Excavations at Troy, 1938 9781463221249

Citation preview

Excavations at Troy, 1938

(m

\\\

A n a l e c t a Gorgiana

249 Series Editor George Kiraz

Analecta Gorgiana is a collection of long essays and

short

monographs which are consistently cited by modern scholars but previously difficult to find because of their original appearance in obscure publications. Carefully selected by a team of scholars based on their relevance to modern scholarship, these essays can now be fully utili2ed by scholars and proudly owned by libraries.

Excavations at Troy, 1938

Carl Biegen

gorgias press 2009

Gorgias Press LLC, 180 Centennial Ave., Piscataway, NJ, 08854, USA www.gorgiaspress.com Copyright © 2009 by Gorgias Press LLC Originally published in All rights reserved under International and Pan-American Copyright Conventions. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise without the prior written permission of Gorgias Press LLC. 2009

1

ISBN 978-1-60724-478-3

ISSN 1935-6854

Extract from The A^merican Journal of Archaeology, vol. 43 (1939).

Printed in the LTnited States of America

CORRIGENDA

p. p. p. p. p. p. p.

207, 208, 208, 215, 217, 222, 228,

1. 19, for southwestern, read southeastern. fig. 4, should be figure 14; caption correct for figure 4. fig. 5, read: Tower V I h from E a s t . fig. 10, read: Pillar House from East, Showing Foundations of South Wall, fig. 14, should be fig. 4 ; caption correct for fig. 14. fig. €1, read: Lower Temenos with Archaic and Hellenistic Altars, 1. 1, for Arkan, read Arikan.

EXCAVATIONS AT TROY,

1938

A SEVENTH c a m p a i g n , extending f r o m t h e beginning of April to t h e 10th of J u l y , 1938, b r o u g h t t o a conclusion t h e excavations a t T r o y which h a v e been c o n d u c t e d annually since 1932 b y t h e archaeological expedition of t h e University of Cincinnati. 1 Preliminary r e p o r t s of t h e results obtained in preceding years h a v e been published a f t e r each season in earlier n u m b e r s of this JOURNAL; 2 a n d in order to complete t h e series a s u m m a r y of t h e work done in t h e final c a m p a i g n is offered here. I t is fitting on this occasion t o refer once more to t h e initiative of Professor a n d M r s . W . T . Semple in t h e founding of our enterprise and to their unfailing interest a n d s u p p o r t , accorded in b o u n t i f u l measure, which alone h a v e m a d e t h e completion of t h e u n d e r t a k i n g possible. I t is likewise a n agreeable d u t y to recall again o u r obligation to t h e G e r m a n Archaeological I n s t i t u t e a n d its generous a t t i t u d e in waiving its own p r o j e c t a t T r o y in f a v o r of ours. N o report of our own excavations would be complete w i t h o u t a g r a t e f u l acknowledgment of our enduring indebtedness t o Professor D o r p f e l d , whose brilliant discoveries m a n y decades ago laid t h e firm f o u n d a t i o n for all s u b s e q u e n t research in T r o j a n problems a n d whose cordial enc o u r a g e m e n t a n d help h a v e been c o n s t a n t l y extended to us during t h e p a s t seven years of our work a t t h e site. T h e greater p a r t of our t i m e was devoted to intensive s t u d y of t h e m a t e r i a l previously recovered, m u c h of which, especially on t h e ceramic side, still awaited a t t e n t i o n ; b u t digging was actively continued in some eight different areas. On t h e eastern side of t h e citadel we h a d already in 1937 removed a v a s t a m o u n t of e a r t h a n d debris a n d some later constructions in order to expose properly to view t h e massive, well preserved fortification of t h e Sixth S e t t l e m e n t . A t t h e end of t h a t season a long stretch of the town-wall, extending s o u t h w a r d f r o m t h e E a s t G a t e , stood f o r t h in all its impressiveness, together with t h e greater p a r t of t h e eastern f a c e of T o w e r V l h . 3 I n t h e recent campaign, u n d e r t h e supervision of M r s . Blegen, who was assisted b y M r s . Hill, excavation was continued s o u t h w a r d a n d westward, in squares J - K 8, with t h e o b j e c t of completing t h e t a s k b y disengaging t h e southern side of t h e tower a n d t h e adjoining section of t h e fortress wall. T h e area measured 1 T h e regular staff of the expedition comprised John L. Caskey, Lewey T . Lands, Dorothy Rawson. Marion Rawson, and C. W. Blegen, Field Director. Mrs. C. W . Blegen and Mrs. B. H . Hill assisted during t h e greater part of May, M r . and Mrs. N.-G. Gejvall from M a y 12 to J u n e 12, Mrs. J. L. Caskey from June 9 to July 10, Dr. B. H. Hill for a short time in June, M r . and Mrs. John L. Angel in July and August. Our Commissioner, representing the Turkish Government, was M r . H a y d a r Siimerkan. K a n i Barin of Erenkov, our veteran foreman, again served us competently and faithfully throughout the campaign, during which the number of workmen employed reached a maximum of 46. The various tasks in addition to the supervision of digging were distributed as in former years: M r . Caskey again acted as cashier and accountant and kept the epigraphical record; Miss Dorothy Rawson made t h e catalogue of miscellaneous objects, Miss Marion Rawson the inventory of t h e pottery, and M r . Lands, our architect, completed all the plans and made innumerable drawings of sherds, pottery-profiles and small objects. 1 AJA. xxxvi, 1932, pp. 131 ff.; xxxviii, 1934. pp. 223 ff.; xxxix, 1933, pp. 6 if., 550 ff.; xli, 1937, pp. 17 ff., 553 ff. 3 The designation is taken from Professor Dorpfeld's plan, Troja und Ilion, pi. iii: cf. AJA. xli, 1937, pi. xix. 204

T H E ARCHAEOLOGICAL I N S T I T U T E O F AMERICA

EXCAVATIONS

AT

TROY,

1938

205

roughly 10 by iO m., a n d as t h e deposit varied f r o m 3 t o 5 m . in d e p t h , a formidable q u a n t i t y of e a r t h had t o be shifted, most of it being carried off t o the n o r t h e r n edge of the hill. I n t h e course of these operations it proved necessary to demolish t h e ruins of some later buildings t h a t encumbered t h e ground, t h o u g h representative portions of all t h e s t r u c t u r e s removed h a v e been left in place to indicate their character. T h e principal obstacle here was t h e substantial f o u n d a t i o n t h a t once supported the south wall of the great R o m a n colonnade, called b y D o r p f e l d I X M , which passed directly across t h e southeastern corner of Tower V l h . T h e l a t t e r h a d indeed suffered no little d a m a g e when t h e R o m a n builders were preparing t h e bedding for their masonry. T h e demolition of t h e western p a r t of this wall sufficed t o free t h e angle a n d the southern face of t h e tower. Four parallel lines of t e r r a c o t t a water-pipes (fig. 1), sloping downward f r o m west t o east just outside t h e f o u n d a t i o n s of t h e colonnade, belonged to a water-system of late R o m a n times, when a small reservoir h a d been installed within t h e citadel just above T h e a t e r B, t h e Bouleuterion. T w o h e a v y f o u n d a t i o n walls indicated t h a t a room had been added a t some time against t h e south end of the colonnade; and to t h e west of it a rough stairway of three steps led t o a sunken area below t h e level of t h e floor of t h e portico. F a r t h e r to t h e southwest a short piece of a n earlier wall—of Hellenistic d a t e also h a d t o be t a k e n a w a v ; it once formed p a r t of a large building, perhaps a colonnade, t h e remaining ruins of which still lie beneath t h e northeastern q u a r t e r of t h e Bouleuterion, as shown on D o r p f e l d ' s plan. FIG. 1.—TERRACOTTA WATER PIPES W h e n all these later and intrusive elements o r ROMAN TIMES h a d been eliminated t h e character of the underlying pre-classical deposit became a p p a r e n t . I t consisted mainly of accumulated debris heaped u p against the town wall and sloping downward toward t h e east, although in this direction subsequent disturbance had p e n e t r a t e d more deeply t h a n alongside the Sixth City wall a n d m a y h a v e stripped away some of t h e deposit. A t t h e western end of the area, immediately a d j a c e n t t o the fortification, portions of several small rooms of one or more houses c a m e to light, some of which had already been seen a n d recorded on his plan b y Professor Dorpfeld. T h e floors were p a v e d with cobbles, and t h e walls, built of rough unworked stones, were marked by t h e use of small o r t h o s t a t e s in the base course. T h e a t t r i b u t i o n of these remains t o period V l l b was f u r t h e r supported by t h e p o t t e r y recovered, which included some sherds of Buckelkeramik and which permitted t h e reconstruction of a large twohandled jar in gray ware.

FIG.

2.—FRAGMENT

OF B O W L

WITH

PLASTIC

OF HUMAN HAND. TROY

REPRESENTATION*

VI

FIG. O.— CYPRIOTE AND MYCENAEAN

SHERDS

EXCAVATIONS AT TROY, 1938

207

Immediately below this habitation layer we came upon a stratum, ca. 0.60 m. deep, containing mainly potsherds of period Vila. At the bottom of it were two roughly circular, stone-paved areas, about 1.30 m. in diameter, which may have been hearths. They were bedded on a deposit, ca. 1 to 1.25 m. deep, which showed several striations, varying somewhat in color and composition, but was clearly a single layer, referable to one period. I t contained a good deal of carbonized matter, quantities of animal bones, and a great mass of shattered pottery, almost all of which could without hesitation be assigned to the final phase of Troy VI as represented in the extensive "earthquake deposit" discovered in 1936 in square J 6. A few sherds were evidently of earlier date, and occasionally fragments appeared that looked like the characteristic wares of Troy V i l a . We concluded that the deposit as a whole must comprise the débris thrown out over the town wall early in period V i l a in the course of levelling operations made necessary after the destructive earthquake that brought the Sixth Settlement to its end. The lower limit of the débris was marked off by an underlying bed of crushed murex shells, like that noted inside the citadel between House VIG and the fortification wall. The shells in turn rested on a firm layer of brown earth which yielded potsherds of the earlier phases of Troy VI; and the sequence of occupation-levels at the southwestern edge of the acropolis is thus gratifyingly complete. As might be expected, the broken pottery thrown out as rubbish proved to be sadly incomplete, and though we devoted considerable time and effort to the task, in view of the excellent quality of the material, it was not possible to put together many vases. The shapes represented are particularly numerous: plates, bowls, rounded and angular cups, the cylix, goblets, jugs, basins, craters, deep jars, and the stirrup vase were recognized. The most remarkable piece (fig. 2) is a fragment in gray ware of a plain bowl (probably one of several attached as in a kernos to a central stand), which was supported underneath by a fairly well modelled representation of a human hand. The fingers are all carefully distinguished, though the thumb is turned at an unnatural angle, and even the finger-nails are indicated in some detail. I know no counterpart to this peculiar vessel. More than 200 fragments of Mycenaean and Cypriote pottery were recovered from this deposit (fig. 3). Most of them are small and not many are of especial value in themselves, but they are highly useful in the chronological evidence they provide. At least two-thirds of these sherds are from imported vessels, the majority brought from metropolitan Mycenaean centers, others apparently of provincial manufacture; and the rest seem to be local Trojan imitations and adaptations. With the exception of one or two earlier pieces, the whole collection must be attributed to the fourteenth century B.C., and our dating of the end of Troy VI based on the evidence from the cemetery and from the "earthquake-deposit" is thus strikingly confirmed. The most manifest result of our work on the eastern border of the citadel, however, lies in the revelation to a general view of a considerable part of the handsome fortification wall of Troy VI (fig. 4), along with its massive abutting Tower, Vlh (fig. 5). From an observation-platform, which has been arranged on the high bank of earth to the east, the whole monument may now be viewed in its simple lofty

F I G . 5 . — T O W B B Y I H FROM E A S T

EXCAVATIONS AT TROY, 1938

209

grandeur, and no one who sees it can fail to be impressed and filled with admiration for t h e architectural genius of the rulers of t h e Sixth Settlement at Troy. Outside the east wall of T r o y VI, in square K 7, a test pit was dug in order to permit a study of t h e stratification. I n this area we had in 1937 uncovered some remains of a house of period V i l a , which had been partly destroyed when t h e massive foundations of the R o m a n colonnade I X M were laid in their deep trenches cut down to native rock. J u s t beneath the floors of the house we had encountered a layer of relatively loose earth containing enormous quantities of animal bones and of broken pottery of the kinds with which we h a d become familiar in the " e a r t h q u a k e d e p o s i t " inside the acropolis. This layer now proved to have a thickness of ca. 1 m.; exactly like the deposit found by Mrs. Blegen in squares J - K 8, it is surely the debris spread out when the settlement was being reconstructed at the beginning of period V i l a after the seismic disaster. I t rested on a sort of hard pavement of whitish earth, clay and pebbles, which I think certainly gives us the ground-surface t h a t was used after t h e erection of t h e fortification walls, in the latest phases of the Sixth Settlement. Continuing on down we brought to light an earlier habitation-layer, some 0.80 m. deeper, marked by a b u n d a n t traces of burning, ashes, carbonized matter, and red-baked earth. T h e pottery, with small carinated cups and stemmed goblets as characteristic forms, is clearly assignable to an early stage of VI, long preceding t h a t represented by t h e whitish pavement; and we must consequently assume t h a t an intermediate layer is missing in the stratification here. Perhaps the ground was cut down considerably when the fortress was built. A still earlier occupation-level was reached some 0.40 m. below t h e burnt layer, and this apparently belongs to the very beginning of the Sixth Settlement. T h e pottery is mainly M i n v a n ware, gray, red or buff in color, of an early undeveloped type with angular shapes. Goblets on stems are common, and t h e stems occur both in plain and in ringed varieties. Some of the red ware is hardly distinguishable from t h a t of T r o y V. Beneath these earliest remains of occupation in Troy V I our pit revealed a thin sterile layer of decomposed hardpan which rested on the rough and irregular native rock itself. A much larger trench to afford a test of the stratification was dug inside the citadel in square J 7, running from east to west across the space between House VIG and the rear wall of Tower V l h . I n previous work carried out here by Mrs. Blegen in 1936 and 1937 habitation-layers of V l l b and V i l a had been successively removed, together with a subjacent fill of " e a r t h q u a k e d e b r i s " ; a late ground-level of T r o y V I had been identified and below it a thick deposit assignable to the penultimate stage of the Sixth Settlement had yielded an immense harvest of pottery. This rich deposit came to its end a t ca. 4 m. below our datum, where a hard-packed layer of earth marked a floor or a pavement. Beginning our trench a t this point in 1938 we dug steadily downward until native rock finally appeared at a depth of ca. 8.70 m. T h e ground at all times sloped downward from west to east, as shown by the m a n y successive strata clearly distinguishable in the side of our trench. I n this deep accumulation it was possible to recognize six major lines of division, presumably indicating as m a n y consecutive stages of h u m a n occupation and activity. Measured at the middle of the trench these lines came a t approximately 4.60 m., 5.75 m., 7 m., 7.40 m., 8.20 m., and 8.70 m. below our d a t u m .

210

CARL W. BLEGEN

T h e thin layer, lowest of all, which rested on the rock itself and filled a basin-like cavity of no great size, hollowed out in hardpan, contained only a few small nondescript potsherds, some of which can be identified as fabrics of T r o y I, while a few others m a y perhaps be attributed to I I . T h e sherds found immediately above this layer in the last few centimeters before h a r d p a n was reached, include a few fragments of red ware, probably belonging to T r o y V. B u t with these exceptions, the pottery recovered appertains exclusively to T r o y VI, and it is clear t h a t the whole of the deposit, some 4.50 m. deep, was laid down during the time of the Sixth Settlement. Indeed, when we include in our calculation the layer, 0.80 m. thick, assigned to t h e penultimate phase of VI, which was dug in 1937, we obtain a formidable total of not less t h a n 5.30 m. for the depth of the accumulation of earth and rubbish t h a t m u s t have been formed in this one period. Counting only the principal lines of division, floors, pavements and the like, we can differentiate no fewer than seven subs t r a t a in this huge fill of debris. T h e numerous potsherds carefully Collected from all these seven layers constitute a real ceramic series, exhibiting a slow regular development, reflected both in fabric and in form, from the sturdy vigorous wares of the initial phase of the Sixth Settlement to the more refined and elegant products of the final " e a r t h q u a k e period." I have no doubt t h a t from evidence such as this one m a y safely deduce a concurrent social and economic change in the community, as with a rise of material prosperity it slowly evolved from a hardy Spartan simplicity to a softer and more luxurious mode of life. T h e seven substrata undeniably betoken an equal number of ceramic phases through which t h e settlement passed, b u t there is no sharp change from one to another. In t h e absence of architectural evidence for some of these phases, and without still further tests in other areas, it would be unsafe to generalize and postulate a like number of divisions extending throughout the whole mound. Though we cannot therefore maintain t h a t T r o y V I is to be divided definitely into exactly seven subperiods, we m a y with perfect safety conclude t h a t it was a very long period, and the evidence from square J 7 thus corroborates fully what we had previously inferred from t h e great depth of t h e deposit in squares F 9 and C 8 along the southern edge of the acropolis. A detailed discussion of t h e pottery from our trench m a y be left to a f u t u r e occasion. Here it will suffice to remark t h a t t h e earliest fabrics of T r o y VI, characteristic of the initial phases of t h a t period, belong unmistakably to t h e various categories of M i n v a n ware. At the beginning red and buff wares are more common than gray, a circumstance t h a t m a y perhaps be ascribed to the influence of the ceramic tradition of the F i f t h Settlement, with which the E a r l y Bronze Age comes to its close. Later in the period gray ware becomes vastly predominant; and finally, in the concluding years of t h e settlement, just before t h e earthquake marks t h e passing of T r o y VI, red ware in a characteristic orange-red variety again takes an important place beside the gray. Throughout the period sober monochromv prevails, and t h e T r o j a n potters seem never to have a t t e m p t e d to create a style of painted decoration of their own. I n the later phases of T r o y V I decorated pottery, imported from Mycenae and from Cyprus, is certainly not uncommon, and, as we have seen, it frequently enough gave rise to local copies and imitations. B u t in the earlier half of the period, corresponding

EXCAVATIONS AT TROY, 1938

211

to t h e M i d d l e Bronze Age, i m p o r t e d ware is exceedingly rare. I n our t r e n c h a f t e r a d e p t h of 4.80 m . below our d a t u m h a d been reached only three p a i n t e d sherds c a m e t o light. One, a t - 6 . 8 0 m., is a piece of M a t t p a i n t e d W a r e ; t h e second, a t - 7 . 2 0 m., is a small b i t bearing a b a n d in black glaze; a n d the third, a t - 8 . 1 0 m., is a f r a g m e n t of a flat-rimmed hole-mouth jar, crudely finished, with a reddish slip on which t h e p a t t e r n was p a i n t e d in c r e a m y white. All t h a t remains of the design is p a r t of a c u r v i n g stripe, p e r h a p s t h e b r o a d rim of a spiral, bordered b y a row of dots. T h e vessel f r o m which this bit was broken certainly c a m e f r o m a b r o a d , b u t its place of origin h a s not y e t been determined. A s u p p l e m e n t a r y examination m a d e b y M r . Caskey just south of House V I F revealed t h e o u t e r face of t h e great doorway which originally gave access f r o m this side to t h e lower floor of t h e building. T h e c h a r a c t e r of t h e opening was long ago observed b y Professor D o r p f e l d , although it h a d been walled s h u t (fig. 6) in one of t h e later periods of occupation of t h e house. T h e exterior edge of t h e threshold, which was composed of several flat stone slabs, was uncovered, a n d outside it were n o t e d remains of t h e stone p a v e m e n t of a street or passage coming f r o m t h e east. I t was n o t possible to determine t h e w i d t h of this roadway, as t h e area t o w a r d the south was occupied b y t h e wall of a large house of period V i l a . T h e j a m b s of t h e d o o r w a y were deeply grooved, p r e s u m a b l y for t h e reception of a h e a v y wooden casing or door-frame. T h e door itself m u s t h a v e h a d two leaves swinging on pivots t h a t revolved in stone sockets, set j u s t inside t h e room. I n t h e region of t h e South G a t e some f u r t h e r investigations were u n d e r t a k e n in 1938 in t h e hope of elucidating certain unsolved problems. A considerable a m o u n t of digging h a d been done in previous seasons j u s t outside t h e g a t e w a y , where we h a d endeavored t o ascertain t h e c h a r a c t e r of a building t h a t a p p a r e n t l y stood directly opposite t h e projecting T o w e r Vli, facing t h e intervening stone-paved street. T h e n o r t h a n t a of t h e edifice, owing its preservation to t h e protection afforded b y a superposed R o m a n f o u n d a t i o n , is a h a n d s o m e t a p e r i n g pillar (fig. 7), carefully c o n s t r u c t e d of well-worked blocks in the best architectural style of T r o y V I . T h e n o r t h wall, extending eastward, alongside a n d almost parallel to t h e fortification wall, h a d been exposed to a length of ca. 8 m., b u t its end h a d not y e t been f o u n d . T o w a r d t h e south, where Hellenistic a n d R o m a n constructions h a d obliterated most of t h e earlier remains, a b a d l y ruined wall h a d come to light t h a t might h a v e formed t h e s o u t h e r n side of t h e s t r u c t u r e , b u t no t r a c e of a corresponding a n t a could be recognized. F u r t h e r m o r e , t h e level of t h e floor h a d n o t been satisfactorily determined, a l t h o u g h we h a d m a d e a test of t h e stratification down t o h a r d p a n . Some of these problems h a v e n o w been clarified. T h e eastern end-wall has been disclosed (fig. 8), a n d deep trenching has revealed a good p a r t of t h e s t u r d y subs t r u c t u r e of t h e south wall. T h e ground plan has t h u s emerged, a n d w h a t we h a v e is a long, n a r r o w r e c t a n g u l a r building, with over-all dimensions of 1(5.55 m . f r o m east to west a n d 5.26 m . f r o m n o r t h t o south. I t shows no t r a c e of interior partitions, a n d if a western f r o n t wall ever existed, it has completely vanished. B u t a t this end is a two-coursed stone p l a t f o r m , ca. 1.80 m . wide, which has a distinct edge t o w a r d the street, projecting slightly beyond t h e face of the n o r t h a n t a , a n d which t e r m i n a t e s t o w a r d t h e east in a straight transverse line, as if it h a d once been laid against a

FIG.

6. — SOUTH

DOORWAY BY

OF

HOUSE

V I

F

CLOSED

F I G . 7. — A N T A

OK H O U S E O U T S I D E S O U T H

GATE

WAXT,

F I G . 8 . — E A S T W A U L OF " A N T A - H O U S E "

F I G . 9 . — L I K E OF E A R L Y W A L L B E L O W INSIDE SOUTH

GATE

STREET.

EXCAVATIONS AT TROY, 1938

213

cross-wall. T h e platform might then be the pavement of an original portico, flanked by antae to north and south. Owing to the repeated activity of Graeco-Roman builders in this particular spot, the evidence is unfortunately inconclusive, b u t it looks as if t h e building had continued to exist without a front wall, its façade entirely open toward the street. J u s t inside t h e " p o r t i c o , " well to the south of the longitudinal axis, lies a massive stone block, ca. 1.25 m. long a n d 0.50 m. wide, the purpose of which we could not determine, though it establishes clearly the level of the floor. T h e broad foundations of the south wall of the Bouleuterion, which run diagonally through the middle of our building, have destroyed any other furniture t h a t m a y once have been placed in the center of the room. In spite of the new information gained, we are still left somewhat uncertain as to the real character and purpose of this peculiar edifice t h a t stood just outside the main entrance to the citadel in the time of the Sixth Settlement. As pointed out in our preliminary report for 1935, it seems definitely not to have been an ordinary dwelling-house; and we therefore venture to repeat our conjecture made at t h a t time t h a t we have before us a public building of some kind. T h e situation itself, affording easy control over those entering or leaving the town, would be eminently suitable for a guard-house, or possibly even for an octroi-post, if such a suggestion should not appear over-fanciful; b u t I think it altogether more likely t h a t the whole area immediately outside the South Gate was a cult-place, and t h a t our structure should really be recognized as a public sanctuary, perhaps associated in some way with the monumental stone pillars erected against the south face of Tower Vli. Inside the fortress supplementary tests were made in m a n y places along the course of the main street ascending from the South Gate. I n its upper section, as seen in squares G 7-8, the accumulating rubbish must occasionally have been removed, while t h e adjacent ground was steadily rising, so t h a t in late VI and the beginning of V i l a the roadway had become a deeply sunk passage, bounded by high retaining walls on either side. During period V i l a much débris was allowed to remain, bringing about a rapid rise in the level of the route. T h e lower p a r t of the area, in squares G 8-9, was often the scene of construction, destruction and reconstruction, as a result of which the orderly sequence of layers has been considerably disturbed. Remains of two sets of massive walls t h a t certainly antedate the final phases of the Sixth Settlement constitute our most interesting discoveries of the season here. I t was observed t h a t underneath the narrow street which in period V i l a branched off in a westerly direction from the plaza just inside the gate, the lower p a r t of the heavy terrace wall " A , " supporting the platform in front of the Pillar House, continued on toward the south some 2.50 m. until it reached a well built corner. T h e position of the latter indicated t h a t our structure was not designed with reference to the Pillar House, but must be older. On the other side of the main roadway, beneath the west wall of House 700, a similar wall, " B , " came to light, turning eastward almost directly opposite the angle of " A " already mentioned. We may thus recognize the corners of two large buildings which stood some 2.50 m. apart, with the street passing between them. I t has not yet been possible to determine whether they belonged to separate houses standing within t h e citadel, or constituted together the actual gateway of an early fortification.

214

CARL W. BLEGEN

When the existing stone pavement of the street was cleaned and swept a plainly marked line of division appeared, extending obliquely to a length of ca. 7 m. along the western side of the roadway (fig. 9, p. 212). Further investigation soon revealed the eastern face of a huge wall, " C , " constructed of immense blocks of stone in the Cyclopean style. The top of it, as preserved, had been utilized to form part of the later pavement and the stones had suffered no little wear from the traffic that passed over them. At its south end the wall turns sharply westward and disappears beneath the structure we have called " A . " A second and equally massive wall, " D , " running parallel to " C , " came to light at a distance of 2.40 m. toward the east, beneath the doorway of House 700 and under the floor of the latter building. I t is more than 2 m. thick, and it terminates (or turns eastward) opposite the corner of wall " C . " There can be little doubt that we have here yet another gateway, preceding " A - B " in date, in the series of fortifications built to guard the southern entry to the acropolis during the time of the Sixth Settlement. Unfortunately these structures are for the most part so effectually covered and concealed by later monuments that it is not possible to examine them in detail; but it is obvious that they confirm our view of the long duration of Troy VI and its steady growth in power. Mr. Caskev devoted much time and attention to a careful study of the great Pillar House uncovered during the past seasons in squares F-G 8-9. B y means of numerous pits and soundings the foundations of the east and south walls were exposed in many places and traced almost continuously to their junction; and the ground plan of the huge structure has now been recovered virtually complete. The southeastern corner had suffered much more serious damage than the rest of the building: the superstructure was undoubtedly shaken down by the earthquake that ended the Sixth Settlement, and later, in period V i l a , even the foundations were to a great extent removed to make way for a new street now laid out, mounting steeply westward from the open space inside the South Gate. Farther toward the west the level of the street rose so high that the underlying stonework was for the most part spared. The immensely thick south foundation of the Pillar House (fig. 10) is peculiar in its construction. The western half of it is built regularly, provided at intervals of ca. 5 m. with relatively deep vertical off-sets, while the latter are lacking in the eastern half, which furthermore bends inward at an appreciable angle. Interesting results were gained by a stratigraphic test of the deposit contiguous to the southwestern corner, where it was possible to distinguish clearly two layers that had accumulated after the building of the wall. These strata contained no sherds belonging to the latest phases of Troy VI, as represented by the pottery from the floor of the Pillar House itself, from the cemetery, and from the "earthquake deposit," or as exemplified by the bulk of the wares found in House VI F and adjacent areas, but yielded only material that must be attributed to antecedent stages of the Sixth Settlement. Consequently we may take it as certain that our massive south foundation was erected, long before the Pillar House itself was planned, in one of the middle periods of Troy VI, when it may have formed part of an early system of fortification, such as those we have had reason to deduce from the evidence of the successive

F I G . 1 1 . — S U C C E S S I O N O F P A V E M E N T S IN C H A M B E R AT W E S T E N D O F PILLAK

HOUSE

216

CARL W. BLEGEN

gateways in square G 9. Subsequently, perhaps at the time the existing city-wall was constructed, we may assume that the Pillar House also arose, employing as its south foundation the earlier wall which had now become superfluous. An examination of the three small rooms at the western end of the Pillar House shed some new light on the history of the building (fig. 11). Originally the chambers were entirely open toward the pillared hall, forming three deep recesses or bays, each provided with a good pavement of solid stone slabs. Later a floor of crude brick was added above the stones, while the recesses still remained open. Next, coincident with the laying of a second floor of brick, a transverse wall of stone was built across the front of the bays, shutting them off as rooms, while a final readjustment is represented by a new pavement of irregular small stones applied on the bed of brick. A thin layer of debris rested on each pavement, indicating that they were actually used as floors. Regarding the purpose of the recesses at the inner end of the great hall, we have no very clear evidence; a comparison with the three shrines which often appear in the temples of Mesopotamia and other regions naturally suggests itself, but a discussion of this problem may be left to another occasion. In squares E - F 9-10, under Mr. Caskey's supervision, a considerable part of a small theater, or odeum, of Roman date was brought to light. I t was first discovered long ago by Professor Dorpfeld, who dug a trench across it from northwest to southeast, cutting through sections of the cavea, orchestra and stage buildings, and who called it Theater C. His trench had later become filled again, but its course could easily be recognized. The theater is an interesting monument of its kind, meriting a brief description here. Its builders took advantage of the contour of the ground (as they found it), placing their structure so that the orchestra fell in the hollow below and outside the ancient citadel, while the auditorium rose across the fortification wall which thus served in part as its sturdy foundation. Little or no mortar was used in the masonry, save perhaps in the outside retaining wall. The cavea (fig. 12) was divided by four narrow stairways into five cunei, or sections, that at each end being considerably broader than the three in the middle. The seats were made of limestone blocks, flat on top, with their risers hollowed out beneath a plain band. The upper part of the auditorium has disappeared, and in the lower only portions of some eight tiers of seats remain, most of them in a bad state of preservation. The eighth row of the middle section was occupied by a more elaborate seat of honor in marble: it seems to have consisted of five blocks, only one of which was found in situ. Two had fallen down, and have now been replaced, while the other two have apparently been shattered completely. The orchestra (fig. 13), comprising more than half a circle, was deeply sunk below the auditorium, separated from the latter by a wall, at one time revetted in marble, of limestone slabs set upright. Only the outermost stairway on each side continues down in a series of shallow steps to the orchestra level. The latter was once paved with thin marble slabs, most of which are now missing. Two flights of steps led up from the orchestra to the stage, the front of which was further embellished by five symmetrically spaced niches, presumably intended for statuary. The central recess, much larger and deeper than the others and terminating in an apse, was flanked to

F I G . L I . — T H E A T R E C IN SQUARES E - F 9 - 1 0 , C A V E A FROM SOUTH

F I G . 13. — T H E A T R E C . ORCHESTRA FROM W E S T

F I G . 14. — F A L L E N BLOCKS M A R K I N G D E S T R U C T I O N IN L A T E R O M A N T I M E S

218

CARL W. BLEGEN

right and left by a small rectangular niche; then came the stairway on each side, with another rectangular niche beyond it. F r o m the m a n y architectural pieces recovered it m a y be possible to reconstruct on paper the whole façade of t h e structure in which the Ionic and the Corinthian orders were combined. T h e epistyle bore an inscription in large, well cut letters; it is, however, too incomplete to give us the full n a m e of the benefactor a t whose expense t h e edifice was constructed. At the rear there were apparently five doorways leading out to the south, only four of which are now preserved. T h e scene-building has suffered so much damage t h a t m a n y important details of its construction and plan are probably irreparably lost. T h e a t e r C was evidently still in use in late R o m a n times, in the second and third centuries A.D. T h e date of its original erection can perhaps be more closely determined when the architectural, epigraphical and numismatic evidence has been adequately studied. T h e most formidable undertaking of our final campaign was t h e completion, under the direction of Miss D o r o t h y Rawson, of t h e excavation of the sanctuary she h a d previously discovered on the southwestern slope outside the citadel. At the end of t h e season of 1937, although t h e southern corner had been exposed in a trench and the eastern had been reached by a tunnel, a considerable strip along the southeasterly side of the precinct still remained deeply buried in débris, the removal of which was left to the present year. I t was first necessary to cut a broad shelf into a towering bank of loose rubbish thrown out from Professor Dorpfeld's excavations about Building V I M , and next to take away a deep underlying accumulation of firm earth which seems gradually to have covered this area in post-Roman times. After some weeks of such preliminary labor Miss Rawson finally reached the level of t h e sanctuary in its R o m a n phase, and continued on down into the earlier layers, with careful attention to t h e study of the stratification. Two layers were conspicuously marked by fallen stone blocks which had clearly been hurled down in some occurrence of violent nature, one such event probably to be attributed to the late imperial period (fig. 14), the other to t h e first century B.C. AS already noted in our report for 1937, it was evidently this earlier disaster t h a t led to the complete reconstruction of the sanctuary at a much higher level in t h e Augustan Age. Confirmatory evidence for the dating came to light this year. T h e southeasterly wall of the Hellenistic precinct proved to have been almost completely demolished in ancient times, no more t h a n a single pair of orthostates resting on their euthynteria having been spared a t the end adjoining the eastern corner. Throughout the rest of its course only the foundations had been left in place, where they had been m a d e to serve as t h e floor of a large drain t h a t carried off water from t h e high terrace on the northeast. T h e drain (fig. 15) had been built as an underground passage with stone walls, finished on the inner side, rough on t h e outer, covered by flat slabs of limestone, laid transversely. T h e channel was completely blocked with earth and rubbish, containing vast quantities of tile-fragments and potsherds, as well as a few badly corroded bronze coins which may possibly provide useful chronological evidence, if they can be cleaned sufficiently for identification. T h e construction of the drain cannot, in any case, be dated earlier than Augustan times and it should probably be placed later.

FIG.

15.—DRAIN

B U I L T OVER SOUTH-

EASTERLY W A L L OF

FIG. 1 7 . — E A R L Y

FIG. 18.—"SACEIFICIAL

TABLE"

IN S A N C T U A R Y TO THE E A S T

OF THE

SANCTUARY

W A L L S ALONG SOUTH-

E A S T E R L Y SLDE OF SA.NCTCAKY

FIG.

19.-STRATIFICATION SANCTUARY

OF K O A D

BETWEEN

AND W A L L OF T R O Y

VI

ALTAR

220

CARL W. BLEGEN

At t h e Hellenistic level no f u r t h e r m o n u m e n t a l remains were discovered when the easterly side of the precinct was cleared. A large flat stone slab, p a r t of which h a d already been exposed just to t h e east of t h e altar-podium, m a y h a v e continued to serve some purpose in connection with t h e cult at t h a t time; b u t when t h e whole of it was revealed we found it to be originally a construction of an earlier period. Of impressive dimensions, 1.90 m. b y 0.9,5 m., t h o u g h relatively thin, t h e slab h a d been set horizontally, s u p p o r t e d b y four or five somewhat irregularly spaced rough stones, which functioned as crude legs to raise the " t a b l e " some 0.50 m . above t h e ground. T h e space u n d e r n e a t h h a d become filled with soft loose e a r t h . T h e "sacrificial t a b l e " (fig. 16) s t a n d s m u c h too high, however, to be c o n t e m p o r a r y with t h e apsidal altar which lies below t h e Hellenistic podium, and it should probably be regarded as bearing some relation to t h e rectangular, stone-lined, ash-filled s h a f t incorporated b y t h e Hellenistic builders in their s t r u c t u r e a n d representing a n intermediate period. Digging deeper along t h e southeasterly side of t h e s a n c t u a r y (fig. 17), Miss R a w s o n b r o u g h t to view f u r t h e r extensive portions of t h e early precinct walls recognized in 1937, some of which are mentioned in our report for t h a t year. T h e latest wall t h a t m a y be associated with t h e apsidal altar can now be traced continuously from t h e extreme western angle (where there was probably a predecessor of t h e later gateways) to the opposite eastern corner, where it t u r n s n o r t h w e s t w a r d b e n e a t h t h e high Hellenistic retaining wall. Along t h e southwestern side it was p a r t l y demolished in t h e f o u r t h c e n t u r y B.C., when Well A was cut through i t ; f r o m this point it curves eastward a n d t h e n continues in a straight line t o w a r d the n o r t h e a s t . I t was trimly constructed, chiefly of r a t h e r small stones, fitted together in polygonal style, with a well finished exterior face. A low sixteen-sided base of poros, with concave profile, still s t a n d s outside t h e wall, near t h e eastern corner, a n d it is likely t h a t a g a t e w a y once p e r m i t t e d access to the s a n c t u a r y f r o m the southeast a t this place, t h o u g h most of t h e evidence was destroyed when t h e enclosure was recons t r u c t e d shortly before 300 B.C. I n its next earlier period the precinct was considerably smaller, surrounded b y a h e a v y curving wall, built of large unworked stones. This, too, was more carefully laid on its o u t e r face t h a n t o w a r d t h e interior, a n d , like its successor, m u s t h a v e been intended p a r t l y as a support for t h e higher ground within. T h e only piece of this wall now preserved bends eastward f r o m Well A a n d t h e n n o r t h w a r d to a point behind t h e sacrificial table already mentioned, rising gradually in level as it proceeds. N o t r a c e of it came to light in t h e n o r t h e r n q u a r t e r of t h e sanctuary, where it probably stood so high t h a t it h a d to be removed when the ground was levelled in Hellenistic times.

EXCAVATIONS AT TROY, 1938

221

The oldest enclosure for which we have evidence is represented by an irregularly curving heap of stones, definitely edged toward the outside, which can be recognized for some distance, forming an arc about the southern side of the apsidal altar. Between it and the latter is a narrow cobble-stone pavement, on which lay a deposit of black earth and ashes, containing gray ware of a pre-orientalizing period of Troy V I I I . Altogether a good deal of ceramic material and other objects (fig. 18) was brought to light in the course of our work along the southeasterly side of the precinct, and Miss Rawson, who was assisted for a short period by Dr. Hill, has recovered a substantial body of evidence bearing on the early history of the shrine. For its later

F I G . 2 0 . — G E N E R A L V I E W OF U P P E R AND L O W E R S A N C T U A R I E S FROM S O U T H

history valuable supplementary observations were made in an investigation of the area that lies between the precinct and the fortification wall of Troy VI. This narrow space was completely filled to a depth of more than 6 m. with earth and debris, in which successive layers are distinctly visible (fig. 19). A study of the stratification has disclosed the series of changes and modifications undergone by a street, which, through a long period of years, provided an ascent toward the west in this place behind the sanctuary. In the fourth century B.C. its level was low, and the rear wall of the temenos was a free-standing construction, only one course thick, meant to be seen from front and back. In the course of time the level rose until the weight of the accumulating rubbish began to threaten the precinct wall, or perhaps actually to thrust it out of line. An almost complete reconstruction on a far more massive scale was then undertaken, and the great retaining wall which is still so impressive today came into being. On the same occasion the fill behind was greatly increased and the roadway correspondingly raised. The process of accumulation then began again in the street, and eventually the upper part of the reconstructed

222

CARL

W.

BLEGEN

wall in its turn came to be endangered. In order to protect it a new terrace wall was built behind it, facing northeastward toward the street. Once more, however, the process of filling up resumed its course and at length the debris rose over the new terrace wall, too, which ultimately came to be buried more than 1 m. beneath the final street-paving of Roman times. In the whole deposit some six consecutive pavements of this kind can be seen; and with the help of the pottery and a few bronze coins recovered here we hope a fairly precise dating of the sequence of layers may be secured and their exact relations with the later periods of the sanctuary established. During the past season Miss Rawson was also able to continue and conclude her exploration of the smaller temenos (fig. 20), lying to the south westward of the other at a lower level, in squares A 8-9. She had previously discovered here the bases of two altars, apparently of Hellenistic date, and parts of the foundations of the two lateral enclosing walls of the precinct, which in our report for 1987 were erroneously attributed to Roman times. The whole of the temenos has now been cleared. Backed up against the higher sanctuary, it has a peculiar trapezoidal shape, with its shortest side toward the southwest, where there was probably an entrance. The en„, , r closing wall has completely 1>IG. 2 1 . — L O W E R

I EMENOS WITH ARCHAIC AND I J O W E R

vanished, but its heavy substructure, built of large blocks of soft limestone, still remains intact, and it is unquestionably a work of the early Hellenistic Age. To this same period we may no doubt safely assign the construction of the larger of the two altars, the long rectangular base of which lies to the south of the other. No architectural pieces of the altar itself were recovered, and its original appearance can only be conjecturally restored. An examination of the smaller altar, which evidently continued to be used in Hellenistic times, showed that it was really built in a much earlier epoch. It is not merely the foundation for a (missing) superstructure, though its top was certainly HELLENISTIC ALTARS

EXCAVATIONS AT TROY, 1938

223

utilized as such t o w a r d t h e end of t h e f o u r t h c e n t u r y B.C., b u t it is t h e a c t u a l altar itself, virtually complete, as it was erected in t h e archaic period (fig. 21). T w o long a n d t w o short slabs of limestone, smoothly dressed a n d carefully fitted, were set on edge t o f o r m t h e box-like s t r u c t u r e , s t a n d i n g on a well finished projecting e u t h y n teria course, which in t u r n was laid on a rough f o u n d a t i o n . T h e c o n t e m p o r a r y ground-level of t h e s a n c t u a r y was m a r k e d b y a t r o d d e n e a r t h p a v e m e n t , a b o u t even with t h e b o t t o m of t h e e u t h y n t e r i a blocks. T h e n o r t h w e s t e r n p a r t of t h e precinct was excavated down to this early floor, a n d t h e intervening deposit between it a n d t h e Hellenistic level, nearly 1 m . deep, yielded a rich h a r v e s t of s h a t t e r e d p o t t e r y , f r a g m e n t s of t e r r a c o t t a figurines, and o t h e r objects. A n i n t a c t archaic female b u s t is p r o b a b l y t h e m o s t i m p o r t a n t single item (fig. 22). T h e b u l k of t h e p o t t e r y consists of i m p o r t e d C o r i n t h i a n ware (fig. 23), b u t R h o d i a n ware is also c o m m o n (fig. 24). M o s t of t h e vessels are small—cups, bowls (fig. 25), saucers, plates, aryballoi, etc. —doubtless b r o u g h t t o t h e shrine as votive offerings. I n t h e t i m e available t o us it was not possible t o work over all this m a t e r i a l a d e q u a t e l y , b u t already we h a v e been able to p u t together sufficiently for restoration in plaster some seventeen p o t s t o a d d t o our inventoried collection. T h e discovery of so g r e a t a mass of i m p o r t e d Corinthian a n d R h o d i a n p o t t e r y is n o t w i t h o u t significance for our s t u d y of t h e history of T r o y V I I I . I t is n o w clear t h a t worship was actively carried on in t h e t w o adjoining sanctuaries on t h e southwestern declivity of t h e hill t h r o u g h t h e eighth, seventh a n d six centuries B.C., a n d Miss R a w s o n ' s researches here h a v e done m u c h t o fill t h e g a p in our information concerning t h e life a n d t h e culture of t h e o c c u p a n t s of t h e site in t h e period of t h e E i g h t h Settlement. I n spite of t h e a b u n d a n c e of t h e material, however, a n d t h e occurrence of several inscribed vase-fragments, no conclusive evidence c a m e t o light to settle t h e question as t o t h e i d e n t i t y of t h e shrine. T h e n u m e r o u s t e r r a c o t t a figurines f o u n d in 1936, 1937 a n d 1938 are indecisive, a n d we h a v e no epigraphical records t o help us. Accordingly b o t h t h e lower a n d t h e u p p e r sacred precincts m u s t provisionally still remain nameless, although t h e f u r t h e r s t u d y of our d a t a m a y y e t afford some d e t e r m i n a t i v e clue. T h u s , for example, t h e discovery within t h e u p p e r s a n c t u a r y of skeletal remains a n d t e e t h of m a n y large carnivorous animals, such as t h e bear a n d some m e m b e r of t h e leopard family, m a y n o t be w i t h o u t i m p o r t in this connection. T h e identification of t h e bones in question we owe to M r . N . - G . Gejvall, who spent a m o n t h a t T r o y as a m e m b e r of t h e expedition a n d who continued t h e imp o r t a n t s t u d y he h a d b e g u n in 1937. A v a s t s u p p l e m e n t a r y accumulation of animal bones, in which all periods were a b u n d a n t l y represented, was subjected t o his methodical, p a i n s t a k i n g examination. T h e relative (and for some areas t h e actual) n u m b e r s of individuals of all species t h a t could be differentiated were recorded as carefully as possible, a n d all evidence relative t o t h e domestic a n d wild animals, birds, fish, a n d shell-fish, k n o w n in each of t h e successive settlements, was n o t e d and t a b u l a t e d . T w o full representative collections of selected typical bones, comprising specimens of all t h e periods f r o m T r o y I t o T r o y I X , were formed, a n d all notable pieces were n u m b e r e d a n d inventoried. A m o n g t h e more r e m a r k a b l e items m a y be m e n t i o n e d t h e m a j o r p a r t of t h e skull of a large bear (from T r o y V I ) ,

F I G . 21. — A R C H A I C F E M A L E H E A D A N D B U S T

F I G . ¿ 3 . — C O R I N T H I A N A R V B A L L O I VBOM L O W E R

F I G . 2 4 . — R H O D I A N AND E A S T G R E E K W A R E FROM L O W E R

FIG.

-

IÌOWL OF E A S T G R E E K

PROM L O W E R

TERRACE

STYLE

TEMENOS

FIG.

25B

TEMENOS

EXCAVATIONS AT TROY, 1938

225

which M r . Gejvall succeeded in fitting together from innumerable small fragments, and p a r t of t h e jaw of a lion or a leopard. All this material, of great potential value for an understanding of t h e economic and cultural setting of ancient Troy, will be f u r t h e r studied by Mr. Gejvall, who has undertaken to treat it fully in a section of our final publication. H u m a n skeletal remains antedating the Hellenistic period have come to light with extraordinary rarity a t Troy. Almost no ancient site of corresponding size and importance has yielded so little in the way of the actual physical remains of its inhabitants. In spite of persistent searching Schliemann and Dorpfeld were unable to find a real cemetery; and save for the discovery of a burial ground of T r o y VI, containing numbers of ash-filled cinerary urns, our own experience has been t h e same. Here and there about the site graves of children have occasionally been uncovered, and in a few instances we have encountered adult human bones, in casual single fragments or in larger groups. All these remnants, among which each of the m a j o r periods a t T r o y is a t least barely represented, were of course collected and preserved; and having been transported to Istanbul last July, they have now been carefully examined and studied by Mr. John L. Angel, whose collaboration as anthropologist to the expedition we were fortunate enough to secure. For the facilities placed a t his disposal in carrying out this work we are greatly indebted to the Director, M r . Aziz Ogan, and the other officials of the Istanbul Museum, as well as to Dr. Bittel, Director of the local branch of the German Archaeological Institute. Notwithstanding the incomplete and fragmentary state of the material, M r . Angel was able to make a series of interesting, detailed observations, which will have to be duly considered in any a t t e m p t to interpret the anthropological background of T r o j a n history. We are counting on M r . Angel to present in full the results of his investigations in a chapter of our projected work on the excavations at Troy. As already mentioned, our chief occupation during the season of 1938 was a painstaking study in detail of the vast quantities of pottery recovered in earlier campaigns. All t h e members of the staff took p a r t assiduously in this time-consuming task, which of necessity concerned itself largely with minutiae; in the course of it m a n y thousands of separate lots of sherds were patiently scrutinized, their particulars analyzed and recorded. Notable fragments were numbered and listed for all the periods. Miss Marion Rawson completed t h e long inventory of Early Helladic and related sherds; and similar lists were drawn up for Middle Helladic pieces and for the more important fragments of Mycenaean ware. Much of the information thus laboriously gained will be of use in the preparation of the full account we shall ultimately a t t e m p t to give of the growth and development of ceramic art in Northwestern Anatolia. Throughout the campaign our force of menders, usually six in number, was kept busy, joining, repairing, and completing in plaster all vessels t h a t we judged worthy of restoration. T h e n u m b e r of pots thus recovered, which were entered in the inventory and described by Miss Marion Rawson, totalled 233. Deserving of special mention are some large storage jars of Troy I I (fig. 26) and Troy IV (fig. 27), a huge stemmed crater of T r o y VI (fig. 28), and, belonging to the same period, a series of capacious stirrup vases decorated in imitation of t h e Mycenaean style (fig. 29); also a group of pots of an early phase of Troy V H b , and the

F I G . 26. —STORAGE J A R OF T R O T

II

P I G . 27. —STORAGE JAR OP T R O Y

F I G . 2 8 . —LARGE CRATER OF T R O T

VI

FIG. 2 9 . — T H R E E LARGE STIRRUP VASES, T R O T

VI

IV

EXCAVATIONS AT TROY, 1938

227

imported Corinthian and Rhodian vases from the lower sanctuary on the southwestern slope. Miscellaneous objects of metal, stone, bone, terracotta, etc., were relatively few in 1938: among t h e 126 items catalogued by Miss D o r o t h y Rawson t h e most interesting are three archaic terracottas, namely a head, a kore, and a well preserved bust of a female figure, the facial and other details of which were rendered in a lively m a n n e r in dark paint (fig. 22). Our study of the similar objects found in former years was greatly facilitated through t h e kind helpfulness of t h e Director, M r . Aziz Ogan, and his staff, permission being generously granted to have t h e m all brought back from the Istanbul Museum to Troy, where we were able to spread them out and examine t h e m at leisure on the tables in our workrooms. I t was thus possible to obtain a general survey of t h e whole mass of material, and to compare objects of the same categories from all the seven campaigns of digging, and our task of classification and correlation was thereby enormously lightened. I n m a n y instances regroupings of objects could be made for photographic purposes, and "closed g r o u p s " from certain areas, the excavation of which had occupied more t h a n one season, could be assembled completely. All the implements and utensils of bone were inspected by M r . Gejvall, who identified the material more exactly whenever possible. In sorting out the m a n y thousands of objects t h a t had to be handled and in subsequently repacking t h e m in their proper boxes, Mrs. Caskey rendered services of signal usefulness. At the end of our campaign, early in July, all the inventoried pots, the miscellaneous objects, t h e h u m a n skeletal remains, and the animal bones required for further study were shipped to the Museum in Istanbul. Following instructions from the Ministry of Education, our more important sherds, in 200 cases, were transported to (Janakkale for temporary storage in the local museum, together with one of the collections of animal bones, packed in ten boxes. At Troy itself we have left in our spacious workrooms a complete and extensive series of potsherds, arranged in chronological order and representing abundantly all the periods of settlement; also several large pithoi and numerous heavy stone objects, such as querns, grinders, pivot-sockets, etc. All portable inscriptions on stone, our few fragments of sculpture, and some of the smaller architectural pieces have likewise been deposited here. I t is our hope t h a t the buildings may continue to stand as a small museum a t the site, where the interested visitor who desires more t h a n a view of the citadel and its ruins m a y find some of the other evidence of the changing conditions of life in the successive settlements. I n our concluding campaign we had again the benefit of much assistance and advice from m a n y friends. We take pleasure once more in expressing our thanks to M r . and Mrs. Francis Bacon and to M r . and Mrs. Godfrey Whittall of Qanakkale, who through all these years have ever been ready to lend a helping hand; also to Director K. Bittel, Head of the German Archaeological Institute in Istanbul, for good counsel and aid, as well as for his kindness in extending t h e hospitality of t h e I n s t i t u t e building to members of our staff; and to Dr. H. H. von der Osten of Ankara, who has likewise earned our cordial regard. We are grateful to the Government of the Turkish Republic for the permission to continue and to complete our excava-

228

C A R L

W.

B L E G E N

tions at Troy. H . E., the Minister of Education, Mr. Saffet Arkan, has had the interests of our expedition at heart, and we have always been able to count on the friendly understanding and cooperation of Dr. Hainit Kosay, Director of the Archaeological Service of the Ministry. M r . Aziz Ogan, Head of the Istanbul M u seum, has already been mentioned several times in this report, and we have a b u n d a n t reason for offering our warm thanks for his m a n y courtesies and good offices. We are also glad to have this opportunity to acknowledge our obligation to M r . L&tif Oktem, representative in Qanakkale of the Ministry of Education, and particularly to M r . H a y d a r Sumerkan, our Commissioner, who was unfailingly courteous, helpful, and ready to assist us in times of need. Our excavations a t Troy have now been brought to their end. Seven years of work in the field and of preoccupation with T r o j a n problems have given us a sense of intimate familiarity with a site, the peculiar charm of which has grown steadily stronger, and it is with a feeling of deep regret t h a t we are taking our leave. B u t the main objective of our undertaking, as we set it before us in 1932, has been attained. T h e whole sequence of layers has been independently re-examined. One by one each s t r a t u m has been intensively studied, wherever it could be "isolated," usually not in a single place, b u t in several areas; and every one has yielded a generous a m o u n t of " c e r t i f i e d " material. New architectural monuments have been revealed for several periods, and with our formidable collections of pottery and miscellaneous objects of metal, stone, bone and terracotta, we have a fresh and voluminous set of h u m a n documents to illustrate the progressive evolution of culture and to fix a series of landmarks for the charting of otherwise unrecorded history. I n our second major aim, too, we may claim at least a partial success. Although burials of earlier periods eluded us, as they had our predecessors, a persevering search for tombs led to the discovery of a cemetery belonging to the Sixth Settlement, in which, it was definitely established, the practice of cremation prevailed. T h e site of T r o y has not yet been exhausted. M a n y areas remain t h a t still invite examination, and no one can predict what surprises future exploration may bring to light. B u t for the moment, and in the present state of archaeological knowledge and method, we believe we have done enough. If t h e results of our researches contribute toward the elucidation of some of the problems t h a t confronted us, we shall feel t h a t our labors have been adequately compensated. In the meantime we have left a b u n d a n t opportunities for further investigation, verification, and doubtless rectification, for successors who may some day, possessed of wider knowledge and more perfected methods, desire to resume the excavation of ancient Troy. CARL W . U N I V E R S I T Y OF CINCINNATI

BLEGEN