Home >> Encyclopedia-britannica-volume-11-part-2-gunnery-hydroxylamine >> Hirschberg to Homicide >> Homer



HOMER ("Qs pos), the traditional epic poet of Greece. The name is especially applied to the author of the two great epic poems, the Iliad and Odyssey (rl 'IXtas 7rorlacs, rl rrano•Ls), which were recited or performed by rhapsodes at the Panathenaea at Athens, every four years during the classical period. An analysis of their contents will be found under the title HOMERIC POEMS, together with an account of the rest of the Homeric cycle.

The great books which come from an age before literary record are sometimes anonymous, like the Eddas, sometimes attributed to famous figures in the tradition, as the Psalms in mass to David and the Pentateuch to Moses, sometimes to a def inite but unknown person, as the Chanson de Roland to Thorold, sometimes to gods or mythical characters, like the poems in Book X. of the Rigveda. The special difficulty about Homer is that, whereas David and Moses have an independent existence, whether or no they wrote the works ascribed to them, Homer has not : he is nothing but the author of the Homeric Poems. The poems are facts and "Homer" a hypothesis to account for them.

Lives of "Homer."—There are indeed certain traditional "Lives of Homer." Eight are published in the Oxford text of Homer, vol. 5, together with the Contest of Homer and Hesiod. The longest and best is composed in Ionic Greek under the name of Herodotus. The dates of these lives are hard to determine, particularly since the latest may contain nuclei of ancient tradi tion. But it seems likely that the demand for a life of Homer arose at least as early as the general interest in literary biography in Greece, i.e., about the time of Aristotle; and that, when it arose, there was no material for satisfying it beyond romance and conjecture. The existing lives seem to be made up out of deduc tions from certain passages in the Homeric poems and from proverbs and popular verses. The Lives quote many of these, sometimes lines about particular places—Smyrna and Kyme (Epigr. iv.), Erythrae (Epigr. vi., vii.), Mt. Ida (x.), Neon Teichos (i.), sometimes about particular trades or occupations—potters, sailors, fishermen, goatherds, etc. The verses are attributed to Homer, and occasions are invented for his uttering them. Simi larly various characters in the epics, Mentes, Mentor, Tychius etc., are explained as real persons whom he rewarded or punished by putting them into his books. There is romance in the concep tion of the divine poet, blind, old and poor, wandering from city to city; and it may well be real history which, amid the many varying stories, always brings Homer from Ionia, and almost always from that part of Ionia which was originally Aeolic.

Seven cities are recorded as claiming to be Homer's birthplace: "Smyrna, Chios, Colophon, Salamis, Rhodos, Argos, Athenae," but it is not always the same seven. Kyme, Ithaca, Ios, Pylos come in some lists, while Sparta, Egypt and Babylon also com pete. Many of these fall away as obvious inventions; Sparta comes from the Lycurgus romance; Egypt and Babylon belong to the conception of Homer as "all-wise," full of the learning of the east. Smyrna and Chios have the most support, followed by Kyme and Colophon.

As to Homer's date, the pseudo-Herodotean Life puts his birth at I102 B.C. Elsewhere the estimates vary from 685 B.C. (the historian Theopompus) to I159 B.C. (certain authorities in Philostratus). Herodotus considers that "Homer and Hesiod" made the Greek pantheon "not more than 400 years before me": i.e., about 830. Aristarchus puts his floruit at 1044 B.C. These dates seem to be based on the date of the Trojan War and a conjecture of Homer's relation to it.

Sometimes one can place a poet's date and nationality by his dialect. But the Homeric dialect has no contemporary parallel by which it can be judged, and as it comes to us in the poems is clearly not a dialect that was ever spoken, but one created by the epic tradition, and even by the hexameter metre, for its own purposes. In the main it is an Ionic dialect, like an early stage of that written by Herodotus; but the surface of the poems has been unconsciously Atticized, an inevitable result of the Pan athenaic recitation and the Athenian book trade. When this surface corruption is removed we find many non-Ionic forms which have all the appearance of being Aeolic—the dialect of Thessaly, Lesbos and Aeolis. It is also clear that, where two forms are metrically equal, the Ionic form is generally preferred.

This would seem, by all analogy, to show that'either the poem itself or at any rate the poetry which created the Epic dialect, had once been Aeolic, just as the people of Smyrna, Chios and the neighbourhood had. (The phenomenon is not uncommon in traditional poetry. Thus the English poem, Sir Degrevant, is shown to be taken over from a Norman original by the fact that, while the hero is normally called Degrevaunt, and of course re mains so when required to rhyme with "avenaunt," "recreaunt," he becomes "Degrivauns" when he has to rhyme with "counte nauns" and "Frauns." In Norman, "Degrevans" was nominative and "Degrevant" accusative.) But beneath both the Ionic main stream and the Aeolic inci dents there is an element of very old Greek, comprising many expressions which were not understood in classical or Alex andrian times, and many more which needed a commentary (aµ€vi va K6p7va, &aKTOpOS OVTfs). What is more remark able, there are some forms (such as FavaE, KEpaµos) which are found in Cyprian and Arcadian, and nowhere else. That is, they belong to the ancient undivided language which survived in these two isolated and mutually remote dialects. The problem may be raised whether the Homeric dialect does not contain an element of the pre-Greek language of the Aegean, derived per haps from pre-Homeric Minoan poetry. Doubtless there is such an element but it does not seem noticeably stronger in Homeric than in classical Greek (cf. Haber, De Lingua antiquissimorum Graeciae Incolarum, The Text of the Poems.—These observations, which seem at first sight so confusing, nevertheless yield a result. The man "Homer" cannot have lived in six different centuries nor been born in seven different cities; but Homeric poetry may well have done so. The man cannot have spoken this strange composite epic language, but the poetry could and did. Let us turn back therefore from the unknown man to the known book.

The works of Homer "and no other poet" (Lycurg. in Leocr. p. 209) were recited at the Panathenaea in accordance with a law attributed variously to "our ancestors," Solon, Hipparchus or Peisistratus. The "works" were the Iliad and Odyssey. They had to be recited E or E inroXrt/iews, i.e., by giving or taking a cue; there was a fixed order, so that where one man left off another began. The custom continued at least to.the times of the author of the pseudo-Platonic dialogue Hipparchus, and seems to have started with the foundation or re-foundation of the Panathenaea by Peisistratus.

If there was a fixed order, there must have been something like a written text; and as a matter of fact we find in the scholia an accepted tradition, supported by certain corruptions in the text, that the poems were once written in the old Attic alphabet, an archaic and awkward script with no double letters or long vowels. This tradition has been considered incredible on the grounds that the new, or Ionic, alphabet was used in Athens for literary purposes from very early times, the old script being purely for official documents; that Homer must have come to Athens in Ionic script; and finally that several passages in tragedy which describe the writing of heroic times always take the Ionic script for granted. (Eurip. Theseus, fr. 382 N.; Agatho fr. 4, Achaeus fr. 33, Theodectes fr. 6.) The conclusion seems to be that, if the Panathenaic text of the Iliad and Odyssey was ever written in the old Attic script, it must have been an official document, drawn up by order of the state.

There is some evidence that this was so. A tradition which Wolf described as vox totius antiquitatis speaks of a text or an arrangement, or recension, of the works of Homer by Peisistratus, tyrant of Athens between 56o and 527 B.C. The oldest witness is one Dieuchidas of Megara in the fourth century B.C., to whom may perhaps be added the Aristotelian Dicaearchus and the grammarian Crates, who founded the art of grammatike in Rome. (He was detained there by breaking his leg when on an embassy from Attalus II. to the Senate.) The clearest are writers of the early Roman period and various scholia and lexica. Cicero puts it that Peisistratus primus Homeri libros, confusos antea, sic disposuisse dicitur ut nunc habcmus (is reported to have been the first to arrange in their present order the books of Homer, which were previously in confusion) ; Josephus says that Homer left nothing in writing; the poems were afterwards put together from the varying memory of different places; Suidas, the Lives, Eustathius and the Townley Scholia explain that Homer sang his poems airopiu5i v, "in a scattered way," as he travelled about, and that Peisistratus afterwards collected them. Pausanias, Aelian, Diogenes Laertius, mention or assume the tradition; indeed one may say that it is the accepted view of the gram marians with the exception of Aristarchus, who held that Homer himself was an Athenian.

Such authority is of course not conclusive. But there is very little of our grammatical record that has more ancient credentials, and even if there were no tradition at a11, it is hard to conceive how the Panathenaic recitation could have been established with out some editing of the poems, under the orders, presumably, of Peisistratus or his artistic son, Hipparchus.

There must then have been an authoritative Athenian text in the second half of the sixth century B.C. That is the earliest solid fact in the record of the Homeric poems. The curious thing is that this text did not maintain itself. There is little evidence affecting the text of Homer in the fifth century, after the fall of the Peisistratidae, but as soon as quotation becomes common, in the fourth and third centuries, it is clear that the current texts varied considerably both from one another and from our vulgate. Still more striking is the evidence of the early papyri, numbering about 200 which have been discovered since 1890. Eleven of these are earlier than 150 B.C.; the rest later. Roughly speaking those later than 15o ac. generally agree with our present vulgate texts; of the eleven supposed to be earlier ten differ "wildly" from our vulgate and one (Pap. Ox. 1388), has proved to be later.

The Establishment of the Text.

Evidently something hap pened about 15o B.C. which reduced the current texts of Homer from disorder into order, and the grammatical tradition enables us to see what it was. It was about 15o B.C. that the famous Aristarchus of Samothrace, then head of the Alexandrian library, published the edition, or rather the two successive editions, of the text of Homer which were afterwards recognized as authoritative. This fact was for long obscured by the language of the scholia, which dwell on Aristarchus's extreme "caution" i'ealt eta) in treating the text and often seem to be in doubt what his reading really was. When he does not "condemn" a spurious line, that normally means that it had not yet made its way into the text. But it is clear that his caution did not prevent conjectures and large condemnations; and the occasional un certainty as to his reading only shows that the current texts a few generations afterwards were not infallible and could not be trusted without further question. It seems probable, especially after the recent work of Cauer, Wecklein and Bolling, that the present vulgate text was approximately, at any rate, the text of Aris tarchus. It is still disputed how far he created that text, and whether it is really some old text—that of Peisistratus, or even of Homer himself—re-established by the great critic out of the mss. which he collected.

It seems clear that from the time of Aristotle and Aeschines on to that of the earliest papyri the texts of Homer were in an extremely fluid state. (Aristotle for example cites from Homer five completely unknown phrases or passages; he quotes many passages in a strange shape; he found several passages not where we have them but in another context; and in order to show how short a resume can be he remarks that Odysseus' story to Penel ope in Od. xxiii. "occupies only 6o lines," while in our texts it only occupies thirty-three.) The first great Alexandrian scholar, Zenodotus of Ephesus (b. 325) set himself vigorously to cope with this confusion. He rejected freely passages which he thought spurious, and probably he found mss. which were considerably shorter than ours. He had not reached the wonderful knowledge of Homeric idiom which was the distinction of Aristarchus; but he had the advantage of catching the text at an earlier stage, and thus was spared some interpolations which had not yet been made, while he preserved many ancient readings and forms (47) tthilara B 144, 3a ra A 5, KvvvAayµov 43 575, &qi aird 'IwK,icov y 307). His text must have been at least a thousand lines shorter than that of Aristarchus, and it was his vigorous pioneer work which made the "caution" of Aristarchus possible (see infra, "Study of Homer in Antiquity").

The Poems Before Aristarchus.

We find thus two stages of something like terra firma in the history of the Homeric text: the present vulgate, established (approximately) by a process which set in about 150 B.C. and was probably due to Aristarchus. Before that a long period of uncertainty, and then earlier, about 550 B.C., the so-called "Peisistratean recension," by which we can only mean the text in the old Attic script which was composed or used for the purposes of the Panathenaic recitation. This is the earliest stage of which we have any record; any suggestion about the previous history of the poems must rest on conjecture, and conjecture supported by little external evidence. Yet it is clear that the year 55o B.C. is an extremely late stage in the history of the Homeric poems, or at least of Homeric poetry. As in almost all departments of Greek literature only the latest growth was preserved; the last stage of the epic like the last of choric poetry, and the last of tragedy. We can at best try by different roads to pierce into this darkness.

I. Were the poems recited elsewhere before they came to Athens? It would seem at first sight probable. There certainly was a good deal of epic recitation; but of the Iliad and Odyssey no public recitation is recorded, and it is noteworthy that when Zenodotus and Aristarchus ransacked the Greek world for mss. they seem never to have lit on any fragment of a pre-Peisistratean ms. No Ionic Iliad or Odyssey seems to have been in existence in the third century B.C. This seems to show either that the poems had not been committed to writing before they came to Athens, or else that the poems written in Ionic, and recited for many generations at Ionian gatherings, were in some way completely outshone and driven out of the market by the Attic poems. At any rate there is no trace of an earlier written text of the Iliad or Odyssey as we know them.

2. For what occasion can the poems have been intended? As epics for reading they are wonderful, yet it seems certain that there was no large reading public in Greece in the sixth century B.e., much less in that far more remote period when Homeric poetry first took shape. As lays for ordinary recitation they are not suitable. A recitation ought not to last more than an hour or two, but the Iliad would last about twenty hours. And the poems are decidedly unities; they are not strings of separate lays or cantos. They absolutely refuse to fall apart into separate lays, as is proved by the subsequent history of Lachmann's famous attempt to make them do so. There were rhapsodes reciting Homeric poetry all over Greece in the sixth century; we hear of them at Sicyon, Sparta and Syracuse. But they can hardly have been reciting the Iliad and Odyssey. Those gigantic wholes must have been meant for some very great and rare oc casion, such, for example, as the Panathenaea—the solemn gathering of all the Ionian cities to their Metropolis once every four years. If the poems were first written down in the time of Peisistratus it may well be that they first received their present form at that time. Indeed such would be the natural conclusion. For the rhapsode, at each performance that he gave, regularly drew both upon his memory and his invention. He knew the style, the language and the facts, and could begin his recitation from any point he chose, as bards in the Odyssey do. (7-Cop &µ69€v ye, Od. i. Jo. iv9€v Od. viii. 50o.) For the new Pan-Athenian festival which he was founding Peisistratus would naturally have a special version or arrangement made.

3. Thus we cannot be certain that the poems in their present form existed before the sixth century, but we can be quite cer tain that everything but the form is vastly older. By form is here meant the selection and arrangement of the incidents. The subject matter, the language used in describing it, the poetical style, and above all the manners and customs of the society depicted belong to an age which can in part be dated and in part reaches back to the dimmest antiquity.

Internal Evidence.

In facing this problem the first thing to realise is that the poet is not naďvely describing the world as he himself knew it, but, like almost all narrative poets after him, describing with deliberate artistic purpose a long past age which he knew from earlier poets and from tradition. This is shown not merely by the occasional contrast between the heroes and "men as they now are" (oioc vvv f3porol €iac, cf Il. xii., 14 sqq. xx., 302 sqq.), but much more by the fact that a number of cus toms are mentioned in metaphors and similes which are carefully excluded from the narrative itself.

Thus in the narrative there are no trumpets, no garlands, no eating of fish (except by starving men, Od. xii. 331, iv. 368), fowl, vegetables, milk or boiled meat; but most of these things were evidently known to the poet and occur in similes or descrip tions. Paintings are not mentioned, nor yet coins; a statue, and a sitting statue, is once implied but not directly mentioned (Il. vi. 303); the art of writing is left doubtful, but any direct mention of it is skilfully avoided and the scholiasts left to take divergent views (Il. vi. i68, vii. 175 with schol.). Cavalry is never mentioned (Od. xviii. 263) though riding is known: the use of cavalry was extremely old, but the chariot was more "heroic." In a similar archaizing spirit there is no mention of colonies, nor of the great "modern" Ionian trading cities like Miletus, just as there are no Dorians; yet the mask slips, for we hear of an old Sicilian woman (Od. xxiv. 211, 366, 389) possibly of a colony (Od. x.io8 ?Ar takia), and a good deal about certain Heracleidae, who were the Dorian chieftains (Il. v., 628, ii. 653) and still more about their ancestor Heracles.

There are abundant "arts and crafts": a famous cup, brooches, belts, necklaces of amber and gold, stained ivory, ivory reins, and a shield from the divine smithy which surpasses the handi work of man, but is doubtless modelled on it. As to food, the scholiasts remark that "the heroes" took three meals a day, and each meal consisted of the roast flesh of the larger quadrupeds. The recurrent phrase "unspeakable flesh and sweet strong wine" (Kpia liar-era Kai µE6v 7)515) betrays the admiration felt by a later generation for such heroic tastes. The dead are burned. The bale-fires make a vivid stain in the scenery of the epic. The Greeks had always the custom of burial, and had mostly not enough wood for pyres. The weapons are made of bronze (XaAKbs).

Homer like Hesiod knows of a time when "black iron was not," except indeed as a material for tools, clubs, arrowheads and other easily workable objects. He knows, as Aristotle knew, that the classical Greek custom of the marriage-dowry was recent, and that in the old days "they bought their women" by giving bride gifts (Ebva). It is wonderful how clearly these last three customs are realized, and how carefully the illusion is kept up. There is a slip in the use of a late proverb "iron of itself draws a man on" (Od. xvi. 294, xix. 13) where "iron" is used for "a weapon"; and once or twice the word ibva is used as if it meant "dowry" (Od. i. 277, ii. 53,196). But in the main the picture of the heroic age is wonderfully well maintained. The schools of bards must have had a fine training in the tradition, though we must re member also that our present text has passed under the eyes of many jealous critics before its final revision by Aristarchus.

For the fact is that the Scholia, or remains of ancient com mentaries, with which the Iliad is richly provided and the Odys sey not deficient, are more occupied with the distinction of the Homeric from the non-Homeric (or "more recent") than with any other problem. Aristarchus and his disciples have lynx eyes for subtle points of language, belief and custom. It is no exag geration to say that modern scholarship had no conception of many subtle uses of words in Homer until the Scholia to the ms. called Venetus A were published by Villoison in 1788. (E.g., that 46(3os is "flight," bins "fear," Tp€Iv is "retire," not "to tremble," 90aac "to cense," not "to slaughter"; that 4p/s w is "to point out," not "to say," cr a is "a dead body"; axibov "near," not "almost," 7raXcv "backwards," not "again," TaXa "quickly," not "perhaps," ypa4ecv "to scratch," not "write".) And the ob servation of customs was almost equally acute. The explanation seems to be clear. It is not to be supposed that either Peisistratus or Aristarchus had access to a flawless ms. many centuries old by which variations could be checked; but there were traditional texts and a traditional knowledge, both in the schools of bards and to a less extent among the critical public, by which all varia tions or new compositions could be tested and kept true to stand ard. It is most instructive to see how those remains of ancient epic which were not selected for the great recitation and thus fell into neglect break all the Homeric rules in language, verse and custom alike, while on the other hand they never approach, as many passages in Homer do, to the lyrical hexameters char acteristic of the fifth century.

Study of Homer in Antiquity.

The critical study of Homer, that is, of the mass of epic literature concerned with the heroic saga, began in Greece, as Monro says, "almost with the begin ning of prose writing." The first names recorded are those of Theagenes of Rhegium (6th century B.c.) and his pupil Metro dorus of Lampsacus. In the fifth century educated men from the impoverished islands and cities of Ionia sought a living in Athens by teaching literature and especially Homer. A typical instance is the adventurer Stesimbrotus of Thasos. From this time on ward the study of Homer was a regular feature of Greek mousike.

It is noteworthy that, from the very first, exegesis took two forms: explanation of the glossae, or hard words, and allegorical interpretation and justification of the subject matter. The prin ciple set out in the Homerica Problemata of Heracleitus (sat century A.D.) "If Homer used no allegory he committed all impieties" (el µrA&v i7XXnybpnaev iravra cr f3 ev) was, as one can see from Plato, accepted early, and regarded as a natural thing. On this system Hera was the Air (rip-a= a-'qp), the adultery of Ares and Aphrodite was the combination in the world of Strife and Live, and so on.

The interpretation of glosses is mentioned in Aristophanes, Daitales frag. 222, where a father asks his son the meaning of Kbpvµ43a and aµEVflva Kappva. Many strange interpretations, recorded in the Scholia as coming from "the glossographi" serve to illustrate these painful beginnings of the science of gramma tike. The phrase vEKpovs Karareevnwras ("the dead corpses") was objected to as tautologous. "No," said the glossographi, N€Kpoc=s4pc, "young men," with Kp "added for euphony." Toios was interpreted as aya86s, a meaning of course which it some times implies (Toiov 'yap Kai erarpbs Od. iv. 206, and even rollo 7Epovros as ayaeov yipovros Il. xxiv. 164). It is noteworthy that Aeschylus actually uses robs as "good" (Suppl. 400), and so, it is said, did Callimachus.

Many individuals are recorded as having made a text of Homer. Probably it was a common thing for students and poets, such as Antimachus and Philetas, instead of buying a book from a shop, to write out their own text. We hear that Plato's library passed on eventually to the library at Alexandria, which may account for the close resemblance of the Homeric quotations in Plato to our vulgate. Aristotle makes many interesting criticisms on the poems from the aesthetic point of view, but seems to have had a text widely differing from ours. Among fourth-century critics was Zoilus `Oµppoµaarc , who shocked public opinion by his attacks on Homer.

The first approach to a really scientific treatment of the poems was made by the great scholars of Alexandria, Zenodotus (d. 26o? B.c.), Aristophanes of Byzantium (about 262-185) and Aristarchus of Samothrace, who resigned his post as head of the Alexandrian library in 137 B.C. The work of Zenodotus is still something hard to appreciate. We can see that the texts of Homer in his time fluctuated wildly and in order to establish a fairly "correct" text he had to cut his way boldly through a jungle. His rejections, as recorded in the scholia, are very ex tensive; his Iliad must have been very much shorter than the present vulgate. He frequently rejected passages for "unseem liness," bca Td airpe nos (see below, "Expurgation") . We often hear of lines which Aristarchus or some later scholar doubted, but Zenodotus dypa4ev. This used to be explained as a sign of his recklessness; but, as Professor Bolling of Ohio has pointed out, there has been, as far as our record reaches, a constant tendency to add lines to the Homeric text, and probably the statement that Zenodotus "did not even write" a line means that it did not exist in his day.

Zenodotus had nothing like Aristarchus's delicate and intimate knowledge of Homeric idiom: he thought that the dual (which had disappeared by his day) was merely another form of the plural; he was capable of such a metrical solecism as irpocAaveiaas ibwv (Il. viii. 378). On the other hand he seems often to have pre served ancient forms and readings which were rejected after wards by Aristarchus (e.k acrd cFwic cov Od. iii. 307; 4)7)=6s 11. ii. xiv• 499; fee KvvvXavyµdv ?uovap Ii. xxi. 575).

Our knowledge of Aristarchus comes chiefly from the rich scholia to the Venetus A, first published by Villoison in 1788. They quote largely from four scholars Aristonicus (Augustan era), Didymus (who wrote against Cicero De Republica), Hero dian and Nicanor (2nd century A.D.). They are sometimes in doubt about Aristarchus's views, and evidently possessed no ms. which could be held as authoritative. We learn from them of his aiµeia or signs, which frequently occur in the Venetus A itself. One would have expected, considering the very imperfect state of Greek ypaµµara at the time, with no system of punctuation or even of dividing word from word, that any new "signs" would have been devoted to supplying this want; but Aristarchus's signs deal chiefly with the genuineness or spuriousness of partic ular lines. There was also a sign (diple, like a V on its side) to give the reference to a note; and another to call attention to mistakes of Zenodotus (diple with dots).

His critical material comprised many mss. Kara

rr6Aecs and several Kar' avbpa. The latter were the work of individual scholars, the former apparently those used by certain cities for the public official recitations. He never mentions the Athenian recension, evidently because he assumes it as his basic text. He speaks of the "better" or "more careful" texts, and the "com mon," "ordinary," and of one as ) 'iroXvorLxos ("of many lines"). He evidently had no access to any text or fragment of text earlier than the Athenian recension, and shows no sign of having any which went back to the sixth or even the fifth cen tury. With this imperfect material and the help of Zenodotus' pioneer work Aristarchus produced a text which immediately won recognition as the classical recension of Homer in antiquity and was probably, though the point is still disputed, the founda tion of our present vulgate. Certainly, of our existing papyri, those written before the time of Aristarchus (15o B.e.) differ widely, while those later are almost uniform.

Aristarchus did even more remarkable work in exegesis. He not only corrected the errors of the glossographi in explaining the obsolete words, he observed with great acuteness the peculiar Homeric uses of common words (see above). He distinguished dual and plural; he saw that in Homer o TO was a pronoun, and he rejected the theories of Homer's allegory of universal knowledge. On the other hand his grammatical explanations often betray the infancy of the science (the first Greek grammar was the work of his disciple Dionysius Thrax). For example he uses freely the theory of "enallage" or "change of case": `Iirirora is "the vocative instead of the nominative"; in Toy 8i CTKOTOS KaXvifiEv, rev is "instead of Tov." The Heroic Age.—The poems, even as they stand now, ap proximately in the form fixed for them by Aristarchus, nearly a thousand years later than the events they profess to describe, have an almost entirely consistent style and phraseology and give an almost entirely consistent description of the age about which they write. Of course "consistent" is by no means the same as "true." Yet, though we have practically no direct information about the Homeric age, we can, by analogy from other literatures better supplied with contemporary history, understand its general character and see that the description seems to be based on fact. The poets described a society which they did not themselves know and could not well have invented, but which corresponds closely to a type known to have existed elsewhere under given conditions. It was easy, of course, for a romantic Greek bard to make his ancient heroes ignorant of iron weapons and the art of writing, or to imagine an idyllic princess washing the royal clothes in the river.

But there are other characteristics of the age which can hardly be inventions. The heroes of Homer belong to a Wanderzeit. The kings had, in general, no national territories and no strong ties of kindred; they depended on their "companions" (Eraapot, comites) or personal followers, who were largely adventurers like themselves. They are untroubled by the sanctities and duties attaching to the tribe or to the buried tribal ancestors. They practise no arts but song and war. They carry their wealth about with them in the form of brooches, belts, rich armour, necklets and caparisons. They burn their dead, they build no temples but worship at altars in the open air: and they worship a collection of gods who are just like themselves, gods who feast and sing and conquer and utter judgments, but who never work, have never created a world—only conquered and divided it—and who are to an extraordinary degree rootless and international. They stand or fall by their personal qualities; their strength and courage, their faithfulness and military honour, and the fact that since they fear no one they never tell a lie.

The above picture needs qualification if we take account of details; e.g. the Odyssey is a romance of the return of the wanderer, and implies the existence of a real home; the catalogue of ships in Iliad ii. gives geographical kingdoms to all the heroes. But in the main it is true, and it corresponds to the type of society that we find in many parts of the world under particular historical conditions. It occurs where an old and rich civilization is in process of being broken up by barbarian conquerors. It is the type called by Professor Chadwick a "heroic age," and can be traced in northern Europe during the break-up of the Roman empire, in the Balkans during that of the Greek empire, and in other parts of the world that have produced heroic poetry. The description of Attila's camp by the Byzantine historian Priscus is curiously Homeric in its details. (See Gibbon, ch. xxxiv.) Discoveries.—This discovery has changed the character of the problem of the Homeric age. We saw above that the poets were consciously describing an age long past and notably different from their own. We now see that they were not merely describing a sort of "youth of the world," remarkable chiefly for a rude idyllic simplicity in which the contrast is between "Homer" and of vECWTEpOG. Nor yet were they, as Schliemann's great dis coveries first suggested, depicting the great age of Troy, Mycenae and Cnossus. It was a particular period in history, to be con trasted both with the rich decadent civilization that went before and the peculiar classic Hellenism which came after.

The first palaces of Crete must have been built about 200o B.C., and not long after there are signs of that movement of the Aryan tribes which was to have such incalculable effects on human his tory. The second town of Troy was laid under seven feet of ashes; at Orchomenos, Corinth and Argos citadels were destroyed and rebuilt, apparently by Northern conquerors. The northern "megaron," or hall with a porch and central hearth, begins to be found in Greece. Hitherto the "Kefti" or Men of the Isles have been in close connection with Egypt, but about 17oo, when the Hyksos dynasty largely cut that country off from civilized commerce, the Cretan cities begin to turn more to Greece. Argos learns to grow the vine and olive; it adopts the Cretan woman's dress and the Cretan frescoes, while it still keeps its northern beard and tunic and megaron. A century or so later perhaps the Aryan invaders, or Achaioi(?), take to the sea. They had first called the sea by an "Aegean" or pre-Hellenic word, 46.Xao.va; then they called it &As, "the salt"; it may be significant that at last they used the name irbvros, a good Greek word meaning "path." It was a path that led both to Crete and to Egypt. About 145o, Phaestus and other Cretan palaces were destroyed. For some 5o years Cnossus ruled alone: then comes the destruction of Cnossus itself—a sudden attack, it would seem, in the midst of some great court ceremony—and the end of the Minoan empires. After about 1400 B.C. only Mycenean work is found in Egypt, not Cretan.

Achaioi from Mycenae seem to take the vacant place. In Crete the royal tomb becomes a common pit; portraits of chief tains wear the northern beard and moustache. New towns are called by Achaean names; there is a decline in art and a cessation of wall-painting. It is significant, too, that after this date there is no further mention of the Kefti in Egyptian records.

Civilization has gone back but is by no means destroyed. Writing remains and there are signs of abundant wealth and mass production for commercial purposes. Speculative historians have fancied that in the Tell-el-Amarna letters and other records of this period they find some of the great names of Greek saga, an Alaksandus of Uilusa who reminds them of Alexandros of Ilion, an Attarissyas of Ahhia or Ahhijava who sounds remotely like Atreus the Achaean. With much more probability we begin now to identify names of various peoples, or at least hordes of men, who are known to Greek tradition.

At the battle of Kadesh in the Hittites had in their motley host Iliouna, Dardanoui, Masa, Pedasa, Kerkisha, Loukki and Danaouna. If the terminations -na and -sha denote "peoples" or "hosts," as Egyptologists suggest, we can hardly be wrong in recognising at least the people of Ilion, the Dardanoi, Lukioi and Danaoi. About 1230 and 1225 Merneptah of Egypt defeats in the Delta "uncircumcised Akhaiusha" "from the sea-lands," "who fight to fill their belly daily." With them are Toursha, Shardina, Loukki and Shakalasha, wanderers who may have given their names afterwards to the Tyrseni, the Sardinians, Lycia and Sagalassus. About 1194 Rameses defeated a similar locust-like multitude advancing "by sea and land, with women and children in ox-carts." They had destroyed the Hittite empire and "no people had stood before them." Troy.—Most of the cities they destroyed have perished without record, but Greek tradition tells us that the great sixth city of Troy, Homer's Troy, was sacked about this date. It was sacked, after a long siege perhaps, by a host that may be called indifferently Akhaiu-sha or Danaou-na—Homer has also a third name, Argeioi—and who are led by a "king of men" who is "lord of many islands" and has doubtless gathered in his thou sand or twelve-hundred ships a great force of "the peoples of the sea." (Greek legend also makes Agamemnon and Menelaus joint kings of Sparta, or one of Mycenae and one of Sparta; it also involves them in some ritual king myths; but it is perhaps not rash to detect an element of real history in the Homeric picture.) The fall of Troy seems to be the sign or cause of the final crash. About 1200 the infiltration of further West-Greek tribes becomes a regular invasion. The pirates, or "peoples of the sea" receive an increase both from new "Dorian" invaders and from the broken men of the old Minoan or Achaian kingdoms, but by now there is little rich plunder left. It is a time of ovveXeis rrpare6a6 Kai µeravao rfoeas, "constant warfarings and uproot ings" (Diodorus v. 8o: cf. Thuc. i. 2) .

Homer's narrative of the siege of Troy is doubtless related to the real siege much as the Norse or German or French poems are related to the actual brutalities of the age they describe. The facts have been idealised and transfigured by memory, and confused with abundant myth, folklore and fiction; but some of the truth can be descried, as in a palimpsest, beneath the poetry. We can see in the first book of the Iliad a plague-stricken army, or mob of mixed peoples under diverse leaders, pinned to the barren sea shore, the narrow space choked with dead dogs and mules, the piles of burning corpses, the best fighting force in mutiny be cause of a quarrel about a captured woman; and the various bands "fighting to fill their bellies daily" by raids on the ex hausted neighbourhood. At the end of Book VII., on the arrival of a wine-ship, we see the soldiers selling all they have, bronze, iron, shields, cattle and captives, for liquor and lying all night long on the beach under a thunderous sky.

We can also see, not from direct statements in the Homeric poems, but from implications there and direct statements in the rest of the poetic tradition, that the fall of Troy was the begin ning of a dark age. If Troy fell the rest of Achaian Greece fell with it, including Thebes, Argos and Mycenae. The great kingly houses disappear. The epos knows little of the after history of the Homeric heroes except a disastrous series of so called "returns"; the conception belongs to a later age which conceived these wandering hordes as the armies of regular Greek cities. The sons of Agamemnon, Achilles and Ajax and Odysseus are just remembered, but after them there is darkness.

We are justified therefore in recognising the siege and des truction of the sixth city of Troy, about the year i 200 B.c., as forming the historical nucleus of the poems, just as the defeat of Charlemagne's rearguard by the Basques in the valley of Roncesveaux is the historical nucleus of the Chanson de Roland. The real Hrodland, count of the march of Britanny was killed in that battle on Aug. our existing Chanson de Roland was written soon after 1o66, three hundred years later. But we know that songs or lays on the subject were soon composed: they are mentioned as widely known (vulgata) in a book written less than 6o years after the battle (Life of Louis I., in Pertz SS. ii. 6o8). And an analysis of the existing Chanson shows, beyond question, that it is a re-treatment (remaniement) in large part built up out of previous poetic treatments of the same subject.

Using this comparison as it stands, we may observe that the gap between the fall of Troy and the earliest text of Homer known to us is much greater than 30o years: our present text probably dates from Aristarchus about 150 B.C., and even if we assume that that is in the main identical with the text of Peisis tratus we only get back to the latter part of the sixth century B.C. and Troy fell at the beginning of the twelfth. There must have been, during those centuries of oral transmission, uncon trolled by any fixed record or any learned class, much reshaping of the poetry and much transformation of the historical facts.

Date of Composition.

Innumerable attempts have been made to fix some time and place at which "Homer" may have lived, or at which the Iliad or the Odyssey or both, in more or less their present form, may have been composed. All have failed, and it is well to consider the reasons for their failure.

In the first place, we do not yet know what question to ask, and until we do we can hardly expect the right answer. Yet, from the nature of the case, to know the right question implies an understanding of the whole problem which is at present beyond us.

Next, we have no contemporary record to guide us. Even the political history of Greece between '200 and 600 B.C. is extremely scanty and unreliable; and the history of literature before the time of Peisistratus is practically non-existent. Callinus in the early seventh century is said to have mentioned the Thebais, an epic now lost, as the work of Homer. Herodotus (v. 67) tells us that, in the early sixth century, Cleisthenes the tyrant of Sicyon put a stop to the contests of rhapsodes because the Homeric poems "glorified Argos and the Argives." This shows that there were recitations of Homeric poetry in Sicyon before the time of Peisistratus, but nothing of what that poetry was. Our extant Iliad and Odyssey do not glorify the Argives in any special sense, but they do use the word "Argives" for the Greeks as a whole, which may have been enough to disturb Cleisthenes. There is a general reference by Xenophanes, an early contemporary of Peisistratus, to the immoral stories about the gods in Homer and Hesiod. There is no other pre-Peisistra tean evidence. It is indeed conjectured that the Simonides who quotes a line of the Iliad (vi. i 26) as the work of "a man of Chios" was perhaps the old seventh-century poet from Amorgos; but both Plutarch and Stobaeus, who are our sources for the informa tion, clearly meant the famous Simonides of Ceos, who belongs to the fifth century. One would like to know whether by "a man of Chios" he really meant the traditional Homer or was merely referring to some contemporary Homeric rhapsode, such as Cynaethus of Chios.

The Homeric Tradition.

The other early references chiefly use the name "Homer" as covering the whole heroic or epic tradition. Aeschylus is reported to have described his plays as "slices from the great banquets of Homer"—which is true if "Homer" means the heroic tradition, markedly untrue if it means the Iliad and Odyssey. Simonides refers to Homer as telling how "Meleager surpassed all the youths in spear-throwing across the wild Anauros"—which is not in our Homer: Pindar (Nem. vii.) says that Homer praised Odysseus—which may mean, but does not by any means prove, that Pindar read the Odyssey as we have it. He makes two other references to Homer, neither of them quite irreconcilable with our poems but certainly not sufficing to prove that he knew them. A late schoolmaster, Theodorus (see infra), made a sort of wall-picture which he called TiL cv —an arrangement of the events of the whole early heroic tradi tion, in which the Iliad and Odyssey form only a fragment. An "arrangement of Homer" of a different kind was needed for the great Panathenaic recitation, and took the form—so we have definite evidence—of our present Iliad and Odyssey.

The effect of the recitation and the published text which seems to have accompanied it was quickly seen. By the latter part of the fifth century, especially among Attic writers, the Iliad and Odys sey are well-known and often mentioned : they are accepted as the only canonical works of Homer, though other poems—epics, hymns, epigrams and the like—are occasionally mentioned. Herodotus for instance is doubtful about the Epigoni and is against the authenticity of the Cypria because in his time it differed from the Iliad in its account of the travels of Paris. (It was altered later and made to agree: see Proclus' epitome.) Aristotle treated the comic Margites as genuine. It may be noted, also, that Attic tragedy, which drew its material from the heroic saga, i.e., from "Homer" in the old sense, carefully avoided the "Homer" that was recited at the Panathenaea. It never trespassed on the Iliad and Odyssey.

Lastly it is worth remarking that, while all early quotations from Homer are apt to vary considerably from our texts, the non-Attic writers vary much more than the Attic. Hippocrates, the Coan physician, who lived chiefly in Ionia, has many quota tions which do not occur at all in our Homer, including one which he says is "frequent"; the same is true of Aristotle, born in Stageira and trained in the Ionic tradition. And it is note worthy that a speech of Syagrus the Spartan in Hdt. (vii. not only quotes a hexameter line which is not in our Homer, but seems to quote a Homer who made Agamemnon a king of Sparta. In our Homer he is king of Mycenae, though traces of Sparta seem occasionally to cling to him (e.g., Od. iv., 514ff)• The "Homer" that was "sung in scattered bits" about Greece before the time of Peisistratus may well have been very different from the Iliad and Odyssey.

Thus one may be fairly sure from the external record that there existed before the time of Peisistratus masses of heroic poetry, reputed to be very old and to be in general the work of "Homer." As to its form we can say little. It probably consisted normally of lays, of a length suitable for recitation; the length perhaps of the Catalogue or the Doloneia (400-600 lines). Whether there were already any great epics, too long for ordinary recitation but suitable for some great "panegyris" such as the Panionia or the four-yearly gathering of the Ionians at Delos, we have no evi dence. Many scholars have concluded that the Peisistratean recension provided the first written text of the Homeric poems. Certainly there is no record of any written text of the Iliad and Odyssey previous to that; yet it seems almost impossible that no lays at all should have beeen written down, and the composi tion of the two great poems does at times seem to suggest the use of written sources.

A Crystallization of Tradition.—Thus while it seems prob able that songs or lays about the siege of Troy and other achieve ments of the heroic age came into existence very soon after the events themselves, the first long epics in written form of which we have any knowledge appear in Athens in the sixth century, some 600 years later. The poems that we possess represent the last stage of epic creation, though, as far as we know, the most per fect. The history of epic poetry between and 55o is entirely obscure, and conjecturable only by indirect internal evidence. Some conclusions, however, can be drawn from the language, metre, subject matter and composition of the two poems. For example (1) it is clear from the composition that each poem is a unity, but a unity imposed on a variety of sources. The unity is made up by combining different lays or parts of lays and smoothing away the discrepancies. It is clear also that in the last shaping of the poems Athenian influence counted for much.

(2) The subject matter shows that the poems cannot be ex plained as creations of some one age midway between Troy and Athens. Some elements of custom, story and diction go back to remote antiquity, and some again are not earlier than Athens of the sixth century.

(3) The language shows traces of Attic, not only on the surface but also a little below; a great body of Ionic; traces of a definitely Aeolic dialect; and remains also of some very ancient Greek, not definitely assignable to any particular dialect and already unintelligible in classical Athens.

(4) The metre shows signs of long development, and has, to a degree perhaps unexampled in the history of literature, con ditioned and almost created the Epic dialect. Many centuries of hexametric or at least dactylic composition must have pre ceded the present form of Homeric verse.

To take these points in detail: the old explanation of the poems (Hermann and Grote) as an original nucleus plus inter polations or late additions, and of the occasional discrepancies as due to the interpolators, must by now be given up. For ex ample, there is the discrepancy between books ix. and xvi. of the Iliad, the Embassy and the Patrocleia. In xvi. Achilles, seeing the defeat of the Greeks, breaks into a splendid complaint that if only Agamemnon would seek his friendship and offer atone ment, the Trojans would soon fly and choke the trenches with their dead. Yet the whole of book ix. has been occupied by Aga memnon's offers of princely atonement and Achilles' rejection of them. The discrepancy is manifest; but it is not well explained by supposing that a late poet invented Bk. ix. and interpolated it.

Why should he so upset the story? It is explained at once if we suppose that a poet engaged in building up the great epic out of old material found both the Embassy and the Patrocleia in existence, and, not liking to sacrifice either, wove them both in and smoothed out the more alarming difficulties. Similarly he found—so the Scholia tell us—Bk. x., the Doloneia, as a sepa rate lay. It is somewhat foreign in style and it makes some little trouble, but it was too good to throw away. So again some parts of the Iliad imply the existence of a wall round the ships while others ignore it. It is much easier to suppose that the poet-editor found some lays with a wall and some without, and chose to combine them, than to imagine an "interpolator" who made gratuitous trouble by putting in a wall here and there.

Just so in the Odyssey, there is much confusion as to Odysseus' disguise: he is sometimes disguised and sometimes not. Also there are three incidents in which something is thrown at the dis guised Odysseus by one of the suitors. The three do not form a climax or show any relation to one another (p 462, v 394, v He is twice insulted by one of his own servants. In one place the offender is Melantho, daughter of Dolios (a, r), in the other it is Melanthios, son of Dolios (p, v, 4, X). The Melantho part never mentions Melanthios, nor the other part Melantho though it speaks of other wicked maidservants. Again, in the long story of adventure which Odysseus relates in the first person we find one interruption of third-person verbs (c, 54, 55), and some passages which perhaps show traces of having been once written in the third person. In all such cases the hypothesis of "interpolations" or "late additions" is of little help. The probable explanation is that two or more different sources have been com bined by the—shall we call him poet or editor? He is both; but the Greeks called him bai/.'ddos, or "a stitcher of songs." So far the divergent sources that we have considered have been merely different versions of the same body of saga. The Embassy was always an incident in the Wrath of Achilles—or some similar Wrath; both Melantho and Melanthios had a part in the adventures of Odysseus, or somebody very like Odysseus. But we also find in both poems lays or long passages of quite extraneous origin.

For example, the Catalogue of Snips in Iliad ii. is, by general agreement, an old document originally composed for a different context. It describes the mustering of the ships at Aulis, not their stations on the coast of Troy, and in various ways it shows signs of adaptation, and imperfect adaptation, to its present place. Further there are passages in Odyssey iii. and iv. which give ab breviated accounts of the Homecomings, or Nostoi, of various Greek heroes. Presumably they are derived from the Epic (or collection of lays) called The Nostoi. There are fragments derived from some Heracleia, or poem on Heracles, and it seems likely that some of the Androktasiai or "slayings of men" by various heroes may contain faded memories of real fighting in different parts of the Aegean world. If this is so, we must not reject the possibility of further "liftings"; notably a strong case has been made out for the suggestion that the exploits of Diomedes in the Iliad are in part taken from those of his father Tydeus in the Thebais, and that Hector himself, whose grave was shown in Thebes, was originally the defender of Thebes, not of Troy. The business of the song-stitcher was to make a great epic for the Panathenaea or some similar great occasion, and he had all the riches of "Homer" to draw upon.

Subject Matter.

Most of the typical Minoan or Mycenaean objects seem to have disappeared from the mainland of Greece by 1400 B.C., some 200 years before the fall of Troy. Some of these Homer does not mention (e.g., figure-of-eight shields, inlaid sword-blades, long rapiers), but he mentions elaborate palaces with bath-rooms (au j.uvOos, a pre-Greek word), Nestor's cup with the doves drinking (Il. xi. 633), a helmet with boar's tusks (x. 261-5) and the inlaid metal-work on the shield of Achilles. These are very ancient indeed but may be explained by traditional knowl edge helped out by actual relics. If we ourselves have seen a Mycenaean cup like Nestor's presumably a classical Greek may have done so. Even Euripides describes a Mycenaean inlaid swordblade (Electra 476, cf. Heracles 418) . The big shield which took the place of body armour and is often called Mycenaean remained in use, as Herodotus tells us, till about 65o. It is found in Tyrtaeus and in a relief representing Ionian or Carian mer cenaries approaching Sennacherib (705-680). A rather similar problem is presented by the discovery among Minoan remains of thirteen rings from Thisbe ; the genuineness is avouched by Sir Arthur Evans. These are engraved with scenes or motives which are prominent in the poems or in later Greek saga: a man clinging under the body of a woolly ram, like Odysseus in Od. ix., a young prince slaying a king in a chariot in a narrow defile, as Oedipus slew Laius; a creature like Scylla with many heads attacking a ship, as in Od. xii. 245; a young prince, like Orestes, killing two guilty lovers. In Cnossus itself we find repeatedly the motive of a siege of a city, the hardest work of ancient warfare, which took poetic shape in the Thebais and the Iliad.

The existence of such motives in very early pre-Homeric times gives no evidence as to the date of any particular treat ment of them in poetry. On the other hand the Iliad, though not the Odyssey, also clothes the heroes in the bronze panoply which came in not long before the time of Peisistratus, and often refers to the hoplite tactics which go with the panoply. More than that, Homer dresses both men and women, not in Myce naean bathing drawers nor flounces, but in the "old-fashioned Attic style." He describes—in a book remarkable both for "lateness" and for beauty--a procession taking a peplus to Athena in exactly the manner of the Panathenaean peplus; he makes Athena leave Odysseus and go home to the "strong house of Erechtheus," or Erechtheum, on the Athenian acropolis. It also seems significant that the adventures of Odysseus in the Odyssey are turned into a contest between Poseidon and Athena, with the latter victorious, one of the most characteristic local Athenian myths, while a sympathetic character who protects and helps Telemachus is called "Peisistratus" and made the son of Nestor, from whom Peisistratus of Athens claimed descent. One need not dwell on the supposed "Athenian interpolations" which were noted in antiquity. The omissions are equally sig nificant. In the Catalogue, for instance, Thebes, Aegina, Megara are omitted entirely; Salamis is suppressed and Corinth belittled. Athens, and no other city but Athens, has reasons for just these suppressions. .Of course, the symptoms are never gross. The poems are as far removed from flattery or boastfulness as the rest of classical Greek literature.

Language.—Apart from problems of local dialect, the dialect of Homer is an early or "primitive" form of the language which we know as Greek in the classical age. This can be shown by comparing the grammatical formation and syntax of Homer with those of Attic. (The comparison of the vocabulary is in the nature of things less conclusive.) I. The first aorist in Greek being a "weak" tense, i.e., formed by a suffix (-act), whereas the second aorist is a "strong" tense, distinguished by the form of the root-syllable, we expect to find a constant tendency to diminish the number of second aorists in use. No new second aorists, we may be sure, were formed any more than new "strong" tenses, such as came or sang, can be formed in English. Now in Homer there are upwards of 8o second aorists (not reckoning aorists of "verbs in µc," such as garnv, E(3iv), whereas in all Attic prose not more than 3o are found. In this point therefore the Homeric language is manifestly older.


While the whole class of "strong" aorists diminished, certain smaller groups in the class disappeared altogether. Thus we find in Homer, but not in the later language: (a) The second aorist middle without the "thematic" a or o: as El3Xn-ro, was struck; E49c-ro, perished; aX-ro, leaped.

(b) The aorist formed by reduplication : as

Sacco', taught XEXa(3aOac, to seize. These constitute a distinct formation, generally with a "causative" meaning ; the solitary Attic spec imen is iyayov.

3. It had long been known that the subjunctive in Homer often takes a short vowel (e.g., in the plural, -owl), -Ere instead of -wµEV, -Y)TE, and.jin the Mid. -oµac, etc. instead of -wµac, etc.). This was generally said to be done by "poetic licence," or gratin. In fact, however, the Homeric subjunctive is almost quite "regular," though the rule which it obeys is a different one from the Attic. It may be summed up by saying that the sub junctive takes w or i when the indicative has o or e, and not otherwise. Thus Homer has Z-µ€v, we go, i-o-µ€v, let us go. The later i-w-µ€v was at first a solecism, an attempt to conjugate a "verb in µc" like the "verbs in w." It will be evident that under this rule the perfect and first aorist subjunctive should always take a short vowel; and this is the case, with very few exceptions.

4. The article

(o, i, ró) in Homer is chiefly used as an in dependent pronoun (he, she, it), a use which in Attic appears only in a few combinations (such as o dv . . . o SE, the one . . . the other). This difference is parallel to the relation between the Latin ille and the article of the Romance languages.

5. The prepositions offer several points of comparison. What the grammarians called "tmesis," the separation of the preposi tion from the verb with which it is compounded, is peculiar to Homer. The true account of the matter is that in Homer the place of the preposition is not rigidly fixed, as it was afterwards. Again, "with" is in Homer avv (with the dative), in Attic prose µEra with the genitive. Here Attic poetry is intermediate; the use of avv is retained as a piece of poetical tradition.

6. In addition to the particle av, Homer has another, KEV, hardly distinguishable in meaning. The Homeric uses of av and KEV are different in several respects from the Attic, the general result being that the Homeric syntax is more elastic, and that the less common combinations of the earlier period were disused altogether in the later.

7. In the vocabulary the most striking difference is that many words appear from the metre to have contained a sound which they afterwards lost, viz., English W, written in some Greek alphabets by the "digamma." This letter, however, died out earlier in Ionic than in most dialects, and there is no proof that the Ho meric poems were ever written with it.

In many epics the nationality of the author, or of both author and scribe, can be fixed by the dialect. Thus the author of the Chanson de Roland was probably a native of the Ile de France, while the scribe was an Anglo-Norman. Similar conclusions may be drawn about many of the medieval German poems. But Homer's is no definite spoken dialect—it is a traditional dialect, regularly used for epic, shaped and twisted to an extent probably without parallel in literature by the needs and conveniences of the epic hexameter. In the main the poems are Ionic, with an Aeolic under-current. That suits the tradition, almost universal in antiquity, that the poems came from Ionia, and that "Homer was born" in Smyrna, Chios or Colophon, in regions where an Ionic population had superseded an Aeolic. We cannot how ever answer simply that the dialect is that of Smyrna or Chios at a given date; for it is clear that some process of transmutation has taken place. In general, Aeolic forms are left (ArpELBao, Ken, I oawv, MvpµcSovEaac) where the Ionic form was not metrically equivalent; wherever the two are metrically equivalent, the Ionic is preferred. An exception like 9Efi is left because there was no Ionic Oei7; OE6s was the form used. Nnos has always its Ionic form because the older poetry did not, apparently, mention tem ples, but only altars. Other exceptions are due to various acci dental causes. The attempt has been made to argue that these "aeolic" forms were not really dialectic at all, but merely be longed to an older form of the Greek language from which both Aeolic and Ionic afterwards developed. The digamma, for in stance, r, a w-sound which gradually disappeared from the Greek language like h in French or w in whom, wrong, in English, is practically a living letter in Homer. (Roughly 3,354 places imply the F while 617 ignore it.) By the sixth century F was dead in Ionic though alive in both Doric and Aeolic, and the ancients actually called it "the Aeolic letter"; but two centuries earlier it may well have been current in all forms of Greek.

This view however is disproved by the presence of new forma tions, which occur in Aeolic and in Homer, but are certainly not "Old Greek." There seems to have been some definitely Aeolic period in the development of the epic. On the other hand there is abundant evidence of "Old Greek," dating from a time prior to any Aeolic or Ionic of which we have written record. There are many phrases which were not understood in classical times, or at least had to be taught in school: ajEViva Kaprlva, S1.&KTOpos py€(.u6vrris, 147611/20S, roios (cf. Ar. Fr. 222 Daitales). And, equally significant, there are words which, in ordinary speech, only occur in Cyprian and Arcadian, such as Fava , 7rroXcs, KFpaµos (a prison). Arcadia, isolated in the mountains of the Peloponnese, and Cyprus, isolated among the Semites of the eastern Aegean, preserved in common fragments of the very oldest Greek, which had perished elsewhere.

We have here phenomena corresponding to the Mycenaean cups and primaeval motives ; while on the other hand the Attic influence on the language is equally conspicuous. It occurs in two forms. First, there are numerous Atticisms which are shown to be wrong by the metre, and must therefore be simply mis takes by the Attic scribe. Some are unmistakable like "Ews rag' Wpµacve, where Ews must be a mistake for r)os. Others are almost certain but not quite : S' 'AyEAFws µETfEt7rEV offends against Homeric idiom and has probably supplanted 'AyiXaos EEL7rev ; r)v rot) Ec/€Vpnc is pure Attic and is more likely to be the mistake of an Attic copyist or reciter for ai KEY &PEUp7)t than an original un-Homeric effort of a poet. But beyond a few clear cases we cannot be sure. If we remove drastically all the obvious "surface corruption" there remains a more deep-seated Attic element ::Acos, aµoOev yE, j3E13waa, E65a4Opoc (as a dactyl Il. xxiii.) are isolated, but o7rws (for Aeolic ihrirws, Ionic 6Kws), E7rEcOac (for inr14'ECv) are common ; and one has to re member that these Atticisms represent merely what has been left after generations of grammarians have gone over the text carefully to remove all that was not "Homeric." Metre.—Lastly, this dialect has been re-shaped by the needs of the metre. To mention only a few of the instances: wherever three or more short syllables came together, or where a short came between two long, the word had to be changed before it could get into the hexameter; hence forms like yavopezos, T ELpeoias, Ely iypoi)c, ELei 6Upqw with Et for E; or else tlrEpELUtOS ; iarir), 7rpoOvµi7)tac, 'A7riroXAwva, ELAiiAovea. Analogy of course plays a part in these changes, e.g. TLe1VLEVOS in place of 7.0&uevos because of rLOs)µc, µaXECOµEvos from the rhythm of µaXr)aoµEVOS, contrasted with 7rEpi 71-TOAcos µaxeovµ€Vos r)SE yvvaLKwv (CO 113, X 403) because of 265 iXAa 7rEpi 7rTOAtos TE µaxicr rac r)Si yvvaLKwv. Liberties are taken both with acci dence and with syntax in order to obtain forms that are metrically convenient especially conspicuous is the effort to obtain words or phrases which fill the final dactyl-spondee after the Bucolic diaere sis: ivcoxi)a acc. of r)vtoxos, AiOto7ri)as of Ai9Lo7r€s, Zapiri)Sovros for /apiri)Sovos, Evp a, 7r6vrov (from Evp& 7rovrcp); the alternation of singular and plural in such words as i7r7roavvns L7r7roavvawv, vr)7ripae, K0PL1j KovLperc, that of active and middle in €&OpoWPTE5 Fhop0OvraL, /2717-1.6wvr€S µflrcaaaOat are to be explain ed thus. Similarly the gen. in ow is prevalent in choriambic words 'AvrtµaXoco, hvc0XOto, whereas for obvious reasons we always have 'AELov, Ev, oov, Eraipov, KEAEbOov, and also MevEX tov, ivuavroi, µEyaOcµoy. In spondaic words (a'ypow, 0TicOV, crCrov) -ov is much commoner than ow.

Thus the metre both preserves ancient forms and invents new forms. As to contraction of vowels, Epic prefers the old un contracted forms but admits the shorter forms, usual no doubt in contemporary speech, for special reasons. Metrical necessity produces 7rFAciyF1., 41AEC (and OLAEC), 6,ya7ras, cbo(3EtTac, irapeo µEVOS, etc.: the convenience of having all the cases of a noun metrically identical has some effect (TvoELoew, airaaEwv, 11µEar, vµEas); while analogy also plays its part (ava(3)L for avaf3imc from avaj3as, virfpOvµov LL0µ)SEa from virfpOvµos Acoµi)Sr)s), sometimes assisted by an actual mistake: 7rwv hy' oiwv must have come from some rhapsode who pronounced F 198 os ocwv .dya 7rwv wrong. Similarly a bard who said rpEir' in vtKrw &AA' avroi TpfET' aa1rerov (11. xvii. 33 2) proceeded to say rp€iv Ea HaXXcis (Il. v. 256). The fact that the Greek language had thus to be stretched and twisted in order to satisfy the needs of the hexameter has led to the suggestion that the hexameter must be a foreign metre, made to suit a language other than Greek (Meillet, Origines Indo-Europeennes des Metres Grecs, 1923). The conclusion is not probable. After all no known language suits the hexameter half as well as Greek, and the variety of forms of words was satisfactorily explained by the ancients as due to the "singing" of early epic verse (Athenaeus xiv. p. 632d.), and Greek was spoken in the Aegean by 1400 B.C. (Buck, in Class Phil. xxi.).

The misuses of the old language by rhapsodes who were ac customed to something much later may be compared with ar€Vro for "stood" in Od. xi., 584 of 6ESOV7rbTOS Oi&L7roSao for "having died" (Il. xxiii. 679), of the misunderstanding of ESva "bride-price" as if it meant "dowry" in Od. i. 278, ii. 194, or the introduction of the proverb "iron itself draws a man on (Od. xvi.

294, xix. 13)• The Homeric Style.—These considerations may be thought to lead to a rather chaotic result, if not a purely negative one. Yet, though they do not encourage attempts to discover the age or birthplace of "Homer," they do help us to understand the great qualities of the Homeric style. If we compare the Homeric poems with the Aeneid or Paradise Lost we feel in them much more directness and vital force. They have the quality of the heroic age, of the Volsunga saga or of Beowulf. If we compare them with these latter, we feel them to be far richer in language, larger and closer in construction, nobler in artistic form. They unite, to a degree that is perhaps unique in literature, primitive force with artistic dignity and accomplishment. Poems like Iliad vi. and xxiv., for example, are hardly conceivable except as the work of minds as civilized as that of Aeschylus on ma terial as rude and fierce as the End of the Niblungs.

The same considerations throw light on another quality of Homer, which persists all through the higher Greek poetry. The grand style concentrates on the main subject, not on detail. It is not precise or realistic. It deals in types more than indi viduals. The main characteristics of the main characters, the main interest of the main story, are intensely clear, but the mass of not strictly relevant detail which contributes so much to the lifelikeness of modern fiction is mostly omitted. There are many ships described in Homer, and described vividly, but no one is ever distinguished from another. They are merely "black," "hollow," "swift," "well-balanced," "red-cheeked" or the like.

There are two quite distinct styles of fighting and of armour: first the classical battle of hoplites in phalanx with small bronze shield, breastplate and backplate, which has been introduced into many parts of the Iliad; and second, below this, the fighting of particular champions (7rpoµaXoc) in Mycenaean style with the great leather shield, reaching from neck to ankles and making body armour unnecessary, while an ill-armed mob behind, with little protection beyond goat-skins (Aa ol'ica) help with stones, darts and arrows. Yet the epithets seldom make clear what sort of shield is meant; the general descriptions often leave obscure what style of fighting. The same with the topography.

This quality enabled the bards freely to compose poetry about times and places of which they had no personal knowledge, provided only that they observed the traditional epic manner and were permeated by its spirit. A modern artist, especially a novelist, generally has two aims which seem to us normal, but which are entirely alien to the spirit of Homeric poetry. First, he aims at detailed descriptions of phenomena; while "Homer" aims straight at the emotional effect with a minimum of detail. For example, the similes taken from lions in the Iliad might always have been written by one who had never seen a lion, but never by one who had not been imaginatively thrilled by poetry about lions. Secondly, the modern artist is generally conscious of himself as opposed to his audience, and aims at producing something "original." The book he writes is printed and preserves its sharp edges, whatever the taste of the reader may be.

But the ancient bard was, in the main, performing things al ready well known, the common possession of himself and his audience. An ancient bard could not be eccentric or go against the prevailing taste. A poem so written would simply not live. It would not fit its environment, and the next reciter would alter it. It is essential, in trying to understand ancient oral poetry, to realize this fact: the successful oral poem, like the speech of a popular orator nowadays, must be the joint product of the artist and his public. Thus a rhapsode describing Ithaca or Troy or Achilles must avoid any sharp clash with the public conception of those subjects. He must not contradict people's expectations and memories. The real facts, as they would appear in a modern book of reference, do not much matter, though, in order no doubt to avoid awkward clashes, he does generally abstain from precise detail. When there is detail it is still generic.

There is an oak on the plain of Troy; oaks are common on plains. There is a fig-tree growing in or beside the wall, making a place where the wall is climbable (Il. vi. 433) . That may well be evidence for the existence of a poem describing the scaling of the wall by help of the fig-tree; it is no evidence that there was actually a fig-tree in the wall of historical Troy. The one detailed statement about the rivers of the plain of Troy, viz. that they meet (Il. v. 774), is rejected as geographically impossible even by those scholars, like Leaf and Robert, who try to identify the local details. In the descriptions of Ithaca, some points can be found on the island, and doubtless could on other islands also; some seem to suit Leucas but not Ithaca, while in some again the detail seems to be the remnant of a myth. The isle of Odys seus is described as "low down, farthest away in the sea toward the darkness: the others are separate, toward the dawn and the sunlight." That description is utterly unsuitable to the real geography, but sounds like a description of a mythical Isle of the Setting Sun, in the far west, which possibly lingered in the minds of poets and audiences from an older poem.

It is worth observing that this vagueness of detail, this use of the generic and typical rather than the hard particular, is characteristic of the higher Greek poetry in general, and is prob ably due to the same cause—the dependence of the poet on the tradition as known to, or accepted by, his audience. Thus Aeschylus in the Agamemnon makes his beacon-signals start from Mt. Ida, though in fact the top of Mt. Ida is some 3o miles in the wrong direction. He does so because Mt. Ida is always in tradition the mountain of Troy. Both audience and poet take it for granted. For somewhat similar reasons, the tragedians give no names to their subordinate characters: they are merely Servant, Messenger, or Nurse.

more precarious than the geographical iden tifications are attempts to extract history from the Homeric narrative. It is quite likely that there are hard nuggets of history preserved unchanged in the poems, but it is impossible to dis tinguish history from myth, folk-lore and fiction in such a slowly grown and beautifully welded whole. If Odysseus burns out the eye of a one-eyed giant, if he escapes from the cave by clinging under the belly of a ram while the Cyclops stops and feels the ram's back, one cannot say that such things never happened, but one should not forget that they occur in dozens of folk-tales. If Odysseus had 36o boars, one of whom died every night, one cannot but think of the sun or the year, as described in ancient riddles: one cannot help noticing with interest that the day on which Odysseus and Penelope met was the exact day in which ancient astronomers considered that the sun cycle and moon cycle coincided: the day called "the meeting of sun and moon" in Meton's Eikosieteris or 20 year cycle. Such solar material may quite well be blended with a story of real life. One must simply wait for the evidence.

Again, there is nothing at all impossible in the story that the cause or pretext of the Trojan War was the rape of a princess. Yet the fact that a statement is possible is no proof that it is true: and one must remember that Helen was, in historical times a marriage goddess in Sparta, and as such must herself go through the marriage ceremony, which comprised in Sparta the carrying off of the bride. Hence Helen in saga is constantly "carried off" —to Sidon or Troy by Paris, to Egypt by Hermes, to Deceleia by Theseus or the Apharetidae, to Parmon by "a robber." There is ritual significance also in the fact that she is generally restored by twins, either her brothers, the Dioscuri, or the gemini Atridae; and that, as with other marriage-goddesses, the story sometimes explains that it was not the true goddess but only an "image" that the ravisher seized.

There will be little difference of opinion as to the existence in Homer of real history, mixed up with fiction, folk-lore and myth; it is also clear that we have not at present any body of evidence by which to sift out the history. It would moreover be a great error of method to suppose that in such a combination there is always, or usually, a true nucleus and a fictional or mythical penumbra. In the Niebelunglied, for example, the myth seems to be the nucleus, which has attracted into it some historical figures like Atli and Dieterich of Berne (Attila and Theodoric). The same would hold of most historical novels.

Nor should much weight be attached to the fact that the in cidents of the Iliad are generally of the possible sort; marvels and monsters are kept outside the story. That is purely a question of style. The Iliad is tragic, and serious. The Odyssey veers towards the fairy-tale in the Story of Odysseus (ix. to xii.) and comes back to reality in the later books. The Argonautica was TEparcidfls, "full of marvels," throughout. On the other hand the actual names borne by the chief heroes may well be historical or at least derived from history. Atreus, Paris, Helene, Odysseus (or Olytteus) do not seem to be Greek words; Achilleus, Aga memnon, Menelaus look like non-Greek words twisted into a semblance of Greek. If so, they are probably not invented but derived from real persons who bore names more or less like them, while Diomedes (Zeus-counsel), Neoptolemus (War-new), Hec tor (Holder) and perhaps Priam (King) seem like pure Greek names which anyone could invent.

poems, homeric, greek, text, iliad, poetry and bc