with which he has referred objects to the third, fourth, or the fifth stratum. For all purposes of scientific discussion, the objects found in this middle zone must, therefore, be considered as belonging to these three settlements indifferently. But the uncertainty does not stop here. In the case of the terra-cotta with the helmeted headan object so adverse to Dr. Schliemann's "prehistoric" theory of the stratum to which "Troy" referred it—we have the "find" suddenly transferred from the lowest part of the middle zone to the historic stratum above it! And, conversely, an entire series of 794 objects, given by "Ilios" to the middle zone, is shifted by "Troja" to the prehistoric statum below it! Yet we are expected to believe that each of the three settlements in the middle zone is prehistoric on this ground alone that to each of them Dr. Schliemann has attributed with unfailing accuracy those objects, and those only, which, in that remote past, actually belonged to each!

That the oldest objects dug up at Hissarlik go back to a high prehistoric antiquity, few would dispute. I, for one, have never doubted it. What I deny, and what most people accustomed to weigh evidence would agree with me in denying, is that the unverified and unverifiable assertion of Dr. Schliemann, or of one of his overseers, can be accepted as proof that a given object was found in precisely that position which would constitute it a document in favour of the prehistoric character of settlement 3, 4, or 5. This is a matter of which we can judge for ourselves from a comparison of his own

statements.

From the registration of positions, I now turn to another question -the description of the objects found at a greater depth than 6 feet. Here there are three points which I desire to bring out distinctly.

(i.) The great mass of the objects claimed as evidence for the prehistoric age of settlements 3, 4, and 5, are bowls, cups, plates, vases, and such similar vessels or implements as might have been in common use in households of a humble class. Mr. G. Nikolaïdes, the able and learned author of Ιλιάδος Στρατηγική Διασκευή remarks on these (p. 163):

"The earthenware vessels found (according to Schliemann) in the strata below the topmost, indicate poverty rather than antiquity-just as poverty, not antiquity, is indicated by the cheap and rude earthenware vessels, fabricated at the present day in the villages of Tuscany, as compared with the contemporary masterpieces of ceramic art which come from the workshops of Florence and Lucca. As for the other miscellaneous objects exhibited to us as proofs of remote antiquity, or as bearing the tokens of a mysterious religion, they are simply children's toys, ornaments of poor tombs, or weights of fishermen's nets, marked with rude letters and figures."

Rousopoulos, also, in his Εγχειρίδιον Ἑλληνικῆς ̓Αρχαιολογίας, remarks on the rude character of most of the objects found at Hissarlik, and on the absurdity of comparing them with the works of finished art described in the Homeric poems.

(ii.) In the bulk of the ceramic objects which he refers to settlements 3, 4, and 5, Dr. Schliemann recognises a general community of character, broadly distinguishing them from the objects found in the two lowest cities (1 and 2). See, e.g., "Ilios" p. 510, note 2 :—

"There can be no mistake in the pottery of the two lowest cities, the types being so vastly different from each other, and also from the pottery of all the following cities."

Again, at p. 420, the general community of ceramic type between cities 3, 4, 5 is remarked, and various particular instances are noticed passim-e.g. p. 535 (compared with p. 371), pp. 354, 373, 537, 543, 558, 559.

(iii.) In his “Céramiques de la Gréce Propre," M. Dumont holds that the prehistoric pottery of Hissarlik is of a high antiquity, but finds mingled with it an element which is not distinguishable from the characters common to Hellenic civilisation in the historic age. Thus, at p. 8, speaking of large amphoras found at Hissarlik below six feet:

'Les grands amphores n'ont pas de caractères originaux. Elles pourraient appartenir à toutes les époques; la forme est déjà celle que nous trouverons durant toute la civilisation hellénique.'

Again, p. 69:

'Plusieurs formes ne sont pas différentes de celles qui prédomineront en Grèce à l'époque classique; les autres procèdent des mêmes principes.'

Here, then, in the strata of the 3rd, 4th, and 5th settlements, we have pottery which might belong "to any epoch." Its forms are those "which we shall find throughout the whole course of Hellenic civilisation;"" those which we shall find prevailing in Greece in the classical epoch." The most reasonable explanation of the phenomenon is to be found in the inference that the 3rd, 4th, and 5th settlements were Hellenic. This inference is definitely strengthened by particular facts. At twenty-six feet was found a terra-cotta ball, marked with parallel lines. Dr. Schliemann himself recognised these lines, both in "Troy" and in "Ilios," as meant to indicate "the climates of the globe." Mr. Philip Smith, the English editor of Dr. Schliemann's "Troy," pointed out (p. 188) the plain indications of the torrid and temperate zones on this ball. Now, as Dr. Brentano observed, the mathematical proof of the spherical form of the earth was first given by Eudoxus of Cnidus, circ. 370-360 B.c. From that time dated the division of the globe into zones. Since the objects referable to strata 3, 4, 5 cannot, as Dr. Schliemann admits, be accurately discriminated, this terra-cotta ball, found in the stratum of 3, may have belonged to 4 or to 5. That is, if I am right, it may have belonged to the Greek Ilium or to the Macedonian Ilium. This would agree

with the chronological inference from the zones. Not being able to meet this argument, Dr. Schliemann resorts, as he too often does in "Troja," to invective. This proves nothing: the inference from the terra-cotta ball remains as firm as if Dr. Schliemann had been logical

1

and courteous. Then, at a depth of twenty feet, was found another terra-cotta ball ("Ilios," p. 563; Nos. 1228, 1229), scrawled over with a letter which, as Dr. Schliemann himself perceived, is the Greek p. Here is another Hellenic trace at this depth. And next, I would invite attention to another very interesting piece of evidence, which, so far as I am aware, has not yet been noticed. Object No. 520, figured in "Ilios," p. 424, is thus described:

"Piece of Ivory, belonging to a Trojan seven-stringed Lyre."

This was found at a depth of twenty-six feet, i.e., in the stratum of the Third City. The seven holes pierced in the ivory prove that the lyre really had seven strings. Now, as is well known, the lyre, which in an older form had only four strings, is said to have been first made a heptachord by the Aeolic Terpander of Lesbos, about 660 B.C. And whether the tradition is or is not correct as to the man, there can be little doubt that it is nearly right as to the age, since the date agrees with the rise of that poetry which presupposes a developed lyre. The lyric poetry of the Aeolic and Dorian schools. had its beginnings in the seventh century B.C. Suppose, now, that the Third City was indeed the earliest phase of the historic Iliumthe Aeolic colony. Nothing could be more natural than that we should find there the newly-improved instrument of Aeolic song. Though Dr. Schliemann's ivory lyre was never touched by Paris or Cassandra, it is quite possible that it may have vibrated to melodies of Sappho or Alcaeus.

Thus, then, stands the question as to whether the third, fourth, and fifth settlements are prehistoric or Hellenic. The evidence of particular objects found in the diggings could not be brought to bear with precision except on an assumption which is absolutely unwarrantable, and which Dr. Schliemann's own books decisively exclude, viz., that both the depth and the position of each several object were accurately registered by him, or by his workmen, at the moment of discovery. On the other hand, we find ascribed to strata below six feet not only pottery which is presumably older than Hellenic life at Hissarlik, but also pottery which, as M. Dumont says, might have belonged to any epoch of Hellenic civilization. Further, we find in the lower strata an object marked with Greek letters, another object which should not be older than the seventh century B.C., and another, not older than the fourth century B.C. In regard to the architectural remains, an eminent expert reports that he cannot prove them to be Hellenic, but neither can he prove them to be prehistoric. And on the supposition that they are Hellenic, their present state agrees with historical notices of successive phases in the architectural life of the

(1) "Dr. Schliemann is mistaken (“Troja," p. 241) when he says that he has 'answered in full Professor Jebb's and Brentano's fallacies about the terra-cotta ball.' The 'fallacies' were his own. The inferences of his opponents were logically unassailable."—Saturday Review, Dec. 8, 1883.

town which actually stood at Hissarlik, as Dr. Dörpfeld himself has expressly acknowledged.

Fair-minded readers will now be able to weigh the probabilities for themselves. If we suppose these three settlements to be Hellenic, then the settlement which preceded them, though prehistoric, is not thrown back into so remote an antiquity as to exclude the idea that it may have been a town which was taken by Achaean invaders of the Troad, and which gave rise to the legend of Troy. If, on the contrary, we suppose them to be prehistoric, then we have to make chronological room between our "Troy" and circ. 700 B.C.—when the Greek Ilium was founded-for the rise, existence, decay, and fall of three successive prehistoric cities. On this view, it would be a moderate estimate which referred the archetype of Troy to about 2300 B.C. But a town which perished in a pre-Hellenic antiquity cannot have been the town which gave rise to the legend of Troy, since that legend requires Hellenic invaders.

So far of the actual remains at Hissarlik. "Troja" has corrected the stratification given in "Ilios," and has also adopted the "second city " as Troy. And it is probable, as has been seen, that only two settlements of any importance are to be recognised-viz., (1) this prehistoric town, now called Troy, and (2) the historic Ilium, in its Greek, Macedonian, and Roman phases.

We now turn to a further question. What is meant by "finding Troy"? In what sense is it possible to find it?

Dr. Schliemann set out from the fixed belief that the poetry relating to Troy had the value of history. Even in "Troja" he quotes an utterance of Cassandra in the "Agamemnon" of Aeschylus as an argument for the precise course of the river Scamander.1 Mr. Sayce, on the other hand, finds this faith no longer so easy. He has lately discovered that the "Iliad" is a cynical poem composed by a late Greek sceptic who had caught the spirit of Aristophanes and anticipated that of Cervantes. "To me," writes Mr. Sayce, "the general tone of the Iliad' sounds like that of Don Quixote." "2 A critic of so much literary insight could not be so inconsistent as to assume that this mock epic of circ. 430 B.c. was also a chronicle, trustworthy for the minutest facts of circ. 1180 B.C. Hence the wavering and haziness which pervade the references of "Troja" to this awkward topic. It has, indeed, been a perpetual cause of confusion in the theories of

(1) "Troja," p. 67. Cassandra says: "Scamander, river of my fatherland! From the days of my childhood I grew up near thy banks." (Agam. 1156 ff.) "This passage seems to prove that, in the opinion of Aeschylus, the Scamander flowed at the foot of Troy, and consequently that it was to be held identical with the immense bed of the small rivulet (Kalifatli Asmak), which really flows at the foot of Hissarlik"!

(2) Journal of Philology, Vol. xii., pp. 37 ff.-"The Odyssey seems to me to breathe the spirit of Aiskhylos, the Iliad the spirit of Aristophanês. . . . . To me the general tone of the Iliad sounds like that of Don Quixote: there runs through the greater part of it a mocking laugh, which holds up to scorn all that once claimed the deepest reverence of the Greek people"!

Dr. Schliemann and his allies. I shall endeavour to put the points of the case distinctly.

I. The Homeric poems are the ultimate sources of knowledge concerning Troy. If an oral tradition had preserved the facts of a real war, the Homeric poet might have used it, but he might also have altered or embellished it. The story, in the shape in which we have it, must be considered as a poetic creation. The romance of Charlemagne embodies the historical fact that an Emperor once ruled Western Europe from the Eider to the Ebro.1 It also departs from history in sending Charlemagne on a crusade to Jerusalem, because, when the romance arose, a crusade belonged to the ideal of chivalry. Analogy might suggest that an Achaean prince had once really held a position like that of Agamemnon; also, that some Greek expedition to the Troad had occurred, whether this Achaean prince had himself borne part in it or not. Both inferences are probable on other grounds. Some memorable capture of a town in the Troad had probably been made by Greek warriors; beyond this we cannot safely go. Myth could deal with oral tradition as freely as romance with written history. Indeed, it would be natural to expect that the liberties of myth should be even bolder than those of romance, since the control was less definite. Pompeii was buried, and was rediscovered. The difference between the case of Pompeii and the case of Troy is not merely in the degree of the evidence, but in the kind That Pompeii (1) existed, (2) existed there, are facts as well attested as any in history. For Troy all the evidence is, in its nature, only mythical. It depends on poetical fancy playing around unwritten legend.

II. The Homeric data for the site of Troy (of the buildings we shall speak presently) fall under three heads. (a) First come those more general data which are implied in the political attributes of the town, and in the story of the siege, as conceived by the Homeric poet. Troy is described as the wealthy capital of a wide and powerful realm. Priam is a mighty king, with numerous tributaries and allies. Troy stands a siege of ten years by the united forces of Greece. All this implies that the site of Troy-as imagined by the Homeric poet-was one of natural strength. No town could easily have had a weaker site than the low mound at Hissarlik, in the most exposed part of the open plain, and close to the Hellespont. The historic Ilium was always, in a military sense, very weak: it was taken by Dercyllidas-by Charidemus-by Fimbria, and abandoned even by the warlike Gauls precisely because it was untenable. Half the force of the Homeric Greeks could have cut off its supplies, and starved it out,-unless they had preferred to storm it,-in as many

(1) Freeman, Historical Essays, First Series, Essay I., "The Mythical and Romantic Elements in Early English History," p. 30. The whole essay is well worth study in connection with the Trojan question.