Scots Gaelic, and Welsh with a confused potpourri of Semitic and Egyptian.
23 Another work pervaded by an ill-informed Celtomania is The White
Goddess ( 1948), subtitled "a historical grammar of poetic myth," by the well-
known English poet Robert Graves. Unlike Bernal, Graves is only marginally
interested in attacking the received view of the Greeks, but in his talent
for generating unsound etymologies in support of bizarre hypotheses he is
nearly Bernal's equal. It is among authors like these that we believe Bernal
has earned his place in the history of scholarship. As far as "the linguistic
evidence" goes, Black Athena is nothing more than a White Goddess with a
different axe to grind.
We would like to thank Anna Morpurgo Davies, Gérard Diffloth, Ives Goddard, Sheila Jasanoff, and James Weinstein for valuable help in the preparation of this
article. All errors, unless specifically credited to others, are our own.
We follow the standard practice of using square brackets to represent phonetic
transcriptions. The asterisk denotes a reconstructed form.
Rendsburg 1989, by a Semitist and avowed "dear friend" of Bernal, is at best a
very partial exception to this statement.
Scholars who believe that the Anatolian languages split off from the rest of the
family before Italic, Germanic, Greek, etc., began to diverge from each other often
prefer the term "Indo-Hittite" to "Indo-European." Bernal follows this usage; we
will retain the traditional "Indo-European" here.
The symbol *h1 represents a PIE consonant -- one of the so-called "laryngeals" -- whose phonetic value is uncertain. There is reason to believe that it may have
Readers interested in verifying this for themselves may wish to compare Emout-Meiller 1985 for Latin and Puhvel 1984- for Hittite. The standard Greek etymological dictionaries are Chantraine 1968-75 and Frisk 1955-72. Bernal seems not to
know the latter work.
athē + ́r, like patē + ́r "father," but unlike most other nouns in -ēr, shows inner-
paradigmatic stem variation ("ablaut"), with short -ĕ- in case forms like the acc. sg.
(athĕ + ́ra). pélōr, like hùdōr "water," belongs to the restricted class of nouns in -ōr that are
neuter rather than masculine. dipsā + ́ is one of the very few contract verbs in which the
3 sg. ends in -ē + ̂i rather than -ā + ̂i, -eî , or -oî -- showing that the -a- of the precontracted
forms was originally long.
C. H. Gordon's decipherments of Eteocretan and Linear A as Semitic ( 1962b, 1966) have found almost no acceptance outside his immediate circle; the same applies
to his efforts to find Semitic inscriptions at various locations in the New World (see,
e.g., C. H. Gordon 1982). Note that place names in -(s)sos (e.g., Halikarnassós, Telmēssós)
are also found in Anatolia.
Where possible, Semitic forms are cited from the closely related Northwest