Transcript
361 The Chronology of the Philistine Monochrome Pottery: An Outsider’s View Susan Sherratt Ashmolean Museum Professor Amihai Mazar will perhaps forgive me if I take the opportunity in thisshort contribution in his honor to offer a few thoughts on a debate in which hehas been closely involved: that is, the question of the chronology of the so-called Philistine monochrome pottery (also known as “Mycenaean IIIC:1b”)produced locally at a number of sites in the southern Levant during Iron Age IA.As an outsider to the archaeology of this area, I have followed the debate, whichhas been carried on in extremely lively fashion almost entirely by archaeologistsworking in Israel, with interest and a certain amount of bemusement, at thesame time heartily thankful that I need not become involved. It seems to me,however (and perhaps rashly), that the time has now come to chip in with aview from the outside. The Current Debate The debate and its history are by now well known. 1 Its starting point was the dis-covery in the 1970s and 1980s that, at sites such as Ashdod and Tel Miqne–Ekron, identified with two of the cities of the Philistine pentapolis, the bichromepottery traditionally called “Philistine” was preceded by locally made mono-chrome pottery with a strong resemblance to similar pottery produced in theearly stages of Late Cypriot IIIA and classified by Dikaios as “Mycenaean IIIC:1b”(1969–71: 485). At first, this pottery was regarded as evidence of an earlier waveof seaborne (“Sea Peoples”) invaders preceding the arrival of the Philistines,themselves long identified with the Peleset of the Medinet Habu inscriptionsand thought of as settling in Canaan immediately after the events of Year 8 of Ramesses III’s reign (T. Dothan 1982: 289–95; 1983; 1989). However, since themid-1980s, a consensus has gradually emerged that this monochrome pottery isbest seen as an earlier product of the same Philistines who produced itsbichrome successor, and that its first appearance—rather than that of Philistine 1.For recent summaries see, for example, Naªaman 2000; Bunimovitz and Faust 2001. Author’s note : This small contribution is dedicated, with friendly greetings, to Professor AmihaiMazar. Having enjoyed several arguments with him in the past few years over the interpreta-tion of Philistine monochrome pottery, I am delighted to be able to discuss an aspect of it onwhich I believe we are in complete agreement. Susan Sherratt 362 bichrome—should be associated with the aftermath of Ramesses III’s Year 8, thusgiving it a neat terminus post quem (Mazar 1985; Singer 1985; Stager 1995; cf.T. Dothan 1998). 2 According to this view of its dating, the absence of Philistinemonochrome at sites outside the Philistine pentapolis, particularly those such asLachish that apparently remained under Egyptian control until the end of orafter Ramesses III’s reign, is not seen as a problem, except insofar as it may castdoubt on the old idea, extracted from a certain reading of Papyrus Harris, thatRamesses III settled some of the Philistines as mercenary forces in his garrisonsin southern Canaan (cf. Finkelstein 1998a: 143). Only Philistines could have ause for Philistine pottery, and the presence or absence of this ware provides amaterial expression of their territorial boundaries and the lack of interaction be-tween neighboring groups (Stager 1995).However, consensus among archaeologists rarely lasts long—particularly inthis part of the world. And there are some—beginning already in 1985 with Us-sishkin (1985; 1998) and followed by Finkelstein (1995; 1998a)—who have seengrounds for doubting the validity of the chronology thus established and theneat ethnic, cultural, and political boundaries that the absence of Philistinemonochrome outside presumed Philistine territory presupposes. Finkelstein inparticular has questioned the plausibility of casting pottery in the role of such arigid ethnocultural indicator, particularly in the case of sites located relativelyclose to one another that can hardly be imagined not to have interacted in someway. The absence of Philistine monochrome pottery at Lachish Stratum VI,which seems still to have been going strong during the reign of Ramesses III, andat other sites with dating evidence directly relatable to even later 20th-Dynastyreigns, rather indicates to Finkelstein and Ussishkin that the appearance of thispottery was not only substantially later than Ramesses III’s Year 8 but probablyalso later than his entire reign as well as possibly those of his immediate succes-sors. In the extreme version of this position, Finkelstein (1995; 1998a; 2000) andUssishkin (1998) have argued that Philistine monochrome does not appear untilafter the completion of the Egyptian withdrawal from Canaan at the end of orafter the reign of Ramesses VI, that is, approximately half a century or more afterthe date conventionally attributed to it.I have to confess to some admiration for the logic and unrelenting consis-tency of some of Finkelstein’s arguments and to a certain amount of sympathywith his motives for wishing to lower the chronology. Like him, I have problemswith the idea that something with as little perceived status and self-containedideological potential as pottery seems generally to have had in the second-millennium Near East was naturally confined within rigid ethnic boundaries.However, I do not believe we are necessarily faced with the stark choice with 2.It should be noted that although T. Dothan, between her publications of 1989 and 1998,has gradually moved toward regarding the monochrome pottery as a product of the first Phil-istine settlement, she is still inclined to date its first appearance to just before the reign of Ramesses III, placing it at around 1200 bce on a high Egyptian chronology and 15 years lateron a low chronology (1998: 151; cf. 1989: Fig. 1.6). The Chronology of the Philistine Monochrome Pottery 363 which he presents us between strict ethnocultural or ethnopolitical demarcationon the one hand and lowering the chronology of the monochrome pottery onthe other. Various alternative explanations might be suggested, among them, forinstance, the idea of some sort of politicoeconomic quarantine or embargo exer-cised by the Egyptian authorities and aimed not so much at pottery itself butmore generally at the kind of uncontrollable, and potentially subversive, decen-tralized commercial activities going on in the centers in which this pottery wasproduced. 3 I also appreciate one of the most striking effects of Finkelstein’s low-ered chronology, which is to decouple the link between Ramesses III’s account of the events of his Year 8 and the appearance of either Philistine bichrome ormonochrome pottery, which has been the basis for tying in the archaeologywith the Egyptian historical records to create the traditional narrative accountsof Philistine settlement. Indeed, I would be inclined to go further and questionwhy, having broken this link, we should continue to think it necessary to equatea particular pottery style with Philistine “settlement” at all. 4 That said, however, I do not wish in this essay to embark on a discussion of how the pottery might be interpreted. Nor do I have any desire to become em-broiled in arguments over the site-by site minutiae of the archaeology of thesouthern Levant as a whole. What I would like to do is to comment on the ques-tion of the relative and absolute chronology of Iron Age IA and its pottery in thePhilistine area and to present the reasons why, as a more or less dispassionateoutsider, I find myself continuing to support the higher chronology for this pe-riod defended most energetically by Mazar (1997; Gitin, Mazar, and Stern 1998:184–85). This is not to say that I cannot envisage at least theoretical room forsome revision or adjustment (or at least elasticity) within the chronology of thelong period between the beginning of Iron Age IA and the later part of Iron AgeII in the southern Levant in general, of the sort that has been suggested for indi-vidual Iron Age II strata at Megiddo or Hazor (Finkelstein 1998b; 1999). This issomething that, as Finkelstein has argued, need do no serious harm to biblicalhistoricity in any profound sense but merely to the attempt to match the archae-ological record too minutely and perhaps uncritically to specific episodes of bib-lical political history, and it is something on which more good sequences of plentiful radiocarbon dates based on short-life samples from reliable contextsmay eventually help to cast further light (cf. Mazar 1997: 162; see also Finkel-stein 1998b). Nevertheless, the absolute parameters within which room for suchflexibility might exist seem generally fairly firmly fixed. At the lower end, theyare provided by historical Assyrian cross-links, and I shall argue that equallywell-dated cross-links (although this time of a purely archaeological nature) arealso provided at the upper end, at the turn of the 13th to 12th centuries. Thus,while there may be scope for adjustment in the form of stretching and compres-sion lower down the early Iron Age chronology, there is very little scope for 3.For another alternative explanation, albeit from a point of view that continues to stressquestions of identity, see Bunimovitz and Faust 2001.4.Cf. Sandars 1978: 164–70. Susan Sherratt 364 moving the initial date of what we call Philistine monochrome pottery down-ward by as much as half a century from its traditional date of roughly 1200 bce .If anything, there might even be some argument for raising it very slightlyhigher. The Eastern Mediterranean Context To an outsider, one of the most bemusing aspects of the chronology debate, par-ticularly as it affects the Iron Age I period, is the almost total lack of concern onthe part of the advocates of the lower chronology with the implications this hasfor the chronology of other parts of the eastern Mediterranean. If the Philistinemonochrome pottery, with its acknowledged Aegean (or more accurately Cyp-riot) 5 ceramic links, is to be down-dated, then logically we either have to re-datethe Cypriot sequence (and, by extension, also make adjustments to the Aegeandating) or to conclude that both Philistine bichrome and Philistine mono-chrome reflect styles and forms of pottery that have already died out on Cyprusby the time these appear in the southern Levant. The first might be the moreattractive possibility, were it not that we now have persuasive independentsupport for the conventional Cypriot chronology, which sets the end of LateCypriot IIC and the beginning of Late Cypriot IIIA at roughly 1200 bce . Thistakes the form of an extensive set of calibrated radiocarbon dates, including theresults from a number of short-lived samples, derived from a series of well de-fined Late Cypriot IIC–IIIA contexts at seven different Cypriot sites (Manning etal. 2001). It demonstrates fairly convincingly that the beginning of Late CypriotIIIA is most likely to coincide with the first two decades of the 12th century. Thesecond possibility has always seemed a counsel of last resort, not only because itis extremely difficult to explain, 6 but because it nullifies one of the basic workingprinciples on which comparative relative chronology is traditionally based.Furthermore, not only Cypriot chronology but also that of Ugarit needs to betaken into account. The date of the fall of Ugarit is admittedly not set in stone,but it is the result of closely argued reasoning of an independent nature, basedon the texts (Singer 1987; 1999; Yon 1992; cf. Yasur-Landau in press). The con-clusion is that it hardly seems possible to date it much beyond the very begin-ning of Ramesses III’s reign, and possibly somewhat earlier.The reasons that the independent dating (by quite separate means) of theCypriot sequence and Ugarit has a particular bearing on the date of the earliestPhilistine monochrome pottery are to be found at Ashdod Area G Stratum XIIIb.This stratum, according to current interpretations, succeeded the destroyed Ca-naanite level (Stratum XIV) and is now widely regarded as the stratum associatedwith the initial settlement of Philistines at that site following the events of Ramesses III’s Year 8 (Mazar 1985; 1997; cf. Stager 1995). At any rate, what is 5.Cf. Mazar 1985: 104; Killebrew 1998.6.Cf. Mazar in Gitin, Mazar, and Stern 1998: 184. The Chronology of the Philistine Monochrome Pottery 365 more to the point is that Ashdod Stratum XIIIb is characterized by the appear-ance of Philistine monochrome (“Mycenaean IIIC:1b”) pottery and a number of small rooms or workshops, including one with a store of pottery known as the“potter’s workshop,” the floor of which covers the destroyed remains of part of the Stratum XIV city wall (M. Dothan and Porath 1993: 53–60; see also T. Dothanand Dothan 1992: 165–68). The succeeding Stratum XIIIa produced, in additionto the monochrome pottery, the first of the Philistine bichrome that appears tohave replaced it by Stratum XII. Because Philistine monochrome is generally en-visaged by the proponents of both the high and low chronologies as covering arelatively short time span (little more than 30 years), the implication is that Ash-dod Stratum XIIIb was a particularly short-lived stratum, lasting no more than acouple of decades at most. Since 20 years or less is of little consequence in termsof either noticeable changes in pottery style or indeed the lives of individualpots, it follows that any pottery securely associated with Stratum XIIIb and datedby cross-reference to absolutely datable contexts elsewhere falls within the ac-ceptable range of error of a date for the introduction of Philistine monochromeat the site.Pottery from the Stratum XIIIb potter’s workshop (M. Dothan and Porath1993: 12, 55–58, Figs. 14–15, Pls. 35: 9–16, 36, 37: 1–4) has some especially closeparallels in White Painted Wheelmade III pottery from Late Cypriot IIC or IIC/IIIA transitional contexts, including Tomb 9 at Kition. These include types that,in Dikaios’s ceramic terminology, belong not to the “Mycenaean IIIC:1b” cate-gory of what we now call Cypriot White Painted Wheelmade III Ware, but to the“Late Mycenaean IIIB” category, which Dikaios regarded as the Late Cypriot IICpredecessor of his Late Cypriot IIIA “Mycenaean IIIC:1b” (1969–71: 249–52). 7 Among these are a number of handleless conical bowls that find good parallelsin the upper burial of Kition Tomb 9 (Karageorghis 1974: 86–87, Pls. 155; 156:158, 175, 225, 260, 342, 349; 157: 19, 99, 105, 115), which its excavator dates tothe end of Late Cypriot II (Karageorghis 1974: 93–94) and which, by any ac-count, cannot be later than Kition Floor IIIa, the earliest Late Cypriot IIIa level,since the tomb was covered by the construction of Floor III (Karageorghis andDemas 1985: Pl. 28: 2 [Section F–F 1 ]). This shape is also found in Late Cypriot IICcontexts at Enkomi and occurs already in the Plain White Handmade and PlainWhite Wheelmade I Wares of Late Cypriot I–II (Keswani 1991: 101–2, Fig. 11.1:E–F); and there is little evidence of its White Painted Wheelmade III versions per-sisting long into Late Cypriot IIIA (Kling 1989: 135–37). 8 Also represented in the potter’s workshop at Ashdod are a number of unusu-ally small skyphoi (M. Dothan and Porath 1993: Fig. 14: 9–15, 17–18, 20),which, like the accompanying larger linear-decorated skyphoi (M. Dothan andPorath 1993: Fig. 14: 16, 19, 21–23), also find their best parallels in Late CypriotIIC or transitional Late Cypriot IIC/IIIA pottery (Sherratt 1990a). Equally inter-esting is the presence also in Area G Stratum XIIIb, from one of the adjoining 7.For an explanation of the confusing Cypriot ceramic terminology and the contexts inwhich it was formulated, see Kling 1991; see also below.