Black Athena: Afroasiatic Roots of Classical Civilization; Volume III: The Linguistic Evidence

Chapter 20: Geographical Features and Place-Names

[CH. 20] GEOGRAPHICAL FEATURES AND PLACE-NAMES 485 CHAPTER 20 GEOGRAPHICAL FEATURES AND PLACE-NAMES INTRODUCTION P lace-names are, if anything, more durable than words. They can tell one as much or more than language about both prehistory and history. Unlike other words, they often survive the disappearance of the languages that formed them. Their linguistic provenance can provide important evidence on the languages of the populations or rulers in the distant past. In the United States and Canada, for instance, eighteen states and three provinces have names not amenable to analysis in any European language. As they stretch from Quebec and Massachusetts in the east to Utah and Idaho in the west, one could deduce— without historical or current linguistic evidence—that Pre-Columbian populations existed across North America before the arrival of Europeans . On the other hand, four states in the Southwest (California, Nevada , Colorado and Montana) and one in the Southeast (Florida) have Spanish names indicating Spanish settlement, or at least influence, in these regions. Typically, however, political names are less conservative than those of natural features or even than those of towns and villages. In England, apart from London and those to which the Latin-castrum was attached, very few Celtic town names remain. Nevertheless, the most frequent river names—Avon, Derwent, Dart, Don, Ouse and Trent—are mainly British, 486 BLACK ATHENA while some like the Thames (Temis) and Severn are even pre-Celtic. Both facts are striking since Britain became Celtic-speaking more than 2,500 years ago and because British (Welsh) has not been spoken in England , apart from Cornwall, for at least 1,400 years. Absence can also be indicative. We know that England was conquered by Normans in 1066 CE and ruled by French speakers for the next three hundred years. Yet, very few French toponyms can be found; the migrations of Saxons and Danes, whose languages have provided most placenames , were far more substantial. In the Balkan peninsula as a whole nearly all toponyms can be derived from Indo-European. Greece is the great exception—virtually none of its ancient place-names can be explained by Indo-European. For this reason classicists have almost completely given up trying to understand or write about them. No book-length study on the subject has appeared since Adolf Fick’s Vorgriechische Ortsnamen als Quelle für die Vorgeschichte Griechenlands came out in 1905.1 Fick’s book has very little phonetic and absolutely no semantic discipline. His only detectable system was a refusal to consider the most obvious Semitic etymologies, as with, for instance , his explicit denial that the Greek river name Iardavno" could be derived from the Canaanite Yarde\n, Jordan “descending river.”2 For Jasanoff and Nussbaum’s attempt to discredit this obvious loan, see below. Since Fick’s attempt and the ill-fated venture of Blegen and Haley described in Volume 1, classicists have left Greek place-names strictly alone.3 Jasanoff and Nussbaum justify this failure when they write, “names, in principle, can mean almost anything.”4 I cannot accept this approach because I believe that, while names are often repeated simply as names, the originals nearly always had a meaning, particularly in the case of place-names. Frequently, we fail to understand the meaning of a placename simply because we do not know, or are not aware that we know, the language from which it was constructed. Nevertheless, before supposing that a specific name derives from an unknown language we should check to see if it can be explained by one that is known and understood. I am convinced that the reason why Greek toponymy has not progressed is that nineteenth- and twentieth-century scholars have seen these place-names as remnants of unknown, lost pre-Hellenic languages. Seldom , if ever, have they considered the possibility that many toponyms can plausibly be explained in terms of Ancient Egyptian or Semitic. My approach here is not “undisciplined,” as my critics complain. In looking for the etymologies of Greek place-names, I insist that (1) there [CH. 20] GEOGRAPHICAL FEATURES AND PLACE-NAMES 487 must be a good phonetic fit and either (2) the proposed Afroasiatic etymon should be attested as a place-name or (3) the proposed etymon should fit the physical features of the place named. Previous scholars have not considered this last criterion. NATURAL FEATURES Islands As with other toponyms, island names sometimes slip...