LINGUIST List 15.1878

Mon Jun 21 2004

Review: Historical Linguistics: Vennemann (2003)

Editor for this issue: Terence Langendoen <>

What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in.

If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Sheila Dooley Collberg at


  • Hayim Sheynin, Europa Vasconica - Europa Semitica

    Message 1: Europa Vasconica - Europa Semitica

    Date: Wed, 9 Jun 2004 10:24:22 -0400
    From: Hayim Sheynin <>
    Subject: Europa Vasconica - Europa Semitica

    AUTHOR: Vennemann, Theo, gen. Nierfeld EDITOR: Patrizia Noel Aziz Hanna TITLE: Europa Vasconica - Europa Semitica SERIES: Trends in Linguistics: Studies and Monographs 138 PUBLISHER: Mouton de Gruyter YEAR: 2003 ANNOUNCED IN:

    Hayim Y. Sheynin, Gratz College, Melrose Park, PA.

    DEDICATION To the memory of great linguist Robert Larry Trask.

    SYMBOLS AND ABBREVIATIONS USED CONSONANTS ? alif, glottal plosive laryngeal stop � ayin, voiced fricative pharyngeal continuant sonorant guttural c_ non-emphatic velar plosive stop ^d dhal, voiced interdental fricative coronal continuant d! dad, emphatic voiced dental plosive h! ha/heth, voiceless fricative pharyngeal continuant sonorant guttural q! kof, emphatic uvular plosive stop ^s Hebrew shin, voiceless high alveopalatal fricative sibilant s! sad, emphatic sibilant coronal fricative ^t tha/thaw, voiceless interdental fricative coronal continuant t! ta/teth, emphatic voiceless dental plosive coronal stop coronal x! haf/ha, voiceless high velar fricative continuant

    VOWELS � shwa (reduced low central vowel) #a ultra-short a in Hebrew and Aramaic, short in Indo-European #e ultra-short e in Hebrew and Aramaic, short in Indo-European #i short i in Indo-European #u short u in Indo-European

    NOTATIONS * reconstructed word or phoneme + Proto-European or Proto-Semitic reconstructed and attested word /.../ encloses examples from Basque and Afro-Asiatic languages [ ] encloses my notes in the body of the review

    ABBREVIATIONS AA Afro-Asiatic Akk. Akkadian Ar. Arabic, Arabian Arc. Aramaic Ass. Assyrian B.-Arc. Biblical Aramaic Bq Basque B.-Sl. Balto-Slavic Cel. Celtic Eg. Egyptian Eth. Ethiopic Gmc. Germanic Gr. German Gk. Greek Gur. Gurage Ha. Hausa Hb. Hebrew HG High German Hmc. Hamitic HS Hamito-Semitic Iber. Iberian IE Indo-European L. Latin M-Ar. Modern Arabic NG North Germanic Nrw. Norwegian O.-Akk. Old Akkadian O.-Ass. Old Assyrian OE Old English OHG Old High German Pct. Pictish PG Proto-Germanic PIE Proto-Indo-European Ph. Phoenician PS Proto-Semitic Rs. Russian S.-Ar. Epigraphic South Arabic Smc. Semitic Sp. Spanish Src. Syriac Sw. Swedish Tg. Tigre Tgr. Tigrinya Ukr. Ukrainian Ug. Ugaritic WG West Germanic WIE West Indo-European W.-Smc. West Semitic Yid. Yiddish

    INTRODUCTION The current reviewer writes this review with mixed feelings. He is expected to evaluate the Lebenswerk of a scholar who holds an important place in European Comparative and Historical linguistics, who studied in good universities of Germany and the United States (and who studied with exceptional teachers), who has chaired the University of Munich Department of Germanic and Theoretical Linguistics since 1974, and who was visiting Professor of Linguistics in a number of universities in the United States and Austria. His work for the last three decades is known in many areas of general and Germanic linguistics as well as historical phonology, historical morphology, word order studies, typology, syntax and semantics, phonology and morphophonology. Since his Ph.D. dissertation "German Phonology", University of California, Los Angeles, he published a score of articles and conference papers, several books, and now a big book which includes articles united by a particular idea, but devoted to research of many diverse genetic language families.

    For about two decades Professor Theo Vennemann gennant Nierfeld (henceforth V.) has worked diligently to uncover and explain Germanic and Indo-European words which didn't have clear etymology. He started his research from revising existing theories about the origins the names of European rivers and other toponyms. Already in early eighties he came to certain conclusions and proceeded to work in the area of etymology, steadfastly propagating his theories and many times repeating and restating his conclusions. Results of this endeavor were numerous articles scattered in various academic journals and proceedings of scholarly conferences. Now he collected his unchanged articles into a big and well edited book and published it in a very prestigious series "Trends in Linguistics".

    Unfortunately after finishing the book the reader remains with more questions than he had before starting the book.

    There is no way in a reasonably short review to deal with each separate article or chapter, nor with many aspects of the book under review. Therefore I will try to separate several areas covered in the book, the areas I consider to be indispensable, to describe the problems associated with each of these areas, and to give a critical evaluation.

    DESCRIPTION OF THE BOOK'S PURPOSE AND CONTENTS The book contains 26 papers V. published between 1984 and 2000 and one article previously unpublished (chapter 17), i.e. 27 chapters altogether. Most of the articles were written in German (18), the rest in English. In the present book the English abstracts were added preceding the German articles. Among the indexes, it calls to attention the lack of the general index or an index of private names. The articles are reprinted without changes, even when negative criticisms were known to the author -- it is clear that the negative reactions to his work were known to the author, since he himself acknowledges this, see p. x-xi and 577-78, and also refers in notes to some of these critical reactions).

    Since the book concentrates on substrata and superstrata in the most ancient period of the Indo-European or Proto-Indo-European the present reviewer finds it necessary to declare that he accepts the general principle of existing of substrata and superstrata and their role in development and growth of the languages in condition of language contacts. Our presumption is only that the existing of such contacts should be proven either by external or internal evidence beyond reasonable doubts.

    The general ideas which unite the articles in the book are as follows.

    1. About one third of the Proto-Germanic vocabulary has no Indo- European etymology. V. proposes that these unexplained words may be owed to prehistoric substrata. According to him [I copy his account almost verbatim, mostly from the Introduction, although inserting passages from other parts of his work where necessary, and beg the readers for pardon for such a long exposition of his ideas, which is necessary, however, for understanding the order of the author's thought and for his logic], Indo-European agriculturalists, possibly originating from the area of the Pannonian Basin (in chap. 2, from the Carpathian Basin), migrated further into Europe in the sixth millennium BC and did not arrive in Scandinavia before the fourth millennium BC "After the last ice-age, the Vascons spread from southern France to almost all of Europe north of Alps, where they were subsequently submerged by Indo-Europeans immigrating from the south-east of the continent." (In chapter 26 "Grundfragen der Ortsnamenforschung ...") Since the climate in Europe (after the last glaciations) had begun to improve much earlier, Indo-European settlers encountered other, non- Indo-European, peoples in Europe who had started to settle there from the eighth millennium BC onward. The first settlers were forced to name European rivers, lakes, mountains, settlements, etc. for reasons of orientation, which is a fortunate coincidence for linguistics: The oldest water names are probably the oldest "linguistic documents" in Europe north of the Alps. V. utilizes Hans Krahe's collections of hydronyms and reinterprets Krahe's etymologies. Whereas Krahe assumed Indo-European provenance of Old European hydronomy, V. proposes that new analyses of the data point to an agglutinating source with initial accent, without vowel quantity, and with the predominant vowel a. The author states that such a structure does not support an Indo-European origin, while the structural and substantive similarities between the language of the Old European hydronomy and Basque (!) are compelling. V. finds similar geographical names throughout Europe (England, Germany, Norway and Sweden) and attributes them to Basque etyma, the Basque language being the only surviving descendant of the Vasconic language family. [Later on V. finds that the ancient language of Raetia (Alpine area of today Switzerland), see pp. 737-751, was related to Basque and therefore he claims that there was another ancient Vasconic language which gave some terms for husbandry and dairy products to some Italic (specifically Latin) and to some Germanic (specifically German) languages.]

    Traces of substratal Vasconic influence in the West Indo-European languages (particularly the shift to initial word-accent in early Italic, Celtic, and Germanic) are more or less systematic. West Indo- European remains of vigesimal counting, and words permitting Vasconic etymologies. Of the latter, reflexes of a Vasconic word for '(young) woman, lady', preserved in Basque andere, are cited in Celtic, Greek, the Romance languages, and German. Basque handi 'big' and Latin grandi- 'big' are both derived from a Vasconic word +grandi- 'big', the Latin word being a prehistoric substratal borrowing.

    To find appropriate Basque words Vennemann uses a number of dictionaries, most frequently Agud-Tovar (1989), Azkue (1984), Arbelaiz (n.d., ca. 1978) and L�pelmann (1968).

    2. Apart from Vasconic place names there are several toponyms on the Atlantic littoral which are, according to V.'s opinion, neither Vasconic nor Indo-European. V. calls the language of these toponyms Atlantic and assumes that it is the language of the seafarers who influenced the Indo-European languages of the Atlantic littoral of Northwestern Europe from c. 5000 BC onward. (NB. In chap. 10, p. 363 V. states that Atlantic people reached Southern Sweden no later than the third millennium AD [chap. 10, p. 363, n. 9 ; chap. 11, p. 384]). On the basis of etymological reconstruction, he identifies the toponymic roots as Hamito-Semitic or rather Semitidic, since they generally show a closer affinity to Semitic than to Hamitic languages. He says that an early Hamito-Semitic substratum on the British Isles was discovered in the 19th century by Morris Jones (1900) and has been corroborated by Pokorny (1927-30), Gensler (1993), by himself (Vennemann 2000b, 2001c, 2002c and 2002d -- articles that are not included in the current book), and others. Semitic influences may have lasted well into the Phoenician period. Likely examples of toponyms without Indo-European, but with Atlantic, namely Semitic, etymologies are: The Solent (Coates 1988a), Solund, Uist (an island), the Isles of Scilly (sf. Smc./s-l-�/ 'rock, cliff'), Tay, Taw (cf. Ha. /tagus/ 'river') and the Pit-names of Pictland (cf. Smc. */pitt-/ 'area, region', Akk. pittu 'administrative district'(cf. chapters 15 and 16). Smc. loan-words in WIE, esp. in Gmc., are interpreted as traces of Smc. influence, superstratal in the case of Gmc., exerted along the Atlantic littoral rather than in the East. The WIE apple word and the Romance and Gmc. baron word are given special attention.

    Referring to Coates (1988b) Vennemann identifies Uist as Ibiza with the Semitic etymon /ai-b-^s-m/ or /?i-b�sim/, /?�-b�sem/ or /?�-besim/ (p. 222).

    V. expresses highly interesting although unexpected and speculative opinion that the tribes of Picts and Vans were belonging to this Atlantic (=Semitidic) society (chap. 11). However the presentation of this statement lacks any linguistic evidence. V. is aware of this lack, since he cites a passage from Newstead 1979: "Also the Picts disappeared as a separate people after a crushing defeat in 843 and nothing remains of their language except a few dubious inscriptions, a precious fragment of the legend that developed around the royal name of Drust is preserved in the tenth-century recension of the Irish saga, The Wooing of Emer."

    All the evidence of Vans' influence on Gmc. tribes is based on exposition of Vanir's cycle of Gmc. myths and the same source is used to identify their Smc. origin. We do not deny that myths might preserve some historic information in somehow unclear or distorted form, but we do not see how V.'s interpretation can be proven it has any relation to the problem of Smc. Provenance of Picts.

    Another two findings here: 1) the Germanic term for rulers, Old English oe^del- (P.-Gmc. +a�al, Gr. Adel). V. claims that M�ller (1911) noticed the striking resemblance between Proto-Germanic +/a�al/ and the Arabic root /?-^t-l/ and interpreted the words as the same etymon of a shared Nostratic proto-language (!!!) [We should note that neither H. M�ller nor S. Levin, both of whom had a number of predecessors, interpreted Semitic words they found resembling Indo-European as Nostratic; they just presupposed that the consonant roots composed of similar consonants and having similar general meaning must originate from the common source. Nostratic theory presupposes more elaborated phonetic laws and more rigid hierarchy of phonetic transformations.] V. himself find such an explanation problematic, but he uses the word and his Smc. cognates as Atlantic influence on Gmc.

    2) P.-Gmc +sibjo: 'family' (English sib, German Sippe, etc.) may be compared to the Semitic root /^s-p-h!/ the meaning of the root according to V., is 'family'. About this interpretation see below. V. going so far, as to explain the extraordinary systematization and functionalization of the inherited IE verbal ablaut in Gmc as Smc influence.

    V. lists a large number of Smc. and AA words from many languages as sources for IE words. Some of them were identified by his predecessors, others he found searching in a number of dictionaries, most often M�ller (1911), Pokorny (1927-30), [Here we need only to mention that the number of both the Bq and the Smc etymologies of IE words in the grand Pokorny's dictionary is negligible (see Walde-Partridge's indexes to Pokorny (1927-30), 3rd ed., p. 481 (10 Semitic and 2 Basque words). Even if we take for account that Iberian etymologies in Pokorny, according to V., could be Basque, it would add three more words to Basque etymologies. In fact all these Basque and Iberian material belongs to two Indo-European roots andh-/ anedh- and #atos /atta)] and Soden (1965). Since 1995 he most heavily relied on Orel (1995). The last mentioned dictionary was a first attempt of the reconstruction of the AA lexicon based on all existing linguistic information. Unfortunately, as it reflected in many reviews, including those written by the outstanding Afrasian linguists, I. M. Diakonoff and L. E. Kogan, this attempt was not particularly successful. The dictionary contains a lot of errors and mistaken reconstructions. [See e.g. I. M. Diakonoff (1996):25-44; and Kogan (2002):183-202.]

    The number of AA etymologies in V. is very significant, most of them attested in Smc. languages [contrary to position of Orel and Stolbova in Orel (1995) who predominantly rely on African component]. Prior to publication of V.'s book there were in existence several linguistic publications on reconstruction of P.-Smc. roots which could have been helpful to the author but are not mentioned there. [See e.g. Djakonov (1981-86) [in Russian], rev. version in Eng. Djakonov (1994-97); Shevoroshkin (1989), especially contributions by A.B. Dolgopol'sky; Renfrew (1999); Dolgopol'sky (1999); Militarev (2000).]

    One would wonder why the IE words of Smc. provenance were not discovered by the great linguists in the 19th and 20th centuries. V.'s answer to this question is that most of these words are coming from AA languages which were not learnt in the schools. However this explanation cannot be true. Most of examples quoted in V.'s magnum opus are from Akk., Hb., and Ar., all three languages widely represented in the curricula of European and American universities and studied in different academic settings in Oriental, Linguistic or Theological departments.

    We can observe that the long list of Smc. etymologies in V. is united only by an incidental similarities and in no way better than popular etymology. In many cases V.'s Smc. etyma do not share meaning with his Gmc. or IE word or explained by a forced metaphoric extension of meaning.

    To support his theory V. refers to extra-linguistic evidence.

    Thus in support of the idea of prehistoric people's expansion [i.e., according to V., speakers of ancient Vasconic languages] from the south of France to Central Europe he mentions facts of genetics (cf. Cavalli- Sforza 1994) Maps of rhesus negative are almost perfectly convergent with maps of the Old European hydronymy [as based on Vasconic theory]. Furthermore, blood group "0" contributes to the overall picture. It has its densest distribution in the Bq. Country (Mourant 1954). [Cf. Renfrew (2000), esp. chs. 1-3, 11-14, 21, 40-41.]

    The IE blood group "A" is dominant north of the Danube, while south of the Danube, and even more so close to the Alps, blood group "0" is dominant... Today, "0" as the dominant blood group of the Basques is found particularly in those areas of Europe where the people mixed the least for geographical reasons: e.g. mountain regions were not attractive for the IE agriculturalists, large rivers were difficult to cross.

    Here we should mention that prehistoric archaeology did not supply enough data to determine who were first settlers or autochthons in Europe; there are still discussions on Mediterranean and Asia Minor influences, and even directions of migrations are in dispute. So V.'s ideas pertaining to archaeology cannot be verified.

    Exactly as V. tries to support his Vasconic theory by reference to extra-linguistic evidence, he does the same for his Atlantic (=Semitidic) theory. In this respect, he turns to the megalithic monuments of Western Europe and suggests that they may be vestiges of an Atlantic culture. Then he turns to Germanic mythology and finds in the myth of the Vanir the matrilineal society. He notes that such pattern of societal organization is well known to have existed and still to exist among Hamito-Semitic peoples.

    In many places he states that Atlantic (Semito-Hamitic) people were seafarers, experts in architecture, astronomy and navigation, and warriors. However a big group of examples indicate that they were beekeepers and animal breeders in addition to being town dwellers and builders of fortified settlements. Later, when V. needs to explain that the word for apple was borrowed by Germanic and Balto-Slavic languages from Atlantic, he writes (p. 622):

    "How could the apple-word fall so far away from its original language tree right into our Northwest Indo-European garden of words? ... According to my theory of the linguistic prehistory of the Atlantic regions of Europe, the apple did not have to fall far from its tree. The people that brought apple raising and the apple-word lived in close proximity to the northwestern-most Indo-Europeans. In the case of the pre-Germanic Indo-Europeans, they even lived in the same territory and became part of the Germanic people..."

    So the general description of these Atlantic people that emerges from V.'s work, that they were superior to Indo-Europeans in seafaring, warfare and economy and contributed to the Indo-European civilization in general and to Germanic society in particular and even played role in German ethnogeny.

    CRITICAL EVALUATION The two main points outlined above picture great linguistic ideas, which if proven to be true would provide new understanding of development of Indo-European languages, new understanding of pre- historic Europe, would contribute to the theories of migrations and would be helpful in many other aspects of humanities.

    However there are many difficulties with these theories.

    The criticism of the problem of Vasconic substratum in V.'s theory was offered in Trask (1995) who states that V.'s presentation of Michilena's reconstruction [of P.-Bq.] is seriously inaccurate. Formation of the Bq. article -a most probably took place in post-Roman times. None of the roots or suffixes listed by V. for Old European looks like anything in Bq., save for the root *Is- (according to V., identical with Bq. root *'iz- 'water', and Michilena in several places dismisses this putative root as a phantasm. A sizeable number of the roots identified by V. have forms which would have been impossible in P.-Bq., e.g. the roots with initial m-, p-, r-. Trask likewise mentions that the Indo-European nature of the 'Old European' river names is defended by Schmid (1987), who notes that many of the morphs that appear to be present in these river names can readily be identified with well known IE roots and affixes. In continuation Trask dismisses V.'s etymology of German toponym M�nchen ('Munich') (ibid., p. 72-73)

    Later Trask repeats his reaction to all the theories on connection of Bq. with other languages [Trask (1997), Chap. 6, pp. 358-429, esp. on V.'s theory, pp. 365-367.] Here among other things Trask objects V.'s assumption that an ancestral form of Bq. possessed a series of 'weak' consonants which, following the IE tradition, he terms laryngeals. (ibid., 180)

    In Trask's view there is no base to connect Bq. with so called 'Old European' substratum. He shows problems associated with all V.'s examples one after another. In general Trask dismissed V.'s theories as non-serious and lacking both the Vasconist and historical linguistic background.

    However it seems that V. sticks with his position because after all the criticisms he publishes the article where he restates his theory even in greater detail (chapter 17, written ca. 2000), rejecting critical notes as if running counter to general principles of language contact.

    Even greater scrutiny of V.'s ideas on 'Vasconic' or Old European substratum of the Indo-European attempted J.A.Lakarra [Lakarra (1996) (appeared not before 1999, since works published in 1999 cited there)] who offers a detailed analysis and the devastating criticism of V.'s theory of Vasconic substratum. We will cite only a short passage from the English abstract of the mentioned article: "Despite the flattering prospect of us the Basques as the only remnants of Old Europe, the reading of his article [Vennemann 1994 and later works-chap. 4-9 in the book] does not lead us to conclude that it provides the considerable advance that we could have expected for our field of study. In this paper ... we show that (i) V.'s linguistic reconstruction does not correspond, in very important respects, to what we reasonably may know about Proto-Basque structure, and (ii) consequently, on the basis of such a reconstruction one can hardly genetically relate Old European with Basque." This is the most general conclusion. Further Lakarra is going into detailed analysis of the shortcomings in V.'s work. She continues, "V. uses to his own benefit the attested phonological and lexical similarities between Iberian and Basque, but does not mention other well-known differences such as the existence of aspiration and lenis-fortis oppositions in sonorant in Basque. V. also takes for granted the Basque-Iberian hypothesis which goes against both the specialists' opinion and the evidence that it has no effect on the deciphering of the Iberian. V. clearly manipulates the distribution of phonemes in modern Basque so that a and vowel initial syllables seem to be overabundant (closer to the phonemic distribution he holds them to have in Old European), and taking this alleged abundance for granted he claims that e and o are secondary in Basque. V. does not mention the absence of |p|, |m| and |r-| in Proto-Basque, and claims that |h| is in free variation in the modern dialects which still have it."

    We will not continue citing the charges against V.'s linguistic accuracy [they are too many], but together with Lakarra express opinion that V.'s reconstruction of Old European is a complete fallacy.

    V. in the process of preparation of the collection of his articles for the book under review reacted to some criticisms of R.L. Trask in hope that the former would withdraw some of his charges (e.g. p. 198 n. 82; p. 199, nn. 85, 87; p. 596). This never happened.

    Even before J. A.Lakarra's publication evaluating V.'s linguistic theories, Peter R. Kitson demolished V.'s analysis of European hydronomy [Kitson (1996)] writing a paper not specifically dealing with V.'s theories, but expanding on E. Ekwall's English River-Names. Oxford University Press, 1928 in the light of H. Krahe and his followers. He reveals that Krahe's attribution of the river names to the Indo- European sources is upheld against V.'s arguments. Kitson devotes some place to describe some features of so called alteurop�isch which are the parts of a particular linguistic system. "The linguistic material of these names is Indo-European. Not only are most of the roots readily etymologized on this basis, but the suffixes also are those that were productive in Indo-European." It was clear already to Krahe that the suffixes -st- forming superlatives and -nt- forming participles are Indo-European productive suffixes. Kitson shows a particular order of phonetic elements prevalent in the alteurop�isch. "Krahe's list contains at least 20 names Almana, Elmina, Armenos, and the like, with the same order as in Greek participles in -menos, yet there are not any *Elnama, *Arnomos, etc., with the reverse order."

    P. 82: "A recent attempt by V. (Vennemann (1994):232) [=Chap. 6 in the book] to show that suffixal combinations were not ordered ignores these. The only suffixes he proffers that 'reverse their relative order' are -st- and -r-. If they did it would resemble variation in double gradations of adjectives in some Indo-European languages, reflecting partly changes in linguistic fashion over substantial periods of time. V.'s argument depends on a tacit collapse of diachrony. Moreover nearly every one of his examples is suspect as one or more of: falsely segmented, not 'Old European', or not even a river- name. And with so many corners cut, V. still is not able to show any single root with which two suffixes do occur in both orders. Altogether his argument is fairly trivially invalid, and the second-order deduction [from a maximum of two suffixes!] making the language of the hydronymy an agglutinating one is a fortiori.

    On pp. 86-87 et passim Kitson mentions correspondences between ancient IE (Hittite, Tokharian, Old Iranian and Old Indian) words for river, stream, channel and the present rivers' names in entire Europe from Atlantic coast to Russia. All this is running against V.'s theories on Vascon provenance of European hydronomy.

    P. 95: "The Indo-Europeanness of alteurop�isch names was obvious to Krahe and his colleages from the beginning. Occasional attempts to prove otherwise depend on ignoring a lot of evidence presented above and falsifying some of it. A recent such exercise, that of V. [Vennemann, ibid.], parades a technical linguistic (specifically morphological) virtuosity that may mislead the unwary but lacks proper control in several directions." On pp. 96-97 Kitson criticizes V.'s failure to take in account semantic links when positing etymological links and his way of systematic segmentation of rivers' names deliberately manipulating them in order to demonstrate non-Indo- European character of the alteurop�isch. On p. 97 Kitson describes V.'s failure as historical phonologist (cf. p. 111, n. 74). With all the criticism that completely dismisses V.'s theory of the Vascon character of the 'Old European', Kitson finds kind words for V., saying: "Still Vennemann deserves thanks for supplying what had been a gap in the literature and showing us what a seriously worked up attempt to analyse the alteurop�isch linguistic material as non-Indo-European would like." We would like to reject even those kind words, because we understand that from intellectual aspect V.'s work was a non-serious manipulative attempt which was rejected by serious scholars and experts working in all the pertinent areas of linguistics.

    At the end of his paper, Kitson analyses the existing opinions on the Urheimat of the Indo-Europeans or at least the place from which they migrated to Europe, and particularly to Britain. He accepts the original starting point of migration place in North Central Europe (perhaps Eastern Germany or Poland) and states after McEvedy 1967, that the earliest Brits - British beaker folk - came from the Rhine-Elbe region. All this is also a point against V.'s theories. We should mention also that Kitson's analysis moves the chronological frame of migration of the Indo-Europeans several millennia later than V. assumes. So the migration to Scandinavia should be postponed at least for two millennia, i.e. not earlier than second millennium BC Depicting V.'s criticism by Kitson, exactly like the case with Trask's and Lakarra's, we paid attention only to the main points, in no way exhausting all the charges. We think taking in account the criticism by three important experts, we are relieved for mounting new charges against the idea of 'Europa Vasconica'. This leaves us with 'Old European' being completely Indo-European or at least without a shade of Vasconic substratum.

    Now let us to turn to Europa Semitica.

    V. in several chapters matches a big number certain PIE, IE, as well as L. and Gr. words with the etyma coming from different languages of AA language bundle. Many of these languages were known to the linguists for centuries, but there was almost no research connecting AA languages with the Indo-European ones with an exception of M�ller (1911), Nostratic linguists (mainly since early 60s, while some initial works were written since H. Pedersen's 1903 paper), S. Levin (since 1971) and some pseudo-scholarly authors who tried to show that Biblical Hebrew was mother of all existing languages. [See e.g. books of Jacobowitz, J. Monogenesis of language: selected essays based on the first four letters of the Hebrew alphabet. Jerusalem: R. Mass, 1968 (in Hebrew); and Mozeson, Isaac E. Word: The dictionary that reveals the Hebrew source of English. New York: Shapolsky Publ., 1989; 2nd. ed., Northvale, NJ: Jason Aronson, 1995; new ed., New York: SPI Books, 2000. One can see similar statements in a variety of pietistic and fundamentalist Jewish and Christian sources, exactly like fundamentalist Muslims consider this language was Arabic. We can say that the biblical story of Babel tower gave ideas to some linguists and some pseudo-linguists, exactly as 18th century division of languages and ethnic groups as descendents of three sons of biblical Noah (Semitic, Hamitic, Japhetidic)]. As far as one can understand V. matches mostly Gr., Eng., and L. words with the AA words, which he finds in M�ller (1911), Levin (1995) and V. Orel and O. Stolbova [further Orel (1995)]. It was shown recently by I. M. Diakonov and L. E. Kogan, that the work Orel (1995) is insufficient and misleading in many entries. Since V. accepted the results of this publication uncritically, he inherited mistakes of this dictionary in his AA etymologies of the IE vocabulary.

    Checking V.'s list of Smc. words which is presented as an evidence of "Atlantic" superstratum shows mere ad hoc sound similarities. As we already mentioned, plethora of examples in V.'s book are not original. One can find most of them in Levin (1995), some of them were known even earlier. V. does not acknowledge this (he acknowledges only those borrowed from M�ller (1911), Coates (1988), Soden (1965) and Orel (1995), seldom from Brunner (1969). We endeavored to check each one of the words presented by V. as evidence of 'Semitidic' superstratum / adstratum and we cannot accept them on one or another reason. The complete list of these words when we rewrote them from different parts of the book took about seven pages, but we will not bother the reader with the entire list. We illustrate this review only with one dozen randomly selected examples.

    1. to wake (V.'s p. 358) w-q-y Ar., Eth. 'bewahren' (to keep, preserve) Akk. /(w)aq!�(m)/, O.-Ass./waq!a:'um/ 'warten, harren, bewahren, achten auf, aufmerken' (to wait, wait for, keep/preserve, attend to, pay attention); original meaning of the Ar. root/ w-q!-y / is 'to keep, to preserve, to keep watch, caution'. We should point that it is wrong to compare the Eng. word with initial /w-/ to the Akk. form where the initial /w-/ is presupposed on the basis of Ar. /w-/.

    Initial PS *w- had zero reflex in Akk., see table of correspondences of the consonant phonemes in Dolgopolsky (1999), pp. 16-18; the same applies to example #4. Even there is some sound resemblance, one would wonder why such word should be borrowed from Smc. (because this semantic field doesn't reflect tangible realia).

    2. Eng. sib, Germ Sippe (PG +sibjo: 'family') (V.'s p. XVII, 936) Smc. root /^s-p-h!/ (the meaning of the root according to V., is 'family'). In fact, Ar. s-p-h! / Hb. ^s-p-h! means 'to shed, to spill', and etymology of the 'family' in Smc. probably comes from either ejaculation of semen, or by popular understanding of conception as spilling blood of man into the vagina of his female sexual partner [see Gesenius 1921:856 who translates Ar. safah!a 'ausgie�en' and renders Ar. example /sa:fah!aha:/ 'effudit cum ea sc. semen']. Nowhere in Smc. the form similar to Gmc., i.e. without nominal preformative, and with meaning 'family' is attested. Moreover the Hb. word /mi^spa:h!a:/ was known to European scholars from B.-Hb. and in the form /mi^sp�x!a/mi^sp�x!e/ (with h! > x!) from a Gmc. Jewish language Yid. for centuries.

    I should note, however, that V. only mentions this example in the book without an argumentation. Most probably he published the article in which he discusses the example in time it was late to include it in the book [Vennemann, Theo. "Germania Semitica : *sibjo:," In Heizmann, Wilhelm and Van Nahl, Astrid, eds. Runica - Germanica - Mediaevalia. Berlin: Walter de Gruyter, 2003 (Erg�nzungsb�nde zum Reallexicon der Germanischen Altertumskunde, 37) : pp. 871-891. I didn't see the article.]

    3. V. claims after M�ller (1911) the striking resemblance between PG +/a�al/ and the Ar. root /?-^t-l/ as the Gmc. term for rulers (OE oe^del- ; PG +/a�al/, Gr. Adel). This root was also known in a corresponding Hb. form /?as!�l/ 'noble'. Analyzing semantic background of Ar. /?-^t-l/ we should mention that the original meaning of it was 'to take root, to become rooted'. The meaning of 'noble' and further 'wealthy' is probably developed with a later semantic extension that happened during social development of pre-Islamic Arab tribes. In Hb. it remains 'noble'. Nowhere it corresponds to a semantic field 'rulers'. For the period V. is concerned with this Semitic word couldn't mean 'ruler'.

    4. to ward (V.'s p. 360) Smc. /w-r-d / y-r-d/ Akk. /(w)ara:du(m)/ 'to descend', /(w)ardu(m)/ 'slave, servant (also of kings, in palaces, of gods, in temples)', 'a (special kind of) craftsman, (perhaps) master-builder', /(w)ardutu(m)/ 'slavery, servitude; service (also of vassal, or in politics)'; /(w)ardatu(m)/ 'girl, young woman (also said of goddesses and female demons)'. This historical reconstruction is wrong on account of w-, see example #1. We need to stress that the general meaning of the root w-r-d / y-r-d in all Smc. languages is 'to descend' and therefore has nothing to do with 'to ward'.

    5. Ru� (V.s p. 256) Common Smc. /q!-t!-r/ or /q!-t-r/ 'Rauch', Akk. /q!utru/ 'Rauch / smoke', Hb. /q!�t!�ret/ 'R�ucherwerk', Arc. /q!itra:/; S.-Ar.-Eth. /q!eta:r�:/ 'R�ucherwerk' N.-Ar. /q!ut!a:r(un)/ 'smell of cooked meat, of aloeswood'; /qut!r/ and /qut!ur/ 'aloeswood'. It seems to me that smoke was known to the Indo-European people before their contacts with "Atlantic" people. It barely resembles the mentioned Smc. word acoustically. Moreover V. overlooked that the smoke denoted by the cited root /q!-t!-r/ is not a general smoke, it is a smoke, an odour of (burning) sacrifice or a smell of alloeswood or incense. In B.-Hb. the root /q!-t!-r/ as action belonging to ritual is juxtaposed to the root /�-^s-n/ 'smoke'. When in modern Smc. languages the meaning of root /q!-t!-r/ q!-t-r/ appears as regular smoke, this occurs because of the modern semantic extension, as for example in M.- Hb. /miq!t!'eret/ '(smoke)pipe', but 'to smoke pipe' - /l� -�a^s^s�n miq!t!'eret/. Thus V.'s approximation if not impossible, but at least unproven and suspicious.

    Therefore it would be much more logical to adopt Pokorny's root er-/ or-/ r- Erweiterung reu-s- [Pokorny (1927-30), p. 332]. Also this Gmc. root looks akin to Gk. reo: / reu:ma / re�somai 'flow'

    6. Earth / Erde (V.'s pp. 254-55, 559, 614) PS +?ard! Akk./ers!etu/ 'Land, Erde', Hb. /?'eres!/, Arc. /?ar�a:/, S.-Ar. /?ard!(un)/ 'earth, land' (NB: all the etymologies cited from "the school Semitic languages"). The Proto-Semitic form of this word is */?aras!/ , see Dolgopolsky (1999), p. 25, #44. NB. Dolgopolsky used slightly different phonetic symbols.

    Since we cannot imagine ancient Arabs living in pre-historic Germanic lands, V.'s reconstruction of etymology for Gmc. earth / Erde is impossible. The th / d of earth / Erde are wrong phonemes in this reconstruction. In our opinion, Pokorny (1927-30), p. 332 has a satisfactory etymology for this word sub v. er- (er-t- / er-w-) and there is no need to look for a Smc. etymon even it seems very similar. The word /?'eres!/ is a common word in Hebrew Bible as in modern Hebrew, the common Ar. /?ard!(un)/ is frequent in all the periods and in various dialects. Nevertheless this is irrelevant for the etymology of earth / Erde.

    7. HS +/?abol/ 'genitals' in WIG (V.'s pp. 466, 564, 619-24) Orel 1995 No. 8 (cf. Gk. phall�s m. 'penis'); elsewhere in V.'s for 'balls' and 'apple'. V. cites also root */q!ol/ which in a number of Hamitic languages means 'testicles' or 'penis' Orel 1995:no. 2067, exactly as */q!�d/ 'genitals' ('penis, testicles, anus, vulva' Orel 1995:no. 1617). The sample has no phonetic similarity to proposed etyma. Elsewhere V. uses the same Hmc. word as an etymon for apple, which has more phonetic similarities, but nevertheless objectionable on extralinguistic grounds.

    8. WG and NG +folk- 'Kriegsschar, Volk' (V.s p. 665) 'division of an army', Gk. pe'lekys, Rs. polk; according to V., derived from Smc. root /p-l-h!/ 'spalten' Cf. Akk. /pilakku/ 'Spindel'. There is no semantic basis for this suggestion. Also both Akk. and Hb. consonant /p/ stems from PS */p/ (see Dolgopolsky 1999, p. 16-18), Ar. reflex /f/ is of later provenance (innovation in Arabic) and do not have any relation to the sound f of WG and NG +folk.

    9. Heer 'multitude, host, army' (V.'s p. 266); V. cites Brunner 1969:40, Nr. 166 who connects the word with Akk. /q!ar�bu/ (=Hb. /q!�ra:v/) 'Krieg, Kampf', /naq!rabu/ 'Kampf', /q!ar�^su/ 'Heer, Lager'. This is an obviously ridiculous approximation. There is no any phonetic or semantic basis for this suggestion. The original meaning of Smc. root /q!-r-b/ is 'to be near, to be close'. The secondary meanings are derived from the original one: 1) 'to be relative, kin; 2) to come close to God or to the altar - to offer sacrifice, from this we have Hb. /q!orba:n/, Ar. /q!urba:n/ both mean 'sacrifice, victim'; when two hostile armies come close, they initiate battle; from this derived the noun Hb. /q!�ra:v/ 'battle'. If it does mean sometimes in Akk. 'Heer, Lager', it is only by association with 'battle, war'. From here 'Heer, multitude, host, army' is still long way, especially because there is no similar phonemes. The Smc. root /q!-r-b/ has in common with Gr. /Heer/ only phoneme /r/. We do not see any possibility for the final consonant of the root /b/ to disappear with no traces or the strong emphatic velar plosive /q!/ to shift to /h/. If one wants to show the validity of his claim for the proposed etymology, he should prove these two phenomena, at least by providing parallel examples.

    10. Haus, house (V.'s p. 260) Akk. /x!us!s!u / Ar. /x!us!s!(un)/ 'hut, booth of reeds, tavern'. As we can see it is known from both Akk. and Ar. There was no need to demonstrate Ar. word. Although Pokorny (1927-30), p. 953 under 2. (s)keu-, (s)keu�: (s)ku: does not have a satisfactory etymology of this word and on p. 534, s.v. ca:s-, c_�s- has equally unclear etymon for Lat casa. There is no need to look for a Smc. etymon, the word attested in almost all Gmc. and IE languages. Moreover, there is no reasons to unite Haus and casa as derivatives of the same primary root even one seems very similar, the same can be said about Gk. /oikos/ 'house'. OG forms exhibit a variety forms such as Nrw. Hu:se, OHG hu:s, Sw. Hydda, which suggest affinity with Gr. H�tte and Eng. hut. Cf. Rs. and Ukr. /khata/ which provides an excellent example for a good illustration of the real connection of 'hut' and 'house'.

    As for L. casa, contrary to Tronsky 2001:110 who considered this a loan word in L., we can suggest the L. etymon capsa 'receptacle, repository, chest, box, container' [from Lat verb capio:, cepi, captum, cap#ere 'take, catch'] [root cap- + nominal suffix -s-+ nominal suffix of n.f. -a: (capsa > cassa > casa, i.e. regressive assimilation of /p/ preceding /s/ ). The metaphoric extension of the meaning from 'box' to 'hut' and further to 'house' is not too hard to imagine.

    11. (Unge)ziefer, Opfer (V.'s p. 266-67) Akk. /zibu/, Hb. /zevah!/, Arc. /devh!a:/, S.-Ar.-Eth. /zebh!/ Arab. /d!ibh!(un)/ all 'Opfer' (Bergstr�sser 1989:190) As we see the root is known in virtually all Smc. languages. The sounds' similarity is too slight. Besides the imagined Atlantic people would not need to loan a general term for sacrifice of animals to Indo-Europeans. Remember, according to V., the Vascons were the shepards, Atlantics were seafarers and warriors. The term /z'evah!/ in Semitic languages refers to slaughter of the sacrificial animals. Gesenius:1921:192 translates /z'evah!/ 'Schlachtopfer' as an opposition to other types of sacrifices, such as /minh!a:/ and /� o:la:/.

    12. +farh- WG 'Pig' (V.'s p. 664-65) V. notes that raising pigs was brought to the Northern Europe by the megalithic Semitides (Vennemann 1997b: chap. 11, sect. 3). A number of Ger words the author assumes to be of Semitidic origin, but he doesn't give any Semitidic evidence. (pp. 662-669). V. never brings the Sem. etymon, but he probably has in mind Hb. /h!#az�r/ or Ar. /x!inz�r /. Curiously enough, without any evidence V. attributes also the IE word for wild pig Ever/Eber to Sem etymon (Ar. /�ifr/ 'Eber, Ferkel' Akk. /app�rru/ 'Wildschwein' (V.'s pp.252, 560, 614). Since Ar. example is from much later period and it is most similar acoustically to WG word, V. had to find more ancient example exhibiting sound /f/ in this word, i.e. he had to chose rather examples from SA or from African Semitic languages (if this word is attested in them): the reflex of Proto-Semitic */p/ in them is /f/.

    It is known that boar hunting and pig domestication in Northern Europe is known from the Neolithic period (see e.g. in Encyclop�dia Britannica, 1971, v. 17, p. 1068, s.v. Pig). Atlantic people who arrived to Europe, according to V. in the post-Neolithic period, definitely didn't introduce this breed to Europe. If they didn't bring these animals to Europe, there was no need to name pigs and boars by Atlantic settlers.

    Even greater difficulties are found in V.'s persistence that Pct. was a descendent of Atlantic (=HS) languages and in the British toponyms that according to him originated from Pct. and that following Coates 1988a, 1988b, and 1988c (names of the strait The Solent, England; of a mountainous island Solund, Norway; and the Isles of Scilly) he explains from Hb. +s-l-� or Pl. +s-l-�-m 'rocks, cliffs'. Also after Coats 1988b:21 he explains the name Uist, Ibiza from Hb. /?�-b�sem/ [island of balsam or spice?] or /?�-besim/ [if one substitutes s for s! 'island of eggs'?]. The same way non-convincing is his etymology of the Hebrides (Ptolemy's Aibo�dai or Pliny's Hebudes from Akk. /pux!adu/ 'lamb' or Ug. /p-x!-d/ as 'Lamb Isles, Sheep Islands, or Isles of Fright')

    There are many more to say about the use of Smc. materials by V., but this does not change our conclusion that Smc. etyma supplied by V., do not depict the language of the "Atlantic" substratum of the IE or Gmc. languages.

    If V.'s idea that Atlantic people migrated from Africa along Atlantic coast holds true, he should present us etyma from the level of HS, i.e. AA, corresponding to the fifth millennium BC, not from the languages of the second and first millenia at the best. Even Akk. etyma copied from Soden 1965 are too late (into this category fall so called Pit- names of Pictland. Only one etymon in V.'s book is derived from Ha. (/tagus/ 'river') for rivers Tay in Pictland and Taw in England, which V. copies from Stumfohl 1989:137, in spite there were other suggestions for their etymologies, while /tagus/ is not registred on AA level. Among several dozen appellations for river in Orel 1995 /tagus/ is not mentioned. He explains the L. apis 'bee' after Brunner (1969) from Ancient Eg. as reduced form /af/ from /?fj/ (pp. 713-14, 723, 727) and IE root +bi- or *bhi- by a different Ancient Eg. form /bj-t/ 'honey bee'. On the bases of combination of Semitic +HVm- 'Volk, people' + bi- he reconstructs compound word Imme / imbi and its relatives in the WG languages. If this reconstruction is true, the question arises why only WG languages retain this word. If the Afroasiatics migrated from Africa along Atlantic coast northwards why these Imme-words didn't leave traces in any of the languages spoken south and west from the areas of WG tribes.

    In general the opinion of R.L. Trask about impossibility and silliness of linguistic reconstructions going back millennia before the first written records holds also for Atlantic (or Smc.) part of V.'s theory.

    >From a mere technical point of view, we should mention that the form of presentation of material could be greatly improved, had the author endeavored a re-writing of the book in the form of a monograph. This way he would be able to eliminate numerous repetitions, to smooth some inconsistencies and changes of opinions which came on different stages of his research, as well as to consolidate the topical subjects and groups of words in more organized form. It would be also for advantage to publish the book in one language, either German or English instead to mix German and English articles in one book. We do not believe, however, that the book could be improved conceptually, because the practical weaknesses of linguistic reconstruction do not help to support the theories propagated in the book.

    This book teaches us this lesson: However good a theoretician of linguistics one is, it is a paramount importance that he would master the languages he operates in his work. As far as it is known, and as it was stated by V. himself, he is no Vasconist, no Celticist and no Semitic or Afroasian linguist, he doesn't have special preparation in all these areas, neither in onomastics. To base such a big work on the etymologies taken from works of other scholars and from the large library of dictionaries is permissible only on condition that the author understands the linguistic implications connected to each word, each morpheme and each phoneme, the implications which are not usually supplied in the dictionaries. The author should follow the process of language change for each of the examples he provides as a proof of his theory. Speaking frankly, even the great linguists would not attempt historical reconstruction going back to the fifth millennium BC, especially in the condition of absolute lack of linguistic material going back even to the fourth or third millennia, not to speak that the linguistic material used in the dictionaries is of much later time. We should mention that V. failed in this book not only as comparative linguist, or etymologist, but even in his narrow specialization as a Germanist. [See Kitson (1996): 83, nn. 14-15]. Remember, in the beginning of this review I mentioned, that V.'s Ph.D. dissertation was on German phonology. Kitson (1996):110 shows that V. failed to give a rational explanation to the prevalence of o-grades over 'full' or 'normal' e-grades. He failed to recognize the Indo-European phonetic form of suffixes. He failed to treat existence of vowel a in the Indo- European roots.

    In short we consider the book a complete failure.

    We are not going to check V.'s extra-linguistic data; because we believe that first the linguistic facts should be proven. Now when we are sure that the linguistic base of V.'s theories is faulty beyond the reasonable doubt, there is no need to check his extra-linguistic evidence. However we can predict that exactly as both the P.-Bq and the PS reconstructions and approximations are faulty, the same will be found with V.'s extra-linguistic evidence.

    In a sense V. predicted his failure, however, formulating his thoughts in the form of three questions at the end of chap. 10. "Some West Indo- European words of uncertain origin" (originally published in 1997). There the author lists 3 important problems with which he doesn't deal, as follows:

    1. Will we be able to correlate the three series of Indo-European plosives at their various stages of development ... with the plosive series of the Vasconic and Atlantic languages ...?

    2. Will we be able to establish partial chronologies of prehistoric changes in the West Indo-European languages by comparing the structure of putative loan-words to their assumed sources...?

    3. Can it be proved ... that the amazingly regular Germanic verbal ablaut came into being when speakers of Atlantic languages shifted from their rigorously ablauting languages to the mildly and irregularly ablauting Palaeo-Germanic system?

    Today we can answer all three problematic questions in negative. We will not be able to correlate the three series of Indo-European plosives at their various stages of development with the plosive series of the Vasconic and Atlantic languages. We will not be able to establish any chronologies of prehistoric changes in the WIE languages going back to the eighth thousand BC or to the fifth thousand BC by comparing the structure of putative loan-words to their assumed sources. And the problem depends both on IE part and on "putative loan- words". We believe that the "putative loan-words" were too putative ones, i.e. they do not have any relation to IE or its Western branch. It cannot be proved that the amazingly regular Gmc. verbal ablaut came into being when speakers of Atlantic languages shifted from their rigorously ablauting languages to the mildly and irregularly ablauting Palaeo-Germanic system. (See pp.625-627). The languages do not borrow general ideas of structure one from another. Structural similarities can be of two origins, genetic or typological [i.e. as realization of one of many possibilities]. In Semitic, where the structure originated from the common proto-Semitic, the extraordinary systematization of ablaut refers to structure of contrasting verbal stems: basic, intensive, causative, passive, reflexive, etc., including such rare formations as to indicate meaning 'to allow to do something concerning one,' or to point to color or bodily defects, 'to ask or consider to do something' [See Moscati (1969):122-131; Wright 1966:198-226]; and the long series of nouns' models. The systematization of rich ablaut in Semitic is insured by the prevalence of tri-consonantal roots. In the IE, especially in Gmc., languages the ablaut forms are used only for distinction of tenses. Much rarer are changes of vowels in conjugation of verbs (like in Sp. dorm'ir / duerme). In Gmc., there are very few examples that can be interpreted as contrastive verbal stems (a kind of sitzen 'sit, be seated' and setzen 'place, set, put', Engl. sit and set), which reminds on relation of base stem and causative stem in Smc. (like Qal and Hif'i:l stems in Hb., or the first and the fourth formations in Arab). If one designates this resemblance as similarity at all, this has to be a limited typological similarity.

    V. instructs Indo-Europeanists to turn from problems of internal Indo-European reconstruction to the study of their contact languages. Any well versed historical linguist would be wondering as to what IE, Nostratic or Eurasian form V.'s followers should compare the reconstructed forms of contact languages.

    We mentioned above that the book was well edited. A great recognition for editing the book deserves the editor Patrizia Noel Aziz Hanna who assembled all the articles, edited them, composed indexes, wrote the introduction and supplied the abstracts. She and several students of Prof. Vennemann have proofread the text of book. Nevertheless occasional mistakes and typos occur, especially in linguistic examples from less known languages, e.g. discussing the name of the alder on p. 327 V. brings Russ. 'ol#icha (< +alisa:). The real Russian word is /ol'kh'a/; on p. 708 V. brings Rs. /serp#u/ 'sickle'. The correct word should be /serp/.

    REFERENCES Agud-Tovar (1989) Agud, Manuel & Antonio Tovar. Diccionario etimol�gico vasco. San Sebastian: Deputaci�n Foral de Guip�zcoa, 1989-

    Azkue (1984) Azkue, Resurrecci�n Mar�a de. Diccionario vasco-espa�ol- franc�s. Reprint edition (with an introduction of Luis Michelena). Bilbao: Euskaltzaindia, 1984. [First published in 1905, 2nd ed., 1969].

    Bergstr�sser (1928) Bergstr�sser, Gotthelf. Einf�hrung in die semitischen Sprachen : Sprachproben und grammatische Skizzen. M�nchen: Max Hueber, 1928. [1989 reprint].

    Brunner (1969) Brunner, Linus. Die gemeinsamen Wurzeln des semitischen und indogermanischen Wortschatzen: Versuch einer Etymologie. Bern: A. Francke, 1969.

    Cavalli-Sforza (1994) Cavalli-Sforza, Luigi Luca, Paolo Menozzi, and Alberto Piazza. The history and geography of human genes. Princeton, N.J.: Princeton University Press, 1994.

    Champion (1984) Champion, Timoty C., Clive Gamble, et al. Prehistoric Europe, by T C. London ; Orlando : Academic Press, (c)1984.

    Cohen (1970) Cohen, David. Dictionnaire des racines s�mitiques. Paris- La Haye, 1970.

    Djakonov (1981-86) Djakonov, I.M., ed., Belova, A.G., et al. Sravnitel'no-istoricheskii slovar' afraziiskikh iazykov. Moscow, 1981- 1986. 3 vols.

    Djakonov (1994-95) Djakonov, I. M., ed., Belova, A. G., Chetverukhin, Militarev, A. Ju. [and] Porchomovskij, V. Ja., "Historical Comparative vocabulary of Afrasian," in St. Petersburg Journal of African Studies, 1994-1995, nos. 2-5.

    Djakonov (1996) Djakonov, I. M. [and] L. E. Kogan, "Addenda et Corrigenda to Hamito-Semitic Etymological Dictionary by V. Orel and O. Stolbova," ZDMG 146/1 (1996), 25-44.

    Dolgopolsky (1999) Dolgopol'sky, A. B. From Proto-Semitic to Hebrew: Phonology : Etymologica Approach in a Hamito-Semitic Perspective. Milano: Centro Studi Camito-Semitici, 1999.

    Gensler (1993) Gensler, Orin David. A typological evaluation of Celtic/Hamito-Semitic syntactic parallels. Unpublished Ph.D. dissertation, University of California, Berkeley, 1993.

    Gesenius (1921) Gesenius, Wilhelm. Hebr�isches und aram�isches Handw�rterbuch �ber das Alte Testament. 17.te Aufl. Leipzig: Verl. Von F. C. W. Vogel, 1921.

    Jackson (1980) Jackson, K. H. "The Pictish Language," in: Frederick Threlfall Wainwright (ed.), The Problem of the Picts. Perth: Melven Press, 1980. [Reprint of the 1955 ed.], pp. 129-166 and appendices I (161-166) and III (173-176).

    Kitson (1996) Kitson, Peter R., "British and European river names," in Transactions of the Philological Society 94 (1996):73-118.

    Kogan (2002) Kogan, L. E. "Addenda et Corrigenda to Hamito-Semitic Etymological Dictionary by V. Orel and O. Stolbova," JSS 47/2 (Autumn 2002):183-202.

    Lakarra (1996) Lakarra, Joseba A. "Sobre el Europeo Antiguo y la reconstrucci�n del Protovasco," in Anuario del Seminario de Filolog�a Vasca Julio de Urquijo (ASJU) 30/1 (1996):1-70.

    Levin (1995) Levin, Saul. Semitic and Indo-European: The principal etymologies: With observations on Afro-Asiatic. Amsterdam: John Benjamins, 1995 (Amsterdam Studies in the Theory and History of Linguistic Science 129.)

    L�pelmann (1968) L�pelmann, Martin. Etymologisches W�rterbuch der baskischen Sprache: Dialekte von Labourd, Nieder-Navarra und La Soule. Berlin: Walter de Gruyter, 1968. 2 vols.

    Militarev (2000) Militarev, A. [and] L. Kogan, Semitic Etymological Dictionary, Vol. 1: Anatomy of Man and Animals. M�nster: Ugarit Verlag, 2000 (Alter Orient und Altes Testament, vol. 278/1)

    Morris Jones (1900) Morris Jones, John. "Pre-Aryan syntax in Insular Celtic," John Rhys and David Brynmor-Jones, The Welsh people: Chapters on their origin, history, laws, language, literature and characteristics. London: T. Fisher Unwin, 1900, appendix B, pp. 617- 641.

    Moscati (1969) Moscati, Sabatino. An Introduction to the Comparative Grammar of the Semitic Languages. Wiebaden: Otto Harrassowitz, 1969 (Porta Linguarum Orientalium).

    Newstead (1979) Newstead, Helaine. "The origin and grows of the Tristan legend," in Roger Sherman Loomis, ed., Arthurian literature in the Middle Ages: A collaborative history. Oxford: Clarendon Press, 1959 [1979 reprint.]

    Orel (1995) Orel, Vladimir E. & Olga V. Stolbova. Hamito-Semitic etymological dictionary: Materials for a reconstruction. Leiden: E. J. Brill, 1995 (Handbuch der Orientalistik, 18).

    Pokorny (1927-30) Pokorny, Julius. Indogermanisches etymologisches W�rterbuch. 2 vols. (2nd vol., register by Harry B. Partridge.) 2nd ed. Bern: A. Francke, 1989. [Reprint of the 1959 ed.; I used 3rd ed. T�bingen and Basel: A. Francke, 1994.]

    Renfrew (1999) Renfrew, Colin and Nettle, Daniel, eds. Nostratic: examining a linguistic macrofamily. Cambridge: The McDonald Inst. For Archaeological Research, 1999.

    Renfrew (2000) Archaeogenetics: DNA and the Population Prehistory of Europe, ed. by Colin Renfrew and Katie Boyle. Cambridge University, The McDonald Institute for Archaeological Research, 2000 (1999 Euroconference of the Human Genome Diversity Group)

    Schmid (1987) Schmid, Wolfgang P. "'Indo-European' 'Old European' : on the reexamination of two linguistic terms," Proto-Indo-European: The Archeology of a Linguistic Problem. Studies in Honor of Marija Gimbutas, ed. by S.N. Skomal & E.C. Polom�. Washington, DC: Institute for the Study of Man, 1987, pp. 322-338.

    Sheratt (1997) Sheratt, Andrew. Economy and Society in Prehistoric Europe : Changing Perspectives. Princeton University Press, 1997.

    Shevoroshkin (1989) Shevoroshkin,Vitaly V., ed. Reconstructing languages and cultures: abstracts and materials from the first international interdisciplinary symposium on language and prehistory, Ann Arbor, 8-12 November, 1988. Bochum: Brockmeyer, 1989.

    Soden (1965) Soden, Wolfram von. Akkadisches Handw�rterbuch. 3 vols. Wiesbaden: Otto Harrassowitz, 1965-

    Stumfohl (1989) Stumfohl, Helmut. "Die wissenschaftliche Position Dominic W�lfels im Jahre 1988," Almogaren 17-19 (1989):113-155.

    Trask (1995) Trask, R. L. "Origin and Relatives of the Basque Language: Review of the evidence," in Towards a History of the Basque Language, ed. J. I. Hualde et al., Amsterdam/ Philadelphia: John Bennjamins, 1995, pp. 65-77.

    Trask (1997) Trask, R. L. The History of Basque. London and New York: Routledge, 1997.

    Tronsky (2001) Tronsky, I. M. Istoricheskaia grammatika latinskogo iazyka: obstcheiazykovoe sostoianie (voprosy rekonstruktsii). 2-e dop. izd. Moscow: Izdatel'stvo "Indrik" , 2001. [Russian]

    Vennemann (1994) Vennemann, Theo. "Linguistic reconstruction in the context of European pre-history," in Transactions of the Philological Society 92 (1994): 215-284 (chap. 6 in Europa Vasconica - Europa Semitica, pp. 139-201)

    Vennemann (1997b) Vennemann, Theo. "Atlantiker in Nordwesteuropa: Pikten und Vanen," in Stig Eliasson and Ernst H�kon Jahr, eds. Language and its ecology: Essays in memory of Einar Haugen. Berlin: Mouton de Gruyter, 1997. (chap. 11 in Europa Vasconica - Europa Semitica, pp. 371-395), pp. 451-476.

    Vennemann (2000b) Vennemann, Theo. "English as a 'Celtic' language: Atlantic influences from above and from below," in Hildegard L.C. Tristram, ed. The Celtic Englishes II. Heidelberg: Carl Winter, 2000. (Anglistische Forschungen, 286), pp. 399-406.

    Vennemann (2001c) Vennemann, Theo. "Atlantis Semitica: Structural contact features in Celtic and English," in Laurel Brinton, ed. Historical Linguistics 1999: Selected papers from the 14th International Conference on Historical Linguistics, Vancouver, 9-13 August 1999. Amsterdam: John Benjamins, 2001. (Anglistische Forschungen, 286), pp. 351-369.

    Vennemann (2002c) Vennemann, Theo. "On the rise of 'Celtic' syntax in Middle English," in Peter J. Lucas and Angela M. Lucas, eds. Middle English from tongue to text: Selected papers from the Third International Conference on Middle English: Language and Text, held at Dublin, Ireland, 1-4 July 1999. Bern: Peter Lang, 2002. (Studies in English Medieval Language and Literature, 4), pp. 203-234.

    Vennemann (2002d) Vennemann, Theo. "Semitic - Celtic - English: The transitivity of language contact," in Markku Filppula, Juhani Klemola, and Heli Pitk�nen, eds. The Celtic roots of English. Joensuu: University Joensuu, 2002. (Studies in Languages, 37), pp. 295-330.

    Wright (1966) Wright, William. Lectures on the Comparative Grammar of the Semitic Languages. Cambridge: Cambridge University Press, 1890; reprint, Amsterdam: Philo Press, 1966.

    ABOUT THE REVIEWER Hayim Y. Sheynin studied General linguistics, Classical philology, Semitic philology and Jewish interlinguistics in Leningrad (now St. Petersburg, Russia), Jerusalem and Philadelphia. He holds Ph. D. degree in the Oriental languages from the University of Pennsylvania. He taught Semitic languages and Hebrew literature in the University of Haifa, Dropsie University and Gratz College. He participated in several projects of Aramaic lexicography, description of Hebrew manuscripts, publication of poetic and philological texts from the Cairo Genizah, etc. The recent interests include lexicology of Semitic and Jewish languages. He contributed reviews of monographs on historiography of linguistics, on Romance languages and on Arabic to Linguist List and to Studies in Linguistics and on Hebrew manuscripts to Jewish Quarterly Review and other periodicals and festschrifts.