Wim
van Binsbergen, ‘Out of Sundaland?’, paper, First Annual Conference,
International Association for Comparative Mythology, Edinburgh, August 2007
62
Challenged by the preposterous,
but apparently successful work by Joseph Greenberg on African languages and, more recently Greenberg, and Ruhlen, on
native American languages and on languages worldwide, all based on mass comparison, specialist
historical linguists have argued that the only form of comparison agreed to yield valid and authoritative
results is that comparing, not actual forms, but reconstructed proto-forms. This ongoing debate has very
important implications for long-range comparative mythology, for there we tend to remain on the side of
mass comparison, without even making that position explicit. Whether we adopt a time-honoured and
prestigious (but Eurocentric and theoretically barren) classification like that of Aarne and Thompson (recently
renewed by Hans-Jorg Uther), or a finely-meshed classification of our own making (like Berezhin does,
distinguishing c. 1,500 carefully defined micro-mythemes), or a broad classification into just a few
dozen Narrative Complexes each encompassing up to twenty or thirty constitutent mythemes (like in my own
Ággregative (!) Diachronic approach) – we are always relying on the equivalent of actual forms, on
recently attested mythemes, and have not developed a method yet to validly and reliably reconstruct the protoforms that may be
presumed to underlie these actual
forms. In other words, we are projecting back into the remotest past actual
mythemes as if these were
self-evidently equivalent to ancient mythological forms. Of course, proposals
are made for the, more or
less intuitive and certainly qualitative and unmethodological, reconstruction
of very broad general ancient
forms, such as Witzel’s distinction between Laurasian and Gondwana
mythologies, and my plea (see
below) for the distinction between two broad types of cosmogonies, one based
on the separation of Water and
Land, the other on the separation of Heaven and Earth – but this is more like
reconstructing entire
lexicons than individual lexical proto-forms. Ultimately we will have to face
the problem, either by putting
up an explicit argument as to why we believe our actual mythological forms are
fair approximations of
mythological proto-forms; or by devising a method to specially reconstruct the
latter.
Back to
linguistics. It is easy enough to propose an etymology, but very few
etymologies do live up to the stringent
requirements professional historical linguists would impose (a. phonological
correspondence; b. semantic
correspondence; c. explicit transformation rules, specific to the language
family/ies involved, that dictate
the phonological and morphological link between the word in question, and the
proposed etymon (Blažek, personal
communication, 2005).
Yet, still in the
linguistic field, it is possible to propose quite convincing Austric (i.e.
Sunda) etymologies for, for instance,
the key divine names in Ancient Egyptian mythology: Osiris and Rec (see van Binsbergen, in prep.). And also for Nuaḥ
(see below; the standard Old Testament Studies etymology in terms of ‘rest’
has little to recommend it) – although the statistical and
etymological findings which I shall present below are not
encouraging for the implication of Oppenheimer’s theory to the effect that
Nuaḥite flood myths have a Sunda connection; and
even for the South East Chinese flood heroes Nű Wa 女娲 (cf.
Nuaḥ?) and Fu Xi
伏羲 (cf. Po-sei[-dōn]?) (van
Binsbergen, in preparation).
Osiris
Rec
