HARALD BICH LMEIER

Sächsische Akademie der Wissenschaften zu Leipzig Martin-Luther-Universität Halle-Wittenberg

Research fields: Indo-European studies, onomastics (place names and water names), Old Iranian studies (syntax).

ON THE LINGUISTIC LEVEL OF RESEARCH ON OLD EUROPEAN HYDRONYMY – A RESPONSE TO A POLEMIC

On the scholarly level of research into Old European hydronymy – a response to a polemic

Etymology is a science and is subject to the same criteria of assessment as other sciences. It can be practiced naively, that is, with little or no regard for the relevant state of knowledge already attained, or on a sound basis, whereby every consideration of relevant knowledge contributes to soundness, so that naivety and soundness constitute a gradual criterion of evaluation.¹

NOTE

This contribution is a response to an article in this journal, written in a comparatively polemical tone and ultimately arguing in an unscientific manner. It points out the unmistakable and considerable gaps in that author's knowledge of historical-comparative linguistics in general and Indo-European studies in particular, and concludes that the work of that author in these fields, using

¹ Vennemann 1999: 275; Fettdruck im Original.

KEYWORDS: Old European hydronymy, Indo-European studies, onomastics, word formation, Proto-Indo-European ablaut, reconstruction, laryngeals.

KEYWORDS: Old European Hydronymy, Indo-European linguistics, onomastics, word formation, Proto-Indo-European ablaut, reconstruction, laryngeals.

Straipsniai / Articles

9

HARALD BICH LMEIER

The results obtained partly on the basis of linguistic views and methods that have been outdated for decades ultimately all require renewed examination using up-to-date linguistic methodology. Only after this examination will it be possible to assess how much of those works and their results will stand the test of time.

ANNOTATION

This article is a response to another article in this journal written rather polemically and arguing quite unscientifically. This article will show the deficiencies of knowledge the other article’s author betrays as far as historical-comparative linguistics in general and Indo-European linguistics especially are concerned. As a conclusion it must be stated that all the results the other author has achieved by using research methods in these fields outdated for decades will have to be reevaluated finally applying modern Indo-European linguistics to them. Only after this task will have been accomplished, one will be able to tell, which of this author’s works and results will be safe for further use.

1. PRELIMINARY REMARK

Unlike Jürgen Udolph, unfortunately I cannot remember the exact date when I became interested in onomastics and the study of hydronyms. In any case, it was connected with the review of a collected volume on onomastics², which I undertook in 2005 or 2006 at Wolfgang Haubrichs’s suggestion.³ My acquaintance with Udolph probably also dates to 2006; I likely first met him at the conference Interference Onomastics. Names in Borderlands and Contact Zones in History and the Present. Saarbrücken Colloquium of the Working Group on Name Research in Saarbrücken, October 5–7, 2006. At this conference, he personally urged me for the first time (and repeated this appeal to me several times at various conferences, as well as addressing it to Indo-Europeanists as a whole—for example, in his paper at the conference of the Indogermanische Gesellschaft, The Expansion of Indo-European. Theses from Linguistics, Archaeology, and Genetics, Würzburg, September 24–26, 2009) that Indo-Europeanists should finally return to the study of Old European hydronymy. I took Udolph’s appeal to heart and thus discovered a new field of research for myself, one in which—as the above-mentioned review had already shown—there was still quite a lot to be done.

² Greule, Janka, Prinz 2005.

³ Cf. Bichlmeier 2007.

10

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

Given that I did precisely what Udolph asked me (or Indo-Europeanists as a whole) to do, his criticism of some of my previous statements, advanced in a rather polemical tone in a contribution to this journal⁴, seems rather incomprehensible. For what else could have emerged, when someone approached material that, up to that point, had apparently been investigated by Udolph et al. only with the methods of the years before 1970, or of the prewar period, than the realization that the methods used to study Old European hydronymy require a fundamental revision—and that the original results must then inevitably also be revised, at least in part—if the material is approached with the level of knowledge and expertise available in 2010?

Significantly, the same article that prompted Udolph’s polemic elicited an almost diametrically opposite response from colleagues open to scientific progress: Albrecht Greule (2011: 125)⁵ emphasizes that, in light of the arguments presented in that contribution, onomastic works, too, should no longer dispense with the use of Indo-Europeanist conventions for notating Proto-Indo-European root reconstructions.⁶ A long-overdue paradigm shift is taking shape here.

Perhaps, however, Udolph’s appeal was merely rhetorical, and he actually wanted to continue publishing about these hydronyms all alone in the field—who knows?

In any case, the accusations made by Udolph in the aforementioned publication, as well as his once again evident lack of familiarity with historical linguistics research in general and Indo-Europeanist research of the past half-century in particular, call for a review of the state of the field and a detailed response.⁷

⁴ Cf. Udolph 2011.

⁵ The statement made in the same place that laryngeal theory is “not readily accessible” remains incomprehensible: there are currently about a dozen introductions in German and English to Indo-European studies or laryngeal theory that cover this subject. Moreover, it is taught at a number of chairs in German-speaking countries; laryngeal theory should generally be part of the curriculum in the second semester, or third at the latest. In my view, it is also not substantially more complex than Verner’s law (there are simply far more examples, in considerably more languages, to which it is relevant), and it was already a standard component of my intermediate examination in 1992.

⁶ Ebenso urteilen Nübling, Fahlbusch, Heuser (2012: 224f. mit Anm. 215) unter Verweis auf Bichlmeier (2011a: 180f.). Leider bleibt die Darstellung aber abgesehen von diesem Hinweis hier ebenso auf dem veralteten Stand wie hinsichtlich der Namen theorie, die kaum über den Stand der 1960er Jahre hinausreicht.

⁷ I thank the editor, Grasil da Blažienė, for inviting me to write this response and for agreeing to publish it in this journal.

Straipsniai / Articles

11

HARALD BICH LMEIER

2. INDO-EUROPEAN LINGUISTICS AND OLD EUROPEAN HYDRONYMS

Within Indo-European linguistics, the tradition of Old European onomastics has largely died out.⁸ The last German-speaking scholar who can be described as an Indo-Europeanist and who studied this material in greater depth was Wolfgang P. Schmid (1929–2010), Udolph’s teacher, who died recently. The fact that, apart from him, scarcely any renowned Indo-Europeanist has engaged with these hydronyms in the last two or three decades is due to two factors:

“First, the material in question is preserved only very fragmentarily. Although the location of its origin is clearly established—the majority of bodies of water, including rivers, are still found today where they were 2,000, 3,000, or 4,000 years ago—the actual age of this linguistic material remains ultimately a matter of speculation. Usually, all one can say is that certain river names are pre-Germanic, pre-Celtic, pre-Baltic, etc.—and no more. However, in many cases and in many areas, because of the discrepancy between the research and naming methods of linguistics and archaeology, it is still not clear exactly when, for example, speakers of Celtic or Germanic arrived at a particular river. It therefore also remains unclear when the pre-Germanic, pre-Celtic, etc. period actually ended—that is, at what point one may expect a pre-Celtic, pre-Germanic name to have been incorporated into Celtic, Germanic, and so on. And let me say at once: I have no answer to this either, and I do not believe one will be found anytime soon.”

Most Indo-Europeanists therefore prefer to work with material from languages for which texts have also been preserved (at present, Anatolian and Tocharian languages are particularly en vogue), whose speakers can be discussed and which can also be dated. Research into the historical development of a language and—something that is also very important in Indo-European linguistics at present—its syntax can be conducted only on the basis of such material.

And—if one is honest—there is another problem: the linguistic material attested in Old European hydronymy often consists of

⁸This is shown by a cursory survey of the relevant specialist journals in Indo-European linguistics (Historische Sprachforschung, Indogermanische Forschungen, International Journal of Diachronic Linguistics, Journal of Indo-European Studies, Kratylos, Münchener Studien zur Sprachwissenschaft), as well as the proceedings of the last approximately 15 years of workshops and conferences of the Indo-European Society: apart from occasional works by Wolfgang P. Schmid and Jurgen Udolph, there is practically nothing on this subject. And the works of these authors, as well as Krahe’s, are likewise cited by other authors only in isolated cases.

12

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

words or names that are very short. Since all further linguistic material (especially inflected word forms) is lacking, one must ultimately admit—and I say this as a convinced Indo-Europeanist—that for many simple river names, such as Sala (> Saale), Sara (> Saar), Savus (→ Save), etc., one cannot actually prove even strictu sensu that they are of Indo-European origin. Their structure is too simple for that. At the same time, however, most researchers have considered these words Indo-European since the beginning of their study and have consequently proposed Indo-European etymologies for them. But if one assumes what one actually wants or needs to prove, one is dealing with a classic circular argument. One ought to proceed much more cautiously and thoroughly than has been the case in recent decades—and to put many of the claims made in the subjunctive.

The second reason for the waning of interest in this material on the part of Indo-European linguistics is that the aforementioned scholars⁹ have practiced and continue to practice, in their ‘Old European’ works, a kind of Indo-European linguistics that has been obsolete within Indo-European studies itself for several decades and must therefore be regarded as outdated from that perspective (about 80–90% of Indo-Europeanists overall and at least 99% of Indo-Europeanists under fifty would probably share this view with me). This, in turn, means that the Old European material, presented in the same unchanged form since Krahe’s time (that is, for more than five decades), is unlikely to hold any appeal for younger Indo-Europeanists in particular—the very people most likely to be called upon to bring new life to the subject—because of its old-fashioned character.

This does not, to begin with, diminish the value of the research findings of the scholars mentioned and their colleagues working in the same field. They certainly accomplished something remarkable, but scholarship moves forward—and this has been true of Indo-European studies in particular in recent decades. If Old European studies is once again to become interesting to Indo-European studies, or to Indo-Europeanists, and to receive their attention, it must try to meet the requirements that Indo-European linguistics today applies to etymologies with regard to historical phonology and word formation. To put it bluntly: of course, one can still heat homes with wood-burning stoves and light them with kerosene lamps—but must one do so now, without necessity (or an excessive inclination toward romanticism), when central heating and light bulbs are available?”¹⁰

⁹ In addition to Udolph and Schmid, Hans Krahe (1898–1965), the discoverer of Old European hydronymy and Schmid’s teacher, should also be mentioned.

¹⁰ Bichlmeier 2012b: 14f. – Dieses längere als solches auch gekennzeichnete Selbstzitat sei mir an dieser Stelle in einer Erwiderung auf Udolph, den Meister des nicht gekennzeichneten Selbstzitats (vgl. Bichlmeier 2011d: 109f.; 2012a: 278) erlaubt.

Straipsniai / Articles

13

HARALD BICH LMEIER

3. COMMENTS ON SELECTED PASSAGES OF UDOLPH’S CRITIQUE OR POLEMIC¹¹

Already in the introductory remarks to his critique, Udolph begins by misrepresenting the facts: the article in question (Bichlmeier 2010a) does not criticize “the current state of the conception of Alteuropäische Hydronymie as outdated,” but primarily the way etymology is practiced within the framework of research on Alteuropäische Hydronymie.

3.1. On laryngeal theory and related matters

To take up straight away a phenomenon that Udolph regularly plays down, let us begin with a subject from the field of phonology, namely the phoneme class of the laryngeals.¹² That the reconstruction of laryngeals as consonantal phonemes is regularly interpreted by non-specialists—and Udolph is to be classified as such—as hocus-pocus or the like (or even disparaged), is primarily due to these non-specialists’ inadequate knowledge of Indo-European studies and their unwillingness to engage properly with the material—which, admittedly, is naturally somewhat more extensive and complex than that of an individual philology or the like—and the methodology of Indo-European studies

¹¹ I thank my colleague Sergio Neri for several comments and additions to this and the following sections.

¹² As far as can be said after 200 years of research, Proto-Indo-European had the following phoneme system:

Tenuis (T) | Media (D) | Media asp. (Dh) | Spirant (N) | Nasal labial: p | b | bʰ | – | m [m, m̥] dental: t | d | dʰ | s [s, z] | n [n, n̥] palatal: ḱ | ǵ | ǵʰ | h₁ | – tektal { velar: k | g | gʰ | h₂ | – labiovelar: kʷ | gʷ | gʷʰ | h₃ | – Liquid(en) (R): r [r, r̥] | l [l, l̥] Vokale (V): i [i, i̥] | u [u, u̥] (perhaps also [ī, ū] in monosyllabic words) e | ē o | ō a | (ā)

The class of laryngeals is also an integral part of this phoneme system; they may have been the fricatives of the three dorsal series, just as /s/ was the fricative of the dental series.

14

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

to familiarize themselves with it. At the same time, however, such outsiders to the field seem again and again to think that they can pass judgment on a discipline and its methods (here, specifically, Indo-European studies), which they do not actually know properly.¹³

In principle, laryngeals should be reconstructed only where they are required by the comparative linguistic evidence to explain the attested language material and the interconnections among evidence from different branches.¹⁴ Generalizations are, of course, then also made on the basis of these established findings. At first, these may go beyond what is warranted, but sooner or later they are usually brought back into line through discussion within the field. The fact that outsiders to the field such as Udolph (2011: 168) “cannot help feeling that laryngeals are introduced whenever it suits” once again clearly demonstrates his unwillingness to accept even research findings that have been established for years.

In any case, this phoneme class is an integral part of the phonological and morphological system of Proto-Indo-European and is indispensable to understanding it. There are a whole range of phenomena in the ancient Indo-European languages that can be explained only by assuming the former existence of this clearly consonantal phoneme class. These include, for example, the likewise consonantal phoneme /h/ <h-, -ḫ-, -ḫh-> in Hittite, or more generally Anatolian, which could not be derived from the vocalic (!) *ə often posited according to an outdated view;¹⁵ a somewhat more controversial case is whether word-initial h- in Armenian can also still be a reflex of Proto-Indo-European laryngeals; the development of the intonations (acute versus circumflex) in Baltic¹⁶ and Slavic cannot be explained coherently without positing laryngeals; the

¹³ Much the same phenomenon, albeit in more extreme form, can be seen in questions of writing and standard language, spelling reform, and so on: practically everyone thinks they can speak authoritatively on these matters.

¹⁴ Vgl. zu diesem Problem nun ausführlich Alfieri 2011.

¹⁵ If, as Udolph once mentioned in conversation, his teacher Erich Neu really did believe that nothing could be said about whether Hittite /h/ was evidence for Proto-Indo-European laryngeals, since /h/ is sometimes present and sometimes absent, this merely shows that E. Neu, too, was not exactly among the progressive linguists of his time, for the basic rules governing the continuation of laryngeals in Hittite had been known since the 1920s or 1930s at the latest. And at least in later years, E. Neu in any case acknowledged that Hittite /h/ is indeed the continuation of a laryngeal (personal communication from Elisabeth Rieken, 23 March 2012).

¹⁶ See, for example, Derksen 1996 passim; Smoczyński 2006 passim.

Straipsniai / Articles

15

HARALD BICHLMEIER

The prothetic vowels of Greek and their threefold character can likewise be most simply explained as having developed from colored anaptyctic vowels alongside original laryngeals; the same applies in that language to the threefold set of reflexes in zero-grade forms of the structure *(-)KHK(-) (outdated: *(-)KəK(-)); moreover, certain phenomena involving nasal-infix verbs in Indo-Iranian, etc., remain inexplicable without positing these consonantal phonemes. For all these phenomena, one may consult, among other things, all introductions to Indo-European studies published over the past two decades, as well as countless other works.¹⁷ In other words, anyone seeking to reconstruct Proto-Indo-European from the ancient Indo-European languages must posit laryngeals for that stage of the language as well. There is no getting by without them.

Folglich bleibt auch der Seitenhieb (vgl. Udolph 2011: 168) auf die im Reallexikon der Germanischen Altertumskunde (vgl. Zimmer 2006) gebotene keltische Etymologie zum Volksnamen der Treverer¹⁸ letztlich irrelevant: Udolph ignoriert hier ebenfalls schlicht wieder die vergleichende Evidenz der altindogermanischen Sprachen und die sich daraus ergebende Notwendigkeit des Ansatzes einer laryngalhaltigen Wurzel bzw. eines Stamms uridg. *ueh₁-r-, *uh₁-r- n. ‘Wasser’.¹⁹ Der von Zimmer a.a.O. angenommene Laryngalschwund in der Komposition, gegen den Udolph hier (wie so oft in seinen Arbeiten ohne jegliches Argument) polemisiert, ist eine in der Indogermanistik gut bekannte und allgemein anerkannte

¹⁷ Consider the good dozen introductions to Indo-European studies, laryngeal theory, or various individual languages that have appeared over the past two decades (in alphabetical order): Clackson 2007, Fortson 2004, 2009, Kapović 2008, Lindeman 1997, LIV, LIV², Mallory/Adams 2006, Matasović 1992, 2010, Mayrhofer 2005, Meier-Brügger 2000, 2002, 2010, Meiser 1998, Müller 2007, NIL, Ringe 2005, Rix 1992, Smoczyński 2006 etc. pp. – Among works published earlier, the most important include: Rix 1976, Collinge 1985, Mayrhofer 1981, 1986. – The last introduction to Indo-European studies that is almost free of laryngeals is Szemerényi 1990. – Also to be mentioned here are nearly all etymological dictionaries published over the past three decades or so (which, as is well known, Udolph generally has not consulted in his work), namely EDG, EDHIL, EDL, EDPC, EDPC Add., EDSIL, ESJS, EWAhd, EWAia, EwN, SEJL, among others. By contrast, LEV and EWD, EWdS²⁴, EWdS²⁵ remain at the scholarly level of IEW, with the regrettable result that the still incomplete EWAhd will continue for years to be the only etymological dictionary of German relevant from the perspective of Indo-European studies. A comprehensive and reliable etymological dictionary of Modern High German that reflects the current state of historical comparative linguistics (and likewise one for Middle High German) remains a desideratum!

¹⁸ For more detailed discussion of this etymology and of other proposals for deriving this ethnonym, see Sitzmann, Gründzweig 2008: 280–282. The two authors deserve the highest praise for their proper engagement with findings from Indo-European research, and should serve as an example to other non-Indo-Europeanists.

¹⁹ See NIL 715–717 for an extensive list of further literature.

16

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur

Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

phenomenon. A comparison here would be, for example, PIE *neḱo-ônḫ₁-o- > *neḱo-ôn-o- > Greek νεογνός ‘newborn’ instead of the †νεογανός (in some researchers’ view, possibly also †νεογώς) or similar that would otherwise be expected. This Greek word has given its name to the entire type.²⁰ The problem in the etymology of the ethnonym *tre-uer- ‘those who cross the water / river’ or similar actually lies elsewhere: so far, no unambiguous and generally accepted etymology has been found for Proto-Celtic *trē ‘through, across’.²¹ From an Indo-Europeanist perspective, nothing else is a problem.

Insofern ist in diesem Zusammenhang auch der Vorwurf Udolphs (2011: 165), Indogermanisten würden die Rekonstrukte für „bare Münze“ nehmen, unzutreffend: Jedem Indogermanisten ist bewusst, dass er hier mit Rekonstrukten arbeitet: Gleichzeitig dürfen sich Indogermanisten aber auch sicher sein, dass sie hiermit eben über das im Vergleich zu den veralteten Konzeptionen, mit denen Udolph et al. arbeiten, bessere Erklärungssystem verfügen.

If Udolph should succeed in all these aforementioned cases in providing a meaningful, coherent explanation that, as made possible by the laryngeal approach, accounts for all the cases mentioned and does not become entangled in casuistic minutiae of analogies that vary from one individual case to another, I would be pleased if he shared it with the scholarly community. This would certainly prompt a broader discussion and, if it were indeed a proposal supported by sound arguments, might even lead to a change of thinking in Indo-European studies.

However, after reviewing numerous of his works, one is more likely to conclude that all we can expect from him is a continued recourse to the explanatory models mentioned above, which have been outdated for decades. He would therefore do both himself and scholarship a great service by refraining from making such statements on this subject.

Closely connected with the fundamental need, just discussed, to reconstruct the laryngeal phoneme class is the hypothesis already proposed by Émile Benveniste in the 1930s, that the minimal structure of at least Proto-Indo-European nominal and verbal roots was KVK-. Since, in word-initial position before a vowel, the laryngeals disappear in Anatolian and, where applicable, Armenian, except in the cases mentioned above

²⁰ Formations with the second element *-dʰh₁-o- > *-dʰo- from the PIE root *dʰeh₁- ‘put, place, lay’ have by now become quite numerous among those identified as obscured compounds; for this type of composition, see Bichlmeier (2011d: 120–123) for a more detailed discussion and relevant literature, and Hackstein 2002 for a foundational treatment of the entire complex.

²¹ Cf. EDPC 388, with discussion and further literature.

Straipsniai / Articles

17

HARALD BICHLMEIER

that is, in Anatolian and, where applicable, Armenian—indeed, the cases that provide clear proof are precisely those in which Hittite/Anatolian shows a reflex of a laryngeal, while the other languages have a word with vowel-initial onset—there is nothing to preclude the general assumption that, when no corresponding etymon is attested in Hittite, or in cases where, as with PIE *#h_₁V- etc., loss of the laryngeal must be assumed in Hittite/Anatolian, an initial laryngeal may nevertheless be posited in these cases. All attempts to prove that words or roots with vowel-initial forms in the individual languages must already have had vowel-initial forms in Proto-Indo-European (and not a laryngeal) are to be considered failures. Positing root-initial, pre-vocalic laryngeals is once again the more economical and more probable assumption in this case. Consequently, it remains entirely unproblematic to work with Benveniste’s hypothesis concerning Proto-Indo-European root structure unless the contrary is proven.

If one reasonably accepts the research findings of Indo-European studies over the past good hundred years set out above, it becomes clear that, from the perspective of contemporary Indo-European studies, a root analysis “*el-/ol-,” with which Udolph and others continue to work (see further discussion below), can of course no longer be considered in keeping with the state of the art and should therefore finally be abandoned.

3.2. On the Sonorant Liquids in Proto-Indo-European

Indo-Europeanists are accustomed to the fact that otherwise open-minded scholars from neighboring disciplines—as just demonstrated—also have problems with the reconstruction of the laryngeals, even though their reflexes (as already mentioned above) are quite tangible as consonantal sounds in Anatolian. However, another statement by Udolph (2011: 165) finally does raise doubts about his expertise in historical linguistics and linguistics in general: he even questions the existence of r sonans! Given that this phoneme occurs, among other places, in Old Indic (where there is even a distinct grapheme for this sound), in two EU state and standard languages (Czech and Slovak), and in other neighboring EU languages (Croatian, Serbian, Bosnian, Montenegrin)—all Slavic languages that should be familiar to a Slavist, as Udolph also is—and that even German has this sound in its dialects (e.g. Alemannic) as a

18

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

phoneme, so that this is a quite well-attested phenomenon, this skepticism is utterly incomprehensible.

And since Udolph has also always worked with ablaut phenomena (so far) in his works, or at least mentioned them (although, of course, he has never addressed the semantic content conveyed by the different ablaut grades, and, judging by his statements, evidently continues to assume that ablaut consists ultimately of freely available variants of roots and suffixes), and consequently regularly posits zero grades (as here, 2011: 172, where he posits *l̥- as the zero-grade form of his “*el-/ol-,” to which we will return below), the question naturally arises as to what else should result in the zero grade of a root of the structure *KerK-, if not *Kr̥K-. Here, too, one would be keen to hear an answer.²² In any case, a system in which /r̥/ does not occur but /l̥/ does (which is what would follow if one actually took Udolph’s remarks in his article seriously) is also highly improbable from a typological perspective.

3.3. On Lithuanian aléti and Related Forms: Further Remarks on the Outdated Approach “*el-/ol-”

Udolph’s repeatedly advanced thesis (here p. 164) that “*el-/ol-” is continued in Lithuanian in aléti ‘to flow, drip; to run, hurry’, nualéti ‘to overflow; to inundate (fields)’ (although the simplex never seems to be said of rivers, and, interestingly, the LKŽ cites as the sole attestation of the simplex a sentence in which it occurs alongside alùs ‘beer’,²³ which is probably a

²² Perhaps the solution to the problem lies outside the strictly linguistic sphere: to the best of my knowledge, Udolph comes from a German dialect area that has not the apical r (a tongue-tip r), but the uvular r (a guttural r). A guttural syllabic r is an extremely rare phenomenon and is indeed difficult to articulate; to the best of my knowledge, it does not occur in German dialects either. Against this dialectal background, the apical r of the languages mentioned above, which can be pronounced syllabically without difficulty, may of course seem very foreign, something that “simply cannot exist.” But here we enter the realm of speculation—and we would naturally prefer the explanation urged above in the text to such speculation.

²³ The LKŽ entry is quoted here: 2 aléti, ãli, -ėjo intr. 1. tekėti, varvėti: Alė́te išalė́jo alus, t. y. išėjo, nēr alaus J. 2. bėgti, lėkti: Ji tik iš vieno kiemo į kitą ãli ãli Lnk. 2 išalėti, išali, -ėjo intr. J ištekėti, išvarvėti. 2 nualéti, nùali, -ėjo 1. tr. nutelkšoti, prigirdyti: Vanduo nualėjo žemesnius laukus Grž. 2. intr. nugrimzti: Numarmėjo, nualėjo Sv. (www.lkz.lt on 15.03.2012)

Straipsniai / Articles

19

HARALD BICHLMEIER

A loanword from Germanic is [cf. SE]L 12; IEW 33f., by contrast, treats it as an inherited word ²⁴], ²⁵ but has at least a homophonous root), raises a problem: SE]L 9f. assigns the verb to a Proto-Indo-European root *h₂al- [< *h₂el-]²⁶; according to SE]L, however, this is merely a transponat, and “no secure connections have so far been found.” In other words, the author of SE]L, Wojciech Smoczyński, a renowned expert on the Baltic languages, especially Lithuanian and Old Prussian, evidently thinks little of the connection preferred by Udolph.²⁷ A Proto-Indo-European root *h₂al- can be reconciled with the outdated Udolphian approach “*el-/ol-” only in cases where a language-specific merger of late Proto-Indo-European *o and *a would have led precisely to the indistinguishability of continuants of a Proto-Indo-European root *h₂al- and a Proto-Indo-European root *h₁el(H)- in the o-grade (that is, Proto-Indo-European *h₁ol(H)-). If they had indeed become indistinguishable in this way, how could one still say with certainty which had originally been present?

LitEW I, 8, in turn, does not cite the verb at all, only the derivatives presumably related to it (at most indirectly), almēs ‘fluid discharged from the body, blood serum’, almuō ‘pus’, etc. It still assigns these, in the manner customary at the time, to “*el-/ol-” (without providing information on word formation), and goes on to offer comparisons with other languages that are partly incorrect and partly at least outdated or in need of modification. If some of the connections proposed in the cited work

²⁴ Since a root reconstruction *alu-, as in the IEW, is of course no longer possible, one may (if the word is not to be assigned to one of the other roots discussed above) abstract yet another root from this word, *(H)al- ‘bitter’ or the like. Since a root with this meaning could readily occur in names of bodies of water as well, it should actually also be included in the discussion. This will not be pursued further, however, so as not to complicate matters still further. In any case, this root would further reduce the likelihood that all the corresponding names of bodies of water mentioned in the literature are to be assigned to “*el-/ol- ‘flow’.”

²⁵ Das SE]L (samt Smoczyński 2008, 2009) ist gegenwärtig das einzig verfügbare etymologische Wörterbuch zu einer baltischen Sprache auf dem aktuellen Stand der Indogermanistik. Ein weiteres etymologisches Wörterbuch des Litauischen ist von Rick Derksen in der Leidener Indo-European Etymological Dictionary Series angekündigt; und in Berlin entsteht gerade das Altlitauische Etymologische Wörterbuch (ALEW), das 2013/14 publiziert werden sollte.

²⁶ It should nota bene be noted here that Smoczyński should not be assumed to posit a root-level *-a-; rather, he evidently belongs to those Indo-Europeanists who date the coloring of *e next to *h₂ so early that they regard it as already Proto-Indo-European, and therefore also indicate this coloring in their reconstructed forms. He is thus simply using a somewhat different notational convention.

²⁷ Moreover, given such a situation—and assuming a penchant for hair-splitting—one may also ask (at least as a thought experiment) whether, in the end, a new Baltic root *al- ‘liquid’ or the like might have been abstracted from the Proto-Baltic loanword *alu- ‘beer’ (interpreted as ‘liquid, drink’).

20

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

were nevertheless correct (for example, those involving the Germanic words cited above or Latin alga ‘sea grass, seaweed’ [cf. EDL 33]), this would speak even more strongly against deriving it from “*el-/ol-.”

As a brief aside, it should be noted here that the IEW, which Udolph otherwise cites as virtually his sole etymological reference work of Indo-Europeanist character, has no entry *el-/ol- ‘flow’: it lists only “1. el-, ol-, _əl-, ‘color root meaning “red, brown”’ (302–304) [modern: *h₁el-], “2. el- ‘rest’??” (304f.), “3. el-: ol- ‘be mouldy, rot’” (305) [modern: *h₁el(H)-], “4. el-, ol- sound root” (306), “5. el-, ol- ‘destroy, ruin’??” (306) [modern: *h₃elh₁-]²⁸, “6. el-, elə-: lā-; el-eu-(dh-) ‘drive, set in motion, move, go’ (306f.) [modern: *h₁elh₂-], “7. el-, elə-, with -k- extension elk-, elək- ‘hungry, bad’ (?)” (307), 8. el-, elēi-, lēi- ‘bend’” (307–309). In other words, even the IEW, by whose time a number of Krahe’s works had already appeared, does not seem to have regarded this meaning as primary for the aforementioned no. 6, which is presumably the root Udolph also has in mind.²⁹

In this connection, Udolph’s evident unfamiliarity with more recent etymological works than the IEW again proves fundamentally problematic (his citation of LIV or LIV² in the work discussed here is probably due more to necessity than to deeper insight). In our context, for example, he cites not a single work on the Baltic languages. Among the other comparanda he adduces, Latvian alots ‘spring’ in particular was already viewed with some misgivings in LitEW I, 8, where it is considered whether it might be a dialectal hypercorrection of Latvian avots ‘the same’. The absence of an entry alots in the LEV, which lists only avots (LEV 1, 93f.), ultimately points in the same direction, although, of course, as an argumentum e silentio it is not conclusive. Relevant discussion for our purposes is found, however, in LEV 1, 65 s.v. ala ‘depression, narrowness, spring’, for which borrowing from Middle Low German hol was also considered. Latvian alogs ‘spring’ would then also be related (< Proto-Baltic *alanga-).

Some problems concerning the Lithuanian ālmes ‘fluid discharged from the body, blood serum’ cited above cannot be resolved conclusively. Here, in my opinion,

²⁸ Cf. LIV² 298.

²⁹ Ebenfalls in einer Fußnote sei ein Kommentar zu Udolphs Behauptung (2011: 166) versteckt, dass ich „überzeugendere Versuche“ zur Rekonstruktion der o.g. Wurzel(n) „in der Fußnote versteckt“ habe (in welcher wird von Udolph a.a.O. nicht gesagt, der Nachweis sei hier für den interessierten Leser nachgetragen: Bichlmeier 2010a: 17, Anm. 31). Auch hier zeigt sich wieder, dass Udolph hinsichtlich der Indogermanistik eigentlich eher ahnungslos ist: Die dort angeführten alternativen Vorschläge dürfen als schlicht überholt gelten und wurden nur der bibliographischen Vollständigkeit halber angeführt: Statt *h₁el- bzw. *el₂- wird nun eben *h₁elh₂- rekonstruiert.

Straipsniai / Articles

21

HARALD BICHLMEIER

It is highly questionable whether the root was “extended by -m-” [as Udolph claims], or whether one should rather assume suffixation: for if there had been an already Proto-Baltic or even earlier ‘extension’ in Udolph’s sense, Lithuanian ālmės could no longer be directly compared with Lithuanian alėti, since there would then be two different (Proto-Baltic) roots. To do so nonetheless would be methodologically unsound.

A further consideration should be added here: if one is already working with root extensions, and thus recognizes them as a genuine possibility in the word formation of an early stage of Proto-Indo-European (the author is generally initially critical of this possibility, whereas other scholars are more open to it), one could also regard Proto-Indo-European *_h₂elh₂- and *_h₂elm- as two extensions of an original root *_h₂el-. If this possibility of etymologizing is accepted, however, the Lithuanian words with and without -m- would again not be directly comparable. ³⁰

Falls nun lit. ālmės seinen Akzent ererbt haben und nicht der métatonie douce verdanken sollte, die bei solchen femininen Ableitungen regelmäßig auftritt (anzusetzen ist urbalt. *ālm’ij̄ā-), würde dies sogar darauf hindeuten, dass die Bildung von einer Wurzel ohne auslautenden Laryngal gebildet worden ist. Es käme hier also wieder der Ansatz *_h₂al- [< *_h₂el-] des SEJL ins Spiel. Angesichts der regelmäßig gerade in Feminina auf urbalt. *-ij̄ā- > lit. -e auftretenden Metatonie³¹ ist das Wort aber letztlich hinsichtlich dieser Frage nicht aussagekräftig. Eine Verknüpfung mit *_h₂elh₂- ‘ziellos gehen’ (das sich problemlos semantisch zu ‘nicht zielgerichtet [aus dem Körper] austreten’ weiterentwickelt haben kann) erscheint somit möglich:³² vorurbalt./uridg. *_h₂elh₂-m-ij̄eh₂- > urbalt. I *ālm’ij̄ā- > urbalt. II *’ālmij̄ā- > ālmė-s ist dann völlig regulär. Und falls nicht schon ein Suffix urbalt. *-mij̄ā- abstrahiert worden ist, ist die Bildung als feminine Ableitung zu urbalt. *al-ma- einzustufen, wobei aber immer noch der semantische Gehalt der ma-Ableitung zu klären bliebe.

All the explanations just given are subject to one qualification: neither LitEW nor, in particular, Udolph indicate which accent class the Lithuanian lexemes mentioned above belong to. Moreover, both overlook the fact that alongside Lithuanian ālmės (4) and almuõ (3)—that is, accusative singular álmenį—there is also álmė (1) (see LKŽ s.vv.).

³⁰ This may, however, simply be another example of Udolph’s occasionally imprecise use of terminology.

³¹ Vgl. dazu Derksen 1996: 55–66. Zur Metatonie im weiteren Kontext der baltischen Wortbildung vgl. Larsson 2003: 33–86; Mikulėnienė 2005 passim.

³² Since the SEJL gives no meaning for *_h₂al-, it cannot be ruled out—and may even be likely—that this is in any case the same root reconstructed in LIV, LIV² as *_h₂elh₂-, except that Smoczyński assumes it had no word-final laryngeal.

22

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteruropäischen Hydronymie – eine Erwiderung auf eine Polemik

Both almuõ (3) and álmē (1) thus have an acute root syllable, which must be considered primary in contrast to the circumflex root syllable in ālmes (4). In the generally accepted view, this acute can be explained only if the formations are based on a root ultimae laryngalis. Put differently, in our context the primary formations are, in any case, those that do not show metatony.

It must then also be pointed out that the meanings of almuõ (3), ālmes (4), and álmē (1) could also quite well allow for a connection with the Proto-Indo-European root *h₁el(H)- ‘to be rotten, decay’ (Pokorny’s no. 3 in the list above). This root is likely also present in other Baltic lexemes, for example Latvian al(u) ksna ‘marshy, boggy place (< Proto-Indo-European *h₂ol(H)-s-neh₂)³³, an adjectival derivative of the Proto-Indo-European s-stem *(s)h₁el(H)-os-, continued, for instance, in Greek ἕλος n. ‘marsh’³⁴, Sanskrit sáras- n. ‘lake, pond’³⁵, etc. This etymology also makes it possible to connect New High German dialectal Eller ‘land not used for agriculture, wasteland’ (from an original meaning ‘marshy land’) with this word family.³⁶

In view of these arguments, another explanation for aléti may be put forward for discussion: if a Proto-Baltic root *al- meaning ‘to be marshy, to be musty’ must be assumed in Baltic, the semantic development of Lithuanian almuõ (3), ālmes (4), and álmē (1) could have taken a quite different course: ‘musty, rotten liquid’ > ‘fluid issuing from the body, blood serum, pus’. In this connection, aléti may originally have meant something like ‘to be marshy’ and only secondarily, following the nouns, acquired the meanings ‘to flow, drip’, and then also ‘to run, hurry’. Nor can it be ruled out here that two original roots and the verbs derived from them merged in aléti as a result of their homophony.

The formation of a secondary Proto-Baltic root *alm- in the sense given above cannot, however, be definitively ruled out; given the derivation in Proto-Baltic *-ijā-, it may ultimately even be more plausible (see above). It remains problematic that in the appellative vocabulary the root occurs exclusively in the form Baltic al-.

The Baltic water names with initial El- cited in the Etymologica referred to above and elsewhere as etymologically related cannot for the time being be regarded as self-evidently relevant

³³ The -k- in Latvian al(u)ksna results from a pan-Baltic tendency to insert velars before sibilants and is therefore secondary.

³⁴ Cf. EDG 415, which offers a different derivation following the traditional pattern. The solution proposed above, with the assumption of an s-mobile, is phonologically impeccable and is supported by comparable cases.

³⁵ Cf. EWAia 2, 708. The conjecture expressed there that this word is unrelated to the Sanskrit root sar- ‘to run’ may now be regarded as established fact.

³⁶ Vgl. zum Ganzen ausführlich Neri, Ziegler 2012: 51–56.

Straipsniai / Articles

23

HARALD BICHlMEIER

zugehörig und aussagekräftig gelten, ³⁷ da sie 1) ihren Vokal dialektaler Herkunft verdanken könnten (dialektale Herkunft des E- müsste also zunächst definitiv ausgeschlossen werden) und – was noch viel schwerer wiegt – 2) schlicht nicht zu beweisen ist, dass es sich immer um dieselbe Wurzel handelt: Neben der Wurzel „*el-/ol-“ (bzw. uridg. *h₁elh₂- und *h₂elh₂) kommen auch noch die Farbwupzeln uridg. *h₁el- ‘rot(braun)’, *h₁₃al- bzw. *h₂al- [< *h₂el-] ‘weiß’³⁸ und uridg. *h₁el(H)- ‘modrig sein, faulen’ in Frage – und das gilt nicht nur für die baltischen, sondern natürlich auch für die anderen Flussnamen, die traditionell zu dieser Wurzel gestellt werden, etwa auch für die schon in den von Udolph (hier 2011: 163) gerne zitierten und wieder abgedruckten Kraheschen Tabellen³⁹ angeführten: In dieser Tabelle findet sich keine einzige Wortform, bei der man aufgrund der enthaltenen Suffixe zwangsläufig damit rechnen müsste, dass die jeweilige Wurzel verbalen Charakters gewesen sein muss. Mit anderen Worten: Nimmt man die Mitteilungen der (neueren) Etymologica ernst und bezieht die Existenz der o.g. Farbwupzeln mit ein, wird man bei den wenigsten der hier interessierenden Gewässernamen definitiv sagen können, zu welcher der genannten Wurzeln sie nun gehören – und das gilt natürlich auch für die von Udolph (2011: 167) angesprochenen Gewässernamen im LVŽ (1, 62): Man kann sie weiterhin im Brustton der Überzeugung an „*el-/ol-“ ‘fließen’ anschließen (was in diesem Falle dann aber eine reine Glaubenssache wird, die dann mit Wissenschaft nur mehr sehr wenig zu tun hat), oder man kann der Realität ins Auge blicken und zugeben, dass man neben uridg. *h₁elh₂- ‘wohin treiben’ (LIV² 235) und *h₂elh₂- ‘ziellos gehen’ (LIV² 264) in den Flussnamen auch mit den beiden o.g. Farbwupzeln mit der Bedeutung ‘weiß’ und ‘rot(braun)’⁴⁰ sowie *h₁el(H)- ‘modrig sein, faulen’ rechnen muss.

In light of what has been said, the most likely basis for the Baltic words (unless, indeed, a new root was abstracted from a loanword, or a merger occurred with a verb from the root *h₁el(H)- ‘be musty, rot’) is perhaps PIE *h₂elh₂- ‘go aimlessly’. If this conclusion proves well founded, LIV² should be supplemented in the cited passage with Lithuanian aléti.

³⁷ The same ultimately applies to Lithuanian almēs = ēlmēs.

³⁸ Vgl. dazu ausführlich Bichlmeier 2012c.

³⁹ So etwa Krahe 1964: Tabelle I und II nach S. 62. – In diesem Zusammenhang ist auf eine weitere methodische Unsauberkeit Krahes hinzuweisen, die Udolph durch die Reproduktion jener Tabelle ebenfalls weitertradierte: Zumindest der dort angeführte Flussname „Elira“ ist nirgends belegt, sondern nur Rekonstrukt (als solches richtig gekennzeichnet Krahe 1964: 37) – wird aber in den Tabellen nie als solches gekennzeichnet. Siehe zu diesem Rekonstrukt weiter unten im Text.

⁴⁰ As is well known, this root is indeed continued in Baltic, for example in Lithuanian élnis ‘deer’, etc.; cf. SEJL 146; PKEŽ 1, 68–70; PJ 1, 77.

24

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

With regard to hydronyms in the Baltic region and elsewhere, the general rule is that only names beginning with El- (whose initial vowel does not result from umlaut [as, for example, in Germanic] or dialectal variation [as, for example, in Baltic]) must belong either to PIE *h₁el- ‘red (brown)’ or *h₁elh₂- ‘drift away’; names beginning with Al- may ultimately be formed from any of the roots mentioned (not least because of the widespread merger of (late) PIE *a and *o as *a): they may go back to PIE *h₁/₃al- or *h₂al- ‘white’, *h₂elh₂- ‘go aimlessly’, the o-grade forms of PIE *h₁el- ‘red (brown)’ or *h₁elh₂- ‘drift away’, or *h₁el(H)- ‘be musty, rot’, that is, *h₁ol-, *h₁olh₂-, or *h₁ol(H)-, as well as, of course, *h₂olh₂- (though, to my knowledge, there is no secure evidence⁴¹ for the o-grade in the roots *h₁olh₂- and *h₂olh₂-). A decision is possible here only in individual cases where it can be made reasonably probable that a particular suffix can be combined only with a particular type of root (verbal or adjectival). However, neither Krahe, Schmid, nor Udolph has yet offered analyses of this kind. If such an assignment cannot be made, no decision can responsibly be reached. Ultimately, then, this is yet another example of how ignorance of alternatives—arising from the systematic neglect of research findings—can lead to supposed certainty in etymological solutions.⁴²

In light of these arguments, it should have become clear that it is simply no longer possible to assign all the river names cited by Udolph and by Krahe⁴³ in the relevant context unequivocally to a root “*el-/ol- ‘flow’.” Anyone who continues to do so is making a mere assertion unsupported by any evidence—and the fact that it is merely an assertion is no different even if it is made by the “Lord of Names.”

4. ON THE HYDRONYM Ulster

At this point, let me also add a few remarks on Udolph’s discussion of the hydronym Ulster: contrary to Udolph’s (2011: 172) claim, this river name is entirely irrelevant to our discussion, although

⁴¹ Both Old Prussian alne ‘bison’ and Lithuanian dialectal álnis, álnias ‘deer’, álnė ‘hind’, and Latvian álnis are more indicative of dialectal opening of the initial vowel than of an old o-grade (cf. previous note).

⁴² In my view, the same problem of the impossibility of determining the root precisely also arises in the etymology of the name of the Elbe; cf. Bichlmeier 2012c.

⁴³ Cf. Krahe 1954: 49, 58.

Straipsniai / Articles

25

HARALD BICHLMEIER

There would naturally be an etymological relationship insofar as one of the five roots presented above is likely to occur in these names as well.⁴⁴

In diesen Zusammenhang gehört auch Udolphs Satz (2011: 166): „Zudem ist eine für Gewässernamen überzeugendere Lösung mit einem Ansatz ohne anlautenden Laryngal deshalb vorzuziehen, weil dieser zu sämtlichen europäischen Flussnamen passt.“ Abgesehen davon, dass man Donau auch dann nicht aus „*el-ol-“ erklären könnte – kleiner Scherz am Rande –, sollten die vorangehenden Ausführungen (besonders die oben unter 3.1 zur Wurzelstruktur des Urindogermanischen) klar gemacht haben, dass das natürlich mit einem Ansatz, der dem heutigen wissenschaftlichen Standard entsprechend mit Laryngalen operiert, selbstverständlich auch alles geht – nur die (vermeintliche) Eindeutigkeit der alten Erklärung muss man eben aufgeben. Denn jedem, der ein wenig Ahnung von Indogermanistik hat, ist selbstverständlich, dass eine Form uridg./vorgerm. *H̥(H)-s(t)-ro- (mit dieser Notation, in der *H als Coversymbol für *h₁, *h₂, *h₃ gilt, ist jede der oben erwähnten Wurzeln abgedeckt) ohnehin ebenfalls urgerm. *ulstra- ergeben würde. Auch hier zeigen sich somit wieder Udolphs grundlegende Kenntnislücken auf dem Gebiet der modernen historisch-vergleichenden Sprachwissenschaft bzw. Indogermanistik. Folglich spricht also auch der Flussname Ulster weder für noch gegen eine der oben angeführten Wurzeln, Udolphs Einwand läuft völlig ins Leere. Es hat etwas mit wissenschaftlicher Ehrlichkeit zu tun, dass man derartiges auch irgendwann zugibt und nicht trotz anderer, neuerer Erkenntnisse, die solches in Frage stellen, so tut, als habe man schon alles erklärt.

Moreover, attention should be drawn to a further problem raised by river-name groups (if the names under consideration here may be described as such) of the type Elster, Alster, Ulster, etc., on which Udolph, to my knowledge, has never taken a position, let alone offered an answer: How can it be that before one and the same suffix the root could occur once in the zero grade, once in the e-grade, and once in the o-grade? All the more so since the suffix is ultimately thematic, and thematic formations, according to the general view, no longer showed ablaut already in Proto-Indo-European. A sensible and coherent solution to this problem, going beyond morpheme analysis—that is, the first step in an etymology—would also have to be presented.

River-name groups of the kind mentioned above cannot, in any case, derive from a common paradigm. The names must therefore have arisen individually and must likewise be explained individually. Consequently, the statement that

⁴⁴ Udolph’s remarks (2011: 170–172) on earlier attempts to etymologically connect water names with the Proto-Indo-European root *h₂el- ‘nourish, raise’ are ultimately just as irrelevant in this context. I have never seriously considered such a connection either, although Udolph’s remarks might perhaps suggest otherwise to an inattentive reader.

26

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

such forms stand “in ablaut with one another” (as is sometimes claimed) is false. Rather, only the root elements used to form the various river names occur in different ablaut grades. This is an entirely different situation. As to the etymological (including semantic) relationship among the names, all this tells us is that a different form of the same root was used in each case to form a particular river name, possibly—but also possibly at entirely different times.

5. FIRST INTERIM CONCLUSION

The foregoing considerations concerning the river name Ulster once again make abundantly clear that Udolph is unwilling to familiarize himself with new⁴⁵ ways of thinking and adopt them—even though they have long since proved their worth: judging by his statements on the subject, he has still not understood Laryngeal Theory. Anyone who wishes to criticize something persistently and in a manner that merits being heard—or who does criticize it, as Udolph does Laryngeal Theory—should first understand it and be able to follow the lines of reasoning under criticism. If one cannot do so, however, all criticism necessarily remains empty talk.

Had Udolph understood Laryngeal Theory and its implications, he would long ago have realized that several proposals concerning various river names, including those I have put forward, were primarily intended to bring his (and other researchers’) entirely outdated and obsolete account up to a standard regarded by Indo-Europeanists today as the norm. In this respect, some further remarks by Udolph (2011: 167) also backfire, especially those concerning Indo-Europeanists’ lack of interest in this material: Indo-Europeanists are simply not interested in the material (as already suggested above in section 2) because Udolph et al. have so far treated it inadequately, because the material has hitherto been handled using such deficient methodology, and because—let this be emphasized clearly here—in the end the gain in knowledge for the core domain of Indo-European studies (namely, the reconstruction of the phonology and morphology of the Proto-Indo-European parent language) is quite insignificant. Matters may be different for Indo-European antiquity and settlement history, provided that results become available on which one can then rely.

⁴⁵ And ‘new’ here means nota bene an explanatory model whose beginnings go back to the nineteenth century and which must be regarded as having become definitively established over the past three to four decades.

Straipsniai / Articles

27

HARALD BICH LMEIER

6. EXCURSUS ON THE SUFFIX URG. *-stra-

At this point, a few further remarks may be added on the Germanic -stra-suffix (already entirely correctly so in Krahe/Meid 1969: 184; in the outdated terminology used by Udolph [1994: 243ff. etc.], usually “str-suffix”—after all, the thematic vowel is part of the suffix, is it not?): despite an extended search, I could find nowhere in Udolph’s works (not even in the otherwise relevant section in Udolph 1994: 243–258) an etymological explanation of the suffix. He refers only to the aforementioned passage in Krahe/Meid 1969. Udolph’s explanations here follow Krahe/Meid loc. cit., although it remains entirely unclear why he emphasizes the suffix’s occurrence after gutturals. Moreover, he seems to have overgeneralized Krahe/Meid’s statements: They speak of “designations of places” in the sense of Nomina loci (the words in question mean ‘holster’, ‘sheepfold’, etc.), not of place names in the onomastic sense. The comparison is therefore flawed unless one can plausibly account for the transfer from the appellative to the onymic domain. Among other things, it would first have to be explained how a word-formation pattern for designating concrete, man-made objects (which the suffix Proto-Indo-European *-tro-, with which, among other things, Nomina instrumenti can be formed [see below], could certainly serve) came to be transferred to ‘places’ that are not man-made, such as bodies of water. Such a transfer strikes me as anything but trivial and would in any case require explicit justification.

Krahe/Meid loc. cit. assume that, in the development of this suffix, the suffix Proto-Indo-European *-tro-, which forms Nomina agentis⁴⁶ and instrumenti, was added to roots ending in a dental stop (the roots ending in -s should of course also be included here; see below), giving rise initially by regular sound change to Proto-Germanic *-stra-, which could then be reanalyzed and separated off as an independent suffix.⁴⁷ This could eventually also be added to vowel-final stems, giving rise to the variants Proto-Germanic *-istra- and *-astra-.

However, this is not the only conceivable way for it to have developed: a secondary addition of a suffix Proto-Germanic *-ra- to stems in Proto-Germanic *-stV-⁴⁸ is also possible, as is a derivation with the same suffix from an underlying

⁴⁶ These formations could be derivations with the thematic vowel from the weak stem Proto-Indo-European *-tr- of Nomina agentis in Proto-Indo-European *-ter-/-tor-, that is, Proto-Indo-European *-tr- + *-o-. Whether such an explanation is also viable for Nomina instrumenti may initially remain an open question here.

⁴⁷ Ebenso Hill 2003: 105f.

⁴⁸ Vgl. dazu Krahe, Meid 1969: 3, 163–170. Dazu auch Hill 2003: 105ff. – Nur wenn man dieser Herleitung folgt, kann u.U. ein Zusammenhang mit den lateinischen Adjektiven des Typs silvestris ‚Wald-, waldig‘ hergestellt werden, wie ihn Udolph in einem Vortrag in Leipzig 2008 oder 2009 einmal po-

28

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteruropäischen Hydronymie – eine Erwiderung auf eine Polemik

lying s-stem. In the latter case, when the r-containing suffix was added to the -s- of the base word, an epenthetic dental would arise between the two, likewise yielding Proto-Germanic *–stra–.⁴⁹ The latter derivation would have the advantage—even given the different ablaut patterns of s-stems attested in Indo-European—of allowing the suffix variants explained above as purely secondary to be derived by regular sound change: Thus, for example, with a proterokinetic s-stem, Proto-Indo-European Nom. *ḰéK-os, Gen. *KK-és-s → *K(e)K-és-os, assuming that secondary derivations were usually made from the weak stem, would automatically yield *K(e)K-es-ro- > Proto-Germanic *K(e)K-is-t-ra-. Similarly, the suffix variants *–stra– (< *–s–(t)–ro–) and *–astra– (< *–os–(t)–ro–) could then be derived either from s-stems with different ablaut patterns or by assuming derivation from the strong rather than the weak stem—a process that, although not frequently attested, does nonetheless occur from time to time.

In other words, all the existing suffix variants could have originated through regular sound change. It remains to be examined whether the respective prototypes can also be reconstructed through language comparison, i.e. whether the corresponding s-stems can still be found for Germanic formations in *–(i/a)stra–—if not in Germanic, perhaps in other Indo-European languages. But that must be left to further research.

Whichever derivation one prefers, or whichever should ultimately prove to be the (only?) correct one, it should nonetheless have become clear what etymological deficiencies still need to be addressed here.

Furthermore, the preceding discussion should also have made clear that hydronyms in Proto-Germanic *–stra– do not necessarily have to share the same etymological origin as the aforementioned Nomina loci: while a derivation of these appellatives from original Nomina instrumenti in *–tro– is entirely plausible, the same cannot exactly be claimed for river names: these are more likely to have arisen either from secondarily thematized Nomina agentis in *–t(e)r– + *–o–,⁵⁰ or else are secondarily substantivized secondary adjectives in *–ro– derived from old s-stems, as illustrated at the end: in my view, the derivation of the two semantically quite distinct classes of nouns

has postulated: for the Latin suffix form can only be explained from Proto-Indo-European *-s- + *-t- + *-rV- or *-s- + *-t(e)ro- + *-i-/*-io- (cf. Leumann, Hofmann, Szantyr 1977: 351f.; Lühr[Ball es] 2008: 126f., Weiss 2009: 320); the other two derivations possible for Proto-Germanic cannot apply there because of the Italic-Latin sound laws. Whether these are individual-language innovations or perhaps a Germanic-Italic isogloss would likewise require separate investigation.

⁴⁹ Vgl. zum Lautlichen Görtzen 1998: 449–451.

⁵⁰ Cf. note 46 above.

Straipsniai / Articles

29

HARALD BICHLMEIER

(namely Nomina loci vs. river names) from originally different morphological structures, which have become homophonous as a result of Germanic sound developments, is considerably more likely than the age-old postulate that all of this had only a single point of origin. Further research is also needed here, however, in order to clarify this state of affairs as definitively as possible.

7. UDOLPH’S TREATMENT OF FINDINGS FROM INDO-EUROPEAN STUDIES: A CASE STUDY ON ABLAUT PHENOMENA

As should already have been clearly established elsewhere, the relationship between reconstructed Proto-Indo-European and Old European hydronymy is practically a one-way street:⁵¹ Only word forms, or in this case hydronyms, that correspond exactly to what is to be expected from the comparison of the attested ancient Indo-European languages with regard to phonology and morphology can initially be considered securely Indo-European and regarded as amenable to sound etymological analysis. Word forms whose suffixes or chains of suffixes cannot be attested in the appellative vocabulary of other Indo-European languages pose a problem, as do word forms whose ablaut grades cannot be explained: it is simply no longer sufficient (as it was in Krahe’s time) to state that a word has one or another ablaut grade; one must also explain (or be able to explain) why it occurs and what parallels there are.

A particularly striking example of the neglect of all findings from the comparison of the Indo-European languages is Udolph’s explanation of the river names Polish Drwęca / German Drewenz < “Druantia,” French Druance < Latin Druentia⁵², etc., a case involving one of the formations that are at times indeed problematic because they look like participles.

Following older studies, including Krahe’s (while correcting him), who posits an earlier form “*Dravantia” for this river, Udolph⁵³ derives these river names from “Druantia.” This approach reveals, first of all,⁵⁴

⁵¹ Vgl. dazu etwa auch Bichlmeier 2012d.

⁵² Cf. e.g. ACS 1, 1320f.

⁵³ Vgl. Udolph 1990: 107–112; dort findet sich auch eine Aufstellung der Belege.

⁵⁴ Cf. Krahe: 1964: 45.

30

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

once again wholly outdated conventions of notation. A better approach would be Old European *dru(u)ant(i)jā-, which, as is ultimately clear from Udolph’s further discussion, is traced back (in current notation) to Proto-Indo-European(?) *dru(u)-ont-ih₂-/-ieh₂-.

But this brings us right to the heart of the problem: according to Udolph, this is a participial form. Yet a form with the ablaut grades found here—namely, the zero grade in the root and the o-grade in the suffix, with a feminine ending—is untenable in this form. Such an analysis is impossible on the basis of comparative linguistics for this root⁵⁵: it begins with the fact that the Proto-Indo-European root *drēu- ‘run’⁵⁶ (perhaps secondarily already Proto-Indo-European ‘flow’) occurs appellatively only in Indo-Iranian. The zero grade of this root is attested only in forms that have no direct connection with the formation under discussion, namely the preverb-only PPP, Sanskrit -drutá- ‘run’ and the plural of the perfect, for example 3rd pl. perf. act. Sanskrit dudruvur.

The present tense of this root shows the full grade and thematic vowel in the few Indo-European languages in which it is attested at all, namely Proto-Indo-European *drēu-e/o-. An active present participle must therefore have been Proto-Indo-European masculine *drēu-ont-, feminine *drēu-ont-ih₂-. A zero-grade verbal formation from which an active participle in -nt- could have been formed would have had to be either a Proto-Indo-European root present †dréu-ti, †dru-énti > †dṛu-énti or a root aorist or thematic aorist—that is, forms for which there is no appellative attestation anywhere. The participle of the hypothetical stems mentioned would have been †dru-ént-, which, according to the phonotactic rules of Proto-Indo-European, would have had to be vocalized as †dṛu-ént-; the feminine form would probably have been †dru-nt-íh₂-, which would have been vocalized as †dru-nt-ih₂-. In other words, to produce the (supposed) participle, one would have to combine the zero grade of the unattested root aorist with the o-grade suffix form, which is found only in thematic formations—in this case, only in the present stem. Moreover, one would have to assume that the regular syllabification of the zero-grade form was remodeled on the full-grade form, that is, †dṛu-ént- was remodeled as †dru(u)-ént-. Such a development is of course theoretically conceivable and may perhaps occur somewhere at some time. I am not, however, aware of any parallel cases that securely show this sequence of developments. We would be grateful to Mr. Udolph if he could provide such examples in support of his etymological proposals.

⁵⁵ The existence of a verb of the so-called tudáti type (*KṚK-ē/ó-) would of course also be theoretically conceivable; the trouble is that there is not the slightest indication of one for this root.

⁵⁶ Cf. LIV² 129.

Straipsniai / Articles

31

HARALD BICHLMEIER

By its nature, however, this would in any case once again be a phenomenon that could only have occurred in an individual language, and certainly not in the proto-language. If this is so, this formation can hardly be attributed to a Proto-Indo-European that was already at most dialectally differentiated, as is often done for ancient European hydronyms; rather, it would have to be assigned to a later stage.

If one posits the two aorist stems mentioned above, further irresolvable problems arise: root aorists are formed only from verbal roots with punctual meaning (in our case, one would have to posit for the root a meaning such as ‘flow away’ or ‘splash (up)’, which is evidently not the case). The root in question here is clearly durative (‘run’, possibly secondarily ‘flow’). It therefore probably never formed a root aorist. It is also possible, but unlikely, for a zero-grade thematic aorist to occur alongside a thematic present. Moreover, thematic aorists belong to a late phase in the historical development of Proto-Indo-European. That is, positing a thematic aorist would, in principle, once again contradict the thesis that ancient European hydronymy represents a direct outflow of Proto-Indo-European.

Finally, one must also consider what meaning an active participle of a root aorist would have had: one would have to assume something like ‘flowing away’ or ‘(once) splashing up/over’. Is that really appropriate for a river? All this necessarily shows that the classical, or Udolphian, explanation cannot be correct as it stands.

One must therefore continue searching for an explanation that is also compatible with the findings of comparative linguistics.

One possibility to consider is whether, in view of the problems mentioned above, it would be better to adopt an etymology deriving from a possessive adjective m. *dru-u̯ént-, f. *dru-u̯ṇt-ih₂-, which would mean ‘rich in wood’ (possibly also ‘rich in forest’): the –e-forms of the earliest river-name attestations, which are feminine, would then of course require assuming paradigmatic leveling after the masculine forms (which, however, would be a trivial process), while—as indicated above—the feminine form, which originally had a zero-grade suffix, could indeed have continued as *dru-u̯ṇt- > Druant-. Phonologically, this etymology could explain the hydronymic attestations with <-uant-> as well as those with <-uent->, and it also presents no semantic problem: a river could be described as rich in forest or wood because it often carried driftwood, or because dense woodland lined its course. Given that this is thus far the only possibility for reconciling the attested forms without contradiction with the morphological system of Proto-Indo-

32

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteropäischen Hydronymie – eine Erwiderung auf eine Polemik

European, this etymology should clearly be preferred to Udolph’s solution.⁵⁷

In the context of the etymology of these river names, another example should be noted of how much Udolph’s Indo-Europeanist knowledge needs refreshing—assuming, of course, that this is not a deliberately distorted presentation intended to lend the author’s own theory apparent support by reference to Indo-Europeanist scholarship: In his discussion of the river-name group at issue here, centered on Druentia/Druantia, he cites (1990: 111f.) an article by Alfred Bammesberger (1981). As is already clear from its title, this article deals only with athematic verbal stems. As explained above, however, only a thematic present is attested for Proto-Indo-European *dreụ-. None of this, therefore, has anything at all to do with the case under discussion here. Moreover, the -ant- forms cited by Bammesberger, loc. cit., to which Udolph refers, are either Old Indic forms (which, owing to the sound laws of Old Indic, do not permit any direct inference in this position as to whether their vowel derives from Proto-Indo-European *e, *a or *o), or they are forms in Greek verbs with Proto-Greek *ā- as their root vowel, which, according to Bammesberger, arose by analogy; according to the view accepted today (and already widespread in 1990), these are in any case more plausibly traced to the coloring effect of Proto-Indo-European *h₂ (Proto-Indo-European *sth₂-ént- > Proto-Greek *stānt-), or perhaps, in this case, better explained by Osthoff’s shortening of a long vowel before a resonant-plus-stop cluster (Proto-Indo-European *stéh₂-ṇt- > Proto-Greek I *stānt- > Proto-Greek II *stánt-), if it should prove to be a formation originating only in Proto-Greek. With uninterrupted sound development, this verb might in any case even have been expected to show Proto-Indo-European *stéh₂-ṇt- > Proto-Greek I *sta.at- > Proto-Greek II *stāṭ-,⁵⁸ in which case the participial -nt- could then have been secondarily restored by analogy in this form (and the long vowel subsequently shortened again in accordance with Osthoff’s law).⁵⁹ In other words, the forms with -ant- owe their existence (both according to Bammesberger’s laryngeal-less account and to the view adopted here, which posits laryngeals in Proto-Indo-European) solely to developments internal to Greek or to

⁵⁷ It should be noted that this proposal must be distinguished from earlier proposals, already cited by Udolph (1990: 109f.), which likewise take a word for ‘tree, wood’ as the basis of the etymology. Those proposals, too, generally reflect outdated views on word formation and mostly assume that the hydronym was formed only later in Slavic.

⁵⁸ If one considers the effect of Lex Rix possible in medial position as well, one might possibly posit Proto-Indo-European *stéh₂-ṇt- > Proto-Greek I *sta.ant- > Proto-Greek II *stānt- > Proto-Greek III *stant-.

⁵⁹ The example of Old Indian shows that such restitutions need not necessarily occur: in some verbal classes, the suffix variants -ant-: -at- (< Proto-Indo-European *-ént-: -ṇt-) remained side by side.

Straipsniai / Articles

33

HARALD BICHLMEIER

Indo-Aryan sound changes. And there is no evidence here anywhere for -ont-. Consequently, the supposed proof for the forms postulated by Udolph (such as, for example, the aforementioned “Druantia”) does not exist. Udolph’s argument must therefore be rejected as incorrect.

To emphasize the point once again: ablaut is not an arbitrary phenomenon, whether in the root or in the suffixes (or in the endings, which, however, is of no further relevance here). Rather, it is subject to fairly strict rules, most of which Indo-European studies have presumably now elucidated. The continued disregard of such findings can only lead, on the one hand, to the loss of scholarly currency and/or to the results thus obtained (i.e. by means of entirely outdated methods of reconstruction) having to be regarded as unusable or irrelevant; and, on the other hand, to the kind of scholarship thus practiced having to be called naive, in the sense of the quotation prefixed to this contribution (in contrast to well-founded scholarship proper, here namely Indo-European studies in their current form).

8. SECOND INTERIM CONCLUSION

And so, as a further interim assessment, we can note the following: Udolph severely criticizes the reconstructions of Indo-Europeanists, even writing at one point, apparently with the intention of discrediting the methodology of Indo-European studies, of “„re“constructed” (2011: 165) forms—but he himself evidently also reconstructs (at least, he too uses asterisks before word or root forms, which normally indicate that a word form has been reconstructed); he simply does so using outdated methods, and therefore less well than any Indo-Europeanist (or student of Indo-European studies).

The fact that—as should already have become clear in the preceding sections as well—scholarly discussion in Indo-European studies is still ongoing in some areas and concerning quite a few reconstructions is, surprisingly, repeatedly construed as a weakness by the same outsiders to the field mentioned above—even though precisely this kind of culture of debate ought rather to be one of the strengths of every humanities discipline. For progress can be achieved only through discussion and the exchange of opinions. Disciplines or fields, by contrast, that are dominated by the views of one person or a small number of people, who may not even be able to renew their own discipline regularly or to take on board new trends (which may sometimes also come from neighboring fields), but do everything “the way it has always been done,” ultimately harm that discipline.

34

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

In this sense, Udolph’s approach to Old European hydronymy is clearly a model whose time has passed, and it is indeed time for Indo-Europeanists to take up the matter again at last.

9. ON UDOLPH’S REMARKS ON THE ETYMOLOGY OF THE RIVER NAME Iller

Kommen wir nun zu weiteren Punkten der Polemik Udolphs (2011: 169), in denen er – wie eingangs schon an einer anderen Stelle moniert – wieder (absichtlich?) meine Darstellung etwas verzerrt wiedergibt:

In my contribution, I clearly pointed out that I consider not only the initial H- in the earliest attestations of the name Iller to be non-etymological (which Udolph apparently also does, so we agree on this point), but also regard the single -l- as at least suspicious, since in both cases one must reckon with what I called a “monkish-etymological” reinterpretation involving Latin hilaris ‘cheerful’.⁶⁰ Udolph does not address this possibility at all. It was also made clear that the geminates that become more frequent in medial position from the 12th century onward must be regarded as linguistically real, but that there is simply no possibility known to me by which a geminate could have arisen, in the transition from Old High German to Middle High German, in a river name supposedly going back to *Elira-. In his current argument, Udolph seeks to bypass this problem by resorting to the trick of supplying a Middle High German form in the first place and never positing an Old High German form. He writes: From *Elira “through the influence of -i- on the initial E-^[61], the regular Middle High German form *Ilere arose; through reduction of unstressed syllables, the attested forms Ilre, Ilr arose. Only afterward did the geminate develop. Laryngeals played no role.” Indeed, laryngeals play no role in the reduction of unstressed syllables in Middle High German—but, to my knowledge, nobody has ever claimed that they did.

Udolph does not address the problem I raised, namely that in Old High German one primarily finds forms of the type (H)ilar- (occasionally also (H)iler- and once (H)il(i)r-): one simply does not get to Old High German (H)ilar- automatically from the postulated “*Elira.” And we are back at the starting point of my argument: if the H- and the -a- can be attributed to the influence of Latin hilaris, then of course this may also apply to the single -l-. And since Udolph himself has now, thankfully, provided an early attestation for the

⁶⁰ Vgl. Bichlmeier 2010a: 27f.

⁶¹ Experts would speak here of ‘raising’.

Straipsniai / Articles

35

HARALD BICH LMEIER

Citing a geminate, “804 Hillergaugensem,” it seems even more likely to me that it is precisely the form with a geminate that is old, and not the one with a single -l-. In principle, it should be much more likely that a Germanic or Old High German geminate was not recorded by monks writing in Latin than that a single consonant was written as a geminate. How the single -l- is then supposed to have become a geminate in the 11th/12th/13th centuries, as Udolph postulates, remains wholly incomprehensible: nothing of the sort occurs, and accordingly it is found in no grammar—tellingly, in his article Undolph cites not a single one of the numerous grammars of Old and Middle High German, let alone any historical grammar of German. Rather, in light of the “attested forms Ilre, Ilr” from the Middle High German period that he cites, it seems plausible to suspect that he is considering whether the geminate might have arisen through the following -r-: something of this kind certainly exists—but 500 to 1,000 years earlier, namely West Germanic consonant gemination before *-r-/-l-/-m-/-n-/-u-.⁶²

If, however, Udolph can still produce conclusive evidence, together with citations from the secondary literature, for the phonetic phenomena in the history of German that he has, in my opinion, entirely invented, scholarship would surely be grateful to him. There is, however, reason to fear that he will not succeed.⁶³

Thus, at this point we have already reached the conclusion that the traditional solution for deriving the name of the Iller, which Udolph vehemently defends (precisely because it is traditional), simply does not work and must therefore be rejected.

In his further remarks on the etymology of this name, now concerning the suffix occurring in it, Udolph demonstrates, on the one hand, a somewhat peculiar deference to authority, while on the other, a need to catch up in the history of Germanic and German sound change becomes evident:

On p. 169 he immediately points out that Adolf Bach already regarded the suffixes “-ra, -ara, -ura, -ira, -era” as self-evident for “feminine river names” in his work on the study of place names. These suffixes are, of course, present at some level or chronological stage—but neither Bach nor Udolph says which one: as one can probably infer from the statements of both,

⁶² Cf. Krahe/Meid 1969: 1, 104–107.

⁶³ As further examples of Udolph’s careless treatment of findings from the historical study of sound, morphology, and semantics in various languages (especially Proto-Indo-European and Germanic), one may cite his etymology of the place name Magdeburg (for which he has invented a Germanic adjective “*magath- ‘large’; cf. Bichlmeier 2011d), or his etymology of place names containing Hall-, which is phonetically sound this time but semantically wholly unsatisfactory (according to Udolph, ‘slope’ or the like; cf. Bichlmeier 2012e, note 12, with references to relevant literature).

36

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

one may conclude, consistently, that this is a Late Common West Germanic synchronic stage (if not often an even later one). In other words, here too no acceptable or even meaningful etymological information is provided: the -r- occurring in these suffixes in river names of Germanic or Germanic-mediated provenance may go back to Proto-Indo-European *-s- or *-r-; among the vowels before -r-, -a- may continue Late Proto-Indo-European *a, *o, perhaps also *ā; -u- could go back to *u, *ū; -i- would consequently go back to *e, *i, and probably *ī as well; and -e- most likely continues *e, provided the word forms are not already attestations reflecting the weakening of unstressed syllables to /ə/, which began in Late Old High German.

Only once all this has been established for each individual name form (or at least for a sufficiently large and therefore representative selection) can one meaningfully address the question of which name forms can actually be compared with non-Germanic or non-Germanic-mediated names that also have “a suffix -ira” (Udolph 2011: 169; if the spelling without an asterisk is meant seriously, Udolph commits another methodological error here, since he then compares actually attested word forms from one region with reconstructed ones from another, that is, he does not argue on the same chronological level)⁶⁴. Here, too, one must first follow the procedure just outlined and clarify what the vowels and consonants contained in the suffix could go back to. This also applies to the river names cited by Udolph (2011: 170), in which he “unquestionably” identifies the suffix *-ira: what is telling here, once again, is that—as so often in this article and in numerous others—he gives no indication of which linguistic stage his “(*)-ira” belongs to. If this is not intentional, meant to conceal his own indecision or lack of knowledge and/or to confuse the reader, it is yet another of the numerous terminological imprecisions in Udolph’s work: if the suffix (complex) were Proto-Indo-European, he would have to write *-ireh₂- or Late Proto-Indo-European *-irā-; if it were Proto-Germanic, it should actually be *-irō-; if it is only West Germanic, *-ira- may be acceptable, but it could then go back to Proto-Germanic *-irō- or *-izō-. One would dearly like to know what Udolph actually means; then one could address his (purported) arguments far more precisely.

⁶⁴ Vennemann takes a comparable approach in many of his works, for instance when, to substantiate his Vasconic etymologies, he compares medieval forms of German place and river names with modern Basque words, instead of comparing, as methodology requires, the Proto-Basque reconstructions of the Basque words in question with the pre-Old High German/Proto-Germanic reconstructions of the place and river names.

Straipsniai / Articles

37

HARALD BICH LMEIER

A sound comparison of all the water names mentioned is possible only once the etymology of each individual name has been clarified. So far, at most initial tentative morpheme analyses are available for most of these names (which are finally to be investigated properly)—nothing more.

Further evidence in this connection that Udolph’s command of the historical phonology of the Germanic languages is still incomplete is his statement (Udolph 2011: 169) concerning rhotacism, that is, the transition of Proto-Germanic *-z- > -r-: “Such a thing is possible only in the Proto-Germanic area, that is, on Proto-Germanic settlement territory.” Quite apart from whether one accepts Udolph’s localization of the Germanic homeland in northern Germany in the broader sense (outside Udolph’s circle and that of his students, this is anything but established),⁶⁵ it is beyond doubt that rhotacism is a comparatively late phenomenon: Gothic (Bible Gothic, late 4th century; manuscript from the 6th century [Codex argenteus]) shows no rhotacism, nor do the early runic inscriptions. As is well known, Gothic even has its own grapheme <z> for the phoneme /z/, cf. Gothic saislep alongside saizlep, or Gothic huzds vs. Modern High German Hort). Although the dating is naturally uncertain because of the scant evidence, this sound change in West Germanic can probably be dated no earlier than the first half of the first millennium AD, perhaps even to its middle. One of the few ways to date this sound change somewhat more precisely might even come from the field of hydronymic naming: if one accepts the etymology of the river name Czech Jizera, German Iser (1297 Gizera) (about 165 km, right-bank tributary of the Elbe), and its derivation from West Germanic */izero-/ < (Early) Proto-Germanic *isa/erā- (or < Proto-Celtic *isaro/ā- [if a Proto-Celtic intermediate stage is assumed]) < Proto-Indo-European *h₁ish₂-(e)ró/éh₂-^{66, 67}, then this name proves that when the Slavs arrived in the Bohemian Basin—which, according to the generally accepted view, did not occur until the 6th century—they heard the name with a voiced fricative in the middle, and not yet with -r-. The German name Iser is to be regarded as a secondary borrowing from Czech. If this is correct, the year 500 AD provides the terminus post quem for rhotacism in West Germanic.

It follows directly from all this that Germanic words, too, could still undergo this sound change if they entered Common Germanic relatively late—

⁶⁵ Cf. Wagner 1994/95.

⁶⁶ Cf. Blažek 2010: 32.

⁶⁷ For a detailed discussion of some hydronyms formed from this etymon, see Bichlmeier 2011c and Bichlmeier 2012b.

38

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

anic, but were nevertheless formed early enough, just as words borrowed by Germanic peoples from other languages could still undergo this change. If we assume that Germanic peoples had arrived in the Iller river basin in considerable numbers by the 3rd century AD at the latest (their first incursions into the Agri Decumates had already occurred from the second half of the 2nd century AD), I see no problem in supposing that Germanic speakers either formed a form *ellizō themselves, or heard and adopted a similarly sounding non-Germanic (perhaps most likely Celtic?) name form, which would then have developed like all other Germanic words.⁶⁸ I still consider it possible that the name of the Iller is indeed of Germanic origin; I have outlined an etymological derivation that, in my view, properly takes account of the historical circumstances and phonological developments. I am open to suggestions for improving this etymology through further research; Udolph’s arguments, by contrast, are a step backwards.

In short, here too there is no trace of etymology in the sense of explaining words at every level; there are merely descriptions of one, or probably even several, synchronic states that have been wrongly treated together (probably from several languages), including an initial morpheme analysis. Apart from the Germanic or Germanic-mediated river names mentioned here, one must also consider the river names that are certainly non-Germanic or not transmitted through Germanic and that display similar suffixes. In those cases, too, it must first be established whether rhotacism is possible and what the various vocalic elements go back to. Udolph’s juxtaposition of all these similar suffixes from broad regions of Europe is methodologically highly questionable as long as the individual forms have not been clarified etymologically, and can really only lead to incorrect conclusions. Here again it becomes clear how necessary it would be to—

⁶⁸ Absolute dating of prehistoric sound changes (i.e. those occurring before a language is attested in writing) remains difficult, of course, but I would like to draw attention here to a more recent internal chronology of Proto-Germanic that Mottausch (2011: 2f., 173f.) has made probable. Mottausch places his ‘Proto-Germanic III’, whose linguistic characteristics correspond to the Proto-Germanic otherwise reconstructed in Indo-European studies, in the period from the 1st century BC to the 2nd century AD; this linguistic stage must have been reached before the Goths migrated away (at the end of the 2nd century AD). This ‘Proto-Germanic III’ had in any case not yet undergone the rhotacism *z > r. Unless this sound change therefore took place immediately afterward, that is, in the first decades of the 3rd century AD (for which there is so far no evidence whatsoever), this chronology also supports the view that the Germanic peoples moving into the Alpine foothills still had the sound /z/ in their language.

Straipsniai / Articles

39

HARALD BICHLMEIER

to review all the relevant material once again, reassess it according to current standards and, above all, before embarking on a major synthesis, first investigate separately the etymologies of the individual hydronyms. In light of all that has been said here, the approach applied so far can only be described as unsound. The results of previous studies must therefore be called into question across the board, since, given their outdated methodology, they can at best have arrived at correct results by chance.

In any case, Udolph’s arguments have done nothing to bring us closer to a solution to the problem of the etymology of the river name Iller.

10. SUMMARY AND OUTLOOK

Taking stock, then, of the foregoing discussion of Udolph’s remarks (in 2011 and elsewhere), one can only reach a single conclusion: It is indeed time for Indo-Europeanists, as specialists in the historical-comparative linguistics of Old European hydronymy, to take it up again. Udolph’s call for this should unquestionably be heeded, and by many. For it is essential that these hydronyms finally be analyzed according to the criteria and methods of modern Indo-European linguistics. In terms of methodology and knowledge of relevant secondary literature in linguistics, especially Indo-European studies, and other research tools, Udolph’s work remains at Krahe’s level—and thus, ultimately, mostly at that of the interwar period. The advances made in Indo-European studies over the past half-century are systematically ignored. Udolph is nowhere near the current state of scholarship in either the phonology and historical phonetics of Proto-Indo-European or those of the individual languages (such as Germanic and Slavic). This lag is particularly glaring in historical morphology, especially word formation: he never gets beyond what is usually, at best, a rudimentary description or naming of the morphemes occurring in a word or name—and even the morpheme analyses he offers, which are merely the first step toward a complete etymology, meet at best the standards of Krahe’s time, owing to decades of disregard for Indo-European research on word formation. Ultimately, neither in the works by Udolph examined here in detail nor, for the most part, elsewhere in his oeuvre, can one find a complete etymology that would do justice to de Saussure’s insight that a word consists of form and content

40

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

(both of which must therefore be explained when an etymology is constructed).

No one should be misled by the equally polemical and self-assured tone of the statements by the “Lord of Names” discussed here at length: they reveal serious inconsistencies in every area of historical linguistics.

In light of the quotation from Vennemann prefacing this article, etymologies à la Udolph must be called “naive,” since he consistently refuses to ground his work in the “relevant findings” provided by Indo-European linguistics.

Given the clear shortcomings demonstrated here in a number of selected etymologies by Udolph, his research findings will have to be treated with even greater caution in the future: since the linguistic analysis of the respective word forms has demonstrably serious deficiencies, it is questionable whether conclusions drawn from material obtained in this way (for example, concerning the original homeland of the Slavs and Germanic peoples) can stand at all.

At the current state of research, no definitive answer can yet be given. It should, however, have become clear that all the material collected and analyzed by Udolph and his predecessors must be re-examined from the ground up using the methods of modern Indo-European linguistics, in order to purge it of etymologies that are clearly at least dubious, if not outright wrong. With the material thus prepared, one can then once again set about considering word-formation patterns in hydronyms and, finally, perhaps also speculate anew about the extent to which conclusions might be drawn from them about possible homelands.

The proponents of Old European studies have thus left Indo-European linguistics a mountain of work, having failed to apply to material familiar to them, in some cases for decades, the research tools provided by Indo-European linguistics. At the same time, Indo-Europeanists made the mistake of leaving the analysis of this material to researchers who were not methodologically sufficiently equipped. Since Old European studies will not be able to make up this scholarly lag in the foreseeable future (not least because there is a lack of trained early-career researchers), Indo-Europeanists will have to shoulder this task. The future will show whether enough younger scholars can be enlisted for it, scholars who will not be deterred by the old-fashioned presentation of the research findings to date—after all, Indo-European linguistics already offers more interesting fields of research than can be properly investigated.

Straipsniai / Articles

41

HARALD BICHMEIER

BIBLIOGRAPHY

ACS – Holder Alfred: Alt-Celtischer Sprachschatz. 3 vols. Leipzig, 1896–1907.

Alfieri Luca 2011: Radici Indoeuropee inizianti in *#V-, l’esistenza di *a- primaria e l’esito di *#h₃e- in Anatolico. – Historische Sprachforschung 123, 2010 [2011], 1–39.

Bammesberger Alfred 1981: Das -nt-Partizip bei athematischen Verbalstämmen im Griechischen. – Kuhns Zeitschrift 95, 1981, 286–292.

Bichlmeier Harald 2007: Rezension zu Greule/Janka/Prinz (Hgg.) 2005. – Zeitschrift für bayerische Landesgeschichte 69/3, 2006 [2007], 1011–1014.

Bichlmeier Harald 2010a: Bairisch-österreichische Orts- und Gewässernamen aus indogermanistischer Sicht. – Blätter für oberdeutsche Namenforschung 46, 2009 [2010], 3–63.

Bichlmeier Harald 2010b: Rōma – Namenkundlich-sprachhistorische Anmerkungen zu einem allgemein bekannten Ortsnamen (Mit einem Exkurs zu Fragen der Chronologie von Lehnwortbeziehungen benachbarter Sprachen am Beispiel von nhd. Haus und seinen Vorformen). – Das Altertum 55, 2010, 175–202.

Bichlmeier Harald 2010c: Arelape, Tergolape, Interlaken – Ein Beitrag zur keltischen Fluss- und Ortsnamenschicht im Alpenraum. – Zeitschrift für Historische Sprachforschung 122, 2009 [2010], 254–267.

Bichlmeier Harald 2010d: Moderne Indogermanistik vs. traditionelle Namenkunde, Teil 3: Traun, Raab und Auders. – Österreichische Namenforschung 38, 2010, 104–113.

Bichlmeier Harald 2011a: Einige grundsätzliche Überlegungen zum Verhältnis von Indogermanistik und voreinzelsprachlicher resp. alteuropäischer Namenkunde mit einigen Fallbeispielen (Moderne Indogermanistik vs. traditionelle Namenkunde, Teil 1). – Namenkundliche Informationen 95/96, 2009 [2011], 173–208.

Bichlmeier Harald 2011b: Rōma – hydronyme « paléoeuropéen » ou désignation d’une « terre agricole »? Tentative de mise au point et brèves considérations sur d’autres propositions étymologiques plus anciennes. – Nouvelle Revue d’Onomastique 51, 2009 [2011], 69–84.

Bichlmeier Harald 2011c: Bayerisch-österreichische Orts- und Gewässernamen aus indogermanistischer Sicht: Teil 2: Isar und etymologisch Verwandtes (sowie Addenda zu dem Beitrag in den BONF 46 (2009), 3–63). – Blätter für oberdeutsche Namenforschung 47, 2010 [2011], 21–31.

Bichlmeier Harald 2011d: Noch einmal zum Ortsnamen Magdeburg. – Namenkundliche Informationen 97, 2010 [2011], 109–132.

Bichlmeier Harald 2011e: Moderne Indogermanistik vs. traditionelle Namenkunde, Teil 2 – Save, Drau, Zöbern. In: Ziegler, Windberger-Heidenkummer 2011, 63–87.

42

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

Bichlmeier Harald 2012a: Rezension zu: Stadnik-Holzer Elena, Holzer Georg (Hgg.): Sprache und Leben der frühmittelalterlichen Slaven. Festschrift für Radoslav Katičić zum 80. Geburtstag. Mit den Beiträgen zu den Scheibbser Internationalen Sprachhistorischen Tagen II und weiteren Aufsätzen. (= Schriften über Sprachen und Texte 10). Frankfurt a. M. etc.: Peter Lang, 2010. – Zeitschrift für Balkanologie 47/2, 2011 [2012], 277–281.

Bichlmeier Harald 2012b: Einige ausgewählte Probleme der alteuropäischen Hydronymie aus Sicht der modernen Indogermanistik – Ein Plädoyer für eine neue Sicht auf die Dinge. Acta Linguistica Lithuanica 66, 2012, 11–47.

Bichlmeier Harald 2012c: Anmerkungen zum terminologischen Problem der ‚alteuropäischen Hydronymie’ samt indogermanistischen Ergänzungen zum Namen der Elbe. – Beiträge zur Namenforschung, Neue Folge, 47/4, 2012, 365–395.

Bichlmeier Harald 2012d: Welche Erkenntnisse lassen sich aus dem lexikalischen und morphologischen Material der ‘alteuropäischen Hydronymie’ für die Indogermanistik gewinnen? – Versuch einer Bilanz. Vortrag gehalten auf dem Gedenk-Kolloquium für Vallija Dambe in Riga am 10.5.2012.

Bichlmeier Harald 2012e: Einige indogermanistische Ergänzungen zur Etymologie des Namens der Thüringer. – Beiträge zur Namenforschung, Neue Folge, 47/2, 2012, 207–224.

Bichlmeier Harald (in press): Notes on the terminological problem of ‘Old European hydronymy’, together with Indo-European supplements concerning the name of the Elbe. – Beiträge zur Namenforschung, Neue Folge 47/3, 2012 or 48/1, 2013.

Blažek Václav 2010: Etymological analysis of toponyms from Ptolemy’s description of Central Europe. – Brozović Rončević, Fomin, Matasović 2010, 21–45.

Brozović Rončević Dunja, Fomin Maxim, Matasović Ranko (Hgg.) 2010: Celts and Slavs in Central and Southeastern Europe. Studia Celto-Slavica III. Proceedings of the III^rd International Colloquium of the Societas Celto-Slavica, Dubrovnik, September 18–20, 2008. (Znanstveni zbornici 2) Zagreb: Institut za hrvatski jezik i jezikoslovlje.

Clackson James 2007: Indo-European Linguistics. An Introduction (Cambridge textbooks in linguistics). Cambridge: Cambridge University Press.

Collinge Neville Edgar 1985: The Laws of Indo-European (Amsterdam studies in the theory and history of linguistic science, Serie IV, 35). Amsterdam – Philadelphia: John Benjamins.

Derksen Rick 1996: Metatony in Baltic (Leiden Studies in Indo-European 6). Amsterdam – Atlanta, GA: Rodopi.

EDG – Beekes Robert with the assistance of Lucien van Beek: Etymological Dictionary of Greek (Leiden Indo-European Etymological Dictionary Series 10. Leiden – Boston: Brill, 2010.

Straipsniai / Articles

43

HARALD BICHLMEIER

EDL – de Vaan Michiel: Etymological Dictionary of Latin and the other Italic Languages (Leiden Indo-European Etymological Dictionary Series 7). Leiden – Boston: Brill, 2008.

EDHIL – Kloekhorst Alwin: Etymological Dictionary of the Hittite Inherited Lexicon (Leiden Indo-European Etymological Dictionary Series 5). Leiden – Boston: Brill, 2008.

EDPC – Matasović Ranko: An Etymological Dictionary of Proto-Celtic (Leiden Indo-European Etymological Dictionary Series 9). Leiden – Boston: Brill, 2009.

EDPC-Add. – Addenda et Corrigenda to Ranko Matasović’s Etymological Dictionary of Proto-Celtic (Brill, Leiden 2009) [Zagreb, December 2011]: http://mudrac.ffzg.unizg.hr/ ~rma-tasov/EDPC-Addenda%20et%20%20corrigenda.pdf; 15.4.2012).

EDSIL – Derksen Rick: Etymological Dictionary of the Slavic Inherited Lexicon (Leiden Indo-European Etymological Dictionary Series 4). Leiden – Boston: Brill, 2008.

ESJS – Havlová Eva et al.: Etymologický slovník jazyka staroslověnského. Prague: Academia, from fascicle 15: Prague: Tribun EU, 1989ff.

EWAhd – Etymologisches Wörterbuch des Althochdeutschen. Volume I: –a – bezzisto. By Albert L. Lloyd and Otto Springer. Göttingen – Zürich: Vandenhoeck & Ruprecht, 1988. Volume II: bî – ezzo. By Albert L. Lloyd, Rosemarie Lühr, and Otto Springer†, with the collaboration of Karen R. Purdy. Göttingen – Zürich: Vandenhoeck & Ruprecht, 1998. Volume III: fadum – fūstslag. By Albert L. Lloyd and Rosemarie Lühr, with contributions from Gerlinde Kohlrusch, Maria Kozianka, Karen R. Purdy, and Roland Schuhmann. Göttingen: Vandenhoeck & Ruprecht, 2007. Volume IV: gāba – hylare. By Albert L. Lloyd and Rosemarie Lühr, with contributions from Gerlinde Kohlrusch, Maria Kozianka, Karen R. Purdy, and Roland Schuhmann. Göttingen: Vandenhoeck & Ruprecht, 2009.

EWAia – Mayrhofer Manfred: Etymologisches Wörterbuch des Altindoarischen. 3 vols. Heidelberg: Winter 1992, 1996, 2001.

EWD – Pfeifer Wolfgang (ed.): Etymologisches Wörterbuch des Deutschen. 2nd ed. Berlin: dtv, 1993.

EWdS²⁴ – Kluge Friedrich, Seebold Elmar: Etymologisches Wörterbuch der deutschen Sprache. 24th, rev. and enlarged ed. Berlin – New York: de Gruyter, 2002.

EWdS²⁵ – Kluge Friedrich, Seebold Elmar: Etymologisches Wörterbuch der deutschen Sprache. 25th ed. Berlin – Boston: de Gruyter, 2011.

EwN – Philippa, Marlies et al. (eds.): Etymologisch woordenboek van het Nederlands. I: A–E. Amsterdam: Amsterdam University Press, 2004. II: F–Ka. Amsterdam: Amsterdam University Press, 2005. III: Ke–R. Amsterdam: Amsterdam University Press, 2007. IV: S–Z. Amsterdam: Amsterdam University Press, 2009.

44

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

Fortson Benjamin W. IV 2004: Indo-European Language and Culture – an Introduction (Blackwell textbooks in linguistics 19). Malden – Oxford – Carlton: Blackwell.

Fortson Benjamin W. IV 2009: Indo-European Language and Culture – an Introduction. Second Edition (Blackwell textbooks in linguistics 19). Malden – Oxford – Carlton: Blackwell.

Görtzen Jens 1998: Die Entwicklung der indogermanischen Verbindungen von dentalen Okklusiven mit besonderer Berücksichtigung des Germanischen (Innsbrucker Beiträge zur Sprachwissenschaft 94). Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft.

Greule Albrecht 2011: Arbeits- und Darstellungstechniken des Deutschen Gewässernamenbuchs. In: Ziegler, Windberger-Heidenkummer, 2011, 117–126.

Greule Albrecht, Janka Wolfgang, Prinz Michael (Hgg.) 2005: Gewässernamen in Bayern und Österreich. 3. Kolloquium des Arbeitskreises für bayerisch-österreichische Namenforschung (Regensburg, 27. / 28. Februar 2004) (Regensburger Studien zur Namenforschung 1). Regensburg: Vulpes.

Hackstein Olav 2002: Uridg. *CH.CC > *C.CC. – Historische Sprachforschung 115, 1–22.

Hill Eugen 2003: Untersuchungen zum inneren Sandhi des Indogermanischen. Der Zusammenstoß von Dentalplosiven im Indoiranischen, Germanischen, Italischen und Keltischen (Münchner Forschungen zur historischen Sprachwissenschaft 1). Bremen: Hempen.

Kapović Mate 2008: Uvod u indoeuropsku linguistiku. Pregled jezikâ i poredbena fonologija. Zagreb: Matica hrvatska.

Krahe Hans 1954: Sprache und Vorzeit. Europäische Vorgeschichte nach dem Zeugnis der Sprache. Heidelberg: Quelle & Meyer.

Krahe Hans 1963: Die Struktur der alteuropäischen Hydronymie (Abhandlungen der Geistes- und Sozialwissenschaftlichen Klasse, Akademie der Wissenschaften und der Literatur Mainz 1962,5). Wiesbaden: Steiner (in Kommission).

Krahe Hans 1964: Unsere ältesten Flußnamen. Wiesbaden: Harrassowitz.

Krahe Hans, Meid Wolfgang 1969: Germanische Sprachwissenschaft. Bd. 1: Einleitung und Lautlehre. Bd. 2: Formenlehre. Bd. 3: Wortbildungslehre. 7. Aufl. bearbeitet von Wolfgang Meid. Berlin – New York.

Larsson Jenny Helena 2003: Studies in Baltic Word Formation. University of Copenhagen.

LEIA – Vendryes Joseph: Lexique étymologique d’Irlandais ancien. A: Dublin: Dublin Institute for Advanced Studies, 1959 (reprint 1981). B: Dublin: Dublin Institute for Advanced Studies – Paris: Centre National de la Recherche Scientifique, 1981.

Straipsniai / Articles

45

HARALD BICH LMEIER

C: Dublin: Dublin Institute for Advanced Studies – Paris: Centre National de la Recherche Scientifique, 1987.

D: Dublin: Dublin Institute for Advanced Studies – Paris: Centre National de la Recherche Scientifique, 2007.

M, N, O, P: Dublin: Dublin Institute for Advanced Studies, 1983.

R, S: Dublin: Dublin Institute for Advanced Studies – Paris: Centre National de la Recherche Scientifique, 1974.

T, U: Dublin: Dublin Institute for Advanced Studies – Paris: Centre National de la Recherche Scientifique, 1978.

Leumann Manu, Hofmann Johannes Baptist, Szantyr Anton 1977: Lateinische Grammatik. Erster Band: Lateinische Laut- und Formenlehre (Handbuch der Altertumswissenschaft, Abt. 2, Teil 2). München: Beck.

LEV – Karulis Konstantīns: Latviešu etimoloģijas vārdnīca diuos sējumos. I: A–O. II: P–Ž. Rīga: Avots, 1992.

Lindeman Fredrik Otto 1997: Introduction to the ‘Laryngeal Theory’ (Innsbrucker Beiträge zur Sprachwissenschaft 91). Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft.

LIV – Rix Helmut et al.: Lexicon of Indo-European Verbs. The Roots and Their Primary Stem Formations. Wiesbaden: Reichert 1998.

LIV² – Rix Helmut et al.: Lexicon of Indo-European Verbs. The Roots and Their Primary Stem Formations. Second, revised and expanded edition, Wiesbaden: Reichert, 2001.

Lühr Rosemarie (Hg.) 2008: Nominale Wortbildung des Indogermanischen in Grundzügen. Die Wortbildungsmuster ausgewählter indogermanischer Einzelsprachen. Band 1: Latein, Altgriechisch bearbeitet von Irene Balles (Schriftenreihe Philologia 121). Hamburg: Kovač. Band 2: Hethitisch, Altindisch, Altarmenisch bearbeitet von Joachim Matzinger (Schriftenreihe Philologia 122). Hamburg: Kovač.

LVŽ – Autorenkollektiv: Lietuvos vietovardžių žodynas. I: A–B. Vilnius: Lietuvių kalbos institutas, 2008.

Mallory James P., Adams Douglas Q. 2006: The Oxford Introduction to Proto-Indo-European and the Proto-Indo-European World. Oxford: Oxford University Press.

Matasović Ranko 1992 [²2010]: Kratka poredb enopovijesna gramatika latinskoga jezika. Zagreb: Matica hrvatska.

Mayrhofer Manfred 1981: Nach hundert Jahren. Ferdinand de Saussures Frühwerk und seine Rezeption durch die heutige Indogermanistik. Mit einem Beitrag von Ronald Zwanziger (Sitzungsberichte der Heidelberger Akademie der Wissenschaften, Philosophisch-Historische Klasse 1981,8). Heidelberg: Winter.

46

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

M a y r h o f e r M a n f r e d 1986 [²2012]: Indo-European Grammar. Volume I/2: Phonology: Segmental Phonology of Indo-European (Indo-European Library: Series 1, Textbooks and Manuals: Indo-European Grammar Vol. 1). Heidelberg: Winter.

M a y r h o f e r M a n f r e d 2005: Die Fortsetzung der indogermanischen Laryngale im IndoIranischen (Sitzungsberichte der Österreichischen Akademie der Wissenschaften, Philosophisch-Historische Klasse 730). Wien: Österreichische Akademie der Wissenschaften.

M e i e r - B r ü g g e r M i c h a e l 2000: Indogermanische Sprachwissenschaft. 7., völlig neubearb. Aufl. der früheren Darst. von Hans Krahe. Unter Mitarbeit von Matthias Fritz und Manfred Mayrhofer (De-Gruyter-Studienbuch). Berlin – New York: de Gruyter.

M e i e r - B r ü g g e r M i c h a e l 2002: Indogermanische Sprachwissenschaft. 8., überarb. u. erg. Aufl. unter Mitarbeit von Matthias Fritz und Manfred Mayrhofer (De-Gruyter-Studienbuch). Berlin – New York: de Gruyter.

M e i e r - B r ü g g e r M i c h a e l 2010: Indogermanische Sprachwissenschaft. 9., durchges. u. erg. Aufl. unter Mitarbeit von Matthias Fritz und Manfred Mayrhofer (De-Gruyter-Studienbuch). Berlin – New York: de Gruyter.

M e i s e r G e r h a r d 1998: Historische Laut- und Formenlehre der lateinischen Sprache. Wiesbaden: Wissenschaftliche Buchgesellschaft.

M i k u l ė n i e n ė D a n g u o l ė 2005: Cirkumflexinė metatonija lietuvių kalbos vardažodiniuose daiktavardžiuose ir jos kilmė. Vilnius: Lietuvių kalbos institutas.

M o t t a u s c h K a r l - H e i n z 2011: Der Nominalakzent im Frühurgermanischen. Konstanten und Neuerungen (PHILOLOGIA: Sprachwissenschaftliche Forschungsergebnisse 159). Hamburg: Kovač.

M ü l l e r S t e f a n 2007: Zum Germanischen aus laryngaltheoretischer Sicht. Mit einer Einführung in die Grundlagen der Laryngaltheorie (Studia Linguistica Germanica 88). Berlin – New York: de Gruyter.

N e r i S e r g i o, Z i e g l e r S a b i n e 2012: Horde Nöss. Contributions to an Etymological Dictionary of the Thuringian Dialects. Bremen: Hempen.

NIL – W o d t k o D a g m a r S., I r s l i n g e r B r i t t a, S c h n e i d e r C a r o l i n: Nouns in the Indo-European Lexicon. Heidelberg: Winter, 2008.

N ü b l i n g D a m a r i s, F a h l b u s c h F a b i a n, H e u s e r R i t a 2012: Names. An Introduction to Onomastics. Tübingen: Narr.

PKEŽ – M a ž i u l i s V y t a u t a s: Etymological Dictionary of the Prussian Language. 1: A–H. Vilnius: Mokslas 1988. 2: I–K. Vilnius: Mokslo ir enciklopedijų leidykla, 1993. 3: L–P. Vilnius: Mokslo ir enciklopedijų leidykla, 1996. 4: R–Z. Vilnius: Mokslo ir enciklopedijų leidykla, 1997.

Straipsniai / Articles

47

HARALD BICHlMEIER

PJ – Toporov Vladimir Nikolaevič, Ivanov Vjačeslav Vsevolodovič: Prussian Language. 1: A–D. Moscow: Nauka, 1975. 2: E–H. Moscow: Nauka, 1979. 3: I. Moscow: Nauka, 1980. 4: K–L. Moscow: Nauka, 1984. 5: L. Moscow: Nauka, 1990.

Ringe Don 2006: A Linguistic History of English. Volume I: From Proto-Indo-European to Proto-Germanic. Oxford: Oxford University Press.

Rix Helmut 1976: Historische Grammatik des Griechischen: Laut- und Formenlehre. Wiesbaden: Wissenschaftliche Buchgesellschaft.

Rix Helmut 1992: Historische Grammatik des Griechischen: Laut- und Formenlehre. 2., korr. Aufl. Wiesbaden: Wissenschaftliche Buchgesellschaft.

SEJL – Smoczyński Wojciech: Etymological Dictionary of the Lithuanian Language. 2 vols. Vilnius, 2007.

Sitzmann Alexander, Grünzweig Friedrich E. 2008: Die altgermanischen Ethnonyme. Ein Handbuch zu ihrer Etymologie (unter Benutzung einer Bibliographie von Robert Nedoma). Herausgegeben von Hermann Reichert. Wien: Fassbaender.

Smoczyński Wojciech 2006: Laringalų teorija ir lietuvių kalba. Vilnius: Lietuvių kalbos institutas.

Smoczyński Wojciech 2008: Uzupełnienia do “Słownika etymologicznego języka litewskiego”. Część I. – Acta Linguistica Lithuanica 58, 53–151.

Smoczyński Wojciech 2009: Uzupełnienia do ‘Słownika etymologicznego języka litewskiego’. Część II. – Acta Linguistica Lithuanica 60, 17–127.

Szemerényi Oswald 1990: Einführung in die vergleichende Sprachwissenschaft. 4. durchges. Aufl. Wiesbaden: Wissenschaftliche Buchgesellschaft.

Udolph Jürgen 1990: Die Stellung der Gewässernamen Polens innerhalb der alteuropäischen Hydronymie (Beiträge zur Namenforschung: Beiheft Neue Folge 31). Heidelberg: Winter.

Udolph Jürgen 1994: Namenkundliche Studien zum Germanenproblem (Reallexikon der Germanischen Altertumskunde: Ergänzungsband 9). Berlin – New York: de Gruyter.

Udolph Jürgen 2011: Alteuropa, Iller, Alster, Elster und alḗti. – Acta Linguistica Lithuanica 62–63, 161–176.

Vennemann (gen. Nierfeld) Theo 1999: Volksetymologie und Ortsnamenforschung. – Beiträge zur Namenforschung, Neue Folge 34, 269–322.

Wagner Norbert 1994/95: Review of Udolph 1994. – Beiträge zur Namenforschung, Neue Folge 31, 184–193.

48

Acta Linguistica Lithuanica LXVIII

Zum sprachwissenschaftlichen Niveau der Forschungen zur Alteuropäischen Hydronymie – eine Erwiderung auf eine Polemik

Weiss Michael L. 2009: Outline of the historical and comparative grammar of Latin. Ann Arbor – New York: Beach Stove Press.

Ziegler Arne, Windberger-Heidenkummer Erika (Hgg.) 2011: Methoden der Namenforschung. Methodologie, Methodik und Praxis (= Akten der 6. Tagung des Arbeitskreises für bayerisch-österreichische Namenforschung (ABÖN), Graz, 12.–15.5.2010). Berlin: Akademie Verlag.

Zimmer Stefan 2006: Treverer, I. Philologisch, § 1: Namenkundlich. In: Reallexikon der Germanischen Altertumskunde 31 (2006), 173f.

On the scholarly level of research into Old European hydronymy—a response to a polemic

ABSTRACT

This article is a response to Jürgen Udolph’s discussion paper (Acta Linguistica Lithuanica 62–63, 2011, 161–176), in which he takes issue with the author of an article published in Blätter für oberdeutsche Namenforschung 46, 2009 (2010), 3–63. In his article, Udolph critically evaluates and refutes some of Bichlmeier’s claims. Nevertheless, his opponent’s objections are not sufficiently well argued. Udolph himself misjudges processes that took place in the language, thereby revealing scant familiarity with works published in recent decades on Indo-European linguistics and its research methods. The following shortcomings can be identified in his article:

1) The phonological system of the Proto-Indo-European language was characterized by laryngeal phonemes. Reconstruction models that omit these phonological elements are outdated and should therefore be regarded as irrational.

2) On this basis, one may argue that the reconstruction of the roots “*el-/ol-” relies on an outdated methodology. Moreover, the Baltic-language examples cited by Udolph do not support a structural model of the Proto-Indo-European root based on that methodology.

3) Udolph’s claim that the Proto-Indo-European language did not have the sonant r is casuistic.

4) The principled disregard for the effects of ablaut sequences on the semantic differentiation of hydronyms should be regarded as a scientific anachronism.

5) The attempt to derive the hydronym Iller from *Elira, argued on the basis of Krahe’s etymological considerations, is unsuccessful. The etiology of the geminate in this word, attested in sources as early as the ninth century, which Krahe and Udolph attempted to explain, remains unclear.

6) Udolph’s attempt to establish the origin of Ulster, Alster and other river names in this group likewise relies on the aforementioned erroneous methodology for explaining the development of sounds from Proto-Indo-European into Proto-Germanic and the individual Germanic languages.

7) It should be noted that Udolph’s conclusions are contradictory with regard to the chronology of West Germanic rhotacism.

8) Despite Udolph’s negative assessment of the new explanation of the origin of the Proto-Germanic suffix *–stra-, it is traditionally treated by Krahe, Meid, and Udolph as a primary affix. Such a position chang-

Straipsniai / Articles

49

HARALD BICHlMEIER

…: this working suffix arose through the convergence of at least two distinct elements of the analysis; it has been proposed that appellative derivatives with this affix be derived from s-stem forms or PIE roots with the endings *-s- and/or *-d/t-, and from the PIE instrumental suffix instrumenti *-tro-. Alongside such formations, Proto-Germanic suffix *-stra- may have arisen in river names with a secondary n. agentis suffix *-t(e)r- + *-o- thematicization, characteristic of PIE roots with the endings *-s- and/or *-d/t-, or of substantivized secondary PIE adjectives formed with the suffix *-ro- from old s-stem forms.

On the basis of arguments 1–7, it can be asserted that Udolph disregards the usual research methods of modern comparative historical linguistics, or Indo-European studies.

In conclusion, it can be inferred that Udolph’s earlier analysis of the very oldest European hydronyms (attributed to Old European, Proto-Germanic, or Proto-Celtic inheritance) must be verified using the methods of modern linguistics. This researcher’s conclusions are based on outdated assumptions and a methodology that is entirely unpromising when assessed from the perspective of contemporary Indo-European studies. Only after all works in this field have been examined will it be possible to determine how much lasting value they have.

Received on 31 May 2012.

HARALD BICHlMEIER

Sächsische Akademie der Wissenschaften

Arbeitsstelle Jena: Etymologisches Wörterbuch des Althochdeutschen

Zwätzengasse 12a, D-07743 Jena

[email protected]

Martin-Luther-Universität Halle-Wittenberg

Orientalisches Institut, Seminar für Indogermanistik und Allgemeine Sprachwissenschaft

Heinrich-und-Thomas-Mann-Str. 26, D-06108 Halle (Saale)

[email protected]

50

Acta Linguistica Lithuanica LXVIII