the poems themselves definite evidence to show that this
is actually what happened.^ There is: and I will ask you to
spend some time in considering it. At this point, unfortunately,
the air begins to thicken with controversy, and controversy
generally obscures understanding. I propose to argue as little
as possible, but merely to make a re-statement of some of the
evidence already observed by various Homeric critics. My case
will be by no means complete. The evidence of language, for
instance, to my mind the most fundamental of all, is not suitable
for discussion in these lectures. But my object all through is
illustration rather than argument.

What we require for our purpose will be a series of cases in
which we already have reason to believe that a change of custom
took place between the Mycenaean and the Classical ages, that is,
roughly speaking, between the thirteenth century B. c. and the
sixth. If the Iliad is, as we have argued, a traditional book,
modified by succeeding generations, we shall expect to discover
some traces of this process. Probably we shall find, roughly
speaking, that on the surface the poem complies with the later
customs, while deeper down there are marks of the older. For
it is, by our hypothesis, an ancient poem worked over from time


to time to suit various new generations, or — to put the same thing
in another form — a comparatively later poem using masses of
ancient material. Let me say at once that we shall find nothing
amounting to demonstration. There is no possibility of demonstra-
tion in the case. We shall only find a number of comparatively
small and inconspicuous phenomena which are quite simple and
normal if the Iliad is a traditional book, and extremely puzzling
if it is not.

Perhaps the clearest case is the change of armour. The Greek
of Classical times was a conspicuous figure in his Ionian panoply.
He was clad in solid metal from head to foot : helmet, breastplate,
and backplate, small round shield, and greaves, all of metal.
When Psammetichus, king of Egypt, was driven from his throne,
he was told by the oracle at Buto to find bronzen ineji who
would restore him. He found them in the shape of Ionian and
Carian mercenaries (Hdt. ii. 152).^

Hdt. I. 171 (ot Kapef) . , . KCLi o\ava dcrni(Ti ovtoI fieri oi noirfadfievoi
■npcoTOC Teats 8e tivev oxdvcov ecj^opeov ras (lo-Trt'Sa? Travres olnep elutdeaav ua-niai
Xpdcrdai, reXapuai crKVTivoiai olrjKi^ovres, Trepi roiai avxecri /cat rotcri dpiarepoiai
wfioicri nepiKetiievot. Hdt. II. 1 52 wr riVty rj^ei. dno daXdaarjs xnXKewp dvdpaiv
(TnffiavfvTav. . . . "lavds T€ ku\ Kdpns. {Circa 650 B.C.) In view of criticism, let
me correct some false impressions. It is not part of my case to deny that
there were round shields and may have been breastplates in Crete or Egypt
in Minoan times: the evidence is doubtful; it depends on the Zakro seals,
which are difficult to interpret. The seal {B.S.A. xii. 241) selected by
Mr. Lang (p. 73) as most conclusive seems to me to represent a person
of uncertain sex carrying, not wearing, a ritual cope like that worn by the
leader of the Harvest Procession on the well-known steatite vase from Hagia
Triada. At any rate it covers the arms, and therefore can hardly be a
breastplate. But in any case Minoan is not Homeric, ov^e iyyvs; it is
pre-Mycenaean, and Mycenaean is pre-Homeric. — My case is that we know
of a big-leather-shield-and-no-breastplate period both from the remains
and the definite statements of Herodotus ; and we know of a classical period
with small round shields and complete metal body-armour. And both these
periods can be traced in Homer. So far the argument is archaeological ;
then comes the philological confirmation, the fact that wherever the thorex
occurs in the poems it is always ' inorganic ' and generally troublesome.
G. Lippold {Milnchener archaeologische Siudiett, 1909, pp. 400-504) ably
argues that ' Mycenaean ' is a misnomer. The big hanging leather shield of
Homer is the Dipylon shield, which he separates from the Mycenaean
and connects with the Boeotian. Dipylon shield-bearers often appear on
chariots, Mycenaean shield-bearers never.

Mr. Lang, besides his valuable argument about the date of the Homeric
breastplate, raises interesting questions about the chiton, and why Homer
does not mention the Kviraacns or archaic bathing-drawers. I will not
attempt to deal with that question now, but I welcome it, as also the question
he raises about women's dress. The general result of such inquiries will be,

L 2


Now the warrior of an earlier generation — we will call him for
convenience ' Mycenaean ' ; but the type lasted much later ; it is
assumed in parts of Tyrtaeus, and Herodotus conceives it as still
normal about 650 B.C.— went to battle in a very different state.
He was not in the least a ' bronzen man '. He had a leather
helmet, sometimes perhaps adorned with bits of metal. He may
have had sometimes a thick waistcoat or jerkin of linen to serve
for a breastplate, and soft leather leggings in place of greaves.
But normally he wore only a loin-cloth ^ and a linen tunic, while
instead of any corselet or body-armour he used the loose skins of
beasts, treated in one of two main ways. The common man got
the best beast-skin he could, the fell of a wolf, a goat, a pard, or,
if he could afford it, an ox ; he tied this skin by the paws round
his neck and let it hang. Then in battle he caught the lower
flapping edge with his left hand and held the skin tight in front
of him. It would keep off stones and arrows and perhaps sword
cuts, and would give him at least one extra chance of dodging
the cast of a spear. For he could whisk the skin aside as the
spear pierced it.

The chieftain or rich man improved upon this simple defence.

He had his ox-hide dried and made stiff and held in position by

cross staves of wood. As to the shape, the hide might be left

roughly in its natural condition, a sort of oblong ; a shield, as

Homer says, ' like a tower.' Such a shield covered the man

admirably from head to foot. But unfortunately it was a little

weak. It could be pierced by a spear- thrust. To meet that

diflSculty you could of course increase the thickness. You could

have two, three, or four hides instead of one. But that increased

the weight very seriously. Aias is said to have had a shield ' like

a tower ' consisting of seven ox-hides and a layer of metal. If

so, it must have weighed rather more than twenty stone ; we

need not be surprised that it was famous, nor yet that no one else

would have anything to do with it. But you could strengthen

the shield without adding to the weight by another device.^ It

in my judgement, that our Iliad, on the surface at least, is merely classical —
it represents the normal expectations of an audience in Athens in the fifth
century. To them, as to us, a hero dressed in bathing-drawers or a Helen in
archaic flounces would be an-pfTrtV. (See on Cretan armour Prof. Burrows,
Crete, pp. 37, 207.)

' See also Mackenzie in B.S.A. xii (1905-6).

"^ This remark I owe to Prof. J. L. Myres, who also suggests that the


can easily be practised on a half-sheet of note-paper. Take a
piece of the rim of the ox-hide about the middle on both sides,
a piece about a foot long, pinch the ends of each piece together
and at the same time draw both pieces inwards. That will make
the shield bulge out, both vertically and horizontally, till it projects
into a boss or point in the centre. It will so be stronger in itself;
it can easily be coated in the centre with a piece of metal ; and,
thirdly, weapons will glance off from it. The price you pay for
these advantages is, of course, that you make your shield narrow
in the middle. This is one reason, says Prof. Myres, why so
many people in Homer get wounded in the thigh or flank.

Now this shield was not regularly held on the arm by a strap
and handle like the later small shields. Its only strap was a long
one which passed over the left shoulder and under the right arm.
In the stress of battle a man's shield-strap ' sweated about his
breast ' {B 388), so that evidently there was no breastplate. The
cross-staves perhaps formed a kind of handle by which you could
move it to and fro at need — s^eer your dry cow^ as Hector
expresses it.^ But you could, if necessary, let the shield simply
swing, and advance on your enemy holding a great spear in both
hands, or two smaller spears, one in each hand. The shield was
so heavy that the warrior usually went in a chariot to the place

shields on the 'warrior Vase' are very likely Mycenaean shields with the
staves taken out, folded up for carrying on the march. They do not fold flat,
of course, hence the concave line at the bottom. The Dipylon shield is so
badly drawn that it is hard to be sure about it, but it is Mycenaean in
general character— large, leathern, suspended by a telainon. The large
leathern shield has left a great mark on poetic tradition. Protesilaos in the
legend was buried in his shield ; it was therefore Mycenaean. Similarly
Astyanax in the Troades is buried in Hector's shield. Amphiaraus when he
drove down to Hades was flying from the battle, and had his Mycenaean
shield hung on his back ; a vase-painter of the fifth century ( Wtejter Vorlege-
bldtter^ 1889, xi. 8), not understanding this, makes him — very awkwardly —
hold a small round metal shield behind his back (see Reichel, Waffen^ p. 64).
The shield in Eur. Electra, 430-80, shows Mycenaean tradition. In Tyrtaeus
I think one can show a clash or blending, much as in Homer ; this is natural
enough. In fr. li the young Spartan is to stand iv nponaxois . . . nrjpoCs
re Ki'Tjfxas re kc'ito) Knl arepva Ka'i u^ovs | danO^oi evpeirji yaarpl KaXv^dfievoi . . .
vjxe^s 6', o) yvfj,vijT(s, vn dcrniSos (iWoBfu aXXos | TTTcocrcrowfy fifydXois (idWere
X^pfiaBloii : a very ' Mycenaean ' picture. In 12, 26, however, there is a dapt]^
and the men fight in lines {(pdXnyya 21). See Wilamowitz, Die Textge-
schichte der gr. Lyriker, in Abh. der Gottinger Gesellschaft der Wiss.^ fhilol.-
hist. Klasse, N.F. iv. 3 (1900).

* viofifjcrai ^u)v I a(aXii]v, H 238. Herodotus uses the metaphor more
strongly of the pre-Carian, i. e. Dipylon or Mycenaean, shield : TtXap.a>ai

a-KVTlfOKTi olr)Ki(ovTfi (i. 171).


where he wished to fight. Arrived there, he dismounted, and
stood with the shield ' like a tower ' in front of him, or ' edged
himself step by step forward ' (vTrao-iriSia TrpoiroSi^cou) into striking
distance, being careful to keep always under cover. Dangerous
moments were those of getting down from the chariot, or getting
up again, or turning to retreat. There was also some danger of
tripping, both when you turned and when you moved forward.
For your shield-rim was close upon the ground, and you could
not safely look so far over the top as to see the earth close in
front of you. When once you were in position, however, the
cover was excellent, and there ensued what Homer calls a siadie
husmhie, a ' standing battle '. If no vital part of your enemy
showed round the edge anywhere, you entered into conversation
with him. A happily directed insult might make him start, lift his
head too high, or expose a piece of his flank. Then you speared
him. If you were a very strong man, you could try to drive
your spear clean through all his layers of ox-hide and reach his
unarmed body. Or you could even, as Hector and Aias some-
times did, by a blow with a huge stone, knock his shield right
back upon him and send him flat on the ground beneath it.

Peculiar and special tactics, as any one can see ; and quite
different from those of men armed with a small shield and
a breastplate. But now let us observe one particular piece of
what I may call the normal defensive drill. Suppose an enemy
threw his spear with all his force against your shield, the proper
plan, since you could not move the heavy ' cow ' swiftly about,
was to edge it as best you could in one direction and yourself
twist rapidly in the other. Then even if the spear came right
through your shield, it probably missed you or only grazed
your side.

Now what sort of armour, and what sort of tactics, do the
Homeric poems describe? It ought to be quite easy to say,
considering how much close description of fighting they contain.
As a matter of fact, if you consult Dr. Reichel, the discoverer of
this whole series of facts, he will tell you that the Homeric
heroes all fight in Mycenaean armour with the large shield and
no breastplate, except for some few late interpolated passages.
If you turn to Dr. Ridgeway, he will explain that the heroes all
have metal breastplates and round shields, except some few


individuals with ' Pelasgian ' antecedents. Neither of these ad-
mirable writers has, I think, faced the fact of the gradual growth
of the poems.^ Each tries to make the poems square with one
style of fighting or the other, and, when they refuse to do so,
proceeds to casuistry or violence. That is not a fair way to be-
have. We must take the poems as they stand. And, as they
stand, the main impression is pretty clear. The surface speaks
of the Ionian or Athenian style of fighting, the heart of the
narrative is something different and more primitive.

By ' the surface ' of the poems I mean such parts as the
formulae of introduction and transition, the general descriptive
phrases, the inorganic lines and some of the perpetual epithets :
all these are full of the Men of Bronze. We hear countless times
of the ' greaved Greeks \^ of ' the bronze-clad Greeks ', of ' the
clash of men in bronzen breastplates' (J 448 =©62), of 'the
whole plain blazing with bronze' (T 156), of how 'men's eyes
were blinded by the glitter of bronze from blazing helms and
breastplates, new-burnished and gleaming shields' (JV 341), of
a warrior whose ' whole body shone with bronze, like the light-
ning of aegis-bearing Zeus ' {A 66), or who ' gleams with the
bronze wherein his body is clad ' (M 463, cf. N 191, X 2>2, 134,
&c., &c.). It is the Men of Bronze everywhere. The gods who
watch the battle look down upon the ' flashing of bronze, men
slaying and men slain ' {A 83). And not only is it ' men of
bronze ' that we find in this sort of passage, but it is the tactics of
' men of bronze ', the movement of ordered regiments of infantry
in line, obeying their officers and making concerted movements,
like the classical Greek hoplitae. ' The Trojans came on, like
lines of waves on the sea, line behind line, flashing in bronze, to-

^ See Robert, Studien zur llias, who makes this same criticism on Reichel
(chap. i). Also Lippold, 1. c.

^ ivKvr)\xi.bes, only once x"^'^"'^^W^^^^i so that Reichel says the word only
means ' with good gaiters '. But gaiters, even when not hidden behind a big
shield, are not conspicuous or exciting objects, whereas the bronze greaves of
a line of men marching would be both, as the legs moved and the bronze
glittered. An epithet of this sort must be taken from something striking. I
am informed by the Hon. Oliver Howard that among the Suras, a tribe
against which he fought in Northern Nigeria in 1907, the cavalry wore
permanent iron greaves fastened on by a blacksmith so that they could never
be taken off, and fitted with a blunt spur on the inside of the calf. They
wore nothing else, except perhaps a loin-cloth. I know of nothing like this
in antiquity, however.


gether with their commanders ' {N 80 1). The Greeks ' advanced
in silence and in order, fearing their commanders, their hearts
set upon supporting one another' (P 1-9, J 427-32). That is
the way in which Nestor from time to time exhorts the Greeks
to fight, ' so that clan shall support clan, and tribe tribe '
(B 362 f.). It is the way which, we are told, the god Ares, as
a professional, especially commended ; that men should advance
vcs. phalanxes^ or lines, in close array, shield touching shield, an
impenetrable wall (IV 126, 130 ff,, 145). It is in this way that
people are said to be going to fight before each great battle be-
gins. But strangely enough it is not at all in this way that they
really fight when the battle is fairly joined, in the heart of the
poem. In the heart of the poem, when the real fighting comes,
it is as a rule purely Mycenaean. It is essentially a battle of
promachoi^ or champions. Usually each champion drives forward
on his chariot, dismounts and stands forth alone behind his big
shield, to engage in a series of duels. At most two or three
occasionally form together in a small group to check a rout or an
advance.^ At certain rare moments they drive their chariots into
the thick of a yielding foe (0 88 f., 348).

We have illustrated enough already the tactics of these
Mycenaean pj^omachoi or ' champions in the forefront '. But
the background of the Mycenaean battle deserves a word in
passing. Behind the great shielded champions there seem to
have lurked, in the real Mycenaean battle — first, individual

^ This is perhaps the movement indicated on the small vase from Hagia
Triada, described by Burrows (p. 38) from Paribeni in Rendiconti, Ace. Line.
xii. 324. See A. Mosso, Escursione nel Mediterraneo, Figs, '^'i,, 34. In
any case the chariots present some difficulties ; see Cauer, Grutidfragen,
P, p. 268 f. Why is the chief epithet of the chief hero ' swift-of-foot ' ? Why,
after the elaborate chariot-scene at his going forth in T 392 fF., does he never
use the chariot in pursuing the Trojans all through the next three books ? It
is only once mentioned, and then in repeated lines in a simile (Y 499-503 =
A 534-7 and 169). The only real chariot-battle, in the full sense, is in the
'horseman' Nestor's reminiscence, A 711-61 ; cf. his advice about chariot-
tactics in A 297-309, advice which seems never to be followed in the Iliad.
Diomedes also uses his chariot to charge Ares and Aphrodite in E. — I suspect
that we have a combination of sources ; for instance, tradition always gives
chariots to the heroes of the Thebais, Adrastus, Amphiaraus, Tydeus, &c.,
which might account for Diomedes (Miilder, Qiiellen, p. 72). Again, why do
27 Trojans have chariots, and only 9 Greeks ? For historical reasons, because
the Trojans are at home on their horse-breeding plain (see p. 36, n. 3), while
the Greeks have crossed the sea ? Or for merely romantic reasons, as Drerup
thinks, because it is glorious to fight on foot against chariots ? (Drerup, I. p.


distinguished archers, sometimes crouching behind the shield of
2i promackos in the very front, sometimes taking cover wherever
it offered ; and secondly, an almost unarmed rabble, shooting
arrows and little darts and stones from the sling or the bare hand,
making as terrible a noise as they could, and defending them-
selves with their flapping laise'ia. Now the distinguished archers
are of course present in the Iliad} but on the whole the bow is
somewhat fallen in repute, and, as one might expect, little is said
of the rabble. We can discern its existence clearly enough. We
hear how the Trojans in one place come on like flocks of birds,
screeching as they come (T 2). We have a good many mentions
of the stones and arrows coming from no specified hand.'^ But
in the main those undignified adjuncts of the ancient battle have
tended to be forgotten or omitted. The later poets were full of
the pride of Bronzen Men and the tough hand-to-hand death-
shock of spear and shield, as we hear of it in classical Greek

Let us stay a moment at this point. ' What ', it may be
objected, ' is this going to prove ? Why should you expect
a mixed army, collected from all parts of Greece, to be uniform
in its accoutrement ? The army of Xerxes contained Persian,
Median, and Assyrian soldiers, with the best weapons that the
century could produce, together with Ethiopians clad in lion
and leopard skins, and armed with stone- pointed arrows, and
Sagartians who carried daggers and lassos. The Chinese army
in the late war against Japan contained some soldiers armed with
the newest rifles, and some with bows and arrows. Early vases
combine Boeotian shields with round shields.'

Let us distinguish. Of course, the varieties in the armour
would not prove much. The poet often comments on the
peculiarities of different races in battle. We hear in H 419 that
the Lycians wore no initrae to their tunics ; in N 712 f. that the
Locrians ' had no bronze helmets nor round shields nor ashen

^ Paris, Pandaros, Teucer, for example. See Lang, Homer and his Age,
136 ff.

"^ Arrows, r 79, A 191, <& 113, O 313, &c. ; stones, M 154, n 774: but in
general scarcely a x^PH"^'"" is mentioned in the Iliad but has its definite
thrower. I suspect that every big stone lying on the plain of Troy had
its legend. It was thrown there by Aias or Hector or Aeneas or Diomedes,
as similar stones in Cornwall have generally been thrown by St. Paul, or
by the Devil.


spears ; they came with bows, you know (apa), and cords of
sheep-gut ! ' So we hear that the Dardanoi were dyy^ifia^rjTaL,
which seems to mean ' fighters in dense formation ', like the
dyxefxdxoL Abantes, who advanced ij,e/xacoT€? opeKrfjcrip ii^Xlrjai^
'straining forward with long ash-spears stretched out.' Pro-
fessor Macurdy ^ shows that the same formation was practised by
the Macedonians with the long sart'ssa, by the Dardani in Livy,
and by the Germans in Tacitus,

There is nothing here to cause difiSculty, or to suggest confu-
sion of dates or places. What makes the difficulty and suggests
the confusion is that the poets themselves seem not to be
conscious of the mixture of Mycenaean and ' Bronzen ' fighting,
or at least not to mean that the audience shall be conscious.
The men are, so to speak, advertised as being about to fight in
one way, and then, without a word of apology, they proceed to
fight in another. The fact is that, in all parts of the poem, it is
understood that, unless otherwise stated, each hero is clad in the
normal armour of the best style of Greek warrior. But in the
old lays that normal style was Mycenaean, or at least pre-Carian;
in the editorial parts it was the style of the sixth or fifth century.

In one or two places an actual correction of the text has been
made. There must have come a time — after Mr. Lang's argu-
ments we must not put it earlier than the age of Pisistratus, and
for myself I should now put it later — when the whole conception
of high warfare was wrapped up in these hand-to-hand battles of
Bronzen Men in full armour. Probably some reciter or editor of
the I/tad found among his sources lays describing both kinds
of fighting, and had to blend them together. Of course some
slight editing was necessary ; many omissions of lines no doubt,
a few simple and rather mechanical additions. For one thing,
the heroes, nearly all, find themselves summarily provided with
corslets, 6copT]K€?. The notion gives one something of a shock ;
it is so hard, in the atmosphere of modern print, to understand
the simple artifices of a Traditional Poem. Yet the fact Is there.
If we knew nothing of archaeology, if we could suggest no
explanation at all of such a proceeding, we should have to
suspect that the thorex had been put into the poem by a later
hand. For, often as it occurs, it is almost always in what is

* Troy and Paeonia^ chap. 3.


called an ' inorganic ' line. That is, the phrasing is such that it
can be dropped straight out without any injury to grammar,
sense, or metre. This is too extraordinary a state of things to be
the result of mere coincidence.^

To illustrate what is meant by ' inorganic ', let us take a fairly
innocent example. There is a passage twice repeated describ-
ing the first clash of battle : —

Together they dashed their ox -hides, together spears and

rages of men
[Clad in bronzen corslets, and bossy shields]
Came one against another, and a great turmoil arose.

J447ff. = 6i ff.

The line in brackets is inorganic. It does no great harm,
except that one does not quite see the difference between the
' ox-hides ' and the ' shields '. But drop it out, and sense,
grammar, and metre are as complete as before. There are
many such lines scattered about the poems, now here, now there,
and the fragments of papyri which have come down to us from
the second and third centuries B. c. often show such lines in places
where our texts omit them, and sometimes omit them where our
texts have them.

Sometimes the inorganic breastplate-line does actual harm.
There are two identical passages where a man performs the
sleight mentioned above.^ An enemy's spear comes through his
shield, but, standing well back from the shield, he twists aside
and the weapon grazes past him. I translate line by line : —

Right through the shining shield the strong spear came
[yi7id drove heavily '^ throicgh the richly -wrotighi corslet\
And straight on beside his flank it cut through his tunic,
That spear did : but he twisted aside and escaped black death,

^ A curious obstacle in the way of further analysis is the fact that we do
not know the derivative or original meaning of the words dtoprj^, daprja-aeadnt.
The verb is fairly common in the poems and was taken in classical times to
mean ' to put on a corslet ', though that is hardly its original meaning, and
there are many passages it does not suit (B 526, 587, 818 ; n 218, cf. 133, &c. ;
see exx. in Ebeling). Reichel thought Owprj^ was a general word meaning
' protection, clothing ', afterwards specialized to a particular kind of protection.
Another suggestion is that the verb means ' to make oneself bold ', and so ' to
prepare for battle'; this suits most of the Homeric passages, and accounts
for the slang fifth-century meaning, ' to get drunk '.

^ r 358, Paris ; H 252, Hector : cf. A 436, Sokos ; and A 136, Menelaus.

^ TjprjpfiarTo, 'was pressed,' or 'driven with weight', to ^Imov ttjs Tr\r]yrjs
7rapa8r)\ol ra rpaxd Toii prjixaros, Schol. BL.


Without the bracketed line the sense is clear. But with it ?
Does not every reader feel some difficulty ? You can twist aside
from a spear that is coming through your shield, but not
from one that has ' driven heavily ' through your breastplate.
Doubtless the audience understood it as a pluperfect : ' he had
twisted aside.' That is quite possible Greek. He had twisted
just before the spear struck, so the spear struck the very edge of
his corslet and, strange to say, instead of glancing off ' drove '
through. Sit down with a good will and you can imagine ways
in which, with exactly the right kind of corslet, such a thing
might conceivably happen : for of course the poets who recited
the Iliad would never leave a stark naked impossibility. only
the thorex can never be the real metal breastplate of dvSpcou
XaXKeoOoopiJKcoy. But how much simpler it would be with that
thorex-line away ! It occurs thus four times, making alwaj^s the
same kind of difficulty.

There is an arrow in /I 134 ff. whose performances are
described at great length, and very puzzling they are. Pandarus
had shot at Menelaus, and Athena was protecting him. She
brushed the arrow aside

And herself directed it where the clasps of his girdle

All-golden joined and a double protection met it.

Down dropped the bitter arrow on the fixed girdle,

And on through the cunning girdle it flew,

\^A nd it drove heavily through the richly -wrought corslet^

And through the initre which he wore to protect his flesh,

a fence against darts.
Which was his greatest defence ; right on through that it went,
And just grazed the man's flesh, &c.

Read this without the bracketed line and it is fairly clear. We
may at worst be a little puzzled by the exact relation between the
initre^ or waistbelt, and the zoster or girdle. Later on (185 ff.)
Menelaus is reassuring his brother about the wound : —

The keen bolt did not fix in a vital spot ; the flashing
girdle warded it off, and lower down the loin-cloth and
7nitre wrought by smiths.

He makes no mention of any breastplate, but says It was the
girdle that saved him ; he is able to say this because he has just
(v. 151) looked— apparently by pushing back the belt — a7id seen


that the string and barbs of the arrow are outside his flesh. All
is reasonably clear.

But now read the passage with the thorex-line in, and all is
confusion. The arrow went right through his breastplate.
What did the clasps of the girdle matter if there was a solid
metal corslet there ? Kow could Menelaus see the wound ? Why
is there so much talk about the piercing of the girdle, and ' the
initre which was his greatest defence ', and not a word about the
much more remarkable piercing of the breastplate ? ^ Other
awkwardnesses occur as one studies the passage : and they all
disappear with the removal of one inorganic line.

These superpositions of armour upon armour are not in-
frequent in our MSS. of early Greek poetry, though we must
always remember that, if a bard liked to have two versions of
a description or a metaphor in his private book, it does not
follow that he used both when he was reciting. one small case
was noticed by the Alexandrians. We are told of the archer
Paris in r 1 7 that he ' fought in front, with a pardskin on his
shoulders and a bending bow ', the natural accoutrement for an
archer, who needs both his hands. Then follow the lines (18-19)

And sword : and brandishing two spears tipped with bronze
He was challenging all the Argives to battle.

Zenodotus, and perhaps Aristarchus too, deleted these lines.
No doubt rightly. The two spears destroy the picture and

^ Mr. Lang bravely tackles the difficulties of this passage, and offers the
explanation that the arrow went, not through the thorex, but between the two
parts of the thorex in the narrow open space in front. Athena had, in fact,
by mistake, directed it to the one dangerous spot ! {World of Homer, p. 76.)

I cannot help suspecting that the ^'Vpr; also is interpolated here, or rather,
that there has been a contamination of two sources, in one of which it was
the tiirpr], in the other the double thickness of the girdle that saved him. It
is worth observing that the 6a>pr}^-\\ne makes a slight grammatical awkward-
ness wherever it occurs : it brings in a kuI clause between n^v and hi.
Possible language : but odd that it should occur always ! Apart from the
above passages the making of the thorex plays a curiously small part in the
Armour-Making, 2 478-613 ; 133 lines are given to the shield, one to the
thorex, one to the greaves, two to the helmet. That is, the shield was
originally all that mattered much. And in Y 259 f. Achilles does seem rather
to forget that he has a breastplate. Again, in the great passage describing
Patroclus' death, n 801 ff., Apollo, by a blow with the flat of his hand, makes
Patroclus stagger, so that his helmet falls off and he drops his shield. That
originally left him unarmed ; but the bard who armed him with a breastplate
has had to add the disastrous line 804 : ' And the Lord, the son of Zeus,
Apollo, also unbuckled his breastplate ! ' (Xwe Se ol duprjKa uva^ Ator vl6s


would prevent Paris from using his bow. It is interesting, too, to
see what happens later when Paris has to fight a duel in full
armour with Menelaus, He borrows the necessary breastplate
from Lycaon (F 330-8) and ' takes ' a sword and a spear. The
lines are, as usual, carefully arranged so as to avoid a direct con-
tradiction with the previous passage. Rut it is worth observing
two facts: a Papyrus (Hibeh 19) of the third century B.C. has
three additional lines here, describing armour, while Zenodotus
on the other hand deleted two of these in our text. We do not
know his reasons : possibly he only meant to delete the sword
and spear in one place or the other. But we see his method,
and can make out from it how an ancient bard or editor avoided

A very clear superposition can be seen in Hesiod's Shield of
Heracles (Aspi's, 139-320). The shield gives its name to the
poem, and has 180 lines of description, the rest of the accoutre-
ment sixteen. But this is not all. Apparently in the groundwork
of the poem the hero had a Mycenaean shield for practically his
whole defence. Then, as in Achilles' case (see note, p. 157),
other armour is added. But Heracles in tradition was represented
not only as a hoplite ; he was also an archer, also a korzmetes or
club-bearer. Consequently in Hesiod, as the text now stands
[Aspts, 122-38), he wears, all at the same time, greaves, breast-
plate, and helmet ; an iron club ; a quiver and arrows ; a spear,
and a Mycenaean shield ! The explanation in its general lines
would seem to be that bards had varying forms in which they
used to recite the battle of Herakles and Kyknos. In one
Herakles had a club and arrows, in another a Mycenaean shield
and spear, in another a full panoply. Some person in his ' book '
had notes of the alternatives, but when the book was published
the alternatives were added together.^

Before leaving this subject, there are two points we should
notice for the sake of their historical significance. In the first
place, while the breastplate and shield have been inserted almost
all through the Iliad^ there is no clear trace of them in the
Dolonela (K) nor yet in the Odyssey. K, we have reason to
believe, was a separate poem and not inserted in the Iliad till
a late date ; how late we shall discuss in Chapter XI. The breast-
^ Cf. Deiters de Her. Scuti Descriptione^ pp. 59, 399.


plate-inserter would seem to have done his work before K was
incorporated. In the Odyssey there was of course less reason to
revise the armour, as the military interest is much slighter than
in the Iliad. But this absence of the breastplate is another
instance of the fact we have noticed before, that the Odyssey
seems to have been altogether less worked over, expurgated, and
elaborated than what many books still persist in calling without
qualification ' the older poem '.

The second point is an observation on the epic style. The
introduction of the breastplate, on almost any conceivable theory,
makes, not indeed an absurdity, but at least some awkwardness,
some blurring of the presentation. The confusion of two styles
of fighting does the same. What we have to realize is that, like
most ancient poetry, the Iliad produces its effect not by accuracy
of detail but by a broad emotional sweep. It does not stimulate
our powers of close attention as do, for instance, the battle-scenes
of Tolstoy : it rather hypnotizes them by its rush and splendour
and stately music. We shall dwell on this characteristic more in
detail in Chapter IX ; for the present we may note one further
instance of it. A mark of the epic style is, as we all know, the
conventional epithet. All objects of interest have descriptive
adjectives habitually attached to them, and among such objects
are, of course, shields. Now you would expect, if the poet
wished to give a clear conception of what he was describing,
that the epithets would show at once whether a particular shield
was conceived as the great Mycenaean tower of ox-hide or the
small round metal targe of later Greece. But in fact it is not so.
When indeed a shield is called ^^olKk^ov, ' bronzen,' there is a
strong presumption that it is of the later type : when it is
d/j.(j>i^p6Tr] or Tro8r]V€Krj9, ' man-enveloping' or 'reaching to the
feet ', it is of the earlier. But as regards the greater part of the
epithets scholars differ. Reichel and Leaf try to make as many
as possible suit the Mycenaean shield. Ridgeway does the
opposite. What is clear is that shields which must from the
tactics have been Mycenaean, which are, for instance, large
enough to cover a man from head to foot, are called ' round ' or
' even in every direction ' or ' orbed ' or ' bossy ' — words which
at first sight seem to apply much more naturally to the later


It is just the same with the geography. Ahnost every
traveller who examines the plain of Troy or the isle of Ithaca
identifies the various landmarks to his own satisfaction, but the
identifications are all different. Let any reader study, for instance,
the discussions by Leaf and Robert about the position of the
Scamander. Both are quite first-rate investigators. They differ
completely as to the lie of the river, and only agree that the one
definite statement into which the poems are betrayed, that the
Simois and Scamander join their streams (E 774), is a mistake
and an interpolation.

A result like this cannot be attained by accident. It is a matter
of deliberate skill, this subtle avoidance of detail so as to con-
centrate interest on the central impression. In that respect it is
closely parallel to the use of types instead of portraits in classical
art ; but the artistic method may well have grown through his-
torical causes. A simplification of issues, a sweeping rejection of
the sort of detailed fact which is easily forgotten or which may
suit one reciter but give trouble to the next, a determined loyalty
to the broad poetic convention, undismayed by research and
untainted by personal idiosyncrasies : these are principles which
naturally suit the work of the ' Homeridae ' in carrying on by
corporate efforts a long-lived poetical tradition.

Let us briefly run through some other cases where the changing

customs of different ages have left their marks upon the poems.

There is the change from bronze to iron. The excavations have

produced no iron at Mycenae, and only two little lumps at Troy.

No weapons of iron have been found in the pre-Hellenic remains

anywhere. And on this subject the epic tradition is very clear

and vigorous. Bronze is the conventional metal of war : Ares

himself is xaA/ceoy, ' bronzen,'and ' the bronze ' proverbially means

' the sword '. Iron is known as a rare and very hard material,

difficult to work, but suitable for ploughshares, for clubs, for

arrow-heads, for axes.^ It is only now and then by accident that

a poet drops into using ' iron ' for a sword or spear, as we should

use * steel '. \ Antilochos is afraid lest Achilles should ' cut his

throat with the iron ' (S 34). Slaughtered oxen ' writhe about

^ Hesiod also thinks of iron in connexion with work rather than fighting.
Erga 150 )i'(iXko) S' f p-yafoj/ro, fjie\as 8' ovk idKe aidrjpof. on the antiquity
of iron see Montelius in Praehisiorische Zeihing, 191 5, esp. pp. 304 f. (Egypt
1300-1200, Troad c. 1 100, Central Europe 1000-875 B.C.).


the iron ' {W 30). So far we might be dealing with a ' transi-
tional' age in which iron was just coming into use as a new metal;
but then comes twice over the proverbial phrase, ' iron itself
draws a man on '—a weapon is a temptation (tt 294, r 13). Such
a saying would only be current in an age to which iron weapons
had been long familiar. Of course, though these mentions of iron
show clearly that the writers knew of iron weapons, the general
use of ' bronze ' and ' bronzen ' is no sign that the writers still
used bronze weapons. The memory of a bronze age happens to
have stamped itself on the language of poetry. That is all.
|; All Greek poetry was archaistic in language because it was
I permeated by a sense of style. It felt that modern words and
^phrases were out of tone with the heroic past. Swords are
spoken of as ' bronze ' down to the latest times of the Greek epic,
when such a thing as a bronze sword had perhaps not been seen
for centuries.

More complex Is the treatment of funeral rites. Burial was
at all times the usual Greek custom, but the flaming pyres of the
great Northmen of the Migrations lived in tradition, so the
Homeric heroes are burned. Yet both story and language are
pervaded by faint memories of a still earlier splendour, when
Mycenae buried and embalmed her kings.^ Hector was not
burned till the twenty-second day after his death. Achilles
himself was not burned till the eighteenth (/2 31, 413, 665, 785 :
(0 65). Surely those touches come from a time when embalming
was practised. The actual word which meant 'preserve' or
' embalm ' {rap^veLv) is used in Homer to denote the ordinary
burying of burned ashes. This is a clear case of survival, though
sometimes, from its very inappositeness to mere burial, the word
gathered to itself a metaphorical suggestion of ' preserving ' the
dead man's memory. ' His brethren and kindred will preserve
him with a mound and a pillar : for that is the honour of the
dead' {TI 456, 674}. The honey once used for embalming is

* on the gradual change from bronze to iron and burial to burning in Crete
see Burrows, pp. 100 f. Even in classical times kings who died abroad
were embalmed : e.g. Alexander, Agesipolis, Agesilaus. It is interesting
to note that in Scandinavia the general testimony of early writers put burning
before burial— the reverse of the truth. See above, p. 74, note. Dorpfeld
believes m a combination of the two, Comptes rendus du Congrh ArcMol. i\
Athhies, 1905, p. 161. But see Drerup, I, p. 160.




still vaguely associated with the last rites, thoug^h its meaning has
been forgotten. When Patroclus was burned upon a pyre they
set leaning against the bier two great jars of honey and unguents
(W 170). And Achilles himself was burned 'in raiment of the
gods and plenteous unguents and sweet honey ' (00 67). The
honey and unguents were useless : but man was reluctant to stint
his beloved dead of any honour that he had once given him.

There is a very interesting development In the forms of wor-
ship. The oldest Greek worship, like the Semitic, seems to
have had no temples and no graven images. You did not make
a god, at least not consciously. You found him : found him
dwelling In some strange rock, some ancient tree. In the water
that came from unknown depths and made the earth fertile.
You found him in the pillar that supported your dwelling, but
might fall, if angered ; In the battle-axe that fought for you so
bravely, but might at any moment wilfully break or miss its aim
or turn In your hand and betray you.^ And where you found
him you worshipped, and gave him sacrifice. Hence come the
' pillars and high places ', the Hebrew bdinoth and Greek boinoi.
At later stages you marked off a little space around the divine
object as specially sacred or haunted : this was a Temejios^ a
Precinct. Later still, as the faithful proceeded to make offerings
to the god at this precinct, you must needs have a resident priest
to act as caretaker ; and eventually, since, in spite of all the most
appalling curses on sacrilege which society could devise, the
offerings, hung on the tree or set in the crannies of the rock,
became too great a temptation to passers-by, It was best In the
end to build a properly walled house for the god and his belong-
ings to dwell in. How the Images of the god arose it Is not
clear. Dr. Relchel ^ believed that in general thrones came before
images. You found on some rock or high place some sign of
the god's habitation, a place where he sat or stepped or the like.
You Improved the seat for him ; In your temple you made a still
better seat, and eventually you put an Image of the god himself to

' See especially Evans, Mycenaean Tree and Pillar Cult, in J. H. S. xxi ;
R. Smith, Religion of Semites, pp. 97, 135, &c. ; W. M. Ramsay on Anatolian
Religion in Diet. Bibl., extra volume. Of course the combination of *ani-
conic ' and * iconic ' forms is common in later Greek religion : Prolegoviena,
pp. 18 ff., and ASHERAH in Encyc. Bibl.

* Vorhellenische Gotterculte.


sit there. The image would always serve an important purpose.
For the very simplest way of getting a god to do something was
to have an image of him and make the image do it. The chief
diflSculty lies perhaps in the transition from the real fetish to the
mere imitation or image. I find it difiicult to see how a purely
artificial image can originally have been worshipped except as
an imitation of something already known or supposed to exist.
Our early Greeks, driven out and cut off from their natural holy
places, would be reduced to making with their own hands imitations
of the god whom they had left behind.

Now it is clear that during the greater part of the Iliad and
Odyssey worship is carried on at High Places or altars in the
open air, ' We were gathered round a spring by the holy altars,
under a beautiful plane-tree, where bright water ran ' : so says
the Iliad of the sacrifice at Aulis, where appeared the wonder
of the birds and the snake (B 305, cf. 238 f.). So in the
Odyssey (^162) the sight of Nausicaa reminds Odysseus of the
young palm-tree which he saw ' growing beside Apollo's altar '
at Delos. It did not grow indoors. You hear normally, not of
the Temple of any god, but of the ' very beautiful oak of aegis-
bearing Zeus' (E 693, U 60, ^ 328, r 297) : of 'Athene's grove
beside the way, all of poplars ; and spring water runs through
it, and meadow-land is all around' (^291): of a grove of
Poseidon, a grove and altars of the Nymphs (B 506, i 200, p 210:
cf. V 278).

Then occasionally we hear of a iemenos, a precinct fenced off
from common life. We hear twice of the ' marble threshold of
the Archer Apollo in rocky Pytho ' (I 404, (9 80) : and lastly,
some seven times in all, we hear of definite temples. In Z there
is a full description not only of a temple and the worship therein,
but of a definite seated statue of the goddess Athena, on whose
knees a robe is to be laid, exactly as was done at the feast of the
Panathenaea, at which the poem itself was being recited. Is not
that a ritual centuries later, one asks, than the sacrifice by the
spring at Aulis } And observe a curious point. Chryses, in
the^'first book^of the Iliad^ is a very antique figure, not exactly a
priest, but rather a^professional ' cursing man ', or areier^ like
Balaam, son of Beor, in the Book of Numbers. And naturally,
when he performs his sacrifice, he does so {A 446 ff.) at an altar

M 2


in the open air. Yet in the introductory prologue he is made to
cry to his Mouse-God with the appeal, ' If ever I roofed for thee
a gracious temple' (,439). It is the same phenomenon which
we noticed in the case of the armour. The writer of that line did
not observe, or did not choose to observe, that in his original there
had been no temple, only an altar. To him and his audience
an altar implied a temple, so he took the temple for granted.

It is the same with another social change, affecting marriage
customs. In the primitive ages of Greece, as Aristotle has
remarked {Pol. 1268 b), ' men carried weapons and bought their
women from one another.' That is, the suitor paid a price,
normally calculated in oxen, to the father of the bride, who thus
became her husband's property. In classical Greece the custom
was just the opposite. The father gave a sum of money with
his daughter to induce the suitor to marry her. Speaking very
broadly, this means that in the early times there were not enough
women for the marriage market, in the later times too many.
It would seem that the first custom arose in an age when, owing
to dire poverty and continual wars, men hesitated a good deal
about rearing their children at all, and especially were reluctant
to burden themselves with daughters. There is something
touching in the frequency with which during the heroic times
you find names of women compounded from bous^ an ox. Oxen
were the gold currency of the time, and these names seem
perhaps to express the excuse which the parents made to them-
selves for venturing to rear the useless female child. ^ The real
reason was simply that they could not bear to kill it. But they
would never allege that. It is not the way with the human race
to avow such motives. We are much too shy. No doubt their
neighbours and the less agreeable of their elder relatives con-
sidered it extravagant of them, foolishly sentimental or ostenta-
tious. Well, maybe it was : but after all perhaps the girl
would bring in a good price some day : so they called her
Alphesiboia, zmniier of kine^ Phereboia, bringer-in-of-kiiie^
Polyboia, worth many ktne, or Stheneboia, Periboia, Eeriboia,
Meliboia, and the rest of the names.

* Cf. Letter in Witkowski Epistulae Privatae (58 = Pap. Ox. 744) from a
man to his wife : ' if it is a boy, rear it ; if a female, destroy.' Apollod. Bib.
3, 9, 2 ('AraXdjTJjS') 6 Trarrjp dppevav rraidui' inidviiSiv (^iOrjKiv ovttjp.


Now the poems as a rule maintain this older conception of the
marriage bargain. Hector bore his bride ' out from the halls
of Eetion, when he had paid countless bride-gifts ' (X 472).
Iphidamas was slain before he brought home his bride, and ' had
no joy of her, though he gave a great price ' (A 243). Othryoneus,
the suitor of Cassandra, gave his services in the war instead of a
bride-gift (JV 366: cf. A 289). Hephaistos in the Odyssey, when
Aphrodite is false to him, vows that he will keep her in prison
till her father returns all the bride-gifts, ' yea, all that I put in his
hand for the sake of his dog-faced maiden ' {6 319). There are
special exceptions which perhaps merely prove the rule. Old
Altes gave a great dower to his daughter Laothoe when she
married Priam (X 51). Agamemnon, among the gifts with
which he vainly sues Achilles, offers to give him one of his
daughters, not only without exacting a bride-gift, but giving her
a dowry as well (J 146 ff,). There is also an intermediate stage
in which the gifts are paid, not to the bride's father, but to the
bride herself, ^ They seem not so much a real gift as a proof of
the suitor's power to maintain a wife.

Now, so far, the evidence might be interpreted in either of two
ways. It might denote a long progress of time during which
customs changed, or it might point merely to an age in which
the custom varied according to circumstances. Two passages
in a late part of the Odyssey decide the question (/? 194, a 278).
' Let Telemachus bid his mother go back to her father's house.
And the folk there shall make a marriage -feast and furnish eedna
in plenty, such as are meet to go with a dear daughter.' A
dowry is meant ; but the word used is ee^ra, ' bride-gift.' The
writer of the lines was accustomed to the later practice of ^epi/?;
or Trpot'^, ' dowry,' and mistook the meaning of ee5j/a because his
generation had forgotten the custom (cf. also /S ^^.

It is the same with the question of the Homeric house. one
reason for the divergent theories of scholars about that elusive
object has been that they tried to work with only one form of
house, and there are really at least three. The house of Odysseus
at the end in the Battle with the Suitors stands by itself. It is a

] C 159 : cf. Schol. n 178 : also cf. Aesch. Prom. 559 eSmy a-yaye? 'V{<Ti6vav
Treiddp. The code of Hammurabi has marks of an intermediate stage,
practically equivalent to this. The suitor paid a bride-price to the father,
and the father also gave a dowry which normally included the return of the
bride-price, but did not always do so. See Ham. 160, 163, 164.


Mycenaean palace, not unlike Tiryns, as Prof. Myres has shown. ^
But the normal house of both the Iliad and Odyssey is quite
different. There seem to have been two types of house in the
Aeg-ean in early times, the Cretan or Southern palace and the
Hellenic or Northern one-roomed ' Megaron '. The Cretan
palace consists of countless rooms leading one out of the other,
and a whole structure so complicated that it has perhaps given
rise to the story of the labyrinth. Its main rooms tended to have
the entrance door or doors on the long wall of the room so that
the southern sun came in through the broad opening. Conse-
quently they had no fireplace.^ The Hellenic house was like a
modern shed or a Greek temple in aniis^ an oblong building
with a door at the narrow end, a porch in front, and a fireplace
in the centre of the big hall, which was called megaron or thalantos.
In the palaces of Greece proper, Mycenae, Tiryns, and Arne in
Lake Copais, this northern megaron has been combined with the
' labyrinthine ' scheme of the Cretan palaces. But in the Iliad
and Odyssey the houses are normally one-room halls. The master
and mistress live in the megaron in the daytime and sleep there
at night ; strangers are invariably given a bed in the porch just
outside the front door. That is where Telemachus is put when
staying with Nestor and with Menelaus (y 395-406, <5 296-307) ;
Odysseus with Alcinoiis (77 228-347), and when he is a stranger
in his own house (u i) ; Priam with Achilles (/2 643-50). Grown-
up sons and daughters have separate ' halls ' or thalamoi built
for them close by (y 413, ^ 2-5). When Hector goes to find
Paris in his thalamos (Z 321 ff,), he finds Paris cleaning his
armour, and Helen with her handmaids spinning, all in the same
room; and the room is certainly the place where Helen and
Paris slept. When the gods are summoned to Hephaestus' house,
they stand in the porch and see from there his bed with chains
like spider-webs drawn round it {Q 304, 325). And Alcinous
speaks of the night being long ; ' it is not yet time to sleep in the
hair (X 373).

That is the normal Homeric practice. But there are other
passages where the master and mistress have a separate bedroom
away from the hall ; Penelope, in particular, and certain young

^ /. H. S., vol. XX, and Monro's Odyssey, Appendix VI.
^ There is a central hearth in the second city at Troy — perhaps owing to
the climate, perhaps to some exceedingly early influx of Northerners.


g-irls dwell in ' well-wrought upper- chambers '. And here, as
before, the poet who brings in the later use does not notice that
he is contradicting an earlier use. So Helen and Menelaus go
to rest in the usual way ' in the inward part of the lofty hall ' ;
but in the morning Helen comes out of her ' fragrant high-roofed
bower '(5 304, 310, 121). In the case where Achilles puts the
aged Priam to sleep in the porch, the later poet seems to be
troubled at such apparent lack of hospitality, and invents a reason,
which no commentator has ever succeeded in understanding, for
not asking him to sleep properly inside (/2 643-76). Apparently
he did not understand the custom which he found implied in
his source.

Other evidence could be added to this : ^ evidence from the
treatment of the gods, a most curious subject ; from the law
about guardianship of a widow ; from land tenure, government,
and, most important of all, from the changes and misunderstand-
ings of linguistic forms. All are involved in a network of small
but ever-thickening difficulties as long as we try to regard the
poems as the work of one man or one age. All begin to clear
and become intelligible as soon as we recognize what the Poems
really are. They represent, not the independent invention of
one man, like some modern books, but the ever-moving tradition
of many generations of men, like almost all ancient books. They
are wholes built up out of a great mass of legendary poetry, re-
treated and re-created by successive poets in successive ages,
unities which have not perfectly assimilated the sources from
which they draw.

* Cases of conscious avoidance by Homer of ' modern ' subjects are given
by Br&l, Pour viieux connaitre Homere, pp. 7-1 1 : e. g. writing, statues,
paintings, money. Cf. also Drerup, I, chap. iv.

Note. — My discussion of the armour is based chiefly on Reichel, Hovie-
rische Waffen, Leafs Appendices to his edition of the Iliad, and Robert's
Studien zur Ilias ; Lippold's valuable article (see p. 147 note) only came
to me while the second edition was in the press. The passages about funeral
customs, bronze and iron, temples and dowries, are taken chiefly from
P. Cauer's admirable Grundfragen der Ho7nerkritik (third edition, 1921, 1923).
Some remarks also are due to Finsler's Ho7ner, and of course Helbig
(Homerische Epos aus den Dettkmalern erkliiri) and Tsountas and Manatt.
From the 'unitarian' side the best discussion of the armour is that of
Andrew Lang in The Wotld of Homer and Hovier and his Age. on houses
I would specially refer to Noack, Homerische Paldste and Ovalhatis und
Palast i?t Kreta, and the four articles by Mackenzie on Cretan Palaces in
the Annual of the B. S. A., xi-xiv. Cf. also Ridgeway in the Athenaeicm for


Nov. 21, 1908. (The gable and pediment are northern; the flat roof and
frieze Mycenaean.)

Miss Lorimer of Somervllle College, in an unpublished paper, has pointed
out that the typical Mycenaean weapons (figure-eight shields, inlaid sword-
blades, long rapiers) seem to have disappeared from the mainland of Greece
before 1400 B.C. ; that is, some 250 years before the Siege of Troy. Homer
does not mention these specific objects, but he uses freely the big ' Mycenaean '
shield, and he describes inlaid metal work in the Shield of Achilles (and
Euripides in a sword-blade. El. 476), elaborate palaces with bath-rooms
{a.(ya\i\.v6o^, a pre-Greek word), as well as objects like Nestor's cup, A 633 f., and
the helmet with boar's tusks in K 261-265, all of which are attested before
1400 but not after. Miss Lorimer considers that this points to the existence
in Homer of pre-Homeric and even pre-Trojan-War poetry, a conclusion not
at all improbable. However, tradition plus relics will explain all but the
shield, and the big shield did continue, as Herodotus tells us, at any rate in
Asia, till about 650 B.C. Mr. Hall has published in /. H. S. xxxi. p. 123 a relief
representing Sennacherib (705-680 B.C.) receiving a deputation of Carians or
lonians with crested helmets and big tower-shields. And on the mainland,
in spite of the silence of the monuments, the language of Tyrtaeus (p. 149 n.)
seems to imply the use of the big shield. See p. 196 n.

In the realm of language two general results seem to me to emerge with
increasing clearness : (1) the mixture of old and new is proved to the hilt;
(2) the task of separating the strata is shown to be much more difficult than
the last generation of scholars imagined ; you cannot simply cut out ' late parts '
and leave the rest uniform (see above, Preface to the Second Edition, p. x).

Fr. Bechtel, Die Vocalcontractioti bei Homer, supports the general results
of criticism, especially those of Robert, and is usefully criticized by K. Meister,
Die Ho7neri5che K^aistsprache, 1 92 1 . Particularly valuable are Wackernagel's
SpracJiliche Untersuchic7tgen su Ho»ier, 1916. Dr. Hentze, in several articles
{Beitr. ztir Kunde der IG. Spfache, xxix, p. 280 fT., Philologies, N. F., xix. 2,
p. 161 ff., Zeitschr. f. vcrgleich. Sprachforschtmg, N. F., xli, p. 356 ff.), has
treated the different stages in Homeric syntax, especially in conditional and
final sentences. Signer Delia Seta, in the Rendicotiti delta R. Accad. dei
Lincei {Classe di sc. morali, etc.), serie v, vol. xvi, pp. 134-210, shows interest-
ing results about the comparative age of the words 'A;^ato/, 'Apyeiot, Aawot ;
'A^ijr//, 'A^ijpair], and"lXtoi/, Tpoirj. The age comes out in the above order. The
late and perhaps Attic form 'Adrjvnlr] occurs oftenest in the Converse of Hector
and Andromache, Z (10 'Adrjvalr], 4 'Adrjvt]), a significant suggestion. Prof.
J. A. Scott of Illinois, in Classical Philology (Chicago), iv. 3, v. I, and
Classical Reznew, xxiv. i, has ably argued that some of the commonly received
differences Of language between the Iliad and the Odyssey, as wholes, are
fallacious, but has carried his own methods, I fear, to still more fallacious
conclusions in The Unity of Homer, University of California, 1921. See
G. M. Boiling in Class. Phil. xiv. 4 (191 9) on Abstract Words in Homer, and
ib. xvi. 4 (1921) on Infs. in -nnv. Miss .Stawell {Hotner and the Iliad, Dent,
1 910, a finely written book) has attempted a new division of the poems into
parts written by Homer and parts added later. (See Mr. Shewan in C. Q., April
and October, 1910.) The problem ot' the mixture of dialects in Homer receives
much light from Thumb's admirable Handbuch der griechischen Dialekte
(Heidelberg, 1909).

In the domain of metre, the enormous importance of which for Homeric
Language was demonstrated in Schulze's Quaestiones Epicae, an article by
Mr. J. A, J. Drewitt on Sca}ision in Hotneric Verse (C. Q., April, 1908) is
remarkable, both for its fine observation and its curious results. Solmsen's
Untersuchungen stir Gr. Laid- und Verslehre (1901) have had much influence,
but to me remain unconvincing : cf. Danielsson, Zur Lehre voin Horn.
Digamma, Ind. Forsch. xxv. 264-284, and especially Witte in Pauly-Wissowa
on Homeros, Sprache und Metrik.



While I was trying in my fourth lecture to draw a general
comparison between the Hebrew traditional history and the
Greek epic as regards their manner of growth, an objection may
have occurred to some of my hearers. The objects compared
are too unlike. The Book of Genesis or of Judges is essentially
a chronicle, a prose record of traditional history, narrated as far
as possible in order of time, year after year, generation after
generation. The Iliad is a definite poem, composed with great
artistic elaboration for an artistic end, beginning in the middle
of the action, and leading up to a skilfully prepared climax. Its
methods are the methods not of conscientious pillar-to-post
chronicle, but of artistic fiction. The time of its main action
amounts to some four days.^

This is true ; and before going further we should try to realize
how the difference has come about. Both books, I believe, are
made from much the same raw material, but they have developed
it in different ways. In the simplest form of the saga there were
probably elements of both prose and poetry — poetry where you
happened to find it, in lyrics or ballads, and prose to fill in the
facts. We find that style of composition in the Book of Judges
and some Icelandic sagas. But Hebrew poetry, as it developed
afterwards, is too impatient and emotional to narrate history.
In a book like Judges, on the contrary, poetry has been con-
quered by prose. The saga has been developed, to the best of
the writer's power, into a systematic prose history, chronologi-
cally arranged and edited with a view to religious instruction.
In the Greek saga it looks as if poetry had things its own way.
Greek poetry developed special forms for telling continuously

^ More exactly, four days of fighting followed by twenty-two of funeral.


the deeds of the past. And it told them as it pleased. The
versified chronicle became more and more of a poem and less of
a history. It meant no harm ; but it had in it from the first a
dangerous and unprincipled element, the poet s sense of beauty,
which in that particular soil grew, and overpowered in number-
less elusive ways the honest spirit of chronicle.

The early French epics were mostly known by the name of
Chansons de Geste^ that is, apparently, Songs of Gesta or
Deeds. This plural Gesta was often used in the title of historical
books, like Gesta Francorzun^ which was interpreted to mean
History of the Franks^ as though ' Gesta ' as a feminine singular
was equivalent to ' History '. The Epics were Songs of History.
The poet found his material sometimes in traditions and popular
songs, sometimes in the direct prompting of monks who read
or showed him their chronicles.^ Possibly some similar origin
should be assumed for most of our Greek epic remains ; but,
here as elsewhere, the diflSculty is that our record begins so late.
We have none of the raw materials left ; we have only finished
poems or fragments of finished poems. We have no Aeolian
epos ; no Ionian ; we have not even the text as recited at the
Panathenaea when it first came from Ionia. We have only, so
^ to speak, a text edited ' from the prompt-books of the best
theatres ' at some time later than the death of Aristotle. Still it
may be instructive to spend a few minutes in trying to think out
something of the processes by which poems could be made out
of saga.

Suppose, for instance, that some early editor of the Book of
Judges had been not a scribe or priest, but a Homeric bard or
rhapsode, how might he have treated his material ? ^ Our Book of
Judges consists mainly of the exploits of four Judges or Heroes
who delivered Israel from oppression : Ehud of Benjamin, who
slew Eglon, King of Moab ; Barak, of the northern tribe of
Naphtali, who defeated Sisera, the general of Jabin, King of
Hazor, and whose story contains the splendid song of Deborah ;

' See Les Ligendes epiqties, Recherches sur la formation des Chansons de
Geste, par J. Bedier. Paris, 1908. Also La Naissance de la Chanson de
Geste, by J. Y\3ic\\, Journal des Savants, vii (1909). — Did the phrase 'rerum
gestarum scriptor' = 'writer of chronicles' help in the change of meaning of
' gestes ' ?

* If I remember rightly, the old scholar Joshua Barnes did actually make
a Latin epic out of the Book of Judges.


Gideon of Abiezer in Manasseh, who overthrew the Midianites ;
and Jephthah of Gilead, who smote Ammon and sacrificed his
daughter. There is added to these an account of Samson, who
did not exactly deliver his people, and was rather a ' strong
man ' of folk-lore than a judge ; and an appendix on the sins
and destruction of the tribe of Benjamin. There are also brief
mentions of seven other Judges who are little more than names.
This raw material is worked up into an appearance of continuous
history with fixed, though fictitious, dates and a special religious

Now what would a Homeric bard have done with it? He
would, we may suppose, select a hero and a centre for his poem.
The choice would lie between three heroes : Gideon, who has
three chapters devoted to him, besides a long account of the
doings of his son ; Jephthah, who has two chapters and a fine
tragic story ; and Samson, who has four chapters. Now my
instinct tells me that he would not choose Samson : and to
choose Jephthah would lead at once to a human sacrifice in the
front plane of the story. It follows that he would probably
choose Gideon. Then he would consider how to draw into
his poem as much as possible of the rest of the book. He
certainly must not lose the Song of Deborah, for instance.
Looking through the record, he would find that at a certain
point (vi. 34 f ) ' Gideon blew a trumpet and Abiezer was
gathered together after him. And he sent messengers through-
out all Manasseh . . . and unto Asher and unto Zebulun and unto
Naphtali ; and they came up to meet him.' There is an opening.
When the herald went to Naphtali, we should be told, he spoke
to the men of Naphtali, and the men of Naphtali wavered, and
did not wish to join the war. They feasted and bade their
minstrel sing to them. And an old minstrel — in Greek saga he
would be a blind minstrel — came and smote his harp and sang
the Song of Deborah, how Jabin the Syrian had oppressed
Israel ; how Barak awoke and led his captivity captive ; how
Deborah arose, a mother in Israel ; how the river Kishon swept
them away, the ancient river, the river Kishon. So the princes
of Naphtali were reminded of the great deeds of their forefathers
and came in their strength to fight for Gideon. All the Song of
Deborah will come straight in.


The story of Ehud, again ; it is easy to get that told by some Ben-
jamite. Then the great story of Jephthah must not be omitted.
It only needs a little boldness. When the embassy comes to the
men of Gilead, we shall be told, their aged chieftain, Jephthah, is
bowed with grief and cannot join Gideon himself, because he is
not yet purified from the slaying of his daughter. He or another
Gileadite tells the story, and he sends his followers with a
blessing. The only real difficulty lies in the dates. Very un-
fortunately, Jephthah seems to have been later than Gideon. If
the chronology is firmly estabhshed, our bard will have to bring
in a prophet who can foretell Jephthah's story. But if the
chronology is not beyond dispute, or if our poet feels that, be
the facts as they may, the poem will be much the better for the
change, he will ignore the dates and let the Muse have her way.

And Samson? Well, one ot two things must be done.
Either we will leave Samson entirely aside, to be celebrated in
separate lays of his own, or, if we must cover that piece of
history too, we may have some character like Nestor in the
Cypria and Iliad^ like Menelaus in the Odyssey^ who can make
a digression and tell the whole story. Gideon's father, Joash,
might do, or his armour-bearer, Purah. Joash can regret that
men are not now as they once were, when he was young and
was entertained at Zorah by Samson : Samson, son of Manoah,
who ... Or he can warn some young man to be prudent, lest
he should fall like Samson, who . . .

And for the rest of the Judges, I believe that a Greek bard,
such as the authors of the Cypria, would have got them all in.
The wise Joash would denounce the weakness of the present
race of men, how unlike to Shamgar, the son of Anath, who
smote with an ox-goad six hundred Philistines ! Or Gideon, in
a great speech refusing to bow down to Baal, would explain that
nothing would induce him to do so, not all the riches of Jair the
Gileadite, who gave to his thirty sons thirty cities and set them
to ride upon thirty asses : not all the still vaster wealth of Abdon
the son of Hillel. And so on.

As a characteristic of the Hellenic races, in contrast with the
Hebrew, there is great significance in this tendency to work up
tradition into an artistic and poetical form. And it does add one


more to the already numerous forces which turn all legendary
history away from the path of truth.^ If you take up the Iliad
as a record of history, you will soon put it down as so much mere
poetry. But if you read it as fiction you will at every page be
pulled up by the feeling that it is not free fiction. The poet does
not invent whatever he likes. He believes himself to be dealing
with real events and real people, to be recording and explaining
things that have value, primarily, because they are supposed to
be true. And again, when you come to the passages that do not
represent real tradition but merely serve to join or to introduce
parts that originally did not belong together, you will inevitably
be struck by the extreme reluctance of the Homeric poets to
trust long to their own invention. It is one of the things that
sometimes irritates an ordinary modern reader in the analysis of
the Iliad or Odyssey to be forced to observe how the later
poets or editors, those responsible for a or 0, for example, will
go to any lengths in patching up centos of old lines, taken from
the most varying places, rather than invent new lines. Perhaps
their command of the epic style was not confidently perfect.
But after all it was not the business of a bard to invent. It was
his business to know, by information from the Muses or else-
where, the history of the past, and to tell it to his new audience
accurately, word for word, as the Muses had told it to him.
Even in the case of new songs, which naturally had their attrac-
tion, the poet's praise is that he knows them and tells them
accurately. ' Accurately ? ' Well, o-a0<5y e/caora ; each detail
vividly and clearly, so that you feel it must come straight from
the Muses, The imagination which he puts into it is merely one
of his best means of persuading people that it is true.

I suspect that the element of conscious fiction comes in first of
all in the formulae of transition and introduction. The writer of
Z, for instance, makes Glaucus tell to Diomedes during a battle
the whole story of Bellerophon. That is merely his way of
getting the history of Bellerophon told. He does mean that the
story is true; but he does not in the least mean to assert that
Glaucus actually told it on such an occasion.^ It would probably

^ Cf. note on p. 193 below.

* We happen to know that some ancient critics transferred the whole
incident to another place; presumably they were following their MSS.
Schol. Z 119.


be a very complicated business to unravel in the Iliad what the
reader is meant to take as history, and what is merely the device
of the poet for convenience in narrative or for dramatic effect.
And I fancy that the instinct of most readers will generally lead
them right without any rules. The important thing is that there
are real masses of supposed historical truth, somehow connected
together, and beautified, as they pass, by the processes of fiction.
The main basis is not fiction, but traditional history. A clear
proof of this lies, I think, in the general agreement as to state-
ments of important fact between all our different sources of
tradition ; the wide range of epic or quasi-epic poems ascribed to
Homer, Hesiod, Stesichorus, and others, and even, where we can
get them, the local legends attached to temples and oracles.
The differences between these various sources are of course
large and numerous ; but the underlying consensus of statement
quite unmistakable. And its significance can only be minimized
by adopting a theory which was universally prevalent a few
decades ago, but which in our present knowledge can only be
described as desperately improbable. According to this theory,
there is really in Greece no traditional history at all : the Iliad
and Odyssey are two primaeval works of fiction, presented as it
were by miracle from pre-historic times ; and all the other epic
tradition is made up out of these two books by the deductions,
imitations, and inventions of ingenious commentators.

In some cases this process has no doubt occurred. In others
it may have occurred. For instance, there existed in the sixth
century a tradition of a marriage between Telemachus and the
youngest daughter of Nestor, Polycaste. Now, in the Odyssey^
when Telemachus goes to Nestor's house, Polycaste is put in
charge of him and, after the custom of the age, gives him a bath.
Did the poet of the Odyssey know the tradition? Did he
perhaps know people who claimed descent from Telemachus
and Polycaste? Or, on the other hand, did the poet of the
Odyssey mean nothing at all when he mentioned this one
daughter by name and put Telemachus in her charge, and is
the supposed tradition a mere embroidery worked up from that
accidental mention ? In that case I hesitate to decide. But in
the great mass of cases one cannot hesitate. The existence of a
real saga behind any particular treatment of it forces itself upon


the mind ot the reader. As a matter of fact, the Iliad and
Odyssey not only refer to other legends as already existing- and
treated by other poets ; that every one admits ; ^ but they often
in their digressions tell stories in a form which clearly suggests
recapitulation or allusion. They imply the existence elsewhere
of a completer poetical treatment of the same subject. Take, for
instance, the story of Bellerophon in Iliad Z. The queen,
Anteia, her love being rejected, falsely accuses Bellerophon to
her husband. (Z 165.)

So she spoke, and fury seized the king for the thing he
heard. Slay him he would not : he had aidos of that in his
heart. But he would send him to Lycia, and gave to him
grisly signs, which he wrote inside a folded tablet, many
and murderous, and bade him show them to his wife's
father, that he might perish. And he went to Lycia wider
the blaineless sliding of the gods. And when he came to
Lycia and the flowing Xanthus the king of broad Lycia
honoured him with open heart : for nine days he feasted
him, and nine oxen he slew. But when the tenth rosy-
fingered dawn appeared, then he questioned him and asked
to see the sign that he brought with him from Proitos his
son-in-law. Then, when he had received the evil sign, first
he bade Bellerophon slay the raging Chimaera (She-goat).
Now she was of birth divine, and not of men : in front a lion,
behind a serpent, and in the midst a She-goat, breathing out
a fearful force of burning fire. And her he slew, following
the signs of the gods.

So on and so on. Bellerophon surmounts all his trials ; the
king of Lycia repents and gives him his daughter in marriage.
He seems to be on the point of living happily ever after.

BzU when he also was hated of all the godSy then verily
along the Plain of Wandering alone he wandered, eating his
own heart, avoiding the footfall of man.

What does it all mean ? Is that the way to tell a new story
unknown to your hearers? one wants more explanation all
through. What ' blameless guiding of the gods ' led Bellerophon
to Lycia } What ' signs of the gods ' showed him how to slay
the Chimaera ? ^ Above all, how did he become ' hated of all the

* Monro, Odyssey, Appendix, p. 294.

* Pegasus is omitted by Homer as a monster: he occurs Hes. Tkeo^. 325


gods ', and go wandering ? And why the phrase ' when he also ' ?
Is it not plain that the poet of Z is in the first place referring to
an existing legend, and secondly, one may almost say, quoting
from an existing poem ? And what can that poem have been ?
Bellerophon was a Corinthian hero. So that when we find that
there did exist an ancient mass of poetry vaguely called ' Corin-
thiaca ', and attributed to one Eumelus of Corinth, which is on
general grounds the obvious source for any Corinthian traditions,
we naturally conjecture that some early form of the ' Corinthiaca '
is probably the source of our particular digression.

Let us follow this conjecture further. Shortly before this
Bellerophon passage there comes in the Iliad (Z 130 ff.) another
digression, telling how Lycurgus, King of Thrace, came to an
evil end because he ' fought with the gods ' in resisting Dionysus,
and the gods hated him. The passage troubles commentators
because Homer usually ignores Dionysus. As Dr. Leaf says,
' Dionysus is an absolute stranger to the Homeric pantheon.' If
we look into the scholia we find that the story of Lycurgus
resisting the god Dionysus was told by Eumelus of Corinth in
the ' Europia '. The Europia, or ' Verses about Europa ', are
presumably the parts of the Corinthiaca or general Corinthian
traditions which dealt with Europa. The same source which we
suspected for Bellerophon ! Evidently Homer — if we may so
name the poet of Z — since he was using the Europia for the
story of Bellerophon, took the Dionysus-Lycurgus story from
them at the same time. And he speaks, you remember, of
Bellerophon also being hated of all the gods. That also has no
meaning where it stands in the Iliad. Apparently in the original
Bellerophon came in a list of such people, following upon
Lycurgus. Lycurgus was hated of the gods and went blind :
' Bellerophon also ' was hated of the gods, and went mad. It is
all clear. If anything were needed to make it clearer still, it
would be that the Verses of Eumelus are quoted as the earliest
known authority for the story of the Argo and Medea \ and the

rr]v fieu n^yaaos elKe Koi eV^Xo? BfWf po(f)6yTr]s, and is mvthologically very
ancient. (The Chimaera, a savage monster in remote lands, is obviously
less incredible than the tame Pegasus in a stable in Corinth.)

' e.g. by Schol. Find. 0/. xiii. 74; Schol. Ap. Rhod. iii. 1372 (six lines
directly borrowed from Eumelus) ; Paus. ii. 3. 10. That is, the most authori-
tative form of the Medea-Argo epic, in Alexandrian times and later, was the


composer of our Odyssey speaks of the Argo as a subject of
which ' all minds are full '.

There has been an extraordinary reluctance among- scholars
to look facts like these in the face, or even to understand the
possibility of their occurring. This comes from two causes.
First, criticism is still beset by the unfortunate phrase ' Cyclic
poets ', and all the false ideas it connotes. When the Iliad and
Odyssey had become canonical some scholar unknown made a
complete ' cycle ' ot epic history based primarily upon these two
poems and, where they failed, on the remains of the various old
traditional epics.^ To call the poets themselves ' Cyclic ' because
others made up a ' cycle ' out of their remains is as unfair as to
call an insect ' microscopic ' because persons over whom it has no
control choose to look at it through a microscope. It suggests,
what is quite contrary to the evidence, that it was the original
poets themselves who made this ' cycle ', deliberately completing
the Iliad and Odyssey. And secondly, Greek scholarship is not
yet familiar, as Hebrew is, with the idea of a traditional book.
The truth, as we have already seen, is that all these poems or
masses of tradition in verse form w^ere growing up side by side
for centuries. Either could quote or be quoted by the other as
easily as the Book of Judges could refer to Samuel or Samuel to
Judges. Both these books, if we are to believe the most careful
Biblical scholars, had begun to exist by 900 B. c. ; but Judges
was only finished a little before 200 B. C, and Samuel not quite
finished then. Or, to take a much stronger concrete instance, to
show how complicated this process of mutual quotation may be.

Corinthian epic of * Eumelus '. It is the habit of the Grammatici to quote
the earliest authority they can find. ' Eumelus ' is, so to speak, the ' Homer'
of the Corinthian-argonautic traditions. So far as we can guess at the date
of any personal ' Eumelus ' he would seem to be a Homer according to Nitzsch,
not a Homer according to Hermann— i.e. not the original inventor but the
late perfector of a floating epic tradition. The Corinthiaca had a most
interesting history and well deserve a new monograph. one can trace in them
(i) old mythical material ; (2) the fables generated by the earliest exploring
voyages to the NE. ; (3) a gathering-up and development of these legends in
Corinth as a centre; (4) late re-editing and abridgement, such as occurred to
the poems that were made into an * epic cycle '. Clement of Alexandria
{Strom, vi. 267, Sylb.) thinks of Eumelus as the man who made the prose
epitome then extant of certain epic traditions. See Appendix H.

* IlepnrovTat t5 eniKos kvkXos, €K dincpopaiv Trotjyrcor crv ^nrX-qpovjxevos
jJ-eXP^ ^^* drro/3a(rfa)s OSuo"(reco? rrj? els 'WaKrjv, iv fj Kai vno tov Traibos TrjXefMiixov
ayvoovvTos KTiiverai, Proclus apud Phot. Bibl. 319 A. See Appendix H, 'The
Epic Cycle.'

2760 JJ


Isaiah, chap. xxx\a-xxxix, is quite full of quotations, sometimes
complete, sometimes abridged, from the Second Book of Kings.
(Driver, L. O. Zl, p. 22^.) on the other hand, the Second Book
of Kings quotes not merely Isaiah but the much later writer,
Jeremiah ; and quotes him not directly but by way of Deutero-
nomy. That is, it takes from Deuteronomy passages which
Deuteronomy has already taken from Jeremiah. (lb. p. 203.)
All the great books were growing up together, and passages
could be repeated from any one to any other.

These facts should guard us against two possible misconcep-
tions. They show that the Iliad, though in one sense it may
be called an independent work of fiction, is also a Traditional
Book, dependent on a living saga or tradition. It was meant to
be history, or what then stood for history. And secondly, that
it is not alone among such books, a great original copied by
a few late and obscure imitators, but one among a great number,
each embodying the traditions specially prominent in their own
circles of influence, and all of them freely overlapping and
intercommunicating as the enterprise of a bard or the interest
of his audience suggested.

I have jotted in the margin of my Ih'ad notes of the probable
sources of the various bits of legend which seem foreign to the
main story of the Iliad or alien to their immediate context.
Many of them have been in ancient times or modern marked
as ' spurious ' or as ' interpolated ' — a phrase which seems
often merely to mean that the critic wishes a line were not
there when it plainly is. one finds in the first few books of
the Iliad: first, the Catalogue of Ships, belonging originally to
some Boeotian source, the school of genealogies and catalogues.
This was known even in antiquity. The ancient title of the
whole passage was ' Boeotia ', and It is omitted in many MSS.^
But we can see that there was an intermediate source before
the Catalogue came into the Iliad. Various points of language
show that the heroes are described, not as they were in the
tenth year of the war, nor even as already disembarked at
Troy, but as in the act of assembling at Aulis. For example

^ In D, T, U, and pap. B, among the best ones (Leaf). Mr. Allen, in his
great collation, cites an even longer list. The similes introducing the Cata-
logue, however, are not omitted.


695-710, 716-717. We have in 716 ff. a description of the men

of Methone, ' whose leader was Philoctetes the skilful bowman,

and in each ship were fifty rowers skilled to fight with the bow.'

So far all is according to the usual formula, and all is correct —

as a description of the fleet assembling at Aulis. But by the

time of the beginning of the Iliad Philoctetes had long been

marooned on Lemnos ; so, without any alteration in the original

version, eight lines are mechanically added to explain that

Philoctetes was not there. Similarly in 695-710 Protesilaus is

described as leader of the contingent from Phylake, and then

a correction is added to explain that he was now dead. Most

conclusive of all, the tenses of the verbs are imperfect : ' He

proceeded to draw up his ships ' or ' his ships moved into line.'

{ay^ vrjas, vies icTTixocouTo.) Now we happen, by the luck of

a quotation, to know that there was an old poem, the Cyprian

Verses, which narrated at length the assembling of the Greeks

at Aulis and also contained a Catalogue. True, our authority

only speaks of a ' Catalogue of the Trojans ', such as forms the

second part of our Catalogue in B. But to any one who has

grasped at all what literature was in the days before the book

trade and the reading public it will seem a strained hypothesis

to suggest that a Greek bard, reciting to Greeks, would give a

catalogue of the enemy and leave out his own people.^ We

^ See Wilamowitz, //om. i/nt., p. 374. on the later Catalogues (Apollo-
dorus' Epitotne ; Hyginus ; Dictys Cretensis, Dares Phrygius) see Allen,
Catalogue, pp. 22-31. In general they put the scene at Aulis ; their numbers
are variable, but often less grandiose than those in b; Hyginus, in the
manner of his time, tries to give more details, while Dictys and Dares
add some names. They are all dependent on B, or at least on the same
source as B and Euripides. See also Appendix J. The curious Cata-
logue inserted in Eur. Iph. Aul. 164-302 is, of course, abbreviated from
some older source, and that source seems to be the Cypria rather than
B. First, the ships are there described at Aulis ; Protesilaus is alive and so
is Palamedes (195-9) ! there is a reference to the Judgement of Paris (181);
all these points would come straight from the Cypria, they would imply con-
scious change if the source was B. Also, it is very interesting that the
problem how to harmonize the positions of Adrastus and Agamemnon — one
being, as Mulder puts it (p. 60 ff.), the great king of Argos in the Thebais, the
other in the Iliad or Ackiileis~is solved in a different way from that followed
by B. In B Agamemnon leads his forces from Mycenae and ' Sikyon where
formerly Adrastus zvas king ' (b 570) ; in Ip/i. Aul. 269 Agamemnon leads the
ships of Mycenae, ^ and with him was Adrastus, as a friend with a frie?td*
(The emendation dSeX^oy is a wilful refusal of light.) The question we
cannot answer is how far the MSS. of the Catalogue may have varied in
Euripides' day, but, as they both stand, it seems that the Catalogue in the
Iph. Aul. is in many points nearer to the source than that in b.

N 2


may fairly suppose that our Catalogue, wherever its origin,
stood at one time in the Cypria.

In any case the Catalogue provides us with an instructive
example of method.^ Whatever the source from which the
Catalogue comes, the poet of the Iliad^ in taking it over, has
taken over not only the facts but the actual words, even when
they did not quite suit their new context. The imperfect tenses,
for example, are certainly not natural as they stand. They are
left standing because the bard did not think it worth while —
or perhaps did not think it right — to re-write the lines. It is
exactly like that ' also ' in the Bellerophon passage.

It is of course hard to get clear instances of this process of
verbal borrowing because the poems which served as sources
are not extant. But sometimes we get a glimpse of one. For
instance, in the fifth book (E 385 ff.) there is a list of the
injuries done to gods by men, especially by Heracles, which
seems to be taken from the Heracleia (cf especially 403 f., with
Leaf's note). We happen to have a quotation from the Heracleia^
as composed and re-formed in the sixth century by Panyassis,
the uncle of Herodotus, and the quotation has a startling verbal
and rhythmical similarity with this passage in £. If the passage
in K could be original there, then Panyassis might have been
merely imitating E ; but the passage evidently is not original
in E. Presumably Panyassis and the author of E are both
adapting the same passage in an older form of the Heracleia?-

Another interesting reference to the Heracleia is in The
Tricking of Zeus (E 249-69), where Sleep mentions how Hera
once before, in the matter of Heracles, bribed him to put his

* Two remarkable books have appeared on the Catalogue since my second
edition: Leaf's Homer and History (Macmillan, 191 5) and T. W. Allen's
Homeric Catalogue of Ships (Oxford, 1921). Both regard the Catalogue as
coming from an alien source, but beyond that they differ. See below, p. 202.

* The lines are, in the Iliad, 385, 392, 395 :

T\r] fxev "Apr]s, ore fxiv Qros Kparepos t' 'E^tdXr/;? . . .
tXt] 8' "HpT/, ore niv Kpanpos irais ' Apcpirpvan'os . . .
tX^ 6' Ai8t]s (V TOLCTL neXcipios wkvv oicTTov,
€VTe pip avros dvqp ktA.

In Panyassis, fr. 16:

tXt] pep AriprjTrip, tXtj de kXvtos apcpiyvfjas,
tXtj 8e Ilo(Tei8da>p, t\?j 8' dpyvporo^os AnoWaif,
ap8p\ TTapci dprjra drjrevepep els ePinvTov.
rX^ 8f Koi o^pipodvpos "Apr]{ vno Trarpos avdyKj],


spell upon Zeus and how he suffered for it ; another, very clear,
in r 95-136 tells how, by a plot of Hera's, Heracles was born
a servant to Eurystheus. There seem to be other fragments
of Heracleia in some of the stories about Pylos. Other passages,
again, seemed to be derived from those poems or groups of
saga stuff which were eventually handed down under the names
of the Cypria^ the Little Iliad^ the Sack of liion^ the Aeihiopis^
the ArgonazUica^ the Battle oj Gods and Titans^ and the
Naztpactia or Aetolian verses.^

But the whole problem of Homer's sources has been brought
brilliantly into the foreground of interest since the first appear-
ance of this book by Dietrich Mulder's Die Ilias tind ihre
Quellen (1909). Mulder conceives the main idea of the Iliad
to be a great united war of all the Greeks against an arrogant
barbarian city, and that the poet builds up this idea partly by
pure invention, partly by free use of existing poems. The
Meleager story, the Thebais^ and a judicious fictional use of
Nestor account for the greater part of the Iliad. The Meleager
poem quoted at length in the Embassy to Achilles (I 527-599)
contains a \ir\vi<s with all the motives of the Achilles /:^i/iy as
told in the Iliad. It is, of course, earlier than the book which
quotes it. It also seems more original and better grounded.
When Meleager abstained from battle because of his mother's
curse, he stayed perforce in his ddXafios in the besieged city,
whereas Achilles might easily have gone away from Troy.
And several peculiarities of the Embassy Story in I would be
natural enough in an embassy to Meleager. This poem, then,
provides the motive of the 'wrath', which is for fictional
purposes highly convenient. The poet can have his chief hero
either present or absent from the war, and thus make room
for the deeds of all the other heroes of legend whom he wished
to draw into his great story. Then comes the Thebais. There
was, according to Mulder, no detailed record of any siege of
Troy, only a tradition of the Aeolic colonization and the actual
remains of more than one burnt city. But there did exist one
great epic describing the expedition against Thebes of a federa-
tion led by Adrastus, King of Argos. This provided a mass
of fighting by mixed Greek champions in a plain against a city
' See Leaf on E 392. Also below, pp. 219 ff.


on a hill, an ancient motive of art which we find already ex-
tant in Middle Minoan frescoes and in a silver vase from
Mycenae.^ This explains why Homer's Greeks are called
'ApyeioL and Aavaoi equally with 'A^aLoi^ which is the proper
name for Achilles' men. It also explains the presence of Greek
chariots in the Iliad. The charioteers of the Thebais are famous ;
witness Adrastus, Amphiaraus, and Tydeus ; while it is not
likely that an early Greek army would have been capable of
transporting horses and chariots in an expedition overseas to
Troy. The actual heroes of the Thebais are not brought over
to Troy. That would have been counter to tradition. But
Diomedes, the son of Tydeus, is there, and brings with him
a tremendous dpicrTeia embodying the only real chariot charges
in the I/iad, and perhaps taken straight from his father. The
place of Adrastus is taken by Agamemnon. He is the leader ot
all Greece, and we are told that he has succeeded Adrastus as
King of Argos, though in most aspects he and his brother
king look remarkably like two normal kings of Sparta, and
it is in Sparta, or the neighbourhood, that, outside Homer,
they have their graves and their worship and most of their
roots.2 It somewhat confirms these arguments when we find
that Hector, the defender of Troy, has quite independently been
shown by Dummler and Bethe to have intimate Theban con-
nexions and to have been worshipped in Thebes as a local hero.
This bold theory needs more careful consideration than it has
received. Probably Mulder errs in underrating the amount of
genuine saga material about Troy which is embodied in the
Iliad\ but I think he has shown that there is in it much more
conscious fiction than the present writer, at any rate, was
formerly inclined to believe; and, further, that the Thebais is
probably one of the poet's most important sources.

On the other hand there are books, and very fine books,
which seem to be pure original fiction. The most brilliant of
all are perhaps Z and /2. Then there is I, embodying, besides
the Meleager story, the realistic and perhaps genuine history
of Phoinix. There is also K, describing a midnight raid by
Odysseus and Diomedes, in which they catch a Trojan spy with

^ Evans, Palace of Minos, I, pp. 302, 314.
^ See below, pp. 210 f.


a fictitious name — Dolon, Crafty — and through him succeed
in killing Rhesus, chief of the Thracians. This looks like a
piece of fiction made up out of two separate traditional sources :
a tradition of the slaying of Rhesus by Diomedes, presumably
in Thrace,^ and another about the midnight expedition of
Odysseus and Diomedes into Troy to carry off the Palladium.^
Of course that is only conjecture. But it serves to illustrate
the kind of material with which we are dealing in the Iliad.

In its actual working up, however, our Iliad has reached a
further stage of development than the ordinary run of poetic
chronicles, if I may use the term. The imaginary epicizing of
the Book of Judges which we discussed some time ago would
land us, not in a poem like the Iliad^ but in one like the Cypria
or the Corinihiaca^ in one of those authorless chronicle- poems
of which we hear so much in Greek literature, and know, at
first hand, so little. It was their fate, first, to be superseded by
the Iliad and Odyssey^ and then, in a later age, to be strung
together in what was called an 'Epic Cycle' by some scholar
or historian. Here again the Odyssey shows itself a stage
nearer to the raw material. And, curiously enough, there is
one quite late poet who, partly by conscious love of the archaic
and partly from the peculiar nature of his genius, has returned
to a type of epic chronicle earlier than either the Iliad or the
Odyssey. I mean the Alexandrian poet of the Argonaut legend,
Apollonius Rhodius.

Let us consider this point more closely. What is the meaning
of the n^me. Iliad^ in Greek 17 'IXlus irorjcns} Ilias is an ad-
jective meaning 'about Ilion'. Poesis means 'verse-writing':
that is, first, it denotes the process of 'making' verses, and
secondly, the result of the process, a mass of verse-writing.
Not, you will observe, a thing quite so definite as a Poenia.
It is ' poetry ', not a ' poem '. The name 'iXias TTo-qais, then,
means ' the poetry about Troy '. That is the traditional name,

^ See below, Lecture VIII, pp. 215 ft".

^ K of course occupies a peculiar position. The Townley Scholia have a
very ancient note: 'They say that this rhapsody was "drawn up by itself"
{Ibia TtraxQai) by Homer, and is not part of the Ilicui, but was put into the
poem by Pisistratus.' The language of K is also in many ways divergent
from that of the rest of the Iliad. See Leaf's Introductory note to K, and
Monro, H. G., p. 234. (See, however, Shewan's Lay of Dolon, which tries to
rebut this observation.) It is a brilliantly written book.


and it is generally felt to be pretty satisfactory. But how does
the Iliad itself begin ? Does it begin, for instance,

I sing of Ilion and Dardania of the swift horses, for
whose sake the Danaans, servants of Ares, suffered many
things ? ^

That would be the natural sort of beginning for an Ilias Poesis.
And the lines did, as a matter of fact, form the beginning of
one of the old chronicle epics: the poem which afterwards
supported a mutilated and obscure existence under the name
of the Little Iliad.

Our Iliad begins with quite a different appeal : —

Sing of the Wrath, O Goddess, of Peleus' son, the wrath
accursed which laid many pains upon the Achaeans.

That is, it professes to tell the story of a fatal quarrel between
Achilles and Agamemnon, which took place in the tenth year
of the war and lasted for a very few days. Nay, it does not
tell even the whole of the Wrath quite exhaustively. It might
have included the capture of the two causes of it, the maidens
of Bresa and of Chryse. The poet appeals to the Muse to ' sing
of the Wrath, beginning iJiere where first there was strife and
sundering between Agamemnon, King of men, and divine
Achilles '.

Now, we can understand this language. It implies the
existence of a plentiful and well-known poetical tradition. It
is the phrase of a bard selecting for purposes of recitation some
special episode out of a longer history. It is the same in the
opening of the Odyssey : ' From somewhere amid those tales ^
O Muse, speak to us also.' It is the same with the bards who
are spoken of in the Odyssey.

And Demodocus called upon the god and made min-
strelsy, beginning where the Greeks had gone tipon their
betiched ships., and were sailing the sea, but Odysseus
and his comrades lay hidden in the market of the Trojans
{0 500).

That is how the Phaeacian bard is described ; and his lay seems
to have lasted for a few hundred lines at most. That is as much

1 Ps. Hdt. Vita Homeri, § 16:

l\iov dfiSoj Ka\ Aapdavirjv evTra>\ov
Tjs TTfpi noWa iraOiiv Aavaoi, OfpairovTes" hpr^os.

vn 'ILIAD' AND 'MENIS' 185

as people will usually endure to listen to. The poet proposes
to select out of a mass of legend the particular episode of the
Wrath, an episode just large enough to make a good ' Lay '.

The incidents of the Wrath are these: Agamemnon, provoked
by the free-speaking of Achilles, puts a dishonour upon him.
Achilles withdraws from the war. Agamemnon fights without
him and is defeated by the Trojans. The Greek ships are in
danger. Achilles is implored to save them. He still will not
fight himself, but sends his bosom friend, Patroclus. Patroclus
is killed by Hector. Achilles, furious with remorse, joins in
the battle himself, slays Hector, and gives Patroclus a splendid
funeral. The subject, as here announced, is not Ilion as a whole,
not even the last war of Ilion ; it is merely a four-days' incident
in the tenth year of the war. And yet the poem is called 'iXias
TTorjcn^^ the ' poetry about Ilion '.

And not unsuitably. For no sooner has the poet explained
in the first book the origin of the Wrath than he leaves that
subject and, roughly speaking, does not return to it until the
eleventh book. He goes back in the second to a catalogue of
all the Greek host, describing the fleet, not as it was in Troy
after nine years of fighting, but as it was in Aulis before it
started for Troy. After the catalogue come various battles,
including a duel or ordeal by combat between the two principals
in the international quarrel, Paris and Menelaus : battles which
are rather curious as they now stand, but fall into place at once
if you realize that they properly belong to the very beginning
of the war. The ordeal by battle was tried first : owing to
some Trojan's treachery it failed, and the two nations sat down
to a ten years' conflict. Then follow further battles ; in J an
obscure duel between two other heroes : ^ in jE a whole brilliant
poem about Diomedes,^ which not only runs the risk of upsetting
the balance of the I/md by dwarfing the exploits of Achilles,
but also shows in itself a definite connexion with another context.
Next, a fine stretch of poetry in Z, which tells of Troy from the
inside and treats Hector as a sympathetic hero, not a hated
enemy. Every line of it is noble : but how is it introduced ?

^ Very possibly pointing, as Bethe suggests, to a form of the legend in
which Aias was the chief hero. There seem to be traces of such a form.
' See note on p. 179. Cf, Mulder, Quellen der Ilias, 1910.


How is Hector brought into Troy ? In the thick of a desperate
battle, when Diomedes is slaughtering the Trojans and Hector
is the only man at all capable of resisting him, Hector leaves the
field to take a message, not in the least of a confidential nature,
to his mother, and to converse with his wife !

I am touching on all these points very lightly. The proof
of each one depends for its validity on detailed and accurate
examination of the words of the poem. I am using them merely
to indicate the sort of process by which the short Lay of the
Wrath of Achilles has been made into the great ' Poetry about
Troy ' : or, to put the case from a different point of view, how
the most diverse traditions of heroic fighting have been joined
together and made fairly consistent by this ingenious device
of the ' Wrath '. I cannot think that the Wrath was mere fiction.
It was an old traditional motive. But it was chosen, I vSuspect,
for its fictional convenience. one chieftain after another can be
the greatest of the Greeks while Achilles is away from the field.^
If another is expressly asserted to be the best, or swiftest, or
handsomest, of all warriors, even that statement can be retained
by the addition of an inorganic line, like

Tcov dWodv Aavacov //er' dfxv/xopa IlrjXetcova,

6(pp' 'A)(iX€v^ firjvuv 6 yap noXv (pipraTO? rf^v :

*of all the Greeks, else, after the blameless son of Peleus\ or
' while Achilles was in wrath. For he was the strongest far '.
The composer, as a matter of fact, has reached out on every
side and collected the most diverse masses of heroic tradition
to insert between the joints of his Wrath-Lay.

The result of this process is that the Ih'ad is really a Lay
which has utterly outgrown its natural boundaries. It professes
to be a Lay, but is so no longer. There are other instances
of this kind of growth in Greek literature. The Homeric Hymns
give themselves out to be npooi/iia, ' Preludes ' ; that is, mere
addresses to a god, preparatory to beginning a real poem : the

^ See Miilder, Homer und die altioiiiscJie Elegie, pp. 19 ff. Also Wecklein,
Sttuiien zur llias. Cf. N 321 ff,, B 673 f,, 768 f., H 1 1 1 ff., 226 ff., contrasted
with B 530, r 227, 229 ; Z 98 ff. ; H 289. These last are perhaps the only
passages where a superlative is applied to another hero without the addition
of some qualifying clause about Achilles. Miilder's arguments are attacked
by Rothe {llias ah Dichtung, pp. 31-8), but not, I think, successfully.


sort of prelude that Demodocus used when he 'began from
a god '. But these preludes have grown in interest and beauty
and length, till now the first five of them run to some hundreds
of lines apiece. They have become, not Preludes to a Lay,
but complete Lays in themselves. Again, the victory songs
performed by Pindar's choruses generally contain less than
fifty lines ; but one of them has over four hundred lines, burst-
ing all its natural bounds. That particular lyric, the Fourth
Pythian, was composed to be a great gift and peace-offering
laid at the feet of the King of Cyrene by an exiled noble. It
was to be a gift such as no other noble had ever given, no king
ever received.

But now comes a difficulty. Every work of art that was
ever created was intended in some way to be used. No picture
was painted for blind men ; no ship built where there was no water.
What was to be the use of the Iliad ? What audience would
listen to the recitation of such a poem .> It contains over fifteen
thousand verses. It would occupy twenty to twenty-four hours
of steady declamation. No audience could endure it, no bard
could perform it, in one stretch. And it is specially constructed
so as not to fall apart in lengths. From Lachmann onward
innumerable scholars have tried to break it up into separate
recitations, and have all failed. It is all one — at least, as far
as its composers could make it so. A single lay could be recited
at one sitting. A chronicle poem, falling easily apart into
separate stories, could be recited evening after evening in several
sittings. The Cypria, from what we know of them, would fall
apart excellently into separate episodes ; so would a good deal
of the Odyssey. It has the ' plots of many tragedies in it ', as
Aristotle has observed, and as we have noticed before. But
the Iliad has been deliberately elaborated on a plan which puts
it out of use for ordinary purposes of recitation. Yet recited
it must certainly have been.^

The late F. A, Paley was so much impressed by this difficulty
that he actually came to the conclusion that the Iliad was a
poem composed for reading, not for recitation, and that con-

' Cf. Br^al, 1. c, pp. 43 ff., who lays stress on the influence of Public Games
on the Iliad. His general conclusion agrees almost exactly with mine.



sequently it was not an early epic at all, but a learned poem
composed in Athens at some time between Euripides and Plato,
when there existed a reading" public. This view, as it stands,
is opposed to much that we regard as certain about early Greek
literature ; but Paley's arguments have never been answered, and
the difficulty is a real one.^

Now here, as it happens, the very first fact certainly known
about the Iliad and Odyssey comes to our aid. They were
performed at the Panathenaea at Athens, just as tragedies were
performed at the Dionysia, They were recited, not by one
bard, but by relays of bards in competition. The order was
fixed, the next bard having to begin where the last left off.
These festivals meant much more in ancient life than any
similar ceremony at the present day. In drama and epos, as in
other things, the ancients were accustomed to rarer and more
prolonged indulgences than we, their weaker descendants. After
waiting four years for a full-dress epic recitation they expected
something a//e/i7rroi/, which no one could call insufficient. The
Panathenaea was the greatest of all Athenian festivals, recur-
ring once in four years and lasting several days. Established
in the sixth century, it formed the occasion for the gathering
of all the lonians from their diverse settlements under the wing
of the great ' Metropolis ' or mother-city, who was their champion
and leader against the barbarian. The Panathenaean recitation
exactly explains what without it would be inexplicable in the
form and size of the Iliad and Odyssey.

The fact may suggest to us a question. What, after all, is
the meaning of the name ' Panathenaea ' ? Who are the ' All-
Athenaioi ' for whom the feast is made ? Not the Athenians
themselves ; that would give no meaning to the ' Pan '. The
answer occurs immediately. Who can the * All- Athenians ' be

' There is some answer to Paley in Geddes, Problem of the Homeric
Poems, Appendix A. My own view will come out in Chapter XII. Roughly
speaking, I think Paley erred because he still operated with a single poet,
who created the whole I/iad about 415 B. C. If he had grasped the conception
of a Traditional Book, and argued that work was still being done upon the
I/t'ad, that it was being edited with a view to readers, instead of audiences, as
late as 415 and even later, I think he would have proved his case. See Paley,
Remarks on Prof. MaJiaffy^s account of the rise and progress of Epic Poetry
(Bell, 1881), Post-Epic or imitative words in Homer (Norgate), Homeri quae
nunc exstant an reliquis Cycli carminibus antiqtiiora jure habita sint (Nor-
gate). Also Sayce's Appendix to vol. i of Mahaffy's Classical Greek Literature.


but the very people whom Athens was then shepherding, and
whose universal character was that they were ' all sprung- from
Athens ' ? Twelve cities in especial called themselves Ionian,
and had their great meeting at the feast of the Panionia at Cape
Mycale. But they were not more Ionian than many other cities,
says Herodotus : ' In reality all are lonians who are sprung
from Athens and keep the Apaturia' — an Athenian festival
(i. 147). only, he observes, many of them, and especially the
Athenians (143), avoid the name, and do not wish to be called
' lonians '. Exactly ; the name ' Athenaioi ' was more honourable ;
it was also wider in range. For it included those various cities
that did not belong to the Ionian Twelve, but admitted that
they were 'sprung from Athens '.^ The informal league of
which Athens was chief, at a time when ' the Ionian race was
of lowest account, and had no city of weight, except only Athens '
(Hdt. i. 143), could have chosen no better name than 'All-
Athenians ' when it gathered for its great festival every fourth
year, exactly at the same time when the great Dorian gatherings
met for the Pythian games at Delphi.

And, to return to the Iliad, what after all is its essential
story? Is it not the story of the battle of All- Greeks
against the barbarians of Asia? ' All -Greeks ' : the won-
derful word rings out again and again in the poems — what
though it comes chiefly in later parts, and against the tradition
of the epic style ? It is a modern formation, markedly out of
tone ; forcing itself in just because it so exactly expresses the
meaning for which the older language had no word. ' Pan-
achaioi ', you will say, or ' Panhellenes ' ; not ' Paniones '. True,
Homer uses generally the older and more dignified term,
' Achaioi ', to denote the whole race whom the Italians called
' Graeci ', the Asiatics ' laones ', the Greeks themselves in later
days ' Hellenes '. The lonians knew this, and even claimed
themselves to be not only ' lones ' and ' Athenaioi ', but also
'Achaioi'. To justify the claim they brought their founders
from Achaia. In later times, at any rate, they had the legend
that, while coming ultimately from Athens, their ancestors had

^ The theory that the lonians were all sprung from Athens may not, of
course, have had much historical foundation. That difificult problem does
not here concern us.


gone quite out of their way and stayed for a time in the little
district of the Peloponnese which was called by that name (Hdt,
i. 145).^ Paniones, Panhellenes, Panachaioi, and at last Pan-
athenaioi ; there is the same conception behind all these names,
only some minor differences of time or of local centre. It is
a union of men of Hellenic civilization against the multitudes
of eastern barbarism.

In many ways the Pisistratean festival of ' All- Athenians '

^ forms exactly the occasion for which our Iliad might have been
composed. The poem is not Athenian in the special sense,
but ' All-Athenian ' in the sense just explained is exactly what it
is. It is Pan -Achaean ; from the point of view of Ionia it is
Pan-Hellenic. If it breathes the spirit of any single city it is

' that mother-city which was claiming to be the champion and
the centre ot all who stood as Greeks against the barbarians
of Asia. We know of no city except Athens which could have
fostered a Hellenism so broad, so utterly un-parochial. Besides
this, if we are to believe some recent researchers, the ordinary
armour of the poems, the ordinary men's dress, the women's
dress, the conception of the appearance of the gods and much
of the actual religion of the two poems, seem to tally exactly
with Athens of the sixth or fifth century, and do not suit any
earlier period of which we have historical knowledge. These
broad facts are so strong and far-reaching that we need not
lay stress on the so-called Athenian interpolations — on the
statement that the almost unknown Athenian, Menestheus, was
the greatest ' marshaller of men and horse ' (B 554) in the array,
that Orestes came home ' from Athens ' (y 307) and not from
Phocis, or that Athena, when seeking her natural abode, went
into the ' House of Erechtheus on the Athenian acropolis ' (77 81) :
we need not debate whether the fact that Nestor's son in the
Odyssey (y, 5, 0) bears the fictional name ' Pisistratus ' is based
upon a compliment, or whether the verse ' Multitude 0/ masters
is no good thing; let there be one master ' (B 204) is a manifesto,
undetected and unexpurgated, in favour of the mild Tyrant
of Athens. Beyond all doubt the influence of the Panathenaic
recitation upon our poems was immense. Yet, though it in-
fluenced it cannot have created. The language was there
^ on this point cf. Wilamowitz, Die lonische Wanderimg and Panionion.


before, pre- Attic and pre- Ionic, already established as the
correct vehicle for epic. As to the Attic forms, apart from mere
errors of copyists, they do seem to prove the existence of Attic
poets, composing in good epic style, but occasionally, in spite of
all skill and care, slipping into their own dialect. It may well
be that the first emergence of a written Iliad which we could
recognize as such took place in Athens in the sixth century ; but,
if so, it was a re-composing of old themes out of old words and
verses and formulae.^

Behind the recorded Panathenaic recitation there must lie
long years of unrecorded recitation at various great Ionian
gatherings. Pisistratus, or whoever he was, must have taken
over to Athens an institution already existing in Ionia. one
thinks first of the Panionia, the great gathering feast of the
Twelve Cities at Cape Mycale. That is the obvious correlative
to the Panathenaea. And there is some confirmatory evidence.
It has been remarked long since that, among the Homeric gods,
there stand out three who are never jeered at or made ridiculous ;
two of them really grand figures, Poseidon and Apollo ; the
third, at least a very faithful and formidable partisan of
the Greek cause, Pallas Athena, who is especially prominent
in the very latest additions to the Odyssey. Athena was the
patroness of Athens in general, and in particular the visible
champion of Pisistratus. Poseidon and Apollo were the two
patron gods of the Panionia at Cape Mycale.

Or one might think of the great four-yearly festival at Delos,
at which the Homeric hymn to Apollo was sung by ' the blind
minstrel of craggy Chios ' to a gathering of all the ' long-robed
lonians'. The gods would suit almost equally well. About
this festival there is a curious passage in Thucydides (iii. 104).
In narrating how the Athenians in 426 B. c. ' purified ' the
island of Delos, he mentions that Pisistratus had purified it
before, though not completely. He had moved only those
pollutions that were in sight of Apollo's temple. He continues
his narrative of the doings of 426 : —

A nd the Four-yearly festival was then celebrated by the
Athenians for the first time since {or, after) the purifica-

• Wackernagel, Sprachliche Untersuchungen zu Homer, 1916. Cauer,
pp. 99-136.


Hon. There used to be in quite ancient days a gathering
of the lonians and the neighbouring islanders to Delos.
They came to the games with their wives and children,
as the lonians still go to the festival at Ephesus. There
was a gymnastic contest and a contest of minstrels, and the
various cities sent dances to the gods.

This seems to say that the Athenians in 426 celebrated the
feast for the first time since the cleansing of the island by
Pisistratus. If so, much would become clear. We could suppose
that, when Pisistratus cleansed the island and made the old fair
or gathering-place sacred and ' untreadable ' (ajSaroi/), the Delia
naturally came to an end, and the contest of minstrels was
transferred to Athens, as the federal treasure was transferred

We sometimes find Homeric critics, misled by mediaeval
analogies, discussing whether the Homeric poems are 'Court
poetry ' or ' popular poetry ', and generally deciding in favour
ol the former. But the parallel does not hold. The poems
do not show consistently the marks of either style. There is
high and prolonged artistic tension, there is no buffoonery or
vulgarity, there is an implication of dignity and culture in the
audience as well as the poet. So far we are reminded of court
poetry. on the other hand, they contain no flattery towards
any patron, no glorification of any special royal house, no hinted
demands for bakshish, no affectation of language. They go
into neither pigeon-hole. And very naturally so. For we
know from history that they were recited to an audience and
on an occasion completely different. They were recited at a
Panegyris, one of those solemn Pan-Hellenic gatherings which
formed the centre of so much of the higher life of Greece.^

There is the conventional mixed dialect, the conventional
generalized religion, best explained by the mixed and pan-
hellenic atmosphere of the whole occasion. There is a sacred
truce between all Greek cities. No Greek is an enemy ; no
race or tribe of Greeks is maligned or satirized. There is war

' The French Epic, Le Pelerinage de CharlcDiagne , was composed for the
Fair at St. Denys known as L'Endit. Bedier, op. cit., thinks that recitation
at the mixed gatherings of the great pilgrim centres explains the conventional
mixed dialect of the Chansons de Geste.


in the air, but it is not a war of neighbour against neighbour
in the common way of the mainland, it is a war of All-Hellenes
against the Barbarian, yet a war in which the Dorian nobles,
the martial aristocracy of Greece, are strangely ignored.^ There
is the religious euphemia or avoidance of evil words, which is
known to have been incumbent at these festivals and which
probably explains not only much of Homer's expurgation of
ancient legends, but his deliberate abstinence from all those
notes of horror and ghastliness which are common in the rest
of early epic tradition as well as in Attic tragedy. Apollo was
more insistent on euphemia than Dionysos.

What a difference, after all, there is between the Greek and
the Hebrew traditional book ! The general process at work was
much the same in both, but a great divergence must have begun
early. The Hebrew reviser, except where religious motives
came into play, tampered so little with his wording. He took his
raw material just as it was, and copied it out, merely inserting his
introductory and connecting formulae, smoothing out contradic-
tions, and correcting the orthodoxy of his authorities where they
needed it. A Homeric scholar cannot but be surprised at the
extreme ease with which interpolations in the Hebrew writings
often betray themselves. They are made quite undisguisedly,
with no artifice and sometimes no regard for grammar.^ No
Greek editor ever dreamed of doing his business like that. For
every Son of Homer was himself a poet, and kept modifying
and working up into poetry everything that he touched.

Consider the ultimate purpose to which the literature was
destined in either case, and most of the differences in form and
spirit will follow. The Hebrew scriptures became, to use the
rather strange technical term, ' books that defile the hands '.
That is, they were holy : after touching them you must wash

^ one might compare, allowing for differences of date and circumstance,
the exhortations of Lysias and of Isocrates at Olympia, delivered to an
audience representing the whole of Greece.

* For instance, the older phrase ' the Ark ' was expanded by later editors
into ' the Ark of the Covenant ', or * the Ark of the Covenant of Yahweh '.
Now an elementary rule of Hebrew grammar is that a noun in the construct
case (i. e., in our terminology, followed by another noun in the genitive) cannot
have the definite article. Yet these pious correctors did not venture to delete
the article before ' Ark '. They preferred to leave the utterly ungrammatical
phrase Dnan plNH (Josh. iii. 14) or {ib. 17) : Hp. nna \T\^r\,

2760 O


your hands before touching any mundane thing. They were
kept sacred and apart. Their purpose was to be read aloud
accurately letter by letter in the synagogue for the instruction of
the people. If a member of the audience was not interested,
more shame to him. No one dreamed of imputing any blame to
the writings.

But the Greek traditions from the very outset were made into
Lays to be recited by bards for the delectation of the company.
If men were not interested, it was the fault of the bard and his
poems. And in the very earliest times of Greece we meet with
that characteristic and only half praiseworthy Greek institution,
the public competitive recitation. The poems became, in the
Greek phrase, ^TnSeiKTLKd, things of display. The bards who
knew the traditions came to recite at the great games and
gatherings. Each recited his own poems — i.e. those that he
'possessed', not necessarily those that he had composed — and
tried to make them more attractive than other people's. He was
bound, of course, not to violate history too grossly ; not to be
ylreuSrj^y or ' false-speaking ', above all, not to be ignorant. But
he might, by the help of the Muses, tell his audience a great deal
more about the heroes than by any human means he was likely
to know. He might transfer incidents from one legend to
another, he might alter names or disregard times and places,
provided the change really made his poem better and did not stir
his hearers to contradiction. He could work up the known
incidents till they became more and more moving, more edifying,
or more pleasing. An element was thus admitted which leavened
the whole lump, an element which, in' the hands of a less wonder-
fully gifted people, must, one would think, have led to bombast
and vulgarity, but which was somehow stopped when it had done
its maximum of good and was only just started on its career
of evil ; I mean that strange mixed passion known to all artists,
which consists, at its higher end, in the pure love of beautiful or
noble creation, and, at its lower end, in conscious strain for the

admiration of an audience.





One of the last letters which I received from Andrew Lang-
before his death contained the words ' The next thing Homeric
critics will go mad about is historicity '. Far be it from me to
say that the prophecy has been fulfilled, but a fashion has set in
somewhat violently in favour of accepting the poems as histori-
cally valid. This makes a new cross -division among Homeric
scholars. Some of the strictest 'Unitarians', naturally enough,
take the sceptic side and treat the poems as mere fiction, and
some of the most radical analysers talk cheerfully of pre-Homerlc
' documents ' and ' archives ', which it would be a sin to doubt.

Euhemerus, as we all know, started a great movement in
ancient theology by the hypothesis that the Greek gods were all
real men, and the myths about them real facts embroidered and
exaggerated, and some of the champions of historicity apply his
method with the same impartiality to the whole Greek tradition.
They are prepared to believe in the reality of Dionysus, Asclepius
and even Hippolytus, and think that very likely Cadmus did
follow a cow and build a city where it lay down ; why should
not he ? I cannot generally agree with these scholars, but I
think it likely that their method when critically applied may
sometimes yield historical results. They profit, for example, by a
comparison of the Egyptian and Assyrian records and the
researches of anthropologists.^ It is at least possible that the
mythical Amazons represent some real race of nomads who
seemed to be female because they were beardless; that the
Golden Fleece may well come from real fleeces which were used
to collect gold dust in the rivers of Phasis ; that the King of

^ See particularly Myres and Frost, Historical Background of the Trojan
IVar, in K/io, xiv (1915).

O 2


Troy may have had a real golden candlestick with seven branches,
which he obtained from the Assyrians or Hittites ; or even that
the Flight of Danaos with his fifty daughters to escape the sons of
Aigyptos may have been connected with the defeat of the Danauna
by the Egyptians about 1200 B.c,^ But this general problem lies
outside our special task. We have to face a more particular one.
Some learned and distinguished writers, such as Dr. Leaf and
Professor Chadwick, take the view that, while the Greek myths
as a whole are more or less futile and negligible, the Homeric
narrative is largely historical and correct. It is argued that it
fairly represents the central thread of Greek tradition ; that it is
in the main reasonable and in accordance with possibility ; that
it is not violently contradicted by itself or by known facts outside
it; and, lastly, that in most myths, amid all the nebula of
exaggeration, fancy, and mere muddle, there is apt to be an
historical nucleus without which the nebula could not have

To take these points in order, in what sense do the Poems
form the main or central thread of Greek tradition .'* We know
that from about the sixth century onwards Homer formed the
staple of Greek education. Every one knew Homer, and all parts
of Greece accepted him. Consequently, any local tradition
which conflicted with Homer tended to die out, or else to be
trimmed and fitted into consistency with him. So much will be
agreed. Now, the Poems formed the dominant Greek tradition,
not in the least because historical research showed them to be
true, but because they were such good and successful poems.
They constituted iho. fable convenne^ and surely, if ever we have
to choose between the fable convenue and some stubborn frag-
ment of inconsistent local tradition which for some reason
continued to maintain itself in the teeth of the prevailing fashion,
at first blush an historian's preference should be for the latter.
I do not, of course, suggest that it is necessarily true. Hecataeus

^ Problems arise about the definitely Minoan objects which occur in
Homeric and Attic poetry: e.g. sword inlaid with racing horses, Eur. El.
476 ; dog-and-fawn buckle, t 228 ff.; Skylla (dog-headed sea-monster), /x 86 ff. ;
man clinging under ram's belly, t 432 ff. ; snake and eagle, Soph, Ant. 126 ;
of. Evans, Palace, i, pp. 274, 666, 698, 715, The explanation is probably
Kft^jjXia in families or temples plus popular tradition, I see no trace of
Minoan * documents '. See above, p. 168,


found the traditions of the Greeks ' many and ridiculous ', but
those of them which answer the above test deserve more respect
than most, and at least compare favourably with ' the lies of the
poets '. But almost more dangerous than the lies of the poets are,
I think, the devices of the harmonizers. For example, if Homer
says that Hector was buried in Troy, and the Thebans in histo-
rical times are found worshipping his tomb in Thebes, we shall
not, if we are prudent, try to get out of the difficulty by accepting
the explanation that Hector was, of course, really buried as
Homer said, but that later no doubt, — though no ancient writer
happens to mention it, — in obedience to an oracle the Thebans
sent to Troy to dig up his bones. Or again, if the Homeric
Catalogue makes the Boeotians the central force of Agamemnon's
army, occupying the whole of Boeotia, while the rest of Homer
almost ignores their existence and Thucydides says that they
were first driven from Thessaly into Boeotia sixty years after
the Trojan War, we shall not solve that contradiction by accepting
the suggestions of old historians that perhaps a small section
(diroSaa-iJio?) were already in Boeotia before the Trojan War, or
that perhaps they all came in before the War and were all
unexpectedly driven out and returned again. Such hypotheses
are generally figments invented to explain a difficulty, and, after
all, do not explain it.

But then, it will be said, the Homeric narrative is in the main
so reasonable and possible, and the Homeric characters make
such an impression of reality! 'Contrast the Iliad\ says
Dr. Leaf, 'with the ArgonatLtica with its shadowy characters
and its abundantly miraculous incidents.' But surely to argue
thus is to put oneself at the mercy of the story-teller. By such
canons Peiidennis and Resurrection would be historical, and the
Song of Roland not. As a matter of fact, the Argonattiica
contains, as no doubt Dr. Leaf would admit, a great deal of
historical deposit : the passage of the Bosphorus, the exploration
of the Euxine, the discovery of strange tribes and perhaps of
gold dust in rivers. But it is composed in the romantic and
miraculous style, while the Iliad is full of acocppoa-vi^rj and veri-
similitude. Thus Professor Chadwick shows by examples that the
carrying off of a princess was in the Homeric Age a quite possible
cause of war, and invites us therefore to believe in an historic


Rape of Helen. Dr. Leaf has made a careful reconstruction of the
walls of Troy and shows that the chase of Hector round them was
nowise impossible. But the critics who doubted the truth of these
stories never, as far as I know, doubted their possibility. It is
not likely that a Homeric poet engaged in fiction would, with all
the world before him, deliberately choose a fiction that was in-
credible. It is a matter of style and of artistic competence.

' But at least the Homeric story is seldom or never contradicted
by known facts.' Here again, as I have already ventured to sug-
gest, critics have been misled by a point of style ; I mean the deli-
berate care with which the Homeric poets, and indeed most poets
in the classical tradition, avoid committing themselves to details.
We have noticed it in the matter of the shields and the armour.
Scholars and poets innumerable have read the descriptions of the
Homeric battles, and in their mind's eye have clad the heroes in
the most diverse styles of armour without ever coming on a flat
contradiction. That could never happen to readers of Manmon
or The Lady of the Lake. It is much the same with the topo-
graphy. The rivers of Troy, Scamander and Simois, are vividly
presented to our imaginations. But no hint whatever is given of
where they lie, and all geographers agree that the one definite
topographical statement which the poems have been rash enough
to make about them cannot be true. They cannot have run
into one another (E 774). Such slips are exceedingly rare. As
a rule, all descriptions are merely generical. Think of the
innumerable ships that are mentioned in the Odyssey. Most o*
them have their suitable epithets. They are 'hollow', 'black',
' even ', ' rowed on both sides ', ' swift ', and sometimes ' red-
cheeked '. Yet not one, if I remember rightly, has any single
definite quality to separate it from any of the others. It is much
the same with rivers, plains, mountains, harbours, and all else
that goes to make up that beautiful and heroic, but strangely
standardized, world. There are indeed some topographical
details in Ithaca, and a vast deal of trouble they have given.
There is far less detail about the Plain of Troy. Dr. Leaf appeals
with much apparent force to the success with which his topo-
graphical reconstructions can be built up without contradicting
Homer, He thinks It Is due to Homer's geography being exact
and true, and would have us believe In the location even of the


Ford and the Wild Fig Tree. His work is so beautifully done
that it is with real regret that I find myself reflecting that other
travellers have made quite different reconstructions with almost
equal success, at any rate, in avoiding any contradiction by the
text of Homer.

The theory of the historical nucleus surrounded by a nebula
of fanciful or mythical additions seems at first sight attractive.
An Indian god, Nikal Seyn, has grown up with mythical appurte-
nances round the solid figure of John Nicholson, Another has
grown round Sir Courtenay Ilbert, or rather, since the name of
the god appears to be lUabuttabil rather than Illabutta, round
the famous Bill associated with his name. We are told in the
mediaeval life of Alexander the Great how Iskander Khan once
flew on an eagle with the Prophet Elijah to obtain the water of
life. And so, it is argued, there is a prima facie case for
believing in the reality of Achilles, Agamemnon, Helen, and the
rest. But, in the first place, what kind of ' historicity ' would this
be .'' What single true fact should we know of Alexander, Ilbert,
or Nicholson, from the stories cited ? Names and nothing more.
Nay, not even the names, for, after all, it was not Ilbert but the
Ilbert Bill that has taken divine form in the myth, and Iskander
Khan is not quite the same name as Alexandros. But further,
what is nucleus and what is nebula ? The historic Alexander is
not the nucleus which has attracted the Eagle story : that story is
the nucleus which has attracted one famous name after another.
It was told of Gilgamesh, the Babylonian, long before Alexander
was heard of, and of a Sumerian, Etanna, before the days of
Gilgamesh. That is a very important fact in large masses of
stories. Even if an historical name does occur, the nucleus is not
history but fancy. If Hamlet was the name of a real Danish
prince, the story of Hamlet was considerably older than his day.
Most of us know nothing of St. George except that he killed the
dragon. Gibbon, indeed, identified him with a fraudulent army
contractor who supplied bacon to certain forces in Cappadocia,
but Gibbon, it seems, was merely yielding to the seductions of
' historicity '. The truth about St. George is said to be quite
different. There was a very ancient bas-relief at Lydda repre-
senting a hero slaying a gigantic monster. The Christians "
called him George, but the Greeks before that called him Perseus,


while the Babylonians were confident that he was their favourite
Gilgamesh killing Tiamat. There is a canon of Pope Gelasius
establishing the canonization of St. George which prudently refers
to him as one of those ' whose names are justly revered among men,
but whose acts are known only to God '. If that is historicity, we
need not grudge it to Agamemnon. And, further, when once
the goddess of fiction has really set to work, surely the quality
of the original peg or nucleus shows very little in the finished
product. There is a character in Kenilworth called Weyland
Smith. He was originally the Norse god, Wielant, Icelandic
Voliindr, but in Kenilworth he has just as solid flesh and blood
as Tony Foster or Amy Robsart, whose bodies now lie in
Cumnor Church. When Attila the Hun or Guthhere the
Burgundian is drawn into the Saga of the Volsungs and Nibelungs
there is no visible quality to show which of those gigantic cloudy
figures were originally made of mist and which of bone and
muscle. Nay, even where a real person is taken over to fiction
and in the fiction remains vividly real and palpable, what guarantee
have we that the fiction, because life-like, is true to life ? Sir John
Oldcastle, the Lollard, was a man of eminent piety and high
character. The report of his trial for heresy is still extant.^
Yet when he became a favourite character on the Elizabethan
stage (where in Shakespeare's later work his name was
eventually changed to Falstaff ) he was an extraordinary and
shining type of all the most un-Lollardish failings, while innocent
as a babe of the crime of heresy.

Modern instances have largely to be taken from fiction. We
do not in modern times find faded gods turning into heroes,
except in odd cases like Weyland Smith. That is merely
because we have no mythological gods. But the Greeks had
numbers of them, and their mythology, as soon as literature
begins, appears as the commonest source for poetical composition
and the ordinary food for popular imagination. In modern India,
under our eyes, as the great gods and goddesses of the plain
conquer new territory, the local gods whom they supersede are
very apt to become their human votaries or door-keepers.
There is certainly no ground in general literature for denying
that gods may become men.

^ In Pollard, Reprints of Fifteenth Century Verse and Prose.


Bearing these cautions in mind we may proceed to our problem.
In most traditional poems there are three fairly distinct elements.
There are masses of mere fiction — that is, stories and personages
deliberately invented by the poet out of his head. There are,
secondly, the shapes of myth and saga, which the poet narrates
in good faith, as he received them, with at least a modicum of
belief in their reality. And, thirdly, there are fragments of
definite history. Take the Nibelungenlied^ for instance. There
the whole web of the story is woven on lines of romantic fiction.
But many of the characters, the Niblungs and Odin and
apparently Sigurd himself, belong to the region of myth.
Again, we have historical persons in Atli, who is the Hunnish
King Attila, and Dietrich of Berne, who is the real Theodoric.

In Homer we may make the same sort of division. There is,
in the first place, much deliberate fiction. The whole frame-
work into which the incidents are fitted, the wanderings of
Telemachus in the Odyssey^ the Embassy to Achilles in Iliad I,
are evidently mere inventions of the poets. on the other hand,
such beings as Zeus, Hephaestus, Bellerophon, Typhoeus, the
Chimaera, clearly belong to the realm of myth. And, thirdly,
the excavations have proved the historical reality of the great
towns of Troy and Mycenae. As to the persons, it is a different
matter. Professor Chadwick persuasively argues that Heroic
Ages are rich in striking and picturesque personalities, who find
a place in poetry not because of their historical importance but
because of some purely romantic or adventurous interest. If so,
they will obviously become almost indistinguishable from the
heroes of fiction. If there are any Attilas and Theodorics hidden
among the various gods and tribal heroes, there is unfortunately
no independent historical document by which to identify them.

Now, as to the fictional parts of Homer, I do not wish to dwell
upon the value of fiction as indirect history. one might point
out that fiction, to adopt a phrase of Aristotle's, if it does not tell
you what did take place on a given occasion, constantly shows
you what might well take place. And even where the main
subject of the fiction is romantic or marvellous the background
or setting in which it is placed is very likely to be drawn from
normal life. The Cyclops, for instance, is a fictitious monster ;
but his processes of dairy-farming are real and historical. And


that kind of information is sometimes what helps us most toward
the understanding of a far-off state of society. If the Iliad and
Odyssey were all fiction we should still learn from them a great
deal about early Greek customs, about practices of war and of
government, about marriage, land-tenure, worship, farming,
commerce, and, above all, the methods of seafaring. Let any
one read thoughtfully the story which Eumaeus the swineherd
tells of his life in Odyssey o, and then consider how much history
of the life of the Aegean, about the seventh century B. c, he has
learnt from three pages of poetical fiction. In the same way,
even if the main story of the I/t'ad is fiction, it is significant that
the social world in which it is placed is curiously like the real
world of an ' Heroic Age '. Figures like Diomedes, Ajax, and
Odysseus, as they appear in the I/md, wandering chieftains with
a band of followers and some ships instead of a kingdom or even
a fatherland, are typical ; and Achilles and Agamemnon hardly
less so. We have here the greatest of the general differences
between the Catalogue in B and the rest of the Iliad. For the
Catalogue starts with a conception of Greece as a settled country
divided into clans and kingdoms, and has fastened each Homeric
chieftain down to a fixed geographical district.

The historical value of the Catalogue has been recently dis-
cussed in two striking books.^ It is by general agreement an
alien document, written by some non-Homeric author or authors
for a non- Homeric context, and afterwards, with omissions and
additions, incorporated in the Iliad. Mr. Allen, like an advocate
fighting for a client s life, endeavours with great learning to show
that there is no flat contradiction in terms between the Catalogue
and the rest of the Iliad., and no statement which is demonstrably
untrue of an age before our records begin. It is pretty certainly
of Boeotian origin and connected with the school of Hesiod ; and
there is reason to suspect that it had been used in the Cypria
before it was put into the Iliad? But here arises an interesting
question of principle. When it differs from the rest of the Iliad.,
which is more likely to be historically true? The dull list ot
names and figures, with not much apparent reason for existence

* Leaf, Homer and History, 1915 ; T. W. Allen, The Homeric Catalogue
of Ships, 1921. Also Leaf, Troy, 1912, on the Catalogue of Trojan Allies.
^ See p. 179 and cf. Appendix J, note.


except as a record of fact, or the brilliant and moving narrative,
exposed to all the temptations of fiction ? one would say the
Catalogue; and probably this in some sense holds good. It
probably does represent some genuine belief, based on tradition
and perhaps helped out by guesswork, held by a school of bards
in Boeotia about the condition of Greece at the time of the
Trojan War. Certainly it is not a deduction from the Iliad. It
is a deduction from other sources and, as such, an early historical
document of great interest.

Dr. Leaf has shown, in my judgement beyond the possibility
of refutation, that the whole point of view of the Catalogue is
very different from that of the Iliad. The prominence of
Boeotia, the gathering of the fleet at Aulis, the strange Corin-
thian kingdom which is all that is left to Agamemnon after
Argo shas been taken away and given to Diomedes, the large
navies provided by the Arcadians and other inland peoples who
never appear again in the Iliad., are all ' un-Homeric '. Also, we
may be sure that the document was not put into the Iliad early.
It was put in late. The fitting-in is mechanical and shows the
joints : it has not been either re-created by a poet or frequently
worked over by skilled rhapsodes. Of course, to be ' un-
Homeric ', or even to be ' late ', is not necessarily to be untrue.
Yet on most of the points of difference the Catalogue version
does seem improbable.

Truth must have been lost in the course of the Catalogue's
history in at least three ways. The original Boeotian authors
doubtless tended to exaggerate the importance of Boeotia ; the
Athenian reciters, who had the poem in their hands from the
time of the Panathenaic recitation onwards, are shown both by
external and internal evidence to have affected the text a good
deal, more perhaps by omission than by interpolation. Twelve
passages In the Catalogue are known to have been marked as
spurious by ancient critics, most of them for reasons in some way
connected with Athens. Modern critics have detected others.
Again, if one looks for the important places which are 'sup-
pressed ' or belittled in the Catalogue they form a list which is
hard to understand until we ask what places q tactful reciter
would be likely to suppress when competing for a prize before
an Athenian audience ; then it is intelligible : Aegina, Megara,


Thebes, Delphi, and to some extent Salamis and Corinth. Then,
again, the adaptation of the Catalogue to its place in the Iliad
must have involved some changes of statement, made in the
interest of harmony and not of truth.

And, even when the document was intact, before any of these
perversions had begun, what was it ? The other catalogues give
lists of goddesses who had human children, of heroines who had
divine lovers, of ancestresses of various tribes, of rivers, of
Nereids, of lucky and unlucky days. The Catalogue of Ships,
and even that of the Trojan Allies, so brilliantly expounded by
Dr. Leaf, was perhaps based on less speculative foundations,
but can hardly claim an impeccable authority.^

I do not think that we have in the Homeric Poems any
' document ' or ' archive ' which can claim on its own authority
a right to be accepted as true. But I wish at present to deal
with a different question, viz. the origins, historical or otherwise,
of some typical characters or incidents in the Poems. First, let
us take a character of pure fiction. ^ Many might be cited : the
herald 'HnvTLSrj^, the bard ^T^/zioy TepindSrjs, many of the
Phaeacians in ^ ii ff. and the Nereids in ^ 39 ff., with their
transparent names. The most striking, perhaps, is Brise!^, the
maiden who is taken by Agamemnon from Achilles and thus
becomes the passive heroine of the Wrath. She has no father
or mother : no history apart from the one incident for which she
is invented and which is presumably an invented incident; as
before mentioned, she has not even a real name. For Kotire
Briseis only means ' Maiden of Brisa ',^ the Aeolic form of Bresa,
a town in Lesbos, taken by Achilles in the course of the war.
It is worth noticing, indeed, that, like other characters In good
fiction, Brise'is eventually acquired independent legendary life
and even rose to some importance In Chaucer and Shakespeare,
under the name of Cresslda. (Cresslda is the accusative ' Bri-
seida' slightly corrupted, and confused with the name of the
other maiden, Chrysei's.)

^ The Trojan Allies lie along an ancient trade route ; Leaf, I.e.

^ For the following, cf. E. Bethe, Homer utid die Heldensage, from Neiie
Jahrbiicher- f. d. klass. Alt., 1902; F. Dlimniler, //".j-^Y^r, ^Anhang ii to
Studniczka's Kyrene, 1890. Also Bethe's Miirc/iett, My thus imd Sage, and
Radermacher, Die Erziihlmigen de}- Odyssee (Vienna, 1 91 5).

^ Wilamowitz, Horn, U/it., p. 409 ; Tiimpel, Lesbiaka, p. 106.


But in the Iliad Briseis is a shadow, a figment of the poet.
Contrast her, for instance, with a real saga-heroine, Helen.
Helen appears in the Troy legend, but was certainly not created
for it. She dominates other legends as well. We know her
parents and her home. She was an important goddess, a
marriage-Kore, in Sparta. Her temple at Therapnae has been
excavated ; she and her husband Menelaus were worshipped
there together, ' not as heroes but as gods ', and she had sacred
trees and wells in many places.^ In most of her functions she
resembles the other Korai who preside over the affairs of women,
but Helen has certain special characteristics. She is a Swan-
child : ^ her mother may be Leda or may be Nemesis, her father
Zeus or Tyndareus, but she is always the daughter of a white
swan. Her brothers, the DIoscori, Castor and Polydeuces, are
twins and also stars. She and they appear to sailors in time of
storm as those balls of the fire that is called ' St Elmo's '. It
used once to be ' St Helen's '. The chief point in her saga is
quite clear : she is always being carried off by a ravisher. And,
I venture to suggest, the origin of this trait is so clear that one
can hardly understand its having so long escaped notice. A
Spartan bride-goddess was bound to be carried off, since in
Spartan marriages the carrying off of the bride was part of the
ritual. The same set of ideas explains two other peculiarities of
Helen, why she is always brought back from her ravishment and
why she is so much associated with an Eidolon or Image," which
is like Helen but is not the real Helen. In the ritual the Image
of the goddess would probably be carried off, but would have to
be secretly restored afterwards to its place in the temple. We
have no detailed information about the rite at Sparta, but in
Samos, where Hera was the marriage-goddess, her Image was
regularly carried off in the Feast called Tonaia, and then secretly
recovered and brought back.

' Isocr. X. 72 ; Paus. iii. 19, 9; Theocr. xviii. 43 ff. &c.

^ Euripides revels in this white-swan atmosphere. Helen is o/^t/na KVKvotitipov
Ka\\o(Tvvas, Or. 1386. Cf. /.A. 793 ; //e/. 214.

^ The ei8o)\op occurs first in Hesiod : then in Stesichorus, Euripides,
Herodotus, Plato, &c., and plays a great part in the later Greek literature :
see reff. in Bergk, F.L.G., under the Pali7todia of Stesichorus. The old
view that the Eidolon story was a fiction of Stesichorus is clearly untenable.
It is cited from Hesiod and has all the marks of genuine myth. {npcoTos
'Halobos 'EXeJ'Jjf to eiSwXor Trapijyaye. Schol. to Lycophron 822.)


A third characteristic of Helen has also a marked ritual or
sacral appearance. When she is restored, she is always restored
by Twins or something very like Twins. She was carried off
to Egypt by Hermes, to Sidon or else to Troy by Alexander-
Paris, to Deceleia by Theseus or the Apharetidae ; but always
recovered either by the so-called ' Twin ' Atreidae or the genuinely
Twin Dioscori. vShe was also carried off to Mount Parnon by a
robber, and we do not happen to hear who brought her back.
But I should be much surprised if the rescuers were not Spartans
and Twins.

Passing from the rich saga-figure of Helen, let us turn to the
comparative barrenness of Achilles, Apart from the fine psycho-
logical treatment of his character in the later books of the Iliad^
Achilles in saga is little more than a typical heroic Kouros, as
Hermes and Apollo are divine Kouroi. Like all the Kouroi he
is young, swift, tall, and beautiful ; like Balder, Hyaklnthos,
Hippolytus, Adonis, Linos, and many others, he is cut short in
his youth and ritually lamented : in the saga by a choir of
Nereids, in actual life by choruses of women. The odd story of
his dressing in girl's clothes has pretty obviously a ritual origin.
Even his many ' wraths ' or ' strifes ',^ which might look at first
sight like the real memories left by a hot-tempered soldier, seem
only to be derived from the regular ' strife ' of the Kouros with
his enemy, the Summer with the coming Winter, the New Year
with the Old. The real history that we can legitimately extract
from Achilles is, I venture to think, what we may call tribal
history. only we must remember that It was not exactly a tribe
that Achilles led at Troy ; It was a coutitattis^ a mixed band of
clansmen and strangers who followed the chief's fortunes. But
we know from Greek history that these mixed bands, when they
founded cities, generally formed themselves into fictitious tribes
and adopted common ancestors. The men who counted Achilles
as their chief, Hellenes of Achaia Phthlotis, and the strangers
who joined with them, did all that Achilles did. They left their
home on the mainland: they stayed first at Skyros, till they

^ Wraths with Agamemnon, one about Brise'is, one because of a late invita-
tion to dinner ; with Diomedes ; with Odysseus ; with Thersites. All seem
to have some oracular significance, as when Achilles strove with Odysseus
and Agamemnon ' rejoiced in heart because the bravest of Achaioi were at
strife ' — remembering an oracle. {6 78.)


were grown strong : they conquered and occupied Lesbos.
They fought on the Thracian coast. They eventually went
through the Hellespont and Bosphorus up to the Black Sea, and
made settlements which bore the name of Achilles in later ages.
But there is something to be learned from studying the various
places where Achilles was worshipped. The worship in Thessaly
was, we are told, ordered from Dodona (Philostr. Heroictts^
p. 741, quoting an interesting hymn to Thetis). This agrees
well with Achilles' prayer to Zeus of Dodona (JT 233). It is
natural enough, too, that he should be worshipped at Sigeum,
at Skyros, at Mytilene, in the island of Leuce, and that inscrip-
tions should be found at Olbia and Odessos calling him Poni-
arches, 'Lord of the Pontus.' But he had worship in other
parts of Greece too. He was worshipped in Laconia, says the
scholiast to Apollonius (iv. 815), citing Anaxagoras. Pausanias
saw a great Achilleion, or shrine to Achilles, on the road from
Sparta to Arcadia. There was worship at Brasiae ; in Elis ; in
the island of Astypalaea ; probably in Cos, since the Aeacidae
in general had a shrine there. And in Tarentum there were
shrines both to the Aeacidae in general and to Achilles. What
does this mean ? Does it not destroy our conception of Achilles
as a special tribal hero ? No : it only serves to illustrate a point
of cardinal importance for the understanding of prehistoric
Greece, the extreme mobility and the frequent scattering of the
various tribes. It is the natural result of that time when all
Hellas was dvdaTaTo^, ' driven from its home ' ; the time of the
* constant war-paths and uprootings of peoples '. There were
fragments of tribes cast away in the most diverse parts, and
where they were strong enough they carried their tribal gods
with them. The Achaioi, who settled in the Peloponnese and
migrated again beyond it, naturally took with them the worship
of Achilles.

If any one would have a conception of the way in which tribes
and races can be scattered, when in a mobile condition of life, I
recommend him to look at some map of the linguistic stocks of
the North American Indians.^ If the Iroquoian or Siouan or,

^ e.g. Jfelisde Reclus, Giogr. Univ., AmMqiie, ii. 40 f. Or, to take a less
remote parallel, the Germanic tribe called Eruli ' are first mentioned in the
third century A. D., at which time they appear almost simultaneously on the
Black Sea and the frontier of Gaul '. Chadwick, Othin, p. 33.


still more, if the Athapascan-speaking races had been in the
habit of building shrines to their tribal heroes, in what extra-
ordinarily diverse parts of the vast continent we should find the
heroa ! And the Iroquoians would have made the Algonquins
worship him too. The result would completely dwarf any
strangeness which we may at first feel in the scattering of the
shrines of Achilles from Tarentum to Odessos. He remains the
hero or kotcros of the people who followed him from Thessaly,
of whom we can only say in Homer's words that ' Myrmidones
were they called and Hellenes and Achaioi '.

The figure of Agamemnon presents unsolved problems.
Famous as he is, there are comparatively few legends about him,
and most of these belong to well-known mythical types. Yet
the existence of some great federated Achaian expedition against
Troy seems almost as certain as the destruction of Troy itself,
and it presumably had some chief leader. The position of
Agamemnon, in the Iliad,, as a ' king ' with authority over all the
other ' kings ', based not on law or hereditary position, but on an
ad hoc oath which they have all sworn, to serve him till Troy is
taken, seems to correspond exactly with what we should, with
our present knowledge, expect to be true of the Heroic Age.
The Anax Andron looks like real history, whatever his name
may have been. He would be a ' Lord of Ships ', leading the
Akhaiusha and the ' peoples of the sea ' or some offshoot of them.
It is likely enough that his predecessors, with or without their
coinitattiS,, at some period had settled in Phrygia, which would
explain why Agamemnon's family are Pelopidae, sons of Pelops
the Phrygian.

He was a King of Kings. As such he steps into the inheri-
tance of the regular King myths. Especially the old sequence
of Ouranos-Gaia-Kronos, Kronos-Rhea-Zeus, is repeated in
Agamemnon- Clytemnestra-Aigisthos. The Old King is slain by
the Young King, who is helped by the Queen and then wedded
to her : the Old Year slain by the Young Year, while the same
Earth-Mother is wedded to all of them.^ An old beam or trunk
of wood worshipped at Chaeronea becomes Agamemnon's
sceptre (Paus. ix. 40, 11), which even in the Iliad is a thing

^ See my Shakespeare Lecture to the British Academy : Hamlet and Orestes.


of tradition (A 234 ff,). As King of Kings, also, he inevitably
becomes involved in the Human Sacrifice of his daughter,
Iphigeneia. It is regularly the King's daughter who has to be
sacrificed. Naturally, too, he is specially associated with Zeus,
the greater King of Kings : his attendant and sacrificer (Aesch.
C/io. 255, 358 ff.) and in various ways his special charge.
Nay, we find a worship of Zeus- Agamemnon at Sparta, like that
of Zeus-Lakedaimon (Hdt. vi. 56) or Zeus-Pelops.

The peculiar manner of his death seems to ask for explanation.
The legend tells that he was killed with an axe, in his bath, after
being wrapt round and round in an enormous or ' endless ' robe.
Now the word for bath, SpoLTt], also denoted the stone sarcophagi
in which great potentates in pre- and post -classical times were
wont to be buried. Can it be that in the Dark Age or later,
when some of the royal graves at Mycenae were opened by
peasants or robbers, a royal figure was found, its skull perhaps
broken from a wound in battle, wrapped in a rich and royal
winding-sheet and lying in a SpoiTr] ? The finders would weave
a story round it, as they did about the Tomb of Midas or the
bas-reliefs of Sisyphus and Tantalus.^

One is not surprised, if the historical King of Kings was
a chief of the ships of the Akhaiusha, to find some difficulty in
locating Agamemnon's home. The Greek tradition wanted
naturally to give him an empire on land. In the //tad he
generally refers to his home as Argos ; once or twice it is
Mycenae rich in gold ; but his empire is over ' many islands and
all Argos '. True, in the Catalogue Argos proper is taken away
from him in order to provide for Diomedes ; that may well be
a mere mistake by the Catalogue-maker, as Dr. Leaf argues.
But the word Argos itself seems to mean ' a watered plain \ and
in Homer it appears to have at least three meanings. It is the
Argos of Thessaly, the Argos of the Peloponnese, and it is also
a general name for Greece, especially when combined with
Hellas— ar' 'EXXdSa Kal f^eaou "Apyos. The point has been
worked out by Cauer,^ who thinks that Agamemnon was
originally Thessalian. Outside Homer, however, the local habita-

* See Salomon Reinach, ' Sisyphe aux Enfers et quelques autres Damnes'>
Ctdtes,, et Religions, vol. ii, pp. 159-205.
'^ Grundfragen^ , pp. 284 ff.

2760 p


tion of Ag-amemnon admits of little doubt. He and Menelaus are,
quite simply, joint kings of Sparta ; the dual monarchy of Sparta
exactly suits them. The Spartan Syagrus, when it was pro-
posed that Sparta should yield up the leadership of Greece
to Gelo of Syracuse, answered by a hexameter line, pre-
sumably Homeric, though it does not occur in our texts :
• Bitterly would Pelops' son Agamemnon groan ' if any but
Spartans led the host (Hdt. vii. 159). His tomb was shown at
Amyclae near Sparta among the other ancient Spartan kings.
Pindar mentions that he died at Amyclae ; Stesichorus and
Simonides both refer to him simply as King of Sparta.^ And,
curiously enough, there is a passage in the Odyssey which points
in the same direction. It is apparently a quotation from the
Nostoi, or Homecomings, and describes Agamemnon's voyage
home in language that is unintelligible If he is bound for Mycenae
or Argos or Corinth. When he ' was nearing Cape Malea ' he
was driven to sea by a storm ; but why did he ever approach
that dangerous cape if he only wanted to make Argos or My-
cenae.' If he was bound for Sparta it was his natural way.
And the poet, though he has suppressed the mention of Sparta
because the Odyssey in general put Agamemnon in Argos, seems
to let out his secret when he makes Telemachus ask about
Agamemnon's death : ' Where was Menelaus ? Why did he not
help his brother ? ' That certainly suggests that both kings are
expected to be together. Probably, therefore, the Nostoi, like
Pindar, Stesichorus, Simonides, and most of the non-Homeric
tradition, had located the King of Kings of the Sea-People in
the natural position for any supreme commander-in-chief of the
Greek forces ; they made him the senior of the two Spartan
kings. Why the Attic-Homeric tradition removed him from
there we can only conjecture, but it is not difficult to see possible

Odysseus, though now prominent in the IHad^ seems as a saga
figure to have had little or nothing to do with Troy. In the
Odyssey he is mostly a folk-lore hero with folk-lore adventures,
though, of course, one can never be sure that these adventures
have not been attached to a historic name. Some of them, at
any rate, were actually told in the Middle Ages of Iskander Khan.
' Find. Pytli. xi. 32, Ne7n. xi. 34 : Schol. Eur. Or. 46.


But myth has been at work also, and myth of a pronounced and
curious kind. The point has not yet been noticed and needs
a fairly full statement. It is a matter of the solar and lunar

Time has been g-enerally measured by the * eniautoi ', or
repeated circuits, of the moon and the sun, i, e. by the month
and the year. The object of a scientific calendar has always
been to find a period in which the two circuits should correspond ;
the New Moon should coincide with the winter solstice, and Sun
and Moon begin their life together. (As a matter of fact, the
two circuits are incommensurable, but that is a recent discovery.)
The lunar month is twenty-nine days plus a fraction ; twelve
months make 354 days plus some hours ; a solar year equals 364
days and a little over. Various cycles were tried with poor
results. The simplest was a Trieteris, or double year ; next
a Penteteris or period of four years, such as regulated the
Olympian and Pythian Festivals. This period had fifty and forty-
nine months alternately, and came fairly right every second time,
in what was called the Ennaeteris. This ought to have come out
at eight lunar years of 354 days each plus three intercalary
months of thirty days, that is 2,922 days; the same figure as is
given by eight solar years of 365^ days ; unfortunately the
fractions were wrong, and there was an error of about a day and
a half in eight years.

Hence came the greatest effort of ancient calendar- making,
Meton's Eikosieteris, as it was called, or Grand Eniautos of Nine-
teen Years. on the last day of the nineteenth year, which was
also by Greek reckoning the first of the twentieth, the New Moon
would coincide with the New Sun of the Winter Solstice ; this
was called the ' Meeting of Sun and Moon ' {'^vvoBo's 'HXiov kol
^^X-qv-qs) — a thing which had not happened for nineteen full
years before and would not happen again for another nineteen.

Now when did Odysseus return to Penelope ? The date is
given with a precision most unusual in epic poetry. He returned
to Ithaca 'just at the rising of that brightest star which heralds
the light of the Daughter of Diawn ' {v 93). He rejoined his wife
' on the twentieth year ' ; i. e. he came as soon as the twentieth
year came, as soon as the nineteenth was complete (\|a 102, 170 ;
p 327, /? 175). He came at the New Moon, on the day which the

P 2


Athenians called ' Old-and-New \ ' when one month is waning
and the next rising- up' (r 307, ^ 162). This New Moon was
also the day of the Apollo Feast, or Solstice festival of the Sun
(v 156, 258), and the time was winter.^ The ^vyoSo^ 'OSvaaeoos
KOL nrji'eXoTn]^ exactly coincides with the Hvi/0809 of the Sun
and Moon.

There was an ancient Wise Man, Cleobulus, who showed his
wisdom by making conundrums. one, recorded by Diogenes
Laertius, runs thus : * one father, twelve children ; each of them
with thirty children, partly black and partly white ; and though
immortal they all die.' The answer is ' The Year ', or rather the
Eniautos, with its 360 day-and-night periods, which all pass.
Can we be surprised to learn (^ 20) that Odysseus had just 360
boars, and that one of them died every day ? Or can we any
longer neglect the other solar characteristics that seem to cling
about Odysseus : that the Sun is his rival and enemy ; that he
goes under the world in the West, visits the realm of the dead
and comes up in the extreme East, ' where the Daughter of
Dawn has her dwellings and her dancing-floors and the Sun his
uprising ' (yn 3) ; that he is brought back home asleep, in a magic
boat, like the Sun in Mimnermus,^ by Phaiakes ('Dark ones? ')
who do even the furthest journeys in twenty-four hours (77 326) ;
that he lives in an island in the sea, ' low-lying, apart from others,
furthest of all towards the sunset' (i 25 f.), a description that can-
not be twisted so as to suit Ithaca; that he is a Far-darter of
arrows ; that his death comes to him out of the sea ; and that,
like most Year-kings, he is doomed to be slain by his son ?

To turn to another type, let us consider one of Achilles' parti-
cular enemies ; to wit, Thersites. Every reader of the I/t'ad
remembers his brief and inglorious appearance in B, where he
rails at Agamemnon with unseemly words, and is thrashed with

* $ 457 vv^ (TKOTOfjirjvios, Odysseus freezing in his rags: p 25 there is frost;
Odysseus cannot face the morning cold ; t 64, 319 a great deal about piling
up the fire; (^ow? efxev r]3e depfadai; cf. a 332. Wilamowitz, //. U., p. 87;
also 114. Cf. Carl Fries's Stiidie7i zur Odyssee (}-,€v^z\g, 1910, 191 1): also
Sseck, Qtiellen, pp. 53 fif., 265 fif. on Ithaca cf. A. B. Cook, Zeus, \, p. 328.

^ The cattle of the Sun are in seven herds of fifty each, total 350. Hippo-
crates divided the year into seven seasons (Galen, vol. xvii, I, p. 19; Kuhn,
Cojufn. in Epidem. 7), and Aristotle recognized the 350 as the days of the
lunar year. See Eustathius, ad loc.


a staff by Odysseus. He was the ugliest man in the Greek army,
bald, and hump-backed, with one leg longer than the other.
Let us remember that ; and then notice what Odysseus threatens
to do with him. He will strip him naked and drive him away
from the company of men (dyopfjOeu) with blows. Does it not
remind one at once of the pharmakos or scapegoat, the ugliest
man in the community, who was made into a sin -offering and
driven out from the city ? But let us look further.

The name Thersites has all the appearance of a fiction. It is
derived from Thersos^ the Aeolic form of Sapao^, ' courage ' or
' impudence '. And the poet of B evidently meant the name to
have this latter meaning. It is rather a surprise to find that
Thersites is really an independent saga-figure outside Homer
with a life of his own and very distinguished relations. He was
a son of Agrios, the savage Aetolian king, and first cousin once
removed of the great Diomedes. His mother was Dia, a name
which suggests a goddess. Returning to Homer, we find that
Thersites was {B 220) ' to two of the Greeks especially most
hateful, Odysseus and Achilles '. Odysseus' enmity needs no
further explanation : Odysseus beat him. But why should
Achilles be his enemy ? Because Achilles, in the ordinary story,
killed him. It happened in this way. When Achilles was
fighting with Penthesilea the Amazon, and had given her a
mortal wound, he was suddenly struck with remorse and love as
he looked upon her dying face. Thersites saw this and grossly
jeered, so Achilles very properly slew him, some say by a spear-
thrust, others, by a heroic box on the ear. He was purified for
this manslaughter by Odysseus. Diomedes, however, Thersites'
kinsman, took up a feud against Achilles in consequence.^

Another story is given in the old chronicle writer Pherekydes
(fr. 82) and the poet Euphorion (fr. 131). Thersites took part in
the hunt of the Calydonian boar, and, tor showing cowardice,
was thrown by his cousin Meleager over a rock. (He is made to

^ So the Aithiopis : followed by Chairemon's tragedy, Achilles Thersito-
ctonos. Hence Apollodorus, &c. Cf. Istros about the man called Pharmakos
who was stoned vttci twj/ Trept tov ^ kyChXia, Harpocrat. s. v. For the feud
with Diomedes see Tzetz, Posthom. 199, 206 : also Lycophr. 999, Quint.
Smym. i. 767 ; Schol. Soph. Phil. 445, Dictys Cret. iv. 3. This late
Latin book goes back to ancient sources. An earlier Greek version of Dictys
has lately been discovered, dating probably from the second century A. D.
Tebtunis Papyri.


recover, much injured, in order to be slain again by Achilles.)
Throwing from a rock, it may be remembered, was one of the
regular modes of getting rid of a phannakos.

The evidence so far points towards some connexion with
a human sacrifice of the pharinakos type, that is, a purgative
sacrifice to cleanse the community ; also to some special con-
nexion with Achilles. Can we take it a step further ?

Professor Usener, the author of that stimulating book,
Gotternainen^ points out a more strange coincidence.^ Thersites
is found as a name elsewhere in Greece : and derivatives of the
same stem are common, Thersias, Therson, and the like. Now
in the Lacedaemonian ^ dialect this word would probably take
the form Theritas, Qrjpira? : as U^pa-e^oveia becomes in
Laconian, UTjpeipoueta. And Theritas in Lacedaemon is a god
of whom we know something. Pausanias saw his temple
between Amyclae and Therapnae. Pausanias says that he was
the same as Ares ; Hesychius, perhaps more accurately, says he
was Enyalios— another war-god. He had a nurse — or mother —
called Thero.

Now the old sacrifice of the human scapegoat had in Sparta
died down to a curious form, to which, however, there are many
parallels elsewhere. It became an annual fight in a plane grove
between two bodies of Ephebi, or Spartan youths. They fought
with no weapons : only fists and feet. The plane grove was
surrounded by a moat, and they threw the defeated, if they could
manage it, into the water — another regular way of disposing of
the phannakos? And before this annual batde the Ephebi
^^cioxva^'^ a sacrifice to Enyalios at a place called the Phoibeion,
and a sacrifice to Achilles at his temple on the road to Arcadia
(Paus. iii. 19. 7; 20. 2 ; 20. 8; battle of Ephebi, 14. 8). If
Enyalios is Theritas, as Hesychius tells us, we have here the
ritual form of the old batde of Achilles and Thersites. What
that battle in its primitive religious significance really was lies

^ Der Staff des gr. Epos, in Siizungsber. Wiener Akad., phil.-hist. Kl.
1898, p. 47.

' In strict Spartan ^rjpirai. erjpiras would be the Doricized 'Achaian'
dialect of the Perioikoi, if Meister is right. See his Dorer und Achder,
pp. 24 ff.

^ In the Thersttes-Penthesilea story in Dictys, Diomedes has Penthesilea
thrown, still living, into the water.


beyond our scope. Usener thinks of the common annual rites of
the slaying of Winter by Summer, or of one vegetation god
by another.

Different, again, is a hero like Telamonian Aias. He has no
tribe, no home, no belongings. only a shield which no one else
can bear, and a father whose name is Telamon, ' Shield-strap'.^
The lines connecting him with the island of Salamis are of the
latest description. But he has another characteristic. Himself
an immense man and fabulously strong, he constantly goes about
with a companion, as brave as himself but small. The two
together are called ' Aiante ', ' the two Aiases '. The name of the
other varies. As the Utad now stands, this companion is gener-
ally Aias's half-brother, by name Teukros : sometimes he is Aias
the Less, a Locrian and son of Oileus.^ These persons require,
of course, separate study. one of them at least, Teukros, seems
to be a real saga-figure. But, like the more shadowy son of
Oileus, he has been pressed into service as the Great Aias's lesser
twin. The Aiante are figures of folk-lore, and no doubt of
primitive worship, parallel to the other sets of divine twins, the
Tyndaridae, the Aphareidae, the Dioscuri, the Anake, the
Leucopolo, the Aktorione Molione. It is worth noticing that
Pick considers this twin-worship as characteristic of the Leleges:
Salamis and Locris are both Lelegian centres. And the name
Oileus is referred to the Lelegian language.

Take again the case of Diomedes. He seems to be a tribal
god or hero, connected with Aetolia and the Aetolian settlements
on the north coast of the Peloponnese, though in the Catalogue
he belongs to Argos and Epidaurus. Originally perhaps an
Achaean, he has been affected by association with these wild
Aetolian tribes, who came from Illyria and expelled the Achaeans,

' It has been suggested by P. Girard, Rev. des Etudes Grecques, xviii.
(1905), pp, 1-75, that TfXo^Licoi^ (' Supporter'), as the father of Aias, is origin-
ally not a shield-strap, but a door-post or pillar. This is good in point of
religion, and would suit excellently with the conception of the Aiante as twins ;
and an inscription (fifth cent.) from the Argive Heraeum uses reXa/icoi' as =
'pillar'. It is also a Roman use — 'Telamones', like 'Caryatides'. See
Herwerden, Lex. Stcpplct. But to the writers of the Iliad Aias is obviously
a shield-hero.

"^ This suggestion was first made by Wackernagel, H. U. 247". Cf. Eur.
/. A, 192 KQTiibov fie hv A'iavTi avvidpo), top OlXiws TeXa/^&i/oy r€ yovov.


reducing AetoHa in historical times to savagery. His kinsman
is ' Agrios ', ' Savage '. His father Tydeus would have been
made immortal, owing to his many merits, had not his own tribal
war-goddess, Athena, seen him eating an enemy's head on the
battle-field, and after that preferred to let him die. However
that may be, we find in Greek tradition two ostensibly distinct
persons bearing the name of Diomedes. There is this hero,
mentioned in the Iliad and Odyssey^ the Epigoni and the
Alkmaeonis, by birth an Argive, but a traveller to Aetolia, Troy,
Italy, and Cyprus. He is a fierce and fiery young warrior, much
associated with horses, but decidedly, if I may borrow a con-
venient phrase from the language of the theatre, ' sympathetic '.
That is to say, we are wont to be on his side, not on that of his
enemies. But there is also another ' unsympathetic ' Diomedes,
a ruflSan and a savage ; a son of the Thracian war-god Ares, and
King of Abdera in Thrace. This Diomedes, who fed his fierce
white horses with human flesh — an evident trace of human
sacrifices — was suitably destroyed by Heracles, and his horses
taken away. Now, as Eric Bethe has pointed out, these two
heroes are probably in origin the same. As soon as you
scratch the Argive Diomedes you find under his Hellenic surface
the mark of the Thracian. In the most diverse localities we find
him connected with the same horses and the same uncanny sacri-
fices. In Cyprus to the far south-east he was worshipped with
human victims. To the far north-west the Venetians sacrificed to
him white horses. In the Iliad Diomedes has been cleared of
his cannibal tendencies, and is left one of the most attractive
figures in the poem, peculiarly brave and modest and wise in
counsel. Yet incidentally we are constantly coming across his
Thracian connexions. In K he slays the King of the Thracians,
Rhesus, and carries off his famous white horses. In E also, I
would suggest, he fights and routs the god of the Thracian
aborigines, Ares : Ares flies to heaven, leaving no horses behind.
But we find that, just before, Diomedes has fought Aeneas and
his mother Aphrodite, and carried off Aeneas's matchless horses.
Aeneas is the son of Aphrodite, and Aphrodite is the goddess
belonging to Ares. Originally, it would seem, a war-goddess
and wife of the war-god, she has passed through the crucibles of
Greek imagination, and emerges identified with a half-oriental


love-goddess, a creature who has no business in battles, and is
merely the paramour of the warrior god (see Schol. on ^ 416).
Also her son in this case has Anchises for his father, not Ares.
This probably is the result of mythological changes and poetic
adaptations. one suspects that originally the hero conquered by
Diomedes, and robbed of his horses, and immediately afterwards
succoured by both Aphrodite and Ares, was a true son of Ares.
Thus the story of Diomedes in E becomes an exact parallel to
that of Diomedes the Thracian tyrant. For, in the processes ot
ancient mythology, to conquer a son of the Thracian Ares and
despoil him of his matchless horses is exactly the same thing as
to be a son of the Thracian Ares who is so conquered and
despoiled. In the one story Diomedes has the passive part, in
the other the active. It is like Dionysus the Bull- Slayer, and
Dionysus the slain bull ; Apollo the wolf, and Apollo the averter
of wolves.

So many and various are the elements of saga and tribal
history which have taken shape in the heroes of the I/md. Of
course we may admit freely the possibiHty that in any particular
hero there may be traces of a real individual. The legends ot
the Middle Ages are full of historical names. And the names
Paris or Hector or even Agamemnon may have belonged
originally to as definite a person as those of Charlemagne or
* Virgil the magician ', Attila or Dieterich of Berne. Professor
Bury has remarked that the name and personality of a great
foeman are apt to remain fixed in a nation's memory. Had
nineteenth-century England been still in the saga-making stage,
she would certainly have mingled ' Boney ' with her ancestral
demi-gods. But, if any of the persons are historical, we cannot
identify them. And if the names are real, it does not follow that
any part of the story really happened to the bearer of the parti-
cular name. None of the mediaeval magician-stories happened
to the real Vergil.



In the spring of 1923 a man brought to the experts of the
Natural History Museum, South Kensington, a strange object
which he wished identified. * It looks,' he said, ' exactly like
a petrified mammalian stomach ' ; and so it did, with a well-
marked pylorus and other details. ' You may be sure,' said the
expert, ' that if it was a petrified mammalian stomach it would
not look like one.' It proved in fact to be a stone ginger-beer
bottle which had collapsed before baking.

I think that, in the search for historical facts among the poetical
traditions of Greece, the warning of that expert should be borne
in mind. Any real historical fact which was in the poetical
tradition by 1000 B. C. would pretty certainly be transformed out
of all likeness by the time it had passed through the hands of
several hundred bards and formed a ficelle in innumerable
different poems. Of course we often find that new discoveries
' confirm ' or at least ' explain ' some epithet or phrase in the
poems of which readers had taken no heed. The ' horse-taming
Hector ' and the Trojans ' mounters of swift horses ', the sack of
Troy by Heracles eVe/c' 'iirTTdnv Aao\ik8ovTos, receive new light
when we learn from Professors Myres and Peake that Ilion II
was very likely destroyed in the third millennium by an invasion
of hordes from the steppes of Russia, who had just discovered
that they could ride horses as well as drink the milk of mares.
The horse-folk no doubt established themselves on the plain of
Troy, and Ilion remained for long a special centre of the horse
industry.^ It throws light on Hector's war-cry

Tpcoes Kol AvKLOL Kol Adp8avoL dy\i/xa)(T]Tai,

^ See F. Sartiaux, Troie^ 1915 : Peake, Racial Elements concerned in the
First Siege of Troy, 1916 : Myres, Catnbridge Ancient History (1923), p. 82 :
and especially Macurdy, Troy and Paeonia, which I have been privileged to
see before publication.


when we are taught by Professor Macurdy that the Dardans and
other Thracian tribes in historical times seem to have origin-
ated the dense infantry formation known as the Macedonian

In these cases history is preserved in the implications of an
epithet ; it may equally be preserved in myths or proper names,
without the knowledge of those who told or spoke them. What
does not seem to occur is a plain statement about the act of
an individual or tribe in which both the names and the facts are
unchanged. So much we have seen in several of the examples
taken in the last chapter. If we throw our net rather more
boldly and look not for the direct record but the indirect
evidence or implication, we shall have a better prospect of
obtaining some tangible catch. As so often in these problems
of early history, we may be able to divine not indeed the exact
thing that did happen, but the sort of thing that must have

Consider the historical background of a case like the following.
There is a fine passage of some seventy lines in Iliad E 627-98
which narrates the slaying of Tlepolemus of Rhodes, a son of
Heracles, by the Lycian Sarpedon. The passage interrupts the
context. It is never referred to afterwards. The Heracleidae
are nowhere else mentioned in Homer. And for divers reasons
editors have marked the passage as a foreign insertion. But
where does the insertion come from } The Heracleid of Rhodes
has no place in the Trojan circle of legends. When one sees
that his adversary is a Lycian, that is, a chieftain of the mainland
just opposite Rhodes, where the Rhodians were constantly
attempting to force a settlement, one can guess what may have
happened. A local legend of batde between the Rhodian and
the Lycian has been torn up from its natural context and
inserted into the midst of the fighting about Troy. The song is
a fragment of the history of Rhodes and Lycia.^

* The Sarpedon passages generally bear the marks of being in some sense
foreign matter, either invented later or transferred from a different context.
For instance, the Sarpedon who was buried in the famous grave-mound in
Lycia must have been slain in Lycia, not in Troy. This was remarked in
antiquity. The passage (n 668 ff.) where Sarpedon's body is carried from
Troy to Lycia by Sleep and Death was considered ' spurious ' by Zenodotus,
with whom Didymus agreed {SchoL ad /oc). Eustathius also (p. 1069, 29)
makes the very plausible surmise that Homer, knowing of the historical


In searching for fragments of real history, like this, in the
Homeric poems, it is probable that our best hunting-grounds
will be in the little backwaters of narrative, where the plot
interest is weakest and the details least important. That is to
say, the poet will have left the history most unchanged in those
places where he had the least motive to falsify it ; and con-
versely. In the case of Diomedes which we have just considered,
for instance, the narrative is in the front plane of interest.
Consequently the original story — if we were right in our sugges-
tions — is hidden away under a mass of ornament and addition.
Not only has the place of Diomedes' battle with the war-god and
his spouse been moved from Thrace to Troy, but the name of
Aeneas has been substituted for some other name. When a story
is mere background, and does not need to be made interesting, it
is less tampered with.

In the same book (43 ff.) we have the following passage : —

Then Idomeneus despoiled Phaestus, son of the Maeo-
nian, even of Borus who had come from deep-soiled Tarne.
Him spear-famed Idomeneus stabbed with his long lance as
he was about to mount upon his chariot, through his right
shoulder. And down from the chariot he crashed and
a horror of darkness laid hold on him.

Idomeneus is the King of Cnossos in Crete, and Phaestus is
otherwise only known to history as the next most famous town

grave-mound in Lycia, invented these lines in order to combine it with his
own story that Sarpedon was killed at Troy. It has long since been suggested,
on other grounds, that influence was perhaps exerted on the Iliad by the
princes of Lycia, who derived their descent from Sarpedon and Glaucus.

Sarpedon, however, seems to have Thracian connexions as well as Lycian —
even if the latter are not entirely an invention of the said princes, who may
well have identified a native ancestor of their own with the famous Sarpedon.
A promontory near Ainos in Thrace was called Sarpedon (Strabo, p. 331, fr.
52 ; cf. p. 319), and Ainos is the home of Sarpedon in one of the Heracles
legends. Ainos was an Aeolic settlement among Thracians ; hence Sarpedon
is the blood-foe of Patroclus. His chosen comrade, Asteropaios (M 102 f.)
is a Paeonian, son of the river Axios (Bethe, 1. c). His other comrade,
Maris, is otherwise unknown, but suggests Maron. Glaucus himself, one may
observe, is guest-friend of the Thracian Diomedes : but Glaucus is a figure
with many ramifications. See Macurdy, I. c.

One may notice, as a further mark of something unusual, that the Lycian
genealogy given in Z 199 does not agree with the one ordinarily given, from
Europa-Minos. And Diodorus says that Sarpedon fought on the side of
Agamemnon against Troy ! (v. 79). Perhaps a mere slip.


in the same island. That is to say, Phaestus zs the town, or the
eponymous hero of the town, and he is asserted to be a ' son '
of the Maeonian or Lydian of the mainland. It reminds us
of Zidon the firstborn of Canaan : apparently this was some
Maeonian settlement at Phaestus. We may well have in this
passage a record of a local battle or conquest in Crete, torn up
from its surroundings and used by the poet to fill in some details
of slaughter in a great battle before Troy.

And what sort of a conquest was it ? Idomeneus, if we inquire
into his antecedents, appears pretty clearly as a northern invader
of Crete. He is a son of Deucalion, which points to Thessaly.
He is a great founder of cities in the north-west, like Diomedes
and Odysseus. The men he fights fall into two groups : ^
Oinomaus and Alkathoiis — who is in some legends one of the
suitors slain by Oinomaus, in others a son of Pelops the slayer of
Oinomaus — these two take us to the Pelops-group of invaders in
the Peloponnese. The others are what we may call Creto-
Asiatic ; Asius, from the Asian plain in Lydia, this Phaestus, son
of the Maeonian from Lydia, and Othryoneus, a name derived
from the Cretan word for a hill {6dpv9, see Fick-Bechtel, p. 421).

Is there not history here, real history, however fragmentary
and adrift from all its moorings ? I think, following a hint of
Bethe's, that there is very likely a good deal of historical fact
contained in certain passages which look at first sight like mere
strings of meaningless names, I mean, the avSpoKTaatai, or ' Man-
slayings ', which constantly fill up the background of a Homeric
battle picture. For instance, at the end of Diomedes' great battle
we have {Z 29 ff.) this passage : —

Then Polypoites, firm in battle, slew Astyalus, and Odys-
seus smote with his bronzen spear Pidutes of Percote,
and Teucros godlike Aretaon. And Antilochus, son of
Nestor, smote Ablerus with his shining spear, and Aga-
memnon, king of men, slew Elatus. (He dwelt by the
banks of fair-flowing Satnioeis, in lofty Pedasus.) And
Eurypylus despoiled Melanthius. And Menelaus caught
Adrastus alive.

And so on.

' I omit Aeneas and Deiphobus. They are obviously not inconsistent with
the above grouping, but I hesitate to offer an explanation of their meaning in
this context. Orsilochos, Idomeneus' supposed son in v 260, looks like a fiction.


There may be fiction, and the emptiest kind of fiction, mixed
up in this. And probably most of the history is at present
untraceable. We will take one case in detail presently. But,
first, let us reflect what constituted a man's chief claim to public
honour among these primitive northern tribes. The greatest
thing, perhaps, was to be Ptoliporthos^ a Sacker of Cities.^
Short of that, a hero was chiefly known by the enemies whom
he had slain. A curious remnant of this interest is to be found
in Hyginus, the mythographer, chapters 1 1 2-1 15. ' Qui cum quo
dimicarunt ', ' Nobilem quem quis occidit ', ' Achivi qui quot
occiderunt ', and so on. It is a well-known feature of heroic
societies. Think of Sigurd Fafnirsbane, Hogni Sigurdsbane,
and the rest. Think of the stories of Heracles, Achilles, Dio-
medes. In each case the main groundwork is a list of the enemies
whom the hero slew. In more civilized times we put on the tomb
of a general a list of the victories which he won. In earlier
times these victories were, or at least were represented as being,
personal duels, man-to-man, and were commemorated, at any rate
in times of migration, not by inscriptions on tombs, but by paeans
or verses current among the tribe. one remembers how the
Myrmidons in Iliad X march back to the ships singing their
paean : ' We have won us great glory, we have slain godlike
Hector, to whom the Trojans in their city prayed as to a god.'

The emotion connected with these various victories would of
course generally become dim with time, but the verses recording
the bare facts would be remembered carefully by the bard of the
tribe or group. Indeed their preservation would be the chief
part of his business. And I strongly suspect that the lists of
men slain by the various heroes in the I/md are, in their origin,
these same quasi-tribal records, condensed into mere lists of
names and, of course, transferred from their original contexts.
In detail fiction may have entered in, and some names may be
pure inventions. But in general, if we only interpret the lan-
guage rightly, I incline to believe that ' Odysseus ' did very likely
slay ' Pidutes of Percote ', and that some people claiming con-
nexion with Agamemnon did very likely take the town of Pedasus
in the valley of the Satnioeis. This last point, indeed, we actually
know from history.

^ Cf. Aesch. Ag-. 472 and Cic. E/>, ad Fatn. x. 13. 2.


But let us follow the story of the last victim in this 'Man-
slaying', Adrastus, who was taken by Menelaus alive and
eventually slain by Agamemnon. He appears suddenly, with
no name of father or country. But his fate is told at length.
His horses took fright, ran into a tamarisk bush, broke the pole _
of his chariot, and flung him out upon his face. So Menelaus
took him, but would have spared his life had not Agamemnon
run up and himself stabbed Adrastus in the flank with his

Who is this Adrastus, and where was this battle really fought ?
The name (' No-escape ') is common in myth. It recurs in JB 830.
But we may notice that Pausanias saw a place near Thebes which
was called Harma, Chariot ; and when he inquired the origin of the
name, he was informed that Adrastus, the celebrated King of Sik-
yon, Leader of the Seven against Thebes, was there cast out of his
chariot, which was entangled in a tamarisk bush, when he was flying
from the battle at Thebes. This cannot be entirely a fabrication
based on the Iliad. It is, in part at least, an independent tradi-
tion, and we can make a shrewd guess at its source. Adrastus
was the leader of the Argives in the T/tedais, and his defeat and
flight one of its crowning incidents. We have found the Iliad
using the Thebais before. (See p. 181 f.) And again, when
Agamemnon's kingdom is described in the Catalogue (B 572)
it includes ' Sikyon, where aforetime Adrastus was king '. That
is, this fatherless and floating Adrastus seems — though the
reciters of the Iliad as a rule did not suspect it — to have been
originally the great Adrastus of the Theban War. And what of
his slaying by Agamemnon ? Does it represent some misty
tradition of a real battle ? one would think so ; but there is
still the likelihood that it may be a mere fiction. In any case, if
there is any real history behind it, that history did not take place
at Troy.

It seems quite possible that few of these battles of the Iliad
did. A line of research Indicated by Eric Bethe in a brilliant
essay on Die Trojanischen Sagenkreise tends to establish clearly
what many of us had suspected before, that much of the fighting
which Homer locates at Troy, in Asia Minor, on the south-
eastern shore of the Hellespont, Is really a reminiscence of old


mainland wars, whose legendary centre, I would suggest, was
once not Troy but Thebes. Dr. Bethe's method is this.
Those heroes who have a real existence in the tradition, apart
from the Iliad^ can in many cases be traced to their diverse
homes or settlements by three trains of evidence: first, their
graves and places of worship ; secondly, their blood-feuds, for a
tribe's blood-feud is usually against a close neighbour ; and
thirdly, their wives, kinsmen, and the like.

Take the case of Achilles. It is quite clear. Achilles is
firmly located in Phthia, in the country between Thessaly and
Locris, on the Sperchelos river. All his kindred are about
him. The temple of Thetis, his mother, is close to Pharsalos.
His father Peleus is associated with Mt. Pelion. His sister was
married to the river Spercheios. And in the same neighbour-
hood we find his blood-foes. Two heroes, celebrated in other
contexts, but in the Iliad reduced to mere names for filling up
an ' androktasia ',^ Dryops and Deucalion, belong to this region.
So does his better-known enemy Cycnus, the Swan-hero. More
than that, there is quoted from the third-century historian Istros
a statement which puzzles Plutarch and directly conflicts with
all the Homeric tradition, that Alexandros or Paris was slain
by Achilles and Patroclus upon the banks of the Spercheios.

In Homer, of course, Alexandros is a Trojan prince who
perhaps never saw the Spercheios in his life, and he is not killed
by Achilles, but on the contrary kills him some time after
Patroclus is dead. It is startling to find him fighting on the Locrian
border. Yet an inquiry into the origin of Alexandros-Paris gives
him also a home in the same region as his enemy, Achilles.
His sister, who like himself has a double name, Alexandra
or Cassandra, was worshipped in historical times in Locris.
(The Locrians had some strange connexion with Ilion. As
late as the fourth century B. c. they supplied periodically two
highly taboo priestesses to the temple there. The natives always
tried to prevent them coming in, and once killed one of them.^)
The heroes with whom Paris fights in the Iliad^ especially those
who have no importance in the story, and are therefore not
inserted for a fictional purpose, are almost all Thessalians, such

» Y 455, 478.

^ Timaeus ap. Schol. Lycophr. 11 55, 1 159: Aeneas Tacticus 31. 24.


as Machaon, Eurypylos, Menesthios.^ He is killed at last by the
Malian Philoctetes.

Andromache, the wife of Hector, comes from Thebe, a town
which is described as 'TTroirXaKiri, or in words which explain
that epithet, ' beneath wooded Placos '. No one in antiquity
knew what or where Placos was, though it was presumed to
be a mountain. Was it not the mountain above that Thebe
which lies between Pharsalus and Mt. Pelion, at the northern
boundary of Achilles' realm ? Andromache in one passage of
the I/md {Z 397 ff.) is made a Cilician ; but in the saga generally
she is connected not with any place in Asia, but with the north
and north-west of Greece. She is the mother of Molossus, the
eponymous hero of a tribe in Epirus called Molossi. In another
legend she is the mother of Kestrinos, eponymous hero of the
Epirot territory Kestrine. This seems to be the real tradition.
It is then harmonized with the Troy-poems by making some one
bring the Trojan queen back to Greece after the capture of her
city. In one legend it is Neoptolemus, the son of Achilles, who
so brings her. In another it is Helenus, her brother-in-law.

And what of Hector himself, the great ' City-holder ' ? Did he
always in legend hold the same city ? As Dummler has observed,
he was worshipped as a hero in Boeotian Thebes. And if we
examine the list of people whom he kills or fights in the Iliad^
their cults and graves and legends crowd round the neighbour-
hood of Boeotia. Leitos (P 601) had a tomb in Plataea : Oresbios
(E 707) lived in Hyle : Arkesilaos (O 329) was buried in
Lebadeia. As for Hector's comrade, Melanippos (O 547-83),
we know that a hero Melanippos was, like Hector himself,
worshipped in Thebes. Hector was a great ' slayer of men ',
and his victims in the Iliad make a sort of road from Thebes
upward to the bounds of Achilles' region. Dr. Bethe mentions
Schedios the Phocian, whose tomb Strabo saw at Daphnus on
the Euboean Gulf (O 515, and again P 306: Strabo, ix. 425) ;
Autonoos, worshipped as a hero at Delphi {A 301); Orestes,
connected in saga with Phocis {E 705) ; Trechos the Aetolian,
who must be the eponymous hero of Trechis (E 706). Trechis
lies at the mouth of the Spercheios on the borders of the realm

* Cf. A 506 ; A 580, B 736 ; H 9.

2760 Q


of Achilles. Patroclus, Hector's greatest victim, belongs to the
heart of that country. Further north he slew Helenos, son of
Oinops (E 707), Epeigeus from the town of Boudeion (IT 571),
and in some legends also Protesilaos. The road has led us even
beyond the blood-foe Achilles, up to Thebe, the city of Hector's
wife Andromache. It is strange. It looks as if forgotten
remnants of old Boeotian saga, or even epos, omitted from
the canonical Thebais, which concentrated on the war of The
Seven, were used for building up the plot of the ' poetry about
Troy '.

Another group of closely united enemies — in these connexions
neighbour and enemy are almost interchangeable terms — is to be
found in Lacedaemon. If the above was the Achilles- Hector-
Alexandros group, this is the Helen group. It consists of Helen,
Agamemnon, Menelaus, Alexandros the ravisher of Helen, and
Deiphobus her third husband. Alexandros, it will be seen,
appears in both groups,^ Helen of course lived in Sparta. Her
husband Menelaus had a grave and a temple at Therapnae : and
at the same place, according to the statement of a late though
well-informed authority,^ both Alexandros and Deiphobus
received divine honours. Perhaps in this statement Therapnae
is a mistake for Amyclae, which suits the geography slightly
better. Also Amyclae is the home of a Deiphobus in the
Heracles legend (Apollod. ii. 6. 2 ; Diod. iv. 31 ; Jahn, Bilder-
chronik^ p. 70) : and in Amyclae also lay the sanctuary of
Alexandros's sister Alexandra-Cassandra, and beside it her tomb,
together wath that of Agamemnon.

I will not pursue the subject further. one may well be surprised
at the tenacity with which these ancient local worships held their
ground through almost the whole lifetime of Greece as a nation.
The tribes which instituted them, and through which alone they
had reality, had long since passed away both from those particular
neighbourhoods and from the face of the earth. They were
often in flat contradiction with that other stream of history
popularized and made canonical by the Iliad and Odyssey. At

' Cf. Agamemnon, of whom the same is true : both Agamemnon and
Alexander are famous as ' Lords of Ships'.

''' The dialogue ' Theophrastus ' on the immortality of the Soul, by Aeneas
of Gaza (fifth century A. p.), cited in S. Wide, Lakonische Kulte, p. 351 ;
Bethe, 1. c, p. 16.


a time when all educated people throughout Greece knew from
their childhood the Helen of poetry, the fairest of mortal women,
the gentle adulteress whose sin is almost forgiven because of her
graciousness, old peasants and pietists continued to worship her
rude and ugly idol at Sparta, praying for happy marriage or for
the health of sick children. Sometimes the two streams of
legend, that of the Iliad and that of the local worship, ran on
without mingling ; more often, of course, ways were invented for
harmonizing the two. That is why, for instance, Cassandra is
brought from Troy by Agamemnon, to be buried beside him at
Therapnae ; why a Locrian hero is made to commit a sin against
Cassandra, to be expiated ever afterwards by the Locrians
worshipping at her sanctuary.

There is perhaps one point here which calls for special
reflection. It seems to our minds unnatural that such vital
poems should be written about mere creatures of the imagination,
like gods and tribal daemons. And Chadwick has convincingly
shown that a ' heroic age ' is an age of great doings and great
individuals. In the German and Scandinavian legends many of
the heroes bear real historical names and represent real persons.
Why should this not be the case in Greek legend also ?

I think the cause of the difference lies in the vastly greater
lapse of time during which the Greek epic grew. The native
epic tradition in Northern Europe was checked and overruled
from very early times by the Latin cultural tradition. The
Latin chronicles kept some guard over history. The rich and
accomplished Latin poetry killed out, at least among the literary
classes, any effective interest in the struggling vernacular poetry.
The Church checked the Pagan imagination. But the Greek
poetic tradition had no elder brother to snub it, no foreign
schoolmaster to run a blue pencil through its errors and drive its
imagination underground. It grew in its own soil, luxuriating
in invention and building up on its own native lines an ever
statelier beauty.

The study of Greek mythology shows beyond serious question
the enormous part played by village ritual and custom in form-
ing the great legends. The aboriginal peasants of Greece seem
to have sung and danced about their field-magic, their fertility

Q 2


cycles, and their Year-Daemons, from the earliest antiquity.
Then came the Migrations with their stirring events, their strong-
armed magnificent adventurers. The focus of interest was
changed. Instead of thinking of the crops and the customs,
people talked of the last great deed of some famous condoitiere^
the last great downfall of a Minoan city or castle. The epic
lays that then broke into being were doubtless full of these
historical persons and events. And if fate had arrested their
development about the year looo B.C., we might well have had
poems as historical as — say — the French epics concerned with
Charlemagne. Dr. Chadwick's four stages of Heroic Poetry ^
do not happen to fit the course of Greek literature ; but, roughly
speaking, Homer seems to represent what Stage II would
have grown into, if it had had centuries of free life, and had
been perfected neither at Kings' Courts nor at village inns,
but at great Pan-European Festivals, like the Panegyreis of

The epic developed, but the individuals of the Heroic Age
lost their definite settings. The old myths and ritual stories
sprang up again and overgrew them. Their real names perhaps
remain in fair number ; but their deeds are the deeds of mythical
beings and they themselves indistinguishable from so many
creatures of mythology or fancy.

Even in historical times gods and spirits were present to the
minds of the Greek to a degree that we can hardly realize.
When they defeated the Persians in the full light of history,
their general's comment was : ' It is not we who have done this,
but the gods and the heroes' (Hdt. viii. 109). That is not
perfunctory piety. It is not even the mark of a specially
reverent and beautiful humility ; for the speaker was Themistocles,
a hard-bitten and scheming man. It is a remnant of the feeling
which permeates our record of the warfare of the Heroic Age,
and which is found still among primitive peoples. Think how it
pervades the Old Testament. Think of the many stories in books
,of anthropology telling how a savage who has succeeded or

^ Heroic Age, p. 94: 'I. Court poems of the Heroic Age itself. II. Epic
and narrative poems based on these. III. The popular poetry of the eighth
and following centuries. IV. The German poems of the twelfth and following
centuries, when heroic subjects had again come into favour with the higher


failed in catching his prey explains that his inana^ or even his
ioievi^ has been on that particular occasion strong or weak.
There is an early inscription extant in which the people of
Selinus celebrate a successful battle, in which presumably
various individuals had in the normal ways distinguished them-
selves. We moderns would have mentioned their names. But
the inscription of the Selinuntians runs thus: 'Thanks to the
following gods we of Selinus have conquered : Zeus Nikator,
Phobos, Heracles, Apollo, Poseidon, the Tyndaridae, Athena,
Malophoros, Pasikrateia, and the others, but especially thanks to
Zeus' (/. G. A. 515). We know how the gods Castor and
Polydeuces fought for Rome at the battle of Lake Regillus, and
for the Locrians against Croton. We know how the Greeks
before the battle of Salamis sent a ship to Aegina to fetch ' Aias
and Telamon and the other Aeacidae ', including Peleus and
Achilles, to lead them against the Persians (Hdt. viii. 64). They
are doubtless included, if not specially meant, in Themistocles'
words, attributing the victory to ' the gods and heroes '. The
same Aeacidae had been lent by Aegina to Thebes on a previous
occasion, about which the less said the better. For the Thebans
were defeated, Aeacidae and all (Hdt. v. 80), and told the
Aeginetans that next time they would prefer a regiment of men.
Now, suppose the battle of Salamis had been fought, not in the
full light of Greek history, but in the misty dawn of the Epos,
what sort of a story should we have had ? Would it have been
all about Themistocles and Eurybiades and the Corinthians .^ I
suspect it would have been Aias and Telamon and Peleus and
Achilles who defeated Xerxes. That, at least, is the way
in which many early Greek traditions seem to have been

It remains to consider another point. Why do the Homeric
battles all refer not to any warfare that was going on at the time
of their composition, but to warfare of forgotten people under
forgotten conditions in the past ? The fact is certain. Even if
the analysis made in this essay be all wrong, there will remain
just the same problem. For the poems were certainly for many
centuries in the hands of Ionian and Attic bards, who are shown
by all the evidence to have largely added to them. Yet, with all
their additions, they never brought in any celebration of their


own immediate present. There is no mention of the Asiatic
colonies, of the great Ionian cities, of the later groupings of
tribes. The few exceptions to this rule are mere accidents.
There is all through the poems a distinct refusal to cheapen epic
poetry by the celebration of contemporary things. If men
wanted to celebrate the present they did so in other forms of

What shall one say of this ? Merely that there is no cause for
surprise. It seems to be the normal instinct of a poet, at least of
an epic poet. The earliest version of the Song of Roland which
we possess was written by an Anglo-Norman scribe some thirty
years after the conquest of England. If the Normans of that age
wanted an epic sung to them, surely a good subject lay ready to
hand. Yet as a matter of fact their great epic is all about Roland,
a not very important chieftain dead three hundred years before,
not about William the Conqueror. The fugitive Britons of
Wales made no epic to tell of their conquest by the Saxons ;
they turned to a dim-shining Arthur belonging to the vaguest
past. Neither did the Saxons who were conquering them make
epics about that conquest. They sang how at some time
long past a legendary and mythical Beowulf had conquered
a monstrous Grendel and Grendel's mother and a dragon.

Yet this past of which epic poets make their songs, what
exactly is it ? It is not the plain historical past. It is the past
transformed into something ideal, something that shall be more
inspiring or more significant. In the case of the Iliad the old
traditional fighting is all concentrated into one great war, and
that a war for the possession of the very land which the pro-
fessed descendants of Agamemnon were fighting for in historical

I . It is difficult to sum up this scattered evidence, but we seem
to find In the historical background of the poems the following
elements. There were extant the ruins of a great fortified
stronghold at Ilion In the Troad, at least twice destroyed, and a
tradition of the moving wars of those peoples from the North
who troubled and eventually destroyed Aegean civilization In the
third and second millennia B. C. As to dates, the Troad seems
to have been Invaded by the horse-folk about 2300 B. c, but we
hear most about those ' Peoples of the Sea ' and others who


raided as far as Eg)^pt between 1300 and 11 50: Akhaiuasha,
Danauna, Luka, Dardenni, T'akarai and the like.^ It follows
that this fighting did not all take place at Troy, And it
seems almost certain that, as the fame of the Trojan War grew
greater, alien heroes were drawn into it, as Atli and Theodoric
were drawn into the Nibelungenlied.

2. The chief fighters in this warfare, like Menelaus, Odysseus,
Agamemnon, bear names apparently Greek, though markedly
different in formation from the Greek names of historical times ;
others bear regular Greek names, but of types that in later ages
occur chiefly in Macedon, Thrace, and Epirus.'^ It seems likely
enough, after Dr. Chadwick's work on the subject, that under
some of these names lie the names of real Heroic Age adven-
turers, while others belonged originally to divine or mythical
beings. This is a question only to be solved by external evidence,
and no external evidence is to be had, but it seems clear that
even where these names once belonged to real men, their sagas
had by the sixth century been covered over by a vast growth of
myth and even of folk-lore, to say nothing of pure fiction.

3. The poet is always conscious of a great difference between
his own age and that of the heroes, and maintains with few
lapses a remarkably consistent convention of what the Heroic
Age was like. We must conclude that there was a fairly con-
tinuous tradition from the Heroic Age downward, kept alive no
doubt in part by the ruins of Mycenaean castles, by ancient gems
and jewels such as are still discovered, by the ancient armour and
other relics which we know to have been preserved in temples,
but also by sagas and lays. At any rate it is certain that in the
poems as recited in the fourth century this heroic convention is
preserved and consciously protected. Any lapse from it is
noticed in the Alexandrian scholia. Yet the Heroic Age must
have been over by about 1 100 B. C.

4. It must always be remembered that the Ih'ad and Odyssey
are not poetic chronicles ; they are elaborate and highly wrought

^ Cf. 'AxaiF°h ^at'oo'j Avkioi, Aapdaroi, TevKpoi The terminations -sha and
-na seem to be the same as occur in Lycian. Hall, Oldest Civilization,
pp. 178 ff. Cyprus was raided by Lukki about 1400 B. c. For other records of
these tribes see Breasted, Ancient Records, iii, pp. 136, 157: Cowley,
Hittites (1920), pp. 13-23-

2 See Bury in Quarterly Review, July 1922 ; also Macurdy, 1. c.


works of fiction using traditional material. The material is
often very old, the finished poem in its present state comparatively
late. For example, the ram-adventure of Odysseus in Odyssey l
is depicted on an early Cretan gem and must have been current
about 1500 B. c. ; whereas the reunion of Odysseus and Penelope
after Meton's eikosieteris implies an astronomical discovery which
is attributed to the year 430 B. c. or thereabouts. (See p. 21 1.)

5. Into this legendary past were projected the wars of the
Greeks who in historical times established themselves in the
region called Aeolis, just south of Troy, and perhaps a real
historical meeting of hosts from the Peloponnese with hosts from
South Thessaly— of Agamemnon with Achilles ; while at a
later time again the whole tradition was reinspired by the great
war of all Hellas against Persia and the East, and knit up with
the Spartan ritual myth of the beloved Helen, for ever ravished
and rewon.

If we consider this fifth observation more closely, we ought to
receive some light upon that question which so vexed antiquity,
the birthplace of Homer. IHon is a fixed and known place ; the
Aeolian tribes also belong on the whole to a definite area. They
were driven from South Thessaly across the North Aegean by a
direct bridge of islands : Ikos, Skyros, Lesbos — and there was the
south-west extremity of the Troad immediately in front. The
meeting of Achilles and Agamemnon, if we could be sure what it
meant, would be more conclusive still. Achilles, though he had
worship in the Peloponnese, is mostly Thessalian : Agamemnon,
though he had Thessalian connexions, is mostly Peloponnesian :
and if we look for some great traditional meeting-place for the
people of Agamemnon from the south, and the people of Achilles
from Thessaly, the first place to suggest itself is the island of
Lesbos, It was also about a ' girl of Brisa ' in Lesbos that the chiefs
quarrelled. The fourth point is hardly needed, it points to the
same result. If the ravishment of Helen now takes a new
direction towards the Troad, that fits in with a movement of Helen's
Peloponnesian worshippers towards the same place. The time
and place which originally provided the main strands of the frame-
work of the Troy-saga are fairly clear. The time is the Aeolian
migration, the place is Lesbos or some early settlement on the
shore of Asia. If we take Homer as the author of the Troy-epic


the area known as Aeolis is his first birthplace. This conclusion is
exactly borne out by the ancient Lives of Homer, which, however
apocryphal and romantic, represent at least a tradition.

Further, our historical argument fits with the argument from
language. True, the Iliad and Odyssey, as given in all our
MSS., appear in a quasi-Ionic dialect. But it is beyond ques-
tion that the dialect has been in some way changed. The epos
has been worked over into its present Ionic from some other
speech. What that speech exactly was is open to dispute.^

Professor Fick, in his epoch-making editions of the two poems,
argued boldly that it was ordinary Lesbian Aeolic, and that both
poems had been definitely translated into Ionic by the rhapsode
Kynaithos of Chios about the year 540 B. c. He showed that
the poems were full of ' Aeolic ' forms in the midst of the Ionic,
and these Aeolic forms had the peculiarity, nearly always, of
being metrically different from the corresponding Ionic forms.
That is : the poems for the most part were simply wrought over
into Ionic word for word, but when the proper Ionic word did
not scan, the older Aeolic form was left. The practice is common,
one may almost say regular, in traditional books. Many English
ballads occur in northern and southern forms, many old French
poems in French of Paris, Norman-French, and Picard. And
this general conception of an ' Aeolic ' stage of the Homeric
poems has been accepted by almost all advanced critics.

Yet it needs an important correction. Pick's full theory, with
Kynaithos and the sixth century included, has had few supporters.
And if we abandon that definite date and person, the linguistic
arguments rather change their character. For the two most
characteristic distinctions of the Ionian speech, the loss of w-
sounds and the turning of a into ?;, can be shown to have occurred
later, and perhaps considerably later, than the first foundations
of the cities in Asia Minor. So that the w and the long a sounds
of Homer were as much the property of Proto-Ionic, if we may
use the term, as of Aeolic. The language of Homer — omitting
for the moment the numerous false forms and modernisms of our
present texts — is markedly based upon an older stage of the

^ To the ancients it was chiefly a question of pronunciation : cf. rr]v iroiija-iu
auayiyfoxTKeadai d^iol Za>nvpos 6 Mdyvrjs AtoXt'St SiaXeKTio' to Se dvro Koi AiKaiapxP^
(Osann, Atiecd. Rom. 5).


Greek language than either the Ionic of Herodotus or the Lesbian
of Sappho.^ This is illustrated, among other facts, by the
curious affinities between the Epic dialect and two dialects utterly
out of the range of epic influence, Arcadian and Cyprian.^

What can there be in common between Arcadia, the central
highlands of the Peloponnese, and Cyprus, the remote Greek
island in the gates of the Semite ? Nothing, one would say, but
their isolation. They were both so cut off from the normal
currents of progressive Greek civilization that they retained more
than other communities of their original speech, as the French in
Canada retain peculiar elements of the language of Louis XIV.
And consequently they show curious agreements with Homer,
whose dialect, for reasons easily intelligible, clung to the oldest
form of speech that was capable of being comfortably understood.
It is not, therefore, accurate to say siinpliciier that Homer has
been translated from Aeolic into Ionic, if by Aeolic we mean
sixth-century Lesbian, or the group of which Lesbian is the type,
Lesbian -Thessalian-Boeotian. one has to allow for the existence
of an old poetic dialect, ancestral both to the Aeolic of Lesbos
and to the Ionic of the ' mixed multitudes ', the language of old
Greece, ' before the sons of the Achaioi came '.

There is certainly a strong Lesbian element in Homer, as was
recognized in antiquity. There are certain forms of words which
are definitely Lesbian, and not primitive, dialectical peculiarities
which first originated in the Lesbian- Aeolic dialects ; falsely
formed datives in -^crdi, falsely formed perfect participles in
-K(cv^ -KovTos, a preference for Kev over dV, and various forms
like aXXvSi9, vvfi(pa, &c. The Lesbian form dypeco is generally
altered to aipeoo, but in the imperative, where it was not recog-
nized, it is left. Still, the main texture of the earlier Homeric
language is not Lesbian- Aeolic, but some earlier and more widely
diffused speech. What does this mean In history ?

It is just what we should have expected from our analysis
of the raw material of the poems. It is the speech of these
immensely old tribal or quasi-tribal traditions which, as we have
seen, form the ultimate historical content of the Iliad. They

* See the valuable Appendix to Monro's edition of Odyssey xiii-xxiv,
pp. 455-88. He seems to me to underrate the Aeolic element.

"^ Thumb counts Cyprian as 'Achaean', meaning thereby ' pre-Dorian '.


come in part from Southern Greece, but mostly from the
country traversed by various groups of wandering ' Northmen
in the Heroic' Age, from the Vardar and Struma valleys, from
Thrace and Macedon and Thessaly. What took place in Aeolia
or Lesbos was the first collecting of them into a Trojan setting.
It is interesting in this connexion to notice that the scenery of
the similes is apt to be Thessalian and not Asiatic: that the
Muses come from Olympus and the vale of Pieria in Thessaly,
and the gods, wherever they may wander, still keep their
' Olympian houses '.^

And there is this to obser\^e : that however loosely the various
masses of legend floated, there was in very early times some
feeling that they formed a whole, or at least a series of wholes.
There was some conception of a consecutive saga or history.
Each bard is understood to begin his lay — ei/6€v iXcou, or t^i/
aiioOev ye— at some particular point in the great story.'- There
must have been some great deed or experience in common, some
impulse to history writing, some breaking down of family and
tribal barriers. That experience, it seems clear, must have been
the Heroic Age.

The next birth of Homer was certainly in Ionia. We have
seen that the colonists of Lesbos had some pretensions to unity
of race. The place from which the exodus came was so close.
The bridge from Mt. Pelion to Aeolis, by Skyros and Lesbos, is
so straight and complete. There may also have been some
unity of race in the extreme south of Asia Minor, where the
group called itself ' Doris '. The Dorian tribes were at any rate

1 The Mysian Olympus may have been regarded locally as the seat of the
gods : but the ' Homeric ' gods evidently dwell in the Thessalian Olympus.
The 'Twelve Gods' seem to have been a Thessalian institution ; the first
altar to them was built by Deucalion (Schol. ap. Rhod. iii. 1085. Cf. ii. 532
with Schol.).

^ one would like to know when these lays became (l) continuously metrical
and (2) hexametric. The hexameter as it stands in Homer has been thought .
to show traces of having originated in two dactylic trimeters with anacrusis,
what the Greek metrists call Paroimiacs (see Van Leeuwen, Enchiridio7i,
pp. 1-24), and it is curious that the extant Aeolic poets hardly use the
hexameter at all. (Sappho 93, 94, 95, are instances.) The Stesichorean
metre, sets of dactylic trimeters mixed with trochaic (or iambic) metra

—^yj—\u\^ , — w — -), is perhaps older than the finished hexameter. If

there were evidence to show that the hexameter was specifically Ionic, some
clear conclusions could be drawn. ('There is. It was the metre of the
Paean, and was invented by Olen of Lycia, i. e. at Delos.'— J. A. K. T.)


the leaders of their communities. But all along the great stretch
of coast between these two little groups there seems to have been
no definite unity or common descent. They may have Inherited
a good deal from Crete,^ but every city wall contained a
av/€iKrou TrXrjdos, a ' mixed multitude ', They could merely be
classed together as ' lawones ', Sons of Javan, and even that name
is given them by foreigners.

It looks as if these ancestors of the lonians had in the extreme
stress of their migrations lost hold upon their earlier tradi-
tions. At any rate, it was only in later times, and only by
turning to their northern neighbours, that the lonians obtained,
or recovered, their heritage in the Epos. It came to them then
as part of a regular process. For it is just these central settle-
ments, these most tribeless and fugitive of the Sons of Javan,
that built up the greatest achievements of Greek civilization
before the rise of Athens. In historical times the Ionian Greek
is always prevailing over the Aeolian, ousting him, outstripping
him, annexing his cities and his possessions. The Ionian poet
Mimnermus, early in the seventh century, narrates how a party
of lonians from Colophon and Pylos set forth and captured
Aeolian Smyrna.^ The same thing can be shown to have
happened in Chios, though there the memory of the conflict
was forgotten, and the island counts as simply Ionian. And
these cases may be taken as typical. The Aeolic settlements
belong to an earlier, ruder, and more chivalric stage of culture,
and were superseded by the higher intelligence and practical
adaptability of the lonians. And besides their walled cities, the
Aeolians were robbed also of their Homer.

How did this process take place ? There may conceivably at
some time have been a definite authoritative change of dialect ;
but it seems more likely that the Epic dialect gradually changed

' See Prof. Burnet's monograph JV/io were the lonians ?

^ Mimn. 9. He makes no apology ; but we have beside his verses a more
defensive Ionian account of the affair, explaining that they were not the
aggressors. Strabo, xiv. 634. The town first belonged to the Leleges ; they
were driven out by 'Smyrnaeans' from Ionic Ephesus : they were expelled
by Aeolians, but returned with help from Ionic Colophon, and ' regained their
own land '. The story illustrates first the tendency of Ionia to outstrip and
thrust aside Aeolis; and secondly, the existence of a certain feeling of shame
in thus expelling a city of brother Greeks. To drive out Leleges was of
course fair hunting. Cf. Wilamowitz, Paniofiion, in Sitzungsber. Berlin.
Akad. 1906, iii.


as the spoken language changed. As more and more Greek
cities, and those the richer and larger ones, began to drop the
letter Vau and to pronounce Eta instead of long Alpha, the
bards who recited Homer in those cities naturally changed their
pronunciation too. Such a change would be as unconscious as
the modern English change in the pronunciation of tea or room.
But there was another and a decisive motive of change. We
have seen already that, though a short lay may be recited round
a camp fire or a banquet board, a poem at all approaching the
length of the Iliad or Odyssey can only be recited on some great
public occasion, lasting over several days, and consequently can
only have been created with that sort of occasion in view. Now
though our information is imperfect, it seems certain that the
greatest gatherings in the Aegean were Ionian. Bards who
wished to compete at the Panegureis at Delos, at Panionion,
at Ephesus, must almost of necessity recite in Ionic Greek, and
change their method of pronunciation as the spoken pronuncia-
tion changed. We know that there was a Chorus of Maidens at
Delos which could sing or speak in whatever Greek dialect was
required of them : each bard or composer ' would say it was
himself speaking' {Hymn. ApolL 163). But it is the audience,
not the performer, that eventually determines matters of dialect
and accent. The performer must adapt himself and the audience
need not. The Olympian Muses, if their ambition insisted upon
a great poem and a great audience, must perforce abandon their
native accent. And later, when the centre of culture passed on
from Ionia to Athens, they had to repeat the process.

Aeolis was left in a backwater. And when it emerged, it
spoke in tones as different from those of its old Homer as can
well be conceived. Poetry in Lesbos became Traditional Poetry
no more. We must leave it aside and return to the development
of Homer in Ionia.




We still stand under the spell of the Iliad. Amid the deepest
strands that are woven in the thread of our Western civilization
there is more than one which is drawn originally from Greece and
Greek literature. And at the fountain-head of Greek literature
there stands, naturally enough, the dateless traditional book, not
indeed sacred as in other lands, but still unapproachable, and
far removed from the possibilities of human competition. This
was the position of the Iliad in Alexandrian Greece. Rome
took over the conception, and it has passed on, for the most part,
to be part of the intellectual heritage of the Western world.

Criticism has, of course, in some respects, shattered the
Alexandrian view to pieces. Instead of the primaeval and all- wise
poet. Homer, we are left with a kind .of saga-figure, similar
to that of Achilles or Agamemnon, or the mighty flashing-
helmeted Son of a Shield-Belt. The name Homeros may con-
ceivably be a name once borne by a living person. But if so, we
know nothing of him, except indeed that he did not, in any
complete sense, write the Iliad and Odyssey. It seems on the
whole safest to regard Homeros as the name of an imaginary
ancestor worshipped by the schools of bards called 'Oiir^piSaL or
^Ofirjpov naiSes, a name parallel to Ion, or Doros, or Hellen, or
even Amphictyon.^ The exact form of theory which we accept

^ This line of thought has been brilliantly and to my mind conclusively
developed by Prof. J. A. K. Thomson in Chapter x of his Studies in the
Odyssey. The Aoidos, or Bard, and afterwards the Poieth, is evolved from
the Exarchon or Leader of the magico-religious dance. Homeros is especi-
ally the ideal Aoidos or Exarchon of the Dalian Maidens, as Apollo is of the
Muses. They sang (Hes. Theog. 39) <^<iivr\ ofx-qpfvcrai; Homeros is a ritual
name, generated from this ' homereia ' and the ' Homeridae ' or ' Homeristae '.


is of little moment. There is a broad general agreement between
most of the followers and correctors of Wolf and Lachmann. I
wish in the present lecture to advance no theory of my own, but
merely to consider what effect this scientific analysis has, or
should have, upon our general enjoyment and understanding of
the Iliad as a great poem.

Mr. Mackail, in his Li'fe of William Morris^ remarks in pass-
ing that in the Iliad we have a second-rate subject made into
a first-rate and indeed incomparable poem by the genius of
a great poet. I think this view would probably be widely
accepted. Many scholars would agree, with a pang, that the
subject of the Wrath was not quite in the first rank of nobleness.
The Wrath against Hector after Patroclus' death may be a great
subject. But the Wrath with Agamemnon about a personal
slight is not. The fact that in the loss of Briseis it is almost
entirely the personal slight, not the loss of a beloved being, that
matters to Achilles, puts all the emotion several degrees lower.
So much many scholars would admit, and then console them-
selves by asserting the splendid perfection of the poem and the
genius of the incomparable poet.

Now over this incomparable poet there is much high feeling
and, in my opinion, some confusion of thought. He certainly
did not write the whole Iliad : so much I may take as generally
admitted. (Though if even that were denied, one might ask
what is meant by 'the whole Iliad\ Is it the Oxford text?
Or is it the text of our earliest papyri, probably some thousands
of lines longer } Or is it the text of Zenodotus or Aristarchus,
some thousands of lines shorter?) What then exactly did he
write that is so vastly better than the work of his collaborators
and followers ?

The Alexandrians proceeded by the method of ' obelizing '
certain lines, passages, or whole books. These were ' spurious ' :
all the rest was the work of the one Homer. Is this a satis-
factory method ?

No one would now analyse the Pentateuch by cutting out as
'spurious' the parts that cannot have been written by Moses,
and leaving Moses author of all the rest. No one would cut out
all the psalms that cannot have been written by David, and leave


David author of all the rest. one cannot even apply such
a method to Isaiah, where it would be much more legitimate.
Isaiah is a definite historical figure. We know when and where
he lived. We know his circumstances and his policy. We have
some criterion for telling what he wrote. Yet even in his case
this method has completely broken down. The processes through
which the Book of Isaiah has passed are far too complex for
a mere division into ' genuine ' and spurious '.^ Yet this method
at its crudest is still apt to be applied to the Iliad.

As soon as one has grasped the idea of a Traditional Book, it
is clear beforehand that mere ' obelizing ' will lead to no good
result. It means stripping off one by one the contributions of
all the poets who have worked at the Iliad. It is like the old
attempts at restoring the original language of the original kernel,
only far, far more desperate. And in practice, too, it refuses to
work. For as you analyse the poem back towards its source, it
proves not to have one source but many. The Catalogue and
the Doloneia are almost universally recognized as coming into
the Iliad from elsewhere ; the Embassy^ the book of all others
which is most quoted in antiquity and seems most to have im-
pressed the imagination of Greece, is also one of the parts most
markedly foreign to its present framework. I will not multiply
instances. Very little reflection is needed to convince us that
a mere process of stripping off the ' non-original ' will not auto-
matically leave us with the pure work of the incomparable poet.
If we want to discover him we must search for him.

And how shall we search ? What criteria have we ? In the
case of Isaiah we have that prophet's date, his residence, his
recorded political activity. What have we for Homer } The
tradition supplies us with plenty of competing birthplaces, with
a date which fluctuates between the twelfth century and the sixth,
and half a dozen confessedly mythical lives. It is hard to make
much use of these. If we try to discover criteria of our own,
well, Fick considers that Homer was an Aeolian, and only those
parts which will turn back into Aeolic are his genuine work.
Some old English scholars thought he lived in Thessaly, and got
a criterion out of that. Neither criterion has been successful, for

^ See, for instance, Prof. Kennett's Compostiion of the Book of Isaiah
(British Academy), 1910; or Box's Isaiah (1908).


reasons which we need not go into here. one reason was that
they chose as their field of operation the supposed first origin of
the poems, where our knowledge is almost nil. Obviously that
is not a sound method. Beginning at the later end, where there
is more hope of a safe result, Wilamowitz has forcibly suggested
that one definite individual can be discovered in &. He wrote
0, the Broken Battle^ in order to make room in the Iliad for
the Embassy and the Doloneia^ I and K. If we accept that
result, we have at any rate one poet whom we can isolate. Bethe,
again, has come near to persuading us that the man who wrote
the Converse (Z) also wrote the Ransoming (fl), and did a great
deal towards the general shaping and arranging of the I/i'ad.
Such a view would perhaps come near to satisfying Miss Stawell ;
and such a poet, if one felt sure about him, might almost deserve
the name of ' Homer '. Yet not quite. He would be a magni-
ficent poet : of that there is no question. But would he be
incomparably better than various others? Than the author of
the Embassy^ for instance ? Or can we confidently say that the
man who put the Bellerophon passages — or the Sarpedon
passages, or the Shield-malving — into our Iliad v^^s, incomparably
better than the unknown persons who seem to have originally
written them for different contexts ? Can we say that the Iliad
owes incomparably more to him than to them } I confess that in
the present state of our knowledge all such confident language
about the Poet seems to me unwarranted. We have got the
Poem, and we can puzzle out a good deal about its probable
manner of formation. We have not discovered any one historical
poet. He is at best only a hypothesis. There may of course
have been a man called Homeros, as there doubtless was a man
called David. But we know nothing about him, not his date nor
his birthplace nor what he wrote. And the Homer of our
imaginations is not he, but a projection of our own feelings,
a result of our habit of always thinking in terms of persons,
a mythical Maker to account for the thing made. What we
really know is not a man but a poem ; let us focus our thoughts
upon that and try to understand its greatness. I believe we
shall find among the causes of that greatness something nobler
and more august than the genius of any individual man.^

^ Compare the case of the Heracleia. There were evidently many versions

2760 R


I wish first to consider patiently this difficulty. It is, I suppose,
quite clear that the Iliad is a good poem. Most people have
only to read it to feel quite sure of the fact : and if any particular
reader does not feel sure by his own instinct, there is enough
authority on the subject to convince any but the most self-
confident that his doubts are ill-grounded. Now why is it that
the Iliad is a good poem when it has so many of the character-
istics of a bad one }

In the first place, as we noticed above, the subject is second-
rate. The horrid phrase which describes Achilles as ' sulking in
his tent' is not very far from the truth. And sulking is not
a noble, nor yet a poetical, state of mind. Achilles, again, is not
a very sympathetic hero. His eloquence is amazing, and we are
ready to believe in his dauntless courage and prowess and swift-
ness of foot. But, if it were not for his mere misery and repent-
ance at the end of the poem, I think that most readers would
actually dislike him for his crude pride and self-absorption, his
cruelty and lack of love. Even his love for Patroclus never
impresses one as having unselfishness about it : it is not like the
love of Orestes and Py lades.

Again, there Is a test which most people apply instinctively to
a modern work of fiction, and which is most powerful in separat-
ing the good from the bad. I mean the amount of finish and
conscientiousness in the more hidden parts. What we call
' showy ' or ' flashy ' work is generally work in which the

of that epic, and their 'Homer' is sometimes referred to as *he who made',
sometimes as ' they who made ' the Heracleia. (Eratosthenes ap. Strab.,
p. 688 ot T^v 'HpuKXeiav Tron'](TavTes. It means, I suppose, ' the various people
who put the Heracles-saga into verse'. Cf. Schol. V on tt 57 01 t(ov Kvirpicov
TToirjTai.) But the interesting thing is that among them we know of three
distinct individuals : Pisander of Rhodes, Peisinoos of Rhodes, and Panyassis
of Halicarnassus. There were doubtless others as well. Now Paul
Friedlaender {Philologische Utttersuchingeft, vol. xix) has made a brilliant
study of the Hefacleia problem. By analysis of the myth and the local data
he succeeds in tracing several stages in the development of the Heracles-
saga : an epic poem, the Dodecathlos, made not in Argos nor yet in Boeotia,
but in Rhodes, with the Rhodian goddess Alektrona-Electryone as the hero's
mother ; an expanded Rhodian form ; and a later Samian and Coan
reshaping. These three stages correspond fairly well with the three authors,
two Rhodian and one from Halicarnassus ; and if the Heracleia were extant
we could probably separate and appraise their respective contributions. We
have no such data for the Iliad and Odyssey. A good attempt at finding
personal qualities in the poems in Cauer, Gnmdfr?, pp. 548 fif. Also, though
with too much dogma, in Rothe, on Repetitions and Contradictions.


momentary effect of particular scenes is strong, but which will
not bear looking- Into. If you look close you find weaknesses,
inconsistencies, contradictions. Now, notoriously, this is the
case with the Iliad. The wall round the Greek camp alone,
though the writing about it is always good and stirring, will
provide half a dozen glaring instances of this sort of inner flaw.
It is built at the end of H in the tenth year of the war. Yet
a phrase in the description of the camp later (H 31) implies that
it was built — as it naturally would be — in the first. In M 10-33
it remains ' steadfast ' {e/XTreSou) till the end of the war and is then
destroyed by floods ; but in O 361, before the death of Patroclus,
it is swept away by Apollo like a child's castle of sand on the
sea-shore. Its towers had been broken in M 399. In M and N
the wall is sometimes present and sometimes absent. Also two
separate heroes, Hector and Sarpedon, are mentioned in different
places, and in exactly the same words, as being the first to get
over it (M 438, 11 558). There is a fearful fight when the
Trojans are attacking the wall to get to the ships : when they
retreat in panic there is generally no wall there. All this is
explained in detail in Dr. Leaf's commentary.^ It is pretty clear
that there were two versions of the fighting extant, one in which
the camp was unfortified, and one in which it was provided with
a wall and moat. And brilliant episodes are borrowed from one
or the other as the minstrels pleased.

Again, there is the cardinal instance of the contradiction be-
tween Books n and I. In Book IT, Achilles, as he sees the
routed Greeks, breaks into a splendid complaint that if only
Agamemnon would seek his friendship and offer him amends the
Trojans would soon fly and ' choke the trenches with their dead '.
He sends Patroclus forth to help the Greeks, but warns him not
to go too far in pursuit, lest Agamemnon should feel too secure
and should fail to offer atonement.

Obviously, then, Agamemnon has not offered atonement. Yet

there is a book before this which is occupied from first to last

entirely with Agamemnon's offers of princely atonement! one

sees what has happened. Both passages lay before some com-

^ The late Professor Earle — anticipated, I find, by Hermann — shows
reason to suspect that Thucydides used an //lad which did not contain the
account of the Wall-building in H. Earle, Collected Essays, pp. I42ff. See
Chapter XII, p. 295 n.

R 2


piler of the Iliad. They were not consistent, but each was too
good to lose. He put both in, sacrificing, like a bad artist, the
whole to the part.

Thirdly, there is the same sort of fault running through many
of the descriptions. Even the battle scenes, vivid as they are,
will sometimes not bear thinking out. As we saw in the case of
the breastplate, the poet has not fully thought out the words he
was using. It sounds well. It is exciting. But it is not real.
It is like a battle composed by some romantic poet, who furnishes
his warriors with gleaming morions and resounding culverins,
but is not quite sure what things they are.

Apply the same test even to the language, the miraculous
heaven-sent language which has been the wonder and the awe
of all poets afterwards. Is it not full of such ' morions ' and
' culverins ' ? Do you not find upon every page fair-sounding
words, whose meaning seems to have been far from clear to the
poets themselves who used them? Of course it is rare to find
a definite substantive of which the meaning is quite unknown,
though even such occur : for instance, in the case of epithets
of the gods. *EpfjLeia9 aKCLK-qra, SiaKTopo? apye'L(p6vTrjS, not one
of the epithets is understood. There are also a few words which
are used in two senses, of which we can fairly say that one
is a mistake.^ But it is more often the form of the word or
sentence that shows a lack of understanding. There are crowds
of words which, as they stand, are no words but only mistakes,
old forms first miswritten and then wrongly recorrected so as
to fill up the metre. There are words first wrongly divided,
like i/r]Svfxo9, and then wrongly explained.

^ For example, 8ovTrrj(rai means 'to make a noise* {= ■^o(j)rj<Tai say the
Lexica), but owing to the phrase Sovnrjaev 8e rreaaiu, ' he crashed as he fell ',
the old Glossographi, who explained the hard words in Homer in pre-
Alexandrian times, interpreted it as simply ' to die '. Aristarchus has to
correct them {ol ykuxraoypafpoi to Bovwiicrni eu dvO evos avrl rov dnodaveiv).
But the writer of ^ 679 uses the phrase Sebovnoros OlSmoSao for ' when the
son of Oedipus (?) was slain', TedvrjKiWos. That is, he misunderstood the old
usage, just as the Glossographi did. (See Lehrs, Aristarchus, p. 103 f.; of
course there are ways of escape suggested by the grammarians.) Again, the
word uTivrai, oTTfi'To, looks as if it meant 'stands, stood', but really meant
' intended', or perhaps 'boasted'. So Aristarchus (Lehrs, p. 98 f.). But in
X 584 it is used as ' stood ', areiiro Se 8i^c'ia)Vy TTueiv S' ovk hx^v fXeadai. And
Aristarchus can only condemn the lines : "lararo vvp enl nohdv' KexprjTai 8e rfj
Xe^ft 6 8ia<TKfva<TTr]s ivapa rqv tov ttoitjtov (Tvvrjdeiav is Aristonicus's note.

These are not isolated phenomena.


Now, of course, a great deal of this is mere ' surface corrup-
tion '. Many mistakes are only due to the latest rhapsodes, who
recited the Ionic poem in Attica, and thus inevitably introduced
Attic elements into the language, and even misunderstood the
older Ionic forms. You can largely remove the Atticisms and
obvious errors. Editors like Van Leeuwen and Piatt and Rzach
have corrected them by the hundred, with most useful and
instructive results. But the process of correction is never com-
plete. Clear away the Attic surface and there rises beneath
another surface with another set of corruptions, where Ionic
rhapsodes have introduced just the same elements of confusion
into an Aeolic, or at least a pre-Ionic, language. The confusion
of tongues is deep down in the heart of the Homeric dialect, and
no surgery in the world can cut beneath it.

Of course one must not judge a poet as one would a gram-
marian. Yet this confusion of tongues has a certain weight as
evidence. It seems to be part of a general vagueness of treat-
ment, a lack of precision and of grip.

We often find, too, that descriptive phrases are not used so as
accurately to fit the thing described. They are caught up ready-
made from a store of such things : perpetual epithets, front
halves of lines, back halves of lines, whole lines, if need be, and
long formulae. The stores of the poets were full and brimming.
A bard need only put In his hand and choose out a well-
sounding phrase. Even the similes are ready-made. There
must have been originally some poet who saw the spring of some
warrior in battle, and was struck by its likeness to the leap of
a lion. But that was long before our Iliad. The poets of our
Iliad scarcely need to have seen a lion. They have their stores
of traditional similes taken from almost every moment in a lion's
life : when he is hungry, when he is full, when he attacks the
fold, when he retires from the fold, when he is wounded, when he
is triumphant, when he is scared with torches, when he walks
ravening in the wind and rain. Every simile is fine, vivid, and
lifelike ; but a good many of them are not apposite to the case
for which they are used, and all have the same ready-made air.

Consider in detail this fine simile (M 41) :

As in the midst of hounds and men that are hunters,
a boar or a lion wheels, glaring in his strength ; and they


set them like a wall {7rvpyr)86v) and stand against him, and
the spears fly fast from their hands ; yet his proud heart
trembles not nor flees, till his daring is his death, but swiftly
he turns and turns, making trial of the lines of men ; and
wheresoever he charges, the lines of men give way.

The description of the boar or lion is splendid : but what
situation does it seem to describe ? A hero left alone, hard
pressed by enemies but refusing to retreat ? That is what one
thinks of. That is probably the situation for which it was
originally written. But, as the passage stands, the Greeks are
flying and Hector pursuing them back beyond their wall. The
passage continues : ' Even so Hector, going up and down the
throng, besought his comrades, urging them to cross the trench.'
Hector, urging on his conquering comrades, is really not parti-
cularly like this surrounded and baffled lion, ' whose daring is
his death '.

Now at a point of the action immediately before this — there is
a digression between — in yl 551, there is a hero very like indeed
to this boar or lion, to wit, Aias, who has been up to the last
moment standing alone against the advance of the Trojans and
protecting the Greek retreat. At the end Zeus sent into him
also a spirit of flight.

He moved backward, searching with his eyes as a wild
beast searches, back toward the throng of his comrades,
half turning again and again, slowly changing knee for
knee. Even as a red lion draws back from a yard of oxen,
frighted by hounds and husbandmen keeping vigil all night
long, who suffer him not to take out the fat of the oxen ;
and hungering for flesh, he charges but wins nothing ; so
fast fly spears from brave hands to meet him, and flaming
torches, which he shrinks from for all his fury ; and at dawn
he goes away alone with misery in his heart : so then did
Aias go back from the Trojans, unwilling and with misery
in his heart. For he feared for the ships of the Achaeans.

There follows instantly another simile, slightly strange perhaps
to our conventional taste, but very vivid and good :

Even as an ass going beside a field overpowers the boys
who drive him, a dull ass about whose back many a staff
is broken ; and he enters the standing corn and ravages it,
and the boys smite him with sticks, but their strength is
feeble, and scarcely do they drive him out when he has had


his fill of the corn. So then about Aias the tall, son of
Telamon, high-hearted Trojans and allies famed afar
followed thrusting, &c., &c.

Now think of our first simile, the lion or boar surrounded
and confronted by a wall of men and hounds, but refusing to
retire. Does it not seem to belong here rather than to its
present context ? Did it not perhaps describe the state of Aias
just a moment earlier, while he still stood alone and Zeus had not
yet sent into him that ' fear for the Achaean ships ' ? I think,
agreeing with Leaf and others, that this must have been the
original place for which the simile was written. The rhapsode
who was composing our eleventh and twelfth book found in
various MSS., that came somehow into his hands, no less than
three different similes applied to Aias covering the Greek retreat.
He put two of them straight in together, the midnight lion and
the ass in the corn. The other was far too good to lose, so he
kept it by him to use at the first opportunity. Early in the next
book came the mention of a wall, which checks for a time the
rush of the Trojans ; it so happens that the hounds and hunters
of the simile were said to be like a wall. That place will do.
The incongruity will be decently masked. So he puts it in there ;
and at present the triumphant advance of Hector is compared to
the stubbornness of a baffled boar or lion refusing to retreat.

Does this explanation fail to carry conviction ? Demonstra-
tion is, of course, impossible in these questions of criticism. But
take another case in the same book. When the Trojans
(M 131 ff.) are charging at the gate of the Greek wall, they find
there standing in front of the gate two heroes of the race of the
Lapithae, Polypoites and Leonteus.

They two in front of the high gate were standing like
high-crested oaks on a mountain, which abide the wind and
the rain through all days, firm in their long roots that reach
deep into the earth.

A moment after we are told of these same two men : —

Out then they charged and fought in front of the gates,
like wild boars on a mountain, who abide the oncoming
throng of men and hounds, and charging side-long break
the underwood about them, tearing it root-wise up, and
through all else comes the noise of gnashing tusks. ... So


came through all else the noise of the bright bronze upon
their bodies, smitten with shafts in front.

People who stand firm in front of a gate, like oaks, are not very
like wild boars that rush out and tear up the undergrowth,
making a noise with their tusks. This may sound captious : but
the difficulty is quite real, and was felt in ancient times.
Different solutions are offered, for instance, by Porphyry and
Hephaestion. Did not the last compiler of M find in two
different books two different accounts of this fight at the gate ?
In one the two Lapithae alone stood like oaks. In another a
mass of Greeks charged out, led very possibly, but not certainly,
by the two Lapithae. Both similes were too good to lose. He
followed the story of the oaks, yet he was reluctant to lose the
wild boars. So observe his mode of procedure. He puts in the
wild boars, and then, at the end, soothes the imagination of any
hearer who is puzzled at the lack of resemblance, by explaining
that the point of similarity lies in the noise. The contradiction is
masked. Boars' tusks make a noise, and so do shields struck
with spears ! ^

Another simile, fifteen lines later, makes of this hypothesis
almost what in this atmosphere of conjectures may be called
a certainty. Asius, who is leading the Trojan attack, cries out
that ' these men are like a swarm of bees or wasps who have
built their nests beside a rocky path, and pour out to fight with
hunters to protect their young '. That comparison can scarcely
have been invented to describe two solitary heroes standing in
front of a gate. It may well have described a great mass of
Greeks pouring out through the gate. But that was part of the
rejected story. It belonged to the same version as the rushing
wild boars.2

^ An idiom by which ' a mere detail in the original scheme of the simile is
made the base of a fresh simile ' (Leaf) has many parallels in Homer, but
hardly in such an extreme degree as this. The passage O 618 ff. is very
similar, and probably has a similar history. Hector's onset is compared
(i) to waves falling on a rock, which stands immovable ; (2) to a wave
crashing down upon a ship, which is badly shaken ; then comes v. 629,
' even so was the spirit of the Achaeans shaken within their breasts.' I sus-
pect that these two similes come from separate sources ; the minstrel felt
them to be not quite consistent, so he added v. 629. It is worth remarking
that the five lines just preceding are inconsistent with their present context,
and were condemned by Zenodotus and Aristarchus.

'^ Breal, 1. c, p. 115, traces the double names in the languages of gods and


These are mere illustrations. The force of the argument, of
course, depends upon the number of such cases.

The conclusion is hard to resist, and it is one that seems to
detract enormously from the high value of the poems as original
poetry. Even the similes, the very breath of life of the poetry of
Homer, are in many cases, indeed usually, adopted ready-made.
Their vividness, their closeness of observation, their air of
freshness and spontaneity, are all deceptive. Nearly all of them
are taken over from older books, and many of them were origin-
ally written to describe some quite different occasion.

All these qualities, which we have arrayed in a catalogue, have
one common characteristic, and that one which is generally
considered fatal to any art which claims to be what we call
' original ' or ' individual ', a thing created by a particular man.
I do not say that Homer has no other flaws. But as to these
already mentioned, I venture to think that we only find them
vicious because we are judging by wrong standards. We are ]
applying to a traditional poem, the creation of whole generations
of men, poets and hearers, working through many ages, canons
which only apply to the works to which we are accustomed in
modern literature, original poems, made at a definite date by
a definite self-conscious author. ^

The subject is a difficult one, and I am not sure that I see
clearly through it. But I will try to give the result of my

First of all, I think that we are apt to confuse originality with
a much less important thing, novelty. A story about motor-cars
or wireless telegraphy possesses, or once possessed, novelty ;
but whether it ever possessed originality depended entirely on
qualities in the author's mind.

Of course, there was originality in conceiving the notion of
bringing the motor-car or the wireless telegraph into the realm
of art. A very small modicum of originality, but still some.
And I would not say that such originality was contemptible,
because one of the ways in which art advances is by the opening

of men to the same multiplicity of sources. one source said Bptapewr, another
Myal<x>v (A 404). So also Miilder, Quellen, pp. 65, 139, 223.


up of new regions to its influence, or, in other words, by the
discovery of beauty or interest in new places. Also, the man
who conceives or executes a thing for the first time is no doubt
apt to do so with a freshness and intensity which make his work
not only novel but original. But the difference between the two
qualities is clear. Mere novelty is a thing external and acci-
dental. It depends upon dates. It wears off. For instance, the
Hippolytus seems to have been the first love tragedy in European
literature. In that sense it was novel, but its novelty has worn
off during these last two thousand three hundred years. Yet its
originality is living still and felt vividly.

Origo means a spring, a rising of water. And, though it is
generally a mere waste of ingenuity to tie the sense of a word
down to its supposed derivation, I suspect that the most fruitful
way of understanding the word ' originality ' may be to re-
member this meaning. We do call a work of art original when
it produces the impression of a living source, so that one says :
' Here is beauty or wisdom springing ; not drawn through long
pipes nor collected in buckets.' This spring-like self-moving
quality is a thing which does not depend on novelty, and there-
fore cannot grow stale. I remember examining in Florence a
MS. of Euripides, which was very hard to read, blurred with
age and sea-water and exposure to the sun. And as I pored
over it, there gradually showed through the dusty blur the first
words of a lyric in the Alcestis. It was as old as the hills, and I
had long known it by heart. Yet the freshness of it glowed
through that rather stale air like something young and living.
I remember a feeling of flowers and of springing water.

This quality has not much to do with novelty. Probably it
does imply that the poet has in some sense gone himself to the
fountain-head, that his emotion is a real first-hand emotion, self-
moving and possessed of a life of its own, not merely a derivative
emotion responding to the emotion of another. Yet I doubt
if even so much can be fairly demanded, that a poet, to be
original, must himself go to the fountain-head. The words are
ambiguous. It would be preposterous to demand that a writer
shall experience personally all that he writes about. And it is
very noteworthy how many great poets seem to have drawn
most of their inspiration not directly from experience, but deriva-


lively from experience already interpreted in other men's poetry.
Think of Burns's vsongs. There is almost no poetry so original
in the impression it makes. And yet we have detailed evidence
that a great deal of Burns's most beautiful and spontaneous work
is really a working up of old traditional material. He thought
over the words and rhythm of an existing country song while
his wife sang the air, and thus gradually he modified the existing
verses and added others, till a song was produced, a song both
new and old, derivative and yet highly original. I suspect that
the mistake which we are apt to make is to apply a merely
external test to something that depends on the most intimate
workings of a man's imagination. The thing that is of impor-
tance in a poem, given the necessary technical power, is not
mere novelty, nor yet personal knowledge or experience, but
simply the intensity of imagination with which the poet has
realized his subject. And that intensity may be the product
of a thousand things ; of which personal experience may, but
need not, be one. Almost the first characteristic which one
notes in what we call a * man of genius ' is his power of making
a very little experience reach an enormous way. This sounds
very different from Carlyle's definition of genius as an infinite
capacity for taking pains. But in reality that capacity for taking
pains is itself dependent on an intense and absorbing interest.
So long as you are really interested, you cannot help taking
pains. As the interest fades, you first begin to be conscious of
the pains, and then cease to take any more.

In the same way, when we blame a work of Art as ' con-
ventional ' or ' laboured ' or the like, we are often using language
loosely. A laboured work is of course not a work on which the
man has worked hard : it is a work in which the labour is more
manifest than the result, or in which one is somehow conscious
of labour. Pains have been taken, but some other factor of
success is not there. A conventional work is not a work com-
posed according to the rules of some convention or other. All
art is that. It is a work in which other qualities are lacking, and
the convention obtrudes itself

Intensity of imagination is the important thing. It is intensity
of imagination that makes a poet's work ' real ', as we say ;


spontaneous, infectious or convincing. Especially it is this that
creates an atmosphere ; that makes us feel, on opening the pages
of a book, that we are in a different world, and a world full of
real beings about whom, in one way or another, we care. And
I suspect that ultimately the greatness of a poem or work of ima-
ginative art depends mostly upon two questions : how strongly
we feel ourselves transported to this new world, and what sort
of a world it is when we get there, how great or interesting or
beautiful.^ Think of the first scene of Hamlet^ the first page of
the Divina Coimnedia^ the first lines of the Againeimwn ; how
swiftly and into what wonderful regions they carry you ! And
if you apply this same test to the Iliad or Odyssey^ the response
is so amazing that you understand at once why these poems have
so often and in such various ages been considered absolutely of
all the greatest. Open the book anywhere (^33). 'So spake
he, and the old man trembled and obeyed his word ; and he
went in silence by the shore of the many-sounding sea, and
prayed alone to the Lord Apollo, whom fair-haired Leto bare.'
Turn the pages (^ 'Sl'^- ' -^"*^ ^ herd he wrought thereon of
straight-horned kine. The kine were wrought of gold and of
tin, and lowing they wended forth from the byre to their pasture,
by the side of a singing river, by a bed of slender reeds.' Turn
again (X 356). ' I look upon thee and know thee as thou art.
I could never have moved thee, for the heart is iron within thy
breast. Therefore beware lest I be to thee a wrath of god, on
that day when Paris and Phoebus Apollo shall slay thee in thy
valour at the Scaean Gates.' ^

^ Of course, in proportion as art becomes more realistic the ' new world '
in question becomes more and more closely the present world more vividly
felt and understood.

^ Of ((peiT, e^eiatv 8' 6 yepoiv Koi iTTtidero fj.v0a>'

I3rj S' atcecoi/ napa diva 7roXv0Xoto-/3oio daXdaarjs'
noWa 8' eneir dnavevdf Kia>v Tjpaff' 6 yepaios
' AnoWoivi avaKTi, Tov rjijKonos reae A.t}to). (A 33 ff')

'Ef 8' dyeXtjv noiTjaf jBocov opdoKpatpdcov'
al Se /Sdes \pv(Toio TeTfvxdTO Kaacrnipov Tf,
fjLVKtjdpw 8' diTO Konpov eTrecraeiiovTO vopovde
nap TTOTiipov KiKdhovTa, rrnpa poSavov So^oK^n. (2 573 ^-J

H (t' €v yiyuaxTKav TrpoTioacropai, ovS' lip epeWov
TTtiaeiv ' Tj yap trot ye cribTjpeos ev (f>pfa\ dvfMos'
(ppa^eo vvv, prj rol ri 6(a>v prjVipa yevcopai
rjpari ra ore Kev at Udpis Kal Ooi/3oy 'ArroWmv
eaOXov iovT oKfaaxriv eVt ^Kaifjcri TrvXijaiv. (X 356 ff.)


How irresistibly do the chance words bear one away, and to
what a world ! We can stand apart and argue and analyse, and
show that the real world portrayed in the poems was one full
of suffering- and injustice, and that the poet was sometimes over-
lax in his moral judgements. Yet the world into which he
takes us is somehow more splendid than any created by other
men. Where were there ever battles or heroes like these, such
beauty, such manliness, such terror and pity and passion, and
such all-ruling majesty of calm ? There are many strong men
and fair women in other stories ; why is it that, almost before a
word is spoken, we feel in our bones the strength of these
Homeric heroes, the beauty of these grave and white-armed
women ? You remember, in the Old Testament, the watchman
who stood upon the tower in Jezreel, when they saw men and
chariots approaching in the distance, and sent out one horseman
after another to inquire their purpose. 'And the watchman
answered and said : He came even unto them and cometh not
again. And the driving is like the driving of Jehu the son of
Nimshi ; for he driveth furiously.' We knew nothing about the
driving of Jehu before. We hear no word more about it after-
wards. But the one sentence has behind it just that intensity of
imagination which makes thoughts live and vibrate like new
things a hundred, or a thousand, or two thousand, years after
their first utterance. And that is the quality that one finds in

Think how the beauty of Helen has lived through the ages.
Like the driving of Jehu, it is now an immortal thing. And the
main, though not of course the sole, source of the whole concep-
tion is certainly the Iliad. Yet in the whole Iliad there is
practically not a word spoken in description of Helen. As
Lessing has remarked in a well-known passage of the Laokoon^
almost the whole of our knowledge of Helen's beauty comes
from a few lines in the third book, where Helen goes up to the
wall of Troy to see the battle between Menelaus and Paris. ' So
speaking, the goddess put into her heart a longing for her
husband of yore and her city and her father and mother. And
straightway she veiled herself with white linen, and went forth
from her chamber shedding a great tear. , . .' The elders of
Troy were seated on the wall, and when they saw Helen coming.


' softly they spake to one another winged words : " vSmall wonder
that the Trojans and mailed Greeks should endure pain through
many years for such a woman. Strangely like she is in face to
some immortal spirit^ ' ^ That is all we know. Not one of all
the Homeric bards fell into the yawning trap of describing Helen,
and making a catalogue of her features. She was veiled ; she
was weeping; and she was strangely like in face to some
immortal spirit. And the old men, who strove for peace, could
feel no anger at the war.

Now this intensity of imagination can be attained by many
writers at their most exalted moments. Their imagination can
follow the call of their emotions. But one of the extraordinary
things in the Iliad is the prevalence of this intensity all through
the ordinary things of life. ' As riseth the screaming of cranes
in front of the sunrise, cranes that have fled from winter and
measureless rain, screaming they fly over the streams ot
ocean, bearing unto the dwarf-men battle and death.' - Who
that can once read Homer freely, untroubled by difficulties of
language, can ever forget the cranes ? And not only the cranes,
but the swarming bees, the flies about the milk-pails, the wolves
and boars and lions and swift dogs, and the crook-horned swing-
footed kine 1 It is a fairly wide world that the poets lay open to
us, and every remotest corner of it is interesting and vivid,
every commonest experience in it, the washing of hands, the
eating of food, the acts of sleeping and waking, shares some-
how in the beauty and even in the grandeur of the whole.
Mr. Mackail ''• has observed how full the poems are of images
drawn from fire : the bright armour flashes like fire, the armies

^ "flf ftTToOcra Bih. yXvKvp Ifiepov ffi^aXe ffvfia

dp8p6s T€ Trporepoio Koi acrrfos jjSe tqktjwv'
avTiKa di" apyfvvfjai. KoXv^afievr) odovTjcriv
wpnar fK daXdfxoio reptv Kara dciKpv x^o^<^(t- (^ ^39^-)

H*ca npos aXXrjXovs (irta iTTepoeur nyopfvov'
'* Ou vifxecns Tpcoas koi evKvrjiJ.i,8as 'A)(aiovs
TOitiS" dp.(f)\ yvvaiKi ttoXvv xpd^op aXyea nua-xeii''
ctluS)S ddavdTr](n. derjs els a>7ra eoiKfv.''' (T I55ff')

HuTf ntp KXayyt] yepdvoav TreXei ovpavodi npo,
ai T fTrei ovi> ;^ei/xcoi'a (pvyou koi ddecrfpaTOV on^pov,
KXnyyfj rai ye neTOVTai en 'Q.Keavoio podcof,
nv8pdai Ilvypaioicri (f>6vov koi Krjpn (fiepova-ai. (F 3 ff.)

^ In one of his lectures as Professor of Poetry at Oxford.


clash, ' even as destroying fire that falls upon a limitless forest ' ;
a hero's ' hands are like unto fire and his wrath unto red iron ' ;
and the men ' fight together, a body of burning fire '. The whole
poem is shot through with this fire, which seems like a symbol
of the inward force of which we have been speaking, a fiery
intensity of imagination. Given this force within, and the
Homeric language as an instrument for its expression, a language
more gorgeous than Milton's, yet as simple and direct as that of
Burns, there is no further need to be surprised at the extra-
ordinary greatness of the Iliad.

But now comes a curious observation. We who are accus-
tomed to modern literature always associate this sort of imagina-
tive intensity with something personal. We connect it with an
artist's individuality, or with originality in the sense of ' newness '.
It seems as though, under modern conditions, an artist usually
did not feel or imagine intensely unless he was producing some
work which was definitely his own and not another's, w^ork
which must bear his personal name and be marked by his personal
experience or character. one element at least in the widespread
admiration of such authors as Browning, Meredith, and Walt
Whitman, has been, I think, a feeling that their work must some-
how be particularly real and spontaneous, because they have
insisted on doing it in a way in which, according to most well-
constituted judges, it ought not to be done. And conversely,
poets like Tennyson or Swinburne have been in certain circles
despised as a little tame, conventional, uninspired, because they
seemed to be too obedient to the ideals which poetry had
followed before them. I do not specially wish to attack this
modern prejudice, if it is one. I largely share in it : and its
excesses will very likely disappear. But I do very greatly wish
to point out that artistic feeling in this matter has not always been
the same. Artists have not always wished to stamp their work
with their personal characteristics or even their personal name.
Artists have sometimes been, as it were, Protestant or Iconoclast,
unable to worship without asserting themselves against the
established ritual of their religion : sometimes, in happier circum-
stances, they have accepted and loved the ritual as part of the
religion, and wrought out their own new works of poetry, not as


protests, not as personal outbursts, but as glad and nameless
offerings, made in prescribed form to enhance the glory of the
spirit whom they served. With some modifications, this seems
to have been the case in Greece, in Canaan, in Scandinavia,
during the periods when great traditional books were slowly
growing up. Each successive poet did not assert himself against
the tradition, but gave himself up to the tradition, and added to
its greatness and beauty all that was in him.

The intensity of imagination which makes the Iliad alive is
not, it seems to me, the imagination of any one man. It means
not that one man of genius created a wonder and passed away.
It means that generation after generation of poets, trained in the
same schools and a more or less continuous and similar life,
steeped themselves to the lips in the spirit of this great poetry.
They lived in the Epic saga and by it and for it. Great as it was,
for many centuries they continued to build it up yet greater.
• What helped them most, perhaps, was the constancy with
which the whole race — to use a slightly inaccurate word — must
have loved and cherished this poetry. Amid the chaos that
followed /zera ra TpcoiKci, when the works of art, the architecture,
the laws of ordered society, the very religions of the different
centres, were all lost, for the most part never to return, the germs
of this poetry were saved. The fugitives left their treasures,
their gods, and their wives behind, but the sagas were in their
hearts and grew the richer for all their wanderings. They
carried their poetry as other nations have carried their religion.
How strange and significant a thing, after all, is that which we
speak of as either ' the Epic style ' or ' the Epic language '. It
seems more than a style, though, as we have seen, it cannot quite
be treated as an organic spoken language.

For many hundreds of years this wonderful mode of speech
was kept alive to serve nothing but the needs of poetry. The
ordinary audiences must have understood it as well as, for instance,
our audiences understand the authorized version of the Bible,
though the differences between Jacobean and Victorian English
are utterly trifling compared with those between Homer and the
prose speech of classical Ionic inscriptions. And how wonder-
fully the poets themselves knew it ! Even under the microscope
of modern philology the Epic dialect appears, in the main, as a


sort of organic whole, not a mere mass of incongruous archaistic
forms. Van Leeuwen and Monro can write consistent grammars
of it. And this language has been preserved and reconstructed
by generations of men who never spoke it except when they
recited poetry. It was understood by audiences who never
heard it spoken except when they listened to poetry. And not
a man among them had any knowledge of the laws of language ;
1^ they had only a sense of style.

But to meet the special difficulties raised above, let us consider
especially the later generations of these bards and the task
that lay before them. They were poets, 'makers' as well as
' singers ' ; but, much more than that, they were Homeridae^ or
Hoineroti, Paides^ the sons and servants of the greatest of the
poets. None of them dreamed of vying with Homer ; only of
exalting and preserving him. Other people no doubt might
wish for a new style of poetry, for lyrics, for elegies, for iambic
and personal verse. The old Epic language was becoming less
known and more remote. The meanings of some of the words
were taught in schools, others had been forgotten. And the
last bards had before them various books, not very many, it
seems, telling the great legends.

I am not looking for the work of any particular compiler or
harmonizer; I am merely trying to understand the spirit in
which any one of these later poets — how great or how small a
poet matters little — seems to have set about his task. He could
have written an epic poem himself, of course : but who wanted
him to write one ? How should he dare to ? The world was
not yet reduced to such straits as that. There was plenty of the
old poetry still in his power. He knew it by heart, and he
possessed scrolls of it, poetry of men far greater and wiser than
he, voices of those who had talked with gods. Diligently and
reverently he wove it together. He had before him — let us
imagine — a Wrath in which Agamemnon offered no atonement,
and he found besides a lay telling of the Embassy to Achilles ;
or he had before him some battles around an unwalled Greek
camp and found another version with the storming of a wall ; or
perhaps he merely found fragments of other epics too good to
lose and not too firmly rooted in their context to transfer.
Diligently and reverently, with a good deal of simple cunning,

2760 S


he arranged his scheme so as to make room for all. He put
inconsistent passages far apart ; he altered a few words to mask
awkwardnesses and get rid of stark contradictions. He added
lines, when he needs must, to connect or to explain ; always
unobtrusive lines, making no dissonance, borrowed word by-
word, phrase by phrase, from the old poems themselves. And,
amid all this gentle and lowly service, when he rehearsed his
great recitation, when he went over the lines of some tremendous
passage that shook all his being, then, it would seem from the
evidence, there came into him the spirit of the ancient men, and
a voice as of Homer himself The lines that he spoke became
his own. He had always belonged to them, and now they
belonged to him also. And in the midst of them and beyond
them he too had freedom to create.

And we critics, we mete to him a hard measure. When he
creates, we call it interpolation. When he preserves with careful
ingenuity all the fragments that he can save of his ancient Homer,
we call attention to the small joints in his structure, the occasional
incongruity of a simile which he loved too well to let die. If we
knew his name, I suppose we should mock at him. But he has
no name. He gave his name, as he gave all else that was in
him, to help, unnoticed, in the building up of the greatest
poem that ever sounded on the lips of men.

There is, outside and beneath the ordinary rules of art,
a quality possessed by some great books or pictures and denied
to others, a quality of attracting sympathy and causing the
imagination of the reader or spectator to awake and co-operate
with that of the artist. It is a quality that sometimes irritates
a critic, because it acts fitfully and often depends upon accident.
It puts the efforts of art at the mercy of prejudice. Yet, in
a clear air, when prejudices can be laid aside and forgotten, this
quality is seen to be, despite its occasional connexion with very
third-rate things, itself a great thing, like the power of attracting
or not attracting love. And in the last analysis, I suspect, one
will find that this sympathy, like love in general, mostly goes to
the man who both wants it and will duly pay for it. A poet who
strikes his reader as perfect — of course none ever are so — who
makes the impression of having entirely succeeded in saying


what he meant to say, so that he requires no help from others, is
apt to be treated with some respectful indifference. If he actually
seems self-satisfied, then it is much worse. The reader becomes
lynx-eyed for weaknesses, anxious to humiliate, like Ruskin, for
instance, in his criticisms of Guido and the later Renaissance
painters. And there are other poets or artists whose work has
the power of appeal ; the nameless charm and wistfulness of
a thing not perfectly articulate, which means more than it can
ever say, possesses more than it can ever impart, envisages more
than it can ever define. It is the beauty of the ruin, suggesting
the wonderful building that once was ; of the unfinished statue,
suggesting the splendour that should have been.

Of course this conception must not be used as an excuse for
bad workmanship. It is in the essence of the contract, so to
speak, that this appeal to the imagination of others only begins
to act when the artist himself has taken all the pains he can. It
is only the intensity of his imaginative effort which kindles ours
into action. And that intensity will, under normal circumstances,
have made him work his best. only it so happens that the
' greatest imaginings and desires of the human mind are beyond
the greatest powers of words or paintings to utter. And the
best artist, when he has used the very utmost of his skill, is left
at last dependent on the sympathetic imagination of others. If
that fails him, he dies with his meaning unexpressed.

It is in this spirit of sympathetic imagination that we should
^ read most ancient traditional books. And, as a matter of fact,
we generally do so. They are all markedly imperfect, but we
hardly notice the imperfections. How few of us, for instance,
ever noticed that there were two different accounts of the Creation
in Genesis before we were compelled .? How few scholars were
troubled by discrepancies between Iliad I and IT ? How little
we resent the half- inarticulate quality of ancient vocabulary and
syntax ? Nay, we admire them. For the best things that these
books are trying to express are not to be reached by any correct
human words. With all the knowledge in the world at our
disposal, we must needs sooner or later throw ourselves on the sea
of imaginative emotion in order to understand or express these
greatnesses. And the reason why we are willing to do so in
these cases, and not in others, is, I think, ultimately the intensity

S 2


of the imagination behind. The driving of Jehu, the weeping
face of Helen : these have behind them not the imagination of
one great poet, but the accumulated emotion, one may almost say,
of the many successive generations who have heard and learned
and themselves afresh re-created the old majesty and loveliness.
They are like the watchwords of great causes for which men
have fought and died; charged with power from the first to
attract men's love, but now, through the infinite shining back of
that love, grown to yet greater power. There is in them, as it
were, the spiritual life-blood of a people.



There is a well-known list of the seven cities which claim to
be the birthplace of Homer. There are always seven ; but the
names vary so that the actual claimants mentioned amount at
least to ten. ' Smyrna, Chios, Colophon, Salamis, los, Argos,
Athenae ' ; but instead of ' los ' we find ' Rhodos ' and ' Pylos ',
instead of ' Salamis ' sometimes ' Ithake '. Now, without going
into the rather transparent pretensions which have placed some
of these cities on the list, we may notice two points. First,
antiquity in general is quite agreed in regarding Homer as an
Ionian, and it knew the poems only In the Ionian dialect.
Secondly, the two cities which have, in the mere statement of the
tradition, the strongest claim, are also the two of which we know
that they were first AeoHc and only long after Homer's time
Ionian : Smyrna and Chios.^ In both of these Homer was
worshipped as a local hero. Thirdly, the two chief Ionian cities,
Miletus and Ephesus, are never mentioned in the list of birth-
places. That Is to say, the chief Ionian birthplaces prove, on
examination, to be not Ionian at all ; and the tradition, even
while it received and read its Homer in Ionian form, instinctively
felt that the spirit of Ionian civilization at its ripest development
was alien to the spirit of Homer.

The traditional birthplace of Homer floats from Ithaca to
Colophon. His date varies from 1 159, given by some authorities
quoted in Philostratus, to 686, the year assigned by the historian

* The evidences for Chios are: Pind. Nem. ii. i, and Schol. ('O/ir/piSai) ;
the Hymn to Apollo rvcfjXos nvrjp, oiKfl de Xia eVi TrnirraXoea-crr] • cf. Thuc. iii.
104 ; Simonides, ev 8e to koXXkttov X'los (eintv dvrjp, meaning Homer and
quoting Iliad, Z 146; the anonymous Life of Home}-. For Smyrna: a local
sanctuary {'Ofirjpeiov) and statue; Strabo, p. 646; Cic. pro Arch. 8 ; a native
tradition which showed (and still shows to-day) the cave by the river Meles
where Homer was born. See Proclus, Vila Horn., and ' Plut.' Vila Hom.^
Paus. vii. 5. 6, and cf. the name MeXijaiyeVijf.


Theopompus. But he is never born in either of the two greatest
Ionian cities at the time of their power.

The rise of the Ionian civilization is in many ways the most
wonderful phenomenon In Greek historj- j Every kind of intel-
lectual advance seems to have its origin in Ionia, The greatest
works of colonization and commerce, the first banks, the first
maps, and the first effective Greek fleets come from there. The
first prose ^ historian mentioned by tradition is ' Cadmus of
Miletus ' ; the first who has real substance and influence is
Hecataeus of Miletus. The first Greek philosopher is Thales
of Miletus, the second and third are Anaximander and Anaxi-
menes of Miletus. Consider for a moment the strangeness of this
figure of Thales. Before the end of the seventh century, while
the Ionic portions of our I/tad are. still taking shape, Ionia
seems to have been ringing with the fame of this new kind of
great man, not a king nor a warrior, nor even an adventurous
merchant prince, only a (7o0oy di/^p, a wise man : a philosopher,
who has quietly rejected all the myths about gods and theogonies ;
an engineer, able to divert the river Halys from its course ;
a mathematician and an astronomer, able to predict the eclipse
which occurred on May 28, 585 B. C. And this man is not
persecuted like Galileo or Priestley, not dependent on power-
ful protection, like Leibnitz or Descartes. He is an acknow-
ledged leader of his people, a man to consult in crises, when
other nations performed a human sacrifice or took the inarticu-
late and dangerous advice of a sacred snake. A generation
or so later, about 540 B. C, just about the time when the I/z'ad
and Odyssey were taken over to Athens to be recited at the
great national festival, we meet another strange Ionian figure,
a Colophonian this time. Xenophanes is a professional reciter
of his own poetry, whose zeal for the expurgation of ' Homer'
has become so great that he traverses Greece denouncing the
falseness and immorality of the very poems from which his own
performances were originally drawn. All the myths are false.
There is only one God, infinite, all intellect, without bodily parts.
Homer and Hesiod ' tell lies, attributing to the gods all that
among men is a shame and a rebuke, thievings and adulteries^

^ See Radermacher in PhiloL Wochenschrift, igoy, No. 10.


and deceivings one of another '. And another philosopher, not
otherwise sympathetic to Xenophanes, remarks in passing that
' Homer and Hesiod ought to be whipped '.

Now one must not suggest that the tone of these Philosophers
represents the ordinary state of mind of the educated Ionian
public. Thales and Xenophanes, and still more Heraclitus, were
exceptional men. But the existence of an extreme view or
a great advance of thought among a few people is nearly always
good evidence for the prevalence of a more moderate view or
a feebler advance among a much larger number. Before
Xenophanes arose to denounce the moral atmosphere of the
Epos altogether, there had probably been others improving that
atmosphere from within. The spirit of expurgation, which we
studied in the fifth lecture, had already begun its long work of
removing the traces of primitive cruelty and brutishness from the
heroes of Homer.^ It could not make its work quite complete.

^ The limits of date within which expurgation went on are hard to deter-
mine. Some bold Paleian might argue that all the expurgation is a late
Attic process, on these lines: (i) We find it still going on in the time of
Zenodotus (see p. 124 note); (2) Attic tragedy, being early, mostly follows
the unexpurgated versions of the sagas ; (3) The argument below, that
Aeschylus seetns more primitive than Homer, may be interpreted as simply
showing that Aeschylus was so, and that the ' Alexandrian ' treatment of the
Gods in Homer really is Alexandrian — i. e. belongs in date to the fourth or
third century, and that Zenodotus was right in considering the passages in
which it occurs as spurious. To this we may answer: (l) Expurgation is a
normal and constant process, always acting when the next teller or hearers
of a story have any different standard of sensitiveness from the last.
There is no reason it should stop until the text is fixed firm. We find as
early as Xenophanes not only a spirit which must have produced expurgation,
but a standard of ethical criticism so exceedingly high and severe that it can
scarcely have been the first, or anything like the first, of its kind. I mean :
before people got to complaining that Homer's gods were in human shape,
they must in all probability have complained of more obviously objectionable
things. (2) This is important, but easily answered. Attic tragedy is in
a different convention. It takes its legendary material comparatively un-
expurgated because its characters are (comparatively) unidealized. Achilles
can torture Hector in the Ajax or the Andromache, Odysseus and Agamem-
non can slaughter Iphigenia in the Agamemno7i or the Iphigenia Taurica,
because those heroes are not set up as models of chivalry ; in many cases
they are definitely meant to be 'unsympathetic', and within limits, the
wickedv^r they are the better. The Iliad, on the other hand, was the ' mirror
of chivalry ', a recognized instrument of moral education because it repre-
sented an idealized heroic age. Agamemnon in the Iliad could not sacrifice
Iphigenia, just as Tennyson's King Arthur could not burn Guinevere alive
or tear her between wild horses, whatever the old legends might say in
either case. Such deeds would be out of the convention, and shocking. (3)
The frivolous treatment of the gods in serious or romantic literature is a
convention which probably, like much else, comes to Alexandria from Ionia.


Yet if it had done for the gods what it did for the human beings
there would not have been much ground left for the indignation
of Xenophanes.

But there seems to be always a limit to these processes of
expurgation and reform from within. A progressive nation
with a rich legendary tradition must from time to time wake up
to look upon its legends with fresh eyes. They are regarded as
something authoritative, unquestioned, indisputably edifying.
And yet in them there are here and there details which seem
hard to believe, harder still to admire. They are explained,
allegorized, altered, expurgated. For the moment all is well.
And then quickly there appears another crop of difficulties
requiring the same treatment. The process is repeated. The
amount of hard thinking and of emotion which mankind has
again and again expended — perhaps wisely — in trying to patch
the fragments of some great system of false beliefs, which often
has nothing valuable about it except the emotion with which it
happens to be regarded, is one of the most profoundly charac-
teristic things in human history. It was widely prevalent in
Greece, especially after the classical period. But a moment is
apt to come, sooner or later, at which men begin to wonder
whether after so much jettison there is really anything true to
save, whether a bridge so extremely full of rotten planks is worth
such repeated mending. The point at which this stage is reached
seems to depend on a certain proportion of qualities in the minds
of the persons affected, the proportion between their critical
intelligence and boldness on the one hand, and their reverence
and depth of emotion on the other. Now Ionia in the sixth
century was full of intelh'gence and daring ; it was adventurous,
critical, scientific, rationalist, and self-confident. It was not, like
Thrace, Crete, Athens, South Italy, a centre of religion or
reactionary dreaming. It produced indeed some mysticism ; but
a peculiar scientific and speculative mysticism of its own, more
concerned with the properties of the Infinite [to direLpov) than
with the traditional anthropomorphic gods.^

(Aristophanes is of course quite different.) This subject is too large to
discuss here. But we know that Zenodotus and Aristarchus regarded the
frivolous scenes as aTrpsTri) ; that is, they were not natural and suitable
according to Alexandrian taste.

^ See Schulz, lofiische Mystik. Also Die Gotterwelt Homers by Paul
M&ytr, Jahfesbef. ilber d. K, Klosterschide zu Ilfeld, 1907.


This scientific and critical temperament among the people of
Ionia was met by a special weakness in the Homeric religion.
It was not really religion at all. The beautiful Olympians whom
we find in Homer forming a sort of divine family, and whom we
know from statues, do not represent the Gods worshipped by
any particular part of early Greece. They represent an en-
lightened compromise made to suit the conveniences of a
federation. Each local god had been shorn of his mystical or
monstrous characteristics ; of everything, that is, that was likely
to give offence. And it is nearly always the mystical or monstrous
elements of a belief which seem to have excited the keenest
religious emotions of an ancient people. The owl Athena, the
cow Hera, the snake-man Cecrops ; the many ghosts and shapes
of terror ; the mystic bull Dionysus, who is in some strange
sense the beast which he himself tears to pieces alive, and from
whose blood our souls are made : these things are cleared away
from Homer's world, or else humanized and made to tone in with
his general serene anthropomorphism. This anthropomorphism
happened to suit the art of sculpture, which became highly im-
portant in Greece, and for that reason among others the Homeric
gods have dominated the later tradition. But the real worship of
Greece before the fourth century almost never attached itself to
those luminous Olympian forms. There were many ecstasies of
enthusiasm and outbreaks of superstition in Greece, but they
all depend on deities of quite a different sort. There was
enthusiasm for Orpheus and Dionysus: enthusiasm for the
mysteries of the Mother and Maid at Eleusis. There was
religious feeling about the local pre-Hellenic festivals, like the
Thesmophoria. There was superstitious terror in Athens about
the mutilation of the Hermae. But those Hermae were no
images of the handsome young Homeric god ; they represented
the old divine boundary stone, whose unedifying form has been
entirely expurgated from the Homeric epos. The failure of
Nikias in his retreat from Syracuse was due to reverence for no
Homeric Artemis, but for the ancient and unhumanized holiness of
the Moon. Even the goddess who led Pisistratus back to Athens,
Pisistratus tov 'O/xrjpiKcoTaToy, was originally not so much the
Homeric daughter of Zeus as the ancient pre-Homeric ' Athenaia
Kore '. And the temple of Zeus, which the same Pisistratus, in the


spirit of his Homeric policy, proceeded to build with so much
pomp, was left all through the classical times unfinished. All the
treasures of Athenian building went to Athena and Poseidon,
the native Earth-Maiden and the native Sea. Of course Athens
may have been a specially ' Pelasgian ' community : but fnutatis
intitandis the same observations could probably be made of any
Greek town of which we possessed adequate records.

One can see then what was likely to happen to the Homeric
gods. They had been made, up to a certain standard, very
beautiful, highly anthropomorphic, not in the least poverty-
stricken, barbarous, or grotesque. But in the process they
had lost their special hold on the worship of any particular
community. They had forfeited the powerful support of un-
critical local superstition : and, after all, in the eyes of an
educated and sceptical Ionian, would they quite bear thinking
about ? This serio-comic Olympian family, with its permanent
feud between the husband and wife, in behalf of which we can
but lamely plead that the wife's unamiability is the natural
result of the husband's extreme unfaithfulness, and the husband's
unfaithfulness almost excused by the wife's monstrous unamia-
bility ? The lame son at whom the other gods laugh .-' The
pretty daughter, always in scrapes and tears.'' To a reverent
spirit these things can be allegorized. To a scientific historian
they possess an historical origin and explanation. But to the
critical Ionian, whose eyes are no longer blinded by the sacred
past, who patronizes while he loves, they tended to take a curious
form. It is a form hard to characterize or to understand, unless
perhaps it is an imitation by the taste of a refined and sceptical
age of the simple-minded burlesque of sacred things which is
often found in primitive ' ages of faith ' : the form which reaches
its highest, or perhaps I should say its lowest, point in Ovid, or
before Ovid in the Alexandrians. The gods are not by any
means rejected. They are patronized, conventionalized, and
treated as material for ornament. Their traditional character-
istics, roughly speaking, are preserved ; Zeus is royal, and
Apollo is musical, and Athena is a warrior or a spinster: and
the late Ionian poets believe in them not much more effectively
than Pope believed in the sylphs who tire his heroine's hair in
the Rape of the Lock. There is a depth of unbelief profounder


than any outspoken denial. Pope would not have troubled to
deny the existence of sylphs. When you take the gods in such
a spirit as this it is not worth while to furbish up their moral
characters. They are more amusing as they stand ; they may
even be, in a certain external and shallow sense, more beautiful.

I think that in this matter of the Homeric or Olympian gods
one can notice three distinct stages. There is a primitive stage,
represented best by the earliest strata of Hesiod's Theogony :
a stage in which, for one thing, men did not use their critical
faculties at all on this sort of material, and, for another, a great
many of the myths which afterwards became shocking or ridicu-
lous still preserved some remnant of their original meanings.
At such a time, for instance, the quarrels between Zeus and
Hera may still have been felt consciously as part of the old and
respectable feud between the conquered native goddess and the
invading patriarchal god.^ Secondly, there is a long middle
stage of expurgation, of rejection, of humanizing. When it
began we can hardly guess, nor how the expurgations gradually
came to be accepted and canonized in the official texts ; but the
process must, in some form or other, have lasted through a great
part of the life of the poems. Thirdly, there is the late Ionian
stage of which we have just spoken, in which the Olympians have
ceased to have any genuinely religious significance, but serve to
provide expedients to the story-teller, and afford material for a
kind of half-licentious humour.

Presently, I think, we shall see reason to add a fourth stage,
that of the acceptance of the Homeric system by non-Ionian
Greece, a stage in which the more primitive Greek communities,
beginning to feel uneasiness at the muddle and crudity of their
own local superstitions, receive with reverence and enthusiasm
the comparatively orderly and civilized system of Homer. In
the sixth century, when Ionian culture spread in a great wave
to the mainland of Greece, Ionia was probably already blasde
to the theology of which she was the chief centre. And the
Zeus whom Aeschylus accepted from Ionia and Homer was a
widely different being from the Zeus of whom the men of Miletus
made merry tales.

^ J. E. Harrison, Pri/ner 0/ Greek Religion.


At the very outset of that interesting branch of literature
which culminated in the Greek Novel, we hear of the Milesian
Stories. Light tales they seem to have been, much in the style
of Boccaccio. A typical one is the tale of the inconsolable
widow of Ephesus, who used constantly to frequent her hus-
band's tomb — from mixed motives ; partly from devotion to his
memory, partly because there was a fascinating young soldier on
guard there. The first collector of such stories whose name is
known to us, Aristides, belongs to an uncertain but much later
date. But two or three tales in Herodotus bear the same stamp :
among them some, like that of the wife of Candaules, which were
certainly not first told by Herodotus. And besides, the very fact
that Aristides called his collection ' Milesian Stories ' seems to
mean that the type of story was already recognized as Milesian.
It was a name like ' Contes gaulois '. And I think one can
see this spirit, a mocking, half-licentious, Boccaccio-like spirit,
already at work in the later, and not the very latest, parts of
the Iliad.

We will take two detailed instances. But first, let us be clear
about the issue. As we have seen before, the human beings in
Homer always maintain their dignity and self-respect. No hero
is a liar^ or a coward. None is drunken or loose-lived or
vicious. None tortures his enemy. But the gods : that is quite
a different matter. They are capable of anything. They not
only practise torture — the gods of most nations have had a
weakness in that direction — but they lose their dignity. They
are cheated, beaten, imprisoned. They lie and are found out.
They are routed by human beings. They howl when wounded.
Their father ' bangs ' them ' about the house '. That, you may
say, is characteristic of all simple and primitive religions. Does
not Ouranos swallow his children and again vomit them up }
Does not the Babylonian Apsu, in the primaeval chaos, cut his
wife Tiamat in two, to make one half of her into heaven and the
other into earth .^ Yes. Those are simple and savage stories,
visibly allegorical, dependent in part on the mere helplessness
of primitive language. The Homeric passages in question are

^ Of course a disguised hero in the course of a dangerous adventure tells
the necessary lies to avoid detection. That is in the essence of all romances
of adventure.


totally different from that. They are not primitive, but smooth
and sophisticated. They mock with easy scepticism at the
indecorousness of the primitive beliefs.

But let us take our two instances. There was in Greece a
widespread tradition of the Wars of the Gods. Zeus somehow
holds his power by conquest over other beings, vaguer, older,
and darker shapes, belonging to some old order, or, perhaps, to
the chaos that preceded all order. We hear of many treatments
in early epic of the Titanomachia, Theomachia, Gigantomachia.
And in our Hesiodic collection we have preserved, imperfectly
and with many repetitions, due apparently to a conflation of two
sources, a long fragment of a Titanomachia. It tells how Zeus
gained the victory over the Titans by freeing and calling to his
aid certain primitive beings whom the Titans and Ouranos had
oppressed {Theog. 617 ff.).^

Briareos and Kottos and Gues, their father Ouranos con-
ceived hatred of them in his heart, being afraid at their wild
valour and their looks and tallness, and he bound them in
bondage deep beneath the wide-wayed earth. And there
they dwelt in anguish under the ground at the ends of the
great world, seated on the verge of things, a very long time,
amazed and with great mourning in their hearts. But Zeus
and the immortal gods, by the counsel of Earth, brought
them again to the light.

Zeus asked them to help him in the long war against the Titans,
and they consented. The gods stood on Olympus and the
Titans upon Othrys ; and they had fought already for ten years.
So they joined battle :

And the Titans opposite had made strong their lines, and
both sides put forth their might. And there was a terrible
cry from the boundless sea, and shattering of the earth, and
the broad sky groaned, and high Olympus was shaken from
his foundations with the rush of immortal things : and the
quaking and the noise of feet upon the steeps came down
unto cloudy Tartarus. . . . And the armies met with a great
shout, and Zeus held back his fury no more. Down from
Olympus and heaven he came in one sweep of thunders

^ If Briareos is a fifty-oared ship, as seems likely, he must have been
introduced later into this story. But perhaps the Fifty-oar was rather identi-
fied with an already existing Briareos, and thus Briareos identified with


that ceased not : and the bolts went winged from his mighty
hand, and the life-bearing Earth cracked with the burning,
and around him the fathomless forest roared in fire, . . . And
foremost in that bitter stirring of battle were Kottos and
Briareos and Gues, unsated of war, who cast from their
hands three hundred great stones, one on another, and
darkened the Titans with their castings, and drave them
down and bound them in bitter bondage, for all their pride,
as far beneath the earth as the sky is above the earth. For
a bronzen anvil cast from heaven would fall nine nights and
days, and on the tenth night would come to the earth. And
from earth a bronzen anvil would fall nine nights and days,
and on the tenth night would come to cloudy Tartarus :
whereabout there is driven a bronzen fence, and around it
Night is shed, Night in three floods. And over it the roots
are planted of the earth and the unharvested sea.

Now the exact merit of this as poetry may be a matter of
dispute. It may be a little incompetent, a little bombastic. But
it is at least genuine and reverent. If we are to describe these
primitive battles of gods, that is the kind of way in which to
conceive them.

Now turn to the battle of the gods in a late part of the Iliad

It was shield-piercing Ares who began, and sprang upon
Athena with his bronzen spear, and uttered a word of insult :
' Wherefore again, thou dog-fly, dost drive the gods to strife ">
Rememberest not the day when thou didst let loose Diomedes
to wound me, and thyself in sight of all didst grasp the spear
and drive full at me and tear my fair flesh ? Now I warrant
me thou shalt pay for all thy doings ! ' So saying he made
a lunge at her aegis tasselled and terrible^ which not the
thunder of Zeus can make to fall. There bloody Ares lunged
with his long spear. But she started back and caught up in
her stout hand a stone lying upon the plain, a big black
jagged stone, which men of old had put to be the boundary
of a field ; and she hit Ares on the neck with it, and his
limbs gave way. He reached over seven furlongs as he fell,
and his hair was filled with dust and his arms rattled about
him. And Pallas Athena laughed aloud, and boasted over
him with winged words. ' Fool, hast thou not learned yet
how far I am thy better, that thou wilt dare to match thy
strength with mine .? That is the way to fulfil thy mother's
curses, who plans anger and mischief against thee for desert-
ing the Greeks.'


Presently Aphrodite, who was in love with Ares, came and
took him by the arm to help him up, while he made a great
groaning, and began gradually to come to. Hera saw, and
called to Athena :

* Here is that dog-fly ' — the poet has an affection for that
word — ' coming to help Ares. Chase her ! ' So Athena,
rejoicing in her heart, flew at Aphrodite, and drove her in
the chest with her stout hand, and her limbs and her dear
heart gave way beneath her. And there the two of them
lay together on the many-nurturing Earth.

Later on, towards the end of the battle, Artemis is facing
Hera :

To her in wrath spake the reverend spouse of Zeus :
' What seekest thou, shameless she-dog, standing against
me ? ' . . .So spake she, and with her left hand gripped both
the hands of Artemis by the wrist, while with her right she
took the bow and arrows off her shoulders ; then with the
bow and arrows whipped her about the ears, and laughed as
she dipped her head this way and that. And the arrows
kept dropping from the quiver. And the goddess full of
tears fled like a wood-pigeon.

' one of the few passages in the Iliad^ says Dr Leaf, ' which
can be pronounced poetically bad.' True, yet the badness lies
entirely in the taste, not in the execution. The verses are
admirably written, incomparably better than those of Hesiod's
Titanomachia. But the poet was not writing about anything
that he felt as real or as mattering much to anybody's feelings.
He was almost writing parody or mock-epic. And he made it
quite pretty !

Let us take another instance. Among the old traditional
subjects of semi-religious Epos was one which our extant remains
of Greek literature leave rather obscure, the mystic marriage of
Zeus and Hera. This may have been in its origin a sort of
marriage of Heaven and Earth, or of the two greatest divine
beings, from which all things arise. It may conceivably have
symbolized the union of the two races and two religions — the
patriarchal Zeus of the Northerners, being united with Hera, the
Argive Kore. It may have been one of those naive recognitions
of the mystery and divinity of the processes of life, which often


shed such high dignity upon the external grossness of primitive
religion. Whatever its origin, it was a subject treated by divers
poets with reverence and mystery, as we can tell by the allusions
in Pindar, Aeschylus, and Euripides.

Now, how is this subject treated in the Fourteenth Book of
the Iliad ? Absolutely in the spirit of Boccaccio : I might almost
say, of a Palais Royal farce. The passage is sometimes much
praised, and is certainly admirably written : ' radiant with
humour, grace, and healthful sensuousness,' is the criticism of
Dr. Leaf. But what is the story ? Its name is almost enough :
it is called by ancient writers The Tricking of Zeus.

The father of gods and men was sitting on the top of many-
fountained Ida, watching the war. The gods had offended him
by giving secret help to the Greeks, and he had arranged that
the Trojans should win the present battle. So he went himself
to sit on Mount Ida, and see that all proceeded as he desired.
His wife Hera, a partisan of the Greeks, saw him sitting there —
aTvyepos Si ol eTrXero 6vfXM — ' and thought how much she disliked
him ! ' She determined to outwit her lord and master. So she
went to her room, washed, anointed, and scented herself, and
put on her best immortal raiment, including ear-rings with three
stones in them. Next she went to Aphrodite and begged for
the loan of her Cestus, or embroidered girdle, which acted as a
love-charm. She explained — falsely, of course — how she wished
it in order to reconcile an old married couple dwelling at the
end of the world, who had unfortunately quarrelled — Okeanos
and mother Tethys, in fact. Having obtained the Cestus, she
proceeds to find the Spirit of Sleep, and with some difficulty
bribes him to come and be ready to charm the eyes of Zeus at a
critical moment. The bribe has to be high, since Sleep had
done her the same service once before, in the old Heracleia^ and
had suffered in consequence. Finally, she repairs to Mount Ida,
to ask in most dutiful language the permission of Zeus to make
her expedition to Okeanos and mother Tethys. She does not
like to go so far without her lord's approval. Remember that
all this edifying story began by her thinking how much she dis-
liked Zeus ! I can find no dignified word to describe adequately
her provocative conduct towards her victim. However, she
succeeds in entirely engrossing his attention, and so rouses his


passionate admiration that he compares her favourably with no
less than seven other females towards whom he has entertained
similar feelings. He quite forgets the war. He goes to sleep
in her arms. And Hera sends a message to her allies that they
can do what they like now : Zeus is safe !

Now, were I required to subscribe half a crown to save
Aristides of Miletus and all his children from everlasting death,
I do not say that I would outright refuse. In its own place
this kind of literature has a certain value, and seems to have
served as a stimulus to better work in others. But not all the
riches of Egyptian Thebes could, I think, ever atone for the
injury done to the human race by the invasion of this Milesian
spirit into what is perhaps the greatest poem of the greatest
nation of poets that the world has known. It has defiled its own
beautiful world. It has ' slain the image of God, as it were, in
the eye '. For the poets who actually wrote these passages there
is a great excuse. Their cause was, perhaps, on the whole,
rather a good cause than a bad. But historical circumstances
combined to catch and stereotype the epic at the moment when,
perhaps just after the zenith of its glory, it had caught this
mocking infection. Rightly sceptical towards the authorized
gods and their legends, it had not the serious courage simply to
seek truth and reject falsehood in what are generally regarded
as the highest regions of human thinking. It neither denied its
gods nor remade them. It degraded them further, and used
them for ornament and amusement, to make a good tale the
merrier. I had almost written, to make a good tale into a bad
one. When once this infection has crept into its blood, the Epos
as a form of living and growing poetry was doomed.

Consider what that meant for the history of Greek literature.
Greek literature starts from an immense wealth of Saga traditions,
and the need of an instrument for expressing them ; to meet that
need it created the Epos. It had been a costly and a rare
creation ; a metre, a style, a whole language almost. And now
that part of the Greek people which had done all this for the sake of
the Saga had outgrown the Saga, and was beginning to parody
what it had formerly adored.^ Had Ionia been the whole of

' Monro allows quite a large place to the mock-heroic in the second part
of the Odyssey, Telemachus' sneeze which ay-epboKiov Kom^rja-e (p 542), the

2700 T


Greece, not only the Epos, but the whole heroic tradition, might
have died during the sixth and fifth centuries. But Ionia was
not the whole of Greece, and the Saga found a new utterance in
Attic tragedy.

I always hesitate to use the antithesis of northern and native,
or Hellenic and pre-Hellenic, as applied to the whole of any con-
crete fact. The rule is that everywhere you find northern and
native elements, but nowhere do you find a purely northern or
purely native community. Yet in contrasting the Epos with
tragedy that antithesis cannot but occur to one's mind.

When the ancestors of the Aeolians and lonians fled across the
seas — a mixed set of races, often under Achaean leaders — they
were compelled, as we observed in the second lecture, to leave
behind them their sacred places, most of their tribal and family
institutions, and notably the graves of their fathers. The prestige
of the Achaean chiefs, the partial return to migratory life, the
convenience of the Northern institutions of the Saga and the
Bard, combined to give to the Epos its prevailing Achaean tone.
But on the mainland of Greece during all this time, even where
the northern occupations were most tyrannous, there remained
always some fragments of the old population, peasants and serfs
and outlaws for the most part, who still clung to their old
objects of worship, their Earth-Maidens and their harvest magic,
especially their tribal initiations and their sacred tombs. A
downtrodden people they must have been for many generations,
worshipping by stealth and in fear. But as the populations
became more mixed, which was the case everywhere on the
mainland, the result was that the old pre-Hellenic stratum of beliefs
and emotions re-emerged. How the initiation rites led to the
formation of an initiation-god Dionysus, the Zeus-Child who

pigsty described in language borrowed from Priam's palace (| 13 ff.)> the
TTOTvta iMTjrrjp of the beggar Irus (o- 5), &c. He gives some fifteen alleged in-
stances in the index under ' Parody '. Miilder goes much further, Que//en,
pp. 287 ff., 347 ff-

Exactly the same spirit occurs in the Pelerinage de Charlemagne, which,
however, belongs to a quite early and good period. See G. Paris, PoSsie du
M. A., \. pp. 119-49. It can be shown on other grounds to be connected
with the neighbourhood of Paris (e.g. it mentions no towns except St. Denys,
Paris, Chartres, and Chateaudun, with no word of Aix or Laon), and the critic
regards its heroi-comic character as ' le plus ancien produit de I'esprit parisien '.
Perhaps the Demodocus lay, which looks exceedingly ancient, occupies the
same place in ' I'esprit mil^sien '.


died and rose again, the God who showed the candidates for
initiation to their dead ancestors and led his rout of masked and
dancing ghosts ; how this worship of Dionysus, combined with
the old custom of performing rites round the tomb of a dead
hero, narrating his deeds and sufferings and invoking his return
to his people : that story is too long and intricate to attempt
here. In even the latest works of Attic tragedy the Masquers of
Dionysus are rarely dissociated from some sacred tomb. In this
severe, earnest, keenly emotional atmosphere, touched with
mysticism by the shadow of present death, the Greeks of the
mainland kept up in their separate cities and villages their own
local fragments of the heroic saga.

Now about this time of the decay of the Epos, Athens had
thrown off her ages of Pelasgian slumber and was just coming
into intimate contact with Ionia. To her young and groping
genius the high civilization and intelligence of Ionia, the magnifi-
cent form of the Epos, the broad sweep of Homeric pan-Hellen-
ism, the clean and lordly northern spirit, came as a world of
inspiration, and quickened the ancient ceremonials of worship at
the tomb to the splendid growth of Attic Tragedy.

Turn from that late Homeric story of the Trtcktng of Zezt,s
to the earliest, crudest, most incompetent tragedy which we
possess, though, in its way, one of the most beautiful, the
Stippliant Wome7t of Aeschylus. It is not only that there is
a marked change of atmosphere, but it seems like a change
backward, not forward, towards an older, a simpler and a
grander, world. The very first words of the play strike a key-
note : Zei)? ix\v d(f)i<TO)p, ' Zeus the Suppliant '. Would any of
those clear-headed Homeric bards have ventured on that ancient
phrase? They knew of a Zeus who, on a far-off mountain
throne, observed and avenged suppliants. But this Zeus of
Aeschylus is himself the suppliant ; the prayer which you reject
is his very prayer, and in turning from your door the helpless or
the outcast you have turned away the most high God. The
belief was immemorially old.^ It was doubtless in a thousand of

^ The discovery of the Hymn of the Kouretes enables us exactly to under-
stand Zevf 'AcpiKTap. He is a 'projection' of the rite of Supplication; a
conception generated from the band of human suppliants just as Zeus Kouros,
or simply 6 Mijia-Tos Kovpos, is generated from the band of Kouroi, Silenus
from the Silenoi, Pan from the Panes, or, a very clear case, Amphictyon from

T 2


its ramifications foolish and absurd. And the Ionic Epos had
made all its beliefs sensible.

I will venture to read you a strange Aeschylean lyric about
a deed of this same Zeus. It is a story far too primitive and
monstrous for Homer : the tale of lo, the Argive maiden beloved
of Zeus, who was turned into a cow, forsooth, and watched by
the hundred-eyed Argos, and driven over the world by a gad-
fly ! A cow-shaped, or even a cow-headed, maiden ! And a
cow-headed maiden beloved by Zeus! To a cultivated Ionian
such conceptions must have belonged to the very lowest regions
of ' Pelasgian ' folly. They had been expurgated from Homer
generations before. Yet out of that unpromising material
Aeschylus extracts something which is not only genuine reli-
gious thought, but, to my feeling, even somewhat sublime
thought. The love of Zeus leads its object through unearthly
shame and suffering to a strange and overwhelming reward.
We cannot understand. But Zeus is bound by no law but his
own supreme will. He has always his own great purpose, and
he moves towards it by inscrutable ways.

I should explain that to the mythologist lo is probably one of
the many shapes of the horned Moon, the wanderer of the sky.
She was identified by the Greeks with the Egyptian Isis, and her
son — conceived miraculously by the touch of the hand of Zeus —
with Apis, the sacred Egyptian bull The speakers are the
daughters of Danaus, descendants of lo, returned to her native
land, Argos, and praying protection from their pursuers, the sons
of Aegyptus {Snppl. 524 ff.) :

Lord of lords, blessed among the blessed, of perfections
most perfect strength, O happy Zeus, hear us, and let it be !
Shield us from the pride of man, whom thou righteously
abhorrest, and whelm in the dark-blue deep our black
prison-house.^ Look upon the woman's cause ; look on
the race born of old from the woman whom thou didst love,
and make new the joyous tale. Be a rememberer of many
things, O thou whose hand was laid on lo, Lo, we are
beings born of thy race, though sent from this land to dwell

I walk again in the print of ancient feet, where our

the Amphictyones. See references on p. 76, note. ' Le dieu est le desir
(collectif) personnifie,' Doutte, Magic dc rAfriqtie dzt Nord, p. 601.
* i. e. the ship of their pursuers.


mother was watched, moving among the flowers; the
meadow of kine, whence lo fled, sea-tossed by a burning
pain, knowing not her desire, to pass through many tribes
of men. . . .

Her wide wanderings are then described, across the Helles-
pont, through Asia southwards, till she reaches at last * the all-
pasturing garden of Zeus, the snow-fed meadow visited by the
whirling giant of the desert-sand, and the water of Nile un-
touched by sickness '.

Do you observe how deeply and simply serious it all is?
Aeschylus accepts the whole story. But because he is simple-
minded and great-minded, and has not a grain of lewdness
anywhere in him, this old, barbarous, pre-anthropomorphic
superstition has become to him a great and strange thing ; and
the spirit passes from the poet himself to his reader. He throws
no veil over the cow-shaped heroine. The transformation is
part of the mystery, and he emphasizes it. The poem continues :

And men that had then their habitation in the land, their
hearts were shaken with fear at the strange sight, a Being
agonized half-human, part of the race of kine and part of
woman. They marvelled at the mystery. Who was it that
brought her peace in the end, her the far-wandering, the
afflicted, the gadfly-goaded lo ?

He who ruleth through ages of unresting life, Zeus [to
whom years are as yesterday]. The unwounding strength
of a hand, the breath of a god, gave rest to her, and her
heart flowed in a sad tenderness of tears. The word of true
promise became a divine seed within her, and she bore a
blameless child, through ages long perfect in happiness.

Whom of gods shall I praise for works more justified ?
Father, planter of the garden, worker with the hand, and
Lord, thinker of ancient thought, great builder of our race,
Zeus, whose breath maketh all accomplishment !

He hasteth not at the command of another. Being
stronger than all, he maketh great the weak. None sitteth
above him, and he honoureth none. And the deed and the
word are present as one thing, to dispatch that end whereto
the counselling mind moveth.

The story which Homer rejected has become the vehicle of a
theology higher than Homer's, or, if not higher, at least based on
deeper thought and involving the reconciliation of vaster conflicts.
The mind of Aeschylus was possessed by one of the problems,


perhaps the most dreadful problem, of human evolution. He
sees the higher asserting itself gradually over the lower in the
process of years ; but he sees also, what many people blind their
eyes against, that the so-called higher often achieves its end at
the price of becoming something more evil than the wild beasts.
It is good that the white man should supersede the red and the
brown ; but what things the white men have done in the process !
For Aeschylus the contest was probably present in two forms :
a conflict, externally, of Greek against barbarian, and in Greece
itself, of what we may call Achaean or Olympian against
' Felasgian '. Zeus was in each case the spirit of the higher
power ; and, to Aeschylus, probably, if anything on earth
specially typified Zeus, the new conqueror and orderer of heaven,
it was the new Dominion of the Athenian Empire.

It was unlike a Homeric bard to have such thoughts at all.
It is still more unlike him to express them in the language of
the Saga. He was a trained artist, and would not dream of so
violating his convention. He kept his poetry in one compart-
ment ; his speculation, if he had any, in another. But for
Aeschylus they are both one. Two of Aeschylus' earliest
trilogies seem to deal explicitly with this subject. Both trilogies
are represented to us by one play each, the Suppliant Women
and the Prometheus. In the two isolated plays which remain,
the sympathy is entirely on the side of the weaker : it is for the
suppliant women against their pursuers, and for Prometheus
against Zeus. Yet we know from other sources that in the com-
plete trilogy the ultimate judgement was for the stronger, so
soon as the stronger would consent to merge his strength in love.
The story of lo is prominent in both plays. It is only loosely
connected with the main plot, but it typifies in each case the
religious meaning of the whole. Zeus did to lo what seemed
like monstrous wrong ; professing to love her, he afflicted her
and ceased not, and the end was that he brought her to a
perfect joy which — so she is perhaps at the end willing to
believe — could not be attained otherwise. And even while
Prometheus and lo are mingling their griefs against Zeus, it is
shown that a child descended from lo is to be also the deliverer
of Prometheus {Prom. 'j'j2^ 871 ff.). That too is part of Zeus'


We know Shelley's magnificent treatment of the Prometheus
Saga. Shelley was too passionate a friend of the oppressed
ever to make terms with a successful tyrant, be he man or god.
In Shelley's Proinethezis Unbound the prophesied catastrophe
which is to hurl Zeus from his throne actually occurs, and the
tormented Universe, awakening to a life of peace and love, finds
uncontrolled that inward perfection of order which leaves no
place for external government. But in Aeschylus we know that
the end was different. Zeus the all- ruler must always rule.
Does not each one of us know, as a matter of fact, that Zeus and
not Prometheus is now governing the world ? But Zeus, who
came to his throne by violence,^ learns as the ages pass that
violence is evil. For all his wisdom he grows wiser still. Nay,
it seems that even from the beginning, in his cruelty to Prome-
theus, as in his cruelty to lo, he had a great purpose in the
depth of his mind, and that purpose was peace. Prometheus is
unbound, not by a turning of the tide of war, but by the atone-
ment, after ages of pain, after the suffering by which alone
wisdom is born, of a noble rebel and a noble ruler. The Zeus
who could be himself a suppliant, who even in the most ancient
legends forgave and set free his conquered Titans, was capable of
this crowning strength also. I do not suggest that this solution is
ultimately tenable or satisfying. But it at least represents intense
thought, and thought naturally expressing itself in the medium of
poetry. It is just this which Ionia never gave us. It is peculiarly
the gift of Athens.

We have tried to follow, in a very imperfect and sometimes
inconsequent manner, the life of Traditional Epic Poetry in
Greece. We have seen the first fragments of what was after-
wards the Greek race gathering behind their bare walls on
islands and desert capes in the Aegean ; we have caught glimpses
of ancient and diverse memories of tribal history, of great deeds,
of rich palaces and mysterious kings, meeting and parting and
re-joining again into the numerous heroic poems now lost, and
the two, more highly wrought than the others, which still survive.
We have noted how, of these two poems, one again was more
* Homeric ' than its companion ; more carefully purified and

> Cf. Verrall on Ag. 192 ff.


expurgated, more tensely knit and gorgeously worded, while at
the same time the heroic and ancient atmosphere was more
sedulously protected from the breaths of commoner or more
recent life. We have looked as best we could, much helped by
Hebrew parallels, into the strange processes of growth and com-
position which have made the Iliad what it is, and have tried to
analyse some part of its poetical greatness. Lastly, we have seen
how the races which built up ' Homer ' at length outgrew him,
and found other subjects than the Heroic Saga in which to express
their ideals and satisfy their intellectual thirst. Homer did not
die ; on the contrary his greatest fame, his most secure enthrone-
ment among poets, was still before him. We shall see in the
next chapter something of what Athens did for Homer, and shall
perhaps be forced to recognize that the text which we possess is
not a thing of pre-Pisistratid, almost pre-Ionian, antiquity, but
actually, as a text, less ancient than the Agamemnon or even
the Bacchae. But whatever work Athens may have done for
the Iliad and Odyssey it is extraordinary how strictly she kept
up the old Homeric convention, the old language, the old manner,
the old subjects and rules of thought. The preservation of the
Ionic Epos in Athens throughout the fifth century is a cardinal
instance of that sensitiveness to style and tradition which is one
of the deepest characteristics of all Greek art. But, after all, it
was tradition rather than creation : when we seek the great
creative work of the fifth century we find it in other paths, with
which Ionia has little to do.^

We have moved into a sterner land, more Interested in truth
and less In romance ; Into a language less beautiful, more
Intellectual, more highly differentiated ; a language which has
elements of hard prose mixed with Its poetry, and has lost that
splendid and careless gleam by means of which Homer was
accustomed to set all themes in the world aglow. Homer's
poetry was so easy, the sympathy was so clear, the Imagination
was roused so instinctively, that we must leave It with a sigh.
And this new poetry is of a kind which will not yield Its

^ Professor Wheeler of Columbia University calls to my notice the very
similar contrast between the mocking boisterousness of the Ionic vase-
paintings and the severity of the early Attic. See also Mr. Cornford's
remarks in Thucydides Mythistoricus on the difference between the Ionic
Herodotus and the Attic Thucydides.


treasures without hard thinking, without somewhat intense and
vigilant use of the imagination. The poets, for the most part,
are no longer merely singing to please us, according to methods
which have been tried for generations and proved effectual.
They are men not exactly less cultured — intellectually they are
far greater — than the Ionian bards ; but they are less accom-
plished. They are imaginatively nearer to the primitive earth-
born tangle of desires and wonders. Their feet are set in places
lower than Homer's feet ; their thoughts strive towards heights
and obscurities which his poetry dared not penetrate. They
have fought at Marathon, and their hands are reshaping the
world. The bitterness of truth is mingled with their dreams of
beauty ; the passion of men searching gleams through the
stiffness of their majestic conventions. Conquerors of the Mede ;
builders of free Athens ; first makers to the world of tragedy and
of comedy : it is a rare combination.

But there begins the second great chapter in Greek literature.



The main exposition of this book has proceeded in historical
order, starting in times of extreme darkness and working slowly
towards the beginnings of clear and well-lit history. Of
necessity, therefore, the argument has rested cheifly on analogies
and general considerations, not on documents : it has had to be
very cautious, aiming at probability, not certainty, constantly
suggesting, not professing to demonstrate. It will, I think, be
convenient now, at the end of the book, to reverse this process,
and trace briefly such actual recorded facts as we possess about
the history of the poems backward from the known to the
unknown. The two inquiries will just meet in the middle, I
have hopes that this chapter, if not very inspiring to the general
reader, may be of some use to students, helping them perhaps to
clarify their conceptions of the whole Homeric problem and free
their minds from the fatal glamour of false knowledge diffused by
the printed text.^

We start from what we may call the modern vulgate, that is,
the text as ordinarily printed at the present day apart from the
special views of any particular editor. This text is remarkably
uniform, almost as much so as that of Vergil, far more so than
that of Shakespeare. Also it is based upon an extremely large
number of MSS. True, no complete copy is older than the
tenth century A. D., but there are large fragments much earlier,
and indirect evidence carries the Vulgate back a little before the
Christian era.

* For good remarks on the habits of ancient scholars in dealing with their
books, and the remains of fluidity even in the mediaeval MSS. of Homer, see
T. W. Allen, The Text of the Odyssey^ Papers of British School of Rome, V


We also find in the Scholia, or ancient commentaries, a great
deal of information about the texts published or approved by-
certain Alexandrian scholars, especially Aristarchus (fl, 160 B. C.)
and Zenodotus (fl. 285 B. c). It would almost be possible, from
the statements of the Scholia, to reconstitute the whole text
according to Aristarchus, and Dr. Roemer at one time promised
to do so. For Zenodotus our knowledge is not nearly so full,
but we can make out much about his critical method.

It is significant that these two critics invented for their editions
certain special signs. Zenodotus apparently used only one, the
obelus ( — ), to mark lines as spurious. Of Aristarchus's signs
the two commonest are, first the obelus, then the diple ( >), which
is merely a mark for reference like our asterisk. Other signs
denoted that lines were repeated more than once in the poems,
and that in some places they were right, in others wrong.
Others probably showed where the genuine Homer left off and
where he began again, the part in between, as far as we can
make out, being spurious. Aristarchus had also one sign which
meant that he was referring to a note already made by

When you think of the pressing need there was, according to
our ideas, for the invention of a decent punctuation and proper
divisions between words, it becomes the more striking that the
first need these scholars actually felt was for signs to mark
spuriousness. Except for the diple almost the whole apparatus
of signs seems devised for the casting out of spurious matter.

Now Aristarchus's own rejections are by our standards
extremely vigorous : he rejected, for instance, all the last book of
the Odyssey at a blow. But, compared with Zenodotus, he was
celebrated for his Trepiaafj euXa/Sem, his ' excessive caution '.
Some critics indeed have maintained that Aristarchus never
under any circumstances made a conjecture of his own, but
always had some MS. authority for even his smallest deletions.
I do not agree with this view, but the question does not for the
moment affect us.^

The method of Zenodotus was by the standards of a modern
critical editor amazing in its vigour. He hacked away like
a woodman clearing an overgrown forest ; and it is clear that he
* See Cauer, Grund/ragen, ed. 3, pp. 57 ff.


relied largely on his personal feelings. We can see that he
regarded the texts of his day as containing, in every part of the
poems, whole masses of stuff that was not ' Homer '. He
collected many MSS,, but seems not to have had any that he
considered authoritative. He is the author of the traditional
division of the poems into twenty- four books denoted by letters
of the alphabet, the Iliad having capital letters, the Odyssey
small. Being himself an epic poet he used his critical faculty and
rejected much merely because it was ' unseemly ' ; it is possible
that he even rewrote some passages out of his head. The
freedom of the old bards was not entirely dead in the first of
the critics.^

Thanks to the brilliant pioneer work of Zenodotus, Aris-
tarchus was able to proceed with more caution. The ground
had been cleared for him, and, besides, the Ptolemies had been
for some generations zealously collecting MSS. But it is note-
worthy that when Aristarchus does cite a MS. authority for
some reading, he never shows knowledge of any particular
authoritative MS. nor of any widespread and authoritative
tradition. His authorities are such as 57 Hii/mttlki], tj Maacn-
XicoTLKrj, 17 Kara ^Piavov, rj Kar 'Ai/Tifiaxov, at KOLvat, al 8r]fj.a)8ei9,
at \apua-T^paL, Tives Tcou iraXamv ktX. one of these, 7/ Koiurj, it
may be said, is exactly ' the Vulgate ', Possibly ; but, if so, the
' vulgate ' of that day differed demonstrably from ours, and
what is more important, was regarded by Aristarchus with some
contempt. He speaks of al kolvui or ai SrjficoSeis as one might
speak of ' the cheap editions '.^

This seems to show that (i) Zenodotus found the text in
a state of great disorder, and (2) neither he nor Aristarchus had
any authoritative MS. tradition by which to correct it. The one
recension which Aristarchus thought worthy of a special critical
sign was not an ancient vulgate but the edition of Zenodotus.

This conclusion is vehemently opposed by many conservative
critics. Obviously those who wish to maintain that our present
I/i'ad and Odyssey were written, approximately as they stand, by

^ Literature in Susemihl, Alexandr. Literaturg. i. 333 ; see especially
Roemer. I omit the work of Aristophanes (fl. c. 200 B. c.) for the sake of
simplicity. Susemihl, i. 428-48.

* Ludwich, Homervulgata, p. 49.


one great poet in the eleventh century B.C., cannot possibly admit
that the text was still in a very fluid state so late as the third
century. The position of Ludwich, for instance, is that, roughly
speaking, our present vulgate was in existence as an authoritative
text from the very earliest ages, and passed unscathed through
the illiterate centuries of early Greece, through the creative fer-
ment of the fifth century, through the chaos of the pre-Zenodotean
texts, and lastly through the fires of Alexandrian criticism, always
unmentioned but universally recognized, to emerge in triumph
in our post-Christian MSS.

Observe that there are two questions at issue. First, did there
exist at all in pre- Alexandrian times a text like our traditional
one ? Second, was this text, if it existed, an authoritative vulgate ?
To the second I think the answer is a confident No : as to the
first 1 can find no conclusive evidence. But let us consider what
there is. We shall find it in two places. First, in such frag-
ments of MSS. as have come down to us from the times before
Aristarchus ; secondly, in the quotations made from Homer by
classical writers. In the history of this controversy the evidence
of the quotations came first. The great Wolf, who entirely
denied the existence of any text like ours in pre- Alexandrian
times, mentions as a certain fact,

quod apud Hippocratem Platonem Aristotelem et alios istius
aetatis scriptores non solum singulorum verborum varietates,
sed etiam plures insignes versus legimus, quorum nee in
textu nostro nee in Eustathio veterrimisque et doctissimis
scholiis ullum indicium superest. {Prolegomena^ p. 37.)

It might have been more prudent to write Aeschinem instead
of Platonem^ but in the main I consider this statement just in
itself and signally confirmed by recent discoveries.

But quotations are slippery witnesses. It will be best to start
with the more positive evidence, that of the pre-Aristarchean
papyri. We should remark at the outset that in the case of
Euripides and Plato, and, one may say, practically every classical
author except Homer, the early papyri, where they exist at all,
confirm to an extraordinary degree the accuracy of our MS.
tradition. In no case are there any large differences. How does
the case stand with Homer ?

I. I. There are altogether, according to Dr. Hunt's estimate.


some two hundred fragmentary papyri of the Homeric poems,
the Iliad being about twice as well represented in them as the
Odyssey. Of these eight were written earlier than 150 B.C., and
therefore have a direct evidential value for the present question.

The first of these to be discovered was the Flinders Petrie
papyrus (Dublin, 1891) in two fragmentary columns, which
contained A 502-37, the ends of 502-17 in the first column, the
beginnings of 518-37 in the second. The main conclusions
are given thus by Ludwich. Of the ends seven out of twenty
are different from our vulgate, of the beginnings four out of
nineteen. There are altogether thirty-nine lines instead of the
thirty-six of the vulgate, the number being made up by the addi-
tion of four lines, hitherto unknown, and the omission of one.

This extraordinary result was accepted by some scholars as
showing that our vulgate text was merely a product of
Alexandrian criticism ; by others it was brushed aside as the
accident of a single eccentric or ' wild' MS. Such a MS., they
held, could not be a fair specimen of the pre -Alexandrian texts.
Since that time, however, our specimens of such papyri have been
slowly growing both in number and size,^ and they all show in
varying degrees the same general features. They all tend to
have additional lines and to leave out some lines that we know.
And where the lines coincide with the vulgate, the readings
inside the line, as far as we can judge from the fragments, seem
often to have been different. The papyri in question are as
follows : the sign -f denotes additional lines found in the papyrus,
— denotes lines omitted. The number in brackets is that given
in the apparatus criiictis of Mr. Allen's Oxford critical text of

I. (8 Allen) P. Petrie, beginning of second century B.C.^
containing A 502-37 (39 verses: -f4— i; at least 11 variant

II. (5) P. Genavensis, early second century. A 788-M 11
(70 verses : -f 13 — o ; many variant readings).

III. (41) P. Grenfell II, 3, and Hibeh I, 20. Parts oi T A E
{66 verses : -f i — 3 : ' differed widely from the vulgate ').

* Dr. Hunt informs me of three more early papyri : one a fragment of
Z, which approximates to our text [now published as Pap. Ox. 13S8] ; two of
e, both wild.


IV. (7) P. Grenfell II, 2, and Hibeh I, 21, 17-258 (97 verses :
+ 28—0: between 52 and 66 there are + 21).

V. (12) P. Grenfell II, 4, and Hibeh I, 22. Between ^ 387
and ^281 (190 verses : + certainly 11, perhaps 20; — ?).

VI. (40) Hibeh I, 19. Between B 174-830 and P 277-371
(105 verses : +13: many variants).

VII. (19 in Odyssey list) Hibeh I, 23, v 41-68 (30 verses:

+ 3-1)-

VIII. Rylands 49. Beginnings of U 484-9 ; six beginnings,

one of them different.

Lastly, two Heidelberg fragments, known to me by the kind-
ness of Dr. Gerhardt, the learned editor of Phoinix of Colophon^
who has since published them :

Heid. IV. 191 ff., 16 lines, from the same MS. as IV (16
verses +4).

Heid. V. 183 lines from ^ X ¥^, from the same MS. as IV
(roughly something making the average about +7 — 2 per cent.
This is the nearest to the vulgate that has been found).

Rather later in date but similar in character is a papyrus of the
first century B.C. in Berliner Klassikeriexie, v, p. 18, containing
the end of 5", with the description of the shield of Achilles.
This is so instructive that I cite it in full.

1, 596-602 agree with the vulgate ; then it runs :

603 TToXXoy 5' l/xepoevTa xopoi' irepua-TaO' o/ii\o^

604-5 TepirofiepoL' Solco Se Kv^iaTr]TTJp€ Kar avToiis

606 fxoXnfj^ k^dp^ovres kSivivov Kara fxiaaov^.
606"^ eif S" tcraav avpLyye^, ecrai^ KiOapts re Kal avXoi.

607 iv 8' kriOii TTOTafjioio fiiya adipo? 'flK^auoTo

608 dpTvya nap rrvjidrriv adK€09 nvKa ttoitjtoTo.

608* iv 8e XifiTj}/ erirvKTO iavov KacraLTepoio {Aspis 207-8)

'' KXv^OjJLivCO LKiXoS' 80LO) S' dua(f)V(TlOCOl^T€S (2O9-I l)

^ dpyvpeoi SeXcfiii/e^ kcpoiveov eXXona^ Ix^^^- (^^2)
^ Tov S' VTTO \dXK€ioL Tpiou i\Bvis' avTccp eTT* uKTah (213)
(I accept the editors' restorations : they are generally pretty
certain and do not affect the argument.)

Observe : 604-5 ^^^ ^^^ together. In our vulgate they

TepTTOjievoi' fxeTo. Se (T(f)iv eixeXTTsro 6€7o9 doiSb^
(popfii^coy, SoLOi) 8e ktX.


But our vulgate has here behaved oddly. Editors have
forsaken the MSS. and inserted a phrase from the Odyssey
(S 17-18) on the evidence of Athenaeus (p. 180 c), who says
that the lines in question originally belonged to S and not to the

606^ is a new line. 608* ^ '^ ^ are not known to us in Homer,
but a passage closely similar, though slightly longer, stands
in our text of Hesiod, Aspis 207-13 describing the shield of

What is the meaning ot such a phenomenon as this? A
passage known to our tradition as part of the Hesiodic Shield of
Heracles appears in this MS. as part of the Homeric Shield
of Achilles. It is clearly not the mistake of a copyist. It is, as
Diels and others have seen, the deliberate variation of a rhapsode,
who preferred his 'Shield' in that form. He shortened the
expression a good deal and he got in the description of a
harbour with plunging dolphins. Whether his judgement was
wise may well be disputed; the point is that apparently he
thought he had a right to make it. The text of this passage was
not absolutely fixed as canonical even by the time this MS.
was written — when Aristarchus had perhaps been dead fifty

The same explanation seems to me to apply to all the facts
about these pre- Alexandrian MSS. The text was still very fluid,
at any rate in places. For, as Grenfell and Hunt have pointed
out, the additional verses are not scattered evenly all over the

* This is not an isolated phenomenon. The Townley Scholia on Q. 804,
the last line of the Iliad, mention that instead of

Qp ot y' afi^Uuov Tacftou "Ekto^os tTnroSd/xoio

some MSS. read

iif ot y aficf^Unop Td(f)ov "EKTOpos' rjXde 8 'A/xafwr,]os 6vyaTt)p fi€ya\T]Topos avdpocpovoio.

That is, they ran on from the end of our Iliad to another story, the Aethiopis,
about the Queen of the Amazons. And in some cases such a mixture of
sources has actually become canonical. The end of the Theogony in all our
MSS. is mixed up with another poem, The Catalogue of Women who were
loved by gods. The MSS. of our Shield of Heracles have attached tha^
poem to one of the eoiai, or used the eoie, so to speak, as a peg. See also
the striking Fayum fragment (Allen 53) giving the Chryseis episode (A 486 ff.)
in the words of the Hymn to Apollo^ 503 ff. Cauer, Grundfr.^, pp. 44 ff. Cf.
Boiling in A.J. P. 1914, pp. 125 ff.


poems, but are concentrated in particular parts. They come
where the texture of the narrative is loose : where inorganic
verses can easily be added, or whole formulae of two or three
lines inserted. To put the same fact from a different point
of view, some parts of the poems were specially well known and
canonical ; others were still fluid and indefinite — the less
interesting-, the merely transitional, the parts perhaps which were
not often chosen for recitation, though they had to exist in any
professedly complete text.

There is, for instance, perhaps no part of the poems which has
been more ' suspected ' by scholars than ©. According to
Wilamowitz it was largely composed very late in order to make
room in the I/tad for I and K. And a glance at the list above
will show the extraordinary ' wildness ' of the three fragments of
the papyrus containing &. We shall find a similar wildness
about in the quotations.

We may also observe that the new lines seem generally, though
not always, to be made up of lines or half-lines or phrases which
occur elsewhere in the poems ; very few seem to have been
original or vital poetry. The Alexandrian critics were wise in
the use of their obelus.

2. Let us now take the quotations.

At the first blush we can see one thing. There are a good
many small fragments quoted from Homer by various authors
which do not occur in our text. Of fifth-century authors, Pindar
observes that Homer says that a noble messenger gives dignity
to any business. Our Homer never gets nearer to that than to
say that it is a good thing when a messenger is tactful. Hippo-
crates mentions that Homer knew that cattle suffered In winter ;
that is why he wrote o)? S' onor dcnrdaLov 'iap rjXvOe ^ovalv
'4Xl^i. Our Homer writes nothing of the sort. In the fourth
century Aeschines says that ' Homer says several times in the
Iliad (Prifii] S' ey aTparov rjXOe ' : the phrase never occurs in our
Iliad. Xenophon cites from Homer the phrases ydvvraL Si r'
aKovcoi^ and ttvklucc ^peal fnljSea e/5c6?, which do not occur.
Aristotle, who uses Homer a great deal, quotes quite a number
of lines unknown to our texts ; ^ nap yap kfxol BduaTo?, "EKTopa

^ Find. Pylh. iv. 277, cf. O 207 ; Hippocr. n-epi I'lpdijap iii, p. 146 K. (p. 62,

2760 U


S' alSa)^ eIXe, "EKTCop kcTt' d\\o(j)pov^aiv^ fivcrev 8h nepl f^porocao-'
(oTeiXrj, Zev^ yap 01 j^e/xicracrx' or dfi^LVOvL (fxcTi fxd\ono.
Besides these completely unknown lines, he quotes known
passages in a strange shape ; he found diSo/xep Se ol ev^o? dpea-Oai
not in $ 297 where we have it, but in ^ 15 ; he found S 567 in a
shorter form; he found our lines p- 2ig ff., or something very
like them, in a speech of Calypso ; he found part of our descrip-
tion of the Cyclops in the ninth Odyssey as a description of the
Calydonian Boar in the tenth Iliad', he expressly says that
Odysseus' story to Penelope {W 310-41) occupied 'only sixty'
lines: in our text it occupies thirty-three. In Ethics^ p. 11 16
b 24, there are four phrases quoted from ' Homer ', two incorrect
and two unrecognizable. It is also worth noting that Aristophanes
says that Homer describes Iris in words which in our text apply
not to Iris but to Hera and Athena ; or that Plato read pijrrjp
instead of'EKrcop in Z 402, making a change not only in wording
but in a statement of fact.

This list is not complete, but, even apart from the evidence ot
the papyri, it seems to me quite conclusive. There must have
been current in the fourth century texts of Homer very different
indeed from ours. Make handsome allowance for slips of memory
and the like, the testimony of these unknown lines is not to be
overthrown, and cannot even be shaken by any but the most
overwhelming evidence on the other side.

That evidence Ludwich has tried to produce. He collects
a great list of Homeric quotations in authors of the fourth
century or earlier, covering some 480 lines, and urges us not
to concentrate our attention on the ' wild ' lines which reject our
text, but on the great majority of tame' lines which conform to it.

Let us consider this plea. The evidence of quotations is
always hard to use, as certainly an editor of Euripides is not
likely to forget. The quoter may err in memory ; he may adapt
the words of the poet to his own purpose ; he may intentionally
omit lines. He will quote chiefly what is striking and interest-
ing. In the special case before us, what we have to make out is

Erni.) ; Aeschin. i. 128 (Blass) ; Xen. Sy^np. 8. 30; Aristot., pp. 1285 a 10,
1230 a iS, 404 a 29, fr, 167 Rothe, 1387 a 32 ; 162 b 7, 943 b 21, 1109 a 30,
57i5 b 2, 1417 a 12 ; Ar. A^f. 575 ; Plato, CraL 392 b. See also fragments in
Allen, vol. v, pp. 146 ff.— Hippocrates and Pindar may have used ' Homer ' in
a wider sense, see p. 298.


whether each quotation in the ancient authors seems most likely
to come from a text practically identical with our vulgate or from
one like the pre-Aristarchean papyri.

Now, in the first place, single lines or bits of lines which
agree with our text prove nothing. They doubtless also occurred
in the ' wildest ' papyri. Conventional phrases and epic runs
prove nothing for the same reason. Even if there were a general
tendency not to quote the ' additional ' hues much, that would
prove nothing, because the additional lines are seldom striking
or quotable. Mere descriptions of facts or abbreviations of long
passages seldom prove anything, because the differences between
the papyri and the vulgate would scarcely show in them. SHght
variations in language, on the other hand, do not prove much,
nor do omissions of lines. They may be mere mistakes of
the quoter. Such things are common in the quotations from
Euripides. Out of the great list of quotations given by Ludwich,
covering some 480 lines of the Iliad and Odyssey^ more than
half fall away at once as non-evidential.

If we take only the quotations of more than three consecutive
lines we have some approach to firmer ground. We may class
them as follows ; Agreeing but not conclusively^ two passages :

A 17-42, referred to by Plat. Rep. iii. 393 d, in a fairly close
indirect description, with many lines omitted.

B 671-4 : three half-lines cited in Ar. Rhet. iii. 12, p. 1414 a 2.

Disagreeing bttt not concltisively^five :

I 497-501 in Plat. Rep, ii. 364 d, one line omitted; wording
slightly different.

I 308-14 in Plat. Hipp. Min. 364 e (cf. 370 a), one line omitted.

J 446-50, roughly cited in Ar. Pax 1273 : not much evidence,
but a much -suspected breast-plate line is omitted {avv S' eyx^a
Kat fxei/e' dv8p5)v y^aXKeoOcopiJKcop).

T 109-13 in Plat. Rep. ii. 363 b, one line omitted.

V 351-7 in Plat. Ion 538 e, one striking line omitted and word-
ing slightly different.

Clearly agreeing, perhaps twelve (occasionally with some
verbal variation): Z 289-92 in Hdt. ii. 116; Odyssey^ 8 227-30
in Hdt. ii. 116 (cf. Theophr. de Plant, ix. 15. i) ; O 494-9 in
Lye. /// Leocr. § 103 (differences) ; ^ 324-9 in Aeschines, i.
§ 143 ; ^ 42-5 in Ps.-Aristot. de Mtmdo^ 6, p. 400 a 6 ; I 357-63

U 2


in Plat. Hipp. Mm. 370 b (cf. Crito, ^4 b) ; I 650-5, ibidem,
371b; M 200-7 i" P^^'- ^^^^ 539 ^ 5 ^ 335^40 in Plat, /i?;? 537 a
(cf. Xen. Syinpos. 4. 6); co 6-9 in Plat. Rep. iii. 387 a ; i 112-
15 in Plat. Legg. iii. 680 b; H 96-102 in Plat. Legg. iv. 706 d
(slight differences).

Co7iclusively and markedly disagreeing we find seven at
least :

JB 188-202 in Xen. Mem. i. 2. 58 ; six verses omitted,
probably not by accident, as they were counted spurious by

E 391 ff. in Arlst. Pol. iii. 14 (p. 1285 a 10), with an unknown
half-line added, irap yap e/xol 6dvaTo<5.

5* 95-9 in Aeschin. i. 150, markedly different wording.

W 77-9 1 5 ib. 146, with two new lines, one line inserted from
elsewhere, and several differences of wording.

n 10-12 in Plat. Rep. iii. 388 a, considerable differences of

/2 527-32, ib. ii. 379 c, with one strange line substituted for
one of ours.

548-52 in the Platonic Alcibiades ii, p. 149 d, vi\\h four
completely new lines added.

The proportion is just about what it ought to be. The
quotations, where they are long enough to afford a fair test,
instead of lifting a loud protest against the evidence of the
papyri, simply and clearly confirm it.

There is one point more. Grenfell and Hunt, in their
masterly discussion of this question in the introduction to Pap.
Hibeh 19, have shown that if a dividing line be drawn at 150 B.C.
all MSS. earlier than that date differ ' enormously ' both from
our vulgate and from Aristarchus, and all tend to be longer
except possibly Hibeh 20.^ After 150 B. c. the tendency of MSS.
to differ from the vulgate diminishes rapidly, and by the
beginning of the Roman period ' the numerous Homeric frag-
ments published in recent years very rarely contain new verses,
and serve to illustrate only too well the overwhelming pre-
dominance of the vulgate '. Zenodotus had laid the foundations

* This exception is considered by Dr. Gerhardt, in his introduction to the
new Heidelberg fragments, to melt away in the light of later evidence.


of criticism about 280 B. c. Aristophanes and others followed him.
The floruit of Aristarchus, most successful and universally
acclaimed of Homer scholars, is 160 B.C.; the triumph of the
vulgate begins about 150 B.C. The dates speak for themselves.
The predominance of a much-castigated and purified text was
due directly or indirectly to the great critics of the Alexandrian
age. Can we go further than this, and pronounce it definitely
the work of Aristarchus ? That view, put forward by Wolf and
Nauck, and probable on general grounds, has hitherto seemed to
break down on the detailed evidence of the scholia, but is now
almost proved to be right by the acute researches of Professor
G. M. Boiling of Ohio.^ He has collected in the vulgar tradition —
both in the MSS. and the post-Aristarchean papyri — all the
lines of doubtful authenticity, and finds that all the directly
attested differences between the vulgate and Aristarchus are due
to interpolations in the vulgate, while, conversely, no line which
is an interpolation in the vulgate can be shown to have been in
Aristarchus' text. That text contained about 15,600 lines, to
which some 93 have been added by interpolation in the course of
the last two thousand years.

But, granted that the present vulgate had in pre- Alexandrian
times no central and dominant position — and most scholars have
been convinced by Grenfell and Hunt — one question still
remains. Did our vulgate exist at all in classical times, or is it,
very much as Wolf thought, a later creation altogether, a text
hammered out for the first time by the impact of Alexandrian
criticism upon a fluid but rather obstinate tradition ?

The point is a doubtful one, and depends mainly on the
quotations in Plato. They, as may have been seen above, resemble
our text pretty closely. -

On a rough analysis, there are twenty-three'-^'' quotations in
Plato which definitely agree with our text ; there are eigh-

^ A.J. P. xxxvii (1916), pts. I and 4 ; cf. Modern Philology, xvii. 3 (1922).
Cf. Cauer, I.e., chap. 3.

'' ^ Agreeing, sometimes with slight variations: —

A 15 f. Rep. iii. 393 a. M 200-7. ^^'^ 539 b.

A 599 f. Rep. iii. 389 a. P 446 f. Axioch. 367 d.

E 127 f. Alcib. ii. 150 d. 2 23 f. Rep. iii. 388 b.

I 357-63. Hipp. Min. 370 b. T 92 f. Sympos. 195 d.

I 644 f. Cratyl. 42S c. X 414 f. Rep. iii. 388 b.

I 650-5. Hipp. Alitt. 371 b. ^ 103 f. Rep. iii. 386 d.




teen '' of no evidential value, being too short, too vague, or con-
taining mere epic phrases which might come anywhere ; there
are seven '^ which omit lines in the middle ; four that vary consider-
ably in wording and three that vary very slightly ; ^ there are
seven which definitely differ from our text by additional lines or
conflated lines ; ® and there is lastly the perfectly ' wild ' quota-
tion from © in the post-Platonic Alcibiades ii. It needs a bold
man to argue from this that Plato's text was our text. Still it is

•t 308 f.
X 506 f.
"^ 100 f.
i2 IS f.
o 525 f.
X 633 f.

p 383 f.

r 395 f.

X 1-4.

* 335-40. Ion 537 a.

12 80-2. Ion 538 d.

a 32-4. Alcib. ii. 142 d.

y 26-8. Leg.^> vii. S04 a.

t 112-15. I^gS- "'• 680 b.

X 489-91. Rep. iii. 386 c, vii. 516 d

^ Agreeing, but non-evidentially :
A 17-42. Rep. iii. 393 d.
B 813 f. Cratyl. 392 a.
E 221 f. Craiyl. 407 d.
Z 235 f. Sympos. 219 a.
e 14. Phaedon 1 12 a.
n 112 f. Rep. viii. 545 d.
n 856 f. (phrases). Rep. iii. 386 d.
2 108 f. Phileb. 47 e.
Y 64 f. (phrases). Rep. iii. 386 c.

« Omitting lines : —

I 308-14 (om. l).

1 497-501 (om. i)

2 96-104 (om. 6).
X 15-20 (om. 4).
T 109-13 (om. i).
r 173-9 (om. 3).
^ 351-7 (om. 1).

^ Different in wording : —

H 96-102. Legg. iv. 706 d.
a 351 f. Rep. iv. 424 b.

1 8-10. Rep. iii. 390 a.
p 322 f. Legg. vi. 777 a.

A 169-71. Hipp' Min. 370 c (slightly)
n 433 f. Rep. iii. 388 c (slightly).
X 16S f. Rep. iii. 388 c (slightly).

^ Different by additional or conflated lines, (S:c. :
r84-A43i. 7?^/. iii. 389 e.
A 218-19. Rep.'m. ^o^^.
Z 402 {^i]Tr]p). Cratyl. 392 b.
A 639-^630. Ion 538 b.

2 295 f. Rep. iii. 390 b.
O 10-12. Rep. iii. 3S8 a.

i2 527-32 (new line). Rep. ii. 379 d.
e 548-52 ( + 4). Alcib. ii. 149 d (wild)

o 245 f. Axioch. 368 a.

p 347. Charm. 161 a.

p 485 f. i?^/. ii. 381 d.

V 17 f. Phaedon 94 d ; Rep. iii. 390 d

iv. 441 b.
0) 6-9. 7^.?/. iii. 387 a.

Protag. 340 a.
Craiyl. 392 d,
A"^/. iii. 387 a.
7?^/). iii. 391 b.
Axioch. 367 d.
Sympos. 198 c.
AV/>. iii. 389 d.
Rep. i. 334 b.
Ion 535 b.

////>/. Min. 365 a ; ib. 370 a.
A'£'/. ii. 364 d.

A polos;. 28 c.

/i'^/>. iii. 391 a.

Rep. ii. 363 b.
Minos 319b.
Ion 539 a.


clear that Plato's quotations are much closer to our text than
those of any other fourth-century writer.

The simplest conclusion would be to assume that Plato used
a text very like ours. Yet perhaps that would be a mistake.
Among the writings of the first disciples of Aristarchus we find
one by Ammonius, irepl t5>v vtto UXdrcovo? k^ '0/xijpov fxere-
yi]i'eyfxev<iDi', ' on Plato's quotations from Homer '. The purpose
of the book was textual recension. That is, the quotations in
Plato were a recognized authority for the text of Homer in
Alexandrian times. There was a whole small literature on Plato's
relation to Homer. He shared with Herodotus the title of
'OfjLT]piK(tiTaT09y and exercised a quite special influence on the
Alexandrian school. Is it, perhaps, not Plato who agrees with
our vulgate, but our vulgate which, wherever it had the evidence,
tried dehberately to follow the readings of Plato ? It is curious,
at any rate, that the writer whose quotations, few as they are,
come next to Plato's for conformity with our text, is the other
recognized ' Homerikotatos ', Herodotus.^

II. The verbal text, then, was still fluid and subject to change
as late as the fourth and third centuries B.C. What can we be
sure of as fixed } The whole main structure, one would suppose,
the incidents and the order in which they followed one another.
Yet even here one cannot feel absolute confidence, at any rate for
the fourth century and earlier.

For instance, to take an observation made by the late
Professor M. L. Earle of Columbia University: Thucydides,
i. II. I, writes about the Greeks at Troy : 'When they landed
they must have won a battle ; otherwise they would not have
built the fortification round the camp.'^ This shows that
Thucydides (i) knew of the wall round the camp so frequently

* See Sengebusch, Dissert. Frioi-,^ pp. 118-24 ; Ludwich, p. 141, note. In
the next generation Trypho wrote Trepi tP;? apxo-i-o-^ fimyj'&jo-ecoc, which Senge-
busch interprets ' on the readings of Homer shown in the ancient quotations in
general '. Sengeb., p. 124. Cf. Suseinihl, Alexandr. Lifter., pp. 154 and 212.
See also Howes, Harvard Studies in CI. Phil, vi (1895), pp. 153 fif.

^ 'Erret6)7 6' acpiKonevoi fidxv eKparrjadV' ^^Xof de' to yap tpv/ia tc5 arpaTOTredia
ovK av fTdxio^avTo. Thiersch iKpaT-qGrjcniv, ' lost a battle ', which Earle accepts.
The reading does not affect the present argument. The same suggestion, it
is interesting to find, was made long ago by Hermann; Opusada, vol. viii,
P- li^l {iT^)- See Prof. Earle's Collected Essays, pp. 142-4.


mentioned in our Iliads and (2) surmised that it must have been
built at the beginning of the war, after the first battle.

Now in our Iliad {H 337 ff., 436 ff.) the building of this wall
and the exact circumstances which led to it are fully described,
and are not what Thucydides conjectures they ' must have been ',
It was built in a great rush and in picturesque conditions during
a scanty truce in the tenth year of the war. It is noteworthy
that the particular passage in H has been marked by K5chly and
many other critics as ' recentissima '.^

The view we take of this bold suggestion will obviously
depend largely on the presence or absence of other symptoms
pointing in a similar direction. It is always hard to get out
of our minds the associations of printed books, which appear
in definite editions in a complete form, all the copies identical.
But let us look at the direct evidence.

There are still extant many MSS. which omit the Catalogue
in JB, though, curiously enough, they give the series of similes
with which it is introduced. That is, even at the time when the
vulgate became predominant, the Catalogue was not definitely
established as a necessary part of the Iliad.

There are no MSS. now which omit K, but a note in the very
valuable Townley scholia informs us : ' They say that K was
originally placed apart by Homer and is not part of the Iliad.,
but was put into it by Pisistratus.' ^ The statement is repeated
in the learned scholia to Dionysius Thrax and in Eustathius, who
ascribes it to ' the ancients '. That is — to put the case at its
lowest — there was an ancient tradition which knew of, or believed
in, the existence of Iliads without K as well as Iliads with.

We also know that Aristarchus thought the last book of the
Odyssey (00) spurious, and that both he and Aristophanes of
Byzantium considered -v//- 296 as ' the end of the Odyssey \ This
does not necessarily imply that he knew of MSS. without (o
or without the end of -v//-, but it does show that the canon was
still far from certain.

If such large stretches of the poem were not definitely estab-

* Plato's citation of H 321 in J?e^. v. 468 d does not of course affect the

^ ^aa\ TTjV pa^ahiav v(/>' 'Ojirjpov Ibia Tfra)^0ai koi firj dvai fjLfpoi ttJs 'iXtaSoy,
vTTo 8e U(icri(TTpdTov Teraxdiu tls rfji' Troirja-iv. Schol. T on K I. Eustathius says
(Pacriv ol nnkaioi, evidently referring to the same source.


lishcd even in Alexandrian times, it is obviously quite possible
that a passage like the Building of the Wall was not definitely
established in the time of Thucydides. We must not be indignant
merely because such a result would show a conjecture of many
modern critics to be probably riglit.

Is there any other test that we can apply? only one has
occurred to me, rather a curious one.

It is vv^ell known that, for some reason, the Attic tragedians in
choosing their subjects made it a careful rule to avoid the main
subjects and incidents of the Iliad and Odyssey. We know, I
suppose, the subjects of some two hundred tragedies by the three
great writers, and the rule is well kept up. There is, indeed,
one great exception, a lost trilogy of Aeschylus {Mynm'dons^
Nereids, Phrygians) which dealt directly with the subject oi Iliad
I-fl. Its date is unknown ; but it comes very early in the
history of Greek tragedy, and, apparently, the experiment it
made was never repeated. In Satyr plays the rule did not hold.
You could burlesque ' Homer ', as in the Cyclops and in Sophocles'
Washing Girls, or Natisicaa} But you avoided attempting to
treat again in the high style at the Dionysia subjects which your
public already knew in the Recitations at the Panathenaea. I
can only make out two certain exceptions. one is the Rhestis,
which treats in full detail the story of Dolon, Iliad K ; the other
is a Catalogue of the Greek ships in the Iphigenia in Atilis
(164-302). The Doloneia and the Catalogue! Just the two
parts of the Iliad which we know to have been uncanonical !

On the whole it seems to me probable that Thucydides used,
or learnt at school, or heard recited at the Panathenaea, an Iliad
without the account of the Wall-building, Euripides an Iliad
without the Catalogue, the author of the Rhesus an Iliad with-
out K. There is a good field here for further research.

III. In the age of Euripides and Thucydides, then, it would
seem from the evidence that the Iliad and Odyssey differed from
our vulgate not only in the matter of exact words and lines, but
even in large portions of the story. ' Homer ' meant to them, as

^ I agree with Valckenaer, Diatribe 209, and Lessing. Welcker, building
on the far from clear passage in Eustathius, ///W, p. 381, thought the nXi'irptat
a tragedy {Gr. Trag. i. 227), and his view has been commonly accepted. It
was a not unusual subject for comedies.


to us, * the author of the Iliad and the Odyssey \ but we cannot
be sure that either Iliad or Odyssey was exactly what we mean by
those words. If we g-o a century further back, however, we find
that the meaning of ' Homer ' also is different. His name covers
not only the Iliad and the Odyssey^ but much wider and vaguer
masses of epic writing as well. Let us take the quotations.

Kallinus, our earliest witness, in the eighth or seventh century
B.C., cites the Thebaid as Homer's (Pans, ix. 9. 5.). Simonides
— either the great Simonides of the early fifth century or he of
Amorgos in the seventh — quotes a proverbial line that comes in
our Z 146 as the work of ' a man of Chios ' : probably meaning
' Homer '.^ The great Simonides quotes 'Homer and Stesi-
chorus' as describing how Meleager 'surpassed all the young
men in spear-throwing across the wild Anauros '. This does not
come from our Homer ; possibly it came from that old Meleager
epos which is a supposed source for Iliad I. Pindar quotes the
Odyssey in JVe^/i. vii. 20 ; he quotes the unknown line about the
messenger in Pyl/i. iv. 2yy i^ in Isthjit. iii. 53 he seems to say
that Homer has told ' all the virtue of Aias ', including his death.
This could scarcely refer to our Iliad. In fr. 189 he mentions
that Homer wrote the Cypria and gave it for his daughter's
dowry. Herodotus himself, when he says that Cleisthenes in
his anti-Argive policy silenced the rhapsodes in Sicyon * on
account of the poems of Homer, in which the Argives and
Argos are generally glorified in every way ', has been considered
with some probability to refer to the Thebais. Lastly, when
Aeschylus described his tragedies as merely 'slices from the
great banquets of Homer ', it is perfectly clear that he did not
mean that they were taken from the Iliad and Odyssey — which
they markedly avoid. When we hear that ' Sophocles rejoiced
in the Epic Cycle ', and when Proclus tells us, quite correctly,
that ' the ancients attributed also the Cycle to Homer ', we can
understand the situation. The ' cycle ', as Wilamowitz and others
have shown, was a compendium of epic history made up out of
various early masses of poetry. Sophocles and Aeschylus both
' rejoiced in ' and took ' slices out of ' that same great body of
poetry, all of which was ' Homer '.^ They did what the vase-

' Callinus 6, Simonides 85, 53, in Bergk's fourth edition.

^ See above, p. 2S9. ^ Hdt. v. 67 ; Ath. 347 e ; ib. 277 e.


painters did : these also probably considered that they drew their
subjects from Homer, but, with few exceptions, they do not take
them from the Iliad or the Odyssey.

The first of our authorities to reject any of this work as non-
Homeric is Herodotus. He argues that the Cypria are not by
Homer because they contradict the Iliad (ii. 117). He is not
sure whether Homer wrote the Epigonoi^ a sort of sequel to the
great Thebais (iv. 32). By about 350 B.C. the name ' Homer ' is
normally used in our traditional sense, for the author of the
Iliad and Odyssey and no other epics besides. Yet there are
still isolated exceptions, as when Antigonus of Carystus cites
the Thebais as Homer's,^ or Simmias — possibly— the 'Little
Iliad '. A great bas-relief full of scenes of epic tradition from
the War of the Titans onward, intended for educational purposes
and composed by one Theodorus in the first century B.C., is
superscribed ©eoScopeiov /xdOe rd^iv 'Ofirjpov.^ Even as late as
that, in certain phrases at any rate, the whole epic tradition could
be called ' Homer '.

IV. How is this change to be explained? What force was
working between, say, the years 500 and 400 B. C. to put the Iliad
and the Odyssey in a separate and privileged position, as the
only true works of 'Homer', and thus far greater and better
known than the rest of the epic traditional poetry ? one cause
suggests itself at once : the public Recitation at the Panathenaea.
Let us sift the statements of our authorities on this subject.

First, we know for certain that Homer was recited at the
Panathenaea. The orator Lycurgus {in Leocr., p. 209) says:
' Your ancestors considered Homer so noble a poet, that they
made a law that every four years at the celebration of the
Panathenaea his poems and his alone should be recited by
rhapsodes. ' ^ There is a similar statement in Isocrates attributing

^ See Wilamowitz, Homerische Unfersitchun^en, 350 ff., from whom most
of this argument is taken. An attempt to overthrow part of it by Hiller, Rh.
Mtis. N. F. xlii, pp. 321-61.

^ Jahn-Michaelis, Bilderchronik.

' Lycurg. in Leocr., p. 209 [§ 102, Bekker] liovKofiai 6* ii/itr Ka\ rbv "OfiTjpov
Trapao'XiO'Qo-i- inaivSiv' ovtco yap i)7reXa/3oi' vp-av oi narepfs (T7rovba7ov eivai Troirjrrjv,
axTTe vofiov edevro Ka6^ (KiiaTrju TrevTaerripiba rdv Hapadrjvaio)}' /ioi'ov tojv <iW<i)v
noirjrSiV pa-^adelo-dai ra enr]. Cf. Isocr. Paneg., p. 74 olyiai Se Kai Tqv 'Op.r)pov
TToirnTiv pfi(,a Xn^e'iv So^nr, on K(i\ios mis 7ro\ep.r]TavTas rols ^ap^apois fveKUfiiaaf,


the institution to ' our ancestors '. The fact, therefore, is certain :
there was a long-established rule at the Panathenaea of reciting
* Homer and Homer only '.

But what does ' Homer ' in this context mean ? Is it the whole
epic tradition or is it the Iliad and the Odyssey ? I think pretty
certainly the latter.^

The conclusive evidence lies in the words of Lycurgus. He
says ' Homer and Homer only ', and no one will dispute that
in his time {c. 331 B. c.) that meant the Iliad and the Odyssey —
unless possibly trifles like the Margites were admitted also.
The language of Isocrates is almost equally clear. And such
indirect evidence as we have points in the same direction. The
rhapsode Ion, for instance, in Plato's dialogue about him, speaks
definitely of reciting the Iliad and Odyssey^ and never suggests
reciting anything else. Further, some of our witnesses state
particularly that the law ordered the recitation to be 'in order',
one reciter beginning where the other left off. It is obvious from
the state of the text in the fourth century that this ' order ' was
not interpreted very rigorously. The very idea of exactitude in
such matters is a product of a later age. But it is certainly easier
to understand a rule that the Iliad and Odyssey should be recited
in order, than to imagine any such attempt made upon the whole
mass of epic saga.

If then we take Lycurgus's words in their natural sense,
the whole development becomes intelligible. During the fifth
century ' Homer ' gradually gets to mean the author of our two
epics and no others; the chosen poems are known in a fixed
order and gradually acquire a fairly fixed text ; the other epics
gradually fall out of general knowledge, and are used mainly as
quarries of tradition from which the dramatists and others can
carve their works. The rejected epics deteriorate in style and
retain all their barbarities. The chosen two, still fluid and
occupying a central position in an age of splendid and exuberant
poetical creation, tend still to become better and better written,
and morally more and more idealized.

Kca 5ta ToZro ^ovXrjdrjvai rovi npoyovovs fjficov ej/Ti/ioi/ airov TTOirjaai rqv re^i'tjv ev
re Tois Tijs fiovcriKrjs ("lOXois kci\ ttJ naidevaei twv vnoTepatv.

^ The other view is upheld by Dr. Verrall in The Bacchants of Euripides,
pp. 175 ff. With almost all of Dr. Verrall's argument in this essay on ' The
P'irst Homer' I cordially agree.


Can we make out at all why these two should have been
selected ? A certain kind of critic is ready with his answer,
an enthusiastic description of the incomparable poetic merits ot
these two poems and their immense superiority to all the other
poetry of which we know nothing. But the public acts of
statesmen are not often swayed by considerations of poetry.
If these two poems were felt in some special way to represent
in public opinion the crown of the old Ionic poetry, that would
be a real motive. If there was in them already some moral
superiority, that would be a real motive. They were constantly
used for purposes of edification. But I incline to suspect that
Isocrates instinctively discerned the main reason :

I believe that the poetry of Homer won greater glory
because he nobly praised those who warred against the
barbarian, and that this was the reason why our ancestors
conceived the desire to make his art honoured both in the
contests of the Muses and in the training of young men.
{Paneg., p. 74.)

Isocrates was, no doubt, thinking chiefly of the Iliad', but the
Odyssey has its national character too. The Iliad typifies the
national heroes who warred with the Mede, the Odyssey the
national colonists and adventurers who, trusting only to their
brains and their courage, searched strange seas from Pantica-
paeum to Tartessos.

V. We can perhaps make out a little more about the text used
at this official recitation.

The first thing to notice is that to some extent the surface
of Homer has in our tradition been Atticized. To what extent
it is hard to say, since the actual spelling which has come down
to us has passed through a further influence, that of the post-
classical Koine^ or Common Greek. But in any case there are
numbers of lines which run perfectly when the Ionic forms are
restored, and are visibly wrong as they stand at present. The
poems were generally recognized in antiquity as Ionic poems.
Yet all our MSS. and the Alexandrians behind them unite in
giving us the Attic forms. There is no suggestion in the
Scholia of any other view. There are also some few obvious
' Athenian interpolations ', and no doubt many more that are not
obvious. But though some scholars in antiquity suspected them,


there is no statement that any old MSS. left them out. What
does this mean ? Of course a great predominance of Athenian
MSS. would surprise no one ; the literary supremacy of Athens
would ensure that. But this is much more. It means that when
the Alexandrians were searching for ancient MSS. by which to
correct the text, and collecting copies of various sorts in places
ranging from Marseilles to Sinope, they could not apparently
find a single Ionic MS. worth their notice. The Attic versions
had completely superseded the Ionic. We can understand why
the great collector of MSS., Aristarchus, decided that Homer
himself must have been an Athenian.

Zenodotus was an Ionian, and Ionian influences were strong in
Alexandrian literature. Yet we have to admit that either there
were no Ionic texts of Homer at all, or, if there were, they
were so unlike and so inferior to the current Attic texts that
critics would not consider them. Either case confirms our
previous conclusion that the Athenian recitations exercised an
immense influence. Cauer, indeed, argues that perhaps there
never had been any Ionic texts at all ; that epic poems had never
been written down till they came to Athens. But this supposi-
tion is difficult in detail. There is much detailed work in both
Iliad and Odyssey^ which one cannot imagine a poet carrying
through except by careful comparison of different MSS. And
the fate of the Samaritan scriptures shows us how completely,
in the days before a reading public, a book might be killed.
We need only suppose that the MSS, used in Ionia were still the
half-secret possessions of professional bards, and that none
amounted to a complete Iliad or Odyssey, in our sense.

There is lastly a curious phenomenon about which it is hard
to form a confident judgement. We find in the Scholia a clear
tradition, backed up by a number of fairly certain corrections of
the text by modern scholars, that at some time or other the
poems were transliterated from the Old Attic ^ alphabet into the

' See Seeck's Quellen der Odyssee, Verrall's essays in The Bacchants of
Etiripides, and pp. 175 f., 179, above.

''■ Why Attic, it may be asked ? Why not some primitive Ionian alphabet,
of the days before Pisistratus ?— Athens had been the home of the poems for
the last three hundred years; the MSS. in the hands of the Alexandrians
seem, as we have seen, all to represent the Attic recension ; and no Ionian


new. The new is the Greek alphabet that we know : the old —
to speak roughly — used no double letters, made no distinction
between the three E-sounds or the three 0-sounds, and used H
to denote the aspirate.^

Now this tradition is only mentioned by the scholiasts In order
to support conjectural changes, and it may be a conjecture itself.
But it looks rather as if it were a true one^. It does explain with
perfect simplicity some confusions that are otherwise difficult.
And if it is true, we are led to a curious and interesting result.

It has been made out pretty clearly that, though Athens did
not adopt the new alphabet for official documents till 404 B.C., it
must have been in use in literary circles very much earlier,
probably as far back as the days when letters were exchanged
between Solon in Athens and Mimnermus in Ionian Colophon.
For literature at that date was an Ionian accomplishment, and
the new alphabet was the Ionian alphabet. How then could it
happen that, at a time when the new Ionian alphabet was already
used in Athens for literary purposes, the great Ionian book
should be deliberately rewritten back into the awkward old
Athenian script? There is only one obvious explanation. It
was written in the official script as an official text for the per-
formance at the Panathenaea.

An official text dating back probably to the sixth century B. c. :
yet we saw that in the third there was apparently no official text !
The critics can appeal to none such. The papyri and the
quotations show that the poems were still fluid. Is this not
a contradiction .''

Not necessarily, I think, for two reasons. In the first place,
granted there was an official text made for the Panathenaea In
the sixth century, I think it In the last degree improbable that at
that date a reciter would be kept to it. It might be stored up,
it might be used for show and for reference. But the whole
notion of keeping a rhapsode to his written text, Instead of

alphabet known to us satisfies all the conditions. The very earliest Ionian
inscriptions all have H for long E and nearly all have <o for the long open O.
Doubtless at an earlier date there may have been a rudimentary Ionian
alphabet, but, as far as I know, the Alexandrians never show any knowledge
of it. To them the ' Ionic alphabet ' means the * new alphabet '. See Cauer ^,
p. 126, and Fick in Bezzb. Beitr. 30 (1906), p. 297, there cited.

* As Wilamowitz puts it, ENAE0IK02I might mean eV 6' eoiKoo-/, or *\v 8f)
oiKuxTi) or eV 8e oiKovai. See Appendix I.


letting- him give you the best he has in him, was in my judgement
an invention of the second half of the fourth century, and would
have seemed a stark absurdity in the sixth. But apart from that,
if there was in sixth-century Athens a government strong enough
and academic enough thus to strangle the poetical powers of the
bards at the Panathenaea, we know that that government did
not survive the year 510 B.C. The Tyrants' authoritative text
may well have fallen with the Tyrants.

But have we any right to suppose that the recitation and the
supposed recension, either or both, were the work of the
Tyrants .-' Well, if there were no tradition at all, that is the con-
jecture most people would make. The Panathenaea was prob-
ably founded, at the least it was restored in special splendour,
by Pisistratus, The policy of making Athens the head of Ionia
was especially that of Pisistratus. And, apart from the Pisi-
stratidae, the choice is really not large among sixth-century
statesmen. But, as a matter of fact, we have at this point the
help of a definite tradition, the oldest trace of it coming from
Dieuchidas of Megara in the fourth century B. c, the clearest
from some good authorities of the Roman period. Unfortu-
nately there is a lacuna in the quotation from Dieuchidas, so we
do not know what he said. We only know that he somehow
connected Pisistratus and Solon with the text of Homer. Our
earliest full witness is Cicero, a particularly well-informed man
of letters writing in the second great period of ancient scholar-
ship. He speaks of the literary fame of Pisistratus, ' who is said
to have arranged in their present order the works of Homer,
which were previously in confusion '. And the tradition is
mentioned by many writers of the early empire.^ I see that

' Duntzer, yi:j/z;A/. Philol. xci, pp. ^-^Z ff., argues that Cicero's authority
was Dicaearchus, a first-rate witness. 1 subjoin the chief texts: cf. Wolf,
Prolegomena, Cap. xxxiii.

Cic. de Orat. iii. 34 ' Ouis doctior iisdem ilhs temporibus, aut cuius
eloquentia litteris instructior fuisse traditur, quam Pisistrati t qui primus
Homeri hbros, confusos antea, sic disposuisse dicitur, ut nunc habemus.'

Pausanias vii. 26. 13 neto-iarparoff eV/j to 'Ofirjpov di.e<nraa-fi€va re Koi uAXa
aWaxov fivrjuovevoufpa rjdpoi^fTo.

Vitae Hovieri IVand V in Westermann, Bioypa^ot (= Allen, v, pp. 245, 248).

Ta Sf noiTjfjiaTa alrov ra dXrjdj] cnropddrjv irporepov ahofifvu, HeuricrTpaTos
Adqvalos (TVveTa^e. — Ufpiicov ras iroXds COnrjpos) fjdt to. TroirjfiaTa' vartpov 8e
HdaicrTpaTos aiira avvjjyayev, wr to (niypafiiia tovto SrjXol, ^Adrjuijaiv eVtyeypa/i-

flil'OV iV (IKOVl aVTOV Toi IldaHTTpUTOV'


these writers are called ' late authorities '. But there is very-
little of our grammatical record that has more ancient credentials :
a strong tradition in the age of Didymus or Herodian, a faint
trace in the age before the Alexandrians.

If we inquire into the probable sources of Cicero and the
other Romans, the indications point to Crates, the head of the
Pergamene school and the great rival of Aristarchus. He had
gone on an embassy to Rome about the year 168 B. c, on behalt
of Attalus II. We happen to know that — fortunately enough,
as it turned out — he fell into a drain near the Palatine and broke
his leg, which detained him in Rome longer than he intended,
and ' throughout all the time both of his embassy and his illness '
he gave constant lectures and industriously explained his views '.^
It is perhaps curious that the remains of Aristarchus make no
mention of Pisistratus, nor of any Attic recension. The remains
are not nearly full enough to justify us in assuming that he never
wrote of the question at all. But he had less need than most
people to speak of it because he held the theory that Homer was
himself an Athenian, not an Ionian, and that consequently the
crudest Athenian forms needed no explanation.

The testimony is not quite uniform. Most of the authorities
agree with Cicero. one text speaks of Hipparchus, the son ot
Pisistratus. This is hardly a contradiction : the policy was the

Tpi's [le TvpaWTjaaira ToaavriiKH e'^eSt'co^e

S/J/xo? 'Ep«;^^ei6a)i', kui rpls eTreaTrdcraTo,
Tov fifyav fv (BovXa'is UncriaTpaToV bs tov' Ofiijpov

rjdpoicra, (TTTopaBrji) to nplf iuiS6p.€voif.
tjpiTfpos yap Kflvos 6 ;^pi^creof rjv noXirjTrjs,

finep Adrjvaioi 2p.vpvav eTrcoKiVa/xei'.

Diog. Laert., i. 57 ''" ''■^ 'Op;pou e^ Ino^oXiis yeypacpe pn\lrco^€iadai, oTov ojrcv
6 npaiTos eXrj^tv (ipj^eadai top e\6neiiop. paXXop ovv ^oXcov "Oprjpov i(^a)Ti<T(v rj
YleLcricTTpaTOS, oi? (pr]ai Aifv^'-^as iv TrefinTUi MeyapiKaiv' . . . i)v 8e fidXiaTu tu inr}

ravTi' oX 8' ap^'Adrjvas elxov Km to. e^rjs. (A well-known Athenian interpolation,
B 546 ff.) .

Ps. Plat. Hipparch., p. 228 B ('iTrnapxos) to. 'Ofirjpov eni] TrpwTOs fKofiiaev fts
Ttjv yrjv TavTT]in, Koi i]vdyKacre tovs pa\f/-(o8ovs Uavadrjvoioii e^ xmoXrjy^tcis ((pe^rjs
avra biUvai, axjirep vvv en otSf noiovaiv.

Aelian, V. H. xiii. I4"Y(7Tepoi/ Ileto-io-rpoTos awayayidv dn(<f)r]ve Tfjv'lXui8ci Kal
TTjV '08v(T(Ttiav.

Suidas, V. "Opr/poj : ^ YcrTepov avviTidq Ka\ avveTii)(dri inro noXXap, Ka\ fiaXuTTa
viro neKTidTpi'iTov, TOV Tap A6r]Pata>u Tvpdvpov.

Eustathius p. 5 on ev pev TL (TMpa avvfX^s 816X0V Ka\ evappoTTOP 17 tJj?
'iXtdSos 7roi7]ais' 01 Se avvdipevoi TavTt]v, Kar (niTayi'iv, toy (fxiai, IliKJiaTpdTov tov
Tiov ^Adqvaicov Tvpdvvnv . . , KnTiTepov avTo fis KoXXd (i.e. divided it into booksj.

^ Suet. GrcDmn. et rhet. ii, p. 100.

2760 X


policy of the Pisistratid family. But Dieuchidas says it was vSolon
who ordained the recitation, and ' thus threw more light on
Homer than did Pisistratus who . . .' and there comes the gap in
the text. The words imply some knowledge of the Pisistratus
tradition, and apparently some criticism of it, and they attribute
the recitation law to Solon. on the face of it this does not seem
probable. In Solon's time there was very likely no such thing
as the Panathenaea ; pretty certainly there was not yet an
authoritative Pan-Ionian policy ; and we must remember that the
name of Solon, as ' the lawgiver ' par excellence^ had a habit of
attracting to it the credit for all good laws whatever.^

On the whole, the Pisistratus tradition stands its ground. It is
by no means certainly true ; it is not very clear in its statement.
But it accords with the general probabilities of history ; it is fully
as clear as a sober scholar would expect in a tradition about
mere literary history in an age before the annals of literature
had begun. And I am bound to say that the more I study the
traditions of the good Scholia or the Grammatici of Roman
times, the less am I inclined to suspect them of gross carelessness
or wilful invention. In the history of Drama we give credence
to many texts far later and less strongly attested. In any case,
the Pisistratus tradition marks the utmost limit of our Homeric
record. That last little glimpse of firm land may, of course, be
only an illusion. Beyond it, at any rate, we must steer our best
on a sea without a shore.

The study of these great poems is still involved in confused
and sometimes in curiously bitter controversy. This means, of
course, that no advanced critic has yet completely solved the
problem before him ; probably no wise critic ever for a moment
imagined that he had. It may be that the most helpful solution
will be something which no one has yet thought of. But in the
meantime, without expecting agreement about results, we might,
I think, try to agree about our approach to the Homeric Question.
We might distinguish the data from the problem.

' The romance about the travels of Lycurgus of Sparta, in which he meets
Thales and Homer and collects the wisdom of the Egyptians and the secrets
which Rhadamanthys learned from Zeus, ought not by any critical scholar
to be brought into this connexion. Strabo, p. 482 ; it has the compromising
support of Heraclides Ponticus, Pol. ii. 2 (=F. H. G., ii, p. 210).


The Homeric Question can never be solved by start ingf from
the question, Who was Homer and when did he live ? ' Homer '
is a hypothesis and not even a clear hypothesis. The Iliad and
Odyssey are known facts, and it is from the facts that we should
start. The Iliad and Odyssey are two poems dealing with
ancient heroic material selected for public recitation in Athens at
the g-reat festival of the Panathenaea, at or just about the founda-
tion of that festival by Pisistratus, and after that time regularly
recited every four years during the classical period. So much
we can say with reasonable certainty, without going beyond
the bounds of our evidence ; and if we grasp this first fact firmly
we shall find a great deal of light thrown upon many remoter

We have seen that the Homeric poems ' and no others ' ^ were
recited at the great Panathenaea in the generations before and after
Plato. We are definitely told that Pisistratus was the founder of
the great Panathenaea, and have evidence ^ to the effect that
either Pisistratus, or his son Hipparchus — who was much given
to poetry and the arts — instituted the recitation, and fixed the
order in which the poems were to be recited. Before that the
poems were ' preserved by memory ', and were ' different in
different places', so that Pisistratus may be said to have
' arranged in their present order the books of Homer which
before were in confusion '.

The law commanded the poems ' payjrcoSeiaOaL \ to be ' treated
by rhapsodes ', and the word rhapsode was generally interpreted
as a 'song-stitcher' or, in Pindar's paraphrase, a 'singer of
stitched lays'. That is, the songs or lays were already in
existence, but the rhapsode ' stitched them together ' into wholes,
longer or shorter according to the occasion and the time avail-
able. These lay-stitchers were also called Homeridae, ' sons of
Homer', or Homeristae, ' Homerizers ', and presumably specialized
in the works of their 'divine ancestor'. But it needed a very
special effort of lay-stitching to produce out of their old epic
material unities so large and finely knit as the Iliad and

That the Iliad and Odyssey^ and not the Thebais or the poems
of the so-called Cycle, were the ' Homeric Poems ' selected for
^ But see below, p. 309. ^ See p. 304 f. note.

X 2


the Panathcnaea, may be taken as practically certain.^ And the
selection of these two poems for the Panathenaea seems to be
the simplest explanation of their survival, in complete form, when
all the rest of the abundant epic litei^ature perished. Lastly,
this hypothesis alone seems to explain the otherwise astonishing
fact that Attic tragedy, while drawing almost all its material from
the heroic saga, carefully abstains from using the Iliad and
Odyssey. These two poems were performed at the Panathenaea ;
consequently the performances at the Dionysia had to take other

It is quite possible that the name is actually preserved of the
chief rhapsode who worked up the Iliad and Odyssey into their
present shape. There is a note by a certain Hippostratus,
a chronicler of the Alexandrian age, quoted in the valuable
scholia to Pindar, Neinea 2, i. 'The name Homeridae was
first applied to Homer's descendants who performed his poems
in succession to him in Chios, but afterwards to other rhapsodes
who did not claim descent from Homer. Distinguished among
them was the school of Kynaithos (0/ ir^pl Kvi^aiOoy), who are
said to have interpolated much of their composition into Homer s
poetry. Kynaithos was a Chian, and is supposed to have made
the Hymn to Ai)ollo which bears Homer's name. This
Kynaithos was the first person to perform (paylra)Se?i') Homer's
poems at Syracuse about the 69th Olympiad (504 B. c.).'

I am disposed to think that this is probably true. Kynaithos
' made ' the Hymn to Apollo in the sense in which Athenaeus
speaks of ' all those who have made the Heracleia ' ; that is, he
did a poetical version, a noir]aii\ of it. If ' much ' of our Homer
is the ' composition ' of him and his school, perhaps it was they
who put together for Pisistratus the present shape of the Iliad
and Odyssey. When the Pisistratidae fell, in 510 B. c, Kynaithos
sought refuge elsewhere, and in a few years' time is found
reciting at Syracuse, doubtless at some great festival, the
Iliad and Odyssey which he had once recited for Pisistratus.
In this he was doing the normal work of a rhapsode, though
a specially distinguished example of it. For, as the same
scholion proceeds to say : ' At a time when the poetry of Homer
was not yet brought into one body, but dispersed, variable, and

^ See above, p. 300.


divided in parts, the rhapsodes who produced it at performances
did something equivalent to "joining " and " stitching" when they
brought it into a unity. That is how Pindar understands the
word.' And a Httle later : ' They knew by heart and repeated
the poetry of Homer in its scattered state ; and they did great
violence to it '— that is, altered it greatly as they worked it up
into new unities.

We saw above that this action of Pisistratus falls excellendy
into line with his general policy. He wished .to assert the
claim of Athens to be the Mother-City of all the lonians and the
champion of Hellas against the Orient. It is significant that the
only poem which ever received the glory of being recited, or
read aloud, at the Panathenaea together with Homer was the
Perseis, or epic of the Persian War, by Choirilos.^ And,
secondly, he wished to make Athens the centre of Ionian
culture in place of Miletus and the other Ionian cities which had
recently fallen a prey to Croesus of Lydia (about 560 B. C).
The Panathenaea served both ends.

This, then, is the recorded history, in itself consistent and
credible. Is it of any help towards understanding the Iliad and
Odyssey ? In the first place it solves the most pressing of all the
problems : it tells us at last what the two poems are. They are
not court-poems, nor yet folk-poems : they are panegyris-poems,
a kind which does not exist anywhere outside Greece. The
puzzle has always been to understand what the Iliad and
Odyssey were meant for, or how such immense poems came to
be composed. To be read ? But there was no reading public ;
and the tradition always speaks of them as ' sung ' or recited.
To be recited as wholes ? But, by ordinary standards, they are
enormously too long. To be recited in bits, like lays } But they
do not fall apart into lays. one of the most certain results ot
Homeric criticism since Lachmann is that all attempts to divide
the poems into recitable lays have ended in complete failure.
The Iliad and Odyssey are unities ; elaborate, well-constructed
unities, composed with infinite pains out of discrepant materials.
All those pains were not only wasted but were positively self-
defeating if the poet's object was to recite his poem in parts
round the camp-fire or in the banquet-hall. If he had to do

^ The Perseis avv toIs 'ontfpov avayiyvMaKiddai e^r](pi(T6i]. Suidas, S. V. XoipiXos.


that, it was lays he wanted: lays about the length of an ordinary
tragedy, those very lays in fact which he had just spoiled in build-
ing his two great works of art ! The two facts given us are :
first, that the Iliad is a poem originally meant for recitation ;
second, that it does not fall into separate lays and that the whole
would take some twenty-four hours to recite. Conclusion : the
poem must be intended for some extraordinary occasion, demand-
ing even greater enthusiasm and powers of endurance than the
annual celebrations of tragedy at the Dionysia.

The same consideration solves another ot the principal
problems over which controversy has raged since the time ct
Wolf. The old practice, still largely current, was to divide
Homeric critics into Unitarians and Separatists. It was a wrong
division ; because both views are right. The poems are unities ;
while various parts of the unity are discrepant in substance, and
divergent in date and style. For several generations Homeric
scholars tried to treat this problem by supposing that the unity
came first and the discrepancies were ' late additions ' or ' inter-
polations '. There was an original poem of manageable length
by ' Homer ', containing no flaws or contradictions ; then came
interpolating rhapsodes and diaskeuasts, added to it and spoiled
it. The method may be seen at its best in Dr. Leaf's great
commentary. But even at its best it has always failed. For one
thing, no amount of excision has ever resulted in producing the
required uncorrupted poem, not even if allowance is made for
supposed omissions. For another, it has generally been possible
to show that wherever a passage was inconsistent with the rest
of the poem conscious pains have been taken to explain away or
bind together the discordances.^ Again, it was often difficult to
understand the motive of an ' interpolator ' in adding something
of his own which simply made trouble. For these and other
reasons, scholars have been driven to try the alternative
hypothesis : that the discrepant versions were there first, and the
unity has been imposed upon them by the ' song-stitcher '. It is
not a new hypothesis. It is simply that described in the scholia
to Pindar. The ' work of Homer ' — and all epic poetry was the

^ e.g. the scruple of Diomede in Z 129 lest Glaucus should be a god, after
the assurance of Athena in E 124-32. See the long note by Porphyry in
Schol. Ven. B.


work of Homer — was ' scattered ' and ' different in different
places ', recited from memory by ' Homeridae ' ; and the song-
stitchers put them together, ' doing great violence to the poems '
in the process.

For example, Book I of the Iliad^ in which Agamemnon
beseeches Achilles to accept atonement and is refused, seems
inconsistent with certain statements and situations, especially
in Book II, which imply that Agamemnon had made no such
offer. If we say that Book I is a ' later addition ', we do not
explain at all why the maker of the addition should go out of
his way to compose a lay contradictory to the rest of the poem.
But if we suppose that the 'song-stitcher' found in his store
some good material which implied one situation and some which
implied another slightly different, and that both versions were
good in themselves and both already dear to parts of his
audience, it is only natural that he should wish to use both.
Again, in our present Odyssey there is a motive twice repeated
in which the disguised Odysseus is injuriously treated by one of
his own servants. In one part, o- and r,^ the treacherous servant
is a maid Melantho, daughter of Dolios ; in another,^ p, i;, and
especially and x? it is a man, Melanthios or Melantheus, son of
Dolios. The Melantho part never mentions Melanthios ; the
Melanthios part never mentions Melantho, not even when it
comes to the hanging of the wicked handmaids in x 47° ff. It is
surely most unlikely that one is original Homer, and the other a
' late addition '. Who would deliberately add such a monotonous
doublet? If both were varying versions, current in different
schools or in the stock of the same school of poets, one can
readily see how the stitcher took in both, and sewed them

On the other great problems also light is thrown by the same
conception : they all show new use of old material. The Homeric

^ a 321-40, T 65-92.

* p 247, 369; V 173, 255 ; (/) 176, 181, 212, 265; X 142, 152, 159, 161, 182,

19s, 474.

^ In the same way there are three cases where Odysseus has something
thrown at him by the suitors : in p 462 Antinoos hits him hard with a stool
(^/j^vvy), in a- 394 Eurymachus misses him, but fells a cupbearer, with a stool
(ff^fXcis), in V 299 Ktesippus misses him with a cow's foot. The passages do
not form a climax and seem to have no relation to one another. Independent
stories have apparently been combined.


language is very old Greek, some of it actually unintelligible to
Attic ears, worked over and over by Ionic and Attic rhapsodes,
or in the last instance by Ionic rhapsodes and Attic copyists.
The Homeric culture and civilization are not the naive presenta-
tions of a primitive age by a primitive poet, but the conscious
descriptions of an idealized heroic age by one to whom that age
is far past. This is proved by the fact that various modern
objects are mentioned in similes, but not in the body of the poem
(p. 121 f ). And this heroic past as described in Homer has two
marked characteristics, which at first sight seem contradictory :
it shows, with some lapses, remarkable unity and consistency, as
if it were the description of a real historic age, and, on the other
hand, it entirely refuses to correspond with any historical age
known to us. It is not Minoan or Mycenaean ; it is not, within
the limits of our knowledge, post-Mycenaean. It is an idealized
past, when men were heroes and lived, ate, dressed and talked as
heroes should. The unity is not an original fact, marred and
confused by subsequent interpolations ; it is a unity subsequently
imposed on confused materials by conscious art.

The amount of Attic influence actually visible in the poems
has been variously estimated. It seems to me just what one
might expect : quite comparable, except in the matter of dialect,
to the amount that is visible in such a purely Attic creation as
tragedy. In tragedy for the most part we find the great myths
of the heroic tradition treated simply for their artistic value :
Agamemnon, Orestes, Oedipus, Ajax, Heracles are not made
into material for patriotic propaganda any more than Hamlet is
by Shakespeare. At the most we find a few local Attic myths
chosen as subjects, one or two praises of Athens in the choral
odes, an occasional appearance of Theseus where a non-
Athenian poet might not have thought of him.

Even the dialect shows more than the mere mistakes of Attic
copyists. Forms such as ijXios by rjiXios, eVecr^ai by orrd^^Lv,
OTTooi &c. for Aeolic okk- and Ionic ok- : d/ioOev ye, ^e^axra, tJuto,
Ik€?i/to seem to reveal the hand of Attic poets. And is it not
significant to meet the form ioia^opos in the description of
a dawn which ' rises over the sea ' } Significant also that in the
Odyssey in its present form the adventures of Odysseus are
turned into a contest between Poseidon and Athena, a contest


which forms one of the most central and characteristic of Athenian
myths, and, as in the Attic myth, Athena is the true friend and
has the best of it ; that Athena is extraordinarily prominent in
what seem to be the later parts of the poem, so that Odysseus is,
as it were, annexed by the Athenian Kore ; that a sympathetic
minor character who protects and helps Telemachus is actually
called ' Pisistratus ' and is made the son of the aged Nestor — from
whom Pisistratus of Athens claimed descent; that, in order to
prevent any mistake, when Athena leaves Odysseus and goes
home, we are told that she flies straight off to the ' strong House
of Erechtheus ' — in which she lived on the Athenian acropolis !
Such instances can be multiplied.^

It is the same with the Iliad. Take, for example, the
Catalogue of Ships in B. By general agreement that is a docu-
ment of respectable antiquity, not composed for its present place,
but inserted there at some time when the Iliad was nearly
complete, and by additions, omissions, and alterations, accommo-
dated to its surroundings. If we compare it with the rest of the
Iliad, taken as a whole, two classes of peculiarity strike us.
First, there are peculiarities of origin. It is in the style of the
Boeotian school, and magnifies the Boeotians; it also divides
Greece on a principle quite different from that usual in the poem.
Then, secondly, there are peculiarities of revision : additions
made to the original text or omissions from it. If we consider
the spurious additions, most of them are merely editor's alterations
to fit the Catalogue into its context, but some are markedly
Athenian. There is the startling statement that Menestheus, the
obscure Athenian chief, was the best general, for both horse and
foot, in the whole Greek army; the significant single line,
covering apparently a large omission, which states that Ajax, the
hero of Salamis, drew up his ships among those of the Athenians.
Salamis had just been conquered and annexed by iVthens in the
time of Pisistratus.

The omissions, so far as we can trace them, point the same
way : Thebes, Aegina, Megara are omitted entirely ; Salamis is
suppressed, Corinth belittled. These five cities are the principal
enemies of Athens. It is also worth noticing that in the account

y (o}(T<p6pos^ 226, cf. il 12. See p. 191 n. Erechtheum »; 80 f : Pisistratus
y 36 ff. &C.


of the Athenian contingent no town in Attica is mentioned
except Athens. There is no word of Eleusis or Marathon or
Acharnae or Brauron : that must be the work of an Athenian,
writing long after the synoikismos of Attica. The result is
much what might be expected— except indeed by those who
talk always of Attic ' forgeries ' and ' Athenian vanity ' and
imagine that those are the motives by which great poetry is
produced. The Athenian colour is present, but it is not gross
or crude. one handsome mention of the troops of Athens, and
some not very conspicuous silences about the glories of her
principal enemies.

There is one further difficulty which is solved, or greatly
reduced, if once we grasp the circumstances of the Panathenaic
Recitation. If the text of the Iliad and Odyssey was fixed by
Pisistratus, and an official copy written out in the official script,^
how comes it that the text fluctuated so greatly afterwards.''
How, for instance, can we account for the great divergences
between our text and Aristotle's or Aeschines' quotations, and
the ' wildness ' of the early papyri .-* Let us remember that the
Pisistratean text was made for a great recitation, which occurred
only once every four years. Such a recitation would influence
but could not dominate the ordinary tradition, which would
continue to be formed by the frequent everyday recitations or
public readings of Homer, in lays or episodes or extracts. The
complete continuous text would not become impressed on
people's minds until two influences had time to work: the
teaching in boys' schools and the rise of an educated reading
public. It is quite conceivable, though not very probable, that
the Vulgate Text of Homer is really an attempted restoration of
the Pisistratean. But at least we can see why that text was not
likely to fix immediately the text of the ordinary recitations.

We may observe two further consequences that flow from a
clear acceptance of this tradition. It changes the Homeric
Question from a question of genuine and spurious, of original and
interpolated, to a question of Sources. All poems have sources :
and all poems in nations where there is no widespread reading
public are apt to use their sources just as they please, with no
scruples about what we call ' originality ' or ' literary property '.

* See above, p. 302 f.


The rhapsodes of our Iliad and Odyssey used the Thebais and
the NosioiyN'whoMX. disguise or shame, as doubtless the rhapsodes
of the Thebais or the Nosioi used their Iliad and Odyssey.
This simplifies in one way the process of criticism, while on the
other hand it makes necessary much greater caution in using the
poems as historical documents. There has been, and still is, a
regrettable tendency among historians to use any part of the
poems, unless it is definitely condemned as ' spurious ' or ' a late
addition ', as being ' real Homer ', and therefore a first-hand
authority of about the tenth century B. c. This must be given
up. We can only say of any average Homeric statement:
' This was accepted in the sixth century B. C. as belonging to
the Heroic Tradition ' ; though, of course, in certain cases we
can proceed to argue about its original date. one is reminded
of the problems of Deuteronomy. A sober critic no longer
treats Deuteronomy as a work written by Moses but interpolated
by the priests who ' found it in the Temple ' in the year 621 B. C.
He treats it as a work ' found' in the year 621, which contains
ancient material from dijfiferent sources, both Hebrew and

And the beauty of the poetry ? Is that affected one way or
another ) I confess that, as explained in Chapter XI, the beauty
of the Iliad as poetry has to me that touch of the infinite, that
strictly incomparable quality which results when a beautiful
object is confessedly imperfect and inevitably suggests a beauty
beyond itself. ' Thus I deliver my message, but ah, if you had
heard my master ! ' I have noticed that readers of Homer who
have no theories of their own have the habit of freely ' thinking
away' pieces which they do not like — Athena's treachery in
Iliad X for example, or the superfluous transformations and
re-transformations of Odysseus in the later books of the Odyssey.
There is an instinctive sense that the real poem is somehow
more perfect and beautiful than this version that we happen to
have ; perhaps the same instinct that sent all the early critics
searching for spurious lines which were ' unworthy of Homer '.
And it is a perfectly sound instinct. I do not suggest that, if we
had all the library of Kynaithos open before us and all his
memory as well, we should find any given version which would
satisfy us as being absolutely complete and uncorrupted Homer.


Any traditional poem or drama — whatever may be true of an
' original ' one — exists only in a series of performances or versions,
each of which is an attempt to represent better and better an
ideal which is never reached. This is no high aesthetic doctrine,
but plain fact. Hamlet only exists to the full when performed ;
yet Hamlet is obviously something much greater than any
particular performance by Garrick or Irving or, for that matter,
by Burbage and Shakespeare. The Iliad is greater than any
given version of the Iliad. Kynaithos, let us say, did his best ;
and Kynaithos was evidently a brilliant poet with a wide know-
ledge of the epic literature. But Kynaithos knew that nothing
that he could re-create could ever be the full thing that was meant.
And it seems as if some effluence of that knowledge still reached
us as we read, and as if the rhapsode had thereby added to the poem
something which could not have been present in any really original
work by a self-conscious creator. He has added to it that touch
of spiritual hunger which creates a beauty beyond visible beauty.
Whether 'Homeros' is a mythical ancestor of poets, like
Amphictyon or Hellen, or whether there was once a bard called
' Homeros ' who acquired a mythical reputation, the Homer to
whom Kynaithos ascribed the Iliad and Odyssey^ but a more
perfect Iliad and Odyssey than could ever afterwards be
recovered, belongs now to the company of his own heroes, not
born in Chios or Athens or Smyrna, but cov d^l kv ovpaviwi tottcol.



As there has been a tendency of late, perhaps started by Rohde
{Psyche, p. 367, n. 4), to make out that the pharmakos rite was a real
human sacrifice in the full sense, it may be well to give verbatim the
more important texts on which Rohde based his opinion.

I. Ancietit Texts.

{a) Hipponax, several fragments : especially

4. TToXiv KaOaLpeiv koI KpaZrjcn /SdXXecrdai.

5. fSdXXovTes eV Xei/x,covt kol paTri'^ovTcs
Kpd8r](n Koi o-KLXXrjcriv, wcrirep (papfiaKOv.

6. Sei 8' avTov es (fiapp.aKov iKTroirjcraa-Oai.
7- Ko.(f)r] 7rape^€iv liTxd.Sa<; Tc Kat yua^av

Koi Tvpbv olov 1(t6lov(Tl (f)apixaKoi.
g. Xifiia yevrjTat ^po5, iv 8e Tw 6vfjL(Z
(f>apfxaK6'; d.)^6€l^ cTTTttKts paTTLcrOeir].
37. 6 8 i^oXiaOoiv iK£T£D€ Tr)v Kpdp-jBiqv

TTjv iTrrdcfivXXov, rjv [rj MSS.] 9v(.<tk(. ITavSwpT;,
TapyrjXLOiarLV cy^^vTOV Trpo cfiapfxaKov.

These in any case prove nothing about Athens. Hipponax was over
a century earlier than Aristophanes, and Ephesus was a town much
exposed to barbarian influences. But, even as to sixth-century Ephesus,
the fragments prove only : (i) that the Pharmakos-sacrifice was a known
ceremony, as for instance, breaking on a wheel, hanging, drawing, and
quartering, &c., are known to us, but that Hipponax /las to explain if.

(2) That some ceremony or other still went on which could be described
as a 'beating of the pharmakoi ', like our own burning of Guy Fawkes.

(3) It is worth remarking that all these phrases seem to occur in one
context, and the same is true of the passages in Attic Comedy. They
are all comic or rhetorical curses. Now in such curses it is on all
grounds more comic, and more effective, to invoke an obsolete and
imaginative punishment on your victim. The curses in Aristophanes
illustrate this. (Those invoked E<j. 928 ff., Ach. 1156 ff., or the threats
of Ran. 473 ff. have nothing to do with real life.) (4) No fragment


speaks of killing a pharmakos, and fr. 37, obscure as it is, speaks quite
clearly of the dough figure in place of a pharmakos. "Y^y^Tov = ' a cake
in a mould ' ; one of the regular substitutes for a real victim.

(J)) Aristophanes, Ratiae, 732 oicriv t) vroAts tt/jo tov \ ouSe <^a.p\i.a.Koi<Tiv
(LKfi paStws Ixprjo-aT av. This merely shows knowledge of the existence
of such a custom irpo tov, ' once upon a time.'

{c) Eq. 1 135 ff- Toi'crS' . . . wa-TTip S7][xoaiov^ Tpe<^€is . . . cira . . . Ovcra?
iTTtSenn'ei';. It is strange that any one should take this as evidence for
a pharmakos-sacrifice. Who would ' cook and dine on ' a pharmakos ?
The Scholiast (V) explains rightly that Siy/xdo-iot are animals kept and
fattened at the public expense.

{d) Eupolis, Z>e?not\ 120 (K) :

bv XPW ^^ '''^ ''"'^'•5 rpioSois Kttv Tois 6^v$vfiioL<;
TrpouTpoiraiov T^? ttoAcws KaearOai TerpiyoTa.

Merely a comic curse ; perhaps a literary reminiscence of Hipponax.
In any case it proves nothing about contemporary practice.

{e) Lysias vi. 53. 'The right thing would be aTraWaTro/AtVovs
AvSoKiSov Tr]v ttoXlv KaOaipuv koX aTroSLOTrofXTrci(r6ai Koi (fiapfxaKov
dTroTre'/ATTciv '. — Comic abuse, as before. But observe that Lysias thinks
of the pharmakos not as killed, but as ' sent away ', or banished.

II. Explajiatmis of Grammariajis.

A. Much the oldest, Ister : in Harpocration, s. v. t^ap/xaKo'?. Avo
avSpas ^ KOrivqa-iv i^rjyov, KaOdpcria iaofJievovs ttJ<; ttoAcws iv Toi? ©apyT^Xioi?,
cva /A€v vTTcp Twv dvSpwv €va Se vTrcp twv yvvaLKwv. [Originally a man
named Pharmakos had stolen cups from Apollo and virb Ttov -rrepl tov
'A)^iX.Xia KareXevcrOrj.j Kal to, rot? ©apyiyXtois dyop-eva toutwv aTTO/Ai/Aiy-
fiara iariv. lorpos cv a twv AttoXAcovo? linf^aveLwv.

Observe : they did not * kill ', they ' led out ' two people in a
procession ; and the ceremony was an ' imitation ' of stoning" to death.
Such ' imitation ' ceremonies were as common as can be in Greece.
(On the Achilles question see Lecture VIII on Thersites.)

B. Helladius, ap. Phot. Bibl. 1593 e^os rjv iv 'A^i^vats <f)apfiaKov<;
aycLv 8vo, tov fxev vir\p dvSpijjv tov SI VTrep yvvatKoiv Trpos KaOap/xbv
dyo/xeVovs. Kat 6 fxev twv dvSpu)v p-eXuLvas io-;(a8as Trtpt tov Tpd^rjXov etxev,
XevKOLS S' aTepos' orvjBaK^OL Se, (jirjcrLV, <j}vop.d^ovTO. It was an aTTOTpoTriao'/x.os
voo-wv in atonement for the death of Androgeos the Cretan.

This writer agrees with Ister, except that he does not happen to add
that it was a p.Lp.y]p.a. He probably took that for granted. The
imitation cannot have been very close, one would think, if some took it


for a stoning, others for banishment, others for burning. Androgeos
was killed in an ambush on the road to Thebes. We may conjecture
that he in some way f3a\\6/ji€vo^ a-n-iOave. This would give the stoning,
with KpaSat and aKiXXai : then the banishment would be the running
away of the real man ; the burning would be the burning of the eyxyrov
or effigy.

C. Tzetzes on the Hipponax passages : Tzetz. C/iiL v. 726, in case of
special calamity, tov Trdx'TMV AfiopcfyoTcpov yyov ws Trpos Ovcriav' I €19
TOTTOv h\ Tov Trp6(r(f>of3ov OT7/o-avT€? Trjv 6v(Tiav I Tvpov re Sovrcs r^ X^P^ '^^'■
fia^av Koi la^^dSa?, | €7rTa/<ts yap paTrtVavres eKcirov ct? to ttco? |
(TKtXXats^ (ruKais dypi'ais t6 Kat aAAots twv aypiuiv, \ tcXo? irvpi
KareKaLov Iv ^vXoif; toi? dypioL<;. | Kat toi/ (nrohov cts 6a.\a(T(Tav Ippaivov cts
dve'/AOVS. 6 Se 'iTTTrwva^ ktA. (fr. 4-9)-

I do not feel sure what object Tzetzes meant to be supplied to
KareKatov, Did they burn ' him ' or only ' it ', sc. t^v Ova-iav, i. e. the
lyX^Tov or effigy ? It seems to be distinguished from Ikuvov, the man
who ' was led out ' ws ctti Ova-iav, * as though to sacrifice '. But perhaps
Tzetzes did not really understand the source which he was quoting : he
seldom did, being an inaccurate writer, and 1500 years later. So far,
then, there is no single statement that the pharmakoi even at Ephesus,
much less at Athens, were really sacrificed. But now we have two such

{a) Schol. Equites, 1. c. The first part of the note given in the best
MSS. explains quite rightly S-rjixoo-iov-:- Xu-n-ei /3ov<; y Tavpov<;. The second
says erpeifiov yap rivas KQ-qvaloi Xiav dye wet? Kat d^p^crTOVs Kat Iv Katpw
<rvp,(fiopa<; rtvos iTreX6ova-r]<s rfj ttoAci, Xoifxov Xiyw ^ tolovtov tivo?, tdvov
TovTous evcKtt rov KaOapOrjvai rov pnaa-p.aTo'i. And presumably ate them,
as we remarked above !

This note (i) is absent from R and V, the two good sources:
(2) shows itself by its language as belonging to a bad period of scholia,
e.g. the Xoip-ov, Xe'yw, 7] ToiovTov rtvo's : (3) is obviously wrong as an
explanation of the passage to which it refers.

(The note in the good MSS. runs : XetVEi /SoGs 17 ravpov? ^ dXAo n
TOLOVTOV Ovfia. I 87]p.o(Tiov<; Se tov'S Xeyoftevov^ cfiapp.aKov<; oinep KaOaipovai
Ttt? TToAets T<S kavTMV cf>6v(i)' I ^ Tovs 8r][j.oaLa. Kat ivb tijs ttoAcojs Tpefftop-ivovi.
Of these three explanations, the first is obviously right. The second,
' the so-called pharmakoi, who cleanse cities with their blood ', is quite
vague, as well as wrong. It also occurs in Suidas, and probably did
not begin life as a note on this passage. The third is right as far as it

* Probably not the garden squill, but the wild bulb.


(d) Schol. Hanae, 733, one inferior MS., C, has a note : toi"; yap
ff>avXov<; Kal irapa. t^s <^v(TeoJS iTrt/SovXevo/xeyov^ €i? d7raXXayr;v avxp-ov rj
Xl/xov Yf Tivo^ T'2v TOLOvTOiv eOvov, ov<s iKaXovv Ka9dpfJLaTa. Exactly what
one expects in inferior scholia which abbreviate their sources ! He says
Wvov for short, because he was careless. He may have found i^yov evrc
dva-iav or yyov w? evrt OvcrLav. It is not necessarily false as it stands,
since no subject or date is given to Wvov ; but even if it said eOvov rore
ol 'AOrjvoLoi it would be worthless.

The general result is to show that (i) the ancient texts all come to
the same type : ' He ought to be tied on a cart and burnt in a bonfire
like a Guy.' They imply that a pharmakos-sacrifice was known to have
existed at some time somewhere : they suggest that some fxiixtj/xa of it
lived on.

(2) The best grammatical tradition explains that this plp.-qix.a did exist,
and partly what it was like.

(3) The worst and latest grammatical tradition, dropping the qualify-
ing clauses as its manner is, says that ' they sacrificed very ugly people '.

Even without the general considerations of probability advanced in
the text of Lecture I, this evidence clearly points to the Thargelia
ceremony being a ixifx-qixa. [Cf. also Stengel in Hermes, xxii. 86 flf., and
especially Farnell, Cults, iv. 270 ff.]

We give in full the Pelopidas story, which has actually been used as
evidence that the Greeks of the fourth century had no objection to
human sacrifice.

Plutarch, Pelopidas, xxi. (Before the battle of Leuctra, 371 B.C.
Pelopidas was encamped near the grave of certain Virgins who had
been, according to the tradition, violated by Lacedaemonians. They
had died, and their father had committed suicide upon their grave. A
fearful and haunted place !)

' Pelopidas dreamed that he saw the Virgins wailing about their tombs
and uttering curses upon the Spartans, and their father commanding
him to sacrifice to the Virgins a fair-haired Maiden if he wished to
conquer the enemy. The shocking and unlawful (Seivov Kal rrapdvofjioi')
command started him from his sleep, and he consulted his prophets and
officers. one party insisted that the dream should not be neglected or
disobeyed, producing precedents from ancient times, Menoikeus, son of
Creon, and Macaria, daughter of Heracles' [both of these devoted
themselves voluntarily], ' and in a later generation Pherekydes the wise,
who was flayed by the Lacedaemonians and his skin preserved by the
kings, according to a certain oracle ' [a mythical divine king, like


Frazer's Marsyas], ' and Leonidas, who in a sense sacrificed himself for
Hellas by the command of an oracle, and further the men sacrificed by
Themistocles before Salamis to Dionysus Omestes. These actions had
all been approved by subsequent success. on the other hand, Agesilaus
had led an army from the same place as Agamemnon and against the
same enemies ; the goddess demanded of him the sacrifice of his
daughter, and he saw the vision while sleeping at Aulis, but refused,
and through softness disbanded the expedition, which was inglorious
and incomplete.

' The others opposed such a view. No superior and more than
human beings could be pleased with so barbarous and unlawful a sacri-
fice. It was not the legendary Typhons and Giants who ruled the
w^orld, but one who was a Father of all gods and men. As for spirits
(Sat/xoves) who rejoiced in the blood and slaughter of men, to believe in
such beings at all was probably folly, but if they existed, they should be
disregarded, as having no power. Weakness and badness of nature
(^^Xv) w^s the only soil in which such monstrous and cruel desires
could grow and last.'

The arguments on both sides are interesting. The first set shows
what was possible to reactionary and superstitious individuals at a time
of great fear. The others speak the language of ordinary philosophic



This bad business is sometimes misunderstood and grossly overstated.
The torture of witnesses who are suspected of concealing important
facts has only in comparatively recent times been abolished in England
and France. In Athens this sort of torture was forbidden in the case
of freemen, but not in the case of slaves. To say that a slave could
not give evidence at all except under torture is absurd. He could of
course give evidence to a simple fact, e. g. where he witnessed a murder.
And, in a complicated case, Isaeus, Philoct. 16, seems to speak of a
proclamation inviting evidence from relations or slaves. The cases
where a slave's evidence was not good except under torture were those
where the slave had an obvious interest, such as personal complicity or
fear of his master. The typical case is where a man is accused of some
misdoing which his household must have known about. In such a case
the Court cannot seize his slaves and examine them without the

2780 Y


master's consent; but the Accuser can challenge him to hand them
over for examination under torture. The master, if he accepts this
proposal, can stipulate what tortures are to be used ; and if the Court
inflicts any permanent injury or any temporary loss of working power on
the slave, the Court, or the Accuser, as the case may be, has to pay
damages. To Roman or mediaeval torturers such a stipulation would
have made the whole proceeding nugatory.

It is worth observing that; (i) This challenge seems generally to
have been refused. (2) To accept it implied not only a consciousness
of innocence, but a strange confidence in the affection of your slaves.
One would expect a slave in such a situation to accuse his master of
everything that was desired, especially as he could acquire freedom
thereby, if his evidence was believed. (3) I can find no case mentioned
where a witness died under torture. Where torture is really severe such
cases seem to be frequent, from heart failure and other causes. on the
other hand, the Christian use of the word martyr, witness, is terribly
significant. To poor folk in Roman times a witness meant one who
suffered; but, of course, it was implied that the witness refused to
betray his master.

It looks as if this was one of the numerous cases in which Attic Law
preserved in the letter an extremely ancient power which was not much
used, or at any rate not to its full extent. (The scene in Frogs 620 ff.
is perhaps instructive. It is unpleasant and of course unjust, but does
not suggest much real cruelty.) The article Servus in Smith's Diet.
Afitiq. seems very sound.



There is extant a very curious and ancient Greek document which
throws some light directly on this Dark Age which followed the fall of
the Aegean empires and indirectly on the growth of the Epos. It is a
list of the various powers which have exercised what the Greeks called
' Thalassocratia ', or Rule of the Seas, from the fall of Troy up to the
founding of the Athenian League. The list is given by Eusebius with
slight omissions and discrepancies, both in the Chronographia and the
Cafiofies, and was taken by him from Diodorus.^ It bears well the tests

^ See the historical reconstruction by J. L. Myres in/. //. S. xxvi. i ; also
Fotheringham's criticism in/. H. S. xxvii and Myres' answer. Winckler's
discussion is in Der Alte Orient, vol. vii, part 2.


that have been applied to it, and seems to be drawn from authentic
sources, perhaps from a list set up in some Aegean temple.

The list starts with the fall of Troy. That catastrophe, by whatever
coalition of invaders it was immediately produced, is taken as typifying
the final downfall of the old Aegean system, a system which in Greek
tradition is represented by the ancient thalassocratia of Minos. But
what exactly is meant by a thalassocratia, or control of the seas? It
seems to mean something quite definite, not a mere general naval
preponderance, because the dates of the various ' controls ' are marked
off so precisely. And the wars of the Diadochi, in which first Demetrius
Poliorcetes, then Ptolemy, then Antigonus Gonatas are masters of the
Aegean, provide a very suggestive parallel. (See Tarn, Antigonus
Gonatas^ 1913-) The explanation is, I think, to be found in the peculiar
geography of the Aegean, and in the distinctive character of the great
Aegean centres. They were (pp. 36 ff.), generally speaking, fortified toll
stations ; the various cities of Crete commanding all the southern trade
routes ; Troy those of the Hellespont ; Thebes the trafi^ic between its
' three seas ' ; and even Mycenae, which seems so remote, some important
trade routes between the Aegean and the Corinthian gulf. And the
Aegean is so formed that both to the north, the south-east, and the south-
west the necessary routes of trade are well marked and narrow. The
whole of them together could be controlled by a really strong sea power,
though it is not likely that an ancient command of the seas was often
so complete as that. When one reflects on the amount of fighting
which went on in historical times for the possession of, say, the
Hellespont or Naxos, and the constant train of explosive maritime
rivalry, ever ready to burst out in commercial wars, such as that be-
tween Miletus-Eretria- Athens and Chalkis-Samos-Aegina, the conclusion
strongly suggests itself that the prize in each case was the control of
one or more of these five or six great passages or toll stations of the
Aegean, and that such control constituted 'thalassocratia'. A power
became completely ' thalassocrates ' as soon as it could establish a
guard of ships and forts at, say, the Hellespont, the channels of the
Cyclades round Naxos or Delos, the passages on each side of Carpathos,
and on each side of Ogylos, together with certain roads of more local
trade, like the Straits of Euboea.

Now, if we turn to the List of Thalassocrats, we find at the very
outset two phenomena which we might well have expected. First, for
a long time after the fall of Troy there seems to have been no
thalassocracy at all; and secondly, it is a very long time indeed,
certainly 400 years and perhaps 600, before there is a genuinely Greek

Y 2



thalassocracy. The Fall of Troy was dated by the authors of the list —
viz. the tradition represented by Eusebius-Diodorus-Eratosthenes— at
1 184 B. c. The list then runs ' :

Lydi et Maeones 92 years











33 or











?] years

— ?



Now the dates at the bottom of this list can be verified. The
Aeginetan thalassocracy certainly ended in 480 b. c. We work from
480 B. c. backwards, and find a considerable though of course a steadily
decreasing amount of historical confirmation as we go. There are one
or two confusions, notably a grave one at Nos. 10 and 11, the Carians
and Lesbians. These two powers have, in the first place, no specific
time of duration attached to them ; and, in the second place, there
seems to be very little room for either. But whatever we do with these
confused places, it is practically impossible to stretch out the dates
given in the list so as to fill the whole historical period between the fall
of Troy and the invasion of Xerxes. on Mr. Myres' arrangement there
is a gap at the beginning, directly after the Trojan War, amounting
to 128 or 138 years. on any plausible system there is about a century

Now what are we to make of this gap ? I suspect that it really is a
gap, and that after the fall of the old Aegean empires there was no
power strong enough or well enough organized to command much of
the Aegean beyond its own shores. Mr. Myres thinks that the Carians
have been transposed in the list. They are put tenth, where there is
no room for them ; they should have been first, where they are wanted.
There is evidence in Diodorus for this suggested rearrangement, and it
is quite likely to be right. But I would suggest that if we interpret the
language properly a Carian thalassocracy at that date is probably the
same thing as no thalassocracy at all. These race names are apt to be
loosely handled, as we saw in Lecture IL Diodorus and the Greek

^ I take the figures from Mr. Myres' list, marking the more uncertain
figures. The textual criticism of the list is highly complicated ; see Mr.
Fotheringham's article. He considers on purely textual grounds that
Eusebius' text gave Aegyptii 43, Cares 61, and Lesbii perhaps 68. The last
two figures would then be mistakes on the part of Eusebius or his authority.


historians frequently use the word Carian to denote the aboriginal or
pre-Hellenic inhabitants of the Aegean in general. Any rude and weak
creatures whom you drove out of an island were roughly described as
Carian. Take the most explicit passage, Diod. v. 84 :

After the capture of Troy the Carians increased and became
more powerful at sea: getting possession of the Cyclades they
seized some for themselves and drove out the Cretans who were
settled there, while they occupied others in common with the
Cretans who were there before. Afterwards when the Hellenes
increased, it befell that most of the Cyclades were colonized, and
the barbarous Carians driven out.

I suspect that one might put that statement in other words, thus :

After the fall of the Minoan or Aegean empires, under the
influence of the northern invasions, the first effect was not that the
northern invaders began to control the seas. They were not
advanced enough for that. It was that the subject populations in
the islands began to raise their heads, and especially formed a
small piratical power in the Cyclades. The guards of the local
Minoan forts, being cut off from their base, were faced with two
alternatives. They either resisted to the uttermost and perished.
Or they made terms with the natives, and eventually sank to their
level. When the Greeks came into existence as a people, they
found the Cyclades inhabited by populations who were a mixture
of the uncivilized Carian-Lelegian-Hittite natives and the isolated
remnants of the Minoan settlements.

The first thalassocracy mentioned on the list is that of the Lydians
and Maeones. Possibly some federation of the coast people of Asia
Minor arose, under the protection of Lydia, for resisting the piracy
of the Carians in the islands. It is nearly a century later that we find
the first suggestion of a thalassocracy of Northern invaders, and even
that is ambiguous. The Pelasgians, however, are possibly the definite
tribe of that name, the tribe which raided Boeotia during the Trojan
War, and, taking to the sea, made settlements in Lemnos, Attica, and
Crete. They at any rate are succeeded by a real Northern race, the
Thracians, who have left traces in the Maeander valley, in Naxos and
Attica, as well as in Boeotia and Phocis. From what we know of the
Thracians in historical times it is difficult to suppose that their control
of the seas amounted to more than vigorous piracy. Next comes the
first glimpse of something that seems Hellenic : the Rhodians are
thalassocrats from about 800 b. c. for the short space of twenty-three
years. But was Rhodes at that time a Hellenic island ? The settlement
of Rhodes is attributed by Greek tradition to a very early period,
perhaps to the end of the eleventh century. Wandering Dorians,


people from Megara in two relays, people from Crete and from Argos,
seem to have joined hands there. And it is quite likely that when
Rhodes began to use its geographical position, holding the south-east
gate of the Aegean, it deserved actually to be called a Hellenic power.
In any case, it could not long stand, and no other Hellenic power
could support or even succeed it. There follow Phrygians, Cyprians,
Phoenicians, Egyptians, covering some i6o years. The Cyprians were
scarcely Hellenic at this time, and the rest are plain ^dpfSapoi, though
we happen to know that the Egyptian sea-power depended a good deal
upon ' Ionian and Carian ' ships. The Greeks, it seems, could supply
the ships and the fighting material ; they could not yet supply the
permanent basis and organization. But that step was easy to take.
And when Egypt became distracted by the invasion of Nebuchadnezzar
in 604 B.C., the centre of gravity changed from the mouth of the Nile to
the harbour of Miletus, and the Aegean for many centuries to come
remained a Greek sea. Milesians 18 years ; Lesbians 4 ; Phocaeans 44 ;
Samians 17; Lacedaemonians 2; Naxians 10; Eretrians 15; Aegi-
netans 10 ; and then the Athenian Empire.



This central idea of Aidos has various ramifications in the ethics of
early Greek poetry. Most of the Homeric words of disapproval mean
something like 'excess', or 'going too far', and imply that there are
points where a man should check himself. The wicked are arda-daXoi,
'outrageous', vTr€pyj4>avoi, 'overweening', dSuwi, 'away from Dike', justice
or law : most of all, wickedness is "Y/Spa. That word is the antithesis
of a-(x)(fipoavvr] and of aiSws, and like its antitheses it defies translation
into our forms of thought. It unites so many ideas which we analyse
and separate : and it has a peculiar emotional thrill in it, which is lost
instantly if we attempt to make careful scientific definitions. We can
understand it, I think, in this way. Aidos — or Sophrosyne, which is
slightly more intellectual — implies that, from some subtle emotion inside
you, some ruth or shame or reflection, some feeling perhaps of the
comparative smallness of your own rights and wrongs in the presence of

' For a further analysis of these and other ideas of primitive Greek society
see Mr. F. M. Cornford's illuminating book, From Religion to Philosophy^
especially the early chapters : also Glotz, Solidarite de la Famille, and
fLt tides sociales et jtcridiques.


the great things of the world, the gods and men's souls and the portals
of life and death, from this emotion and from no other cause, amid your
ordinary animal career of desire or anger or ambition, you do, every
now and then, at certain places, stop. There are unseen barriers which
a man who has Aidos in him does not wish to pass. Hubris passes
them all. Hubris does not see that the poor man or the exile has come
from Zeus ; Hubris is the insolence of irreverence : the brutality of
strength. In one form it is a sin of the low and weak, irreverence ; the
absence of Aidos in the presence of something higher. But nearly
always it is a sin of the strong and proud. It is born of Koros, or
satiety — of 'being too well off ; it spurns the weak and helpless out of
its path, ' spurns,' as Aeschylus says, ' the great Altar of Dike ' {Ag. 383).
And Hubris is the typical sin condemned by early Greece. Other sins,
except some connected with definite religious taboos, and some derived
from words meaning ' ugly ' or ' unfitting ', seem nearly all to be forms
or derivatives of Hubris.

What relations are there between this group of ideas and the other
great conception of ' Justice ', Themis or Dike ? Both words have a
strong flavour of custom or precedent in them, but their meaning is
different. Themis is the Right Custom, the thing that is always done
and therefore legitimate, inside the group : Dike is the dispute between
two persons, or between the group and another group, and the Right
Decision that is given when such a dispute occurs. A King can utter
Themistes or Dooms, merely declaring the right way of behaviour. An
oracle is the seat of Themis; where the most ancient fountains of
knowledge, the dead ancestors, declare ' what is done ' under such and
such circumstances. This can occur without any argument or conflict,
whereas Dike is the ' thing done ', or the ' right custom ', in trying and
judging a dispute.

Thus Dike and Themis are themselves one of the bonds which Aidos
enables you to feel. You feel ' what a man has a right to expect '. If
your neighbour takes one of your cattle, you will naturally apply to the
judges to make the man give it back, with perhaps something extra for
damages. That is what is always done: what you have a right to
expect. If the judge, having received bribes from your neighbour,
refuses to hear you, then you are aggrieved : that is not Dike, not the
normal course. The judge has no Aidos. The people, and the gods,
will feel Nemesis. ' Dike ' is associated with trials, while ' Themis '
seems rather specially to be concerned with the keeping or breaking
of Oaths.

False Swearing, though it is not mentioned in Hesiod's list of the five


deadly offences, is in general one of the most typical and most loudly
cursed of ancient sins. And its relation to Aidos is very close.

The word Horkos, which we translate an oath, really means * a fence ',
or * something that shuts you in '. The process by which the oath becomes
important is this. You make to a man some statement or promise, and
then he requires some Trto-Tts, some opKos — a Trtorts to make him feel
confident, an o/skos to fence you in. The simplest form of ' Horkos ',
and according to Medea (Eur. Med. v. 21) the greatest, is simply to
clasp hands. With more formality you can, both of you, call upon the
gods, or the daimones who happen to be present in the air about you, to
witness the spoken word. Or you can ensure their presence by calling
them to a sacrifice. And, instead of being satisfied with the general
Nemesis which these divine witnesses and judges will feel if the word is
broken, you and your friend can specify the exact punishment which the
gods are to inflict upon you if you fail. That is the Horkos, the 'sanction '
which binds the speaker. In general, covenant by oath belongs to a
form of society which cannot enforce its judgements. It is ultimately
an appeal to Honour, to Aidos, Of course priests and prophets may
thunder about the vengeance which the gods will exact for a breach of
the covenant which they witnessed : but that sort of vengeance has in
all ages of the world remained a little remote or even problematical.
The real point of importance is that there is no vengeance by men, and
no available human witness. The man who has sworn is really face to
face with nothing but his own sense of Aidos, plus a vague fear of gods
and spirits, who are for the main part only the same Aidos personified
and wrapt in mythology. The thing that makes the perjurer especially
base, or dvaiS???, is precisely his security from danger. I knew once a
perfect case of the simplest Horkos. A certain Egyptian wished an
Englishman to take a quantity of antiquities to Europe and sell them
for him. The Englishman accepted the trust, and drew up a full
catalogue of the articles, with a list of the prices which he might expect
to get for each of them. The Egyptian shook his head at all this com-
plication of securities : ' I would like ', he said, ' if you will sha