Reviewed by Zsolt Simon, Hungarian Academy of Sciences (firstname.lastname@example.org)
Four years after the publication of The Cambridge Encyclopedia of the World's Ancient Languages (ed. by Roger D. Woodard) that aroused much scholarly appreciation,1 a paperback version of the Encyclopedia has come out. Not in one bulky volume, but in five volumes of average thickness, in the usual layout standards of the publisher. The five volumes have been arranged in geographical order, i.e., (1) Mesopotamia, Egypt and Aksum; (2) Syria-Palestine and Arabia; (3) Asia Minor; (4) Asia and the Americas; (5) Europe. It is the last one that is reviewed here.
As the editor has chosen the 5th c. A.D. as a terminus ante quem for 'ancient' languages because of pragmatical reasons (p. xv-xvi), the volume contains the following languages: Attic Greek, Greek dialects (pp. 14-49 and 50-72, resp., both by the editor), Latin (pp. 73-95, by James P. T. Clackson), Sabellian (pp. 96-123, actually only Oscan and Umbrian, since South Picene is underrepresented in the chapter), Venetic (pp. 124-140, both by Rex E. Wallace), Etruscan (pp. 141-164, by the late Helmut Rix), Continental Celtic (pp. 165-188, by Joseph F. Eska), Gothic (pp. 189-214, by Jay H. Jasanoff), Ancient Nordic (pp. 215-229, by Jan Terje Faarlund) and an Appendix on the reconstructed Proto-Indo-European by the late Henry M. Hoenigswald, the editor and J. P. T. Clackson (pp. 230-246). An introductory chapter is devoted to a wide selection of undeciphered and fragmentary languages (pp. 1-13, by the editor).2
The reason for this selection is not quite clear. All literate languages are included -- at least per tangentem. The omission from such an encyclopaedia of non-literate languages known only from onomastic material, glosses or some loanwords is defensible. Yet since the non-literate Illyrian has also been treated here (extensively, at greater length than e.g. Messapic), Scythian or Sarmatian - also known from onomastic material only and from Nebenüberlieferung -- should also have been included (especially because in fact, they are slightly better known, than Illyrian3). And since Illyrian is included, the similarly nebulous Ostalpenindogermanisch (the language of pre-Roman Alpine tribes and the Pannon) should also have a place here.4
Individual chapters provide a short descriptive grammar of the given languages and dialects in a unified and straightforward manner: 1. Historical and cultural contexts. 2. Writing system(s). 3. Phonology. 4. Morphology. 5. Syntax.5 6. Lexicon. 7. Reading List and Bibliography. Some authors have slightly diverged from this general pattern (Gothic lacks 'Lexicon'), the only unfortunate cases are where the section 'Reading List' is missing (Latin, Sabellian, Venetic, Etruscan, Ancient Nordic): as these chapters, of course, cannot replace the detailed standard grammars, an annotated 'Reading List' is of great importance for those who decide to explore a given language further.
As can be observed from this general structure, the main--logical--goal was to give a synchronic description. But Woodard's chapters (Attic Greek and Greek dialects) are overloaded with diachronic information: not only with the changes from Proto-Greek to the dialects (which are, of course, necessary to understand the dialect forms), but also with the Proto-Indo-European background, which is absolutely unnecessary from a synchronic point of view.6
Needless to say, such short descriptions should be up to date and based on the communis opinio of the scholarly community, and idiosyncratic views should be minimized. By and large this volume is in this tradition, though minor flaws can still be frequently found; they have been listed by earlier reviewers and therefore do not need to be repeated here.7 Only the first section on Latin and some parts of the introduction pertaining to the fragmentary languages can be seriously challenged.
J. P. T. Clackson is a well-known follower of the hypothesis of the secondary convergence of Sabellian and Latino-Faliscan, instead of an Italic branch, since 'it has proved difficult to demonstrate conclusively that these similarities result from genetic affiliation, and have not arisen through convergence of separate branches of Indo-European over time. Our present state of knowledge of Sabellian and the early history of Latin is not sufficient to allow a definite answer to this question' (p. 74). However, this is not the case. Handbooks of Latin historical grammar contain long lists of unique shared innovations of Latino-Faliscan and Sabellian (which is the only way to prove the existence of an Italic branch), some of them are quoted by Wallace in the same volume, too (p. 97), so Clackson's view cannot be upheld today.8
Though by adopting a periodization of Latin 'Early or Old Latin, from earliest times to c. 100 B.C.; Classical Latin, c. 100 B.C.-A.D. 14, Post-Classical A.D. 14-c. 400, Late Latin from c. 400 onwards' (p. 74) Clackson is clearly in renowned company;9 this periodization is obviously based on the history of the Latin literature, and not on that of the language itself. At the same time, there are other well-known periodizations, based on the changes of the Latin language.10 Wherever one puts the obviously arbitrary borders, the period before the fundamental changes of vowel weakening, syncope and rhotacism definitely must be separated from the periods after these changes and cannot be treated together under the simplifying heading 'Early or Old Latin'. Another disadvantage of Clackson's periodization that it lacks a date to mark the end of Latin, which, of course, was different depending on the region, but it can be dated by sociolinguistic devices around 750 in Gaul and around the turn of 9th and 10th c. in Italy and in Hispania.11
Contra the introduction (p. 4-5), Early Insular Celtic is known not only from Ogham Irish, but also e.g. from the inscriptions of British coinage.12 And if one accepts the philologically sound argumentation of K. Forsyth that Pictish is actually a Celtic language (Woodard does not mention this work, p. 5), then the scope becomes even wider.13 What is certain, however, is that there was no 'Thraco-Phrygian' (p. 9), since Phrygian is clearly very closely related to Greek (and both belong to Balkan Indo-European / Balkanindogermanisch).14 More than a decade after the discovery (and publication) of the defixio of Pella the treatment of Macedonian (pp. 9-11) as a distinct language, and not a Greek dialect is questionable, and only a mention of this tablet without its consequences is not enough.15
The only main problem of this volume is its up-to-dateness. Even the manuscripts of the hardback edition were completed well before its publishing (as Rix himself noted in the chapter on Etruscan, he couldn't include the results of two books from 1999 and 2000 (!)) and also the past four years have seen the birth of many important monographs.16 The above-mentioned reviews of the hardback edition suggest many corrections. Thus both these corrections and an addition would have been very useful, and this reprint would have been a good chance to include these critical remarks, which unfortunately never happened (a funny consequence is that Maps 1. & 2. (p. 49, 123) are already historical, since they show Yugoslavia as an existing country).
To sum up: this volume is a useful reprint for those who cannot pay $160 for a single book (especially if they are interested only in some chapters), but at the same time a missed chance for updating and making an excellent book even more reliable.
1. See the generally very positive reviews: Yves Duhoux: Les langues de l'antiquité: un important nouveau manuel. Cahiers de l'Institut de Linguistique de Louvain 30/4 (2004) 73-80; Joshua T. Katz: BMCR 2005. 08. 36; K. Jongeling: BiOr 62 (2005) cols. 620-622; Eirik Welo: JIES 33 (2005) 439-450; Jared S. Klein: An Encyclopedia of Ancient Languages. JAOS 125 (2005) 91-97; Raimo Anttila: Diachronica 23 (2006) 448-452; Karl Horst Schmidt: Krat 51 (2006) 65-69.
2. The volume is introduced by technical lists and two prefaces by the editor (pp. v-xiv.) and rounded off with full tables of contents of the former and paperback versions and with indices (pp. 247-261).
3. Cf. Rüdiger Schmitt: Andere altiranische Dialekte and Roland Bielmeier: Sarmatisch, Alanisch, Jassisch. In: Rüdiger Schmitt (ed.): Compendium Linguarum Iranicarum. Wiesbaden, 1989, 92-93, 236-245, resp.
4. See the works of Peter Anreiter: Breonen, Genaunen und Fokunaten. Vorrömisches Namengut in den Tiroler Alpen. Budapest, 1997; Der Ablaut in 'ostalpenindogermanischen' Namen. In: id.-Jerem, Erzsébet (eds.): Studia Celtica et Indogermanica. Festschrift für Wolfgang Meid zum 70. Geburtstag. Budapest, 1999, 23-38; Die vorrömische Namen Pannoniens. Budapest, 2001. For criticism see e. g. Béla Adamik: Review of Anreiter 2001. Acta Antiqua Academiae Scientiarum Hungaricae 43 (2003) 262-268.
5. Syntax is treated in the philologically oriented, traditional and the (post)modern way. The continuous emphasis on head-first/head-final typology is cumbersome, since the authors who cite it need to add the counterexamples immediately (e. g. p. 90, 180, 227.). It is worth noting that although poetic texts are widely held as ill-fitted for syntactical investigations, Clackson cites Vergil's Aeneid I. 109 for scrambling and extreme displacement of the relative pronoun (p. 90.).
6. Since Woodard does not draw a clear-cut line between descriptive and historical linguistics, it is logical that he suggests only historical Greek grammars as 'excellent linguistic overviews' (p. 48).
7. See esp. Duhoux (n. 1.), Klein (n. 1.) 92-96 and Katz (n. 1.). Some additions: The problem of Latin /h/ (which disappeared before rhotacism and was resuscitated later in educated pronunciation only) remains unmentioned (p. 76-77). Feriae, uti, sacer, pius and cena are not restricted to Italic (p. 121), see e.g. Michiel de Vaan: Etymological Dictionary of Latin and the other Italic Languages. Leiden-Boston, 2008, 212-213, 648, 532, 468, 106, resp. The speculative nature of the view that '[t]he nonsyllabic allophones of the first two laryngeals seem to be voiceless, that of the third, voiced' (p. 233) should have been indicated. There was no neutralization of final *-ms and *-ns in PIE (p. 234) and thus acc. pl. cannot be reconstructed as *-ms / *-ons (p. 239) as the Hittite acc. pl. -us < *-Cms / *-oms clearly shows (see already H. Craig Melchert: Anatolian Historical Phonology. Amsterdam-Atlanta, 1994, 180, 185-186). When describing the PIE root structure (p. 237) it would have been worth noting that it does not apply to grammatical words. The view '*so ,,and he" (maintained as such in Hittite)" (p. 240) is false: su in Hittite is a semantically neutral sentence-initial conjunctive particle with preterite verbs of Old Hittite (Joseph Weitenberg: The Use of Asyndesis and Particles in Old Hittite Simple Sentences. In: Onofrio Carruba (ed.): Per una grammatica ittita. Pavia, 1992, 305-353), and for the etymological problems see now Alwin Kloekhorst: Etymological Dictionary of the Hittite Inherited Lexicon. Leiden-Boston, 2008, 772). The augment cannot be projected into PIE time (p. 241) if it is attested only in the closely related groups of Indo-Iranian and Balkan Indo-European languages.-The reading lists sometimes lack important text-editions and monographs: of the Rhaetic (Stefan Schumacher: Die rätischen Inschriften. Geschichte und heutiger Stand der Forschung. Budapest, 1992) and the Messapic inscriptions (Carlo de Simone - Simona Marchesini: Monumenta linguae messapicae 1-2. Wiesbaden, 2002). Helmut Rix's analysis of Rhaetic and Etruscan is a must (Rätisch und Etruskisch. Innsbruck, 1998), the same goes for the dissertation of Joshua T. Katz on PIE pronouns (p. 240). In the case of Mycenean dialects (p. 94), Ivo Hajnal: Sprachschichten des mykenischen Griechisch. Salamanca, 1997 is missing. For the history of Latin, Baldi's somewhat problematic work (see the review by Gerhard Meiser: Krat 47 (2002) 108-115) has been cited, but Andrew L. Sihler: A New Comparative Grammar of Greek and Latin. New York - Oxford, 1995 not (p. 94).
8. Some unique shared morphological innovations can be added to his list, only to quote the firmest evidence: 1) the merge of ablative and instrumental (in abl. sg. and instr. pl.); 2) adverbs on *-ēd from adjectives of the 1st and 2nd declinations; 3) the transformation of the present paradigm of the copula (*esom/*som); 4) the transformation of PIE present-types into the same four conjugations; 5) the *-k-extension of the root *dheh1- 'to do'. For full lists (including the phonological isoglosses, discussions and refs. to earlier literature), see esp. Helmut Rix: Latein und Sabellisch. Stammbaum und/oder Sprachbund? IncLing 17 (1994) 13-29; Ausgliederung und Aufgliederung der italischen Sparchen. In: Alfred Bammesberger, Theo Vennemann (eds.): Languages in Prehistoric Europe. Heidelberg, 2003, 150-156; Gerhard Meiser: Lautgeschichte der umbrischen Sprache. Innsbruck, 1986, 36-38; Veni vidi vici. Die Vorgeschichte des lateinischen Perfektsystems. München, 2003, 30-33; Frank Heidermanns: Nominal composition in Sabellian and Proto-Italic. TPS 100 (2002) 185-202; Peter Schrijver: Review of Meiser 2003. Krat 51 (2006) 46-64; Brent Vine: On 'Thurneysen-Havet's Law' in Latin and Italic. HS 119 (2006) 211-249; de Vaan (n. 7.) 5. Cf. further Gert Klingenschmitt: Die lateinische Nominalflexion. In: Oswald Panagl, Thomas Krisch (eds.): Latein und Indogermanisch. Akten des Kolloquiums der Indogermanischen Gesellschaft 1986 in Salzburg. Innsbruck, 1992, 89-135; Robert van der Staaij: A Reconstruction of Proto-Italic. PhD dissertation. Leiden University; Patrizia de Bernardo Stempel: Kernitalisch, Latein, Venetisch. Ein Etappenmodell. In: Michaela Ofitsch - Christian Zinko (eds.): 125 Jahre Indogermanistik in Graz. Festband anlässlich des 125j ährigen Bestehens der Forschungsrichtung 'Indogermanistik' an der Karl-Franzens-Universität, Graz. Graz, 2000, 47-70.
9. Similarly e. g. Gerhard Meiser: Historische Laut- und Formenlehre der lateinischen Sprache. Darmstadt, 1998, 2; de Vaan (n. 7) 14.
10. Oswald Szemerényi: Latein in Europa. In: Karl Buüchner (ed.): Latein und Europa. Traditionen und Renaissancen. Stuttgart, 1978, 26-31 (= Scripta Minora II. Innsbruck, 1987, 1002-1007); Helmut Rix: Das letzte Wort der Duenos-Inschrift. MSS 46 (1985) 193-194 with n. 6; followed by Heiner Eichner: Reklameiamben aus Roms Königszeit. Die Sprache 34 (1988-1990) 223, n. 32.
11. See esp. József Herman's detailed analysis: The end of the history of Latin. Romance Philology 49 (1996) 364-382.
12. See Patrizia de Bernardo Stempel: Die Sprache altbritannischer Münzlegenden. ZCP 44 (1991) 36-55.
13 Katherine Forsyth: Language in Pictland. The Case Against 'Non-Indo-European Pictish'. Utrecht, 1997. Jongeling (n. 1.) 620 n. 1. already missed this reference.
14. It will be noted that Woodard refers to the Phrygian chapter where Brixhe dismisses Thraco-Phrygian in favour of a close relationship with Greek. On the relationship of Phrygian see esp. Günter Neumann: Phrygisch und Griechisch. SbO+AW 499. Wien, 1988; Gert Klingenschmitt: Die Verwandschaftsverhältnisse der indogermanischen Sprachen. In: Jens E. Rasmussen (ed.): In honorem Holger Pedersen. Kolloquium der Indogermanischen Gesellschaft vom 25. bis 28. März 1993 in Kopenhagen. Wiesbaden, 1994, 244-245; Joachim Matzinger: Phrygisch und Armenisch. In: Olav Hackstein - Gerhard Meiser (eds.): Sprachkontakt und Sprachwandel. Akten der XI. Fachtagung der Indogermanischen Gesellschaft, Halle an der Saale, 17.-23. September 2000. Wiesbaden, 2005, 381-386.
15. The discovery has been followed by lively scholarly discussion, see esp. Laurent Dubois: Une tablette de malédiction de Pella: s'agit-il du premier texte macédonien? REG 108 (1995) 190-197; Claude Brixhe: Un 'nouveau' champ de la dialectologie grecque: Le Macédonien. AION 19 (1997) 41-71; Ivo Hajnal: Methodische Vorbemerkungen zu einer Palaeolinguistik des Balkanraums. In: Bammesberger, Vennemann (n. 7.) 123-124 and most recently James L. O'Neil: Doric forms in Macedonian inscriptions. Glotta 82 (2006) 192-210.
16. E.g. an excellent introduction to PIE linguistics (Benjamin W. Fortson: Indo-European Language and Culture. An Introduction. Oxford, 2004), the new edition of Rhaetic inscriptions (Budapest, 2004, cf. n. 7.) and a detailed analysis of the relationship of Messapic (Joachim Matzinger: Messapisch und Albanisch. IJDL 2 (2005) 29-54).