Aegean Frescoes in Syria-Palestine: Alalakh and Tel Kabri

Woolley was convinced that Minoan fresco painting originated in Syria; other scholars interpreted the Alalakh frescoes as evidence for Minoan influence in Syria. This controversy, also related to chronological problems, remained unresolved until recently.

In 1987, in his excavations at Tel Kabri Aharon Kempinski discovered a painted plaster floor in a Canaanite palace of the seventeenth century BC. Recognising that this floor forms an alien element in the Near East, but has parallels in the Aegean, he invited us to join him in the Kabri excavations. Between 1989 and 1991, the floor and fragments of a miniature fresco comparable to the miniature fresco in the West House at Akrotiri on Thera were unearthed. From 1991 on, fresco paintings showing Minoan iconography were found at Avaris in the eastern part of the Nile delta.

The finds at Kabri and Avaris led us to a reinvestigation of the fresco fragments from Alalakh kept in the Ashmolean Museum, Oxford. Our study showed that they are of Minoan character and led to new reconstructions which are presented in this paper.

We discuss the arguments recently put forward against the suggestion that Aegean artists were involved in the painting of the murals at Alalakh, Kabri and Avaris, and we try to explain the phenomenon of the Minoan frescoes in the Near East.

INTRODUCTION

For almost forty years the wall paintings from Yarim-Lim's palace at Alalakh Level VII of the late (Syro-Palestinian) Middle Bronze Age, excavated by C.L. Woolley and showing a "striking resemblance...to the Minoan frescoes" (Woolley 1955, 228-234, pls. 26b-28b), formed stylistically and technically an isolated phenomenon in the ancient Near East. They were painted in al fresco technique without a binder on the still wet plaster, later details being added in al secco painting (Woolley 1955, 228-230; Barker in Woolley 1955, 233-234; Nunn 1988, 11-12).⁽¹⁾ At that time and until recently, Bronze Age al fresco painting (with added al secco painting) was - with the exception of Alalakh - known only from Crete (see Heaton 1910; 1911; Forbes 1955, 241-242; Cameron, Jones and Philippakis 1977; Hood 1978, 83; Immerwahr 1990a, 14-15), whereas otherwise in the ancient Near East and in Egypt the al secco or tempera technique was used for wall paintings.⁽²⁾ Nevertheless, Woolley stated that "there can be no doubt but that Crete owes the best of...its frescoes to the Asiatic mainland" and "that we are bound to believe that trained experts, members of the...Painters' Guilds, were invited to travel overseas from Asia (possibly from Alalakh itself...) to...decorate the palaces of the Cretan rulers" (Woolley 1953, 74-75).⁽³⁾ The priority of the Alalakh frescoes over the Cretan ones suggested by Woolley has been accepted by several scholars (Naumann 1972, 119; Hood 1978, 48; Sherratt 1994, 238). Other scholars have, however, seen Minoan influence in the Alalakh frescoes (Parrot 1958b, 109; Moortgat 1959, 12; Matz 1962, 113; 1973, 577; Stevenson Smith 1965, 49, 102-104; Schachermeyr 1967, 46; Klengel 1979, 60-61; Nunn 1988, 203; Immerwahr 1990a, 15, 35; Niemeier 1991, 190-195; Negbi 1994, 77-79; Cline 1995b, 269-270). After Woolley's first publication of the Alalakh fragments, our corpus of fresco paintings (with additional painting al secco) in the Eastern Mediterranean has been substantially supplemented by important finds from three sites. From 1967 on, the largest and best preserved assemblage of fresco paintings in the Aegean was uncovered in the early (Aegean) Late Bronze Age settlement at Akrotiri on Thera in the excavations conducted by S. Marinatos and C.G. Doumas (Morgan 1988; Doumas 1992; Televantou 1994).⁽⁴⁾ In the Near East and Egypt fragments of fresco paintings with Aegean affinities were discovered between 1987 and 1991 in the excavations conducted by the late A. Kempinski and us in the late (Syro-Palestinian) Middle Bronze Age palace at Tel Kabri in northern Israel (Kempinski 1988, vii; Kempinski and Giveon 1989, vi; Niemeier 1991, 196-199; Niemeier et al. 1991; Niemeier and Niemeier 1992);⁽⁵⁾ and from 1991 on in the excavations conducted by M. Bietak at Tel el-Dab'a/'Ezbet Helmi, the site of the Hyksos capital Avaris, in the eastern Nile Delta (Bietak 1992; 1994; 1995; 1996a, 73-81; 1996b; Bietak and Marinatos 1995).⁽⁶⁾

The Avaris fragments, depicting - among other motifs - spectacular bull leaping scenes, have caused great excitement and much discussion. They came from three areas, H/I and IV, H/II and H/III.⁽⁷⁾ Most of the fragments were found in secondary contexts, next to the mudbrick platform of a probable palatial fortress in area H/I and from walls and foundation trenches of later repairs and additions (Jánosi 1995). With good reason, the excavators think that the wall paintings to which these fragments belonged had adorned walls of this palatial fortress (Bietak 1992; 1994, 44; 1995, 20; Jánosi 1994, 32; 1995, 68-70). For several years it was dated to the late Hyksos period (Bietak 1992, 26; 1994, 55-58; 1995, 20; Jánosi 1994, 27-33; 1995, 66), but in 1996 Bietak changed the date to the early years of the Eighteenth Dynasty (Bietak 1996a, 67-70; 1996b, 5-11). The fresco fragments in area H/II also came from secondary contexts (Bietak and Marinatos 1995, 49). In area H/III two successive buildings were identified, the earlier one dated to the Hyksos period, the later one to the early Eighteenth Dynasty. In the preserved lowermost part of the facade of the Hyksos period building, lime plaster without preserved figural painting, but "so typical of Minoan wall-painting" was found (Bietak and Marinatos 1995, 49). In the level of the early Eighteenth Dynasty, fresco fragments were found on both sides of a wall and "concentrating" around a portal (Bietak and Marinatos 1995, 49; Bietak 1996b, 14).⁽⁸⁾ From the stratigraphical evidence in area H/III and other evidence, Bietak and N. Marinatos came to the conclusion in 1995 "that Minoan wall paintings existed in Avaris both during the late Hyksos period and the early 18th Dynasty" (Bietak and Marinatos 1995, 49; Bietak 1995, 23). In his most recent publications, however, Bietak denies the existence of Minoan fresco painting at Avaris during the Hyksos period and dates all fresco fragments to the early Eighteenth Dynasty (Bietak 1996b, 14 with n.13). All discussions of the Avaris frescoes have hitherto been based on the initial dating of the fragments from area H/I and IV in the late Hyksos period.⁽⁹⁾ Since the excavations at Tell el-Dab'a/Avaris and other sites in the eastern Nile Delta have demonstrated that the Hyksos were 'Canaanites' from Syria-Palestine (Bietak 1979, 218-268; 1987; 1996a, 10-67; Dever 1985, 71; Redmount 1995), ⁽¹⁰⁾ the majority of the fresco fragments from Avaris have been seen by a series of scholars - among them until very recently Bietak himself - together with those from Alalakh and Tel Kabri as a phenomenon of the Hyksos period (Niemeier 1991, 192-200; 1995, 9-11; Hankey 1993a, 28-29; 1993b; Bietak 1994, 55-58; 1995, 26; Bietak and Marinatos 1995, 60-61; Cline 1995b, 267-270; Mellink 1995, 86-87; Morgan 1995, 29, 44; Warren 1995, 4-5, 13). Bietak's radical redating of all Avaris fresco fragments has therefore caused great confusion among scholars. One of them, E.H. Cline, is now even suggesting "that the rest of the scholarly community might consider giving the Tell el-Dab'a excavators some breathing room" and "wait until they have concluded their excavations in the relevant areas of the site, and have published enough stratigraphical evidence to firmly date the frescoes and other pertinent artifacts, both to their satisfaction and to ours" (Cline 1998).⁽¹¹⁾ If the redating of all Avaris frescoes to the early Eighteenth Dynasty is correct, they have nothing to do with Hyksos/Canaanites and are to be separated from the Alalakh and Tel Kabri frescoes. But we still consider it possible that the Avaris frescoes come from two different periods, Hyksos and early Eighteenth Dynasty.⁽¹²⁾

THE STATE OF THE DISCUSSION OF THE FRESCOES FROM TELL EL-DAB'A AND TEL KABRI

The fresco finds from Avaris and Tel Kabri have given rise to controversy over the question of what sort of exchange of ideas and possibly people between the Aegean and the Near East and Egypt lies behind them. In the case of the Avaris frescoes, Bietak and Marinatos have suggested that Minoan artists, trained in the Knossian tradition, were at work in Avaris (Bietak 1992, 27; 1994, 55; 1995, 26; 1996a, 75-76; 1996b, 21-22; Bietak and Marinatos 1995, 60). In the case of the Kabri frescoes, we have argued that they were painted by itinerant Aegean artisans (Niemeier 1991, 198-199; 1995, 11). Both Bietak and Marinatos, as well as ourselves, have met with agreement,⁽¹³⁾ but also, on the other hand, with scepticism and disagreement. The late V. Hankey thought that it was "premature to attribute the (Avaris) paintings to Minoan artists when very little is known about Near Eastern wall-painting between those at Mari (18th century BC) and those from Alalakh Level VII, Tell Kabri and Tell el-Dab'a (17th to 16th century BC)" (Hankey 1993a, 28). Stylistic and iconographic peculiarities in the Avaris frescoes encourage J.G. Younger to think that "it seems better to exercise caution about...the nationality of their painter(s)" (Younger 1995, 516-518), and make it appear unlikely to M.C. Shaw and P. Rehak that they were painted by Minoan artists (Shaw 1995, esp. 106, 112; Rehak 1996, 41 n.29). If the painters were Minoans, according to Shaw "they have been abroad long enough to have drifted away artistically from the canonic Minoan methods of representation. They would be second or third generation expatriates". For G. Kopcke the frescoes from Tell el-Dab'a are not an Aegean but a Levantine phenomenon, possibly of Aegean inspiration (Kopcke in the discussion following Cline 1995b, 285). S.W. Manning et alii argue against the hypothesis first put forward by us, according to which itinerant Aegean fresco painters were commissioned to decorate Levantine palaces within the dynamics and formal apparatus of the diplomatic practice of gift-exchange (Niemeier 1991, 199-200), on the grounds that "it has yet to be demonstrated convincingly that the Aegean palaces, and especially those of the LM IA/LH I periods, participated in the same level of interpalatial exchange that is evidenced in the Near Eastern palatial archives: such an Aegeocentric approach is unwarranted on present evidence" (Manning et al. 1994, 220-221). S. Sherratt thinks "if Bietak had got to Dab'a, or Kempinski and Niemeier to Kabri, before Evans got to Knossos, I doubt if the question of a diaspora of Aegean fresco artists to the east would seriously have arisen", and sees the relations of the Kabri and Tell el-Dab'a murals to Aegean wall paintings "in terms of the forging of an élite koine - artistic, iconographical, ideological, technological - in the circumstances of the intense maritime interaction between coastal areas of the eastern Mediterranean" (Sherratt 1994, 237-239). For A.B. Knapp, too, the Tell el-Dab'a paintings form part of an eastern Mediterranean koine and have nothing to do with Minoans per se.⁽¹⁴⁾ Those who see Aegean painters at work at Alalakh, Tel Kabri and Avaris are confronted with Sherratt's reproach of having an Aegeocentric view (Sherratt 1994, 237).

The idea of an east Mediterranean koine is not new, however. Close connections between the Aegean on the one hand and Egypt as well as Syria on the other hand were already recognised in the pioneer period of Aegean archaeology between the turn of the century and the end of World War II (Fimmen 1924; Evans and Evans 1936, 44 (Egypt), 208 (Syria); Persson 1942, 176-196). A short time after the end of World War II, H.J. Kantor published a basic monograph on that topic (Kantor 1947), and six more books on it appeared between the 1950s and 1990s (Vercoutter 1956; Stevenson Smith 1965; Schachermeyr 1967; Helck 1979; Crowley 1989; Lambrou-Phillipson 1990; Cline 1994). In studying the Thera frescoes, C.G. Doumas has found that there existed an artistic vocabulary common to all peoples of the eastern Mediterranean (Doumas 1985), and B. Muller (= B. Pierre) has demonstrated that the Bronze Age wall paintings between the Euphrates and the Aegean do not only have common motifs but also common general conceptions (Pierre 1987, 561-576; Muller 1994; 1995). All these and further studies on special issues demonstrate the frequent exchange of objects and ideas between the different civilisations of the eastern Mediterranean in the second millennium BC. In regard to the wall paintings Muller/Pierre has used the term koine (Pierre 1987, 574; Muller 1994, 56). Bietak and we were also thinking in terms of an eastern Mediterranean elite koine when we suggested that the Tel Kabri frescoes were painted by Aegean specialists commissioned within the framework of diplomatic gift-exchange in the eastern Mediterranean (Niemeier 1991, 199-200; 1995, 10-11), and when Bietak suggested that the Tell el-Dab'a frescoes were painted by Minoan fresco painters belonging to the entourage of a Knossian princess married to a pharaoh, first a Hyksos one, then Ahmose, the first pharaoh of the Eighteenth Dynasty (Bietak 1992, 28; 1995, 26; 1996a, 136; Bietak and Marinatos 1995, 61).⁽¹⁵⁾ Sherratt's argument that the question of a diaspora of Aegean fresco artists to the east would not have arisen if the Tell el-Dab'a and Kabri frescoes had been found before the Knossian ones is not a very strong one. Greek vases were found in Etruria before they were excavated in Greece and were commonly called Etruscan during the eighteenth century (Sparkes 1996, 47-56). Vases of the seventh century Wild Goat style were first found in Rhodes and for a long time designated as 'Rhodian' (Cook 1972, 117). But more recent scientific analyses and stylistic investigations have demonstrated that Rhodes was not an important pottery production centre and that there were four producers of Wild Goat style pottery: Miletus, Chios, Clazomenae and a further place in northern Ionia, of which Miletus was the most important (Dupont 1986, 57-71; Jones 1986, 665-671; Cook 1992, 255-266). According to A.B. Knapp, the entire topic of the wall paintings from Tell el-Dab'a, Alalakh and Tel Kabri "needs a thoroughgoing iconographic study set in the context of a prestige-goods exchange system that operated throughout the area" (i.e. the eastern Mediterranean) (see above, note 14). In what follows we shall do this with the fresco fragments from Tel Kabri, which were excavated by us, and those from Alalakh, which we were able to study in the original.

PROBLEMS OF CHRONOLOGY

In investigating the relations between these Near Eastern murals and the Aegean ones, one is confronted with serious chronological problems, since there exist different chronological systems within each of the several regions of the eastern Mediterranean involved, and no agreements have as yet been reached. In the Near East we are confronted with five different chronologies starting from different datings of the First Babylonian Dynasty: the ultra-high, the high, the middle, the low and the ultra-low chronologies (cf. Tadmor 1970, with bibliography 260 nn.5-7). While the highest two and the lowest dates are increasingly ruled out, there is no consensus yet on a choice between the middle and the low chronologies (cf. Dever 1992a, 11; 1992b, 45; Malamat 1992, 122). As for Egyptian chronology, it was thought until a few decades ago to have been securely reconstructed, with the beginning of the Eighteenth Dynasty (which is of importance in this connection) ca. 1575 BC. In recent years, however, serious questions have been raised about the traditional chronology of Egypt within several categories of evidence, and the date for the beginning of the Eighteenth Dynasty now varies between ca. 1570 and 1540 BC (cf. Ward 1992, esp. 53, 56). The traditional chronology of the Aegean Bronze Age as constructed by Evans (see Furumark 1941b, 110 with references to Evans) remained essentially unchallenged until quite recently, and was basically approved by P.M. Warren's and V. Hankey's major study on Aegean Bronze Age chronology published in 1989 (Warren and Hankey 1989). Two years earlier, however, the traditional chronology had come under attack from P.P. Betancourt and S.W Manning who, starting from radiocarbon dates from Thera, suggested that the Aegean Late Bronze Age had begun considerably earlier than in the traditional chronology (Betancourt 1987; Manning 1988). Since then, the controversy has continued (see, for instance, Hardy (ed.) 1990),and today the traditional low chronology and the new high chronology are opposed to each other in modified forms without a consensus.⁽¹⁶⁾ According to the former, the Aegean Late Bronze Age started ca. 1600 BC, according to the latter ca. 1675-1650 BC.

THE FRESCO PAINTINGS FROM TEL KABRI

At Tel Kabri, the fresco paintings were unearthed in the Middle Bronze Age palace in areas D and F. The architectural history of this palace started in early (Syro-Palestinian) MB IIB (see Garfinkle and Peilstöcker 1994, 13-14), depending on the different chronologies used ca. 1800 (Mazar 1968, 97; Matthiae 1975; 1984b), 1750 (Dever 1992a, 10; 1992b, 45), 1725 (Weinstein 1992, 38) or 1700 BC (Bietak 1989, 97; 1992, 62, fig. 24), with a possible predecessor in MB IIA (Garfinkle and Peilstöcker 1994, 12-13). In later MB IIB the palace was superseded by a massive programme of rebuilding (Garfinkle and Peilstöcker 1994, 13-15). This later MB IIB palace saw several changes, mainly the addition of inner buttress walls and the narrowing of some of the rooms (Kempinski 1992a, 4). The palace was destroyed and abandoned during the later part of MB IIB, during MB IIC or during MB III, depending on the terminology used.⁽¹⁷⁾ This happened before bichrome ware came into use in Palestine before or at the beginning of the reign of Apophis (see Kempinski 1983, 131-148, 223), the penultimate Hyksos pharaoh (see Redford 1992, 106-111). Apophis, the opponent of Kamose within a decade of the expulsion of the Hyksos by Kamose's younger brother Ahmose, reigned for about forty years or more (von Beckerath 1964, 128; Habachi 1972, 59). The beginning of his reign is to be dated between ca. 1620/10 and 1590/80 BC, depending on the Egyptian chronology used. Thus the destruction of the palace at Tel Kabri happened around 1600 BC.⁽¹⁸⁾

Unfortunately we were not able to excavate the palace completely before the temporary end of the project in 1993, but the general picture of the layout and the functions of the different areas of the palace is clear (Fig. 1). The palace covered approximately 2000 square metres and had a roughly rectangular plan. The southern wing had cultic affinities, the eastern wing had a domestic character, and the western wing, centred around Hall 611, had a ceremonial function (see Kempinski 1994). This ceremonial wing shows some similarities to the palaces of Alalakh Level VII and Megiddo Stratum VIII (see Kempinski 1992a, 4; 1994). The outstanding feature of Hall 611 is the painted lime plaster covering its floor as well as that of the threshold of the doorway to the east. This threshold was already excavated in 1987. Kempinski, who - as a Near Eastern archaeologist and Hittitologist - certainly did not have an 'Aegeocentric view', realised that plaster floors with similar techniques and designs are known from the Aegean (Kempinski 1988, vii). Therefore he invited us as Aegean archaeologists to join him in the Kabri excavations. Hall 611, which is a square room of 10 x 10 m., was completely excavated in the seasons of 1989 and 1990 (Fig. 2). The floor was found in a rather poor condition. The uppermost layer with the paint was frequently broken away. The floor was cut along the edges by looters in a late stage of the palace in their efforts to rob the orthostates originally lining the lower part of the walls. Nevertheless the general character of the decoration was clear. The floor was painted with a grid pattern of red lines. The squares forming this grid had a lateral length of 40 cms. and were intended to imitate a pavement of stone slabs. This is especially clear in the central area where the decoration of a group of squares undoubtedly imitated in yellow, red-brown and black the marbling of gypsum slabs, as our thorough investigation with infrared photography and 1:1 drawings demonstrated (Figs. 3-4).

Painted storie imitations regularly occur in Minoan art. The oldest known example on a fresco fragment from the 'Loom-Weight Deposit' at Knossos (Evans 1921, 251-252, fig. 188a; Immerwahr 1990a, 22-23, fig. 6f) is to be dated either to MM II (high chronology: ca. 1900/1875-1750/20 BC; low chronology: 19th century-1700/1650 BC) or MM III (high chronology: 1700/1680-1675/1650 BC; low chronology; ca. 1700/1650-1600 BC).⁽¹⁹⁾ A dado with painted gypsum imitation from the east border of the palace at Knossos was dated by Evans on stylistic grounds to MM IIIA (Evans 1921, 355-356, fig. 255). Of LM IA date are painted gypsum imitations from the West House at Akrotiri on Thera, where they formed dados in Rooms 5 (Doumas 1992, 46, 50-51, figs. 14-17) and 4 (Doumas 1992, 49, 86-91, figs. 49-56),⁽²⁰⁾ and a 'chessboard' pattern in the centre of the north-east wall of Room 4 (Televantou 1994, 156-159, fig. 37, col. pl. 14).⁽²¹⁾ In LM IB-LM II we frequently find painted dados imitating gypsum slabs in the palace at Knossos (see, for instance, Evans 1935, 893-894, fig. 873).⁽²²⁾

In the Near East similar painted stone imitations were found at three more sites apart from Tel Kabri: at Alalakh in Yarim-Lim's palace of Level VII, where they were painted in a second phase on the basalt orthostates of Room 5 of the 'Chamber of Audience' (Woolley 1955, 92), and in a house of Level IV (Woolley 1955, 232, pl. 29a-c); at Qatna in the palace (du Mesnil du Buisson 1935, 143, frontispiece; Stevenson Smith 1965, 18, fig. 31); and in Zimri-Lim's palace at Mari on a podium on the east side of Hall 64 in front of the Throne Hall 65 (Parrot 1958a, 105-106; 1958b, 67-69, fig. 54, pl. 15), on the dados of a passage leading into Court 31 of the 'Royal Apartments' (Parrot 1958a, 165, pl. 39:2), and on a plinth near to the north-west corner of that court (Parrot 1958a, 166, fig. 187, pl. 39:1).⁽²³⁾ The example from Level IV at Alalakh is of fifteenth century BC date (see Klengel 1979, 68-75), that from Qatna probably of fourteenth century BC date (see Stevenson Smith 1965, 17-18). Thus both are later than the period here under consideration.

The oldest known Near Eastern examples are those from the palace at Mari. King Zimri-Lim of Mari was a contemporary of the great Hammurabi of Babylon who conquered Mari in the thirty-fifth year of his reign (Kupper 1973, 14, 28; Gadd 1973, 179-182, 189). The dates of Hammurabi's reign vary between 1792-1750 (middle chronology) and 1726-1686 BC (low chronology). The date of Hammurabi's conquest of Mari is either 1757 or 1691 BC. It is supposed that this conquest forms the terminus ante quem for the Mari murals, although there is some doubt whether the palace was destroyed by Hammurabi or only later, by the Hittite king Mursilis I during his famous raid on Babylon, or by the subsequently installed Kassite dynasty (see Stevenson Smith 1965, 20). The list of regnal years of Hammurabi reports only the dismantling of the city walls of Mari in his thirty-fifth year (see Kupper 1973, 28). If we accept Hammurabi's conquest of Mari as a terminus ante quem for the murals, they are contemporary with the late Cretan Old Palace period when using the Mesopotamian middle and the Aegean high chronologies (end of Old Palace period 1750/1720 BC) or contemporary either with the end of the Old Palace period or with the beginning of the New Palace period when using the Mesopotamian and Aegean low chronologies (end of Old Palace period 1700/1650 BC). The Mesopotamian low and the Aegean high chronologies are not compatible (cf. Manning 1995, 219).

The excavator of Mari, A. Parrot, compared the painted stone imitations to those painted on dados from Knossos (Parrot 1958a, 165 n.2). Elsewhere he sought possible connections between the Mari and the Knossos murals and, pointing to the evidence for connections between Mari and Crete provided by the Cretan precious objects mentioned in the Mari archives (Parrot 1958b, 109-110), he tended to see some Cretan influence in the Mari murals, but was constrained by the dating of the Cretan wall paintings by Aegean specialists to not before 1600 BC. Dussaud, Kantor, Stevenson Smith, Schachermeyr and H. Frankfort were positive about some Cretan influence in the Mari paintings. With the exception of Dussaud, they did not, however, see this in the stone imitations of the dados and plinth in Court 31 and of the podium in Hall 64, but only in the spirals framing the latter (Dussaud 1937, 234; Kantor 1947, 31; Stevenson Smith 1965, 18; Schachermeyr 1967, 41; Frankfort 1970, 126). Crowley has interpreted the painted imitation of stone slabs as well as the spirals of the podium as intrusive elements: "Though...these examples pre-date comparable architectural decoration now available to us from the Aegean area, artistically they can only be viewed as Aegean-influenced pieces" (Crowley 1989, 202, see also 195). Muller/Pierre has discussed the imitation of stone slabs at Mari, Alalakh, Tel Kabri, Crete and Thera within the context of the east Mediterranean koine in wall painting (Pierre 1987, 569, 572; Muller 1994, 53; 1995, 133-135). Her statement that the dominating movement of this koine went from the orient to the Aegean at the beginning of the second millennium BC (with which we generally agree) appears to indicate that she sees Mesopotamia as the place of origin of the painted imitations of stone dados and floors. We believe, however, that the idea of painting imitations of stone dados and floors must have originated in an area where stone slabs were actually used for dados and floors. Bronze Age Mesopotamian architecture was almost exclusively constructed of mudbrick (cf. Frankfort 1970, 18). Stone blocks were used only in foundations and only in places where they could easily be obtained (cf. Frankfort 1970, 42).⁽²⁴⁾ Stone dados (orthostates) are known in Crete and in the Levant at least from the beginning of the Middle Bronze Age on (Hult 1983, 38-39, 46-47, 66; Niemeier 1991, 191 with references), but are unknown in Egypt (Hult 1983, 66). Ashlar floorings were common in Crete, although the known examples are all from the New Palace period (Hult 1983, 47-48, 74). But the MM IB painted imitations of ashlar floors at Phaistos (Levi 1957-1958, 332-333; Hirsch 1977, 17 C43) and Mallia (Daux 1965, 1000-1001, figs. 1-2; Hirsch 1977, C22-23, fig. 4; Poursat 1978, 832, fig. 1; 1985, 892; 1987, 463) suggest that Crete had ashlar floorings already by the beginning of the Old Palace period. Ashlar flooring does not seem to have been common in the Levant, but there are some examples from the beginning of the Middle Bronze Age on (Hult 1983, 40, 73 with references).⁽²⁵⁾ In Egypt, ashlar flooring was very rare (Hult 1983, 36, 73). Thus Crete and the Levant appear to be possible candidates for the origin of painted imitations of stone dados and floors. The fact that we know of painted stone imitations in Crete from the beginning of the Middle Bronze Age on suggests that they were first invented on the island. The oldest preserved examples of this motif in the Near East, those in the palace of Mari, imitate gypsum/alabaster (cf. Muller 1995, 134), a material frequently used in Minoan architecture (see Graham 1962, 4-5, 143-144, 207-208; Shaw 1973, 20-23) but not in the Levant.⁽²⁶⁾

If the fragment from the 'Loom-Weight Deposit' at Knossos is of MM II date (high chronology: 1900/1875-1750/1720 BC; low chronology: 19th century-1700/1650 BC), it is older than the Mari parallels or contemporary with them according to both possible combinations of chronologies (Mesopotamian middle with Aegean high chronologies or Mesopotamian and Aegean low chronologies). If the 'Loom-Weight Deposit' fragment is of MM III date (high chronology: 1750/1720-1700/1680 BC; low chronology: 1700/1650-1600 BC), it is roughly contemporary with or somewhat later than the Mari parallels when using the Mesopotamian middle and the Aegean high chronologies or the Mesopotamian and Aegean low chronologies. In summarising all the evidence available, with Crowley we would see the stone imitations at Mari as a Minoan intrusive element in the palace's decoration. At Alalakh and Kabri the imitations of gypsum slabs are also of Minoan influence. At both places the iconographical context is different, however.

The red lines forming the grid pattern of the Kabri floor probably imitate the red plaster filling the interstices between the slabs of Minoan stone floors (see Evans 1928, 683; 1930, 357-358; Pernier and Banti 1951, 46, 73, 266, 284, 291; Halbherr et al. 1977, 72, 80, 82, 87-88, 90-91, 154, 157, 160). Many of the squares forming the grid pattern of the Kabri floor seemed to alternate white and yellow in a sort of chessboard pattern. As we found out, the yellow colour came from much faded, crowded yellow blossoms painted onto the squares. The dark blue flowers painted in the alternating squares were more easily distinguishable. We were able to discern single blossoms (Fig. 5) and chains of stylised linear iris blossoms of the characteristic Minoan 'V'-type (Fig. 6) which first occurred in LM IA fresco and vase painting⁽²⁷⁾ Some squares show an elegantly curved spray of more 'naturalistic' iris blossoms and yellow crocuses (Figs. 7-8) which also have parallels in LM IA wall paintings (see Evans 1928, pl. X; Cameron 1968, 11, fig. 4e, pl. 5: 1). All these flower motifs are alien to Canaanite iconography in which other flowers, like the lotus of Egyptian origin, were preferred (see Kaplan 1980, 33-34, pls. 126-127; Vermeule and Wolsky 1990, 386-387, pls. 182-183; Weippert 1988, 302-307, figs. 3:51-3:53; Aruz 1995, 40-41, figs. 31-32).⁽²⁸⁾ Thus not only the fresco technique of the floor but also its decoration shows strong connections to the Aegean. Sherratt has problems in accepting "that specifically Aegean art and technology were needed to introduce this technique to a region (the Levant) which had a long tradition of doing exciting things with lime plaster since the Neolithic, and where plaster floors and painted plaster occur in even relatively humble contexts in the 19th century BC" (Sherratt 1994, 238). Lime plaster was indeed known in the Levant from the PPNB on (cf. Nunn 1988, 7 with references), but there is no evidence at all for the use of the fresco technique in the Levant before Alalakh VII and Tel Kabri.⁽²⁹⁾ Sherratt refers to an article by M. Artzy, but there only the covering of olive jars with red coloured plaster is mentioned (Artzy 1987). The technique of this colouring has not been analysed (M. Artzy, pers. comm.). The uniqueness of the painted Kabri floor within the Levant has been stressed by all Israeli colleagues who have visited the site, among them M. Artzy (cf. note 28). Surprisingly, the rigid grid pattern of the Kabri floor filled with painted stone imitations and floral motifs has its best parallels in the Mycenaean palaces of the LH IIIB period which are three centuries later (Hirsch 1977, 24-42; 1980). Such floors have been considered a Mycenaean innovation (Hirsch 1980, 453 n.1, 461). As the Kabri floor demonstrates, this cannot be the case (cf. Niemeier 1996). Many Cretan painted floors have been lost or were not recognised in the old excavations (Hirsch 1977, 5-6), and rigid grid patterns were not unknown to the Minoans (and Therans): strip-like plaster bands with missing inlays forming a strict grid were found in two rooms of the west wing of the palace at Zakros (Platon 1971, 84; Hirsch 1977, 21, fig. 11), and the 'chessboard' stone imitation on the north-east wall of Room 4 of the West House at Akrotiri also shows a rigid grid pattern (Televantou 1994, 156-159, fig. 37, col. pl. 14).

The palace of Tel Kabri had another surprising discovery to offer. On the very last day of the 1990 season we found the first fragments of wall painting in the Doorway 698 leading out from Hall 611 to the north. This doorway was investigated by us in the 1991 and 1992 seasons. Here the ashlar blocks of the original threshold, like that of the doorway leading out from Hall 611 to the east, had apparently been robbed. In the very last phase of the palace the hole had subsequently been levelled with stones and lumps of mud lime, and the whole doorway was finally covered by a very coarse lime floor. From the filling underneath this late floor more than two thousand very small fragments of wall painting were recovered. Our work on the fragments in Heidelberg, where we were able to study them for two years, was very time-consuming and often frustrating. Our initial impression during the recovery proved to be correct: the fragments were not found as fallen from the wall, but in a secondary context, completely jumbled and wantonly crushed to be used together with other debris as filling material.

Little by little we were able to make progress in identifying the motifs represented on these tiny fragments. A series of them was painted with spotted brown colour which on some of them runs out in knob-like protuberances (Fig. 9). The knob-like protuberances form a widespread Aegean convention for representing a rocky shore (cf. Hallager 1985, 16 with references). In Aegean wall painting we find the same motif and a similar colouring in the rocky shore to the right of the 'Departure Town' or 'Polis IV' of the south wall of the miniature fresco from the West House at Akrotiri on Thera (Doumas 1992, 71, fig. 36; cf. Morgan 1988, 34; Televantou 1994, 259-260 (Pl. 3, 0-1.80 m.)).⁽³⁰⁾ On two of our fragments a wavy strip is left white. The same enigmatic motif appears in the rugged red-brown terrain to the right of the 'Departure Town' or 'Polis IV'.⁽³¹⁾ With this comparison we had found the key to interpreting the Kabri fresco fragments, which originally belonged to a miniature fresco with a similar theme to that of the Theran miniature fresco. Since we had identified a rocky shore we ought also to have the sea, and on some of the Kabri fragments we found a grey stippling of loop-like dashes (Fig. 10), very similar to the representation of the sea in the 'shipwreck scene' of the north wall of the Theran miniature fresco (Doumas 1992, 62-63, fig. 29). The curved grey-brown stripes narrowing to one end probably belonged to boats,⁽³²⁾ Other fragments of the Kabri wall painting belonged to representations of architecture. They show isodomic masonry in white and blue as well as rounded 'beam ends' (Fig. 11). The same motifs appear in the town representations on the Theran miniature fresco: isodomic masonry in white and blue, also in red and brown (Doumas 1992, 71 fig. 36, 84 fig. 47, 85 fig. 85; Televantou 1994, 264-274, col. pls. 25, 43-44, 53, 56-57, 67-68 (Pl. 3, 0-1.80 m.)) which may represent ashlar masonry, mudbrick work and/or plastered facades painted with imitations of ashlar masonry (see Morgan 1988, 71-74), as well as 'bea mends' above the gate of the 'Arrival Town' or 'Polis V' (Morgan 1988, 75, fig. 108; Televantou 1994, col. pl. 68). Furthermore, representations of flora and fauna also belonged to the landscape represented in the Kabri miniature fresco. The charming swallow (Fig. 12) has no parallel on the Theran miniature fresco, but the swallow was a very popular theme in Theran wall painting and vase painting (see Immerwahr 1990b, 238-241). Two of the Kabri fresco fragments show parallel horizontal lines with rows of alternating filled triangles and dots in between. This is Evans's 'notched plume' motif (Evans 1921, 548-551), which is applied in different media of Aegean art to the wings of griffins and sphinxes (see d'Albiac 1995, 64-68).⁽³³⁾ The S-spiral on another fragment may have belonged to the neck of a griffin. Thus we have hypothetically restored a griffin in flying gallop and with S-spirals on the neck, similar to that on the east wall of the Theran miniature fresco (Fig. 13) (Doumas 1992, 65 fig. 32; Televantou 1994, 252-254, col. pl. 47).

The representation of the rocky shore does not only follow Minoan artistic conventions, but also shows a typical Aegean landscape (cf. Morgan 1988, 34). The Mediterranean coast of Israel looks very different and is almost exclusively formed by flat plains and sandy shores (see Rogerson 1985, 58-59). Representations of isodomic masonry are not restricted to the Aegean, but are also to be found in the Mari murals. But there they are pink-orange and undoubtedly represent mudbricks (Pierre-Muller 1990, 484-485 fragments M 4592-4594, 4599, 524-525), whereas the white and the blue masonry of the Kabri fragments represent other materials (cf. above). A typical Aegean feature of the architecture represented on the Kabri fresco and not to be found in Near Eastern iconography are the 'beam ends' which have a religious connotation (see Morgan 1988, 75-77). Representations of swallows are known from Thera (see above), Crete (Evans 1928, 379, fig. 211; Ruuskanen 1992, 56-57) and the Mycenaean mainland (Karo 1930-1933, 47 no. 24, pl. 21), but not from the Levant. Griffins and Sphinxes are fabulous creatures which certainly did not originate in Crete but were probably introduced to the island from the Levant (Frankfort 1936-1937; Dessenne 1957, 122-129; Bisi 1965, 171-172; Morgan 1988, 50-53; d'Albiac 1995, 64). In Crete they acquired, however, some characteristic features which are peculiar to the island and were adopted on the Mycenaean mainland. These features include the 'notched plume' wings (Dessenne 1957, 130 no. 294, 133 no. 299; Bisi 1965, 193-195; d'Albiac 1995, 64). Thus all preserved motifs of the Kabri miniature fresco have a purely Aegean character.

Where had the Kabri miniature fresco been situated? As mentioned earlier, the fragments were found in a secondary context, in a filling laid down during the very last, non-palatial phase of the palace.⁽³⁴⁾ We think that the fill material, including the fresco fragments, was not brought from elsewhere, but was formed by debris from this area. Thus the miniature fresco probably belonged, together with the painted floor, to the interior design of Hall 611, and ran along the wall in the zone above the doors in a similar position to the miniature fresco from the West House at Akrotiri.⁽³⁵⁾

THE FRESCO PAINTINGS FROM ALALAKH

The conquest of Alalakh by the Hittite king Hattushili I, reported in his abbreviated annals (see Otten 1960, nos. 1-3; 1964, 118-119), is connected by scholarly consensus with the destruction of Alalakh Level VII (see Kempinski 1983, 217-218; Gates 1987, 73-75). The dates of this destruction recently proposed by different scholars lie between the third quarter of the seventeenth century BC and ca. 1550 BC.⁽³⁶⁾ As Kempinski has demonstrated, Alalakh Level VII - like Tel Kabri - is to be dated to the pre-bichrome phase of the late MB IIB period (Kempinski 1983, 80-88, 218, 220-221; 1992b, 70-72). Thus the destruction of Alalakh Level VII must be roughly contemporary with that of Tel Kabri around 1600 BC or a little earlier.⁽³⁷⁾

The fresco fragments from Alalakh are now kept in the Ashmolean Museum in Oxford to which they were handed in 1957 by the British Museum in London. In the springs of 1994 and 1995 we were able to study the fragments and to take new colour photographs of them.⁽³⁸⁾ In Yarim-Lim's palace the wall paintings came from two rooms, Room 5 of the 'Chamber of Audience' (Woolley 1955, 92), and from a large room above Magazines 11-13, according to Woolley the grand salon of the piano nobile (Woolley 1955, 94, plan facing p. 92). The painted dados in Room 5 imitating stone orthostates have already been discussed in connection with the similar motif on the Kabri floor. Of the fragments fallen from the salon into Magazines 11-13 two restored panels (inv. nos. 1957.36 and 1957.37), probably from the north-east part of the salon, show a horizontal violet band framed by two yellow-brown ones outlined in black. Near the left edge of the panel of inv. no. 1957.36 (Fig. 14, dimensions: 0.29 x 0.16 m.) the yellow-brown horn of a bull is preserved, and above it, on the corner of the restored panel, part of the black curved outline of an object. Woolley found it tempting to see in the bull design an analogy with Knossos, but at the time preferred a north Syrian origin for the motif (Woolley 1955, 231 pl. 38b). The strictly upright position of the horn appears to indicate that an isolated bull's head seen from the front (or a bucranium), rather than a live animal, was represented, as suggested by Woolley.⁽³⁹⁾ He pointed to the long tradition of the frontal bull's head/bucranium motif in Near Eastern painting going back to the pottery decoration of the Halaf culture. However, we do not know of any Near Eastern examples of the motif between Halaf and the wall paintings from the palace of the fifteenth century BC at Nuzi (Starr 1937-1939, 143-144, pls. 128-129; Nunn 1988, 94-95, pl. 73) where the motif is possibly influenced by Aegean prototypes (cf. Crowley 1989, 175-176, 188). In Crete bulls' heads or bucrania are to be found on seals from MM IB/II(Yule 1980, 123-124, pl. 4 motif 3A; Weingarten 1985, 170-173; Onassoglou 1985, 120-128), in vase painting from MM III on (Crouwel and Niemeier 1989, 6-7, figs. 3-4). In fresco painting only two miniature representations of frontal bulls' heads from the palace at Knossos are known (Evans 1928, 742 fig. 475; 1930, 40-41, fig. 25 :a,d). According to Evans, they formed ornaments on the garments of large, seated female figures, whereas Stevenson Smith thought that they may have belonged to representations of architecture (Stevenson Smith 1965, 80). 1£this is true, they reproduce much larger prototypes. Woolley thought that the object between the horns of the bull's head was a disc, and compared the head of the bull in the procession fresco from Court 106 in the palace at Mari (Woolley 1955, 231). However, this has no disc between the horns above the head, but a roundel with revolving rays on the ridge between the horns (Parrot 1958b, 20 fig. 18). This appears to be a representation of a natural cowlick of the sort that we also find on the serpentine bull's head rhyton from the Little Palace at Knossos (Evans 1928, 528 fig. 330). Representations of frontal bulls' heads and (more seldom) bucrania with rounded objects between the horns above the head frequently appear in Aegean art. In Minoan vase painting, seal representations and thin golden cutouts from Shaft Grave IV at Mycenae influenced by Minoan iconography, we see double axes between the horns;⁽⁴⁰⁾ on inlaid Minoan metal vessels we find rosettes in place of the double axes (Vercoutter 1956, 306, pl. 35 no. 231; Davis 1977, 118-123 no. 24, fig. 95, 263-266 no. 109, figs. 210-212). Our attempts to reconstruct the motif have shown that the curve of the black object near the upper left corner of the Alalakh fresco fragment inv. no. 1957.36 is rather flat. When reconstructed as a disc with rosette, this would be rather large. Thus we consider the reconstruction as a double axe the more probable one (Fig. 15). If this is correct, it forms strong evidence for Minoan influence. Sherratt has considered "the unusually large scale of the hull's head at Alalakh" an un-Aegean feature (Sherratt 1994, 237). In our reconstruction of the fragment inv. no. 1957.36, the bull's head with the horns is ca. 22 cm. high and thus considerably under life size. It is approximately the same size as the bulls' heads in Minoan vase painting mentioned earlier, which probably follow lost prototypes in wall painting.

The fresco fragments fallen from the main part of the salon in the south-west, beyond the columns that divided the room into two, have a red background (Figs. 16-18). In this preference for red backgrounds at both Alalakh and Avaris, Sherratt has seen a difference from the Aegean, where red backgrounds are rare according to her (Sherratt 1994, 237). However, red backgrounds are very common in Minoan frescoes from MM III-LM I on,⁽⁴¹⁾ and are also to be found in Mycenaean LH IIIB frescoes (see Rodenwaldt 1912, 217; Lang 1969, 79-81 nos. 43 H 6 and 44 H 6, pls. 125, A). A group of fragments, of which only two fitted together, were set in a panel of plaster of Paris. Woolley described these as follows: "On the normal red ground there are represented, in creamy white, tall tufted grasses blown by the wind; the work is sketchy and impressionistic but the effect is admirably represented" (Woolley 1955, 231, pl. 38a). Today, Woolley's additions have been removed from the panel inv. no. 1957.35 (Fig. 16, dimensions: 0.25 x 0.20 m.). Woolley did not know of an exact parallel from Crete, but stated that "it is unmistakably in the spirit of Cretan art". A recent suggestion to turn around the panel by 90 degrees and to see in it the branches of a palm⁽⁴²⁾ is not possible. According to Woolley's description (1955, 231 n.2), the plaster ends both above and below in a straight line, the back of it showing that it was pressed into an angle. Probably the painting occupied the space between two of the horizontal beams in the half-timbered wall. Thus Woolley's identification of the motif is correct. H.A. Groenewegen-Frankfort coined the pertinent term 'absolute mobility' for the central concept of Cretan art, distinguishing it from the Near Eastern and Egyptian arts (Groenewegen-Frankfort 1951, 195-205), and within this concept the movement of plants as in a wind is indeed a characteristic feature of Minoan fresco painting (see Groenewegen-Frankfort 1951, 197; Schiering 1965, 3; Walberg 1986, 89, 98).⁽⁴³⁾ Thus this painting has been seen by several scholars with good reason as of Minoan influence/origin (Moortgat 1959, 12; Stevenson Smith 1965, 103; Schachermeyr 1967, 46; Nunn 1988, 203; Niemeier 1991, 193; Negbi 1994, 77; Cline 1995b, 269).

Two larger pieces found together in situ as fallen from the grand salon were rather hard to study from the black and white photographs in the publication (Woolley 1955, 230-231, pls. 36b, 37b-c), on which up to now all interpretations have relied. Our thorough study of the original fragments in the Ashmolean Museum, inv. nos. 1957.38 (Fig. 17, dimensions: 0.54 m. x 0.305 m.) and 1957.39 (Fig. 18, dimensions: 0.27 m. x 0.70 m.) lead to surprising results. Woolley concluded of the two pieces that "if they do not actually fit one to the other they must have been practically continuous". On 1957.38 the plain red ground of the greater part of the field ends above in a horizontal white band with an irregular wavy lower contour (Figs. 17, 19). This contour is twice repeated in other much faded colours of which Woolley was able to recognise a greenish grey. The upper straight border of the wavy band is accompanied by a string impression. Above it follows a horizontal ladder pattern in much faded greenish grey, from which slender curved lines project in groups of three. Woolley thought they might be the stalks of (missing) flowers. According to him, "at the right-hand end of the fragment the edge of the lower red mass dips down and the more or less triangular space between it and the 'ladder' was covered with a wash of pale greenish-gray on which twigs and leaves were painted in a darker green". This technique was compared by Woolley to that of the Knossian 'Sacred Grove and Dance' fresco with the representation of olive trees (Evans 1930, 66-69, pl. XVIII). The other fragment was in part terribly broken, and Woolley's description of it was not very clear. He thought that we may "have here the trunk and main part of the foliage of the tree of which some leaves are better preserved on the other fragment". Following Woolley's description and interpretation, Stevenson Smith attempted in a drawing to reconstruct the tree imagined by Woolley (Stevenson Smith 1965, 103, fig. 137). Stevenson Smith's reconstruction has been accepted by several scholars and also re-illustrated, by ourselves included (Nunn 1988, 93, fig. 171; Niemeier 1991, 193-194, pl. 46c; Cline 1995b, 269). Nevertheless it is wrong. The fragment inv. no. 1957.39 is much too small in Stevenson Smith's drawing. Moreover, it has to be turned around 90 degrees, as a ladder pattern like that above the white wavy band of the fragment inv. no. 1957.38 demonstrates.⁽⁴⁴⁾ Fig. 18 shows the correct orientation. A closer observation of the supposed foliage at the righthand end of the fragment inv. no. 1957.38 (Fig. 20) demonstrates that the foliage of the tree of the Knossian 'Sacred Grove and Dance' fresco to which it was compared by Woolley looks very different.⁽⁴⁵⁾ There we see an irregular organic system of branches and twigs with leaves on both sides of them. On the Alalakh fragment we distinguish instead parallel straight horizontal lines with rows of filled curved triangles in between. This is again Evans's 'notched plume' motif, here without the dots between the curved triangle.⁽⁴⁶⁾ Thus the roughly triangular space at the right end of the fragment inv. no. 1957.38 filled with 'notched plume' is nothing other than the wing tip of a winged sphinx or - more probably - of a winged griffin of Aegean type.⁽⁴⁷⁾ The wavy white band with the undulating lower border near the upper end of the fragment is an example of the typical Minoan phenomenon called 'concentric composition' by Stevenson Smith, 'umschliessende Bildform' by Schiering, and 'all-embracing landscape' by Immerwahr, in which terrain motifs also project from above (Stevenson Smith 1965, 73-77; Schiering 1965, 3; Immerwahr 1990a, 41-42; cf. also Waberg 1986, 126-130). As Schiering has demonstrated, the multiple wavy outlines of these projections are influenced by the marbling of gypsum slabs (Schiering 1960, 26-34; 1965, 9-11). As stated by Stevenson Smith, the motif is completely different from the wavy bands between straight lines which separate the registers of figures in the Mari wall compositions (Stevenson Smith 1965, 103; see Parrot 1958b, pl. B:b,e). The homogeneous white colour of the terrain motif of inv. no. 1957.38 at first sight looks rather strange, since Minoan terrain motifs normally show different colours. There is, however, an interesting parallel from the Aegean: the terrain motif projecting from above the representation of the lyre player in the throne room of the palace at Pylos (Lang 1969, 79-80 43 H 6, pls. 27, 125-126, A). This is much later than the Alalakh fragment, but many of the Pylos murals remain in a Minoan tradition and show close relations to Knossos (cf. Immerwahr 1990, 98, 136, 218 n.7). Moreover, we think we have distinguished traces of much faded colour on the terrain motif of the Alalakh fragment, so that it (and the terrain of the Pylos fresco?) may have been coloured with additional al secco painting. Ladder patterns frame Minoan and Theran frescoes from above and/or below;⁽⁴⁸⁾ the 'stalks' arranged at regular distances belong rather to the imitation of stone inlays combined in some Minoan frescoes with the ladder pattern frames.⁽⁴⁹⁾ Al secco painting may originally have been added here, too.

As already mentioned, a ladder pattern is also extant on the other fragment, inv. no. 1957.39 (Figs. 18, 21 bottom). Originally we believed it to be the continuation of the ladder pattern on inv. no. 1957.38, but we subsequently distinguished a terrain motif next to the ladder pattern of inv. no. 1957.39, which is different from the continuous wavy white band of inv. no. 1957.38 and shows tight rounded units representing pebbles or boulders (Fig. 21 above the ladder pattern). Similar terrain motifs form the bases of a MM III-LM IA faience panel from the Temple Repositories at Knossos showing an agrimi suckling a kid (Evans 1921, 510 fig. 366),⁽⁵⁰⁾ of the LM IB-II relief fresco with the charging bull from the Northern Entrance Passage at Knossos (Evans 1928, 171-172, fig. 115; 1935, 16-17, fig. 8)⁽⁵¹⁾ and of the bull hunting and bull tethering scenes on the two LM IB gold cups from the tholos tomb at Vapheio in Laconia (Schachermeyr 1964, pl. 49b-c; Marinatos and Hirmer 1973, pls. 200-207; Davis 1977, figs. 1-10, 12). The latter are of special interest in this context, since motifs comparable to the wavy band of the Alalakh fresco fragment inv. no. 1957.38 hang down from the upper edges of their friezes. Davis has interpreted these hanging motifs as representations of clouds (Davis 1977, 20-23), but this interpretation is not certain (Walberg 1986, 129-130). Be that as it may, the 'pebbles' and 'clouds' of the Vapheio cups form interesting parallels to the corresponding motifs on the fresco fragments inv. nos. 1957.38 and 1957.39 from Alalakh. As described by Woolley, the upper part of inv. no. 1957.39 is terribly broken and several pieces were replaced arbitrarily (Fig. 21). The paint is very faded, but on the righthand side one can again distinguish the 'notched plume' motif.⁽⁵²⁾ It must come from the lower, shorter part of the griffin's wing.⁽⁵³⁾ Re-arranging some of the fragments in our drawing we have reconstructed the griffin in a reclining position (Fig. 22), which appears to be the most probable solution. The terrain motifs and the 'notched plume' wing are characteristic Aegean features. We do not see any characteristic Near Eastern element in the decoration of the fragments inv. nos. 1957.38 and 1957.39.

THE FRESCO PAINTINGS FROM TEL KABRI AND ALALAKH WITHIN THE EASTERN MEDITERRANEAN KOINE

During the third millennium BC, the Levant saw the emergence of local polities as a result of the opening up of trade routes by Egypt and Mesopotamia (see Marfoe 1987), and it was possibly these secondary states which "pioneered longer routes which ultimately reached the Aegean" (Sherratt and Sherratt 1991, 367). Evidence for at least sporadic connections with the eastern Mediterranean in the prepalatial period is supplied by imports of hippopotamus and (possibly) elephant ivory (Krzyszkowska 1983, 165-169; 1990, 36-37, 44, pls. 10, 14), ostrich eggs (Dawkins 1903-1904, 202; Sakellarakis 1990, 289-290, fig. 20), Egyptian stone vessels (Warren 1981, 631-635; Warren and Hankey 1989, 125-127) and scarabs (Warren and Hankey 1989, 129), Byblite daggers (Branigan 1966, 125-126; 1967, 119-121) and a Syro-Palestinian silver cylinder seal (Aruz 1984; Warren and Hankey 1989, 127). These 'exotic' materials and objects most probably served the prestige of an emerging elite.⁽⁵⁴⁾

The uniqueness in the Mediterranean of the development around 2000 BC of the Cretan palace system, to the west of the ancient civilisations of the Near East and Egypt, undoubtedly has to do with connections with and stimulus from the east (cf. Lewthwaite 1983; Cherry 1986, 39-42; Watrous 1987; Sherratt and Sherratt 1991, 354-355). Many concepts and ideas were adopted by the Minoans from the Near East and/or Egypt: those of the palace,⁽⁵⁵⁾ of a palatial administration system working with clay sealings (see Fiandra 1968; 1975; Ferioli and Fiandra 1989; 1990; Weingarten 1990, 105-107) and inscribed clay tablets.⁽⁵⁶⁾ As G. Cadogan has put it, "Crete's palaces would appear to be at the West end of a long line of palaces, palace-temples and temples stretching to the East as far as the Euphrates and the Tigris" (Cadogan 1986, 169). The Minoan iconographical repertoire was influenced, from the Old Palace period on, by the arts of the Near East and of Egypt (see Demargne 1936; Crowley 1989, 181-199, 208-209, 298 table 3; Weingarten 1991). As for the art of pictorial wall painting, it goes back in the Near East to the Neolithic period (see Nunn 1988, 34-60), and in Egypt at least to the Predynastic period (see Kemp 1973; Stevenson Smith 1981, 31). In Crete, the existence of ornamental wall paintings cannot be demonstrated before the Old Palace period in the early second millennium BC, and the existence of pictorial wall paintings not before the beginning of the New Palace period in the eighteenth or seventeenth century BC (depending on the chronology used) (see Immerwahr 1990a, 21-22), although Hood has pointed to primitive conventions in Minoan figural wall painting which may indicate that this art has a longer tradition in the island (Hood 1985). Be that as it may, the common features of Near Eastern, Egyptian and Aegean wall paintings with regard to the colour pigments used⁽⁵⁷⁾ - especially the Egyptian Blue, an artificial frit which was discovered during the Old Kingdom in Egypt and introduced to Crete in the early second millennium BC (see Lucas and Harris 1962, 342; Immerwahr 1990a, 16),⁽⁵⁸⁾ with regard to the general organisation of the decoration, to the superimposed figural friezes, and to some of the motifs which are older in the Near East and in Egypt than in the Aegean, suggest Near Eastern and Egyptian influences (see Stevenson Smith 1965, 100, 129-137; Immerwahr 1985; Pierre 1987, 564-572; Muller 1994, 52-54).

However, already in the time of the Old Palaces, Crete had fine 'objets d'art' and prestige objects to offer to its eastern Mediterranean neighbours: fine palatial Kamares pottery (see Kemp and Merrillees 1980, 1-219; Walberg 1983, 144; Warren and Hankey 1989, 134-135; Walberg 1991b; Wiener 1991, 332; MacGillivray 1995), precious metal vessels,⁽⁵⁹⁾ possibly jewellery⁽⁶⁰⁾ and richly decorated textiles.⁽⁶¹⁾ Importantly, a number of texts in the archives of the palace of Mari - contemporary either with the late Cretan Old Palace period or the very beginning of the Cretan New Palace period (cf. above) - list objects of Caphtorian (Cretan) manufacture, including ostentatious weapons, metal vases, clothing and leather shoes (see most recently Cline 1994, 27, 126-128 nos. D.3-12 with references). The fact that some of these Cretan objects were given by Zimri-Lim as gifts to other Mesopotamian kings, among them Hammurabi of Babylon, can be interpreted as evidence for a "down the line palatial gift exchange" in which the Cretan palaces were involved (Wiener 1991, 328). Later, in the earlier Eighteenth Dynasty, emissaries from Keftiu (Crete)⁽⁶²⁾ are shown in the wall paintings of a series of tombs of high officials in Egyptian Thebes carrying precious objects, among them gold and silver vessels with inlays, swords and textiles, many of them undoubtedly of Aegean origin (Kantor 1947, 44-49; Vercoutter 1956, 305-386; Helck 1979, 64-75; Wachsmann 1987, 55-57, 60-61, 64-66, 69-70, 72-73). In the tomb of Menkheperreseneb these objects are given to the pharaoh by the wr (prince or chief) of Keftiu, who is named together with the princes/chiefs of Khatti, of Tulip and of Kadesh (Vercoutter 1956, 64; Strange 1980, 50-51; Cline 1995a, 146).⁽⁶³⁾ As Cline has convincingly argued, the goods carried by the Keftiu in the Theban paintings were not tribute but gifts from the ruler of Crete to the ruler of Egypt (Cline 1995a, 146-147). The word inw is usually translated as 'tribute', but this should not be taken literally since, according to Egyptian ideology, the pharaoh was not only king of Egypt, but actually the divine king of the world (cf. Strange 1980, 47, 52). As A.R. Schulman has argued in a similar case of inw given by the court of Khatti to Tuthmosis III: "The Egyptians, however, with their characteristic egocentric sense of superiority, would have presented such gifts as tribute" (Schulman 1988, 73 n.55).

This - admittedly somewhat fragmentary - archaeological, textual and pictorial evidence appears to indicate that the Cretan palaces participated from the Old Palace period in interpalatial exchanges in the eastern Mediterranean, although archival evidence from them is as yet missing (due to the fact that Linear A remains undeciphered and probably also due to the character of the preserved texts on the Linear A tablets - cf. the silence of the Linear B tablets on that topic). How are the Aegean motifs in the wall paintings at Mari, Tel Kabri and Alalakh to be interpreted in this context? Sherratt thinks "that the direct influence of textiles on wall, floor and even ceiling paintings would account rather better than the idea of a fresco artist's fanciful transmutations of his observations from nature ..." (Sherratt 1994, 239). We would agree that Aegean textiles were an important component of international trade from early in the second millennium, and that there is evidence for the appearance of figured motifs on Minoan textiles from the beginning of the New Palace period on (see Barber 1991, 320-321, fig. 15:6-7). It is not certain, however, that the fresco fragments from Knossos with miniature bulls' heads, sphinxes and griffins come from representations of women's clothing. They may have belonged to representations of architecture (see above). According to Sherratt, "the links... between the floral elements of the Kabri painted floor and Aegean pottery might equally plausibly be seen in terms of an intermediary textile connection" (Sherratt 1994, 239). But what about the imitation of gypsum slabs on the Tel Kabri floor? This motif has nothing to do with textiles. Nor can the Aegean landscape and architecture of the Tel Kabri miniature fresco fragments have been transmitted via textiles.

The main argument for the suggestion that Aegean artistry was involved in the paintings of the palaces at Tel Kabri and Alalakh is the technique, fresco with additions in secco. Although Hankey is right in saying that very little is known about Near Eastern wall painting between those at Mari and those from Alalakh Level VII, Tel Kabri and Tell el-Dab'a (see above), the - again rather fragmentary - evidence appears to indicate that the fresco technique at Alalakh, Tel Kabri and Tell el-Dab'a is an isolated and rather short-lived element alien to local traditions (see above, with note 2). The combination of the fresco technique and the purely Aegean iconography in the wall paintings at Alalakh Level VII and Tel Kabri cannot be explained without direct reference to Aegean artistry. There are various possibilities: the frescoes were painted by travelling Aegean artisans; they were painted under the supervision of Aegean artists with the assistance of Levantine painters trained by them; they were painted by Levantine painters trained by Aegean masters. It is difficult to decide in each case which of these solutions is the right one. We would agree with Betancourt that only a very small percentage of Cretan frescoes has survived and "...that we are touching the tip of the iceberg of a whole series of interrelated workshops, working in Knossos, the Aegean islands, on the coast of western Asia and in Egypt, perhaps travelling back and forth, perhaps occasionally exchanging personnel or going back to Knossos to learn the most recent things" (Betancourt in the discussion at the end of Bietak (ed.) 1995, 129). The close relationship between the wall painting fragments from Tel Kabri and the miniature fresco from the West House at Akrotiri on Thera, ascribed by E.N. Davis and L. Morgan to a Cycladic school of wall painting (Davis 1990, 221-222; Morgan 1990, 253-258, 263-264), has led Negbi to the conclusion that the painters of the Kabri fresco came from the Cyclades and "belonged to the very class of freelance individuals" (Negbi 1994, 87).⁽⁶⁴⁾ She differentiates the Kabri frescoes, painted according to her by Cycladic freelance artists, from the Tell el-Dab'a frescoes, painted according to her by Minoan palatial (Knossian) artists. There are two problems with this hypothesis. First, Negbi apparently follows the idea according to which Thera was an autonomous trade centre, as suggested by Schachermeyr (1978) and Doumas (1983, 129-131; 1986). However, since on estimate only around a twentieth part of the settlement at Akrotiri has as yet been excavated (see Doumas 1983, 45), it is very difficult to say anything about its social organisation. The view of several scholars, according to which Thera was under Cretan supremacy (for instance, Marinatos 1984b; Niemeier 1984, 206-208; 1986a, 248-249; Wiener 1984, 22-25; 1990, 134-137, 140-150, 153-154), is now supported by recently discovered evidence for the existence of a Minoan-type administration at Akrotiri working with Linear A tablets (French 1994, 69; cf. Manning et al. 1994, 221). Moreover, the Theran miniature fresco belongs to an iconographical cycle of Cretan origin showing a town at the sea and going back to the Knossian 'Town mosaic' (Sakellariou 1981; cf. also Polinger Foster 1979, 99-115). We would agree with C.A. Televantou, that "the Minoan origin of the Thera wall-paintings is apparent in the general character, the range of subjects and the techniques. We are clearly not dealing with an independent workshop, but an offshoot of the Minoan School, which acquired a certain individuality, but always kept its leaves turned towards Crete, its spiritual sun, the centre of the great political and cultural events and achievements" (Televantou 1992, 146). Since only a very small percentage of the Cretan wall paintings is preserved, one can imagine that paintings belonging to the cycle in question also existed in Crete, and that the Kabri fresco was executed by Cretan, not by Theran painters.

How can we explain the occurrence of frescoes in Minoan/Aegean technique and style in the palaces of Alalakh Level VII, Tel Kabri (and Tell el-Dab'a/Avaris)? According to Bietak and Marinatos, we have unconvincingly interpreted this phenomenon as a 'fashion' (Bietak and Marinatos 1995, 60-61), think that the frescoes in question served a more or less decorative purpose (Bietak 1996a, 79), and do not take into consideration the ritual aspect of the paintings (Bietak 1995, 26). This is a misunderstanding of our view. We see the phenomenon of the frescoes in the Levantine palaces, probably painted by travelling Aegean artisans, in the framework of diplomatic relations and gift-exchange between rulers in the ancient Near East, in which the rulers of Crete were apparently involved (see Niemeier 1991, 199-200). The Ugaritic myth, according to which the god of handicrafts, Kothar wa-Khasis, has his seat in Crete and has to be brought from there to build a splendid palace for god Baal and to furnish it with precious works of art (see Niemeier 1991, 199 with references in nn.91-92), points to an esoteric dimension to the phenomenon. In this context, M.W Helms's anthropological investigations on long distance relations, trade and craftsmanship are helpful. As she has demonstrated, the knowledge of distant realms and regions and the acquisition of long distance prestige goods are used by elites for their political advantage (Helms 1988, 3-4, 131-171; 1993, 8-9, 160-170). In this context, artisans travel because elites seek craftsmen from distant locales to enhance their chiefly reputations with their presence and the products of their skills (Helms 1993, 34; cf. also Crowley 1989, 263-265). The acquisition of esoteric knowledge from outside has a special role in these kinds of long distance connections (Helms 1988, 16, 19-160,  261). This may explain the fact that many of the Aegean objects carried by the Keftiu in the Theban tombs have a cultic function in their homeland: rhyta of different shapes (bull's head, lion's head and conical rhyta) and bull statuettes (Vercoutter 1956, 311-321, 323-328, 357-359; Helck 1979, 71-72; Wachsmann 1987, 55-61, 69-70), and that earlier the frescoes from the palaces at Alalakh Level VII, Tel Kabri (and Tell el-Dab'a/Avaris), probably painted by Aegean artists or under their supervision, have a religious connotation in their area of origin (Niemeier 1995, 10).⁽⁶⁵⁾

(1).      On the techniques of al fresco and al secco painting, see Nunn 1988, 5-6.

(2).      Near East: Parrot 1958b, 58, 109; Moortgat 1959, 19; Muller 1995, 133. Egypt: Lucas and Harris 1962, 351-353. Both: Forbes 1955, 235-241.

(3).      Woolley argued here that Crete also owed the best of its architecture to the Asiatic mainland and that members of the architects' guilds were invited to build the palaces of the Cretan rulers. Since architecture is beyond the scope of this Symposium, we have omitted the passages relating to architecture and architects. For the problem of Near Eastern influences in Minoan palace architecture, cf. Graham 1962, 231-232; 1964; Margueron 1982, 582-583; Niemeier 1991, 191.

(4).      On the technique, see Noll, Born and Holm 1975, 93; Asimenos 1978, 576-577; Doumas 1992, 17-19.

(5).      For the technique, see Tagliapetra, Rosano and Cornale 1990; Niemeier et al. 1991; Niemeier and Niemeier 1992.

(6).      On the technique, see Bietak 1994, 45; 1995, 23; 1996a, 75-76.

(7).      For the locations of these areas, see the most recent plans of 'Ezbet Helmi: Bietak 1996a, 69, fig. 55; 1996b, 7, fig. 2.

(8).      The only hitherto published fragments of these are: Bietak and Marinatos 1995, 54, fig. 5, 58 fig. 13.

(9).      Cf., for instance, the International Symposium at the Metropolitan Museum of Art, November 3, 1993, Trade, power and exchange: Hyksos Egypt and the eastern Mediterranean world 1800-1500 BC, in which the Tell el-Dab'a frescoes played a major role. The proceedings of the Symposium have been published by Bietak (Bietak (ed.) 1995).

(10).     The problematic term 'Canaanites' (cf. Gerstenblith 1983, 123-124) is here used in the sense of Amiran (1969,167), who termed the area between the 'Amuq plain to the north and the deserts to the east and south during the Middle and Late Bronze Ages 'Greater Canaan', since it formed a cultural unit (but with many regional variations). De Moor (1971, 53) designates 'Canaan' and 'Canaanites' as misnomers, but thinks that they should remain in use until better terms have been coined.  

(11).     We would like to thank E.H. Cline for sending us a copy of the paper before its publication.

(12).     Cf. our discussion of the problem in Niemeier and Niemeier 1998.

(13).     Avaris frescoes: Negbi 1994, 87-88; Warren 1995, 4-5 (who thinks, however, that the Avaris frescoes were the first such Minoan frescoes to be painted, providing a model for similar work in Crete); Morgan 1995, 29, 30, 44. Kabri frescoes: Negbi 1994, 87; Morgan 1995, 44. Both: Aruz 1995, 33, 34; Cline 1995a, 150; 1995b, 267-269; Dever 1995, 114.

(14).      A.B. Knapp, unpublished lecture at Reading, UK, December 1995, summarised by him in an e-mail communication on 'Aegeanet', 7.3.1996. See also Knapp 1998.

(15).     Bietak's idea was accepted by Hankey (1993b). Sceptical are Cline (1995b, 268-269), M.H. Wiener (in the discussion at the end of Bietak (ed.) 1995, 131-132), Morgan (1995, 44), Niemeier (1995, 10-11). For intermarriages between different regions of the eastern Mediterranean, see Crowley 1989, 265-266; Cline 1995b, 278 with references in n.65; Haider 1996, 149-155.

(16).     See for the traditional low chronology most recently Warren 1990-91; for the new high chronology most recently Manning 1995, 200-229.

(17).     On the controversy about the terminology of the final part of the Syro-Palestinian Middle Bronze Age, see Kempinski 1983, 221-224; Bienkowski 1989 (later MB IIB); Albright 1932, §47; 1963, 83-90 (MB IIC); Dever 1992a, 12-13; 1992b, 46-47 (MB III).

(18).     The latest of a series of radiocarbon dates from the palace, 1638±70 BC, fits the supposed destruction date well (see Bonani 1994).

(19).     In Niemeier 1991, 193 n.33 the find context of the fragment was dated to MM II. For the subsequent correction, cf. Niemeier 1994, 81. The 'Loom-Weight Deposit' appears to be a filling with mixed MM II-III material for the construction of new substructure walls in MM III. Thus the fresco fragment can be either of MM II or of MM III date.

(20).     For reconstructions of the entire decorations of the rooms, see Marinatos 1984a, fig. 17 (Room 5); Televantou 1994, 133-142, figs. 27-32 (Room 4).

(21).     For a reconstruction of this within the entire decoration of the room, see Televantou 1994, 133 fig. 27E-Z, 139 fig. 31, 142 fig. 32.

(22).     This dado in the West Porch belongs to the same pictorial programme as the paintings of the adjoining Corridor of the Procession which are possibly of LM IB date (see Boulotis 1987, 145-147).

(23).     Parrot's interpretation of the plinth as 'jeu de palets' is hypothetical.

(24).     The mudbrick base of the podium in Hall 64 of the palace at Mari was covered by a limestone slab which was covered by the painted stone imitation; see Parrot 1958b, 67; Winter this volume, fig. 1.

(25).     Photographs of the ashlar floors in Royal Palace E and the Western Palace of Middle Bronze Age Ebla are illustrated in Matthiae 1984a, pls. 63, 74.

(26).     The orthostates in early Middle Brouze Age Ebla IIIA are of basalt and limestone (see Matthiae 1984a, pls. 49-51, 55, 64), as they also are in later Middle Bronze Age Alalakh VII (see Naumann 1972, 82).

(27).     On the 'V'-type, see Furumark 1941a, 190. As a textile pattern on LM IA frescoes from Knossos: Cameron 1978, 586-588, pl. 3 C.10; and from Thera: Doumas 1992, 140 fig. 103, 156 fig, 120. In LM IA vase painting: Coldstream 1972, 121-122 Θ15, pl. 31; Marinatos 1974, pl. 70.

(28).     As my partner at Kabri, A. Kempinski, and other Israeli colleagues who visited the site and saw the floor, among them M. Artzy, T. Dothan, A. Mazar, B. Mazar and O, Negbi, have confirmed.

(29).     Cf. Forbes (1955, 241), according to whom "the home country of the real 'buon fresco' paintings was ancient Crete, where murals were executed on wet pure lime plaster since 2500 B.C."

(30).     The town next to the rocky shore is designated 'Departure Town' by Morgan (1988, 12-13) and 'Polis IV' by Televantou (1994, 199).

(31).     Cf. Morgan (1988, 34), who thinks that it may represent a dyke for irrigation.

(32).     Cf. the boats of the south wall of the Theran miniature fresco: Doumas 1992,79, fig. 38; Televantou 1994, 199; Pl. 3, 2.60-3.60 m..

(33).     In Cretan and Theran MM III-LM I fresco painting we know of only one representation of a sphinx (Evans 1928, 41 fig. 25d (the 'notched plume' of the wing is not well preserved but existed without doubt according to Dessenne 1957, 130 no. 294)), but several griffins with 'notched plume' wings: Evans 1921, 549, fig. 400; 1930, 41 fig. 25e; Doumas 1992, 158 fig. 122, 165 fig. 128; Pl. 12.

(34).     Evidence for the non-palatial character of this very last phase is provided by the robbing of the ashlar blocks of the threshold in the doorway leading out of Hall 611 to the north and of the orthostates along the walls of Hall 611. Moreover, storage jars were placed along part of the walls of Hall 611 after the theft of the orthostates.

(35).     For the Thera miniature fresco, see the reconstructions by N. Marinatos (1984a, fig. 17), Televantou (1990, 313-314 figs. 4-6). For an attempt to reconstruct the interior design of Hall 611 of the palace at Tel Kabri, see Niemeier 1995, 10 fig. 1:14.

(36).     Third quarter of seventeenth century BC: Muhly 1985, 25-26 with n.38. 1620/15 BC: Kempinski 1983, 216-221. Ca. 1575 BC: Gates 1987, 75-76; Heinz 1992, 203-212. Ca. 1550 BC: Dever 1992a, 14; 1992b, 48-49; 1995, 114.

(37).     Dever's statement (1995, 14) - given without any reasons - according to which Alalakh VII "is surely later than the Kabri frescoes" is not tenable.

(38).     In 1994, we came to Oxford for the Colloquium in honour of M.S.F. Hood, who as a young archaeologist had been Woolley's collaborator at Alalakh (see Winston 1990, 249-250, 255-256). This event gave us the opportunity to discuss the Alalakh frescoes in the Ashmolean Museum and our ideas about them with several colleagues: P.P. Betancourt, N. Marinatos, P.R.S. Moorey, O. Negbi (cf. Negbi 1994, 77-79, 87-88), R. Laffineur, J. Schäfer, w: Schiering and J. Weingarten.

(39).     Cf. the more diagonal horns of the representations of live bulls in a procession scene on a wall painting from Mari (Parrot 1958b, 19-21, pls. 6, B:a) and in Minoan representations, for instance of bull leaping and bull hunting as well as the luring and tethering of bulls: Younger 1976; Cameron 1987, 327 fig. 12; Marinatos and Hirmer 1973, pls. 200-207; Davis 1977, figs. 1-10, 12.

(40).     Vase representations: Crouwel and Niemeier 1989, 7 figs. 3-4. Seal representations: Evans 1928, 619 n.1; Platon and Pini 1984, 14 no. 11; Gill in Pini 1988, 268 no. 259; Kenna and Thomas 1974, 17 no. 15. Golden cutouts from Shaft Grave IV at Mycenae: Karo 1930-1933, 91-92 nos. 353, 354, pl. 44. For the Minoan iconography of these pieces, cf. Vermeule 1975, 48-49.

(41).     See the 'Blue Boy' or 'Saffron Gatherer' fresco from Knossos (Evans 1921, 265, pl. IV; new restoration, Immerwahr 1990a, pl. 11); the 'Lily fresco' from Amnissos (Marinatos and Hirmer 1973, col. pl. 23) (both possibly of MM III date: cf. for the "Saffron Gatherer', Niemeier 1994, 85 with references in n.155; for the 'Lily fresco', Stürmer 1992, 148-149); the monkey fresco from the 'House of the Frescoes' at Knossos (Evans 1928, 447, pl. X; Cameron 1968, col. pl. A1) (MM III/LM IA); the 'Sacral Knot' fresco from Nirou Chani (Xanthoudides 1922, 11 fig. 9, and restoration in the Herakleion Museum); the yellow lotus(?) fragment from stratum IIA of House I at Ialysos on Rhodes (Monaco 1941, 88-89, pl. IX) (for the date, cf. Furumark 1950, 153-166); the white lily fresco from the so-called first building period of Bronze Age Miletus (Gates 1996, 303 fig. 17; Niemeier and Niemeier 1997) (LM IB); the 'Priest-King' relief fresco from Knossos (Evans 1928, 775-790, pl. XIV (frontispiece); for a new reconstruction, see Niemeier 1987) (for the date cf. Kaiser 1976, 284 (LM IB-II)); the Throne Room griffin fresco from Knossos (Evans 1935, 908-913, pls. XXXII-XXXIII (frontispiece) (for the date, cf. Niemeier 1986b, 67-68; Immerwahr 1990a, 84, 96-98 (LM IB/II or LM II/IIIA1)).

(42).     In the discussion at the end of Bietak (ed.) 1995, 132, E. Porada referred to this suggestion of M.J. Mellink.

(43).     Cf., for instance, the ivy of the 'Cat fresco' from Ayia Triada (Hood 1978, 52, fig. 34) and the lilies of the 'Spring fresco' from Akrotiri (Doumas 1992, 100-105 figs. 66-69).

(44).     Although Woolley illustrated the fragment in the wrong horizontal orientation adopted by Stevenson Smith, he apparently realised the correct orientation. He (Woolley 1955, 230 with n.4) described the upper part as terribly broken so that several of the pieces were replaced arbitrarily. This is true of the upper part of the fragment in our orientation.

(45).    Cf. also other representations of olive trees in Knossian wall paintings: Evans 1928, 167-170, figs. 109B, 110, 111, 113; Cameron 1978, 585, pl.3 A-B.

(46).     Other examples of 'notched plumes' without dots on griffin wings: Evans 1928, 41 fig. 25e; Doumas 1992, 65 fig. 32.

(47).     The rare occurrence of sphinxes in MM III-LM I wall painting has already been mentioned (above, note 33). On the relative rarity generally of sphinxes in comparison to griffins in Minoan art, cf. Bisi 1965, 168.

(48).     Knossos: Evans 1921, 549 fig. 400; Immerwahr 1990, pls. 22, 41, 43. Thera: Doumas 1992, 174-175 figs. 136-137. The ladder pattern of the Alalakh fragments has already been compared with Cretan parallels by Pierre (1987, 570). According to her and to Barber (1991, 325-327), it is of textile origin.

(49).     For instance, Immerwahr 1990a, pls. 41-42. According to Evans (1921, 312-314; 1928, 211 n.2), these imitations of stone inlays are a variant of the scale pattern used as a conventional indication of rocks.

(50).     For the dating of the Temple Repositories, see Niemeier 1994, 79.

(51).     For the dating, see Kaiser 1976, 287-289.

(52).     Woolley (1955, 231) saw this motif as "the main part of the foliage of the tree".

(53).     Cf. the wing of the griffin in the 'Mistress of Animals' fresco from Xeste 3 at Akrotiri on Thera (Doumas 1992, 165 fig. 128).

(54).     For evidence for the emergence of an elite during the prepalatial period, see Soles 1988; 1992, 255-258; Sbonias 1995.

(55).     For differences in the layout between the Near Eastern and the Minoan palaces, see Graham 1964; for differences in function, Hiller 1987.

(56).     The Minoan scripts (hieroglyphic and Linear A) are not of Near Eastern origin, however. Their origin is still obscure: see Olivier 1990, and the responses by Palaima as well as Olivier in the discussion following that paper.

(57).     See for the pigments used in the Near East, Nunn 1988, 25-29; for those used in Egypt, Lucas and Harris 1962, 338-351; and for those used in the Aegean, Immerwahr 1990a, 15-16.

(58).     Egyptian Blue has also been identified in the fresco fragments of Tel Kabri.

(59).     Cadogan (1983, 514) thinks that two silver cups with spiral decoration from the Royal Tombs at Byblos (Montet 1929, 191-192, pls. 111 no. 748, 112 no. 749) are Cretan imports. Some of the silver bowls in the Tôd treasure with their affinities to Kamares ware and to a clay mould for a metal vessel from Quartier Mu at Mallia may also be Cretan imports (see Vandenabeele 1980, 89; Poursat 1984, 87; Warren and Hankey 1989, 131-135; Wiener 1991, 332).

(60).     Walberg (1991a) has suggested that a gold pendant from a Middle Kingdom level at Tell el-Dab'a is of Minoan manufacture. This has, however, been doubted by Aruz (1995, 44-46), who considers the piece as of Canaanite origin.

(61).     As copies painted in Egyptian tombs of the Twelfth Dynasty indicate: see Shaw 1970; Barber 1991, 346-347. Later, in the wall painting of the tomb of Menkheperreseneb, High Priest of Amon under Tuthmosis III and Amenophis II, textiles are among the objects carried by the Keftiu/Cretans: see Vercoutter 1956, 361-362, pl. 62 nos. 469-471; Barber 1991, 335, figs. 15, 19.

(62).     We here follow recent research which almost unanimously identifies the country of Keftiu as Crete; cf. Sakellaraki and Sakellarakis 1984, 199-200 with references, to which should be added Wachsmann 1987. For a recent dissentient view, see Strange 1980, 113-184.

(63).     Vercoutter translates wr as 'king'. Cf. also the princes or kings of Keftiu named in the tomb of Rekhmire: Vercoutter 1956, 57; Strange 1980, 46.

(64).     She refers to Cline (1995b, 278-281), who discusses the possibility of the existence of a class of freelance individuals in the Bronze Age Aegean.

(65).     At Alalakh the religious connotation of the fresco fragments is clear from the motifs represented, the bull's head with double axe(?) (cf. Nilsson 1950, 231) and the griffin (cf. Marinatos 1993, 218-220). As far as the Kabri fresco is concerned, the religious connotation is not so clear, but the parallel from the West House at Akrotiri had a ritual function (see Niemeier 1992, 99-100 with references), as had Hall 611 of the Kabri palace, on the walls of which the fresco was probably painted. On the significance of the bull leaping scenes on the Tell el-Dab'a frescoes, see Marinatos 1994.

-------------------------------------------------

-------------------------------------------------