The great strength of the theory that the poems are the work of several ages is the existence in them of various strata of languages, earlier and later.

Not to speak of differences of vocabulary, Mr. Monro and Mr. Leaf, with many scholars, detect two strata of earlier and later grammar in Iliad and Odyssey. In the Iliad four or five Books are infected by "the later grammar," while the Odyssey in general seems to be contaminated. Mr. Leafs words are: "When we regard the Epos in large masses, we see that we can roughly arrange the inconsistent elements towards one end or the other of a line of development both linguistic and historical. The main division, that of Iliad and Odyssey, shows a distinct advance along this line; and the distinction is still more marked if we group with the Odyssey four Books of the Iliad whose Odyssean physiognomy is well marked. Taking as our main guide the dissection of the plot as shown in its episodes, we find that marks of lateness, though nowhere entirely absent, group themselves most numerously in the later additions ..." {Footnote: Iliad, vol. ii. p. X.} We are here concerned with linguistic examples of "lateness." The "four Books whose Odyssean physiognomy" and language seem "well marked," are IX., X., XXIII., XXIV. Here Mr. Leaf, Mr. Monro, and many authorities are agreed. But to these four Odyssean Books of the Iliad Mr. Leaf adds Iliad, XI. 664-772: "probably a later addition," says Mr. Monro. "It is notably Odyssean in character," says Mr. Leaf; and the author "is ignorant of the geography of the Western Peloponnesus. No doubt the author was an Asiatic Greek." {Footnote: Iliad, vol. i. pp. 465-466. Note on Book XI. 756.} The value of this discovery is elsewhere discussed (see The Interpolations of Nestor).

The Odyssean notes in this passage of a hundred lines (Iliad, XI. 670-762) are the occurrence of "a purely Odyssean word" (677), an Attic form of an epic word, and a "forbidden trochaic caesura in the fourth foot"; an Odyssean word for carving meat, applied in a non-Odyssean sense (688), a verb for "insulting," not elsewhere found in the Iliad (though the noun is in the Iliad) (695), an Odyssean epithet of the sun, "four times in the Odyssey" (735). It is also possible that there is an allusion to a four-horse chariot (699).

These are the proofs of Odyssean lateness.

The real difficulty about Odyssean words and grammar in the Iliad is that, if they were in vigorous poetic existence down to the time of Pisistratus (as the Odysseanism of the Asiatic editor proves that they were), and if every rhapsodist could add to and alter the materials at the disposal of the Pisistratean editor at will, we are not told how the fashionable Odysseanisms were kept, on the whole, out of twenty Books of the Iliad.

This is a point on which we cannot insist too strongly, as an argument against the theory that, till the middle of the sixth century B.C., the Iliad scarcely survived save in the memory of strolling rhapsodists. If that were so, all the Books of the Iliad would, in the course of recitation of old and composition of new passages, be equally contaminated with late Odyssean linguistic style. It could not be otherwise; all the Books would be equally modified in passing through the lips of modern reciters and composers. Therefore, if twenty out of twenty-four Books are pure, or pure in the main, from Odysseanisms, while four are deeply stained with them, the twenty must not only be earlier than the four, but must have been specially preserved, and kept uncontaminated, in some manner inconsistent with the theory that all alike scarcely existed save in the memory or invention of late strolling reciters.

Now, even in ages of writing, it has been found difficult or impossible to keep linguistic novelties and novelties of metre out of old epics. We later refer (Archaeology of the Epic) to the Chancun de Willame, of which an unknown benefactor printed two hundred copies in 1903. Mr. Raymond Weeks, in Romania, describes Willame as taking a place beside the Chanson de Roland in the earliest rank of Chansons de Geste. If the text can be entirely restored, the poem will appear as "the most primitive" of French epics of the eleventh and twelfth centuries. But it has passed from copy to copy in the course of generations. The methods of versification change, and, after line 2647, "there are traces of change in the language. The word Ço, followed by a vowel, hitherto frequent, never again reappears. The vowel i, of li, nominative masculine of the article" (li Reis, "the king"), "never occurs in the text after line 2647. Up to that point it is elided or not at pleasure.... There is a progressive tendency towards hiatus. After line 1980 the system of assonance changes. An and en have been kept distinct hitherto; this ceases to be the case." {Footnote: Romania, xxxiv. pp. 240-246.}

The poem is also notable, like the Iliad, for textual repetition of passages, but that is common to all early poetry, which many Homeric critics appear not to understand. In this example we see how apt novelties in grammar and metre are to steal into even written copies of epics, composed in and handed down through uncritical ages; and we are confirmed in the opinion that the relatively pure and orthodox grammar and metre of the twenty Books must have been preserved by written texts carefully 'executed. The other four Books, if equally old, were less fortunate. Their grammar and metre, we learn, belong to a later stratum of language.

These opinions of grammarians are not compatible with the hypothesis that all of the Iliad, even the "earliest" parts, are loaded with interpolations, forced in at different places and in any age from 1000 B.C. to 540 B.C.; for if that theory were true, the whole of the Iliad would equally be infected with the later Odyssean grammar. According to Mr. Monro and Sir Richard Jebb, it is not.

But suppose, on the other hand, that the later Odyssean grammar abounds all through the whole Iliad, then that grammar is not more Odyssean than it is Iliadic. The alleged distinction of early Iliadic grammar, late Odyssean grammar, in that case vanishes. Mr. Leaf is more keen than Mr. Monro and Sir Richard Jebb in detecting late grammar in the Iliad beyond the bounds of Books IX., X., XXIII., XXIV. But he does not carry these discoveries so far as to make the late grammar no less Iliadic than Odyssean. In Book VIII. of the Iliad, which he thinks was only made for the purpose of introducing Book IX., {Footnote: Iliad, vol. i. p. 332. 1900.} we ought to find the late Odyssean grammar just as much as we do in Book IX., for it is of the very same date, and probably by one or more of the same authors as Book IX. But we do not find the Odyssean grammar in Book VIII.

Mr. Leaf says, "The peculiar character" of Book VIII. "is easily understood, when we recognise the fact that Book VIII. is intended to serve only as a means for the introduction of Book IX...." which is "late" and "Odyssean." Then Book VIII., intended to introduce Book IX., must be at least as late as Book IX. and might be expected to be at least as Odyssean, indeed one would think it could not be otherwise. Yet it is not so.

Mr. Leaf's theory has thus to face the difficulty that while the whole Iliad, by his view, for more than four centuries, was stuffed with late interpolations, in the course of oral recital through all Greek lands, and was crammed with original "copy" by a sycophant of Pisistratus about 540 B.C., the late grammar concentrated itself in only some four Books. Till some reasonable answer is given to this question—how did twenty Books of the Iliad preserve so creditably the ancient grammar through centuries of change, and of recitation by rhapsodists who used the Odyssean grammar, which infected the four other Books, and the whole of the Odyssey?—it seems hardly worth while to discuss this linguistic test.

Any scholar who looks at these pages knows all about the proofs of grammar of a late date in the Odyssey and the four contaminated Books of the Iliad. But it may be well to give a few specimens, for the enlightenment of less learned readers of Homer.

Turning to the vocabulary: "words expressive of civilisation" are bound to be more frequent, as they are, in the Odyssey, a poem of peaceful life, than in a poem about an army in action, like the Iliad. Out of all this no clue to the distance of years dividing the two poems can be found. As to words concerning religion, the same holds good. The Odyssey is more frequently religious (see the case of Eumaeus) than the Iliad.

In morals the term {Greek: dikaios} is more used in the Odyssey, also {Greek: atemistos} ("just" and "lawless"). But that is partly because the Odyssey has to contrast civilised ("just") with wild outlandish people—Cyclopes and Laestrygons, who are "lawless." The Iliad has no occasion to touch on savages; but, as the {Greek: hybris} of the Wooers is a standing topic in the Odyssey (an ethical poem, says Aristotle), the word {Greek: hybris} is of frequent occurrence in the Odyssey, in just the same sense as it bears in Iliad, I 214—the insolence of Agamemnon. Yet when Achilles has occasion to speak of Agamemnon's insolence in Iliad, Book IX., he does not use the word {Greek: hybris}, though Book IX. is so very "late" and "Odyssean." It would be easy to go through the words for moral ideas in the Odyssey, and to show that they occur in the numerous moral situations which do not arise, or arise much less frequently, in the Iliad. There is not difference enough in the moral standard of the two poems to justify us in assuming that centuries of ethical progress had intervened between their dates of composition. If the Iliad, again, were really, like the Odyssey, a thing of growth through several centuries, which overlapped the centuries in which the Odyssey grew, the moral ideas of the Iliad and Odyssey would necessarily be much the same, would be indistinguishable. But, as a matter of fact, it would be easy to show that the moral standard of the Iliad is higher, in many places, than the moral standard of the Odyssey; and that, therefore, by the critical hypothesis, the Iliad is the later poem of the twain. For example, the behaviour of Achilles is most obnoxious to the moralist in Iliad, Book IX., where he refuses gifts of conciliation. But by the critical hypothesis this is not the fault of the Iliad, for Book IX. is declared to be "late," and of the same date as late parts of the Odyssey. Achilles is not less open to moral reproach in his abominable cruelty and impiety, as shown in his sacrifice of prisoners of war and his treatment of dead Hector, in Iliad, XXIII., XXIV. But these Books also are said to be as late as the Odyssey.

The solitary "realistic" or "naturalistic" passage in Homer, with which a lover of modern "problem novels" feels happy and at home, is the story of Phoenix, about his seduction of his father's mistress at the request of his mother. What a charming situation! But that occurs in an "Odyssean" Book of the Iliad, Book IX.; and thus Odyssean seems lower, not more advanced, than Iliadic taste in morals. To be sure, the poet disapproves of all these immoralities.

In the Odyssey the hero, to the delight of Athene, lies often and freely and with glee. The Achilles of the Iliad hates a liar "like the gates of Hades"; but he says so in an "Odyssean" Book (Book IX.), so there were obviously different standards in Odyssean ethics.

As to the Odyssey being the work of "a milder age," consider the hanging of Penelope's maids and the abominable torture of Melanthius. There is no torturing in the {blank space} for the Iliad happens not to deal with treacherous thralls.

Enfin, there is no appreciable moral advance in the ODYSSEY on the moral standard of the ILIAD. It is rather the other way. Odysseus, in the ODYSSEY, tries to procure poison for his arrow-heads. The person to whom he applies is too moral to oblige him. We never learn that a hero of the Iliad would use poisoned arrows. The poet himself obviously disapproves; in both poems the poet is always on the side of morality and of the highest ethical standard of his age. The standard in both Epics is the same; in both some heroes fall short of the standard.

To return to linguistic tests, it is hard indeed to discover what Mr. Leaf's opinion of the value of linguistic tests of lateness really is. "It is on such fundamental discrepancies"—as he has found in Books IX., XVI.—"that we can depend, AND ON THESE ALONE, when we come to dissect the ILIAD ... Some critics have attempted to base their analysis on evidences from language, but I do not think they are sufficient to bear the super-structure which has been raised on them." {Footnote: Companion, p. 25.}

He goes on, still placing a low value on linguistic tests alone, to say: "It is on the broad grounds of the construction and motives of the poem, AND NOT ON ANY MERELY linguistic CONSIDERATIONS, that a decision must be sought." {Footnote: Ibid., p. x.}

But he contradicts these comfortable words when he comes to "the latest expansions," such as Books XXIII., XXIV. "The latest expansions are thoroughly in the spirit of those which precede, them ON ACCOUNT OF linguistic EVIDENCE, which definitely classes them with the ODYSSEY rather than the rest of the ILIAD." {Footnote: Iliad, vol. ii. p. xiv.}

Now as Mr. Leaf has told us that we must depend on "fundamental discrepancies," "on these alone," when we want to dissect the ILIAD; as he has told us that linguistic tests alone are "not sufficient to bear the superstructure," &c., how can we lop off two Books "only on account of linguistic evidence"? It would appear that on this point, as on others, Mr. Leaf has entirely changed his mind. But, even in the Companion (p. 388), he had amputated Book XXIV. for no "fundamental discrepancy," but because of "its close kinship to the ODYSSEY, as in the whole language of the Book."

Here, as in many other passages, if we are to account for discrepancies by the theory of multiplex authorship, we must decide that Mr. Leaf's books are the work of several critics, not of one critic only. But there is excellent evidence to prove that here we would be mistaken.

Confessedly and regretfully no grammarian, I remain unable, in face of what seem contradictory assertions about the value of linguistic tests, to ascertain what they are really worth, and what, if anything, they really prove.

Mr. Monro allows much for "the long insensible influence of Attic recitation upon the Homeric text;" ... "many Attic peculiarities may be noted" (so much so that Aristarchus thought Homer must have been an Athenian!). "The poems suffered a gradual and unsystematic because generally unconscious process of modernising, the chief agents in which were the rhapsodists" (reciters in a later democratic age), "who wandered over all parts of Greece, and were likely to be influenced by all the chief forms of literature." {Footnote: Monro, Homeric Grammar, pp 394-396. 1891}

Then, wherefore insist so much on tests of language?

Mr. Monro was not only a great grammarian; he had a keen appreciation of poetry. Thus he was conspicuously uneasy in his hypothesis, based on words and grammar, that the two last Books of the Iliad are by a late hand. After quoting Shelley's remark that, in these two Books, "Homer truly begins to be himself," Mr. Monro writes, "in face of such testimony can we say that the Book in which the climax is reached, in which the last discords of the Iliad are dissolved in chivalrous pity and regret, is not the work of the original poet, but of some Homerid or rhapsodist?"

Mr. Monro, with a struggle, finally voted for grammar, and other indications of lateness, against Shelley and against his own sense of poetry. In a letter to me of May 1905, Mr. Monro sketched a theory that Book IX. (without which he said that he deemed an Achilleis hardly possible) might be a remaniÉ representative of an earlier lay to the same general effect. Some Greek Shakespeare, then, treated an older poem on the theme of Book IX. as Shakespeare treated old plays, namely, as a canvas to work over with a master's hand. Probably Mr. Monro would not have gone so far in the case of Book XXIV., The Repentance of Achilles. He thought it in too keen contrast with the brutality of Book XXII. (obviously forgetting that in Book XXIV. Achilles is infinitely more brutal than in Book XXII.), and thought it inconsistent with the refusal of Achilles to grant burial at the prayer of the dying Hector, and with his criminal treatment of the dead body of his chivalrous enemy. But in Book XXIV. his ferocity is increased. Mr. Leaf shares Mr. Monro's view; but Mr. Leaf thinks that a Greek audience forgave Achilles, because he was doing "the will of heaven," and "fighting the great fight of Hellenism against barbarism." {Footnote: Leaf, Iliad, vol.-ii. p. 429. 1902.} But the Achzeans were not Puritans of the sixteenth century! Moreover, the Trojans are as "Hellenic" as the Achzeans. They converse, clearly, in the same language. They worship the same gods. The Achzeans cannot regard them (unless on account of the breach of truce, by no Trojan, but an ally) as the Covenanters regarded "malignants," their name for loyal cavaliers, whom they also styled "Amalekites," and treated as Samuel treated Agag. The Achaeans to whom Homer sang had none of this sanguinary Pharisaism.

Others must decide on the exact value and import of Odyssean grammar as a test of lateness, and must estimate the probable amount of time required for the development of such linguistic differences as they find in the Odyssey and Iliad. In undertaking this task they may compare the literary language of America as it was before 1860 and as it is now. The language of English literature has also been greatly modified in the last forty years, but our times are actively progressive in many directions; linguistic variations might arise more slowly in the Greece of the Epics. We have already shown, in the more appropriate instance of the Chancun de Willame, that considerable varieties in diction and metre occur in a single MS. of that poem, a MS. written probably within less than a century of the date of the poem's composition.

We can also trace, in remaniements of the Chanson DE ROLAND, comparatively rapid and quite revolutionary variations from the oldest—the Oxford—manuscript. Rhyme is substituted for assonance; the process entails frequent modernisations, and yet the basis of thirteenth-century texts continues to be the version of the eleventh century. It may be worth the while of scholars to consider these parallels carefully, as regards the language and prosody of the Odyssean Books of the Iliad, and to ask themselves whether the processes of alteration in the course of transmission, which we know to have occurred in the history of the Old French, may not also have affected the ILIAD, though why the effect is mainly confined to four Books remains a puzzle. It is enough for us to have shown that if Odyssean varies from Iliadic language, in all other respects the two poems bear the marks of the same age. Meanwhile, a Homeric scholar so eminent as Mr. T. W. Allen, says that "the linguistic attack upon their age" (that of the Homeric poems) "may be said to have at last definitely failed, and archaeology has erected an apparently indestructible buttress for their defence." {Footnote: Classical Review, May 1906, p. 194.}