free web hosting | free website | Business WebSite Hosting | Free Website Submission | shopping cart | php hosting

Alternative models of intercontinental interaction towards the earliest Cretan script

the original article from Black Athena Ten years After: TALANTA XXVIII-XXIX (1996-1997)

Wim van Binsbergen


Alternative models of intercontinental interaction towards the earliest Cretan script

 © 1997, 1999 Wim van Binsbergen


The history of writing systems in the Eastern Mediterranean and the Ancient Near East in general is pertinent to the Black Athena debate in various ways.

                 First, literacy is a dominant feature of the Graeco-Roman classical civilisation, and therefore any exploration of the latter’s historical antecedents is bound to touch on writing systems, their genesis and spread. As a 4th-millennium invention, writing (though not of course the alphabet) at the height of Greek classical civilisation had been in existence for a much longer period than separates us today from the Ancient Greeks. This sobering realisation testifies to the plausibility of the Black Athena thesis claiming extensive ‘Afroasiatic’ (i.e. Ancient Egyptian and Semitic) ‘roots’ for classical civilisation. At the same time it makes it understandable why such a claim, in its generality, had already been contemplated among non-classicist students of the Ancient Near East, for decades preceding Bernal;[2] it is the powerful and synthetic, multidisciplinary phrasing of this claim, with full realisation of its implication for multicultural identity politics today, and with emphasis not so much on Syrian, Canaanite, Anatolian and Mesopotamian but on alleged Egyptian contributions, which mark Bernal’s originality and constitute so many bones of contention.

                 Secondly, Martin Bernal himself has shown an intense interest in the topic, not only in the two volumes of Black Athena published so far,[3] but also in a separate study,[4] which contrary to conventional wisdom in this field argues the early (mid-2nd millennium) introduction of the alphabet to the Aegean and beyond. This ‘by-product’ (p. xi) of Bernal’s project has been rather aloof from the spate of criticism, although it has not escaped dismissal by one of his principal foes, James Muhly.[5] Cadmean letters is in many ways a most interesting product of scholarship. While also here Bernal cannot help engaging in excursions on the sociology of knowledge attending scholarly theories of the history of the alphabet, the overall argument is characterised by such methodological rigour and such command of the entire corpus of relevant data (all ancient alphabetic and syllabic scripts of Southwest and Northwest Asia, North and Northeast Africa, and Europe) that it contrasts with the Black Athena volumes, on whose broad canvas rapid and thin brush strokes sometimes had to suffice — with predictable and justified methodological objections from the specialists. This reminds us of the need to see Martin Bernal’s work as an evolving oeuvre all of whose parts must be taken into account, rather than as a series of disparate works. Against the occasional suggestion of Egyptocentrism of the Black Athena volumes, Egypt scarcely features in Cadmean letters. On the basis of a wave theory of transmission making for multiple, successive centres of transmission and reorientation, the role of the Levant is highlighted, and considerable justice is done to the multicentred nature of cultural exchange in the ancient Eastern Mediterranean. Bernal casts new light on the Greek letter names, and from the erratic distribution pattern of individual letter forms in all the alphabets under consideration, derives the conclusion phrased in his subtitle. In fact, if anything, Egypt features too little in this study, for Bernal’s refusal to go beyond a subtly and convincingly reconstructed (but not systematically tabulated) ‘alphabet of primary transmission’ in what can only be a West Semitic context, prevents him — with considerable tactical insight, no doubt — from raising the ultimate question of origin of the alphabet. The same reticence almost reduces the decipherment of the Sinaitic proto-alphabet by Martin Bernal’s grandfather the leading Egyptologist Alan Gardiner (on the basis of an acrophonic use of Egyptian hieroglyphic signs for Semitic phonemes)[6] to a mere family anecdote worth only a dedication, six lines of text (p. 115), and a chatty appendix based on family papers. This may well obscure from the reader’s consciousness the essentially Egyptianising context of Sinai and Palestine in the first half of the second millennium BCE. A fundamental point of departure for Bernal in Cadmean letters, and one with which I do disagree, is the idea[7] that a script is as old, not as its youngest sign (which marks the completion of that script as a full and integrated package), but as its oldest sign. This assumption allows him to view the history of alphabetic diffusion — with, I admit, impressive success — in an extremely fragmented fashion: as the successive ephemeral resultants of the interlocking diffusion histories of individual signs.

                 Thirdly, the very nature of writing systems as conventionalised systems of signs — as formal systems — allows them (much like other formal systems e.g. board games and divination practices)[8] to extend and ramify widely in time and space beyond the rather more conservative boundaries within which culture-specific and language-specific systems of localised meaning tend to be contained. We may fruitfully study writing systems even externally and ignorantly, looking for formal clues in their distribution, patterning and structuring, even when we are still unable to gauge the specific meanings they may have had for their original users. The tendency of formal systems to cross cultural and linguistic boundaries, as well as their proneness to leave permanent, indelible material traces, make them useful ‘guiding fossils’ in the search for historical interconnections in time and space. All the same we have to remind ourselves of the probability that earliest forms are virtually never preserved; and that specifically the earliest forms of writing may have been on perishable materials (wood, leaves, shells) only later to be replaced by attested carriers which, while of less perishable material, in their external shape may still retain a reminiscence of their vanished predecessors.[9]

                 In the study of ancient writing systems from the Eastern Mediterranean, Crete occupies a pivotal position: by virtue of its geographical situation between three continents; as the principal home of Linear B; as an early meeting ground of Indo-European, Afroasiatic (including Semitic, and for Crete perhaps also Egyptian) and possibly other language groups; and by consequence as an academic battle ground. While Egyptian influences have been recognised (although they have remained somewhat elusive) in such fields as architecture, stone vessels, cultic symbols and practices, and myth,[10] until recently no substantial Egyptian influences were claimed for the oldest Cretan script, as it appears on seals c. 2000 BCE. Even Bernal, with all his searching for Egyptian presences in the Aegean in the third and second millennium BCE and his specific argument[11] on extensive Egyptian presence on Crete, discusses Cretan writing systems without reference to Egypt.[12]

                 Meanwhile, however, Jan Best, in a passionate argument whose style and approach recall Cyrus Gordon’s classic book on the subject,[13] has called the scriptural isolation of Crete a myth, and has proposed[14] to base a provocative and contentious reading of the earliest Cretan script on the identification of three distinct sources for its signs:

    35 signs as derived from Egyptian Hieroglyphic

    30 signs from Luwian Hieroglyphic and

    10 signs from hieroglyphic and linear scripts from Byblos.

                 Best goes on to interpret the political culture of early Palatial Crete by idiosyncratically identifying ‘Egyptianising’ elements on Cretan signs: the bee interpreted as a symbol of kingship,[15] the beer pot interpreted as a symbol of the high-ranking Egyptian court office of ‘butler’, cup bearer or steward,[16] symmetrical convolutions as a ‘streamlined’ symbol of the goddess Hathor with her head-dress of bovine horns,[17] etc. Beyond Bernal’s wildest dreams, Best conjures up a distinct Egyptian mercantile, perhaps even politico-tributary, presence for early Palatial Crete, on the basis of the scrutiny of the multi-sided seals which already afforded Arthur Evans[18] an occasion to apply his proverbially unusual (for as incisive as myopic) powers of perception.

                 The most problematic feature of such an Egyptianising interpretation of the earliest Cretan texts consists in its claim of a direct, unadulterated, unfiltered access to Egyptian lexical, scriptural, cultural and institutional elements by the 20th century BCE.

                 What are the technological requirements for such access in terms of nautical technology? Homer[19] shows that by the early Iron Age direct navigation between the Aegean and Egypt was taken for granted. The Thera frescoes, with what has been interpreted as North African scenery and human physical types as well as Egyptian boats, may carry that suggestion at least half a millennium back into the Bronze Age.[20] Bronze Age nautical techniques in general are reputed to have preferred coastal navigation with nocturnal breaks ashore, so that the sea route from Egypt to Crete would have started along the Libyan coast, followed — from what is today the port of Bard¤iyah and R¤as (Cape) al-Mur¤aysah on the eastern Cyrenaican coast — by the shortest possible north-bound traject;[21] or alternatively, and unlikely for Egyptian ships in the light of political circumstances, all the way along the Levantine and Anatolian coast. Helck & Drenkhahn,[22] however, question this conventional wisdom and argue in favour of direct Egypt/ Aegean navigation across the high seas, which, in view of the constancy of nautical technology since the Neolithic, they think may even be very old; however, they do not enter into a discussion of the techniques for determining a ship’s position, the impact of the growth of astronomy in this connection,[23] etc. The distinct find patterns of Egyptian, Syrio-Palestinian and Anatolian goods on Crete as summarised by Best,[24] with concentrations of geographical provenance concentrating on the parts of the island nearest to these directions, also suggest a plurality of access routes to the island existing side by side.

                 With this in mind, let us try to spell out the implications, both geographically and systematically, of Jan Best’s claim — plausible in itself — concerning the Egyptian, Luwian and Byblian components in the earliest Cretan script.



Model I. Crete as the supposedly unique geographical locus of transformative localisation[25] of the intercontinental contributions towards the earliest Cretan script

The simplest model, as originally envisaged by Jan Best, stipulates that the formative contributions from the three different sources travelled independently, along separate routes, to Crete and only there were integrated, and transformed, so as to constitute Cretan Hieroglyphic. Possible routes are set out in my diagram 1:

     from Egypt, either via Cyrenaica (A); direct (B); via Byblos (C) then direct (D); or via Byblos, and from there by any of the coastal routes (E/ F/ G-H-J);

     from Byblos, either direct (D); or via any of the coastal routes (E/ F/ G-H-J);

     from Anatolia, via the coastal route H-J.

 Diagram 1. Crete as the supposedly unique geographical locus of transformative localisation of intercontinental contributions towards the earliest Cretan script



                 Route C is well-attested and in fact Egyptian influence in Byblos was so overwhelming that Byblian Hieroglyphic[26] unmistakably derived from Egyptian Hieroglyphic. Diagram 1 merely sums up the various theoretically possible routes for intercontinental contributions to the earliest Cretan script. As my argument develops most alternatives will be discarded and one route will emerge as the most likely one.

                 From a nautical and archaeological point of view there would be little objection against the Egyptianising reading proposed by Jan Best. Also Gordon made allowance for occasional hieroglyphic readings of Cretan material.[27]Best’s conclusion ties in with Bernal’s prudent claim that

‘it is possible that the was Egyptian suzerainty over Crete and the Cyclades during the Egyptian Middle Kingdom’.[28]

Best’s reading of the name of a major Cretan seal owner as ‘Cat-Snake’ may even appear in a new light once we realise that the Feline (an epiphany of the sun god Re‘) and the Serpent (as an epiphany of the powers of darkness) form a conventional pair of adversaries in Egyptian mythology.[29] With both feline and snake thus having clear Egyptian connotations, the contentious combination of these arch-enemies in one name might in principle be interpreted as the kind of cosmological bricolage one might expect to find at the early Aegean periphery of a developed Egyptian culture. But on further reflection the interpretation remains highly problematic. Despite Best’s ingenuous invocation of a principle of interpicturality, the ‘cat’ — so prominent on one of the seals that it was honoured to be stamped in gold on the cover of Arthur Evans Scripta Minoa[30] from which most of Best’s material derives — is scarcely if at all visible on at least one seal he discusses in his article. It would have been equally hard to find on the ground, on Crete in the beginning of the second millennium. Although the wild feline (lion or wild cat) as sacred to the goddess Bastet is attested throughout Ancient Egyptian iconography and mythology, domestication of the cat only took place very late, and it is only after 1000 BC that its iconography settled accordingly for a domestic cat. The authority on the history of domesticated animals, Zeuner, states:

‘In the New Kingdom (sixteenth century onwards), however, the cat appears as a domesticated animal, helping to hunt birds and sacred to Bastet or Bubastis, a goddess of the delta. (...) Some archaeologists indeed hold that the cat was domesticated in Egypt from the first dynasties onwards (c. 3000 B.C.) but the evidence is ambiguous. (...) But by eighteenth-dynasty times the cat has become popular and properly domesticated. (...) Following the intense traffic from Egypt across the Aegean Sea (...) cats actually reached Greece from time to time. The earliest record appears to be one from Crete, where a terracotta head of late Minoan age has been interpreted as that of a cat by Bosanquet (...).[31] It comes from Palaikastro and should be earlier than 1100 B.C. At that time domestic cats abounded in Egypt, and this find may provide another cultural link between that country and Greece.’[32]



Model II. Byblos and the North Syrian/ South East Anatolian coast as distinct focal points of transformative localisation of the intercontinental contributions towards the earliest Cretan script

Model I is strikingly implausible for more systematic reasons than Egyptian iconography and the geographical distribution of mammals. In my opinion as an anthropologist, a script is not merely an arbitrary and ephemeral conglomerate of signs that happen to be present at the same time and place, but a systematic and integrated package, all of whose elements have a specific, interdependent function; it can only function culturally by virtue of these characteristics. One can hardly imagine that knowledge of three writing systems arrived independently on Crete, c. 2000 BCE, and only there was arbitrarily and eclectically used to construct Cretan Hieroglyphic with elements from all three. Specifically, if a script combines a sizeable proportion of Egyptian and Byblian signs, our most likely hypothesis is that such a script was formulated in what was, geographically and culturally, the Egyptianising context of Byblos. Then again, if the script incorporates a sizeable proportion of Luwian signs, our most likely hypothesis is that such a script was reformulated in what was, geographically and culturally, a Luwian context.

                 Massive Egyptian influence did travel to and via Byblos. This means that routes A and B in diagram 1 can be discarded as far as Egyptian scriptural influence upon the earliest Cretan script is concerned; of course this does not say anything against the mercantile and cultural utilisation of these routes in general, outside the context of the origins of the earliest Cretan script. It also means that the Egyptian and the Byblian influence did not travel along separate routes to Crete. More probably, they were already amalgamated in Byblos in some provisional, hitherto unattested form (which I provisionally designate ‘*proto-Cretan I’; see below) long before reaching Crete.

                 Another shunting point comparable to Byblos would appear to be the Upper Syrian coast, where according to Woudhuizen[33] Luwian Hieroglyphic originated. It seems most probable that here the provisional package of *proto-Cretan I was transformed as a result of combination with a further substantial contribution from the Luwian Hieroglyphic which by that time (the end of the 3rd millennium BCE) was in statu nascendi. In other words, in terms of my proposed alternative model the intercontinental contributions towards the earliest Crete script, from Egypt, Byblos and North Syria were amalgamated, not in Crete, but (after an earlier stage in Byblos) in North Syria and hence travelled, as a package, to Crete (diagram 2) via coastal navigation.

                 Recent research is meanwhile suggesting an important economic incentive behind what looks like intensified maritime connections between Crete, Luwian lands, Byblos and Egypt around 2000 BCE as shown in this diagram: the general shift to tin-bronze precisely at this time, and the crucial role in this respect of a recently discovered tin mine at Goltepe, South Anatolia.[34]

Diagram 2. Byblos and the North Syrian coast as two distinct focal points of transformative localisation of the intercontinental contributions towards the earliest Cretan script.

                 The model propounded in diagram 2, although greatly narrowed down as compared to diagram 1, insofar as geographical routes of inter­continental contribution are concerned, is still unsatisfactory in that it depicts the three constituent influences on the earliest Cretan script as travelling separately and parallel to each other. In view of my emphasis on the script being an integrated package, a more plausible model emerges from diagram 3, which visualises the genesis of proto-Cretan II (= Cretan Hieroglyphic as attested) as the result of successive and accumulative transformations, first in Byblos (where the unattested *proto-Cretan I was formed), then in coastal North Syria. This resulted — in all likelihood: still on the Upper Syrian coast — in ‘proto-Cretan II’, which however so far has only been attested from Crete, 1000 km to the west, under its accepted designation of Cretan Hieroglyphic; unless we consider Luwian Hieroglyphic (40% overlap with Cretan Hieroglyphic) as a mere variant of the latter.

                 Diagram 4 finally projects the model underlying diagram 3 back onto the map of the eastern Mediterranean:

Diagram 3. An unattested *Proto-Cretan I script phase as an implication of Model II.



Diagram 4. Schematic geographical connections involving a *proto-Cretan I scriptural phase in the formation of the earliest Cretan script (= proto Cretan II).



Conclusion: From Egypt via the Levant, with additional contributions from Anatolia, to Crete

The emerging model of intercontinental interactions towards the earliest Cretan script would appear to have applicability beyond the emergence of Cretan writing alone. Model I would amount to claiming an extensive, direct and unfiltered Egyptian influence on Crete, in terms of which we would be justified to speak (as Jan Best does) of ‘Egyptianising’ — not only with regard to the Cretan Hieroglyphic, but also with regard to the cultural, political and cultic contents expressed in that script and left for us to decipher and interpret. If however, as I have proposed through my increasingly complex Model II, any Egyptian influence on the earliest Cretan script was filtered through two successive transformations effected far away from Egypt in contexts only considerably (Byblos) or even scarcely (North Syria)[35] informed by Egyptian culture c. 2000 BCE, then a very different interpretation presents itself. Egyptian influence is then relegated to a status of remoteness and indirectness, and while it still transpires in the purely formal characteristics of part of the scriptural signs, it can no longer be claimed to largely, let alone fully, determine cultural contents.

                 If Model II is the more plausible one, then the cultural, political and religious meanings expressed in the earliest Cretan script could scarcely be direct, even detectable, reflections of Egyptian institutions at the time. An Egyptianising reading as proposed by Best then becomes implausible. The long (nearly 2000 km) detour postulated by Model II involves substantial transformation and amalgamation of scripts, while along the way these scripts were almost certainly used for languages very different from Egyptian (else Egyptian script would have been retained). Under such conditions, it is highly improbable that such specifically Egyptian semantic complexes like kingship and stewardship (as rendered by strings of Egyptian Hieroglyphic signs often — not always: writing variants are characteristic of the Egyptian script — featuring a bee, a beer pot, etc.) could have reached Crete while retaining much of their original form and contents. Instead, the attending signs are much more likely to have lost such iconographic connotations as they once had in the original Egyptian cultural environment. They must have become highly conventionalised, retaining hardly any reference to Egyptian institutions and to the Egyptian lexical items designating such institutions.

                 Of course, behind this reasoning is the whole intricate question of the nature of hieroglyphic signs as symbols corresponding with the real-life items that many of the individual signs appear to represent, or alternatively as mere conventionalised sings having only phonetic or lexical value. Decipherment of Egyptian Hieroglyphic texts only became possible once it was realised, by the end of the 18th century, that these contained primarily signs, not symbols. However, much of Ancient Egyptian funerary ritual and magic[36] was based on the idea that symbolic qualities remain lurking in the signs and can be activated; similarly, such activation was supposed to be prevented by careful and consistent mutilation of the signs as if of living creatures themselves. It is quite likely[37] that in peripheral, far less literate or even illiterate conditions, e.g. such as obtained when Egyptian script was taken to distant Crete c. 2000 BCE, magical and symbolic elements become stressed over the sign-oriented technicalities of the script. Jan Best clearly takes recourse to the assumption of symbolic qualities clinging indefinitely to the signs, even after diffusion. Considering the very long route of my diagrams 2 and 4, such an argument strikes me as unnecessary.

                 More in general, my model II reinforces the view — which in the context of the Black Athena debate has been expressed by Sarah P. Morris among others — that Egyptian influence on the Aegean was by and large not a direct one, but was mediated via Palestine and Syria:

‘In other words, these two sets of pictorial fragments in Aegean style [from Tell al-Dab‘a, i.e. Avaris, in the Nile Delta; and from Tell Kabri, Northern Israel] clearly reveal the strong connections between Minoan Crete (Keftiu, Kaphtor) and the northern Levant, rather than directly between Crete and Egypt.’[38]

‘Bernal’s view of the ancient equivalent of such a route leapfrogs from Egypt to Greece, disregarding more critical connections via the land of the alphabet’.[39]

                 In a very critical yet fair discussion of the Black Athena thesis, Mario Liverani goes beyond such a programmatic statement and actually presents a dynamic historic model (properly periodicised into Late Bronze, Iron I+II, and Iron III, each with very different structural characteristics) that highlights the specific shifts in regional economies and state systems in the course of which the Levantine rather than directly Egyptian influence on Greece must be situated.[40] Meanwhile we should not exaggerate the difference made by explicit allowance for the Levantine contribution, in view of the fact that cultural influence of Egypt in these regions was considerable — as brought out, for instance, by the hieroglyphic background of the alphabet itself.

                 This is not to deny the importance of the Egyptian, or more in general African, contribution in third and second millennium BCE inter-continental cultural interactions including those leading to the earliest Cretan script, but to call attention to the transformative localisations (involving amalgamation with other influences locally available) this — and presumably other — Egyptian material underwent, before and after it reached the Aegean. Here we encounter the problematic invariably attending diffusionist arguments in the study of culture: the argument of provenance, of diffusion, always needs to be complemented by the argument of transformative localisation once a destination has been reached.[41] Elsewhere in this collection I argue[42] how Martin Bernal has acknowledged this insight under the heading of ‘modified diffusionism’.

                 If, as I argue for Cretan Hieroglyphic, there have been two intermediate destinations serving as focal points of transformative localisation of Egyptian scriptural influence before it could even reach the ultimate destination Crete, such intervening localisation will have substantially eroded and adulterated whatever original cultural contents were there to be diffused. In the process, reference to specifically Egyptian cultural and institutional features most probably underwent shifts to such an extent that the Egyptian elements were reduced to mere formal correspondences, were far from ‘overwhelming’[43] and must not be read as evidence of massive influence of ‘Egyptian culture’ per se.

                 An exploration of non-Egyptian contributions to the oldest Cretan script would add further relief to Jan Best’s argument. Particularly a Mesopotamian contribution is at least suggested by the following elements in his analysis. First there is his claim as to iconographic evidence of a conception of hybrid creatures, ‘Mischwesen’, a prominent and persistent feature in ancient Mesopotamian culture[44] but much less so in ancient Egyptian representations. Moreover the — apparently Luwian — sign consisting of two isosceles triangles, probably should be read not as a reference to any Egyptian ‘pyramid city’. The latter is an abode of the dead rather than of the living;[45] the string of Egyptian Hieroglyphic signs depicting ‘town’ (dm) is totally different — and in general a detailed analysis of ‘Gardiner’ hieroglyphic signs as invoked by Best would reveal such a looseness of fit as to defy any notion of direct and massive Egyptian influence.[46] Instead the triangles may, for instance, derive from the early forms of the Sumero-Akkadian cuneiform writing system, which contains several signs with triple or dual triangles.[47] The Luwian connection, meanwhile, and particularly the emphasis on stag or deer and other animals, suggests a second set of non-Egyptian/ non-African contributions besides Mesopotamia: those linking up with the animal style/ steppe / shamanistic complex and thus with Central and Northern Asia.[48]

                 Although there is no prima facie reason why the antecedents of the earliest Cretan script should provide a widely applicable heuristic model, the possible implications of my proposed Model II for the Black Athena thesis are obvious. Athena (who stands metonymically for the Greek, subsequently European, subsequently North Atlantic and increasingly global civilisation) may not have been autochthonous[49] but that in itself (as the argument of origin and diffusion) does not necessarily give her predominantly or exclusively Egyptian, i.e. African, ancestry — the more likely model providing for multiplex, multidirectional cultural exchanges. Perhaps more important even is the argument of transformative localisation: it is doubtful, as I maintain for the specific case of Crete’s oldest script, whether any such transcontinental ancestry would have been able to decisively determine substance, beyond mere remote reminiscences so eroded as to become conventionalised, superficial, and purely formal. Both the multidirectional element and the transformative element are present in Martin Bernal’s approach, but not always with such clarity and emphasis[50] as to withstand the temptation of reformulations and appropriations stressing unidirectionality and passive reception.



© 1997, 1999 Wim van Binsbergen

[1]I am indebted to Jan Best for generous and intensive discussions from which the present argument originated; to Arno Egberts for Egyptological advice; to Peter Broers for advice on West Semitic languages; and to the theme group on ‘Religion and magic in the Ancient Near East’, Netherlands Institute for Advanced Study in the Humanities and Social Sciences’, 1994-95, for creating an inspiring setting in which I could pursue this topic and related topics.

[2]Cf. Wim van Binsbergen, ‘Black Athena Ten Years After: Towards a constructive re-assessment’ (this volume).

[3]Bernal, M., 1987, Black Athena: The Afroasiatic roots of classical civilization, I. The fabrication of Ancient Greece, 1785-1985, London etc.: Free Association Books, passim (see that book’s index); Bernal, M., 1991, Black Athena: The Afroasiatic roots of classical civilization, II. The archaeological and documentary evidence, New Brunswick (N.J.): Rutgers University Press, passim.

[4]Bernal, M., 1990, Cadmean letters: The transmission of the alphabet to the Aegean and further west before 1400 B.C., Winona Lake: Eisenbrauns; cf. Bernal, M., 1987, ‘On the transmission of the alphabet to the Aegean before 1400 B.C.’, Bulletin of the American Schools of Oriental Research, 267: 1-19.

[5]Muhly, J.D., 1990, ‘Black Athena versus traditional scholarship’, Journal of Mediterranean Archaeology, 3, 1: 83-110, p. 92f.

[6]Gardiner, A. H., 1916, ‘The Egyptian origin of the Semitic alphabet’, Journal of Egyptian Archaeology, 3: 1-16; Cerny, J., 1971, ‘Language and writing’, in: Harris, J.R., ed., The legacy of Egypt, 2nd ed., Oxford: Clarendon, pp. 197-219, 214f and table ‘The alphabet’ at the beginning of that book.

[7]Bernal, Cadmean letters, p. 12; the idea derives from: Ullman, B.L., 1934, ‘How old is the Greek alphabet?’, American Journal of Archeology, 38: 359-81.

[8]Cf. van Binsbergen, W.M.J., 1997, ‘Rethinking Africa’s contribution to global cultural history: Lessons from a comparative historical analysis of mankala board-games and geomantic divination’ (this volume); van Binsbergen, W.M.J., in press, ‘Board-games and divination in global cultural history: A theoretical, comparative and historical perspective on mankala and geomancy in Africa and Asia’, in: Finkel, I., ed., Ancient board-games, London: British Museum Press; van Binsbergen, W.M.J., 1996, ‘Time, space and history in African divination and board-games’, in: Tiemersma, D., & Oosterling, H.A.F., eds., Time and temporality in intercultural perspective: Studies presented to Heinz Kimmerle, Amsterdam: Rodopi, pp. 105-125.

[9]Helck, W., revised by R. Drenkhahn, 1995, Die Beziehungen Ägyptens und Vorderasiens zur Agäis bis ins 7. Jahrhundert v. Chr., Darmstadt: Wissenschaftliche Buchgesellschaft, 2nd rev. ed., p. 15f.

[10]Cf. Bernal, Black Athena II, pp. 154ff and references cited there.

[11]Bernal, Black Athena II, pp. 164-186; for additional arguments cf. Helck rev. Drenkhahn.

[12]Bernal, Black Athena II:

‘This independence [of Crete] is reflected in the fact that palatial Crete did not adopt Egyptian hieroglyphics, cuneiform or a Byblian script, but used its own hieroglyphic and syllabic systems.’ (p. 162)

[13]Gordon, C.H., 1966, Evidence for the Minoan language, Ventnor (NJ): Ventnor Publishers.

[14]Best, J.[G.P.], 1997, ‘The ancient toponyms of Mallia: A post-Eurocentric reading of Egyptianising Bronze Age documents’, (this volume); cf. Best, J.G.P., & Woudhuizen, F., 1988, eds., Ancient scripts from Crete and Cyprus, Leiden: Brill; Woudhuizen, F., 1989, ‘The Cretan branch of Luwian hieroglyphic’, in: Best, J.G.P., & Woudhuizen, F., 1989, eds., Lost languages from the Mediterranean, Leiden: Brill, pp. 65-138; Best, J.G.P., 1997, ‘The Luwian branch of Cretan writing’, Amsterdam: Najade Press, working paper.

[15]Because the word ‘bee’ (bt) and the ‘bee’ sign (Gardiner, A.H., 1957, Egyptian Grammar, 3rd ed., Oxford: Oxford University Press, p. 477 sign L2, cf. p. 50, 51, 73) features in the titles ‘King of Lower Egypt’ and ‘King of Upper and Lower Egypt’. The general, unmarked Egyptian word for ‘king’ however, nsw, as would probably be applicable to rulers outside Egypt e.g. on Crete, for historical reasons implies a reference to Upper Egypt and has no bee connotations whatsoever.

[16]Gardiner, p. 530, sign W23.

[17]Cf. Gardiner, sign C9, p. 449; horns surrounding a circle (iconographically and sculpturally not the absolute prerogative of Hathor) make up only a small part of this sign, which essentially depicts an entire seated woman seen in profile. The Egyptologists Helck and Drenkhahn specifically deny the Hathor connection, o.c., p. 19; Goodison however does see at least the Egyptian goddesses Isis and Nephthys on Cretan seals: Goodison, L., 1989, Death, women and the sun: Symbolism and regeneration in early Aegean Religion, London: Institute of Classical Studies.

[18]Evans, A., 1909, Scripta Minoa, I, Oxford: Clarendon Press.

[19]Od. 3, 285f.

[20]Immerwahr, S. A., 1983, ‘The people in the frescoes’, in 0. Krzyszkowska and L. Nixon, eds., Minoan Society: Proceedings of the Cambridge colloquium 1982, Bristol: Bristol Classical Press, pp. 143-154; Morgan Brown, L., 1978, ‘The ship procession in the miniature fresco’, in C. Doumas, ed., Thera and the Aegean world: Papers presented at the second international scientific congress, Santorini, Greece, August 1978, vol. I. London, pp. 629-644; Marinatos, S., 1969, ‘An African in Thera’, Analekta Archaiologika Athenon, 2, pp. 374-5; Marinatos, S., 1973b, ‘Ethnic problems raised by recent discoveries on Thera’, in R. A. Crossland and A. Birchall, eds., Bronze Age migrations in the Aegean, pp. 199-201; Bernal, Black Athena II, p. 386f; Morgan Brown, L., 1988, The miniature wall paintings of Thera: A study in Aegean culture and iconography, Cambridge: Cambridge University Press, 1988.

[21]Cf. O’Connor, D., 1996, ‘Egypt and Greece: The Bronze Age evidence’, in: M.R. Lefkowitz & G. MacLean Rogers, eds., Black Athena revisited, Chapel Hill & London: University of North Caroline Press, pp. 49-60, p. 54.

[22]Helck rev. Drenkhahn, p. 32f. Useful diagrams of navigation routes in the eastern Mediterrenean during the Bronze Age also in: Liverani, M., 1987, ‘The collapse of the Near Eastern regional system at the end of the Bronze Age: The case of Syria’, in: M. Rowlands, M. T. Larsen and K. Kristiansen, eds., Centre and periphery in the Ancient World, Cambridge: Cambridge University Press, pp. 67-73.

[23]Astronomical analysis suggests that the constellations were defined in the third millennium in the eastern Mediterranean; this may have led to a dramatic improvement of navigational techniques; cf. Ovenden, M.W., 1966, ‘The origins of the constellations’, The Philosophical Journal [Transactions of the Royal Philosophical Society of Glasgow] , 3: 1-18.

[24]Best, ‘Ancient toponyms’.

[25]Cf below, conclusion, and: Wim van Binsbergen, ‘Black Athena Ten Years After’.

[26]Dunand, M., 1945, Byblia grammata: Documents et recherches sur le développement de l’écriture en Phénicie, Beyrouth: Imprimérie Catholique; Diringer, D., 1996, The alphabet: A key to the history of mankind, New Delhi: Munshiram Manoharlal, reprint of the 1947 British edition, pp. 158f; Dussaud, R., 1946-8, ‘L’origine de l’alphabet et son évolution première d’après les decouvertes de Byblos’, Syria, 25: 36-52; on Byblos in general, cf. Helck, W., 1975, ‘Byblos’, in W. Helck & E. Otto, Lexikon der Ägyptologie, vol. I, cols. 889-91; Kitchen, K. A., 1967, ‘Byblos, Egypt, and Mari in the early second millennium BC’, Orientalia, 36: 39-54.

[27]Gordon, legend to Plate XII.

[28]Bernal, Black Athena II, p. 524; italics added. Bernal argues that in the second millennium the Levant was under a more or less diffuse Egyptian cultural influence and the whole of the Eastern Mediterranean tended to form an interlocking, cosmopolitan whole (Black Athena II, pp. 52-56). Then — as he admits — no direct suzerainty but merely rather less focused mercantile and military interactions from the Levant (such as Hyksos invasions to the Aegean, which he claims like Eduard Meyer and Frank Stubbings before him) would rather suffice to account for such equally diffuse and indirect Egyptian influence on the Aegean as the archaeological record seems to indicate. Perhaps the linguistic record (as brought out by Bernal’s numerous but as yet dispersed claims of Egyptian etymologies in early and classical Greek) is more impressive, but mercantile and military interactions involving Egyptian speakers among others would equally account for it. Cf. M. Bernal, in press, ‘Response to John Baines’, in: M. Bernal, Black Athena writes back, Durham: Duke University Press:

‘As I see it, the sporadic nature of the Egyptian dominance in the Aegean and the frequent mediation of the Levant in its contact with Egypt mean that it was possible for a hybrid and distinctive Greek culture to emerge. Such a picture allows for substantial Egyptian cultural and linguistic influence without the massive archaeological testimony of Egyptian presence found in zones of sustained colonization.’

[29]Budge, E.A. Wallis, 1969, The Gods of the Egyptians: Or studies in Egyptian mythology, 2 vols., New York: Dover, republication of the first edition, Chicago: Open Court Publishing Company & London: Methuen & Co., 1904, pp. ii 363f. Hart however makes clear that the Cat/ Apophis connection is typical of the Late Period, so that by the beginning of the second millennium BCE, Bastet was probably still associated with a wild feline, a lion; Hart, G., 1993, A dictionary of Egyptian gods and goddesses, London: Routledge, first published 1986.

[30]Evans, Scripta Minoa, I, cover and p. 270.

[31]Bosanquet, R.C., & Dawkins, R.M., 1923, The unpublished objects from the Palaikastro excavations 1902-1906, British School in Athens Suppl. Pap. 1, p. 54.

[32]Zeuner, F.E., 1969, A history of domesticated animals, London: Hutchinson, first published 1963, pp. 390-392.

[33]Woudhuizen, F., 1989, ‘The Cretan branch of Luwian hieroglyphic’, in: Best & Woudhuizen, Lost languages, pp. 65-138, p. 128f.

[34]Cf. Yurco, F.J., 1996, ‘Black Athena: An Egyptological review’, in: Lefkowitz & MacLean Rogers, o.c., pp. 62-100, p. 97; Yener, K.A., & P.B. Vandiver, 1993, ‘Tin processing at Goltepe, an Early Bronze Age site in Anatolia’, American Journal of Archaeology, 97, 2: 207-38; for a critical view however, cf. Muhly, J.D., 1993, ‘Early Bronze Age tin and the Taurus’, American Journal of Archaeology, 97, 2: 239-53; but cf. Yener, K.A., and P.B. Vandiver, 1993, ‘Reply to Muhly’, American Journal of Archaeology, 97, 2: 255-64. Also cf. Best, ‘Ancient toponyms’.

[35]The Story of Sinuhe (Simpson, W.K., 1984, ‘Sinuhe’, in W. Helck & E. Otto, Lexikon der Ägyptologie, Wiesbaden: Harrassowitz, vol. V, cols. 950-956) suggests that by this time Egyptian migrants were not absent from Syrio-Palestine, but had dependent status. Sesostris I or III’s Asian campaign — which features prominently in the Black Athena debate — still had to take place, regardless of the question of just how much or how little Egyptian cultural influence if may have left behind.

[36]Cf. Borghouts, J.F., 1995, ‘Witchcraft, magic, and divination in ancient Egypt’, in: Sasson, J.M., with J. Baines, G. Beckman & K.S. Rubinson, eds., Civilizations of the Ancient Near East, III, New York etc.: Scribner’s, pp. 1775-1785; Budge, E.A.W., 1971, Egyptian Magic, New York: Dover; orig. ed. London: Kegan Paul, Trench & Trübner, Books on Egypt and Chaldaea, II, 1901; Barb, A.A., 1971, ‘Mystery, myth, and magic’, in: Harris, J.R., ed., The legacy of Egypt, 2nd ed., Oxford: Clarendon, pp. 138-169; Ghalioungui, P., 1973, The House of Life: Per Ankh: Magic and Medical Science in Ancient Egypt, Amsterdam: B.M. Israel, 2d ed.; Pinch, G., 1994, Magic in Ancient Egypt, London: British Museum Press; Wilkinson, R.H., 1994, Symbol and magic in Egyptian art, London: Thames & Hudson.

[37]Goody, J., ed., 1968, Literacy in traditional societies, Cambridge: Cambridge University Press, espec. his introduction.

[38]Morris, S.P., 1996, ‘The legacy of Black Athena’, in: Lefkowitz & MacLean Rogers, o.c., p. 167-175, pp. 170.


[40]Liverani, M., 1996, ‘The bathwater and the baby’, in: Lefkowitz & MacLean Rogers, o.c., pp. 421-427; cf. Liverani, ‘The collapse of the Near Eastern regional system’.

[41]Cf. Wim van Binsbergen, ‘Black Athena Ten Years After: Towards a constructive re-assessment’ (this volume).


[43]Best, ‘The ancient toponyms’.

[44]Cf. Wiggermann, F.A.M., & A.R. Green, 1994, ‘Mischwesen’, in: Real­lexikon der Assyriologie und Vorderasiatischen Archäologie, ed. D.O. Edzard with P. Calmeyer, J.N. Postgate, W. Röllig, W. von Soden, M. Stol & G. Wilhelm, Bd 8, pp. 222-246.

[45]Cf. Gardiner, p. 183 n. 1.

[46]See my notes 11-13 above.

[47]Labat, R., 1988, Manuel d’épigraphie akkadienne: Signes, syllabaire, idéogrammes, 6th ed., rev. F. Malbran Labat, Paris: Geunthner (1st ed. 1948), pp. 166f, 176, 179; Borger, R., 1971, Akkadische Zeichenliste, Kevelaer/ Neukirchen-Vluyn: Butzon & Bercker/ Neukirchener Verlag, p. 64; cf. Evans, Scripta Minoa, I, pp. 223f. Evans acknowledges that the Egyptian Hieroglyphic signs (Gardiner numbers N25 and N26) are similar in shape and meaning, but they are rounded, not pointed.

[48]In this connection it is interesting to note that the three animals identified by Jan Best as symbols of a prominent Cretan seal owner, although belonging to totally different sections of the animal kingdom, all are carnivorous hunters and stalkers:

animal presumable onomatopoeic sound number of legs taxonomic section character
cat mi 4 mammal, vertebrate stalker
snake zi 0 reptile, vertebrate stalker
spider -- 8 insect, invertebrate  stalker


[49]Attica’s culture heroes are claimed to have been just that — emphatically enough to raise our suspicion; Frazer, J., ed., 1970, Apollodorus, The Library, 2 vols. Cambridge, Mass.: Harvard University Press, Loeb, first published 1921, II p. 97f, with relevant copious notes. The Black Athena thesis finds an emblematic illustration in the claim that Athena’s name and cultic persona derive from the Egyptian goddess Neith and her temple or city. In this light it is ironic that Erichthonios/ Erechtheus, the Attic apical ancestor, was almost a child of Athena, but not quite: allegedly the latter, having preserved her virginity, in disgust wiped Hephaistos’ sperm off her thigh with a tuft of wool, and cast this to the Earth, who then conceived, and instantly gave birth to what by a spurious etymology was understood in Antiquity as ‘Wool-Earthy’. However, the latter was immediately adopted by Athena. (Cf. M. Bernal, ‘Responses to Black Athena: General and linguistic issues’ (this volume), for a reference to the same myth.)

In myth analysis, I have generally found two methodological points very fruitful: first, the interpretation must rely on close reading, painstakingly scrutinising every minor detail, every word, for implications that might yield clues; and secondly, contradictions and non-sequiturs are our best guides to whatever kernel of historical truth a myth might contain (cf. van Binsbergen, W.M.J., 1985, ‘The historical interpretation of myth in the context of popular Islam’ in: van Binsbergen, W.M.J., & Schoffeleers, J.M., 1985, Theoretical explorations in African religion, London/ Boston: Kegan Paul, pp. 189-224; van Binsbergen, W.M.J., 1987, ‘Likota lya Bankoya: Memory, myth and history’, in: Cahiers d’Etudes Africaines, 27, 3-4: 359-392, numéro spécial sur Modes populaires d’histoire en Afrique, sous la direction de B. Jewsiewicki & C. Moniot; van Binsbergen, W.M.J., 1992, Tears of Rain: Ethnicity and history in central western Zambia, London/ Boston: Kegan Paul International; and references cited there).

In view of Athena’s Egyptian connotations (which were accorded her in Antiquity even if we do not approve) the mythical play between refused and prevented fertilisation and parenthood, yet adoption, and the interaction between three agents (Athena, Hephaistos, Earth) in the production of Erichthonios, opens up possibilities which have not yet been sufficiently exploited in the context of the Black Athena thesis. We might assume that at at least one level of analysis the myth, like countless others, may be read as a geographical chart of regional interactions. It is of course not sure that such an assumption applies and, if it does, how much weight it should be given; e.g. it is equally plausible that transposed to an Attic context the Erichthonios scenario recounts an imported myth no longer understood. Among likely candidates one would then enlist Egyptian myths featuring the principal gods (including Osiris with his siblings, parents and grandparents, depending on the theological regime and the period); these Egyptian myths often highlight irregular forms of sexuality and reproduction which with the proper structuralist instruments may well be argued to produce, as one of their possible surface transformations, the Attic narrative.

However, if we do try to identify the likely geographical/cultural connotations of the protagonists in this story, the result is striking.

Earth is a crucial deity throughout the Ancient Near East and beyond (the literature is voluminous but dispersed; for some of its symbolic and iconographic connotations, see: van Binsbergen, W.M.J., ‘Rethinking Africa’s contribution to global cultural history: Lessons from a comparative historical analysis of mankala board-games and geomantic divination’, this volume). Yet Earth is always eminently local, and here may be read as a reference to ‘pre-Hellenic’ and/or Indo-European cultural contributions.

Whether we choose to stress Hephaistos’ volcanic, Anatolian/Caucasian, Phoenician or Tyrrhenian connotations (cf. Fauth, W., 1979, ‘Hephaistos’, in K. Ziegler and W. Sontheimer, eds., Der kleine Pauly: Lexikon der Antike, München: Deutscher Taschenbuch Verlag, Bd II, cols. 1024-1028) these all point to the mountainous northern coast of the Mediterranean, not to Egypt.

What shimmers through in the Erichthonios myth appears to be, not down-right Egyptian derivation, but (in view of Athena’s catalytic — or pin-up, if you like — role in what is essentially Hephaistos’ siring of Erichthonios by Earth) a suggestion of a vitally stimulating, but indirect and catalytic Egyptian influence on local material primarily fertilised from the Levant — in other words a model surprisingly close to the Model II advocated in the present paper. The possibly Tyrrhenian dimension meanwhile requires further thought.

But of course, myths are there not only to remember, but also to forget; they can never be proof that something (in this case: direct Egyptian influence with retention of specific Egyptian cultural institutions) was not the case.

Incidentally, a similar theme of denied or missed motherhood is found in the myth of Athena’s own birth from Zeus’ head; and if that myth does have a geographical dimension and does hint at Egypt, it converges with the Erichthonios myth.

Meanwhile, for evidence that the House of Neith/ Athena etymology is as problematic as it is emblematic, cf. A. Egberts, ‘Consonants in collision: Neith and Athena reconsidered’ (this volume), and M. Bernal, ‘Responses to Black Athena: General and linguistic issues’, ‘Response to Arno Egberts’ (this volume).

[50]Thus Liverani, ‘The bathwater’, p. 423, does appreciate that

‘The construction of a new multicentered model is a difficult scholarly task. It is the main historiographical challenge of this generation.’

But he fails to see that this is precisely what Bernal tries to do:

‘Bernal’s historiographical method is severely outdated and naïve. And instead of offering a new, multicentric model he merely seems to suggest an Afrocentric and Levantine model, reverting to the old-fashioned Ex Oriente Lux position’ (p. 424).


page last modified: 20-03-02 09:21:28