Printer Friendly

Sarepta. (4 vols.).

Sarepta I-IV are the first in a series of anticipated final reports on excavations at Sarafand/Sarepta, a low mound on the seashore fifty kilometers south of Beirut. There, from 1969 to 1974 a team from the University Museum of the University of Pennsylvania led by J. B. Pritchard pursued a long-standing objective in the archaeology of the eastern Mediterranean: the excavation of a coastal urban settlement of the late second/ early first millennium B.C. that could provide the chronological key to Late Bronze and early Iron Age Phoenicia. However, the outbreak of the Lebanese civil war in 1975 brought the field work to a premature close. Later, fighting in Beirut hampered efforts to study the finds stored in the National Museum and delayed the printing of the final reports. In the face of such obstacles, it is remarkable that Professor Pritchard and his team have produced these volumes at all, much less in such full detail and so quickly after the end of excavations.

These reports present the results of the 1971-1974 excavations in the two main soundings on the tell at Sarepta, designated Area II, X and II, Y. (Area I, the Roman port one half kilometer south, was investigated in 1969-70.) Sarepta I and II detail the stratigraphy and architecture of the two soundings in conjunction with typological and statistical analyses of the local Bronze and Iron Age ceramics from each. Sarepta III is a thorough typological and chronological study of the Bronze and Iron Age imported wares, chiefly Aegean and Cypriot, from Area II, X. Sarepta IV comprises a descriptive catalogue of small finds from Area II, X, with extended treatments for the inscriptions and post-Iron Age imports and an additional corpus of pottery from that same area.

In what follows, I treat these four books as virtually a single work, since there are throughout frequent cross references between volumes citing descriptions of finds and stratigraphy as well as chronological evidence to support cultural and historical interpretations. Vol. I defines the typology of local ceramics later applied and extended in vols. II and IV. Vol. III provides the illustrations, descriptions and datings of the Aegean and Cypriot imports on which vol. II directly depends. The artifacts catalogued in vol. IV are fitted into the stratigraphic and chronological framework established in vols. II and III, using the vessel typology introduced in vol I. Such close interrelatedness is in the nature of multi-volume site reports, so it is useful to have all four at hand when using any one of them.

In technical respects all four volumes are very well done. There are relatively few typographical or other printing errors; those that do occur will cause no undue confusion to attentive readers. The line drawings throughout are well reproduced, though the small format (17 x 24 cm) gives the plans, plates and appendices a somewhat cramped look. In vols. I and II a few of the printed field photographs lack sufficient contrast and the consequent loss of detail makes them difficult to read (note that fig. 9 in vol. II is inverted). The distribution graphs in the appendices of vols. I and II might have had greater visual clarity had they been rendered as actual figures rather than as typeset replicas of the arrays of "X"s in the original computer printouts. Still, they make their point. The map in vol. III (pp. 177-79) has three misplaced sites: Gezer (no. 12) is closer to the coast than shown; Mesad Hashavyahu (no. 22) and Tell abu Hawwam (no. 31) are coastal, not inland sites; and there are two no. 43s. The bibliographies in all of the volumes are updated only through 1981. This is apparently due to the long delay (quite beyond the authors' control) between submission of the finished manuscripts and appearance of the printed books (see the note at the head of the bibliographies in vols. I, II and IV). Finally, there is an annoying omission from the ceramic descriptions in vol. I: Munsell color designations are expressed only as numbers without the accompanying color names. This is an inconvenience, at least for this reviewer who cannot immediately recall that, for example, "7.5YR 7/4" is "pink" and "7.5YR 5/2" is "brown." These are merely quibbles and only reflect economies demanded by the skyrocketing costs of academic publishing, especially for complicated, highly technical archaeological reports such as these.

There are, however, criticisms that are more than quibbles. As detailed and accurate accounts of excavations in the soundings at Sarepta, these volumes are important contributions to our knowledge of the character and relative sequence of urban Phoenician material culture in the late second and early first millennium B.C. As keys to the absolute chronology of Late Bronze and early Iron Age coastal Phoenicia (and ultimately of early Phoenician foundations in the west), they do not quite open the lock. This is due less to inherent shortcomings of the volumes themselves (though these do exist) than to the still unresolved question of the chronologies of Aegean and Cypriot ceramics in Levanthine contexts.

Import chronologies must be used to determine the absolute datings of the levels at Sarepta; radiocarbon plays only a minor role (on which, see below). The problem is which import chronology to adopt. In Sarepta I (p. 260), II (p. 73), and III (p. 37, 45-46, n. 153) the authors individually are explicit in their acceptance of the "standard" datings proposed originally by Furamark (1941) for Mycenaean, by Astrom (1957, 1972) for Cypriot Middle and Late Bronze Age, and by Gjerstad (1948) for Cypriot Iron Age wares. Of course one may follow the proposed datings, but the accumulated weight of criticisms of the original formulations are sufficient, I think, to warrant caution. The authors themselves, while acknowledging difficulties and even citing critical bibliography (e.g., vol I, p. 260, n. 594, pp. 273-74, nn. 722-28, p. 407, n. 234; vol. II, p. 73; vol. 111, p. 37, n. 77, p. 46, n. 153), nonetheless proceed to date the Sarepta levels on the basis of these questionable chronologies. At several points in vol. I Anderson confronts the question squarely (e.g., pp. 274, 366, 417, n. 273), but ultimately sets it aside as "too involved for detailed treatment in the present study" (p. 440, n. 234). In fairness one cannot criticize this work for failing to be what is not intended to be, namely, a solution to the dilemma of import chronologies for the Levant; but the implications of Anderson's choice to limit the scope of his enquiry are left implicit rather than stated plainly. Therefore, those who use these volumes should be aware of the probable need to adjust the absolute datings of the Sarepta levels in accord with their own judgments about the absolute chronology of Aegean and Cypriot imported ceramics in the mainland Levant. (Anderson, p. 367, and Koehl, p. 27, n. 12, in fact recommend such a course.)

In support of the proposed datings in vol. II, Khailifeh offers ostensible radiocarbon confirmation of the absolute datings already established by the imports. This is a discrepancy here, however. The same radiocarbon sample (P. 1946, MASCA date: 1290 [+ or -] 52 B.C.) is cited in support of two different absolute datings for two different levels. First, the sample is used to date Level 7 in Square II-A-9 to the mid-14th to late 13th century B.C. (p. 102). Then the same sample is adduced as evidence for dating the next layer above, Level 6, to the late 13th century to the mid-12th century B.C. (p. 113). In this instance (p. 1946 is specifically said to have come from a "posthole 0.45 cm deep in floor of Room 74, II-A-9, N of W. 498" (sic, W. 489 must be meant, see. pl. 7). The origin of this confusion is explained in the 1975 preliminary report. There Pritchard makes clear that the posthole in Room 74 (whose charred contents provided the datable carbon for sample P. 1946) was assigned to an upper layer (level 6) on very tenuous grounds. The posthole was actually found in the layer below (level 7) but was assumed to have been dug from above, although this could not be demonstrated stratigraphically (Pritchard 1975: 89). To compound the confusion, Appendix I of vol. IV, which lists the radiocarbon samples collected at Sarepta and their respective datings, locates P. 1946 in "II-A-9, level 7" and gives its calibrated date as "1300 [+ or -] 60 B.C." The slight variation in calendar dates may be only an editorial oversight, and the uncertainty of the sample's provenance does not in itself invalidate the dating of these levels putatively established by the imported. pottery. This confusion does require, however, that this radiocarbon date be discounted as independent corroboration of the proposed absolute dating. Moreover, even if the stratigraphy of the sample were secure, a single radiocarbon date is a precarious peg for an absolute chronology.

The minor problem of the posthole highlights a major problem in the relative stratigraphic sequence, a problem originating in apparent shortcomings of excavation technique at Sarepta. The discussion of field method in vol. I (pp. 38-41) defines the smallest unit of excavation, recording, and artifact collection as a "level." Within structures, the upper and lower limits of "levels" were defined by floors." This suggests (I think) that occupational accumulations on or just above a floor and sub-floor fills contemporary with the next floor above it were, by definition and in practice, recorded as belonging to one "level." Thus artifacts from stratigraphically and chronologically distinct deposits at Sarepta could have been (in fact, probably were) regarded as coming from a single "level." The published plans and sections in vols. I and II offer no help in sorting out these mixed deposits. In the section drawings the floors defining the levels are not specifically labeled, though by comparisons with the corresponding descriptive text, it is possible sometimes to work out which lines probably are the floors. However, the drawings contain other unlabeled deposits which, though they cannot be floors," are not "levels" either. Successive architectural plans depict only wall lines and kiln structures and give no absolute elevations on floors within rooms. Anderson and Khalifeh rank the certainty of attribution of "levels" to the respective phases within each sounding on an ascending scale of "possible," "probable," and "certain," (see appendix A, both volumes); but only in vol. I is this information successfully integrated into discussion of the stratigraphy. In neither case do these evaluations mitigate the failure to separate layers properly. There is, in short, no way of disaggregating the published data from the "levels" into which they have been erroneously conflated. This methodological flaw has significant, negative implications for the integrity of the stratigraphic sequences set forth in Sarepta I and II.

These stratigraphic sequences are crucial to the arguments of Sarepta I and II, since their primary aim is the establishing, within the framework of the site stratigraphy, of a relative ceramic chronology for the local wares by means of rigorous typological classification and quantification. Vols. I and II treat roughly the same amount of material from excavation areas of identical size for, although Sounding X is much larger than Sounding Y, vol. II deals only with a 100 [m.sup.2] subsection of the latter sounding, selected for its intact stratigraphy. The procedures followed in each volume are identical: establish ceramic types, tabulate the numbers of each type by level, then derive statistical profiles delineating first appearance, common occurrence, and residuality for all ceramic types in successive levels. Only such rigorous classification and quantification, Anderson argues, can provide an objective and accurate measure of the variability and continuity of ceramic forms, an important index of cultural and historical change hitherto assessed only with qualitative and often very subjective criteria (vol. 1, pp. 42-43). This statistical rigor is extended to inter-site comparisons and relative datings for which Anderson and Khalifeh expressly reject the "traditional" approach of seeking stylistic parallels between individual forms or broadly defined "ceramic horizons" (vol. I, p. 43, n. 64; vol. II, pp. 5-6).

Here, a gap opens between the stated aims and the actual execution of the authors' research design. In part this is due to the nature of the comparative evidence available. To find formal parallels for the local wares from Sarepta, Anderson and Khalifeh must turn to unquantified assemblages from other sites in the mainland Levant: Megiddo, Hazor, Samaria, Tell Abu Hawam, Tyre. The incommensurability of the material from neighboring Tyre is a striking case in point (vol. I, p. 366, nn. 288-89, p. 439, n. 222, p. 443; vol. II, p. 163-64). Because the Tyre corpus is not quantified in the same way as that from Sarepta, it is not directly comparable except on the basis of individual examples regarded on stylistic grounds or within broadly conceived "ceramic horizons." This is precisely the procedure that Anderson and Khalifeh wish to avoid, but they have no alternative because at no other sites has comparable material been analyzed with similarly exhaustive quantitative methods. Anderson clearly appreciates the contradiction, but is unable to resolve it (vol. I, p. 43, n. 65; cp. p. 366, nn. 8-9).

Such are the pitfalls of innovation. These shortcomings, however, are not grounds for rejecting a quantifying approach altogether. At the time that the authors undertook their work (the mid-1970s), quantification in Mediterranean archaeology was still a novelty. Its pioneering application at Sarepta marked a significant advance in the analysis of archaeological materials in this part of the world. For such a contribution Anderson and Khalifeh plainly deserve praise. Happily, over the past decade there has been a healthy trend toward more quantification (see Tomber 1990 for an assessment) though not yet at sites immediately relevant to Sarepta (but cf. Schwartz 1988). The larger question of the appropriate methodology and ultimate utility of quantification in archaeology - how best to do it and whether it yields more and better data than simpler "traditional" methods - is a separate issue not to be resolved in the space of this review.

Given the potential for quantitative studies to revolutionize our understanding of ceramic change and continuity in Phoenicia during the Late Bronze and Iron Ages, it is all the more regrettable that a better-refined stratigraphic and material cultural sequence is not forthcoming from Sarepta. A carefully separated sequence, enhanced by rigorous quantitative analysis, would perhaps have yielded a more precise relative chronology of local wares. At present, the relative sequences in Area II, X and Y are no more precise than the instruments used to calibrate them, viz., the "standard" import chronologies. In Area II, Y, the tightly phased sixteenth-to-tenth century period (100 years or less per stratum) contrasts strikingly with much broader separation of everything after the tenth century (175-200 years or more per stratum). While this may record actual slower rates of change after 1000 B.C., it is more likely a direct reflex of the asymmetrical periodization in the import chronologies used to obtain the absolute dates. There is a similar pattern in Area II, X: 100-year divisions from the sixteenth to eleventh century B.C. are followed by much expanded intervals for the tenth century and later. Period VII, for example, is 225 years long (ca. 1025-800 B.C.), Period VIII, 450 years long (ca. 800-350 B.C.) (vol. II, pp. 140-56). In the case of Period VIII, there must have been some lapses in the field recording as well. This phase is almost completely missing from the published section drawings. The relevant segments in all but one drawing are labeled "Not Recorded," and the corresponding discussion in the text belies a degree of uncertainty arising from inadequate documentation (pp. 46-52). Whatever the causes, the lack of finer discriminations in the relative sequences for the two soundings in the early first millennium B.C. layers makes it impossible to bring the Sarepta materials to bear on the question of the relative ceramic chronology in the Phoenician homeland during the ninth-eighth centuries B.C., the period of great overseas expansion.

The foregoing are criticisms of analyses of data in Sarepta I-IV. There are, as well, several points in these volumes at which the interpretations of data might have been sounder or pressed further. This is particularly the case for the processes of ancient exchange, which the authors term generally "trade." Running through Sarepta I-IV is the thread of an implicit (and I believe incorrect) assumption that ancient exchange was an exclusively "economic" activity divorced from its social and cultural context. The wide range of non-economic forces bearing on ancient systems of exchange (religious and political ideology, considerations of social status, obligations of kinship and clientship, etc.) are completely ignored. This is most glaring in Koehl's conclusions to vol. III in which absolute numbers of imported sherds are translated directly into levels and trends in overseas trade at Sarepta. Moreover, the sherd counts (there are only 259 imports from all of Sounding X) are compiled irrespective of reliability or representativeness of the excavated sample or of the soundness of the contexts of individuals finds pp..25-26 and ch. IV).

This assumption likewise underlies Pritchard's study of the much later imported lamps from Area II, X. Based on a second-third century A.D. gap observed in the imported lamp sequence, he posits a gap in Roman settlement on the tell, allowing only that the remaining five untyped pieces in the assemblage could fill the missing centuries (pp. 194-95). What he does not acknowledge is the possibility that the "gap" could have resulted from one or a combination of causes unrelated to settlement hiatus at Sarepta. Even considering the question within the limits of Pritchard's own assumptions, the data he cites are incomplete. He omits evidence of third A.D. imports at Sarepta mentioned in the discussion of "Roman and Late Roman Wares" in vol. II which includes several vessels of third-century A.D. date (pp. 224-30: nos. 142, 145-48, 151).

Finally, nowhere in these volumes is there a comprehensive assessment of the significance of the archaeological discoveries at the site in light of Sarepta's place in the wider ancient Mediterranean world. In vol. II Khalifeh at least attempts something like this in his "Concluding Remarks," but these are incomplete and unsatisfying. The excavated remains of Sarepta reported in these volumes are at variance with what might have been expected from a presumably sizable Phoenician city. While the historical references to Sarepta reviewed in vol. I (pp. 35-36) suggest an extensive and important urban center, what the Pennsylvania excavators found in two widely separated areas on the mound was evidence for an uninterrupted series of industrial installations (chiefly pottery kilns) and some rather unimpressive domestic structures. Even the "shrine" in II-A/D-4/9 was a modest affair. This contrast between actuality and expectation, even if not explicable from the data at hand, might at least have been noted for its very inexplicability. Sarepta was "presumably sizable" but it is difficult to gauge the site's overall size from these reports. The published site plan with its superimposed 5 m x 5 m grid suggests that the mound itself might be 1.5 to 2.0 hectare in extent, but as Anderson observes there are 500 meters of unexplored ground between the tell and the Roman port to the south (vol. I, p. 5 1, n. 14). Investigation of this terra incognita must await possible renewed excavations at the site, as must the opportunity to collect the botanical and faunal remains apparently overlooked in the earlier campaigns.

These criticisms should in no way diminish the importance of the results of the excavations at Sarafand embodied in Sarepta I-IV. These reports have contributed immeasurably to our knowledge of coastal Phoenicia in the hitherto ill-documented period of the late second and early first millennium B.C. The continuity of ceramic and architectural traditions revealed at the site provides archaeological evidence for occupation on this part of the Canaanite coast unaffected by the calamities of the twelfth century B.C. Furthermore, elements of this continuity suggest that, at least in cultural terms, the Late Bronze Age in Phoenicia endured longer than in the more southerly parts of the Levant. These are important implications that should be further explored. There is hope that they will be since these volumes are not the final word on Sarepta. We can expect more such comprehensive and extremely useful reports on other aspects of the site not dealt with in the first four volumes. The Iron Age shrine (findspot of an inscribed ivory plaque containing the first unequivocal mention of the deity Tanit in homeland Phoenicia), the score of Late Bronze/Iron Age kilns (which yielded literally thousands of vessels and wasters) and the Roman port (reported preliminarily in Pritchard 1971) all merit extended treatments in their own right. There is also a promised volume on coins by Marcus Arguelles (forthcoming as of 1984, see vol. IV, p. 4). Sarepta I-IV, then, constitute only an initial, but nevertheless substantial and much welcome addition to the all-too-meager archaeological record for coastal Phoenicia of the Late Bronze and early Iron Ages. We eagerly await the succeeding volumes, as we do a return to peace in Lebanon that will permit a resumption of archaeological investigations at this very important and potentially very rich Phoenician city.


Astrom, P. 1957. The Middle Cypriot Bronze Age. The Swedish Cyprus Expedition, vol. IV, pt. 1B. Lund: The Swedish Cyprus Expedition. _____. 1972. The Late Cypriot Bronze Age: Pottery and Architecture. The Swedish Cyprus Expedition, vol. IV, pt. 1C. Lund: The Swedish Cyprus Expedition. Furamark, A. 1941. Mycenaean Pottery, vol. II: Chronology. Skrifter Utgivna av Svenska Institutet i Athen. Stockholm: Svenska Institutet i Athen (reprinted 1972). Gjerstad, E. 1948. The Cypro-Geometric, Cypro-Archaic and Cypro-Classical Periods. The Swedish Cyprus Expedition, vol. IV, pt. 2. Stockholm: The Swedish Cyprus Expedition. Pritchard, J. B. 1971. The Roman Port of Sarafand (Sarepta): Preliminary Report on the Seasons of 1969 and 1970. Bulletin du Musee de Beyrouth 24:39-56. _____. 1975. Sarepta: A Preliminary Report on the Iron Age. Excavations of the University Museum of the University of Pennsylvania, 1970-72. Philadelphia: The University Museum, University of Pennsylvania. Schwartz, G. M. 1988. A Ceramic Chronology from Tell Leilan. Operation 1. New Haven: Yale University Press. Tomber, R. 1990. Roman pottery studies in the Mediterranean: Past research and future prospects. Journal of Roman Archaeology 3:217-24.
COPYRIGHT 1992 American Oriental Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Greene, Joseph A.
Publication:The Journal of the American Oriental Society
Article Type:Book Review
Date:Jul 1, 1992
Previous Article:Old Babylonian Legal and Administrative Texts from Philadelphia.
Next Article:Life Among Indian Tribes: The Autobiography of an Anthropologist.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters