Introduction Author(s): Nicoletta Momigliano Source: British School at Athens Studies , 2007, Vol. 14, KNOSSOS POTTERY HANDBOOK: NEOLITHIC and BRONZE AGE (MINOAN) (2007), pp. 1-8 Published by: British School at Athens Stable URL: https://www.jstor.org/stable/40916594 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms British School at Athens is collaborating with JSTOR to digitize, preserve and extend access to British School at Athens Studies This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms Introduction Nicoletta Momigliano That the pattern [of A. J. Evans's chronology] is too neatly symmetrical is almost self-evident. It also has an empire-building note to it, for the scheme, worked out from the ruins at Knossos, was imperiously extended to the whole of Crete, though it is now certain that at least some of it will not work at all for other sites, such as Phaestos. And why should it? (Finley 1968, 16-17) 'Prehistory' was not a time when people told their own story in terms of changes in pottery styles (Fitton 2002, 25) AIMS AND SCOPE OF THE VOLUME This volume presents, in broad outline, one aspect of the archaeology of Knossos: the development of its prehistoric pottery, from its beginnings in the Neolithic period until the end of the Bronze Age. In other words, this volume offers the most up-to-date overview of the prehistoric pottery sequence of this important archaeological site, and, in addition, includes previously unpublished material (e.g. FIGS. 2.11-13, 3. 12-17, 6.17). The focus will often be on fine decorated wares, partly because of the nature of much of the material available for study, and partly because these wares are usually more sensitive chronological indicators. The site of Knossos needs no introduction. Nowadays, with the exception of the Acropolis of Athens, it is the most visited in Greece, and indeed one of the most visited in the whole Western world (Papadopoulos 1997, 99-101). It has played a central role in the history of Crete throughout several millennia, and also in the history of scholarship on Aegean prehistory. Its prehistoric pottery sequence formed the backbone of Arthur John Evans's well-known tripartite chronological system of Early, Middle and Late Minoan periods, criticised by Finley in the passage cited above, and of his Early, Middle and Late Neolithic periods. Despite being beset with other problems as well (McNeal 1973; Dickinson 1994, 9-12; and also below), Evans's basic system is still in use today and, mutatis mutandis, has been applied to other Aegean regions. Thus, if only for historical reasons, the Knossos pottery sequence, created by Evans and his assistant, Duncan Mackenzie, and improved upon by later generations of scholars, has played a crucial role in Aegean prehistory. In the last few decades, however, new excavations and museum-based studies of Knossian pottery have so thoroughly questioned and modified Evans's dating of ceramic assemblages and definition of his ceramic phases that it seemed desirable to present a new synthesis of the Knossos sequence. This volume could thus claim to be the first publication since Evans's The Palace of Minos (1921-35) that provides, within one cover, an overview of the main developments of Knossian pottery from the Neolithic to the end of the Bronze Age. This volume, however, is not meant to be a corpus of all the prehistoric pottery discovered at Knossos. Its focus is on local ceramic production, and on the most common fabrics, wares and forms or shapes found at this site. By 'local' is simply meant pottery whose production, distribution and consumption appears to be mostly (though not exclusively) restricted to the Knossos-Archanes region, and especially to the Bronze Age palace and surrounding settlement (for possible location of some pottery workshops near Mount Juktas see Day 1988; MacGillivray 1987 and Chapter 4). Since this is not a corpus of Knossian Neolithic and Bronze Age pottery, it cannot supersede the actual detailed publication of deposits from Evans's and from more recent excavations, a task that has been carried out by a number of scholars since the 1950s, with more useful work still to be done. Moreover, its emphasis is on the more common fine wares and their chronological development, which generations of scholars at Knossos have considered to be 'local' because of their high frequency in north-central Crete - an emphasis partly due to the nature of much of the material available for study (see also below). The main focus, thus, will not be on other (equally if not more interesting) aspects of Knossian ceramics such as technology, function or depositional processes, although such topics will inevitably be considered in the following chapters, for they have a bearing on the reconstruction of the ceramic sequence and history of the site. While this volume will not provide answers to all possible questions concerning Knossian ceramics, it is hoped that it will point the reader in the right direction, at least in terms of relevant publications. I am fully aware that '"prehistory" was not a time when people told their own story in terms of changes in pottery styles' (Fitton 2002, 25), and that a ceramic sequence is only a tool, not an end in itself. Moreover, pottery phases are merely taxonomic devices, and as This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms 2 NICOLETTA MOMIGLIANO such are subject to a certain amount of arbitr and interpretation, even if they are based on assemblages that are real physical entities products of past human actions. In other words, deposits are real things - their materiality ine - representing specific points in a continuous ceramic development, created by a variety of sitional processes; ceramic phases, however, cluster deposits with similar features into larger are merely useful classificatory and interpretativ invented by archaeologists, not real entities em in the archaeological record, ready to be (although at times some archaeologists seem to this). In spite of these limitations, however, one stress that most archaeological studies which a wider issues can benefit from a well-defined ceramic sequence, and sometimes can only be built upon this kind of framework (e.g. Warren 1999; Cullen 2001, 9). It is primarily thanks to studies of pottery sequences and synchronisation between various Cretan, Aegean and Near Eastern sites that one can analyse and reconstruct, for example, different patterns of interand intra-regional interaction in Neolithic or early Prepalatial Crete, diverging developmental trajectories in the late Prepalatial period (especially in EM III: Momigliano 1999^), or sequences of construction and destruction which affected those icons of Minoan archaeology, the so-called palaces, with all the implications for the study of the dynamics of power within the island and beyond (Schoep and Knappett 2005; MacGillivray, Chapter 4). It is hoped that the sequence presented here will provide a useful tool to our colleagues for comparative purposes only, facilitating the synchronisation between developments at Knossos and at other sites in Crete and in the Aegean, and stimulating enquiries beyond ceramic phases and chronology. One should also stress that this is not meant to be a scheme to be imposed on other sites: the history of Minoan archaeology has shown more than once that 'Knossocentric' (or 'Phaistocentric') views of Minoan ceramic developments have been rather detrimental to scholarship, and neither the editor nor the contributors to this volume, to echo Finley's words, are in the business of empire-building. Regionalism in ceramic production, discrete historical trajectories of various Cretan regions, and even the simple fact that different sites yield different depositional histories (i.e. stratigraphic sequences), should discourage any simplistic use of this sequence elsewhere. Indeed, I hope that what might be imitated (if anything) is the more basic approach of working out the local sequence and phasing, and understanding its relationship to other assemblages in Crete and elsewhere, before attempting to relate these to the long-established Evans-Mackenzie scheme (see also below). Moreover, even after a century of studies, there are still some gaps in our understanding of Knossos, as highlighted in the following chapters, while current and future research, presenting new material and integrating more fully the study of fabrics, technologies and functions of Knossian ceramics is likely to add to and/or change substantially the picture presented here. THE EVANS-MACKENZIE KNOSSOS POTTERY SEQUENCE AND SUBSEQUENT DEVELOPMENTS (c. 1900-2000) The sequence proposed in this volume is the cumulative result of more than 100 years of excavations and studies at this site. The main stages in its development are briefly described below. For the sake of brevity, I shall not discuss every single work that contained information relevant to the Knossos pottery sequence, but merely summarise, more or less in chronological order, works that have made new and original contributions to the understanding of Knossian stratigraphy and ceramic phasing (more detailed and thematic discussions will also be found in Chapters 1-6). Thus, seminal syntheses on Aegean chronology by Warren and Hankey (1989) and Manning (1995; 1999) are omitted, for these focused on the establishment of accurate calendar dates for existing ceramic phases of the sequences accepted at the time. Also omitted are the Kamares volume by Walberg (1976) and some relevant works by Levi (e.g. i960 and 1976), but for other reasons. On the one hand, Walberg's volume, although original in many respects, was largely based on published illustrations, rather than first-hand knowledge of the pottery, stratigraphy and documentary evidence from Knossos; on the other, Levi's publications include misconceptions about the Knossian stratigraphy and/or Knossian ceramics (as well as other basic methodological flaws), which are so serious that, arguably, they have impeded rather than helped our understanding of Minoan ceramic phasing. The first modern archaeological investigations of Knossos were conducted in 1878 by a member of the Herakleion-educated bourgeoisie, the aptly named Minos Kalokairinos (Kopaka 1995). He discovered remains of a large building attributed to the 'Mycenaean' period, for, under the influence of Schliemann's spectacular discoveries at Mycenae, most remains of the Bronze Age were so dubbed (Karadimas and Momigliano 2004; Cadogan 2006). These were followed by Evans's epoch-making excavations, which started on 23 March 1900 and continued, with various interruptions, until 193 1 (Brown 1983; Hood and Taylor 1 981; Momigliano 1 999*2). Evans's excavations at Knossos, and those made by other archaeologists at various Cretan sites during the first decade of the 20th century, revealed a new 'culture' or 'civilisation', and rivalled in importance those made by Schliemann at Troy and Mycenae a generation earlier. Evans named this civilisation 'Minoan' and the period in which it flourished the 'Minoan Age', terms and notions borrowed from other scholars (Karadimas and Momigliano 2004). Evans also immediately set out This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms INTRODUCTION 3 Table o.i: Summary of the Evans-Ma The Palace of Minos and major subse Evans and Mackenzie Furumark (1941a, 1941^) Various authors (1900-35) (c. 1953-2000) Lower/Early Neolithic Aceramic Early Neolithic I Middle Neolithic Early Neolithic II Middle Neolithic (Furness 1953; Evans 1994) Upper/Late Neolithic Late Neolithic I Late Neolithic II Late Neolithic II / Final Neolithic (Manteli 19930, I993^) 'Sub-Neolithic' EM I EM IA EM IB (Advanced) EM I (Hood 1966, 197 10, 19900) EMU EMIIA EMIIB (Warren 1972/7; Evans 1972) EM III EM III or Pre-polychrome MM IA (Hood 19620) MMIA MMIA Early MM IA Late (with decoration) (Hood 19620) MM IB MM IB MMIIA MMIIA MMIIB MMIIB MM IIIA MM IIIA MM IIIB MMIIIB MM IIIB / LM IA transition (Warren 199 10) LM IA (early and 'mature') LM IA Early LM IA Late (Popham 1977; Warren 1999) LMIB LMlB LMII LMII (subdivided into IIB, and partly c with mainland L LM IIIA LM IIIA: 1 LM IIIAi LM IIIA: 2 LM IIIA2 LM IIIB LM IIIB: 1 LM IIIB (Desborough 1964; P 1965, 1967) LM IIIC LM IIIB: 2 LM IIIC (Desborough 1964; Po 1965, 1967) Sub-Minoan Sub-Minoan Sub-Minoan to systematise the new discoveries and place them within a wider historical perspective. In this taxonomic effort, a significant building block was the creation of a chronological system based on changes in the material culture of prehistoric Crete over time, especially as observed at Knossos. Pottery, because of its abundance and virtual ubiquity, naturally played an important part. Evans's most elaborate descr pottery sequence can be foun The Palace of Minos (1921-35) of earlier classificatory at summarised as follows. Already after the first excavation campaign at Knossos in the spring of 1900, Evans discovered NeoThis content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms 4 NICOLETTA MOMIGLIANO lithic levels, and distinguished between an earlier 'Kamares' and a later 'Mycenaean' palace, both belonging to the Bronze Age (Evans 19000, 6-7, and passim). The former was characterised by the presence of ceramic vessels decorated in white and red pigments over a dark ground, and named Kamares after the Cretan cave where the first examples of this pottery were discovered (Myres 1895; Mariani 1895; Hogarth and Welch 1 90 1, 78), while the latter was characterised by the presence of ceramics decorated in lustrous reddish-brown paint over a buff ground, comparable with the 'Mycenaean' wares discovered by Schliemann and others. Duncan Mackenzie, Evans's field director, further suggested that the 'Mycenaean' palace was also followed by a period of 'decline', characterised by 'decadent' architecture and pottery, usually referred to in subsequent literature as the 'reoccupation' phase (Momigliano 1996). The next major development in the history of the Knossian pottery sequence is represented by Mackenzie's 1903 article in which, for the first time, the Scotsman attempted a phasing of both Neolithic and 'Minoan' pottery into three main stages each, and in which the labels Lower, Middle and Upper Neolithic as well as Early, Middle and Late Minoan were used consistently (Mackenzie 1903). (Interestingly, in the same Journal of Hellenic Studies volume Dawkins (1903, 249) still used the terms 'Kamares' and 'Mycenaean' ware.) Some years later, in the first volume of The Palace of Minos, Evans stressed the 'logical and scientific' nature of these tripartite divisions, for they matched what he saw as an evolutionary sequence of 'rise, maturity, and decay' characteristic of biological and cultural developments (Evans 1 921, 25; see also McNeal 1973; McEnroe 1995). In addition, in the case of the 'Minoan Age', the tripartite scheme roughly, and conveniently, corresponded to the Egyptian Early, Middle and New Kingdom periods. By the summer of 1904, Evans and Mackenzie had already established a further subdivision of Early, Middle and Late Minoan into three sub-phases each (Early Minoan I, II, III, etc), resulting in a total of nine phases, as clearly indicated by Mackenzie's sketch of the West Court section in his 1904 daybook of the excavations at Knossos (Momigliano 19990, 52~4> figs- 19-20). This basic ninefold framework, which according to Evans (19060, 4) nicely echoed and could replace the Homeric 'nine years of Minos'Odyssey 19, 178-9), was first illustrated in the papers that he presented at the 1904 Cambridge meeting of the British Association for the Advancement of Science and at the 1905 Archaeological Congress in Athens (Evans 19050, 19060, respectively). In 1906 Mackenzie published another article focusing on the Middle Minoan pottery of Knossos (Mackenzie 1906), in which he created the first pottery groups (in the sense of groups of ceramic assemblages/deposits) employed to define ceramic and chronological phases at Knossos, based on stylistic and stratigraphical observations. These groups were also interpreted as reflecting specific ev or catastrophes) punctuating the his Further refinements to the ce Knossos, which took place betw publication of the first volume o of Minos (1921), can be found in sources, such as Mackenzie's excav Evans's early draft of his Palace entitled The Nine Minoan Periods The latter shows, for example, t Late Minoan III into A, B and C way by the early 1910s. But it publication of The Palace of Minos and Mackenzie produced the sequence of ceramic (and chr Neolithic (subdivided into 'Lower and 'Upper' or 'Late'); EM I (including SubNeolithic), EM II and EM III; MM IA, MM IB, MM IIA, MM IIB, MM IIIA, MM IIIB; LM IA, LM IB; LM II, LM IIIA, LM IIIB, LM IIIC; and 'SubMinoan' or 'Proto-Geometric' (see TABLE 0.1). Evans (1928, 362-4, 472) also referred to a transitional MM IIIB/LM IA phase, but stressed that this did not really apply to Knossian ceramics. Afterwards, Mackeprang's article (1938, 546-51), though providing a useful summary, did not add much to our understanding of the Knossos sequence. Arne Furumark, on the other hand, made important refinements and modification in his monumental work on Mycenaean pottery and Mycenaean chronology, first published in 194 1 (Furumark 194 10, 1941^), the year of Evans's death. In these volumes, Furumark provided further subdivisions of LM IIIA and LM IIIB into earlier and later stages, producing the following sequence: LM IIIA: 1 , LM IIIA: 2; LM IIIB: 1 , LM IIIB: 2, and Sub-Minoan (Furumark 19410, passim and especially 9, 26-7, 29, 171, 175; Furumark 1941^, 85, 103-9). Furumark's subdivision of LM IIIA into an earlier and later stage did represent a proper exercise in phasing of ceramic assemblages, but his subdivision of LM IIIB could be seen as a rather confusing exercise in re-labelling, for Furumark's LM IIIB: 2 effectively corresponded to (and replaced) Evans's LM IIIC, and was considered by Furumark himself to be contemporary with his Mycenaean IIIC: 1 phase (Furumark 194 10, 178-9; Furumark 1941/*, 106-9). Thus, while his distinction between LM IIIA: 1 and LM IIIA: 2 has found wide acceptance, most scholars have retained Evans's LM IIIB and IIIC labels, as shown in TABLE 0. 1 (Desborough 1964, 167; Popham 1965, 316 and especially 334 n. 44; Popham 1967, 346; Kanta 1980, 3-5). The above-mentioned debate about the LM IIIB2 / LM IIIC label exemplifies one of the main problems inherent in Evans's system for the Cretan Bronze Age, and its exported versions: labels such as EM I, EM II, etc usually have been employed in a combined stylistic and chronological meaning. For instance EM I pottery This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms INTRODUCTION 5 is shorthand for the pottery in vogue dur period, or for the particular style of pot defines or illustrates the EM I chronologic particular site (for pottery styles do, of c spatial and temporal limits). But sometim confusingly, terms such as EM III, LM IB as in the example above, LM IIIB2, have b mere stylistic labels, independent of ch correlations. For example, Evans himself wro III pottery continued in eastern Crete wh already started at Knossos, or that LM overlapped with LM II (Evans 1921, 108; E 322). More recently, a number of scholar the labels 'MM IIIB', 'MM IIIB / LM IA and 'early LM IA' for the same deposits (D Macdonald 1997, 15-23, and esp. Hatzaki, 5 of this volume). In other words, these labe employed as stylistic indicators or descriptor taking into account the chronological impl such terms have for the understanding o sequence and its relative chronology in the w context. To maintain that it does not matter whether we call a deposit, for example, MM IIIB or early LM IA is perhaps naive, for whether one likes it or not, these labels have acquired a primarily chronological significance, and their employment inevitably gives the impression that Minoan Crete has a properly defined MM IIIB, MM IIIB / LM IA transition, LM IA Early and LM IA Late ceramic sequence, with properly defined chronological and ceramic phases, which is certainly not the case, at least at present. All of the above are examples of a system in which, to paraphrase and adapt Renfrew's words (1972, 54) the suggestion that EM III follows EM II seems tautological, and the claim that EM III pottery in eastern Crete could be largely contemporary with MM I in central Crete (or that LM IB is contemporary with LM II) sounds like nonsense (cf. Andreou 1978, 8-10; Dickinson 1994, 9-12). There are other important methodological issues ingrained in the tripartite Minoan system that we inherited from Evans and Mackenzie, and its application as a model elsewhere, such as the (pseudo-) ethnic connotations that terms such as Minoan, Helladic and Cycladic have assumed, at least in the writings of some scholars (Renfrew 1996; Karadimas and Momigliano 2004; and especially Whitley 2006, with further references). Obviously it is well beyond the scope of this volume to suggest a completely new chronological scheme for the whole of the Aegean, but it seems increasingly evident that archaeologists could benefit from a better integrated pan-Aegean chronological scheme, with purely chronological labels such as Early, Middle and Late Bronze Age, in which the deposits and stratigraphic sequences of individual sites might be correlated in a less confusing and more satisfactory way. I would like to suggest that Aegean archaeologists of the 21st century need to rethink the current chronological frameworks and terminologies in a more systematic and radical way, instead of merely tinkering with old schemes and labels (such as Minoan, Helladic and Cycladic), which often no longer reflect the mentalities and objectives of modern practitioners. The excellent SCIEM 2000 project (www.SCIEM2000.inf0/), although providing crucial contributions to the understanding of synchronisation across the eastern Mediterranean, has not yet addressed this more fundamental issue. I have discussed above some of the shortcomings inherent in Evans's system, which become particularly acute when adapted to (or imposed upon) other sites in Crete and in the Aegean, especially when dealing with periods in which strong regional variation can be observed in pottery production and consumption, and when labels such as EM III or MM IA are not employed as stylistic and chronological descriptors. The ongoing debate on the synchronisation of Protopalatial and early Neopalatial Knossos and Phaistos could be used as just one of many examples (Betancourt 1985, 90-114; Levi i960; Walberg 1976; MacGillivray 1998, 15, 97-102; Macdonald 2002, 36-7; Hatzaki, Chapter 5 of this volume). But even at Knossos itself new excavations and further studies of Evans's material have highlighted difficulties which can sometimes be encountered in assigning single vases or even entire deposits to one or another phase in the Evans-Mackenzie scheme. As pointed out by several scholars (Popham 19700, 11; Andreou 1978, 2; Momigliano 1 991, 151; MacGillivray 1998, 16), some of these problems stemmed from the fact that Evans's The Palace of Minos was not a proper site publication, and did not provide a sufficiently detailed discussion of the stratigraphy and of the finds. Much of what was excavated by Evans is still unpublished, and much of what he did publish did not provide adequate information for a standard site publication, even for his own times (Wace 1935; Momigliano 1999^). In spite of some remaining difficulties, since the 1950s new excavations and new studies of Evans's material have considerably improved our understanding of the pottery sequence at Knossos. Thus, in 1953, Furness published an article on the Neolithic pottery of Knossos, in which she produced a new subdivision of Early and Middle Neolithic, on the basis of stylistic and stratigraphical analysis (see Chapter 1). A few years later, Sinclair Hood, during his 1957-61 excavations along the Royal Road and in the palace, discovered important homogeneous deposits and excellent stratified sequences, especially for the Prepalatial and Protopalatial periods. Among the main results of Hood's investigations vis-a-vis the Knossos pottery sequence, one could list the discovery of a key EM I deposit (assigned by him to an early phase of EM I, or EM IA: see, however, Chapters 1 and 2); a properly stratified sequence for EM II - III, which allowed for the first time a clear definition of EM III ceramics at Knossos; an equally well stratified sequence of MM IA-IIA floor deposits; and the first 'pure' LM IB deposit, stratified above 'classic' LM I A (Hood 19620, 1966; see also this This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms 6 NICOLETTA MOMIGLIANO volume, Chapters 2-5). At the time of writing, Hood's excavations remain unpublished, but thanks to the excavator's generosity many scholars have been able to study this material, and their publications of closely comparable deposits have provided more evidence supporting and clarifying his suggestions, especially concerning the EM and MM sequence (Andreou 1978; Momigliano 1991; Cadogan et al. 1993; Wilson 1994; Wilson and Day 1994, 1999, 2000; Momigliano and Wilson 1996; MacGillivray 1998; Knappett 1999^; Momigliano 20000). John D. Evans's 1969-70 excavations in the West Court added considerably to our understanding of the Neolithic sequence. Through the study of these data, Manteli (19930, J993^) established a more detailed phasing of Late Neolithic into LN I, LN II and Final Neolithic, while Peter Tomkins, using both A. J. Evans's and J. D. Evans's material, has now produced a complete revision of Neolithic Knossos, of which a synthesis is presented in Chapter 1. Moreover, J. D. Evans's investigations, combined with Warren's 1972 excavations on the southern side of the Royal Road, also allowed for the subdivision of Knossian EM II into an earlier and later sub-phase (Evans 1972; Warren 1972/*; Wilson 1985), and an earlier suggestion by Hood (19710, 37-8) to subdivide EM II into three sub-phases (EM II A, B and C) was abandoned. M. R. Popham's studies of A. J. Evans's LBA material (Popham 1964, 1965, 1967, 1969, 19700) and his excavations of the Minoan Unexplored Mansion (MUM) provided and continue to provide, even after his untimely death, the most substantial contribution to the understanding of the LM Knossian sequence (see Popham 1984; excavation seasons carried out in 1968, 1972, 1973 and 1977). The MUM excavations, in particular, provided clear stylistic and stratigraphic evidence for a satisfactory definition of the LM II and LM III A 1 phases. Moreover, as shown by Hatzaki (Chapter 6), the MUM also provided stratigraphical evidence to distinguish earlier and later stages for LM IIIB. P. M. Warren's late i97os-early 1980s excavations of the Stratigraphical Museum Extension Site (jokingly abbreviated as SEX) have also provided good stratigraphic evidence for understanding the LBA Knossian sequence, especially for the LM IIIC ceramic phase (Warren 1981, 1983, 1997), and so have the 2001 and 2002 excavations of the Little Palace North Section site by Hatzaki (see Chapter 6). Finally, excavations directed by C. F. Macdonald, N. Momigliano and David Wilson in the early 1990s have produced useful evidence for the stratigraphy and better definition of the Prepalatial and Protopalatial ceramic sequence (Momigliano and Wilson 1996; Macdonald and Knappett, in press). The revisions made to A. J. Evans's sequence in the studies and excavations discussed above have been summarised in the right-hand column of TABLE 0.1. Naturally, this constituted a starting point that could hardly be ignored. But how should one present a new synthesis of this and other ongoing work? One simple option was merely to provide an adequate discussion of deposits that matched closely the existing phases of Evans's modified sequence, as outlined in TABLE 0.1. I felt, however, that this approach would be unsatisfactory for a number of reasons. First, this would add little new information and thus provide no real intellectual justification for the production of a substantial new monograph. Second, I was aware that studies already under way in the late 1990s were already rendering the existing sequence out of date. Last but not least, the problems inherent in Evans's system and labels, as discussed above, strongly suggested the exploration of other avenues. THE SEQUENCE IN THIS VOLUME The approach we adopted is basically that described in Cadogan et al. (1993; see also Renfrew 1972, 53-5; Andreou 1978, 10-11). Our local sequence was first established on the basis of stratigraphic and stylistic observations, without recourse to any labels such as EM IIB or MM IA, and without attempting to fit our data into a pre-existing ceramic sequence. We identified a number of suitable ceramic deposits, which we clustered into relatively homogeneous groups, that is, groups of deposits that shared a large number of ceramic features (not just one or two 'type-fossils') and, whenever possible, a similar stratigraphy, suggesting contemporaneity or close temporal spacing. These groups were then named after the ceramic deposit(s) that was (were) deemed to be the most representative and/or important for other reasons. The groups are formed by deposits excavated during A. J. Evans's and more recent investigations, the latter often providing the best stratigraphic evidence. The limitations inherent in studies of A. J. Evans's material are well known (see this volume, passim, and Momigliano 1991, 154): it is often (but not always) heavily selected, excavated by artificial spits, with materials from different contexts stored together, etc. It would be misleading, however, to infer that this large body of material could not provide any useful information. Sometimes even A J. Evans's excavations have yielded unselected and relatively wellstratified deposits, as, for example, in the case of most tests dug beneath the palace floors and of Neolithic 'Stratum a' in the Central Court (excavated in 1924). Ironically, the ceramic deposits that have suffered most are those from higher levels and often associated with the written documents, which prompted Evans to dig Knossos in the first place. The groups thus established were subsequently placed in a series by means of stratigraphic and stylistic analysis. The result was the left-hand column in TABLE 0.2, which is what appears the most acceptable and satisfactory sequence for Knossos at the time of writing, but will inevitably be susceptible to improvements and changes in the future. This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms INTRODUCTION 7 Table 0.2: The new Knossian ceramic sequence, suggested Neolithic and Minoan labels, and approximate calendar dates. Knossos sequence (2007) Suggested Suggested approximate calendar Suggested approximate Group name Neolithic and dates BC (high chronology for calendar dates BC Minoan labels LBA), adapted from Tomkins (low chronology for LBA) (Chapter 1, this volume), adapted from Warren and Manning (1995, 1999,) Rehak and Hankey (1989) and Warren Younger (200 1 ), Manning and ( 1 998, 2006) Bronk Ramsey (2003) Stratum X IN (Aceramic) 7000-6500 Strata Dt-VIII EN 6500-5900 Strata VII-VIB Strata VIA-V LN I 5300-4900 Stratum IV LN II 4900-4500 Stratum IIIB FN LA 4500-4200 Stratum IIIA FN IB 4200-3900 Stratum IIB FN II 3900-3600 Stratum IIA FN III 3600-3300 Stratum IC FN IV 3300-3000 EM I Well EMI 3000-2650 ~ West Court House EM II A (Early) 2650-2550 North-East Magazines (tests EM II A (Late) 2550-2450 beneath floors) South Front EM IIB 2450-2200 South Front House EM III (Early) 2200-2150 Foundation Trench Upper East Well EM III (Late) 2 1 50-2050 House C / RRS Fill MM IA 2050-1950 20th century Early Chamber beneath MM IB 19 50- 1900 19th century West Court Royal Pottery Stores MM IIA 1900-1850 1800-1750 Trial KV MM IIB 1 850-1800 1750-1700 West and South Polychrome MM IIIA 1800- 17 50 1700- 1640 Deposits KS 178 stone-built MM IIIB 1750-1675 1640-1600 compartment MM IIIB / LM IA transition around 1600 Gypsades Well (Upper LM IA 1675-1580 1600-1510 Deposit) SEX North House LM IB 1 580-1490 15 10-1430 Minoan Unexplored LM II 1490-1430 1430-1390 Mansion (MUM) South Sector Long Corridor Cists LM IIIAi 1430-1370 1390-1360 MUM Pits 8, 10, 1 1 LM IIIA2 1 370-1320 1 360-1330 Makritikhos 'Kitchen' LM IIIB Early 1320-1250 1330-1250 MUM North Platform Pits LM IIIB Late 1 250-1200 1 250-1 190 SEX Southern Half group LM IIIC 1 200-1 100 11 90-1 100 Once we had worked out the relative sequence, we tried to relate our groups to other deposits and sequences both in Crete and in other Aegean regions, whenever feasible. Finally, we attached to our groups the long-established and widely utilised labels of the Evans-Mackenzie scheme, in a way that seemed to fit best not only their traditional use, but also Aegean-wide developments. This has not been employed (or, rather, re-employed) for 'imperialistic' purposes, so that this scheme should be imposed on the rest of Crete, but to show how our sequence correlates to the Evans-Mackenzie scheme, and be we felt this would be helpful to our readers, if because most scholars are more used to and have m chronological correlations with terms and phases as EM I. In addition, one could argue that if ther place where labels such as EM I, EM II, etc ca used with some legitimacy, it is Knossos, where system was created in the first place. Thus, in following chapters terms such as EM I, MM IB, IIIC will often be used, for the sake of brevit shorthand for our pottery groups. This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms 8 NICOLETTA MOMIGLIANO Regionalism and different historical trajectories within the island, and the fact that deposits are often created by events and actions of a very localised nature, cannot produce very precise and harmonious links between the pottery styles and phases of different sites and areas within Crete. Even sites as close as Poros, Archanes and Knossos, some of whose pottery was most likely produced in the same workshops, probably located near or on the hills of Mount Juktas (Day 1988; MacGillivray 1987), have yielded close, but not exactly matching deposits and sequences. Many of the differences between them are undoubtedly functional, that is, due to the different nature and function of these sites (and the detailed and systematic analysis and comparison of different assemblages from Poros, Archanes and Knossos could, no doubt, lead to interesting results). Other discrepancies, however, may be chronological, precisely because different sites have different depositional histories. That ceramic deposits may be the result of pan-Cretan 'events' is probably more often the exception than the rule. Clearly much more work remains to be done in terms of synchronisation, from the micro-regional level (e.g. comparing the histories and relationships of northcentral Cretan sites such as Knossos, Archanes, Poros, Galatas, etc) to the rest of Crete, and from the Aegean to the whole of the eastern Mediterranean, but this updated Knossian sequence should be a first step in the right direction. Undoubtedly some colleagues will challenge some of our EM and LM labels and chronological attributions (or even the inclusion of specific deposits in one or the other group), but we hope that at least our discussion of the stratigraphic evidence and our basic sequence of groups of deposits will be welcomed and less controversial. The results of our work are summarised in TABLE 0.2, and the full discussion of the stratigraphic and stylistic evidence upon which this is based will be found in the following chapters. ORGANISATION OF CHAPTERS The discussion of the ceramic sequence has been organised into chapters which match the expertise of individual authors, and follow, approximately, wellestablished subdivisions of Cretan prehistory such as Prepalatial, Protopalatial and Neopalatial first suggested by N. Platon (19560, 19610) and subsequently refined by other scholars (Hallager 19880, 1988^; Rehak and Younger 1998, 2001). Thus, Chapter 1 covers the Neolithic and ventures into the beginning of the Prepalatial period; Chapter 2 deals with the early Prepalatial phases (EM I-IIB); Chapter 3 is devoted to the late Prepalatial phases (EM III-MM IA); Chapter 4 discusses the Protopalatial period, traditionally spanning the MM IB-IIB ceramic phases, but here also including MM IIIA; Chapter 5 covers the Neopalatial MM IIIB-LM IB phases; and Chapter 6 covers the Final Palatial and Postpalatial periods (LM II-IIIC). Because this is a multi-author volume, and because of the quirky history of scholarship on Minoan ceramics, readers should not expect absolute uniformity of approaches, styles and content throughout the book. The basic information and the basic structure of the chapters, however, follow a strict pattern. Each begins with a brief introduction on the period and history of relevant scholarship. Then, for each pottery group there is a section on the archaeological contexts, discussing the character and nature of the deposits, the relevant stratigraphic and stylistic evidence for their position in the pottery sequence, and including a list of the main deposits. Deposits are normally listed in clockwise order in relation to the Bronze Age palace, starting from the West Court, followed by deposits from the surrounding town and cemeteries. In the list of deposits, numbers such as B.I.i and D.I. 2, preceded by the abbreviation KSM (Knossos Stratigraphical Museum), refer to those marked on the wooden boxes housing the relevant pottery and given in Pendlebury's Guide to the Stratigraphical Museum at Knossos (Pendlebury et al. 1933-5); tne box numbers refer to the other system of numeration marked on the wooden boxes, which was given to Evans's material once it was moved from the original Pottery Archives in the palace to its present location in the KSM (Momigliano 19990, 58, 134). This should facilitate the location of material discussed in this volume for readers who may wish to see it with their own eyes (the only proper way to study ceramics). Archaeological contexts and list of deposits are, in turn, followed by a section on the characteristics of the ceramic group, articulated through a description of most common fabrics (in the sense of ceramic paste), wares (in the sense of surface finish/treatment/ decoration), forms and their shapes (in the meaning employed by Furumark, 19410, 1; Walberg, 1976, 14). As already mentioned, on the whole more attention has been paid to what are usually referred to as fine decorated wares because of the nature of much of the material available for study (i.e. deposits heavily selected by Evans and Mackenzie or subsequent excavators). Moreover, these wares appear to be more susceptible to change and more diagnostic for chronological purposes. Fabrics and other technological features, however, besides providing useful information on other aspects of Cretan prehistory, can also be useful chronological indicators, as well illustrated by Tomkins's chapter, and it is hoped that more studies in the future will focus on these subjects. Finally, there is a section on the relative chronology of the pottery group/ceramic phase, discussing synchronisation with other Cretan deposits and with other Aegean or eastern Mediterranean regions. In addition to the black-and-white illustrations embedded in the text, the volume is provided with a CD, which contains further illustrative material, mostly in colour. This content downloaded from 147.251.69.18 on Thu, 26 Aug 2021 19:22:01 UTC All use subject to https://about.jstor.org/terms