JÜRGEN UDOLPH

Center for Name Studies – Leipzig

Göttingen Academy of Sciences

Research areas: historical linguistics, hydronymy, place names, family names.

HOMELAND AND EXPANSION OF INDO-EUROPEAN TRIBES IN THE LIGHT OF ONOMASTIC RESEARCH

Homeland and spread of Indo-European tribes in the light of onomastic research

ABSTRACT

If one wishes to describe the homeland and expansion of the Indo-Europeans, it is first necessary to determine the original settlement areas of the individual Indo-European tribes. On the basis of hydronyms and place names, as well as the good and fertile soils that make continuous settlement possible, the Slavs can be located on the northern slope of the Carpathians, the Germanic peoples north of the Harz, and the Celts on the western edge of the Alps. Hydronyms belonging to Old European hydronymy further show that ancient connections with the Baltic region and the Baltic languages can be demonstrated from almost all settlement areas. The Baltic region is thus the center of Indo-European names, and nothing speaks against regarding it as the point of origin and homeland of Indo-European expansion. A homeland outside Old European hydronymy, whether in southern Russia, Asia Minor, or the Caucasus, is ruled out.

KEYWORDS: Indo-Europeans, hydronyms, Slavs, Germanic peoples, Celts, Balts, Old European hydronymy, original homeland.

ANNOTATION

If one wishes to characterize the home and expansion of the Indo-European people, it is first and foremost essential to identify the original areas of settlement of the specific Indo-European tribes. On the basis of river and place names and also because of good and

Stripsniai / Articles

173

JÜRGEN UDOLPH

fertile soils which enabled a continuous settlement, one can localize Slavic settlement on the northern slope of the Carpathian Mountains, Germanic settlement north of the Harz and Celtic settlement on the western edge of the Alps. In addition, river names which can be attributed to the Old European hydronymy show that old relations to the Baltic countries and the Baltic languages can be attested. Thus, the Baltic countries are the center of Indo-European names and not much opposes the idea of this area as the exit area and home of Indo-European expansion. A home outside of the Old European hydronymy, be it in Southern Russia, in Asia Minor or in the Caucasus Mountains, can be ruled out.

KEYWORDS: Indo-Europeans, Hydronyms, Slavs, Germanic peoples, Celts, Balts, Old European Hydronymy, Original homelands.

INTRODUCTION

If one asks about the homeland and expansion of Indo-European tribes, one is in every case asking about settlement, colonization, and migratory movements. Both settlement and migration take place on the ground, on the earth. By general definition, a settlement is a place where people live together, usually in buildings or building-like structures, for the purpose of living and working. When people live together in such structures, the need to orient themselves in their environment compels them to name it. Naming serves a functional purpose, and there are noteworthy studies, such as those by Horst M. Müller and Marta Kutas (1997), which conclude that the origins of proper names predate those of language.

From another perspective, Gottfried Wilhelm Leibniz and Jacob Grimm recognized this early on. From a more general perspective, J. Grimm writes: “There is a more living testimony about peoples than bones, weapons, and graves, and that is their languages” (1848: 5); and elsewhere: “Without proper names, every source of the German language would have dried up in the earliest centuries; indeed, the oldest evidence we have for it at all lies precisely in them [...]. For this very reason, investigating them sheds light on the language, customs, and history of our ancestors” (1848: 133). Specifically concerning names of bodies of water, G.W. Leibniz observes: “Et je dis en passant que les noms de rivieres, étant ordinairement ceux de la plus vieux langage et les anciens habitans, c’est pourquoy ils meritent une recherche particulaire” (1882: 264).

When one asks about the homeland and divergence of the Indo-European languages, in my opinion it is essential to take geographical names into account. The following points in particular are what give

174

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

names their particular significance for questions concerning early settlements and migrations:

1) Place names arise only through prolonged settlement. In general, neither an immediate occurrence nor a particular event leads to the establishment of a name, but rather a gradually growing, tacitly developing agreement that takes shape only slowly among the speakers of a dialect (namely, the dialect spoken at the place when the name arose). This has several consequences.

2) A brief conquest and occupation of a country do not lead to the formation of place names in the language of the conquerors and occupiers. Conquering or nomadic peoples such as the Huns and Mongols leave behind names only very rarely, and virtually never names of bodies of water.

3) This applies even to the Roman occupation and settlement of southern and western Germany. Although several place names can be explained from Latin, such as Köln and Koblenz (Confluentes), Latin names of bodies of water are scarcely found.

4) Names of bodies of water are long-lived and tenacious. They especially exhibit a phenomenon of the utmost importance to the question of which peoples once lived in a particular area: when a population changes, such names generally do not disappear, but are altered through the language of the new settlers and adapted to the new idiom. They often persist with extraordinary tenacity; significantly, it is usually the people living in settled agricultural communities who pass the names on.

5) Because of their longevity, geographical names often contain appellatives that disappeared from living languages long ago. This is a phenomenon known from the history of all languages; however, compared with appellatives themselves, evidence from names offers decisive advantages, since it makes it possible to determine the original distribution area of a word much more precisely.

6) It should be assumed that names of bodies of water also arose during the period when Indo-European was formed and during the period of its development and expansion. Tracing these names is a task of Indo-European studies.

Another task of Indo-European studies, as is well known, is to attempt, by comparing corresponding words, forms, and syntagms in the individual Indo-European languages, to establish as persuasive a reconstruction as possible of a Proto-Indo-European foundation from the perspective of linguistic history. One proceeds—and there is no other way—from the individual Indo-European languages.

¹ Repeatedly emphasized by Hans Krahe and Wolfgang P. Schmid.

Straipsniai / Articles

175

JÜRGEN UDOLPH

However, this is precisely the approach one must take when setting out to identify the presumed oldest settlement areas of Indo-European tribes: beginning with a look at the hydronymic layers of individual Indo-European languages, one must try to trace the path from there back to earlier name layers. Since geographical names can also be analyzed as evidence of migrations (Udolph 2008), this method certainly promises to lead to progress. I will begin with Slavic and then turn to Germanic, Baltic, and Celtic.

THE DISTRIBUTION OF INDIVIDUAL INDO-EUROPEAN LANGUAGES IN THE LIGHT OF NAMES

1. SLAVIC

Efforts to learn more about the oldest settlement areas of the Slavs with the help of geographical names are still based essentially on Max Vasmer’s “exclusion method.” He succeeded in delimiting the original Slavic territory by eliminating all regions in which place names reveal a Ugric, Iranian, or Baltic substrate (Vasmer 1971: 101–202, 203–249, 251–534), later supplemented by Vladimir N. Toporov, Oleg N. Trubačev (1962), and O. N. Trubačev (1968). Thus M. Vasmer was able to show, using the exclusion method that remains valid to this day, that Slavic must probably have developed in the area south of the Pripet’ and west of the Dnieper.

We owe the decisive advance beyond M. Vasmer’s method to the research of H. Krahe, although he had hardly concerned himself with Slavic name material. But he recognized the fundamental principles underlying the naming of bodies of water—and this insight applies not only to the pre-language names that H. Krahe preferred to study, but to hydronyms in general and in all languages: “In terms of semasiology and etymology, the primordial and undoubtedly oldest layer of names derives from so-called ‘water words,’ that is, designations for ‘(flowing) water,’ ‘spring,’ ‘brook,’ ‘river’ (or ‘to flow’), ‘course (of water)’ (or ‘to run,’ and the like, with countless finer and

176

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

subtle shades of meaning, which were abundantly available to early humans through their close observation of nature [...]” (1964: 4). In addition, H. Krahe writes on the question of how a people’s earlier homelands can be determined more precisely: “For wherever place names in a particular language are found in greater numbers, that language itself must also have been spoken there, and members of the people who spoke it must have lived there” (1949–1950: 25).

M A P 1. Russia in the last centuries before Christ (M. Vasmer²)

Applying these ideas to the question of the homeland and expansion of the Slavic tribes leads to clear conclusions and, in my opinion,

² Vasmer 1971: 96–97.

Straipsniai / Articles

177

JÜRGEN UDOLPH

definitive answers. I have already published the following discussions in several places³; here, therefore, I offer only a selection from extensive collections and mappings.

First, a further comment on the question of whether anything can be learned from an archaeological perspective about the homeland and expansion of the Slavs. This problem is connected with another question, namely whether one can speak of Slavic tribes already around the time of Christ’s birth. How can they be recognized, or how does one define “Slavs” and “Slavic” for this period? If I have understood Sebasian Brather’s study (2004) correctly, then from an archaeological perspective it is extremely difficult, if not impossible, to identify a people or an ethnic group by means of material culture, or to locate or describe their places of residence and migrations. From the perspective of linguistics, too, one must bear in mind that it is in fact impossible to describe where Slavic tribes settled before the beginning of the Common Era. This is because before the birth of Christ there was no language that we could call “Slavic.” The “Slavic” language—that is, the shared linguistic features of a speech community that emerged from a presumably Indo-Germanic (or Indo-European) dialect area—first had to develop. This was a process that certainly lasted several centuries. I therefore view with great skepticism attempts to establish something about a people from an archaeological perspective, with the help of genetic research, ethnology, or historiography, when that people can ultimately be defined only by means of language. In brief, from my point of view¹, one can approach the question of the homeland and expansion of Slavic tribes fruitfully only by taking linguistic, language-historical, and above all onomastic arguments into consideration. Put differently: the homeland of Slavic tribes should be sought where linguistic similarities developed over a period of centuries within a dialect area of Indo-European speakers, similarities that were peculiar to that dialect area alone. In this I follow an old view of M. Vasmer, who was convinced that “the Slavic original homeland can be established primarily [...] through thorough research into loanwords and place names, and the fullest possible consideration of all ancient historical and geographical sources” (1971: 71 [reprint of 1930!]).

Returning to H. Krahe’s “water words” from the Slavic perspective. How, then, can one identify, within the former and present-day Slavic settlement area—which covered and covers a vast territory—the particular area

³ Most recently, Udolph 2016, 2016a.

178

Acta Linguistica Lithuanica LXXVI

The Homeland and Expansion of Indo-European Tribes in the Light of Onomastic Research

MAP 2. Synoptic summary of 37 distribution maps of Slavic water words (after Udolph 1979: 322, map 40)

that might have been the starting point of the Slavic expansion, whose scale already astonished Byzantine historians? The great expansion of the Slavs, observed by specialists and contemporaries alike (according to a Byzantine source, 100,000 Slavs were to have invaded Thrace and Illyria in 577), must have originated somewhere. This could only have happened north of the Carpathians, that is, in an area that was not readily known to ancient sources.

A synoptic mapping of four dozen Slavic water words and their occurrences in hydronyms, place names, and field names (Map 2) makes clear that the area can be delimited fairly well: since the area south of the Carpathians and the Beskids is excluded by general agreement⁴, the concentrations in Ukraine and Poland are of particular significance.

⁴ On O.N. Trubačev’s thesis that Pannonia should also be considered, see Udolph 1988.

Straipsniai / Articles

179

JÜRGEN UDOLPH

There is a widespread view that Slavic hydronymy displays a certain uniformity. Since the structure and formation of Slavic water names have developed according to the same criteria as the vocabulary, this is entirely understandable, for derivation is highly prevalent in these languages, that is, formation with suffixes. The following elements are widespread in water names (for details, see Udolph 1979: 539–599): formations with *-(j)-ač-, extended with the help of -ov- or -in-, cf. Vod-ač, Il-ač, Gnil-ov-ač, Il-in-jač; *-(j)ak, also occurring as -ъn-ak-, -in-ak-, etc., is found, for example, in Solot-in-ak, Gnil-ak, Bagn-ak, Glin-ak, Vod-n-jak, -at-, often extended as -ov-at-, e.g. in Il-ov-at, Schyl-ov-at, Hlin-ov-at; -ica is typical of Slavic hydronymy, often also extended as -av-ica, -ov-ica, -in-ica, -sk-ica, and formations with -(ь)n-ica are numerous, cf. Bar-ica, Glin-ica, Kal-ica, Vod-ica, Vir-ica, Topolica, Blat-n-ica, Dubr-ov-n-ica, Lis-n-ica, Izvor-st-ica; -ik- is also relatively frequent, sometimes extended as -(ь)n-ik-, -ov-ik-, e.g. in Brn-ik, Bar-n-ik, Brus-n-ik, Glin-ik, Il-n-ik, Jam-n-ik, Lip-n-ik, Il-ov-ik; formations with -in-, -ina-, -ino- are mostly adjectival in origin, as in Berlin, Schwerin, Genthin, and are also found, for example, in water names: Ozer-in, Bolot-in, Vod-in-a, Bar-n-in, Bab-in-a, Dobr-in, Radot-in-a, Slatina, extended with -ev- and -ov- in Bobr-ov-a, Buk-ov-a, Dub-ov-a, Kalin-ov-a, Lip-ov-a, Vugr-in-ov-o, etc. Water names with -isk- occur almost exclusively in West Slavic: Wodz-isk-a, Bagn-isk-a, Zdro-isk-o; elsewhere *-iskjo predominates, appearing as -išč- in East Slavic and otherwise also as -išt-: Ples-išče, Zleb-išče, Ric-išče, Gnoj-išča, Bar-ište, Lokv-išta; -ev-/-ov- also occasionally occurs as a toponymic formative, e.g. in Duna-ev, Il-ov, Borl-ov, Sopot-ov-i, Bagn-iev-o, etc. *-ьc- is very frequent, sometimes extended with -in-, -ov-, -a- and other elements, cf. Izvor-ec, Strunin-ec, Jezer-ec, Blat-ce, Bar-in-ec, Mor-in-cy, Il-ov-in-ce, Hlin-ov-ec, Strug-ov-ec, Brnj-av-ac; the same applies to -ъk-/-ьk-, cf. Potoc-ek, Vir-ok, Dunaj-ek, Ozer-ok, Bagn-ko, Bolot-ki, Vod-n-ev-ka, Il-av-ka, Ozer-ov-ka, Bar-ov-ka, Sigl-in-ka, Zvor-n-ky, Hnoj-en-ki, Klucz-ev-at-ka, Gnil-ič-ko, Kal-n-ic-ki, Reč-ul-ka; adjectival formations with *-ъn-, -na, -no are also frequent: Bar-na, Brez-na, Les-na, Sol-na, Sopot-na, Svib-no, Slatin-ny, Rzec-ny, Hnoj-ny, Il-na, Glin-na, Kal-ne, Zdro-no.

If one looks for formations with these elements in the Slavic water-name corpus, one can already observe different distributions among the water words formed with them that are common and widely known in the Slavic languages, although centers also emerge that require interpretation. Here are a few examples:

To filter out older types from the wealth of Slavic water names, it is possible, as has become clear in recent years (Udolph 1997,

180

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 3. Distribution of place names containing Slavic reka ‘river’. Filled black circles indicate direct formations; white circles, suffixed formations; symbol size is graduated according to water names, place names, and field names (after Udolph 1979: 257, map 26).

1998), to identify several criteria characteristic of old and older formations.

1) They contain archaic suffixes that are no longer productive today.

2) They are older when derived from appellatives that are no longer productive today.

3) They go back to different layers of derivation, whose areas sometimes overlap.

4) They are derived with Slavic suffixes from pre-Slavic, i.e. Old European, hydronyms.

Taking as a point of departure the present-day distribution of Slavic languages and the distribution maps of Slavic water names shown above (maps 2–4), parts of Russia and Belarus, Ukraine, Slovakia, and Poland are of particular interest.

Straipsniai / Articles

181

JÜRGEN UDOLPH

MAP 4. Distribution of place names containing the Slavic pok ‘stream’. Black filled circles indicate direct formations; white circles indicate suffixed formations; symbol size is graded according to hydronyms, place names, and field names (after Udolph 1979: 251, map 25).

For the individual points, I offer further material below, including maps.

1) The names contain archaic suffixes that are no longer productive today. First to be mentioned here is the typical hydronymic suffix -ava, -awa, which we encounter, for example, in Vir-ava, Vod-ava, Il-ava, Glin-awa, Breg-ava, Lut-awa, Mor-ava, Orl-awa and other names. It is probably the characteristic old Slavic suffix in hydronymy and has clear connections with pre-language-specific, Indo-European name formation (Lupawa, Mora-March-Mähren, Orava, Widawa). Occasionally, a reliable distinction is not possible (for details of the material collection, see Udolph 1979: 555–558).

A map of these names (see map 5) shows that the distribution of these old names, too, covers chiefly the area north of the Carpathians—precisely the territory already brought to attention by pok, reka (maps 2 and 3).

182

Acta Linguistica Lithuanica LXXVI

Homeland and Expansion of Indo-European Tribes in the Light of Name Research

MAP 5. Distribution of Slavic hydronyms formed with the suffix -ava (after Udolph 1997: 51, map 3).

Highly archaic Slavic words and names include the so-called -ū-stems, which usually appear as -ev- or -va (not as -ava). From the corpus discussed in detail elsewhere (Udolph 1997: 35–47), I mention here only Buga, Moka, Golva, Mosta, Polka, Branev/Brnev, Maia Padava/ Malapane. It should be noted that these names can often be associated with Slavic words, though some cannot—that is, no connection with the Slavic lexicon can be established. This means that some of them are likely to be older. This is also supported by the fact that formations with very similar suffixes, such as -uv-, -iv- and -(i)uvē, -(i)uvis, occur in the Baltic hydronymic corpus; I will mention here only, very briefly, Daug-uvā, Lank-uvā, Alg-uvā, Ān-uvā, Gārd-uvā, Lat-uvā, Mīt-uvā, Ring-uvā, Týt-uvā, Vad-uvā, Vārd-uvā, Gil-uvē, Aud-uvīs, Med-uvīs, Dit-uvā.

Straisnia / Articles

183

JÜRGEN U DOLPH

MAP 6. Distribution of Slavic hydronyms and place names; ● = so-called (archaic) -ū-stems in geographical names; ○ = Slavic Ponikla, Ponikva, Po-nikiew and similar names, ‘disappearing, underground river’.

The distribution of the names (the Baltic ones are absent) shows (map 6) that once again a belt extending from west to east reaches from central Poland via southern Poland to the area east of Kyiv, without crossing the Carpathians to the south.

I have also included on the map names derived from Slavic ponik, ponikva ‘disappearing river, underground watercourse’ (discussed in detail in Udolph 1979: 239–245). Although these reflect a -ū-stem, they are nevertheless of considerably more recent origin. The map also documents this: the archaic -ū-formations (some of them etymologically transparent) lie at the center, whereas ponik(va) is found notably in areas reached by Slavs at a later date, such as Slovenia (where it is a term for the karst landscape), central and western Poland, and Belarus. Once again, the area north of the Carpathians proves to have been an old Slavic settlement area.

184

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 7

MAP 8

MAP 7. Distribution of place names containing Slavic bagno^' ‘swamp, marsh’; black filled circles = direct formations; white circles = suffixed formations; symbol sizes vary according to hydronyms, place names, and field names (after Udolph 1979: 334, map 42).

2) Slavic hydronyms are of greater antiquity when derived from appellatives that are no longer productive today. In other words, this means that the language from which the corresponding hydronyms were formed must still have possessed the word concealed in the names. These are therefore unquestionably older Slavic names. I have selected a few from the numerous cases available.

The distribution of names such as Bagno, Bahnec, Bagie-nice, Bagienek, as well as the family name Baginski, which is also attested in Germany, is striking; these names are related to Ukrainian bahno, Polish bagno ‘swamp, bog, marsh’, etc. The word has long been absent from South Slavic; it seems to have disappeared from this area early on (for details, the vocabulary in the Slavic languages, and the names derived from it, see Udolph 1979: 324–336). As Map 7 shows, corresponding names are found above all in the West Slavic area.

All the more remarkable is its spread to the southeast, clearly following the northern and eastern slopes of the Carpathians (cf. Map 8), until the word disappears from the active vocabulary, roughly in what is now southern Romania, and the last names formed from it have left their traces at the Iron Gates. This expansion clearly indicates one of the routes by which the later South Slavs entered the Balkans.

Another important observation, decisive for the question of the ancient homelands of Slavic tribes, is connected with the fact that the

Straipsniai / Articles

185

JÜRGEN UDOLPH

Slavic-speaking area is today divided by Austria, Hungary, and Romania: for centuries there have been no close contacts between the West and East Slavs in the north and the South Slavs in the south. The separation of the two Slavic settlement areas naturally led to the development of South Slavic features that remained unknown to the Slavic languages north of the Carpathians. There are cases in which this discrepancy is clearly apparent in the vocabulary, while the name stock presents a different picture.

There are appellatives, including words for bodies of water, that are peculiar only to South Slavic but—and this is the crucial point—also occur in names north of the Carpathians. Before attempting an interpretation of this phenomenon, I present the relevant material.

The long-disputed proto-form of the Slavic word family represented by Old Serbian brna ‘mud, earth’, Bulgarian Church Slavonic brъnije ‘mud, clay’, Old Church Slavonic brъna ‘mud’, Slovenian brn ‘river mud’, etc., is resolved conclusively when the onomastic material is taken into account (for details, see Udolph 1979: 499–514; Udolph 1987: 473–479). A reconstruction *bъrn- is often proposed, but West Slavic and especially East Slavic hydronyms such as Brynica, Brenica, Branica and Bronica, Bronica, Brono argue against it. In addition, alongside the aforementioned reconstruction *bъrn-, the name stock north of the Carpathians also attests the ablaut variant *< *brъn- *bъyn- < *brъn- (Brynica, Brynówka, Brynec). Slavic *bъyn- requires a reconstruction *b(h)rūn- and readily corresponds to Germanic *hrūn- in Low German brūn-, High German braun.

This example is significant in two respects. First, it makes clear that this is a South Slavic word group which, although absent from the Slavic vocabulary north of the Carpathians, has nevertheless left clear traces there in the name stock. There are two possible explanations. One may assume that the word family developed separately in South Slavic and that South Slavic groups then moved north and left their South Slavic features there in the names. This is an exceedingly complicated assumption, especially given that the brn-, bron-, bryn- names are also widespread north of the Carpathians. The second possibility is to take the existence of names north of the Carpathians as evidence that the underlying words were known in the vocabulary of the Slavs settled there and that the names were formed from them. One can only conclude from this: the names attested north of the Carpathians do not originate in any single Slavic language—not even South Slavic—but

186

Acta Linguistica Lithuanica LXXVI

Homeland and Expansion of Indo-European Tribes in the Light of Name Studies

MAP 9. Distribution of Slavic names containing brn- < *brъn- and bryn- < bьrъn- (‘swamp, morass, mud’); the symbols indicate the different developments in the modern Slavic languages (brn-, brьn-, bron-, bryn-), with their sizes graded according to hydronyms, place names, and field names. The outline marks the area in which the long-grade variant brn- occurs (after Udolph 1997: 58, Map 8).

of a precursor of all Slavic languages, that is, Proto-Slavic or Common Slavic itself.

This is convincingly confirmed by the fact that the forms *brъn- and *bьrn- are reflexes of an ancient ablaut (further examples follow below), which likewise must have originated only in an early stage of the Slavic language.

If we now also take into account the distribution of the names and recognize that both ablaut variants occur side by side in the name corpus only in a specific area of the Slavic languages, we gain a further secure argument for the conclusion that here—roughly speaking, in southern Poland and

Straipsniai / Articles

187

JÜRGEN UDOLPH

western Ukraine (in Bohemia, Moravia, Slovakia, and Slovenia only one variant is found)—the homeland and point of departure of Slavic expansion must be sought.

3) Archaic Slavic hydronyms can also be identified by tracing the evidence of ablaut. Ablaut phenomena occur in almost all Indo-European languages; I mention here only sing – sang – sung, offer – offered – offered; English sing – sang – sung; Lithuanian žìas ‘green’ – žõle ‘grass’; Greek lego – logos; Slavic tek- ‘flow’ – tok ‘stream’ (also in potok, see above).

However, traces of ablaut in Slavic—unlike, for example, in Germanic—can still be demonstrated only to a limited extent, so that only a few relics are to be expected in hydronymy as well. These, however, are of particular value, and their occurrence and distribution should therefore receive special attention. I have already discussed *brъn- > brъn-, *brъn- > byrn- above. Further important cases follow here.

For Proto-Slavic, a word *jbz-vorъ may be posited with a meaning such as ‘source, lowland, stream valley, spring, eddy’, which survives, among other forms, in Old Russian izvor ‘source’, Ukrainian izvir ‘small mountain stream’, and Serbian and Croatian izvor ‘source, spring, eddy’. This word contains an archaic compound, since Slavic does have the verb vьrěti ‘to gush’, but no independent *vor- (for the word and name material, see Udolph 1979: 163–170; see also Schmid 1994: 260f.).

The distribution of the names (Map 10) is therefore of particular significance. Given the ablaut inherited from the Indo-European precursor, the assumption that their occurrence in the Carpathian and Beskid region might represent offshoots of a later, South Slavic naming tradition is untenable. Rather, the names in the Dniester and San regions derive from a stage of the language that still knew the underlying appellative. This can only have been a precursor of the individual Slavic languages, in other words, a common or Proto-Slavic stratum of the language.

Moreover, compared with the distribution of the Bagno names (Map 7), this map shows that the later South Slavs not only used a route along the Carpathian arc (cf. Map 8), but also reached the southern Balkans by various routes across what is now Romania.

As with *brъn-/*byrn-, names are widespread that clearly bear traces of an ancient ablaut. These include Belarusian krynica ‘small lake; watercourse that emerges from the earth, source’, Ukrainian krynica ‘source’, and Polish krynica, krenica ‘source, well’, which continue a basic form *krъnica (for details, see Udolph 1994: 367–374). This is a so-called lengthened grade, which in Ukrainian (dialectal) krynycja, kenýc’a ‘source’,

188

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 10. Distribution of place names containing Slavic *jьz-vorb ‘spring, well’; circles and triangles symbolize the different development in the Slavic languages (izvor and zvor, respectively), with symbol sizes graded according to watercourse, place, and field names (after Udolph 1994: 169, map 13).

Old Polish krnica ‘river’, Slovenian krnica ‘deep place in the water, whirlpool, depth of a river’ has its short-vowel counterpart *krūn-. If we consider the occurrence of the krnica names, which cover a broad area, and compare it with the distribution of the short-vowel ablaut variant (Map 11), an area becomes apparent in which both variants occur side by side. The territory that thereby emerges must certainly be regarded as an ancient Slavic settlement area.

Stripsnai / Articles

189

JÜRGEN U DOLPH

MAP 11. Distribution of place names containing Slavic *krun > krьn- or *krūn > kyn (derived forms) ‘spring, well, stream’; filled black circles = direct formations; white circles = suffixed formations; symbol sizes graded according to watercourse, place, and field names. Hatching marks the area in which only the short-vowel variant *krŭn-/krьn- occurs (after Udolph 194: 169, map 13).

Versuche, die Ethnogenese des Slavischen in das Oka-Gebiet (Gołąb 1992), nach Asien (Kustmann 1996) oder auf den Balkan (Trubačev 1983, dagegen Udolph 1988, 1999) zu verlegen, müssen an diesen Verbreitungen scheitern. Es wäre nötig, sich intensiver mit diesen Fakten auseinander zu setzen, zumal sich ähnliche Erscheinungen auch für die Frage nach Germanenheimat und -expansion nachweisen lassen (dazu s. unten). Ganz ähnlich liegt der nächste Fall.

Alongside the well-known Russian appellative grjaz ‘dirt, dung, mud’, which has counterparts in, among other forms, Belarusian hrjaz ‘soft spot on a road, swamp, dirt’, Ukrainian hrjaz ‘swamp, puddle, mud’, and Slovenian grẽ̂ ‘moor, mud’, and which presupposes a Proto-Slavic base *grěz-, Slavic also has the ablaut variant *grǫz-, for example in Ukrainian hruz ‘swamp, moor, bog’, Belarusian hruzá-la, hruzalo ‘dirty place, marshy spot’, and Polish grąz, -grzęzu ‘boggy swamp’ (for a detailed discussion, see Udolph 1994: 142–152). It should already be noted that South Slavic does not have the ablaut variant *grǫz-, and thus has no share in this Proto-Slavic ablaut variant.

190

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 12. Distribution of place names containing Slavic *gręz-/*grǫz- (ablaut forms) ‘mud, bog, swamp’; squares = *gręz-; triangles = *grǫz-; symbol sizes graded according to watercourse, place, and field names; hatching marks the area in which only the *grǫz- variant occurs (after Udolph 1997: 59, map 9).

The distribution in the names corresponds well to this (Map 12): the names are widely scattered, and particular productivity is observable in East Slavic, while South Slavic has a share only in the *gręz- variant.

This conclusively rules out a Slavic homeland in the Balkans (the issue here is Proto-Slavic ablaut variants, whose productivity and effects must be dated long before the incursion into the Balkans). Slavic could have developed only north of the Carpathians, on the basis of this evidence.

In summary, this is supported not only by the word pair discussed above, grązъ/hr̥z, but also compellingly by the previously discussed groups around izvor/vorъti, krynica and, above all, brъn-/brъn-, which is further established in the Proto-Slavic lexicon by its secure connection with a Germanic color word.

4) The discovery that beneath a single-language stratum of water names in Europe (whether Germanic, Celtic, Slavic, or Baltic is immaterial) there lies a network of pre-single-language = ancient European = Indo-European names yields new

Stripsni / Articles

191

JÜRGEN UDOLPH

Possibilities for determining the area in which an Indo-European language developed.

In Slavic territory, this can be recognized above all in a few but important cases: the issue concerns archaic Slavic suffixes attached to pre-Slavic, Indo-European roots, names, or bases. What is then important and decisive is where such names are located.

The largest river in Poland, the Wisła, German Weichsel, bears an unmistakably pre-Slavic name, however one may explain it (most recently Udolph 1990: 303–311; Babik 2001: 311–315; Bijak 2013: 34). For the question occupying us here and now, the San’s tributary Wisłok, about 220 km long, and the Vistula’s tributary Wisłoka, about 165 km long, are of considerable importance.

Sanok, a place on the San southwest of Przemyśl; Sanoka, a hydronym no longer known today, 148 per fluvium Sanoka, near the place Sanok, and with a diminutive suffix to -ok-, a tributary of the San called Sanoczek, are formed with the same suffix (for details, see Rymut, Majtán 1998: 222; Udolph 1990: 264–270). There is also certainty that the San is a pre-Slavic name. The differing views on its etymology do not alter this (most recently and briefly summarized in B. Czopek-Kopciuch 2016: 48, in: Nazwy miejscowe Polski 13, Kraków). The suffix variant -oč also occurs in Liwocz and Liwoczka, river names near Kraków; a mountain range in the Beskids is also mentioned by Długosz as Lywocz.

All the names lie in southern or southeastern Poland, precisely in the area that, through the Slavic words and names discussed so far, has clearly emerged as part of the Proto-Slavic settlement area. In the judgment of the Słownik Prasłowiański (1974: 92), the suffix -ok- represents a Proto-Slavic archaism. It occurs appellatively, for example, in swědok, snъbok, vidok, edok, irok, inok, among others, but its antiquity is also evident, inter alia, from its attachment to archaic athematic stems. It must be clearly emphasized that the existence of pre-Slavic, Old European names in the presumed old or oldest settlement area of Slavic tribes—as was often assumed in the past—does not argue against the supposition that this area was located there; rather, it is the necessary consequence of the fact that the individual Indo-European languages did not develop out of a vacuum, but arose on a broad Indo-European basis from a stratum of Old European names; indeed, one may say, they had to arise in this way.

Since -ok- is an archaic suffix, the names cited here can be assigned to an earlier stage. They are therefore associated with

192

Acta Linguistica Lithuanica LXXVI

Homeland and Expansion of Indo-European Tribes in the Light of Onomastics

MAP 13. Expansion of Slavic Tribes in the Light of Geographical Names (Draft)

with a high degree of probability, to be regarded as a link between Old European and Slavic hydronymy.

Summary – Homeland of the Slavic tribes

My summary of the investigation of hydronyms in relation to the question of the oldest settlement areas of Slavic tribes is as follows: the premises required for addressing the question from linguistic, philological, and onomastic perspectives are met by all the investigated place names and hydronyms with regard to a possible area in a very specific region: it is the pre-map area. In a simple mapping, I have summarized the results of the distribution of Slavic water words and hydronyms (Map 13). The core of the expansion is an area north of the maps, approximately between Kraków and Bukovina. It must, however, be stated very clearly that the boundaries are not as clear

Straipsniai / Articles

193

JÜRGEN UDOLPH

and distinct as the map might suggest. This is an attempt to present the core of the mappings in a simple form. Every mapping of a place name or hydronym varies; there are hardly two maps whose centers or peripheries coincide. Nevertheless, I consider this mapping useful. This is because there is another aspect that I have not yet discussed alongside archaeology, genetic research, linguistics and onomastics, as well as history: I mean soil science.

Soil science, soil quality and the distribution of geographical names

At the beginning of this article, I stressed that my contribution would be made primarily from linguistic and onomastic perspectives. In recent years, however—not only in the collection and interpretation of Slavic names—it has become increasingly clear that there is another aspect which apparently has nothing to do with geographical names, but which, to my knowledge, is of decisive importance for early settlements: soil quality.

As early as the 1920s and 1930s, attempts were made to link certain place names with variations in soil quality. In fact, publications devoted to this subject have appeared repeatedly ever since. I will not list the relevant literature here, mentioning only the work of Otto Schlüter (1952–1958). Recently, I have again come across statements making clear that a connection between soil quality and certain Slavic place-name types must be regarded as established. I drew attention to this emphatically in a review (Udolph 2015) and quote here a few sentences by Walter Wenzel: “In Upper Lusatia we were able to establish that these four [place-name] types occurred only in the central loess areas with the most fertile soils, where the immigrants had first settled [...]. The course of settlement depends to a decisive extent [also] on [...] soil quality, which can vary considerably even over short distances in Lower Lusatia.” And, even more explicitly, elsewhere: This can be “confirmed by concrete soil-quality figures from the Atlas zur Geschichte und Landeskunde von Sachsen [...]. If one compares the distribution of this name [...] with W. Stams’s soil map [(1998)], the causal connection between soil quality and name type is unmistakable” (Wenzel 2014). I have the map in question, and it is impressive to see how the level stretch of land, characterized by fertile soil, from Kamenz via Bautzen to Görlitz can be matched with the mapped distribution of old Slavic place names. In this, a role is also played

194

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 14. Loess distribution in eastern Poland and Ukraine. Mi. = Minsk, Ki. = Kyiv, dark gray = loess more than 5 m thick (after Haase et al. 2007).

whether the soils in question are easy or difficult to cultivate also plays an important role. Chance is ruled out here.

Let us try to apply these by no means new findings to the soils of Poland and Ukraine. Particularly helpful here is a “Loess Atlas” published in 2007 (Haase et al. 2007), from which I present an excerpt here.

It is no effort at all to bring the distribution of loess in the pre-map area into correspondence with the old Slavic hydronyms compiled in this article.

Of course, one must also take into account that loess soils are not always easy to cultivate. My cautious attempt to incorporate findings from the study of ancient settlement landscapes into this question led me to several studies, from which I would like to select just a few passages. As a layperson in the field of soil research, one is well advised to be very cautious here. But the following remarks largely support my thinking. Fritz Scheffer (1978: 16–17) writes: “Since the Neolithic, there has been a considerable narrowing in Central Europe of the areas then suitable for settlement purposes

Straipsniai / Articles

195

JÜRGEN UDOLPH

preferred scherzde areas [..] took place [..]. Up to the present, the heavy soils of the loess regions, the young ground-moraine landscapes, the marshes, and the upland areas have been regarded as more favorable for agricultural production than light soil types, since they contain larger reserves of plant nutrients [..].” And a few pages later Brunk Meyer (198: 119) argues that heavy, potassium-rich soils were preferred during the medieval periods of land clearance, despite the difficulty of working them.

Nevertheless, I would like to add here a related thought, which I draw from a discussion with Kirstin Casemir (Göttingen/Münster) about loess soils and the age of place names. A closer examination of the distribution of place names (e.g. in eastern Lower Saxony; relevant from an onomastic perspective is Casemir 2003) may lead to the conclusion that the oldest Germanic place names, that is, suffix formations, lay “not in the core areas of the loess depressions, unquestionably the oldest settlement areas, [..] but at their margins” (Casemir 2003: 40 after Müller 1952: 144). A similar view is expressed by Gerhard Obeck (cited after Casemir 1997: 49, note 12), according to whom the preference for poorer-quality soils at older settlements was connected with the settlers’ ‘technical’ capabilities in working the soil. The fertile but also heavy soils found at most -būtel settlements could only be cultivated with improved tools. For this reason, the better soils were initially scarcely used and were settled only at a later date.

If we apply this to the distribution of loess in western Ukraine and southeastern Poland, it is striking that the center of Old Slavic names lies in the area where the loess distribution gradually “frays out” (Fig. 14), that is, roughly in the region west of Kyiv, between Kraków in the west and Vinnytsia and Moldova in the east.

Briefly, the distribution of good soils coincides with that of archaic Slavic names. If this is correct, we can assume a cradle area—or, better, a core landscape—of Slavic settlement in the Pre-Carpathian region. The existence of pre-Slavic but Indo-European names and hydronyms whose structure indicates that they originated from an Indo-European base but subsequently developed Slavic characteristics, ¹can now—as already stated above—only be understood to mean that, over a process lasting a millennium, the language group we today call Slavic emerged here from an Indo-European dialect area.

196

Acta Linguistica Lithuanica xxvi

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

Since good soil may apparently have played a role in this, I would like to put forward the following consideration for discussion: good soils lead to better harvests, minimize overall mortality and child mortality, and result in population pressure, which can be alleviated only through the gradual expansion of settlement activity. The same may be observed in the question of the homeland and expansion of Germanic tribes, to which I shall turn shortly.

The intensive study of the geographical names of Eastern and Central Europe leads to the conclusion that Slavic must have developed from an Indo-European dialect in a relatively limited area between the upper Vistula and Bukovina (the Old European hydronymy and Balto-Germanic-Slavic correspondences play a role here); that there can have been no Balto-Slavic intermediate stage; that early, scarcely interrupted contacts with Baltic and Germanic tribes took place; and that, through strong expansion, the later homelands of the East, West, and South Slavic peoples were reached. Very similar expansions have been observed among Celtic and Germanic tribes, although these preceded the Slavic expansion.

2. HOMELAND AND EXPANSION OF GERMANIC SETTLERS AS INDICATED BY NAMES

The development of the Germanic language branch from an Indo-European precursor must probably be dated earlier than in the case of Slavic. By around 500 AD, clear differences in linguistic structures can already be discerned within Germanic, for example between East Germanic (Gothic), North Germanic (Runic, Proto-Norse), and West Germanic. In the view of most linguists, the emergence of Germanic as an independent branch, observable, among other things, in the effects of the First (Germanic) Sound Shift, may be dated to around 500 BC. In its effects, unlike, for example, those of the Second (High German) Sound Shift, it apparently affected all Germanic branches very uniformly and without major differences. I maintain that this process must have taken place in a relatively small geographical area, since otherwise considerably greater differences would certainly have arisen in the reflexes of the sound shift. Apparently, this was not the case. The question is whether the thesis that Germanic

Straipsniai / Articles

197

JÜRGEN UDOLPH

“Urheimat” must be located in a relatively small geographical area; this can be confirmed with the help of the most archaic Germanic settlement and watercourse name types.

When one studies Old Germanic place names and their formation, one is soon led to an old but still valid statement by Jacob Grimm (1826: 403). Almost 200 years ago, he had emphasized: “It is the unmistakable tendency of the later language to abandon derivations and replace them with compounds.” Put another way, derivations, i.e. suffixal formations, are generally older formations in Germanic place names than compounds or compound formations. A study devoted to the earlier settlements of Germanic tribes should therefore turn its attention above all—though not exclusively—to suffixal formations.

In a comprehensive study of watercourse, settlement, and field names of Germanic origin (Udolph 1994), I addressed this subject in its entirety; derivations soon became the focus of further contributions (Udolph 2004, 2010). Studies by Reinhold Möller (1992, 2000) should also be taken into account for Lower Saxony.

Collections and distribution maps of Germanic water words such as hor ‘swamp, marsh’. mar(sk) ‘inland lake, swamp’, Riede ‘stream, flowing water’, of suffixal formations with -ithi-, -ing-/-ung-, -st-, -str- and of place-name generics such as -hude ‘ford, landing place by the water’, *-sējanez/-sēt[j]ōz ‘settlers’, kot ‘settlement’, tun ‘town, settlement’, tie ‘assembly and court place’, sel(e) ‘dwelling, village, settlement’, klint ‘hill, slope’, wedel ‘ford’, and many others clearly show that Germanic developed continuously from a readily identifiable Indo-European substrate within the continental Germanic region—more precisely, in southern Lower Saxony, western Saxony-Anhalt, and parts of Thuringia.

To keep this contribution within bounds, I have selected only a few distribution maps and provided a brief commentary on them.

From this central region, early connections can be discerned both to the west (Westphalia, Flanders, northern France, England) and to the north (Denmark, Sweden, and farther on to Norway). Here are a few examples, first of the connections to the west.

a) germ. *fani/-ja ‘Sumpf, Moor’

An old term for ‘swamp, moor’, but also for ‘low-lying grassland’, occurs in the Germanic word family around Gothic fani ‘mud’; *fanja is usually reconstructed. There are hundreds of Fenn names; in Germany, for example, Ackerfenne, Fanhusen, Fehn, Fehnhusen, Venusberg in Bonn, as well as Venusbruch and Venushügel near Wernigerode, Vienenburg, recorded in 1306 as Vineburch. Also

198

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 15. Germ. *fanja in place names and field names⁵

morphologically older types such as Finne in Thuringia, recorded in 1106 as silva Vin, etc.; Vining and Vineberg near Lüneburg; an -r- derivative in Fiener Bruch near Genthin, recorded in 1178 as palustri silva, que Vinre dicitur; Vinnen (Hümmling), around 1000 Vinum, Finnum; with -str- forms: Vinster (Oberlahnkreis), 893 (copy from 122) Veneter, Wenestre, Uenestre, 1312 and later Vinstern, belong here.

Correspondences are also common in the Netherlands, Belgium, and northern France: Bakeveen; Berken; Diepenveen, including apa names Venep, around 960 Vanapan, Venapen, Venep, 1138–53 Uene, 1144 Vene; compare also Venlo.

Attestations are just as numerous in England: Blacken, Broadans, Bulphan, Coven, Fambidge, Fan, Fanns, Fann’s, Fen, Fenn, Fennes, Fulfen, Gladen, Orsett Fen, Redfern’s, Stringock Fen, Vange, Fencote; place names of the type Fenton are common, recorded in 1086 as Fentone, etc.

⁵ From Udolph 194: 315, Map 31.

Stripsni / Articles

199

JÜRGEN UDOLPH

MAP 16. *hor- in place names on the continent and in England⁶

The map shows a strong occurrence in northwestern Germany, along the Lower Rhine, in Flanders, and in England. Schleswig-Holstein and Jutland are involved only to a small extent. Once again, it is evident that the settlement of West Germanic tribes must have taken place across the Channel.

b) A word that is well attested in German, Dutch, and English has so far received little attention: Old High German hor ‘mud, mash, dirt, dung, earth’, Middle High German hor, hore ‘marshy ground, muddy ground, dung, dirt, mud’, Old Saxon hor ‘dung, dirt’, Old Frisian hore ‘mud, dung’, Middle Dutch hore, hor ‘mud; mire’, Old English hoh, horu ‘filth, dirty’.

⁶ Aus Udolph 1994: 328, Karte 32.

200

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 17. *mar- and *mar-sk- in place names and field names on the continent and in England⁷

German place names such as Haarbach, Haarhausen, Hamke, Horbach, Harbrücken, Harburg near Hamburg, Horb, Horburg, Horchheim, Hordorf contain the word.

From the Netherlands, Belgium, Luxembourg, and France, the following belong here: Althorn near Saarbrücken, 783 Horne; Hoerenkreek (Zeeland); Hoorebeke near Oudenaarde, East Flanders, 1090 Horenbeca; Horsik in Gelderland, and many others.

England accounts for a large share of occurrences in place names: Harborne; Harlick; Hamers; Harpole, 1086 Harpol; Harwood Gate; Harton, 1249 Horton; Hawley; Hollowmoor; Holyport, 1220 Hoport; Horbling, 1086 Hebeling; Horbury, 1086 Here; Horcott, −feld, −wod; Le Hornede; Horemestel; Horeput; Horfield, 1086 Horefele; Horham, ca. 950 Horham; Horish (Wood); Horley, 1374 Horlawene; Great, Little Horned, 1086 Hornede; 1243–64 Horne, with field names Horits and Horyt; Horlake; Horel; Horsell, formerly Horse, Horsel, from Old English horsela; Horwell; Horwood, 1086 Horewode; Warpole; Whaley; Worley’s Fm. Numerous compounds with −ton in Horton, 1086 Hortune; 1086 Hortona; 946 (copy from the 13th century) hore tunge are marked on the map with a special symbol.

⁷ Aus Udolph 1994: 375, Karte 33.

Stripsni / Articles

201

JÜRGEN UDOLPH

The distribution map shows that the decisive connections from the Lower Rhine to England run through the southern Netherlands and Belgium, as well as across the Channel.

c) An old Germanic word hidden in place names, with equivalents and cognates in other Indo-European languages, is -mar-, related to Latin mare and Slavic more (Pomore/Pomern), and also attested in Celtic place names such as Aremorica. It is ablaut-related to German Meer < *mari and German Moor < *mōra (a so-called vṛddhi formation); in Germany it occurs in numerous North and Central German place names such as Behmer, Betmar, Bleckmar, Bothmer, Dilmar, Ditmer, Eschmar, Fletmar, Friemar, Geismar, Gelmer, Gittmar, Görmar, Hadamar, Heumar, Hörstmar, Horsmar, Horstmar, Homar, Hükesmere, Koor, Leitmar, Lohmar, Ostmar, Palmar, Rethmar, Rettmar, Riethmar, Ringmar (cf. English place name Ringmer, Old Hringara), Rottmar, Schötmar, Schöttmar, Vellmar, Villmar, Versmar, Voßmar, Wechmar, Weidmar, Weimar, Weitmar, Wethmar, Wichmar, Widmar, Wiedemar, Wismar, Wißmar, Wismar, Wittmar, Wolmar, Wudmar, Wymeer. Compare the distribution on Map 17.

To the west of Germany, we find the word in Aalsmeer, Alkmaar, Alsmar, Bederme, Bommer, Dossmer, Echemi, Gaasmeer (132 Germ.), Hetere, Hoeme, Homee, Purmer, Schermer, Spilmeri, Wormer, Zonnemere, 119 Sutmera.

It is also very common in England: Badesmere, Blakemere, Boldmere, Bradmore, Bulmer, Colmere, Cuckmere, Dodmore, Falmer, Grasmere, Holmer, Homer, Keymer, Marton, Minsmere, Ringmer, Rugmere, Sledmere, Stanmer.

The distribution shows that two large areas are connected to one another across the Channel: northern Germany, the Netherlands, and Flanders on the one hand, and England on the other. Schleswig-Holstein plays no role.

d) A North Sea Germanic water word, whose distribution and etymology have already been discussed several times, is Riede, in Low German ride, rīde, ren ‘natural watercourse, small river, runnel on the tidal flats’, Middle Low German rīde, rē, rīge (ride, rije, rige) ‘stream, small watercourse, ditch’, Old Saxon ritha, rithe ‘watercourse, small stream’; Frisian riede ‘canal, small river on the tidal flats’, ryt, ryd(e) ‘broad drainage furrow’, North Frisian rīde, riet; Old Frisian red ‘small stream’ and rīth ‘stream’; Dutch rijt ‘watercourse’, rijt, riet ‘channel in unembanked land’, ‘inland water in the marine clay districts close to the coast’, Middle Dutch rijt f., Old Frankish rīth ‘tornes’.

It occurs very early in English, already in Old English rīd, rīde ‘small river’, rīde ‘stream, long, narrow depression, old streambed’, rīd, rīde,

202

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

M A P 18. Riede, ride, rithe, riet, rid in geographical names⁸

rīdġ a small river, English rithe, ride ‘small river, caused by rain’, rigat ‘a small channel from a stream made by rain’, rithe, ride ‘a small stream’.

In Germany there are numerous names, some of them also recent, such as Achel- riede, Aschriehe, Bargriede, Bassriede, Bickenriede, Bleckriede, Bollriede, Borgrie, Boterriede, Brandriehe, Bruchriehe, Brunriehe, Diekriehe, Eilenriehe, Ellerie, Jachelriehe, Janien, Midel Rie (Midels Rie), Exerie, Feldriehe, Feldrie, Flehmanns Rieh, Flamischen Rie, Weckenlands Rie, Fliesenrie, Fluthrie, Fohlenrien, Fuhlenrie, Fuhenrie, Fule Rie, Die Große Rie, Grotr, Haferrie, Hauenrie, Holtrie, among many others.

Older forms include 726 (copy from around 122) Aride (near Cologne), Brücher near Ebe- len, 876 Boratrie, 1290 Bruchrie, also Burchtrie, Borntrie; a Corvey record from ca. 826–876 Hrihem (with non-etymological H-).

In the Netherlands, Belgium, Luxembourg, and northern France I found Bruggerijt, Dieprijt, het Dikke Riet, Munnikenzijlster, Ekkersrijt, Houtrijt, Jutesriet, Peelrijt, Pieperij, Riet, Rijit, Segerijd, among others; some are recent formations.

⁸ Nach Udolph 1994: 393, Karte 34.

Straipsniai / Articles

203

JÜRGEN U DOLPH

England has numerous and in part very old names, such as Aberd; Be- verley Brook, 693 (11th-century copy) beferdi; Blackrith; 972 (copy from 1050) Bordrödig; Chaureth, 1086 Ceuride; Childrey; Coldrey, 973/4 (12th-century copy) (to) coride; Coleread; Cropredy; Cotered, 1086 Clodre; 1228 Ealdnereth; Eelrthe, 680 ad Aelrithe; Efferdy; Erith; Fingrith, 693 water name Fuglerdie; Fulredy; Ful- rithe; Goosery; Henred, 984 Henna rid; 774 Hwelridig; Landrith; Shotery, 699–709 (11th-century copy) Scottärd, among many others.

The mapping (see Map 18) shows that these names occur mainly in northern Germany, the Netherlands, Belgium, and England. Apparently, the settlers migrated to England across the Channel.

e) An old word that links northern Germany and England in particular is German Hude. A German speaker now knows it only from place names, especially Buxehude, Fischerhude, Harvstehude and the Steinhuder Sea. However, the meaning of this word is unknown to a speaker today. Hude is found in northern Germany mainly in settlement names situated by bodies of water; it is still attested in Middle Low German as hūde ‘wood-storage site, stacking place on a waterway, ferry landing’. It is also unknown in present-day English, but is still attested in Old English: hyð ‘a place that receives a ship on landing, a suitable low bank, a small harbour’.

The distribution of the place names is interesting. From Germany I have collected, among others: Altenhude; Aschenhude; Billerhude; Dockenhuden, 1184 Dockenhude; Dodenhude; 1346 Eckhude; Fischerhude, 1124 Wicheshude; Flemhude; Frauenhude; Grönhude; Harnhude; Harvstehude; Heenhude; Huden near Meppen, 1037 -huhun in the record Hlaresuhun; Hodenhagen, 1168 (18th-century copy) de Hode and elsewhere; Hude (frequent), also with umlaut Üde; Huen, around 920 Huhun; Huda; Hudemühlen; Kayhude; Nedernhude, Obenhude; Pahlhude; Ritterhude; Stapelhude, 1258 in loco qui dicitur Stapelhude; Steinhude on the Steinhuder Sea, second half of the 14th century To der Steinhude; Tespehude; Winterhude.

The Netherlands, too, have place names: Oude Hide in Zeeland; Coxde, 1270 de Coxhde; Coxde (Koksie); Hude, 1405 Hude; Hude dries; Hudestra- te; 1359 le Hyde near Dunkirk; Nieuwe Yde near Nieuwpoort/Oostduinkerke, 1277 Noua Hide; Raversijde, 1401 Wilraues hyde; Lombartside, 1408 Lombardsyde; Yde, 1331 in die Hide.

The place names in England are important and old: Adreth, 1169–1172 Alre- heð(a), -hūda; Bablok Hythe; Bladeney, 712 (14th-century copy) ad portam que dicitur Bladenithe; Bolney, 1086 Bolhede; Bulverhythe; Chelsea, 785 Cealchyð, Celchyð, 801 Caelichyð; 1275 Colleshethe; Claythe; Coweithe; Creek- sea, 1086 Criceseia; Downham Hythe, 1251 Dunham hythe; Earith, 124 Her- heth; Erith, 695 Eardhy; Fishythe; Frecnythe; Glanty, 675 (13th-century copy) Glenthe;

204

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 19. German hude, English hyð, etc. in place names⁹

Greenhithe; Heath (multiple); Hidden, 984 (K. around 1240) (innan) Hydene; Hithe Bridge; Hive, 959 (K. around 1200) Hyðe; Hive, 1306 atte hethe; Horsith, 1249 Horsyth(e); Hyde, 1333 atte Hithe; Horseway, 1238 Hors(e)hythe; Hullasey, 1086 Hunlafesed; Huyton, 1086 Hitune; Hythe (Surrey), 675 (K. 13th c.) huþe; Hythe (Cambridge), 1221 Hethelod; Hythe (Kent), 1052 (on) Hype; Hythe (Hampshire), 1248 (la) Huthe; Knaith, 1086 Cheneide, < cneoyþ; Lakenheath, around 945 æt Lacingahið; Lambeth, 1041 Lambhyð; Maidenhead, 1202 Maideheg; Prattshide, around 1250 Pratteshithe; Rackheath, 1086 Racheitha; Rotherhithe, around 1105 Rederheia; Sawtry, 974 Saltrei’am; Small Hythe, 13th c. Smalide; Stepney, around 1000 Stybbanhyþe; Swavesey around 1080 Suauesheda; Welshithe, allegedly 675 Weales hūðe.

The distribution of the names (see Map 19) clearly shows that there is no evidence to suggest that settlers came to England from Schleswig-Holstein or Jutland. Rather, the place names in Flanders must be taken into account, as they evidently form the bridge between the German and English names.

A hundred years ago, Hermann Jellinghaus had already recognized that the -hude names “are strong evidence for the origin of the population of southern England from the Low German plain”¹⁰.

⁹ Nach Udolph 1994: 472, Karte 44.

¹⁰ In: Anglia 20 (1898) 290.

Straipsniai / Articles

205

JÜRGEN UDOLPH

MAP 20. Distribution of hauga- in geographical names (from Bischoff 1975, appendix)

The distribution maps within Germania, as I have already noted above, also show clear connections to the north. These have almost always been interpreted in the past as expansions of Germanic tribes from their northern European homeland toward the south. However, it can be clearly demonstrated that it is much more appropriate to assume an original homeland in continental Germanic territory (for details, see Udolph 1994: 830–918).

206

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 21. Distribution of klint in geographical names¹¹

a) We owe our more precise knowledge of the distribution of the word *haugaz in Germanic—which is attested, among other places, in Old Norse haugr ‘hill, burial mound’—to a study by Karl Bischoff (1975; see also Udolph 1994: 859–863). Note also forms with similar meanings: Icelandic hau gur, Faroese heyggjur, heygur, Norwegian haug, Swedish hög, Old Danish høgh, Danish høj. The word is also found in German place names; Bischoff (1975: appendix) provides a distribution map.

b) klint is distributed in a very similar way; as a common noun, the word is attested chiefly in the north:

Danish klint ‘steep sea cliff’, Swedish klint ‘summit of a hill’, assimilated variant in Swedish dialectal klett, Norwegian dialectal klett ‘mountain peak, steep sea cliff’, already Old Norse klettr ‘isolated cliff’, ablaut variant in Norwegian dialectal klant ‘edge of a cliff, mountain summit’ and Danish klunt ‘block, lump, clumsy person’.

But the name inventory—the “graveyard of words”—also contains it to a considerable extent in the continental Germanic area, as Map 21 shows. Its distribution closely resembles that of hauga-

¹¹ Nach Udolph 1994: 868–881, mit Karte 68.

Straipsniai / Articles

207

JÜRGEN UDOLPH

MAP 22. Distribution of –leben/–lev in geographical names¹²

Also belonging here is the frequently discussed and often cited basic word –leben, North German –lev, attested in approximately 200 place names, for example Uhrsleben, Erxleben, Eilsleben, Eimersleben; Aschersleben, Oschersleben, Eisleben, Barleben, Roßleben, and in Denmark and southern Sweden as Bind­slev, Jerslev, Roslev, Falslev, Tinglev, among others.¹²

Underlying this are Gothic laiba ‘remnant, remainder’, Old High German leiba, Old Saxon lēva ‘remainder, inheritance, estate’, Old Frisian lāva ‘legacy, inheritance, right of inheritance’, Old English lāf ‘legacy, inheritance’, Old Norse leif ‘remnant’, and Old Danish kununglef ‘crown property’. The first element is invariably a personal name in the singular genitive¹³.

¹² Nach Udolph 1994: 503, Karte 47.

¹³ For the latest discussion, see Winkler 2009.

208

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 23. Distribution of –wedel–/*wadil– in geographical names ¹⁴

The unusual distribution of the names (cf. Map 22) has often been discussed, but the major and most comprehensive book on these names was given almost no consideration (Bathe). Including it makes a south-to-north migration more likely than the reverse direction (cf. Udolph 1994: 497–513).

And one final example of name connections between the continent and the Scandinavian north: in northern German place names, Middle Low German wedel, Old Saxon widil ‘ford’ is assumed; it is related to Old Norse vadhell, vadhall, vadhill ‘shallow place in a fjord where one can wade across’, Norwegian val, vaul ‘shallow spot in a fjord’, also found in place names as Voil, Veel, and Danish vedel, vejl, 16th c.,

¹⁴ Nach Udolph 1994: 892–906, mit Karte 71.

Straipsniai / Articles

209

JÜRGEN UDOLPH

MAP 24. Scandinavian words in German place names

Place names Veile, Veilby, Veielgaard, Veilö, Faareveile; in Denmark, the meaning is chiefly ‘passage through marshy terrain’.

The connections with the Scandinavian north briefly outlined here are also evident in a particular feature: appellatives attested exclusively in Nordic are also found in names from the continental Germanic area (discussed and interpreted in detail by Udolph (2000, 2004, 2011)). We now know more than three dozen such cases, including the names Braunlage, Dorstadt/Dorestad, Kösen/Coesfeld¹⁵, Ohrum, Oerie, Rhön¹⁶, Scheuen, Scheie.

They can only date from an Old, Common, or Proto-Germanic period and show that the appellatives underlying them must have belonged to this ancient stratum.

Mapping these place names (see Map 24) reveals a center for these place names, which can be explained only with the help of Scandinavian words: eastern Lower Saxony, together with the neighboring areas in Saxony-Anhalt, Thuringia, and northern Hesse. A very similar picture emerges,

¹⁵ In: Korsmeier 2016: 90–94.

¹⁶ In: Ascher 2015: 203f.

210

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

when one looks for appellatives found only in Scandinavia and the Alemannic area¹⁷ in continental place names¹⁸.

Suffix formations

It is well known that an important change took place in the formation of nouns in the history of the Germanic languages: whereas today compounds are used above all, cf. German Hand-tuch, Auto-reifen, Haus-wand, English main door, forrest path, car tyre, Norwegian hovud dør, skov vej, bildæk, in early Germanic times new words were formed chiefly with suffixes, that is, by derivation: *gab-lo- ‘fork’, *ham-ipja- ‘shirt’. Suffix formations are therefore particularly important when searching for archaic Germanic place names. They are crucial for determining the oldest settlement areas of Germanic tribes. For reasons of space, I select only a few examples here.

a) -s formations.

In ancient times, Germanic partly inherited -s formations in its appellative vocabulary and partly formed new ones. Among the old inherited words, for example, are Gothic aiz, bariz-eins, rimis, as well as some new formations (which may also be based on old s-stems), such as Gothic aqizi, jukuzi and Old High German chuburra. Also characteristic of Germanic are later developments of the suffix -sa/-sō-, including formations with a medial vowel such as Old High German bilisa, elira, felis(a). Hans Krahe and Wolfgang Meid (1967: 137) also provide a brief compilation of “Old Germanic names with an s-suffix”. Alongside personal names such as Gabso, Hariso, Aliso, Thoriso, river names such as Ems and Effze are also mentioned; these are rightly connected with material from outside Germanic. R. Möller (2000) and J. Udolph (1994: 199–218) have investigated and compiled corresponding formations in the name corpus. Here I pass over the fact that -s formations also occur among the pre-individual-language hydronyms of Old European hydronymy (briefly discussed in Udolph 1994: 199–201).

Among others, the following belong here: deserted settlement Blekisi, 826–876 in Blekisi, deserted settlement Degese, 1196 Degese, Devese, Gebesee near Erfurt, 802–815 Gebise; Heerse < Herisi, Hünxe, 1092 Hungese, WgN. Ilse near Boffzen, 1031 Ilisa, Ilvese, 1096 Hil-vise, Klings near Bad Salzungen, 869 Clingison, Leisa on the Eder, 8th c. Lehesi, Lihesi, field name 1692 im Letse, place name Linse near Bodenwerder, first half of the 9th c. Linesi, Meensen near Göttingen, 990 Manisi, Resse near Recklinghausen, 10th c. Redese, Reese near

¹⁷ On the subject, see Kolb 1957 and Alemannien und der Norden 2004.

¹⁸ Cf. Udolph 2004a; mapping of the corresponding and supplemented names in Udolph 2010b: 153.

Straipsniai / Articles

211

JÜRGEN UDOLPH

MAP 25. –s-suffixes in geographical names of Central Europe¹⁹

Stolzenau, 11th c. Raedese, Schlipps near Freising, between 851 and 1130 Slipfes, Slipphes, Sliphës, Schl ipfs, Seelze near Hanover, 1160 Selessen, 1187 Seles se, field name Sötz, marsh forest near Goslar, old Sotisse, Vielshof near Salzkotten, Vita Meinwerci Vilisi.

The mapping (Map 25) reveals a pattern that we already know from other distribution maps: the core area comprises southern Lower Saxony, western Saxony-Anhalt, northern and western Thuringia, northern Hesse, and Westphalia.

b) –st formations

There are numerous formations with a –st suffix, particularly in the Baltic region; among others, I have come across the following: in Latvia, Līlaste, Lubasts, Lubešta, Mùrmasta; elsewhere in the Baltic region, Ab-istà, Akn-ystà, Arv-õstas, Avin-úostas, Debr-estis, Grab-uostà, Iẽž-esta, Lam-istas, Lauk-ystà, Lok-ystà, Malk-esta, Pekļ-ystà, Sav-istas, Taur-ósta, Uol-astà, Varn-úostas, Ým-asta; from Austria, Aist < *Agist(a); from France, Autisse < *Altīssa < *Altīstā; from Dalmatia, Bigeste, Ladesta, Triest, Palaeste, Segest-(ica), Penestae; from Venetia, Este < Ateste.

¹⁹ Nach Udolph 1994: 199–218, mit Karte 25.

212

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

In addition to the two centers that emerge here, namely the Baltic region and the areas around the Adriatic, there is another region in which the suffix accounts for a substantial share of the name inventory: Central and Northern Europe. A more detailed distribution is shown on Map 26, which includes, among others:

866 Alesta, place name near Charleroi, Alst, 12th c. Alest, Aalst, place name in Flanders, 866 Alost, Aalst, place name near Hasselt, 1107 Alost, Alste; Aalst in Gelderland, around 850 Halo-sta, Alst near Leer, 12th c. Alst; field name Apfelstädt, also a place name, 775 Aplast; Arzbach near Gotha, 1049 Arestbach; Beverst near Tongeren, 1314 Beverst; Vita Meinwerci: Bilisti; field name Burdist, 755 Burdist (on the Rhine); *Agist—possibly in Eekst near Assen; Ehrsten near Kassel, old Heristi, Herste; Elst near Nijmegen, 726 Helistê; Ennест near Olpe, 1175 En-nest; Ergste near Schwerte, 1096 Argeste; Euelste near Pattensen, 1246 in Evelste; Exten near Rinteln, 896 Achreste; place name Forst near Holzminden, *Farista; Haste near Osnabrück, 1146 Harst; Harste near Göttingen, 1141 Herste, Harste; field name Idesten or Itesten near Emden; Innerste, old Indrista; watercourse name, 1106 Lammeste near Hanover; place name Landas near Lille, old Landast; Leueste near Hanover, 1229 de Leueste; place name Pretitz near Zingst, 9th c. Bridasti; place name Ranst near Antwerp, 1140 Ramst, 1148 Ranst; Riemst near Tongeren, 1066 Reijmost; Rumst near Antwerp, 1157 Rumeste; Selsten near Geilenkirchen, *Salist; Thüste near Hamelin, 1022 (F.) Tiuguste, Thiuguste; Villigst near Ergste, 1170 Vilgeste; Zingst near Nebra, 1203 Cindest.

Since this collection was compiled in 1994, numerous further parallels have emerged, especially through the comprehensive study of the place names of Lower Saxony and Westphalia.

It should be noted that the place names with the st suffix show a wide variety of forms in their prefixal element. -est- can be identified with some certainty in Al-est, Ar-est, Id-est-, Lameste, Tind-est-; -ast- apparently occurs in Ap-l-ast, Har-ast-, Landast and Bredh-asti; -ist- is very common: Bil-ist, Burd-ist, El-ist-, Am-ist-, An(d?)-ist-, Arg-ist-, Agr-ist-, Far-ist-inda, Har-ista, Lev-ista, Ram-ista, Sal-ist-, Far-ist, Felg-ist.

The diversity of the suffix fully accords with the evidence from the appellative vocabulary. Corresponding forms are readily attested, above all in Germanic and Baltic. For most of the place names assembled here, the Germanic evidence can be compared; this applies, for example, to the following place names, which can be explained quite convincingly using Germanic lexical material: Awist, Beverst, Eext, Ehrsten, Idesten/Itzstedt, Landast, Riemst, Selsten/Zelst/Zeelst, Thüste, Villigst, Zingst.

While names with a -st suffix are also found in non-Germanic settlement areas, the following formative element differs in this respect.

Straipsniai / Articles

213

JÜRGEN UDOLPH

MAP 26. –st-suffixes in the geographical names of Central Europe.

● Hydronyms, ○ place-/settlement names²⁰

c) –str-suffix

A corresponding means of formation is found only in the Germanic languages, for example in the appellative domain in Gothic awistr, Norwegian naustr, Laster < *lah-stra-, Polster < *bulh-stra-, Old Norse mostr < *muh-stra-, Old English helustr, heoloster, Gothic hulistr ‘covering’, Old English gilister, geoloster ‘ulcer’. The names formed with it, too, are confined to the (originally) Germanic settlement area

²⁰ Nach Udolph 1994: 218–243, mit Karte 26.

214

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 27. –strsuffixes in the geographical names of Central and Northern Europe and England. ● Hydronyms, ○ place-/settlement names

(Map 27). According to a list compiled in 1994, examples include Alster, Elster, Alstern, Alsterån; field name Ballestre in England, 940 Ballestran; field name Beemster near Alkmaar, 1083 Bamestra; Beuster, tributary of the Innerste, 1305 Bostere, 1308 Botestere; Emster near Brandenburg, formerly Demster or Emster; Norwegian field name Imstr; Falster; Finster in the Oberlahn district, before 893 Veneter, Wene-stre, Venestre; Fløstr, Scandinavian island name; Gelster, tributary of the Werra, 1246 inter Gelstram; Norwegian river names Jølstra, *Jøstra; Kelsterbach, place and field name near Groß Gerau, 830–850 De Gelsterbach; field name Lister

Straipsniai / Articles

215

JÜRGEN UDOLPH

in the Westerwald, 1532 in der Lyster; English field name Medestre, 940 (on) Medestran; fjord name Ørstr; Seester(au), old name of the Krückau, a tributary of the Elbe, 1141 iuxta fluu ium Ciestere, with place names Seester, Seesterau; Susteren, place name in the Netherlands; 1277 Rususteren, < field name Suster, 714 Svestra; Ulster, field name in the Rhön, 819 Ulstra, Swedish lake name Vänstern, Norwegian field names Vinstr, Vinstra, English field name Winster, 1170–84 (K.) Winster, Wilster → Medem, Wilster Au (Wilsterau) → Stör, Zester, an abandoned field name in the Altes Land, 1197 iuxta Szasteram, with the abandoned place names Zesterfleth, Sestersvlete.

Almost all these names—others have since been added—can be explained from Germanic lexical material. But there are exceptions, for example Alster, Elster and Wilster. Here the single-language Germanic suffix was added to a pre-Germanic base *el-/*ol- or *wil-, which points to a certain continuity of settlement from Indo-European to Germanic in these geographical regions.

d) The suffix -ithi

Dieses Suffix wurde schon mehrfach behandelt (Udolph 1991, 1994: 258–288; Möller 1992; Casemir 2003: 438–446). Es ist ein hochaltertümliches Bildungselement, das vor allem in der Namenbildung deutliche Spuren hinterlassen hat. Es ist heute nicht mehr produktiv und ist uns aus den älteren germanischen Sprachen nur noch in wenigen Wörtern bekannt: got. avithi ‘Schafherde’, angelsächsisch gesylhbe ‘Joch Ochsen’, altenglisch winterfylled ‘Oktober’, althochdeutsch winithi ‘Weideland’, juhhidi ‘Gespann’ und einige andere.

Although it is scarcely still recognizable as an appellative, which suggests a relatively great age, it is very well known in northern and central German place names. Around 200 names belong here, including some quite well-known ones. Here is only a small selection: Birgte, 1088 Bergithi; Bleckede on the Elbe; Bünde, 853 (forgery) Buginithi; Dingden near Bocholt, 1163 Tingethe, from Old High German thing, ding “general assembly of the people”; Döhren, a district of Hanover, around 990 Thurnithi; Essen, 9th c. Astnide; Geesthacht, 1216 in Hachede; Gimbte, 1088 Gimmethe; Grohnde, (1237–47) in Gronde; Helle near Wiedenbrück, late 12th c. Helethe; Huckarde, a district of Dortmund, 947 Hucrithi; Hü sede, 12th c. Husithi; Lengede, 1151 Lencethe; Mengede, a district of Dortmund, 10th c. Megnithi, Mengide; Meschede, 913 Meschede, 1015–25 Meschethi; Sarstedt, (1046–1056) Scersteti, Scerstete, 1196 Scardethe; Sehnde, 1147 Senethe; Sömmerda, 876 Sumiridi and likewise Sumiridi.

West Germanic settlers who crossed over to England from the 5th century onward took the formative element with them and, as Map 28 clearly shows, created a few more place names on the island before the suffix became unproductive and disappeared from the language. These place names include, among others: in Kent, for example, The Brent, Brent Lane, late 14th c. Bremthe; deserted settlement name, 13th c. Bremthe; deserted settlement name, 1286 Bremthe; 4. Brent Cottages, 1359 Brencche; 1206 La Brenithe

216

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

JÜRGEN UDOLPH

MAP 29. Oldest settlement area of ancient Germanic settlers inferred from archaic Germanic place-name types

Beneath the dense network of Germanic place and water names specific to individual languages, large numbers of pre-Germanic hydronyms can be identified in northern and central Germany, for example in Kettner’s study (1972; see also Greule 2014); these can also be found through the series Hydronymia Germaniae by consulting the relevant literature.

If one now attempts to represent the distributions of ancient Germanic names cartographically, then in my opinion the crucial point is that it must be possible to find a common geographical basis both for the name types extending westward (such as formations with the suffixes -ithi, -st- and with the appellatives hor, mar-, Riede, Hude, (h)lar, among others) and for the suffixes and words linking the continental Germanic area with Scandinavia. I have attempted to describe this (Udolph

218

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 30. Soil yield indices in Lower Saxony

1994: 925ff.) and today add a map delineating approximately the area which, on the basis of geographical names, I must regard as the homeland and expansion area of ancient Germanic speakers (Map 29).

I am aware that this mapping invites criticism. I would nevertheless like to emphasize once again that the water and place names clearly support this area. And there is—as already mentioned above in the discussion of Slavic—another important argument repeatedly invoked by historians and agricultural scientists: I mean soil quality and soil types. Maps, for example of soil types in Lower Saxony, show an almost complete overlap between the best soils and the archaic Germanic names mentioned above and their distribution. For this purpose I have selected a map of Lower Saxony which

Straipsniai / Articles

219

JÜRGEN UDOLPH

MAPS 31, 32. Distribution of the -ithi- place names and soil yield ratings in Lower Saxony

the land’s yield ratings show—the best soils appear in dark shades (Map 30)²².

The source from which this mapping is taken offers the following further explanation: “Only along the coast and south of the Mittelland Canal do good soils predominate. In the sea of sand of the Geest and heathland, a few ‘fat eyes’ of good, loamy soil float, like a few fat eyes on a meagre soup (Uelzen; Lüchow; Weser, Leine and Aller valleys; Artland; Hoya-Syke-Goldenstedt)”²³.

The correspondence between soil quality and ancient place names is especially clear in the following comparative mapping, which shows both the distribution of the -ithi- place names in Lower Saxony and the soil yield ratings of Lower Saxony (Maps 31, 32).

We can therefore draw the same conclusions as in the earlier discussion of the Slavic homeland: it is clear that the distribution of the best soils in central and northern Germany corresponds to the ancient Germanic hydronyms and place names. It is well known that the Börde regions (Soester Börde, Calenberger Lössbörde, Hildesheimer Börde, Magdeburger Börde) in the old glacial landscape of northern Germany, on the edge of the central uplands, have black-earth soils that rank among the best in Germany. Every farmer knows what it means when soil ratings of 80 points or more are recorded

²² Taken from: http://www.nls.niedersachsen.de/Tabellen/Landwirtschaft/internetseite2002/hochschulen.pdf.

²³ See note 27.

220

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

of soil is being discussed. As I already noted above in the discussion of the Slavic language, good soil generally produces good and better harvests than poorer soils. The result is lower child mortality and higher life expectancy. The population grows, which in turn leads to a certain degree of overpopulation, migration, and the spread of the language spoken in this region. Spectacular discoveries such as those in the Lichtenstein Cave near Osterode, and the finding that people who lived there 3,000 years ago still have descendants in this area today (demonstrated by DNA analyses; discussed and supplemented from an onomastic perspective by Udolph 2009), show that settlement continuity in the Harz region can certainly be assumed.

3. THE CELTIC

Summaries of the state of Celtic place-name studies indicate that there is agreement that Insular Celtic arose through migration from the continent²⁴. Its origins are therefore to be sought in central or western Europe. For the mainland Celtic hydronyms, the studies by Pierre-Henry Billy (1993), Xavier Delamarre (2003), Alexander Falileyev (2006/7), Ranko Matasović (2009) and Patrick Sims-Williams (2006) are of particular value. The following map is based primarily on archaeological research, but also corresponds to a considerable extent with the assumptions of linguistics.

The decisive step, however, was taken—this is my conviction—by Peter Busse (2007) in a short contribution to a conference of Celticists; these ideas are merely the initial starting points of a larger project. He rightly emphasizes that the study of hydronyms is essential, since this “has not yet been fulfilled with regard to the question of Celtic-speaking settlement” (Busse 2007: 89), for, in contrast to studies of Indo-European, Slavic and Germanic hydronyms, virtually no research in this field has been undertaken in Celtic studies. In his view, almost all researchers avoid this topic.

P. Busse rightly asks whether archaeological arguments can or should play a role in determining the “Celticity” of settlers. He names Simon James (1999) as the leading representative of “Celtoscepticism”.

²⁴ For what follows, see Udolph 2009/2011 for a detailed discussion.

Straipsniai / Articles

221

JÜRGEN UDOLPH

MAP 33. Migrations, territorial acquisitions and the influence of the Celts up to the 1st century BC²⁵

und John Collis (2003), „wobei letzterer ernstzunehmende Einwände gegen die Identifikation von archäologischer Evidenz und ethnischer Zuordnung liefert“. Dieses deckt sich in hohem Maße mit der modernen Forschung in Deutschland, wobei vor allem auf S. Brather (2004) zu verweisen ist.

P. Busse draws the only reasonable conclusion from this: if one seeks Celts from a linguistic perspective, this can only be done using linguistic arguments. But if one does so, one must—and this is absolutely essential—not merely include geographical names, but accord them a very high status. Among these, hydronyms are of particular importance.

In his view, something already attempted on the basis of Slavic and Germanic has become important to the question of the earliest homes of Celtic speakers: the development of an Indo-European language family from an Indo-European pre-individual-language

²⁵ From: http://www.timediver.de/Keltenwelt_am_Glauberg.html (accessed: 9 June 2017).

222

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

If the formation of a dialect area has been a long process—we must certainly reckon in centuries—then in this area one should expect not only hydronyms belonging to a single language, such as Celtic, Baltic, or Germanic, but also Indo-European relics predating the individual languages. This development must be reflected in the hydronyms; or, in other words, an area in which there are only names belonging to individual languages, for example only Celtic names, cannot be the area of the Celtic homeland.

Busse’s project, which he presents only in outline, “aims, by using and bringing together the results of previous studies and projects on the subject of ‘Celticity’ [...] to find an answer to the question of the extent to which hydronymy can provide an informative picture of the early settlement history of Celtic-speaking population groups” (Busse 2007: 91f.). Following H. Krahe, and entirely correctly, the focus is primarily on hydronyms based on words for “water, river, flow,” and so on. Taking into account studies in the Germanic, Baltic, and Slavic fields, he develops the following central questions:

• Can an area of “Celtic hydronymy” be identified that could be regarded as the nucleus of a Celtic expansion, that is, as a homeland?

• To what extent does this nucleus coincide with the extent of the Hallstatt or La Tène culture?

To find answers, he considers, on the one hand, hydronyms located in Celtic territory but of pre-Celtic origin, and, on the other, Celtic names. Among those included are, inter alia, Ainos, Aenus, Dubis fl./Doubs, Douglas and related forms, Devy/Devon, Devoke Water, Old Irish dobur with New French Douvre (1128 Dobra), Douvres (ca. 380 Dubris), Verdouble ca. 79 Verno-dubrum, and so on; Spanish Dobra, among others; German Tauber, English Dover; the family around Glanis, Glanum, Glanon, Glan, Glene, Glane; Gaulish < *gweh₃w- ‘mud’, cf. Welsh baw ‘dirt, filth’; borṃ-/borw-/borb- <*bher- ‘boil up, surge, ferment, boil’ in Borbro (Bourbre), Borvo(n) ‘spring god’, Formio, in France Bormane (Ain), Bourbonne (Aube), La Bourbre (Isère), and many others; brig- < *bhrgh- to *bh̉eregh- ‘high, elevated’, perhaps in the sense of ‘upper course’, in Brigia (Braye = Loir), Brigulos (Saône → Rhône), and many others; esk-/isk- <*peisk-, Middle Irish esc ‘water’, with Esca/Escia/Hisca (Isch → Saar), among others; fruta- <*s(w)rutu-, cf. Welsh ffr-wd, Irish sruth ‘torrent’, in Frudis, a river in Belgica, and so on.

In the conclusion, the results are summarized with the aid of a distribution map: “As a first provisional result, the following can be established on the basis of the map material: Celtic hydronymy is distributed across an area extending along the northern Alpine region, following the Danube, the upper and middle courses of the Rhine and the Rhône, and including their tributaries. The area of origin of the

Straipsniai / Articles

223

JÜRGEN UDOLPH

Hallstatt culture does not coincide with this area, although the most important West Hallstatt and early La Tène finds are located within it” (Busse 2007: 97). I note here only in passing that, in his research dating back to the early years of the twentieth century, the Polish Indo-Europeanist Jan Rozwadowski (partly reprinted in Rozwadowski 1948) had already identified striking correspondences between hydronyms in the Baltic region and those of the Rhône area, such as Visentia, Saane, Brenne, Divonne, Isara, Drome, Ivarus, among others.

This has since been further substantiated. Important name parallels have been identified in France (W.P. Schmid 2004) and in the Moselle region (W.P. Schmid 1988), and clear correspondences between Old European hydronyms in Poland and the Baltic region, on the one hand, and the British Isles, northern France, and southern France, on the other, have shown (Udolph 1990, summarizing on pp. 332ff.) that almost the entire Celtic settlement area (with less certainty, the Iberian Peninsula) contains a substrate of Indo-European hydronyms predating the individual languages.

If we try to weigh previous studies of the hydronyms of Europe against P. Busse’s work on Celtic, the following common points can, in my view, be identified:

Hydronyms are the oldest witnesses to human language in Europe. Their study is indispensable for questions of prehistory and early history.

Comprehensive studies of hydronymy are lacking in areas of former Celtic settlement on the mainland (especially in northern Italy, France, and parts of Switzerland).

Archaeological findings can—unfortunately—be used only to a limited extent to address the question of the ethnic divisions of early Europe (Brather 2004).

Since hydronyms are derived essentially from so-called “water words,” a comprehensive study of the geographical terminology of all Celtic languages is an urgent desideratum (this also applies, incidentally, to the Germanic languages, whereas in Slavic there are exemplary studies, even ones restricted to hydronyms).

In his map, P. Busse has plotted those hydronyms that, in his view, are significant for determining the oldest settlement areas of Celtic tribes.

Based on P. Busse’s study and following corresponding research in the fields of Germanic and Slavic

224

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 34. Mapping of Old European and Celtic names²⁶

one may say: There is much to suggest that Celtic took shape in the region where we can observe a certain “thickening,” an increased concentration of Old European water names. Of decisive importance here is the existence of parallels in the Baltic region and the areas adjoining it, to which J. Rozwadowski (1948) had already drawn attention. Thus, a saying that had already played a role in questions about the homeland of the Germanic tribes also applies to the question of the original Celtic settlements: Ex oriente lux²⁷. Taking this into account, much already points—and one may largely follow P. Busse—in the direction of the region west of the Alps. Research in the countries formerly settled by Celts would do well to turn its attention to water names more intensively than before.

²⁶ Aus Busse 2007: 97.

²⁷ Udolph 1981; Udolph 1985.

Straipsniai / Articles

225

JÜRGEN UDOLPH

4. THE BALTIC REGION FROM THE PERSPECTIVE OF ONOMASTICS

Linguists and onomastic researchers repeatedly emphasize continuity in the settlement history of the Baltic region. Major migratory movements are hardly discernible; what can be discerned, however, is a process that reveals a considerable shrinking of the formerly Baltic settlement area (a map based on M. Gimbutas’s publications, requiring corrections in detail, is provided by W.P. Schmid (Map 35).

MAP 35. Distribution of Baltic water names²⁸

It was essentially Slavic tribes that assimilated the Baltic region from the east, south, and southwest, as is clearly evident in parts of northwestern Russia, Belarus, and northeastern Poland. From a hydronymic perspective, a distribution map of water names is of interest; in J. Rozwadowski’s (1948) view, these are described as “Slavic”

²⁸ Schmid 1976: 15 nach Gimbutas 1963: 30f.

226

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 36. “Slavic water names” discussed by J. Rozwadowski (1948)

(Map 36). But these are by no means Slavic; they are pre-linguistic, Indo-European water names. This is already clear from the headings of individual sections of his studies, for example „dreu- (Drwęca, Drawa [...])“, „Isa“, „Isana“, „Isara“, „Oła i jej grupa: pierwiastek el-“, which in some cases are almost identical to H. Krahe’s much later studies and largely agree with Krahe’s investigations and findings. In Western Europe, they went completely unnoticed.

A recent publication’s map (Map 37) also shows something similar, albeit only in outline, since it clearly shows an increase in pre-Slavic water names in the territories adjoining the formerly Baltic settlement area in northeastern Poland; this undoubtedly continues into the neighboring Baltic settlement areas.

It has since also become clear that the water names J. Rozwadowski regarded as Slavic must be interpreted differently: a small proportion belong to the Baltic substrate in Russia and Belarus,

Straipsniai / Articles

227

JÜRGEN UDOLPH

MAP 37. Oldest hydronyms in Poland according to Z. Babik²⁹

in northern Ukraine and northeastern Poland, but for the far greater part belonging to the corpus of what, since H. Krahe, has been called Old European Hydronymy, a network of hydronyms of pre-individual-language, Indo-European origin (Krahe 1949–1965, 1964; Schmid 1994; Udolph 1990).²⁹

And one further important new finding could be established: within the Old European, pre-individual-language, Indo-European hydronyms of Europe, Baltic has a special status, as W.P. Schmid (1994: 175–192, 226–247) demonstrated. Hydronymy in the former and present-day Baltic settlement area is characterized, on the one hand, by a high density of pre-individual-language names; on the other hand,

²⁹ Babik 2001: 92.

228

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 38. Baltic center in hydronymy³⁰

continuity and consistency in the formation of hydronyms can be discerned from the earliest Indo-European period through to the Baltic period as an individual language; finally, within Old European Hydronymy, only in the Baltic region are there hydronyms with counterparts in many European regions. Examples include Atešys, Atesė, hydronyms in Lithuania: Etsch/Adige (attested in antiquity as Atesis, Athesis), Eisa: Aisė, Limena: Limenė in Lithuania, among others; for mapping, see Map 38.

I was able to establish something similar in my study of pre-Slavic names in Poland (Udolph 1990, summary: 331ff.). The mapping presented here (Map 39) is based on that study, especially the compilations of name correspondences between Poland and various regions of Europe (e.g. the British Isles, Western Europe, Northern Italy, etc., pp. 332–338).

None of this is accidental. As earlier and more recent studies have made clear, the Baltic stock of names shows correspondences in a belt

³⁰ Nach Schmid 1994: 184f.

Straipsniai / Articles

229

JÜRGEN UDOLPH

MAP 39. Name correspondences between water names in Poland and European parallels³¹

of regions extending from the British Isles, France, the Iberian Peninsula, Italy, Switzerland and Austria, and the Balkans to Eastern Europe. In recent times, these commonalities have been identified above all with Germanic, but there are also commonalities between several Indo-European languages, especially between Baltic, Germanic and Slavic.

5. BROADER COMMONALITIES IN THE NAME INVENTORY OF TWO OR MORE INDIVIDUAL LANGUAGES

As has become increasingly clear in recent years, the three Indo-European language groups—Baltic, Germanic and Slavic—are connected to one another to varying degrees in their name inventories. These include, on the one hand, Old European-Indo-European relics, which

³¹ Das Material steht bei Udolph 1990: 332–338.

230

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

stand out within Old European hydronymy because of particular features (morphology, vocabulary), and, on the other, correspondences between two of the three language branches. I present the material below very briefly and refer in particular to the literature cited in the individual sections.

a) Slavic-Germanic

For this area, I attempted to outline the state of research some years ago (Udolph 2005a). From a linguistic perspective, correspondences can also be demonstrated in the name inventory; the first convincing attempts were made by M. Vasmer. Two periods can be distinguished in the relations between Germanic and Slavic languages: 1) In a more recent phase, more intensive contacts arose from about the 2nd century AD, mainly as a result of the expansion of Germanic tribes in eastern Central Europe. 2) The phase of Germanic expansion was superseded from about the 5th century by Slavic settlement movements, which took possession of an area originally settled by Germanic tribes.

Only in recent years has research identified traces of a much older contact in place names within the Old Germanic settlement area, dating to the period when the Germanic and Slavic languages developed from an Indo-European dialect area. The striking phenomenon is that names apparently formed in Germanic can be etymologized only with the help of Slavic words (the following examples are given in Kirstin Casemir 2003, among others). Those cited here include Empelde near Hanover, ancient Amplithi, cf. Slavic *obl- in Polabian wûmbal ‘well’, Bulgarian vumbel, vubel, vubel’, vôbel, ubel ‘well or spring in a valley’, etc.; Hude ‘timber storage site, stacking place on a waterway, ferry crossing, harbour’, attested in numerous place names in northern Germany, probably North Sea Germanic *hūth- and earlier *hunþ-, alongside which the Slavic word *kǫtъ stands in ablaut, in Russian kut ‘the end of a river arm extending deep into the land’, etc.; Ilfeld, Ilten, Ilde, Ilsede go back to *Il-feld, *Il-tun, *Il-ithi, *Il-is-ithi; the saline soil at Ilten and Ilsede is noteworthy. A Germanic interpretation is lacking. Slavic il ‘clay, mud’ offers a straightforward explanation, for example Belarusian il ‘thin dirt of organic origin in water, on the bottom of a waterhole; swampy, grey or white-coloured land’, related to Ancient Greek ἰλύς ‘mud, dung’.

This toponymic material takes us back to an early period of linguistic contact, demonstrating that Germanic must have developed from an Indo-European dialect continuum in relative proximity not only to the Baltic, but also to the Slavic language area.

These observations also compel us, when considering the formation of Germanic

Straipsniai / Articles

231

JÜRGEN UDOLPH

in the relations of Germanic to Slavic (and Baltic), older and earlier connections are to be seen than with Celtic. This fact is still widely overlooked today. From an onomastic point of view, there are no commonalities between Germanic and Celtic.

b) Slavic–Baltic

The close relations between Slavic and Baltic have been the subject of heated and controversial debate for more than a century (one of the latest summaries of the discussions is provided by Pietro U. Dini (2005)). Until now, however, little attention has been paid to the clear fact that, in the fields of hydronymy and toponymy—unlike in vocabulary—there have been hardly any old correspondences or contacts. Baltic, as a clear center of Old European hydronymy, is integrated into the network of Indo-European hydronyms in a very different way from Slavic; the circumstances are scarcely comparable. From an onomastic point of view, Baltic has much closer ancient relations with Germanic than with Slavic.

c) Baltic–Germanic

The realization that especially close relations must have existed between these two branches of language is by no means new. As early as 1863, Ernst Förstemann (1863: 258, 331) emphasized: “No linguistic area is of greater importance to us, if we wish to reconstruct our ancient linguistic and ethnic history, than that of the so-called Baltic languages, which [...] are particularly close to Germanic [...].” “[...] because Lithuanian is in fact, among all languages, genealogically the closest relative of Germanic.” A comprehensive assessment, including the history of research, can be found in Udolph 2000, 2005a; see also Casemir, Udolph 2006; Schmid 1994: 334–357; Udolph 1999a; Udolph 2011b; Udolph 2015a.

Of particular importance here are place names and hydronyms in northern and central Germany that cannot be explained from the Germanic vocabulary, but can be elucidated only with the help of Baltic. Here I mention only briefly Ihme, a hydronym in Hanover, after 1124 in occidentali ripa Himene fluminis, 1351 supra aquam dictam Ymene, etc., < *Eimena, cf. Baltic hydronyms Eimūnis, Ejmenis (variants Eymenis, Eimenys) and Lithuanian eimenà, -õs, eimenas ‘flowing water, stream’.

One of the most important conclusions to be drawn from these striking equations is that Germanic must have developed in relative proximity to Baltic (W. P. Schmid).

d) Baltic–Slavic–Germanic

The closer kinship of Baltic, Slavic, and Germanic within the Indo-European languages has been debated for decades.

232

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

worden (dazu zuletzt: Dini 2005). Ich beschränke mich in den folgenden Ausführungen auf die Welt der Namen und referiere in aller Kürze Beobachtungen und Erkenntnisse, die ich an anderer Stelle schon mehrfach näher diskutiert habe (Udolph 2005, 2011b, 2015).

There are appellatives that can be attested only in the three language groups, sometimes only in one or two of these language groups. But names derived from them are also found outside the corresponding language group and mark—better than appellatives—the area in Europe in which Baltic, Slavic, and Germanic developed out of an Indo-European dialect area. In my opinion, the most convincing examples are the following.

Balge and related forms: Low German balge, balje ‘low-lying, swampy place, watercourse’, balge ‘tidal creek, small ditch, runnel in the tidal flats’, etc., attested as a name, among others, in Balg, 1288 du dru Balge; 1166 Balga; Balge, around 1080 Balga, etc., partly also derived from these with suffixes: 1375 Balghede < *Balg-ithi. Traces are also found in the Netherlands, Holland, and England. Related to these are Eastern European place and water names such as Bloga, tributary of the Pilica, alongside place names Blogie Stare, Szlacheckie; Blogie, a swamp in the former district of Radom; Slovak river name Blh, Hungarian Balog, 1244/1410 Balogh, etc.; Balge, a place name and the name of a part of the Vistula Lagoon (from German?); Balga, a field name in Latvia, also place name Piebalga there; Bologoe, place and lake names in northwestern Russia, as well as the Silesian water name Osobloga/ Osoblaha/ Hottzenplotz.

*dhelbh- in Polish dłubać ‘hollow out, chisel’, Czech dlub ‘depression’, Old High German bi-tel-ban ‘bury’, Old English (ge)delf ‘quarry’, Dutch delf, dilf ‘ravine, ditch, canal’, Lithuanian délba, dálba ‘breaking pole’ is attested in numerous names in northern and central Europe, among others in Delf, 1328 in loco Delf nominato; 1238 locum qui Delue dicitur; Delfen; Delft, Delft heide; Delvenau, early 9th c. Delbende (place name); Thulba (zero grade), 9th c. Tulba(m), Dulba; Delft, 1083 Delf, in Eastern Europe: Delbenen, Dílbas, Dílbi, Dílbene pl., Dłubnia, Dłubina, Dłubała, *Dolobsk-, and many others.

The distribution map of names derived from *dhelbh- (Map 40) corresponds in the spread of the names to that of other Baltic-Slavic-Germanic parallels.

Another important case is represented by place names located in the old Germanic settlement area, but which can be explained only with the help of Baltic and Slavic appellatives. The following place names in particular should be mentioned here: Ilfeld, place name north of Nordhausen, 1154 Ilevelt; Ilde near Bockenem (Hildesheim district), 1065 ad villam Illidi (*Il-ithi); Ilsede near Peine, 1053 Ilisede, < *Il-is-ithi; Miele near Hermannsburg, 14th c. Yle, 1378–79 van deme Yle; Ilten near Hanover, (1225–1247), etc. Ilthene, *Il-tūnā.

Straipsniai / Articles

233

JÜRGEN UDOLPH

MAP 40. Place and water names based on *dhelbh-

The only convincing explanation is found in Slavic and Baltic: Old Russian, Old Polish ilb ‘clay, mud’, Ukrainian il ‘mud, loam, clay, marl’, Belarusian il ‘thin dirt of organic origin in water, on the bottom of a water hole; marshy, gray- or white-colored land’, Russian il ‘mud’, etc., perhaps related to Latvian īls ‘pitch-dark’, but certainly to Ancient Greek ἰλύς ‘mud, dung’, εἰλύς · μελαν (Hesych.).

It is striking that salt deposits can be demonstrated for almost all Lower Saxon names, in some cases in salt domes extending up to the surface.

Thus, the names and their distribution show that the pre-Germanic but Indo-European substrate recognizable in hydronymy is closely connected with the Baltic hydronymic landscape, a phenomenon that we have already observed in the case of Slavic as well. If one may venture a dating: the First Germanic Sound Shift and other processes that created the foundations for Germanic from an Indo-European dialect can be dated to around 500 BC. Names found in the center of Germania that show close connections with Slavic and especially Baltic must date from before that.

234

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

be assumed. However, since these in turn overlaid Old European, pre-individual-language hydronyms, continuity can be assumed back to at least 1,000 BC.

It is a stroke of luck for science that these hypothetical approaches can apparently be confirmed. The finds in the Lichtenstein Cave near Osterode am Harz, the DNA analyses of the 3,000-year-old skeletons and of present-day inhabitants of the surrounding area have shown that continuity of settlement since 1,000 BC may be assumed. Through the investigation of family and geographical names, especially the hydronyms Sieber, Oder, Rhume and Söse, it was made probable (Udolph 2009) that this assumption can also be confirmed from an onomastic perspective. The western edge of the Harz can thus be regarded, from both a genetic and an onomastic perspective, as part of the ancient Germanic settlement area.

Old European Hydronymy and the Homeland of the Indo-Europeans

The research on the hydronyms of Europe conducted above all by H. Krahe, and their incorporation into the considerations of the prehistory of the Indo-European languages made up to that point, were subjected to rigorous criticism by W.P. Schmid, who made a decisive contribution to stricter criteria for determining what characteristics a hydronym must have in order to be awarded the designation “Old European.” Since then, the following conditions have applied:

1) It is a hydronym.

2) The name cannot be explained by the language spoken today on its banks.

3) It contains a lexeme with an Indo-European root structure and the meaning ‘water, flow, etc.’.

4) It must be explained from the entirety of the Indo-European vocabulary and its morphology, usually formed with archaic suffixes (adaptations to the structural and lexical elements of individual languages are to be expected).

5) It is called “Old European” if it has at least one corresponding name in Europe that is archaic and related in root and structure.

6) The names display the structure of Indo-European nouns, adjectives and participles; that is, they are always derived from a root, not from a word.

7) Old European hydronymy presupposes the unity of all Indo-European languages (this is demonstrated, among other things, by East Indo-European appellatives in European hydronyms).

Straipsniai / Articles

235

JÜRGEN UDOLPH

These considerations lead to the conclusion that there can be no doubt about the existence in Europe of a layer or network of pre-linguistic Indo-European hydronyms. The criticism that repeatedly flares up is usually directed at individual names, but in no way undermines the fact itself. Anyone wishing to attack or even reject the concept of Old European Hydronymy (the numerous contributions by Harald Bichlmeier published recently, with their vehement criticism, do not alter the principles outlined above) must, however, make a convincing case on one point: they must know the material that forms the basis for the considerations of H. Krahe, W.P. Schmid, and myself. Not one of the critics does.

This applies to almost all the arguments put forward so far, as well as to some more recent ones, which I will mention here very briefly.

• Recently, Celtic has once again been gaining ground, almost imperceptibly, in the assignment of a hydronym in Germany—for example, in Albrecht Greule’s comprehensive book on the hydronyms of Central Europe (2014)³². Given that East European comparative material has been, and continues to be, used to an insufficient extent, caution is warranted in assigning names to Celtic.

• Pre-Indo-European material cannot be found in many areas. This also applies to Theo Vennemann’s attempts to demonstrate Basque or Vasconic as a pre-Indo-European stratum (for the latest contributions on this, see Udolph 2013).

• From a quite different perspective, Zbigniew Babik (2001) argues against Old European material in Eastern Europe, especially in Poland. He believes that Old European Hydronymy is a unified layer extending across all of Europe, to which Polish hydronyms would not conform. There are decisive phonetic, morphological, and inflectional differences. The only thing that can be said about the old hydronyms in Poland is that they are “old and European” (stare i europejskie). This represents a view that proponents of the Old European hypothesis have long since moved beyond: we know that various suffixes, different ablaut grades, and morphological peculiarities can be observed in hydronymy, and that there is no “uniformity” in the sense of a monolithic block.

In order to find the Indo-European homeland, it is imperative first to determine the oldest settlement areas of the individual Indo-European

³² For this, see my detailed review in Beiträge zur Namenforschung, Neue Folge 52 (2017), pp. 81–105.

236

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

MAP 41. Oldest settlement areas (Slavic, Germanic, Celtic, Baltic), according to the evidence of names

language families. It is my clear view that the search for the ancient homelands of Indo-European tribes, like the (re)construction of Indo-European foundations, must proceed from the individual languages.

I venture to summarize in a map the discussion so far of the homelands of the Germanic, Slavic, Baltic, and Celtic tribes (Map 41).

What can we say about the period in which the individual Indo-European language groups developed from an Indo-European precursor?

Slavic developed from around 500 BC on the northern slope of the Carpathians. Older hydronyms point to closer connections with Germanic and Baltic, and in some cases only with Baltic. These names must consequently be dated to an earlier period, approximately 1,000–500 BC.

On the basis of names, Germanic can be assigned to an area north and east of the Harz, and also to a small extent south of it. In view of the First Sound Shift and other processes, its emergence from a

Straipsniai / Articles

237

JÜRGEN UDOLPH

The Indo-European dialect area can be dated to around 500 BC. Pre-Germanic names must be older and should be assigned to the period ca. 1,000–500 BC.

Celtic is likely to have developed out of an Indo-European dialect in the western Alpine region; its earliest beginnings may—though I know how controversial these proposed dates are—be assumed to belong to around 1,000 BC.

Baltic has undergone the least geographical displacement, although the original settlement area shrank considerably as a result of Slavic expansions. By comparing hydronyms throughout Europe, however, we can assume that, in dating the pre-Baltic remnants, we must go back at least to 500–1,000 BC.

To sum up these findings, we arrive—I think unsurprisingly—at an Indo-European-influenced stratum of hydronyms that can be attested in Central Europe around 1,000 BC and that, with some certainty, shaped eastern France, central and northern Germany, and eastern Central Europe. It nevertheless seems possible to go back one step further.

We know names in Europe that can be explained only with the help of East Indo-European appellatives (Indic, Iranian, Tocharian) and that prove that Old European hydronymy does not belong to a West Indo-European “intermediate stage” (a view consistently upheld by H. Krahe throughout his life and still often accepted today). W.P. Schmid (1994: 129ff.) recognized this and, on the basis of parallels such as sindhu–/*sindhnā – Sinn, Shin, Shannon, indu– – Indura, Indus (Baltic region), vār(i)–, vār–, vairi– – Vara, Vaire, Verna, Warme, War-menau, Wörnitz < *Varantia, etc., adu – Adda, Oder, Attersee, -gau, dānu– – Don, Danube, etc., rightly concluded: “The uniform common language presupposed by Old European hydronymy is nothing other than Indo-European itself” (Schmid 1994: 129). Since the underlying appellatives are known, among other languages, in Old Indian and Old Iranian, but are absent from the European languages, it must be assumed that the vocabulary from which European hydronyms were formed was still familiar to the creators of the Indo-European names. This takes us, with certainty, into the second millennium BC for some of these names. The uncertainty lies in the fact that we do not know when the once-living words and roots ceased to be productive in the Indo-European dialects of that time. Let us recall once again that this relationship between East Indo-European words and European names has parallels in North Germanic appellatives in continental Germanic

238

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

names and in South Slavic words found in names north of the Carpathians.

Taken together, these findings allow us to conclude that large parts of Western, Central, and Eastern Europe are covered by a network of Indo-European hydronyms, to which the Indo-European languages outside Europe and in southern Europe also contributed through their appellative equivalents. It must not be overlooked that the existence of a hydronym dating from the Indo-European period presupposes uninterrupted settlement along the watercourse. Interruptions in settlement lead to the loss of names and to discontinuities, as we can observe in southern Russia and southeastern Ukraine, where even the largest rivers in the region—Dnieper, Dniester, Southern Bug, Don, and Volga—changed their names, especially in their lower reaches, that is, in the steppe and forest-steppe regions. The question of the boundaries of Old European hydronymy remains open; however, an assessment of Europe’s Indo-European hydronyms cannot be made from their periphery (some considerations by Untermann 2009 point in this direction). The periphery can be determined only once conditions at the center are clear.

As regards the question of the ancient settlement areas of Indo-European tribes, the main implication is that these can be sought only within the distribution area of Old European hydronymy. From the period of the individual languages, which we can assess more reliably on the basis of hydronymy and toponymy, we know of no example of an Indo-European language group developing without that process leaving traces in the landscape of names. For theories that posit an Indo-European homeland in southern Russia and Ukraine (Gimbutas; Kurgan culture) or in the Caucasus, this means that traces of a migration of the Indo-European core people from there to Europe or Asia Minor must have been left in the geographical nomenclature. As everyone knows, this is not the case. Instead, I once again point to the secure attestation of East Indo-European appellatives and roots in European hydronyms: clear proof that the roots of the Indo-European tribes settled in Europe and Asia are to be sought in Europe. I repeat: the nomenclature of Europe indicates that the expansion of Indo-European tribes originated on this continent.

To strengthen these theses derived from geographical names and to confirm from another perspective the homeland of Germanic, Celtic, and Slavic that I consider probable, one discipline presents itself—a discipline that has so far played no part whatsoever in the entire discussion of the Indo-European homeland: I mean the study of soils.

Straipsniai / Articles

239

JÜRGEN UDOLPH

MAP 42. Loess distribution in Central Europe³³

I had already drawn on the findings of soil science above, when comparing the Germanic -ithi names with the yield ratings of Lower Saxony (Maps 29–32). A map of the distribution of loess soils in Asia and Europe published ten years ago (Haase et al. 2007), which I had already mentioned above in connection with the question of the homeland of Slavic tribes, shows that a broad loess-soil belt stretches from China all the way to France, across Asia and Europe. How is it distributed in Europe, particularly in the area where I suspect the ancient homelands of Germanic, Slavic, and Celtic tribes to have been?³³

For this purpose, I use an older map of loess areas in Europe, which nevertheless agrees in its results with the more recent map of 2007. On it, I have marked what I consider to be the oldest homelands of the Germanic, Celtic, and Slavic speakers (Map 42).

³³ Source: http://www.geographie.uni-stuttgart.de/seminare/lehrpfad/Pleistozän/Loessseite.htm.

240

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

The loess areas of the Leipzig Lowlands, the Magdeburg, Hildesheim, and Soest Börde regions, and the Golden Aue south of the Harz are clearly identifiable. This is precisely the area that emerged from a detailed study of ancient Germanic place and water names (Udolph 1994) as the center of continuity of Germanic name-giving. The finds in the Lichtenstein Cave correspond to this.

The map shows something similar for the border region between Poland and Ukraine. Here too, on the northern slopes of the Carpathians, loess areas and the oldest Slavic settlement areas strikingly coincide. Soil science thus supports the findings of onomastic research.

The evidence is less clear in the Western Alpine region, where we suspect ancient Celtic settlement landscapes; however, the correspondence with good soils is evident in the presumed homelands of Slavic and Germanic tribes.

This also leads to thoughts on how the enormous expansions—of the Celts, Germanic peoples, and Slavs, in that chronological order—are to be explained and understood. Since we can associate the oldest geographical names with good and very good soils, much speaks in favor of the following thesis: good soils gave people living 2,000–3,000 years ago, who depended on agriculture, livestock raising, fishing, and hunting, better harvests, minimized overall mortality and child mortality, and led to a population surplus that could be mitigated only by the gradual expansion of settlement. As the population grew, its language—e.g. Proto-Slavic, Proto-Germanic, Proto-Celtic, or the corresponding dialects—was transmitted, by virtue of its greater numbers, to neighboring areas and their inhabitants. Such processes are readily understood by comparing them with other known expansions and overlays of languages; one may think of the German eastward settlement, which overlaid West Slavic tribes and their languages, the Slavicization of Finno-Ugric tribes in Russia, and the settlement of America by European settlers and their languages.

From this point of view, I once again venture to emphasize the importance of onomastic research for the question of the oldest settlement areas, and I believe that this branch of linguistics deserves considerably more attention than it has received so far.

Results of the investigation

1) There are hydronyms that can be dated to the period when the Indo-European languages arose and developed. They are located in Europe.

Straipsniai / Articles

241

JÜRGEN UDOLPH

2) The decisive processes in the development and spread of the Indo-European languages took place mainly in the area of Old European hydronymy. Evidence for this includes, among other things, East Indo-European appellatives in European hydronyms. Corresponding parallels are found in Slavic and Germanic.

3) A pre-Indo-European layer cannot be demonstrated in the names of Central Europe. This is surprising, since people who did not speak an Indo-European language must surely already have settled there. I dare not judge whether this is connected with the settled way of life of Indo-European tribes, perhaps in contrast to the nomadic or semi-nomadic way of life of the earlier population.

4) The settlements of Indo-European tribes as well, in the last millennia before Christ, are associated with good and very good soils. This coincides—and this is of considerable significance—with the distribution of the ancient names.

5) The reconstruction of Indo-European proto-forms is based on knowledge of the known individual Indo-European languages. However, Indo-European hydronyms are also attested in the “gaps” between the individual Indo-European languages. It is therefore to be expected that there are differences between the proto-forms of the names and the proto-forms of Indo-European reconstruction.

6) Old European hydronyms have an unequivocal center in the Baltic region and in the Baltic languages. There is much to suggest that this should be seen not only as a center of continuity (W.P. Schmid), but also as a center of diffusion.

7) The homeland of the Germanic tribes lies within the Germanic settlement area, the homeland of the Slavs within the Slavic settlement area, and the homeland of the Celts within the historically attested Celtic settlement area. Everything suggests that this should also be assumed in the case of the presumed homeland of the Indo-European tribes. It follows that:

8) From the perspective of name studies, the proposal that the original homeland of the Indo-European languages was in southern Russia, Asia Minor, or the Caucasus must be decisively rejected. The homeland can only be sought within Old European hydronymy. This is clearly confined to Europe. Taken together, everything points to the Baltic region as the homeland and point of departure for Indo-European expansion.

242

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

LITERATURE

Alemannien und der Norden 2004: Alemannien und der Norden. Internationales Symposium in Zürich, hrsg. v. Hans-Peter Naumann (= Reallexikon der Germanischen Altertumskunde, Ergänzungsband 43). Berlin: De Gruyter.

Atlas zur Geschichte und Landeskunde von Sachsen 1998: Atlas zur Geschichte und Landeskunde von Sachsen. F: Wirtschaft (Landwirtschaft, Bergbau, Gewerbe, Verkehr). IV: Neueste Zeit. Leipzig: Verlag der Sächsischen Akademie der Wissenschaften.

Ascher Diana 2015: Die Siedlungsgeschichte der Rhön aus namenkundlicher Sicht. – Die Rhön – Geschichte einer Landschaft. Petersberg, 201–215.

Babik Zbigniew 2001: Najstarsza warstwa nazewnicza na ziemiach polskich. Kraków: Universitas.

Bijak Urszula 2013: Nazwy wodne dorzecza Wisły. Potamonimy odapelatywne. Kraków: Wydawnictwo FALL.

Billy Pierre-Henry 1993: Thesaurus linguae Gallicae. Hildesheim: Olms.

Bischoff Karl 1975: Germ. *haugaz ‘Hügel, Grabhügel’ im Deutschen. Eine Flurnamenstudie. Mainz: Akademie der Wissenschaften und der Literatur.

Brather Sebastian 2004: Ethnische Interpretationen in der frühgeschichtlichen Archäologie. Geschichte, Grundlagen und Alternativen. Berlin: de Gruyter.

Busse Peter 2007: Hydronymie und Urheimat: Ein neuer Ansatz zur Lokalisierung der Urheimat der Kelten. – Kelten-Einfälle an der Donau. Akten des 4. Symposiums Deutschsprachiger Keltologinnen und Keltologen vom 17.–21. Juli 2005 in Linz. Wien: Verlag der Österreichischen Akademie der Wissenschaften, 89–98.

Casemir Kirstin 1997:Die Ortsnamen auf-büttel. Leipzig: Leipziger Universitätsverlag.

Casemir Kirstin 2003: Die Ortsnamen des Landkreises Wolfenbüttel und der Stadt Salzgitter. Bielefeld: Verlag für Regionalgeschichte.

Casemir Kirstin, Udolph Jürgen 2006: Die Bedeutung des Baltischen für die niedersächsische Ortsnamenforschung. – Baltų onomastikos tyrimai. Aleksandrui Vanagui atminti, sud. Laimutis Bilkis, Alma Ragauskaitė, Daiva Sinkevčiūtė. Vilnius: Lietuvių kalbos institutas, 114–136.

Collis John 2003: The Celts: Origins, Myths and Inventions. London: Tempus Pub Ltd.

Delamarre Xavier 2003: Dictionnaire de la langue gauloise; une approche linguistique du vieux-celtique continental. 2. éd. Paris: Éd. Errance.

Dini Pietro U. 2005: Slawisch-baltische Sprachbeziehungen. – Reallexikon der Germanischen Altertumskunde 29. Berlin – New York, 73–78.

Straipsniai / Articles

243

JÜRGEN UDOLPH

Falileyev Alexander 2006/7: Dictionary of Continental Celtic Place-Names, Aberystwyth = http://cadair.aber.ac.uk/dspace/bitstream/2160/282/5/ContCelticPNDictionary.pdf; Einleitung: http://cadair.aber.ac.uk/dspace/bitstream/2160/282/6/IntroAndElements.pdf; Auswahlbibliographie: http://cadair.aber.ac.uk/dspace/bitstream/2160/282/4/FalileyevBibl.pdf; Karte: http://cadair.aber.ac.uk/dspace/bitstream/2160/282/7/FalileyevMap.pdf.

Förstemann Ernst 1863: Die deutschen Ortsnamen. Nordhausen: Ferd. Förstemann´s Verlag.

Gimbutas Marija 1963: The Balts. London: Thames and Hudson.

Gołąb Zbigniew 1992: The Origins of the Slavs. A Linguist’s View. Columbus (Ohio): Slavica Publ.

Greule Albrecht 2014: Deutsches Gewässernamenbuch. Etymologie der Gewässernamen und der dazugehörigen Gebiets-, Siedlungs- und Flurnamen. Berlin/Boston: De Gruyter.

Grimm Jacob 1826: Deutsche Grammatik. Zweiter Teil. Göttingen: Dieterich.

Grimm Jacob 1848: Geschichte der deutschen Sprache. Leipzig: Hirzel.

Haase Dagmar, D. Haase, J. Fink, G. Haase, R. Ruske, M. Pécsi, H. Richter, M. Altermann, K.-D. Jäger 2007: Vornamen Loess in Europe—its spatial distribution based on a European Loess Map, scale 1:2,500,000. – Quaternary Science Reviews 26(9–10), 1301–1312.

James Simon 1999: The Atlantic Celts. Ancient people or modern invention. Madison Wis.: University of Wisconsin Press.

Kettner Bernd-Ulrich 1972: Flußnamen im Stromgebiet der oberen und mittleren Leine. Rinteln: Verlag C. Bösendahl.

Kolb Eduard 1957: Alemannisch-nordgermanisches Wortgut. Huber: Frauenfeld.

Krahe Hans 1949–1950: Alteuropäische Flußnamen. – Beiträge zur Namenforschung, 24–51.

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

Krahe Hans, Meid Wolfgang 1967: Germanische Sprachwissenschaft 3: Wortbildungslehre. Berlin: Walter de Gruyter

Kunstmann Heinrich 1996: Die Slaven. Ihr Name, ihre Wanderung nach Europa und die Anfänge der russischen Geschichte in historisch-onomastischer Sicht. Stuttgart: Steiner.

Leibnitz Gottfried Wilhelm 1882: Die philosophischen Schriften, hrsg. v. C.I. Gerhard, Bd. 5. Berlin: Weidmann.

244

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

Matasović Ranko 2009: Etymological dictionary of proto-Celtic. Leiden: Brill.

Meyer Brunk 1978: Bodenkunde und Siedlungsforschung. – Reallexikon der Germanischen Altertumskunde 3. Berlin – New York: Walter de Gruyter, 117–120.

Möller Reinhold 1992: Dentalsuffixe in niedersächsischen Siedlungs- und Flurnamen in Zeugnissen vor dem Jahre 1200. Heidelberg: Carl Winter Universitätsverlag.

Möller Reinhold 2000: Niedersächsische Siedlungsnamen und Flurnamen mit k-Suffix und s-Suffix in Zeugnissen vor dem Jahr 1200. Heidelberg: Carl Winter Universitätsverlag.

Müller Horst M., Kutas Marta 1997: Die Verarbeitung von Eigennamen und Gattungsbezeichnungen. Eine elektrophysiologische Studie. – Studien zur Klinischen Linguistik: Modelle, Methoden, Interventionen, hrsg. v. Gert Rickheit. Opladen: Springer, 147–169.

Rozwadowski Jan 1948: Studia nad nazwami wód słowiańskich. Kraków: Nakładem Polskiej Akademii Umiejętności.

Rymut Kazimierz, Majtán Milan 1998: Gewässernamen im Flußgebiet des Dunajec. Stuttgart: Steiner.

Scheffer Fritz 1978: Boden in Mitteleuropa. – Reallexikon der Germanischen Altertumskunde 3. Berlin – New York: Walter de Gruyter, 108–117.

Schlüter Otto 1958: Die Siedlungsräume Mitteleuropas in frühgeschichtlicher Zeit 2: Das mittlere und nordöstliche Mitteleuropa. Remagen: Selbstverlag der Bundesanstalt für Landeskunde.

Schmid Wolfgang P. 1976: Baltische Sprachen und Völker. – Reallexikon der Germanischen Altertumskunde 2. Berlin – New York: Walter de Gruyter, 14–20.

Schmid Wolfgang P. 1988: Zu einigen keltisch-baltischen Namenentsprechungen. – Studia Indogermanica et Slavica. Festgabe für Werner Thomas. München: Sagner, 49–56.

Schmid Wolfgang P. 1994: Linguisticae Scientiae Collectanea. Ausgewählte Schriften. Berlin – New York: de Gruyter.

Schmid Wolfgang P. 2004: Versuch über den Namen der Loire. – Sprache, Sprechen, Sprichwörter. Festschrift für D. Stellmacher zum 65. Geburtstag. Stuttgart: Steiner, 291–295.

Sims-Williams Patrick 2006: Ancient Celtic Place-Names in Europe and Asia Minor. Oxford: Blackwell.

Słownik Prasłowiański 1974: Słownik Prasłowiański 1. Warszawa: Zakład Narodowy im. Ossolińskich.

Straipsniai / Articles

245

JÜRGEN UDOLPH

Stams Werner 1998: Böden nach Bodenwerten. Beiheft zu: Atlas zur Geschichte und Landeskunde von Sachsen. F: Wirtschaft (Landwirtschaft, Bergbau, Gewerbe, Verkehr). IV: Neueste Zeit. Leipzig: Verlag der Sächsischen Akademie der Wissenschaften.

Toporov Vladimir N., Trubačev Oleg N. 1962: Lingvističeskij analiz gidronimov Podneprov’ja. Moskva: Izdatel’stvo Akademii Nauk SSSR.

Trubačev Oleg N. 1968: Nazvanija rek pravoberežnoj Ukrainy. Moskva: Izdatel’stvo Nauka.

Trubačev Oleg N. 1983: Jazykoznanie i etnogenez slavjan (Drevnie slavjane po dannym ėtimologii i onomastiki). – Slavjanskoe jazykoznanie. IX Meždunarodnyj s-ezd slavistov. Doklady sovetskoj delegacii. Moskva: Nauka 231–270.

Udolph Jürgen 1979: Studien zu slavischen Gewässernamen und Gewässerbezeichnungen. Heidelberg: Carl Winter Universitätsverlag.

Udolph Jürgen 1981: Ex oriente lux – Zu einigen germanischen Flußnamen. – Beiträge zur Namenforschung, Neue Folge 16, 84–106 (Nachdruck in: Reader zur Namenkunde 3(2) (= Germanistische Linguistik 131–133), Hildesheim, 1996, 671–692).

Udolph Jürgen 1985: Ex oriente lux – auch in deutschen Flurnamen. – Gießener Flurnamenkolloquium. Heidelberg: Winter Verlag, 272–298.

Udolph Jürgen 1988: Kamen die Slaven aus Pannonien? – Studia nad etnogenezą Słowian i kulturą Europy wczesnośredniowiecznej. Wrocław: Zakład Narordowy im. Ossolińskich, 168–173.

Udolph Jürgen 1990: Die Stellung der Gewässernamen Polens innerhalb der alteuropäischen Hydronymie. Heidelberg: Carl Winter Universitätsverlag.

Udolph Jürgen 1991: Die Ortsnamen auf –ithi. – Probleme der älteren Namenschichten. Heidelberg: Carl Winter Universitätsverlag, 85–145.

Udolph Jürgen 1994: Namenkundliche Studien zum Germanenproblem. Berlin – New York: Walter de Gruyter.

Udolph Jürgen 1997: Alteuropäische Hydronymie und urslavische Gewässernamen. – Onomastica 42, 21–70.

Udolph Jürgen 1998: Typen urslavischer Gewässernamen. – Praßłowiańszczyzna i jej rozpad. Warszawa: Energeia 275–294.

Udolph Jürgen 1999: Die Schichtung der Gewässernamen in Pannonien. – Ural-Altaische Jahrbücher, Neue Folge 15, 90–106.

Udolph Jürgen 1999a: Baltisches in Niedersachsen? – Florilegium Linguisticum. Festschrift für Wolfgang P. Schmid zum 70. Geb. Frankfurt/Main [usw.]: Lang, 493–508.

246

Acta Linguistica Lithuanica LXXVI

Heimat und Ausbreitung indogermanischer Stämme im Lichte der Namenforschung

Udolph Jürgen 2000: Nordisches in niedersächsischen Ortsnamen. – Raum, Zeit, Medium – Sprache und ihre Determinanten. Festschrift für Hans Ramge. Marburg: Hessische Historische Kommission, 59–79.

Udolph Jürgen 2000a: Slavjano-germanskie svjazi v severno-nemeckich toponimach. – Ètimologija 1997–1999. Moskva, 185–191.

Udolph Jürgen 2004: Nordisches in deutschen Ortsnamen. – Namenwelten. Orts- und Personennamen in historischer Sicht. Berlin – New York: Walter de Gruyter, 359–371.

Udolph Jürgen 2004a: Alemannien und der Norden aus der Sicht der Ortsnamenforschung. – Alemannien und der Norden, 29–56.

Udolph Jürgen 2005: Slavisch-Baltisch-Germanische Übereinstimmungen in Toponymie und Hydronymie. – Reallexikon der Germanischen Altertumskunde 29. Berlin – New York: Walter de Gruyter, 64–67.

Udolph Jürgen 2005a: Germanisch-Slawische Sprachbeziehungen. – Reallexikon der Germanischen Altertumskunde 29. Berlin – New York: Walter de Gruyter, 69–73.

Udolph Jürgen 2008: Ortsnamen und Wanderungen der Völker. – Jahrbuch der Akademie der Wissenschaften zu Göttingen 2008. Berlin – New York: Walter de Gruyter, 392–397.

Udolph Jürgen 2009: Liechtensteinhöhle, Siedlungskontinuität und das Zeugnis der Familien-, Orts- und Gewässernamen. – Historia archaeologica. Festschrift f. Heiko Steuer. Berlin – New York: Walter de Gruyter, 85–105.

Udolph Jürgen 2009/2011: Les hydronymes paléoeuropéens et la question de l’origine des Celtes. – Nouvelle Revue d’Onomastique 51, 85–121.

Udolph Jürgen 2010: Morphologie germanischer Toponyme. – Proceedings of the 21st International Congress of Onomastic Sciences 5. Uppsala: Institutet för språk och folkminnen, 254–267.

Udolph Jürgen 2010a: Skandinavische Wörter in deutschen Ortsnamen. – Probleme der Rekonstruktion untergegangener Wörter aus alten Eigennamen. Uppsala: Swedish Science Press, 141–158.

Udolph Jürgen 2010b: Skandinavische Wörter in deutschen Ortsnamen. – Probleme der Rekonstruktion untergegangener Wörter aus alten Eigennamen. Uppsala: Kungl. Gustav Adolfs Akademien für svensk folkkultur, 141–158.

Udolph Jürgen 2011b: „Baltisches“ und „Slavisches“ in norddeutschen Ortsnamen. – Interferenz-Onomastik. Namen in Grenz- und Begegnungsräumen in Geschichte und Gegenwart. Saarbrücken: Kommission für Saarländische Landesgeschichte und Volksforschung, 313–331.

Straipsniai / Articles

247

JŪRGEN UDOLPH

Udolph Jürgen 2013: (Hrsg.) Europa Vasconica – Europa Semitica? Kritische Beiträge zur Frage nach dem baskischen und semitischen Substrat in Europa. Hamburg: Baar-Verlag.

Udolph Jürgen 2015: Rezension zu: W. Wenzel, Namen und Geschichte. Orts- und Personennamen im deutsch-slavischen Sprachkontaktraum als historische Zeugnisse, Hamburg 2014. – Beiträge zur Namenforschung, Neue Folge 51, 242–252.

Udolph Jürgen 2015a: Baltisch, Slavisch, Germanisch – Kontakte und Beziehungen aus der Sicht der Onomastik. – Early Germanic Languages in Contact. Amsterdam – Philadelphia: Benjamins, 39–74.

Udolph Jürgen 2016: Expansion slavischer Stämme aus namenkundlicher und bodenkundlicher Sicht. – Onomastica 60, 215–231.

Udolph Jürgen 2016a: Heimat und Ausbreitung slavischer Stämme aus namenkundlicher Sicht. – Die frühen Slawen – von der Expansion zu gentes und nationes 1: Beiträge zum Schwerpunktthema. Langenweissbach: Beier & Beran, 27–51.

Vasmer Max 1971: Schriften zur slavischen Altertumskunde und Namenkunde 1, hrsg. v. H. Bräuer. Berlin – Wiesbaden: Harrassowitz.

Wenzel Walter 2014: Namen und Geschichte. Orts- und Personennamen im deutsch-slavischen Kontaktraum als historische Zeugnisse. Hamburg: Baar-Verlag.

The homeland and spread of Indo-European tribes in the light of onomastic research

SUMMARY

Drawing on onomastic data, Jürgen Udolph summarizes in this study the results of research accumulated over many years on the Indo-European tribes and their homelands. He draws important conclusions with which one may agree or disagree, but which should not be disregarded. In the author’s view, water names dating to the period when the Indo-European languages emerged and developed are concentrated in Europe. The decisive processes in the development and spread of the Indo-European languages took place essentially within the boundaries of Old European hydronymy. Among other things, this is also indicated by appellatives of the eastern Indo-European languages embedded in European water names. No corresponding parallels are found in Slavic and Germanic. The names of Central Europe do not attest to a primordial Indo-European stratum. This is truly surprising, since people who spoke no Indo-European language must certainly already have lived there. The author does not venture to decide whether this is connected with the sedentary way of life of the Indo-European tribes, in contrast to the nomadic or semi-nomadic way of life of earlier inhabitants. In the first millennia before Christ, the emergence of Indo-European settlements was closely connected with soil quality. The distribution of ancient names also clearly confirms this. Authentic

248

Acta Linguistica Lithuanica LXXVI

The homeland and spread of Indo-European tribes in the light of onomastic research

The reconstruction of Indo-European genders is based on evidence from individual known Indo-European languages. Indo-European water names are attested in individual Indo-European languages. It is therefore likely that there are differences between the genders of authentic names and the reconstruction of authentic Indo-European genders. The centre of Old European water names is unequivocally the Baltic region and the Baltic languages. Much evidence indicates that this was not only an area of continuity (Wolfgang P. Schmid), but also a centre from which names spread. The homeland can be sought only in the lands of Old European hydronymy, which lie in Europe. The homeland of the Germanic tribes is where the Germanic peoples lived, the homeland of the Slavs where the Slavs lived, and the homeland of the Celts where the Celts are attested by history as having lived. It follows that, from the perspective of onomastics, the formation of the Indo-European homeland in southern Russia, Asia Minor, or the Caucasus must certainly be rejected. In summary, one may conclude that the homeland of Indo-European expansion and its area of diffusion is the Baltic region.

Received on 12 June 2017.

JÜRGEN UDOLPH

Zentrum für Namenforschung – Leipzig Peterssteinweg 3, D-4107 Leipzig Akademie der Wissenschaften zu Göttingen

Arbeitsstelle Ortsnamen zwischen Rhein und Elbe – Onomastik im europäischen Raum Robert-Koch-Str. 40, D-48149 Münster [email protected]

Straipsniai / Articles

249