sign in sign up
<<
Home , Languages of Europe, Sprachbund

1492 XIV. Typological characterization of families and linguistic areas

107. The European linguistic area: Standard Average European

1. Introduction guages share structural features which cannot 2. The major SAE features be due to retention from a common proto- 3. Some further likely SAE features language and which give these a 4. Degrees of membership in SAE profile that makes them stand out among the 5. How did SAE come into being? 6. Abbreviations of language names surrounding languages. There is thus no min- 7. References imum number of languages that a linguistic area comprises (pace Stolz 2001a). In prin- ciple, there could be a linguistic area con- 1. Introduction sisting of just two languages (though this would be rather uninteresting), and there This article summarizes some of the main are also very large (-sized) linguistic pieces of evidence for a linguistic area (or areas (Dryer 1989a). Likewise, there is no ) in that comprises the minimum number of structural features that Romance, Germanic and Balto-Slavic lan- the languages must share in order to qualify guages, the Balkan languages, and more mar- as a Sprachbund. For instance, Jakobson ginally also the westernmost Finno-Ugrian (1931) establishes his “Eurasian linguistic languages (these will be called core European area” on the basis of just two phonological languages in this article). This linguistic area features, but of course an area that shares is sometimes called Standard Average Euro- more features is more interesting. As will be pean (abbreviated SAE), following Whorf shown below, Standard Average European (1941) [1956: 138]. The existence of this lin- languages share over a dozen highly charac- guistic area is a relatively new insight (cf. teristic features, so we are dealing with a very Bechert et al. 1990, Bernini & Ramat 1996, interesting Sprachbund. Haspelmath 1998, van der Auwera 1998, Kö- A linguistic area is particularly striking nig & Haspelmath 1999). when it comprises languages from genealog- While the close syntactic parallels among ically unrelated languages (like the South the Balkan languages have struck linguists Asian linguistic area (J Art. 109), or the since the 19th century and the existence of Mesoamerican linguistic area (J Art. 110)), a Balkan Sprachbund has been universally but this is not a necessary feature of a accepted, the European linguistic area has Sprachbund. The Balkan languages are all long been overlooked. This may at first ap- Indo-European, but they are from different pear surprising, because the members of the families within Indo-European (Romance, Sprachbund are among the best studied lan- Slavic, Greek, Albanian), and not all lan- guages of the world. However, it is easy to guages of these families belong to the Balkan understand why linguists have been slow to linguistic area, so nobody questions the va- appreciate the significance of the similarities lidity of the Balkan Sprachbund (J Art. 108). among the core European languages: Since In the case of SAE, three entire branches most comparative linguists know these lan- of Indo-European (Romance, Germanic and guages particularly well, they have tended to Balto-Slavic) belong to the linguistic area. see non-European languages as special and However, here too it is clear that we are unusual, and the similarities among the not dealing with a genealogical grouping, European languages have not seemed sur- because nobody ever proposed a branch of prising. Thus, it was only toward the end of Indo-European that consists of precisely the 20th century, as more and more had be- these three families. On the contrary, Indo- come known about the grammatical proper- Europeanists typically assume a particularly ties of the languages of the rest of the world, close genealogical relationship between Italic that linguists realized how peculiar the core and Celtic (and sometimes even an Italo- European languages are in some ways when Celtic protolanguage), but Romance (the sole seen in the world-wide context. From this descendant of Italic) is inside SAE, while the perspective, Standard Average European may do not belong to SAE. And even appear as an “exotic language” (Dahl since so much is known about the grammat- 1990). ical properties that Proto-Indo-European A linguistic area can be recognized when must have possessed, it is fairly easy to test a number of geographically contiguous lan- whether an SAE feature is an Indo-Euro-

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1493 peanism or not. As was shown in Haspelmath varieties, but Basque seems to show very few (1998), most of the characteristic SAE fea- of them. Somewhat further to the east, Geor- tures (also called Europeanisms here) are not gian in the southern (and perhaps Indo-Europeanisms but later common inno- the other ) shares a vations. surprising number of features with the core Thus, what needs to be shown in order to European languages. These impressionistic demonstrate that a structural feature is a statements should eventually be quantified, Europeanism is but since it is not clear how much weight should be attached to each feature, this is not (i) that the great majority of core European straightforward. languages possesses it; All of the features discussed below are syn- (ii) that the geographically adjacent lan- tactic, or concern the existence of certain guages lack it (i. . Celtic in the west, morphosyntactic categories. I am not aware Turkic, eastern Uralic, Abkhaz-Adygh- of any phonological properties characteristic ean and Nakh-Daghestanian in the east, of the core European languages (cf. Jakob- and perhaps Afro-Asiatic in the south); son 1931: 182: “do six por ne udalos’ najti ni (iii) that the eastern Indo-European lan- odnogo obsˇcˇeevropejskogo … polozˇitel’nogo guages lack it (Armenian, Iranian, In- fonologicˇeskogo priznaka [so far not a single dic); and Europe-wide positive phonological feature has (iv) that this feature is not found in the ma- been found]”). Perhaps phonologists have jority of the world’s languages. not looked hard enough, but at least one ma- Particularly the last point is not easy to de- jor recent study of word prosody in Euro- monstrate for many features because there pean languages has not found any phonolog- are still far too few representative world-wide ical evidence for Standard Average European studies of grammatical structures, so to the (van der Hulst et al. 1999, especially Maps extent that our knowledge about the world’s 1Ϫ4) (but cf. Pisani 1969). A few generaliza- languages is incomplete and biased, we can- tions are discussed by Ternes (1998), but he not be sure about the European linguistic finds that in most respects European lan- area. In this article, I will cite whatever in- guages are unremarkable from a world-wide formation is available, and sometimes I will perspective. Perhaps the only features worth have to resort to impressionistic observa- mentioning are the relatively large vowel in- tions. ventories (no 3-vowel or 4-vowel inventories) The designation “core European lan- and the relatively common consonant clus- guage” for members of SAE is deliberately ters (no restriction to CV ). In these vague, because the European linguistic area respects, European languages are not average, does not have sharp boundaries. It seems but they are by no means extreme either. possible to identify a nucleus consisting of continental West (e. g. 2. The major Standard Average Dutch, German) and Gallo-Romance (e. g. French, Occitan, northern Italo-Romance). European features For this set of languages, van der Auwera In this section I will discuss a dozen gram- (1998a: 824) proposes the name matical features that are characteristic of the Sprachbund. Of the other languages, those core European languages and that together which are geographically further from this define the SAE Sprachbund. In each case I center also seem to share significantly fewer will briefly define the feature and give a few SAE features, i. e. Ibero-Romance, insular examples from SAE languages. Then a name Scandinavian (Icelandic and Faroese), East map, which indicates the approximate loca- Slavic (Russian, Ukrainian, Belorussian) and tion of languages by the arrangement of (ab- Baltic. Even English, a West Germanic lan- breviated) language names, shows the distri- guage, is clearly not within the nucleus. Of bution of the various feature values within the non-Indo-European , Europe. In each case it can be observed that the western (i. e. Hungarian the nuclear SAE languages are within the and Balto-Finnic) are at least marginal mem- SAE , and that the marginal lan- bers of Standard Average European; they are guages tend to be outside the isogloss to a in many ways strikingly different from east- greater or lesser extent. (Part of the material ern Uralic. Maltese also exhibits a number of presented here was already included in Has- Europeanisms not shared by other pelmath 1998.)

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1494 XIV. Typological characterization of language families and linguistic areas

2.1. Definite and indefinite articles 2.2. Relative clauses with relative pronouns Both a definite and an indefinite article (e. g. The type of relative clause found in languages English the book/a book; J Art. 62) exist in such as German, French or Russian seems to all Romance and almost all Germanic lan- be unique to Standard Average European guages plus some of the Balkan languages languages. It is characterized by the follow- (, perhaps Albanian and Bul- ing four features: The relative clause is post- garian), but not outside Standard Average nominal, there is an inflecting relative pro- European. To be sure, their forms and syn- noun, this pronoun introduces the relative tactic behavior show considerable diversity clause, and the relative pronoun functions as (see Nocentini 1996 for an overview), but a resumptive, i. e. it signals the head’s role their very existence is characteristic enough. within the relative clause (cf. Lehmann 1984: The distribution of articles in European lan- 103Ϫ109, Comrie 1998). In English, a rela- guages is shown in Map 107.1. (Abbrevi- tive construction like the suspicious woman ations of language names are given in the Ap- whom I described also displays all these fea- pendix.) tures. Furthermore, in most SAE languages the relative pronoun is based on an interrog- ative pronoun (this is true of all Romance, all Slavic and some Germanic languages, Mod- ern Greek, as well as Hungarian and Geor- gian). (Languages like German, whose rela- tive pronoun is based on a demonstrative, or Finnish, which has a special relative pro- noun, are not common.) The geographical distribution of the relative pronoun strategy is shown in Map 107.2.

------definite and indefinite article present - - - - only definite article present Map 107.1: Definite and indefinite article

In large parts of there are no articles at all (East Slavic, West Slavic, Finno-Ugrian other than Hungarian, Turkic, Nakh-Daghestanian, Kartvelian). Some neigh- boring non-SAE languages do have definite articles (e. g. Celtic, Semitic, Abkhaz, Mord- vin), and Turkish has an indefinite article, but no neighboring non-SAE language has ------relative clause with introducing relative pro- both definite and indefinite articles. The only noun exception among Germanic languages, Ice- - - - - only particle relative clause landic (which only has definite articles like Map 107.2: Two relative clause types in Europe nearby Celtic), is also the most peripheral Germanic language geographically. We can The only other type that is widespread in also be certain that the existence of definite Europe is the postnominal relative clause and indefinite articles is not an Indo-Euro- introduced by a relative particle (Lehmann peanism: The Iranian and Indic languages 1984: 85Ϫ87), which often occurs in the same have generally lacked articles throughout language beside the resumptive relative pro- their . noun type just described (an English example World-wide, articles are not nearly as would be the radio that I bought). Particle common as in Europe: According to Dryer’s relatives of this type exist in most Slavic and (1989b: 85) findings, “it appears that about a , as well as in Scandina- third of the languages of the world employ vian languages and Modern Greek, but also articles” (125 out of a sample of about 400 in Welsh and Irish (Lehmann 1984: 88Ϫ90). languages). Only 31 languages of those in The relative particle is sometimes difficult to Dryer’s sample (i. e. less than 8%) have both distinguish from a degenerate resumptive definite and indefinite articles. pronoun, and in many European languages

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1495 it developed from a relative pronoun through In contrast to the languages just mentioned, the gradual loss of inflectional distinctions. in Slavic, Finno-Ugrian and Armenian the However, this also means that the relative perfect is usually based on a participial clause loses its specifically European flavor, construction with an active participle and a because particle relatives are also attested copula (e. g. Finnish ole-n saa-nut [be-1sg widely elsewhere in the world (e. g. in Per- receive-ptcp] ‘I have received’). Hungarian sian, Modern Hebrew, Nahuatl, Indonesian, seems to lack a perfect completely. In some Yoruba, and Thai, cf. Lehmann 1984: 85Ϫ Nakh-Daghestanian languages (e. g. Lezgian 97). and Godoberi), the perfect is formed on the However, the relative pronoun strategy basis of the past converb plus the copula. clearly is typically European. It is not found Georgian comes closest to the SAE prototype in the eastern Indo-European languages, and in that its transitive perfect is based on a pas- as Comrie (1998: 61) notes, “relative clauses sive participle, but this is combined with the formed using the relative pronoun strategy copula rather than the transitive verb ‘have’, are quite exceptional outside Europe, except so that the perfect has a quasi-passive struc- as a recent result of the influence of Euro- ture, with the agent in the dative case (‘The pean languages … The relative pronoun letter is-written to-me’, rather than ‘I have- strategy thus seems to be a remarkable areal written the letter’). In Welsh, the perfect is typological feature of European languages, formed with the preposition wedi ‘after’ (‘She especially the standard written languages”. is after selling the house’ for ‘She has sold the house’). The eastern Indo-European lan- 2.3. ‘Have’-perfect guages also lack a ‘have’-perfect (for in- Another well-known feature typical of SAE stance, both Persian and Hindi/ have a languages is the (transitive) perfect formed by perfect based on a participle plus the copula, ‘have’ plus a passive participle (e. g. English somewhat like Slavic and Armenian). I have written, Swedish jag har skrivit, Span- Dahl (1995, 1996: 365), taking a global perspective, notes that the ‘have’-perfect is al- ish he escrito; J Art. 59). A perfect of this most exclusively found in Europe. Now one kind exists in all Romance and Germanic lan- might object that this is not a primitive fea- guages plus some of the Balkan languages ture of European languages. Many languages (Albanian, Modern Greek, Macedonian), and do not use a transitive ‘have’-verb for indi- also in Czech (Garvin 1949: 84). These per- cating predicative possession at all, and it has fects do not all mean the same thing, because in fact been suggested that the very existence they are at different stages in the grammati- of a transitive verb of predicative possession calization process: in French and German, is a Europeanism (e. g. Lazard 1990: 246Ϫ47; the perfect can be used as a normal perfective Benveniste 1960 [1966: 195]: “L’expression la past, including the function of a narrative plus courante du rapport indique´ dans nos tense, while in Spanish, English and Swedish langues par avoir s’e´nonce a` l’inverse par eˆtre the perfect has a distinct present-anterior a` … Telle est la situation dans la majorite´ des meaning. What is important here is that they langues.”) The restriction of a ‘have’-perfect all must have had basically the same meaning to Europe would then be just a consequence when they were first created. The geographi- of this (cf. Dahl 1990: 7). However, so far no cal distribution of ‘have’-perfects in Europe published research has documented an areal is shown in Map 107.3. restriction for ‘have’ verbs. From Heine’s (1997: 47Ϫ50, 240Ϫ44) survey of predicative possessive constructions, not much support can be drawn for such a claim. Still, this is an interesting idea to be addressed by further research. If ‘have’-verbs turn out to be typi- cal of Europe, that would fit with the ten- dency of European languages to have nomi- native experiencers in experiential verbs (see the next section). 2.4. Nominative experiencers There are two ways of expressing experiencer arguments of verbs of sensation, emotion, Map 107.3: ‘Have’-perfects in Europe cognition and perception: The experiencer

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1496 XIV. Typological characterization of language families and linguistic areas may be assimilated to agents and coded as European is fairly clear: Indic languages are a nominative subject (e. g. I like it), or it may well-known for their “dative subjects” of be assimilated to a patient or goal, so that experiencer verbs, so again the feature is the stimulus argument is coded as the nomi- not genetic (see also Masica 1976, especially native subject (e. g. It pleases me). In Bos- Map 6, for the areal distribution of dative song’s (1998) typology, the first type is called subjects in and northern ). generalizing, and the second type is called (See Haspelmath 2001 for more discussion inverting. Bossong studies the expression of of experiential predicates in European lan- ten common experiential predicates in 40 guages.) European languages. He computes the rela- tion between inverting predicates and gener- 2.5. Participial passive alizing predicates, arriving at figures between Standard Average European languages typi- 0.0 for English (where all predicates are cally have a canonical passive construction generalizing) and 5.0 for Lezgian (where all (J Art. 67) formed with a passive participle predicates are inverting). By arbitrarily divid- plus an intransitive copula-like verb (‘be’, ing the languages into those showing pre- ‘become’, or the like). In this passive the dominant generalization (ratios between 0.0 original direct object becomes the subject and and 0.8) and those showing predominant the original subject may be omitted, but it inversion (ratios between 0.8 and 5.0), we may also be expressed as an adverbial agent arrive at the geographical pattern shown in phrase. Such constructions occur in all Ro- Map 107.4. mance and Germanic languages, but also in

Map 107.4: Predominant generalization (center) vs. inversion (periphery)

Thus, Bossong’s study basically confirms all Slavic (including East Slavic) and Balkan earlier claims (Lazard 1990: 246Ϫ47, Dahl languages, as well as in Irish. The geographi- 1990: 7) that the generalizing type is charac- cal distribution of such participial passives is teristic of SAE, although some of the fig- shown in Map 107.5. ures are perhaps a bit surprising (e. g. the fact that Hungarian turns out to be more SAE than German or Dutch, and the inclusion of Turkish, but not Romanian or Albanian, with respect to this feature). It is not possible to explain everything here, but we evidently have before us a fairly typical SAE pattern with French and English at the center, Celtic (plus Icelandic this time) at the western mar- gin, Balto-Slavic, Finno-Ugrian and Cauca- sian at the eastern margin, and fairly gradual transitions within the macro-areas. No sys- tematic world-wide studies have been made, but at least the behavior of eastern Indo- Map 107.5: Participial passives in Europe

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1497

No passives exist in Nakh-Daghestanian and be considered further here.) In Haspelmath in Hungarian, and passives of different for- (1993), I examined 31 verb pairs in 21 lan- mal types are found in Turkic, Georgian, and guages and found that languages differ Armenian (stem suffix), in Basque, and in greatly in the way inchoative-causative pairs Celtic (cf. the Welsh ‘get’-passive: ‘Terry got are expressed: Some languages are anticau- his hitting by a snowball’ for ‘Terry got hit sative-prominent, preferring anticausatives to by a snowball’). Finnish and Irish have pas- causatives, while others are causative-promi- sives of a different syntactic type: In this con- nent. It turns out that anticausative-promi- struction, only the subject is backgrounded, nence is a characteristic feature of SAE. In while the direct object remains in its place. my sample, German, French, Romanian, Participial passives are very rare in lan- Russian, Modern Greek and Lithuanian guages other than Standard Average Euro- show the highest percentages of anticausative pean. In Haspelmath (1990) I surveyed a verb pairs (between 100% and 74% of all world-wide sample of eighty languages and pairs that do not belong to the third, non- found that a passive exists only in the mi- directed, type). The percentage in the Euro- nority of the languages (thirty-one). Of these pean languages of my sample are shown in thirty-one languages, only four have a pas- Map 107.6. sive formed from a participle plus an intran- sitive auxiliary, and two of them are Euro- pean languages ( and Danish). The most common formal type of passive is the stem suffix (found in twenty-five languages). Syntactically, the possibility of an adverbial agent phrase is also by no means universal, but it is characteristic of SAE languages (La- zard 1990: 246). It must be admitted that the SAE status of this feature is less evident than that of the first two features because the eastern Indo- European languages also tend to have pas- sives of this type. In fact, in my 1990 study, the two non-European languages with parti------70Ϫ100 % anticausatives ciple-auxiliary passives were Baluchi (an Ira- ---- 50Ϫ70 % anticausatives nian language) and Maithili (an Indic lan- Map 107.6: Percentage of anticausative pairs guage). Thus, one might say that this feature is an Indo-European genealogical feature. However, at least the Celtic languages and By contrast, Asian languages show much Armenian, two non-SAE branches of Indo- lower percentages of anticausatives, prefer- European, do not have such passives, and ring causatives instead (e. g. Indonesian: 0%, Maltese is a non-Indo-European language Mongolian: 11%, Turkish: 34%, Hindi/Urdu with such a passive (calqued from Italian). 35%, Lezgian: 40%). An intermediate posi- tion is occupied by the Finno-Ugrian lan- 2.6. Anticausative prominence guages of eastern Europe (Finnish 47%, There are three ways in which languages can Udmurt 46%, Hungarian 44%) as well as express inchoative-causative alternations such Georgian (68%) and Armenian (65%). In a as ‘get lost/lose’, ‘break (intr.)/break (tr.)’, study involving more languages from , ‘rise/raise’. One is by means of a causative Africa and Europe but less language-partic- derivation (J Art. 66), i. e. a derived verb ular detail, Masica (1976) found a clear dis- based on the inchoative member of the al- tinctive pattern for Europe: few causatives, ternation, e. g. Mongolian xajl-uul- ‘melt (tr.)’, heavy reliance on anticausatives (see espe- from xajl- ‘melt (intr.)’. The second is by cially his Maps 2 and 3). In a recent world- means of an anticausative derivation, i. e. a wide study of 18 verbs from 80 languages, derived verb based on the causative member, Nichols et al. (to appear) report that in in- e. g. Russian izmenit’-sja ‘change (intr.)’, from choative-causative pairs involving inanimate izmenit’ ‘change (tr.)’. (The third type, in participants (i. e. the most typical subtype), which neither member is derived from the the causative is generally favored worldwide other, i. e. non-directed alternations, will not and is strongly disfavored only in Europe.

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1498 XIV. Typological characterization of language families and linguistic areas

Anticausative-prominence is not an Indo- hyena ate the hare’s fish’). This type is not Europeanism: Older Indo-European had a found in Europe at all. Conversely, dative ex- productive causative formation, which lost ternal possessors seem to be very rare outside its productivity in the European branches, Europe (the only case I am aware of is Ewe, but continued to be productive in eastern cf. Ameka 1996), so this is a very robust ex- Indo-European (cf. the low figure of 35% an- ample of an SAE feature. ticausatives in Hindi/Urdu). 2.8. Negative pronouns and lack of 2.7. Dative external possessors verbal negation In König & Haspelmath (1998) and Haspel- The areal distribution of negation in Europe math (1999), we studied the distribution of has been studied in detail by Bernini & external possessors in thirty European lan- Ramat (1996) (see also Ramat & Bernini guages (J Art. 73). We found three main lan- 1990). Here I will single out just one aspect guage types in Europe: (i) those with dative of negation, the cooccurrence of verbal nega- external possessors, e. g. German Die Mutter tion with negative indefinite pronouns. I dis- wäscht dem Kind die Haare ‘The mother is tinguish two main types: (i) V ϩ NI (verb ϩ washing the child’s hair’, (ii) those with loca- negative indefinite), e. g. German Niemand tive external possessors, e. g. Swedish Na˚gon kommt ‘nobody comes’, and (ii) NV ϩ NI bröt armen pa˚ honom ‘Someone broke his (negated verb ϩ negative indefinite), e. g. arm (lit. on him)’, and (iii) those that lack Modern Greek Kane´nas dhen e´rxete ‘nobody external possessors and must express posses- (lit. not) comes’. A third, mixed type might sors NP-internally, e. g. English. The SAE be distinguished in which verbal negation feature, external possessors in the dative, is cooccurs with negative indefinites only when found in Romance, Continental West Ger- the indefinite follows the verb but not when manic, Balto-Slavic, Hungarian and Balkan it precedes it, e. g. Italian Nessuno viene ‘no- languages (Greek, Albanian). North Ger- body comes’, but Non ho visto nessuno ‘Not I manic and Balto- have loca- have seen nobody’. For our purposes we can tive external possessors, i. e. they are some- classify this type as a subtype of (i), V ϩ NI. what peripheral SAE languages with respect The Standard Average European type is to this feature. The geographical distribution V ϩ NI (cf. Bernini & Ramat 1996: 184, Has- is shown in Map 107.7. pelmath 1997: 202). It is found in French (if we disregard the particle ne), Occitan and all Germanic languages, as well as (in the mixed variety) in Ibero- and Italo-Romance and Al- banian (but not in Romanian or other Bal- kan languages). The geographical distribu- tion of the types is shown on Map 107.8.

Map 107.7: Dative external possessor

In the far west (Welsh, Breton, English) and in the southeast (Turkish, Lezgian) of Europe there are languages which do not have exter- nal possessors at all. The eastern Indo-Euro- Map 107.8: Languages lacking verbal negation pean languages Kurdish, Persian and Hindi/ with a negative indefinite Urdu also belong to this type. Outside Europe a fourth type enjoys considerable popularity: All the eastern European languages (Balto- the “relation-usurping” type, where he pos- Slavic, Finno-Ugrian, Turkic, Nakh-Daghes- sessor “usurps” the syntactic relation of the tanian) with the exception of Georgian, and possessum (e. g. Chichewa, a Bantu language, the Celtic languages in the west show the has ‘The hyena ate the hare the fish’ for ‘The NV ϩ NI type. This type is also that of the

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1499 eastern Indo-European languages (Iranian The locative comparatives are all at the west- and Indic), as well as that of the clear major- ern fringe (Breton) or the eastern fringe of ity of the world’s languages: Kahrel (1996) Europe (Finnish, Russian, Nenets, Ubykh, has studied negation in a representative world- Turkish, Laz). The other two types do not wide sample of 40 languages and found only exist at all in Europe Ϫ the exceed compara- five languages with V ϩ NI negative pat- tive is found particularly in Africa, and the terns, one of which is the SAE language conjoined comparative occurs only in the Dutch (the other four are Mangarayi (Aus- and . tralia), Evenki, Chukchi (), and Nama (southern Africa)), as against 41 NV ϩ NI 2.10. Relative-based equative constructions patterns, and seven others. I found a very sim- Comparison of equality (equative construc- ilar pattern in my (non-representative) sample tions) is discussed less often than comparison of 40 languages (Haspelmath 1997: 202). of inequality, and nobody has undertaken a study of equatives on a world-wide scale. 2.9. Particles in comparative constructions Still, there are good to think that Comparative constructions were investigated equative constructions provide evidence for by Stassen (1985) in a world-wide study of 19 Standard Average European (Haspelmath & languages (J Art. 75). Stassen distinguishes Buchholz 1998). In Europe, many languages six main ways in which the standard of com- have an equative construction that is based parison may be expressed: Three kinds of loc- on an adverbial relative-clause construction. ative comparatives (‘bigger from X’, ‘bigger For example, Catalan has tan Z com X ‘as Z to X’, ‘bigger at X’), the exceed comparative as X’ (where Z is the adjective and X is the (‘Y is big exceeding X’), the conjoined com- standard). Catalan com is an adverbial rela- parative (‘Y is big, X is little’), and the par- tive pronoun, and tan is a correlative demon- ticle comparative (‘bigger than X’). The par- strative. A very similar construction is found ticle in this latter type is often related to a elsewhere in Romance (Portuguese ta˜oZ relative pronoun (cf. English than/that, Latin como X, Occitan tan Z coma X), in Germanic quam/qui), and the case marking of the stan- (German so Z wie X), in Slavic (Czech tak Z dard is not influenced by the particle (so that jako X, Russian tak(oj) zˇe Z kak X), in Ro- it is possible to distinguish ‘I love you more mani (kade Z sar X), in Hungarian (olyan Z than she’ from ‘I love you more than her’). mint X), in Finnish (niin Z kuin X), and in As Heine (1994) notes, the six types are not Georgian (isetive Z rogorc X). In the English evenly distributed among the languages of construction, the relative-clause origin of as the world. Of the 18 particle comparatives is not fully transparent synchronically, but in Stassen’s sample, 13 are in Europe, and of diachronically as derives from a demonstra- the 17 European languages in the sample, 13 tive (eall swa Ͼ all so) that was also used have a particle comparative. The distribution as a relative pronoun. In some Balkan lan- within Europe again conforms to our expec- guages, the correlative demonstrative is not tations: Particle comparatives are found in used (e. g. Bulgarian xubava kato tebe ‘as Germanic, Romance, Balto-Slavic, the Bal- pretty as you’), but the standard marker is kans, Hungarian, Finnish and Basque, so this clearly of relative-pronoun origin. (There is is the SAE type. The distribution is shown in probably some connection between the rela- Map 107.9. tive-pronoun origin of equative markers and the relative-pronoun origin of comparative standard markers that we saw in § 2.9.). Non-SAE languages have quite different equative constructions. Many SOV languages in eastern Europe have a special equative standard marker (Lezgian xˆiz, Kalmyk sˇing; also Basque bezain and Maltese daqs), and the Celtic languages have a special (non- demonstrative) marker on the adjective (e. g. Irish chomh Z le X ‘equative Z with X’). In the Scandinavian languages, the word ‘equ------particle comparative ally’ is used on the adjective (e. g. Swedish - - - - locative comparative lika Z som X ‘equally Z as X’). The distri- Map 107.9: Comparative types in Europe bution of the relative-based equative con-

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1500 XIV. Typological characterization of language families and linguistic areas struction in Europe is shown in Map 107.10, in the world’s languages, and strict subject following Haspelmath & Buchholz (1998: agreement is characteristic of a few European 297). languages, some of which happen to be well- known. In her world-wide sample of 272 languages, Siewierska (1999) finds only two strict-agreement languages, Dutch (an SAE language) and Vanimo (a Papuan language of New Guinea). Siewierska further notes that outside of Europe, she is aware of only two additional strict-agreement languages that are not in her sample (Anejom and Labu, two Oceanic languages). Gilligan (1987) reached a similar conclusion on the basis of a sample of 100 languages. The distribution of strict subject agreement markers in some European Map 107.10: Relative-based equative-constructions languages is shown in Map 107.11. Impressionalistically, relative-based equatives seem to be rare in the world’s languages, and the eastern Indo-European languages do not seem to use them in general (however, a counterexample is Punjabi). 2.11. Subject person as strict agreement markers The majority of the world’s languages have bound person markers on the verb that cross- refer to the verb’s subject (or agent). When these subject affixes cooccur with overt sub------languages with strict subject agreement ject NPs (full NPs or independent subject - - - - languages with obligatory subject pronouns, pronouns), they are called agreement mark- lacking verb agreement ers. However, in most languages they can oc- Map 107.11: Obligatory subject pronouns cur on their own and need not cooccur with overt subject NPs. For example, in the Bul- The map shows two non-contiguous areas in garian phrase vie rabotite ‘you (pl.) work’, we see the subject suffix -ite (2nd person plural) which subject agreement suffixes cannot have cooccurring with the independent subject a referential function: Germanic and Gallo- pronoun vie ‘you (pl.)’, showing that -ite is Romance languages with Welsh on the one an agreement marker. But in Bulgarian it is hand, and Russian on the other. Perhaps only equally possible and probably more common the western European area should be thought to say just rabotite ‘you (pl.) work’, i. e. the of as being relevant for SAE; in Russian, subject suffix can have a referential function past-tense verbs do not have subject person on its own. In German, by contrast, this is affixes, so Russian is not a very good exam- not possible: ‘you work’ is ihr arbeit-et. Since ple of a strict-agreement language. In the the agreement suffix -et does not have such eastern Nordic languages (Norwegian, Swed- an independent referential function, the sub- ish, Danish), the subject pronouns are obliga- ject pronoun ihr cannot be omitted. Lan- tory as they are in English, German or Ice- guages like German are often called “non- landic, but the languages have lost agreement pro-drop languages”, and languages like distinctions on the verb entirely (cf. Swedish Bulgarian are called “pro-drop languages”; jag biter/du biter/han biter ‘I/you/he bite(s)’, better terms would be “strict-agreement lan- Icelandic e´gbı´t/pu´ bı´tur/hann bı´tur). These guages” vs. “referential-agreement languages”. languages are thus “non-pro-drop” in a It has sometimes been thought that strict sense, but they are not strict-agreement lan- agreement, as exhibited by German, English, guages. English is approaching this type, as and French, is the norm and that referential the only remnant of subject agreement is the agreement is somehow special. But in fact, 3rd person singular present-tense suffix -s. referential agreement is far more widespread (There are also some languages of this type

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1501 in the eastern Caucasus, and indeed in many striking, but which nevertheless seem good other parts of the world, but they may never candidates for Europeanisms. No maps will have had subject person agreement marking.) be given for these features, and the evidence will be summarized only briefly. 2.12. Intensifier-reflexive differentiation Intensifiers are words like English self, Ger- 3.1. Verb fronting in polar interrogatives man selbst, French meˆme and Russian sam In the large majority of languages, polar in- that characterize a noun phrase referent as terrogatives are marked by interrogative in- central as opposed to an implicit or explicit tonation or an interrogative particle or both periphery (e. g. The Pope himself gave us an (J Art. 77). In his sample of 79 languages, audience, i. e. not just the cardinals (J Ultan (1978) found only seven languages Art. 57; König & Siemund 1999). In many showing the alternative strategy of verb languages, the intensifier expression is also fronting (often called “subject-verb inver- used as a reflexive pronoun, for instance in sion”). Of these, six are European (English, Persian (xod-asˇ ‘himself’: Husˇang xod-asˇ French, Romanian, Russian, Hungarian, ‘Hushang himself’, and Husˇang xodasˇ-ra¯ did Finnish; the seventh language is Malay), so [Hushang self-acc saw] ‘Hushang saw him- that the SAE status of verb fronting seems self’). However, a feature that is typical of beyond doubt. In fact, the large majority of SAE languages is the differentiation of reflex- Germanic, Romance and ive pronouns and intensifiers (König & Has- (plus Modern Greek) appear to have verb pelmath 1999). For instance, German has sich fronting in polar questions in one form or (reflexive) vs. selbst (intensifier), Russian has another. The three European languages for sebja vs. sam, Italian has si vs. stesso, Greek which Ultan explicitly reports that no verb has eafto´ vs. ´ıdhjos. Map 107.12 shows the lan- fronting occurs are peripheral: Basque, Gae- guages in Europe with special reflexive pro- lic and Lithuanian. Furthermore, SAE lan- nouns that are not identifical to intensifiers. guages are characterized by the absence of an interrogative particle. In Ultan’s data, the nine European languages exhibiting a par- ticle in polar questions are all peripheral to a greater or lesser extent: Basque, Irish, Scot- tish Gaelic, Albanian, Hungarian, Lithua- nian, Russian, Finnish, Turkish (and I can add Nakh-Daghestanian). Verb fronting in polar questions was suggested as a Euro- peanism already by Beckman (1934) (cf. Dahl 1990). 3.2. Comparative marking of adjectives

Map 107.12: Intensifier-reflexive differentiation Most European languages have special forms for adjectives occurring in comparative con- structions. For instance, English uses the Intensifier-reflexive differentiation is not an suffix -er in this way (The dog is bigg-er than Indo-Europeanism, because eastern Indo- the cat). Such an inflectional marker of adjec- European languages have the same expres- tives is not common in the world’s languages sion for intensifiers and reflexives (e. g. Per- outside of Europe. Some languages use some sian xod-asˇ, Hindi aap). There are no pub- kind of adverbial particle modifying the ad- lished world-wide studies yet, but it seems jective (‘more’), but perhaps the most com- that non-differentiation is very common mon type is represented by Japanese, where around the world, and while differentiation is the comparative semantics is carried by the also found elsewhere, it is not found in areas standard marker alone (e. g. inu-ga neko yori immediately adjacent to European languages. ookii [dog-subj cat from big] ‘the dog is big- ger than the cat’). 3. Some further likely SAE features Special comparative forms are found in all Germanic, Balto-Slavic and Balkan lan- In this section, I will mention a few features guages (with the exception of Romanian and which are less well-documented than those in Albanian), and most Romance languages § 2, or whose geographical distribution is less preserve at least four suppletive forms (e. g.

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1502 XIV. Typological characterization of language families and linguistic areas

Italian maggiore ‘bigger’, minore ‘smaller’, However, within the and-languages there are peggiore ‘worse’, migliore ‘better’). Compara- several sub-types according to the position of tive forms also exist in Basque (e. g. haundi- the particle, which we may call “A and-B”, ago ‘bigg-er’), Hungarian (nagy-obb ‘bigg-er’), “A-and B”, “A-and B-and”, and “A B-and” Finnish (iso-mpi ‘bigg-er’), and other Finno- (of the remaining logical possibilities, “and-A Ugrian languages. B” seems to be inexistent, and “and-A and- Comparative forms are not completely un- B” occurs only as a secondary pattern). Most known outside of Europe. Arabic has a spe- European languages, and in particular all cial comparative form (e. g. ?akbar ‘bigger’, SAE languages, belong to the sub-type “A from kabiir ‘big’), but it is unique among and-B”. The types “A-and B-and” and “A- Afro-Asiatic languages in this respect. Old and B” are found in some languages of the Indo- had comparative Caucasus and in some , as forms, and the modern Iranian languages well as scattered throughout northern Eu- have preserved them to some extent (e. g. Per- rasia and South Asia (e. g. in Abkhaz, Archi, sian -tær, Zaza -eˆr). But further east, in mod- Persian, Sinhalese, Tamil, Burmese, Korean ern Indic, the comparative does not exist according to Stassen; Stassen also points out anymore, and languages like Hindi-Urdu and that there is a correlation with verb-final Bengali use a construction analogous to the word order here). Furthermore, some periph- Japanese example just cited. Similarly, in the eral European languages make restricted use Uralic languages, the further east we go, the of the with-strategy (e. g. Russian my s toboj fewer comparatives we find. For instance, ‘I and you’, lit. ‘we with you’, and also Old Khanty (a Finno-Ugrian language spoken in Irish, Lithuanian, Polish and Hungarian, western Siberia, i. e. outside of Europe) does according to Stassen). Taken together, these have a comparative form in -sek (e. g. jam-sek data do show that belonging to the “A and- ‘better’), which is used when no standard is B” type is not a trivial feature of the SAE present. But in a complete comparative con- linguistic area. struction, no marking is found on the adjec- tive (e. g. nan ke:se:-n e:welt jam [you knife- 3.4. Comitative-instrumental syncretism 2sg from good] ‘better than your knife’, Ni- In all SAE languages, the preposition that kolaeva 1999: 21). expresses accompaniment (ϭ comitative) also Thus, although this feature is not confined serves to express the instrument role (e. g. to Europe, it is typical of a SAE feature in English with: with her husband/with the ham- that it is robustly present in western Indo- mer). Such languages are said to exhibit com- European and Uralic languages, but gets itative-instrumental syncretism. Stolz (1996) rarer the further east we go in these families. studied comitative and instrumental markers in a world-wide sample of 323 languages and 3.3. “A and-B” conjunction found that this kind of syncretism is typical The feature discussed in this section is less of Europe. Non-European languages more distinctive than the others mentioned so far, commonly possess separate markers for these but I hope to show that it is not at all devoid two semantic roles (e. g. Swahili na ‘with of interest. Stassen (2000) offers the first (comitative)’, kwa ‘with (instrumental)’. As world-wide typological study of NP conjunc- Table 107.1 shows, about two thirds of Stolz’s tion strategies, based on a sample of 260 sample languages are non-syncretic, and only languages (J Art. 82). He distinguishes two one quarter is syncretic. (The remaining lan- basic types, and-languages (using a symmet- guages belong to a mixed type, which I ig- ric particle) and with-languages (using an nore here for the sake of simplicity; thus, the asymmetric comitative marker). Two thirds percentages do not add up to 100%.) of Stassen’s sample languages are and-lan- Two areas diverge significantly from the guages, and since SAE clearly belongs to this general trend: Oceania has far less syncretism type, too, it is not a very distinctive property. than the world average, and Europe has far And-languages cover all of northern Eurasia, more syncretism than the world average. South Asia, the Middle East and northern When we look at the pattern within Europe, Africa, , New Guinea, and parts of it becomes even clearer that we are dealing Central and . With-languages with an SAE feature (as Stolz recognizes, cf. are encountered in sub-Saharan Africa, East 1996: 120). Of the 16 non-syncretic languages and Southeast Asia, the islands of Oceania, in Europe, 10 are Caucasian languages, i. e. and large areas of North and South America. they are clearly outside of SAE, and one is

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1503

Table 107.1: Comitative-instrumental: Syncretic and non-syncretic languages

syncretic (e. g. English) non-syncretic (e. g. Swahili) languages percentage languages percentage

Europe 25 49% 16 31% Africa 20 31% 38 58% Americas 16 21% 54 69% Asia 12 18% 47 71% Oceania 6 10% 54 86% World 79 24% 209 65% only politically, not anthropologically, in This is clearly a very marginal feature in Europe (Greenlandic). Four of the remaining , but it is intriguing that it should five languages are also otherwise not typical show such a clear geographical distribution. instances of SAE (Basque, Finnish, Maltese, Mari). And when we look at the 38 Indo- 3.6. Some other characteristics of SAE European languages in Stolz’s sample, we The features examined so far present the see that syncretism cannot be regarded as an most striking evidence for Standard Average Indo-Europeanism: Of the eight Indo-Euro- European, but there are probably many more pean languages not spoken in Europe, only features that will turn out to be characteristic three show syncretism, while five show non- of the core European languages in one way syncretism. Thus, in Asia Indo-European lan- or another. In this subsection, several such guages behave like Asian languages, and there candidates will be mentioned briefly. The first is no general pattern for Indo-European. few features in the following list are purely negative: At first glance, this may seem odd, 3.5. Suppletive second ordinal but of course the lack of a category that is Most languages have a suppletive form of the widespread elsewhere is no less significant ordinal numeral ‘first’, i. e. a form not de- than the presence of a category that is rare rived from the cardinal numeral ‘one’. An elsewhere. example is German, where ‘1st’ is erster (un- (i) Lack of an alienable/inalienable opposi- related to eins ‘1’), contrasting with other tion in adnominal possession (J Art. 72). In ordinals such as zweiter ‘2nd’ (cf. zwei ‘2’), Nichols’s (1992) world-wide sample, almost vierter ‘4th’ (cf. vier ‘4’), and so on. In Stolz’s half of the languages show such an opposi- (2001b) study of 100 languages world-wide, tion, but no European language does (1992: there are 95 languages with special ordinal 123). More generally, this opposition is rarer numerals, and of these, 78 have a suppletive in the and common in the New word for ‘first’. Thus, languages that say World, but in Europe it is even less common (literally) ‘oneth’ for ‘1st’ are not common. than in Africa and Asia. However, the same sample has only 22 lan- (ii) Lack of an inclusive/exclusive opposition guages in which the word for ‘2nd’, too, is in first person non-singular pronouns. Again, suppletive and not derived from ‘2’ (e. g. this opposition is commonest in the New English second). Thus, most languages have World and in the Pacific , but in (literally) ‘twoth’ for ‘2nd’. The 22 languages Europe it is even rarer than in Africa and that have a suppletive ‘2nd’ word are heavily Asia, as was shown by Nichols (1992: 123). concentrated in Europe: 17 are European (iii) Lack of reduplicating constructions. I languages, and this type is clearly the major- have no systematic evidence to back up the ity within Europe (which is represented by 27 claim that this is a characteristic feature of languages in Stolz’s sample). Of the 10 Euro- European languages, but reduplication is so pean languages that do not have a suppletive common across languages that its almost to- second ordinal, six are clearly outside SAE tal absence in the core European languages (Basque, Turkish, Armenian, Georgian, Lez- becomes striking. (Interestingly, reduplication gian, Greenlandic). Among SAE languages, existed in older Indo-European languages at only some Balkan languages (Romanian, Al- least in one construction, the perfect, but banian, Romani) and German lack a supple- even here it was lost entirely by the Middle tive second ordinal. Ages.)

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1504 XIV. Typological characterization of language families and linguistic areas

(iv) Discourse pragmatic notions such as detail I have to refer the reader to van der topic and focus are expressed primarily by Auwera’s thorough study. sentence stress and word order differences (ix) “Preterite decay”: the loss of the old (Lazard 1998: 116). Only the Celtic languages preterite and its replacement by the former and French give a very prominent role to present perfect. This is a change that oc- clefting, and particles marking discourse curred in the last millenium in French, Ger- pragmatic notions are virtually unknown. man and northern Italian, as well as in some (v) SVO basic word order at the level of the other adjacent European languages (cf. Thie- clause. This feature is of course found else- roff 2000: 285). Its distribution is far nar- where in the world, but in Europe it corre- rower than that of the other Europeanisms, lates particularly well with the other SAE but it is the only feature of those studied by features. The Celtic languages in the west Thieroff whose geography comes close to have VSO order (except for Breton, which is Standard Average European (cf. also Abra- also otherwise more SAE than Irish and ham 1999). Welsh), and the eastern languages have SOV Quite a few additional features have been word order. Interestingly, Balto-Finnic (Fin- mentioned in the earlier literature as charac- nish, Estonian, etc.) and (less unequivocally) teristic of SAE, but earlier authors have Hungarian have SVO word order, whereas sometimes neglected to make sure that a pro- the eastern Uralic languages have SOV. Simi- posed Europeanism is not also common else- larly, the eastern Indo-European languages where in the world. Most of Whorf’s original tend to show SOV word order. (See Dryer examples of SAE features seem to be of this 1998 for more on word order in the lan- kind. For instance, he notes that in contrast guages of Europe.) to SAE, Hopi lacks “imaginary plurals” (such (vi) European languages tend to have just as ‘ten days’, according to Whorf a “meta- one converb (J Art. 83) (cf. Nedjalkov 1998). phorical aggregate”). But of course, we have For instance, Romance languages have the no evidence that such plurals of time-span gerundio/ge´rondif, English has the -ing-form, nouns are in any way characteristic of Euro- and Slavic and Balkan languages have their pean languages. It may well be that they are adverbial participle. The Celtic languages in common throughout the world. (To give the west completely lack such a form, and the Whorf his due, it must be added that he was languages east of SAE tend to have more not interested in demonstrating that SAE than one converb. Otherwise the core Euro- languages form a Sprachbund. He just used pean languages tend to have adverbial con- this term as a convenient abbreviation for junctions (J Art. 63) to make adverbial “English and other European languages clauses. According to Kortmann (1997: 344), likely to be known to the reader”, without they have “a large, semantically highly dif- necessarily implying that these languages are ferentiated inventory of free adverbial sub- an exclusive club.) ordinators placed in clause-initial position”. More generally, they tend to have finite rather than non-finite subordinate strategies (J 4. Degrees of membership in SAE Art. 100), though a multi-purpose usually exists (except for the Balkan lan- Membership in a Sprachbund is typically a guages). matter of degree. Usually there is a core of (vii) European languages usually have a spe- languages that clearly belong to the Sprach- cial construction for negative coordination, bund, and a periphery of surrounding lan- e. g. English neither A nor B, Italian ne´ Ane´ guages that share features of the linguistic B, Russian ni A ni B, Dutch noch A noch B, area to a greater or lesser extent. Hungarian sem A sem B. Again, no world- In order to quantify the degrees of mem- wide study has been published, but such a bership in SAE, a simple procedure suggests negative coordinating construction is rarely itself that was first applied to areal typology reported from languages outside Europe (cf. by van der Auwera (1998a). In addition to Haspelmath to appear). individual maps in which the lines denote iso- (viii) SAE languages have a large number of glosses (as in Maps 107.1Ϫ12), we can com- characteristic properties in the area of phasal bine different features in a single map and adverbials (expressions like already, still, no show the number of shared by the longer, not yet) (van der Auwera 1998b). language. Map 107.13 shows such a “cluster These are rather well documented, but for the map” in which the lines stand for “quantified

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1505 isoglosses” (or “isopleths”). The map com- historical role played by speakers of these bines nine features of § 2.: definite and indefi- two languages both in the early medieval his- nite articles, relative clauses with relative pro- tory of and in the very nouns, ‘have’-perfect, participial passive, da- recent attempt at European unification, this tive external possessors, negative pronouns is of course an extremely intriguing result. and lack of verbal negation, relative-based (b) The southern European languages (both equative constructions, subject person affixes Romance and Balkan languages) are at least as strict agreement markers, and intensifier- as close, if not closer to the nucleus than the reflexive differentiation. The languages in the northern languages and English. This means nucleus (French and German) show the SAE that it is misleading to call SAE features value in all nine of these features. The lan- “Western European features”, as is some- guages in the next layer (Dutch, other Ro- times done. It is true that the Slavic lan- mance, Albanian) show eight features, the guages in the east lack many SAE features, next layer (English, Greek, Romanian) shows but the Balkan languages are generally more seven features, and so on. In this map, the SAE than Slavic, although they are not west- resulting picture is actually very clear, be- ern European. cause the SAE area with at least five SAE (c) England stands somewhat apart from the features stands out from the remaining lan- European nucleus (as noted also by van der guages, which have at most two SAE fea- Auwera 1998a: 823), although it is closely tures. related genealogically to German and has been thoroughly influenced by French. Since English is currently the dominant language throughout the world, it is worth pointing out its somewhat marginal status among its European sister languages. It is important to keep in mind that the fea- tures on which Map 107.13 is based have not been selected randomly and are thus by no means representative of the morphosyntactic features of European languages. They were included precisely because they were known to show a distribution that supports the SAE Map 107.13: A cluster map combining nine fea- hypothesis. Thus, no claim is made that all tures (or even the majority of) features will show a similar distribution. It is perfectly possible Such cluster maps are thus a fairly direct rep- that we will some day discover another resentation of degrees of membership in a lin- Sprachbund, based on a different set of fea- guistic area. But of course, the cluster map tures, that has Russian at its core and extends directly reflects the choice of features that are all the way to western Siberia in the east and combined, and this choice is always some- in the south, but within Europe what arbitrary. Of the twelve features in § 2, comprises only the Slavic, Balkan, and Scan- only nine were selected here because informa- dinavian languages. This area would overlap tion on the other three was incomplete. Ide- with SAE, but it would not contradict it. ally, the features of § 3 should have been Thus, a language may in principle belong to added, too. But it seems to me that the main different linguistic areas, and different lin- results of Map 107.13 would not be changed guistic areas may coexist “on top of” each (this map can also be compared to the very other. Since areal typology is only in its in- similar map in van der Auwera (1998a: 823), fancy, we do not know how common such which combines five adverbial features or situations are, but nothing in the logic of a feature clusters). The most striking features Sprachbund implies that the world should be of Map 107.13 are: exhaustively divisible into non-overlapping (a) The nucleus of Standard Average Euro- Sprachbünde. pean is formed by French and German (a In fact, a number of smaller linguistic finding that led van der Auwera (1998a: 824) areas within Europe have been proposed in to propose the term Charlemagne Sprachbund the literature (apart from the Balkan area, for the nuclear area of SAE). In view of the whose importance is not doubted by anyone),

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1506 XIV. Typological characterization of language families and linguistic areas e. g. by Lewy (1942), Wagner (1959), De´csy The first possibility must be rejected be- (1973), Haarmann (1976), and Ureland (1985) cause the great majority of Europeanisms (cf. also Wintschalek 1993 on a Volga-Kama are innovations with respect to Proto-Indo- area). Currently the most thoroughly studied European. For instance, as far as we know, areas are the Circum-Baltic area (cf. Stolz Proto-Indo-European did not have articles, a 1991, Dahl & Koptjevskaja-Tamm (eds.) ‘have’-perfect, “A and-B” conjunction, strict 2001) and the Mediterranean area (cf. Cris- subject agreement, particle comparatives, or tofaro & Putzu (eds.) 2000). However, no relative clauses with relative pronouns (cf. strong claims about a Circum-Baltic or a Me- Lehmann 1974, Haspelmath 1998). With re- diterranean linguistic area seem to have been spect to Proto-Indo-European, and also with made as a result of these studies. respect to the oldest Indo-European lan- guages attested in Europe (, Old Latin, Gothic), Standard Average Euro- 5. How did SAE come into being? pean is clearly an innovation. Linguistic areas arise through language con- The second possibility, a pre-Indo-Euro- tact, but precisely which contact situation pean substratum in Europe causing the SAE gave rise to Standard Average European is features, would be extremely difficult to de- not immediately clear. And what is the source monstrate, but it might be worth pursuing. It of the various Europeanisms: Who borrowed is intriguing to note that the geographical from whom? A full discussion of the socio- space occupied by SAE languages coincides historical, cultural and sociolinguistic issues fairly precisely with the area of the Old Euro- is beyond the scope of this article, so I will pean hydronymy, i. e. the homogeneous layer restrict myself here to mentioning just five of river names discovered by Hans Krahe possibilities: (see Vennemann 1994 for recent discussion). Vennemann (1994) proposes that these Old (i) retention of Proto-Indo-European struc- European hydronyms were not coined by an tures and assimilation of some non- early prehistoric Indo-European population, Indo-European languages to Indo-Euro- but by a pre-Indo-European people which he pean language structure; calls Vasconic (the only surviving Vasconic (ii) influence from a common substratum language being Basque). Furthermore, the of a pre-Indo-European population in Old European hydronymy is hardly attested Europe; in the British Isles, where the Celtic lan- (iii) contacts during the great trans- guages are spoken, i. e. they could not have formations at the transition from late been influenced by the Vasconic substratum. antiquity to the early in This is in perfect harmony with the well-mo- Europe; tivated hypothesis that the Celtic languages (iv) the (Latin) and the acquired some of their striking features from common European culture of the Mid- a different substratum related to the Afro- dle Ages; Asiatic languages (Pokorny 1927Ϫ30, Gens- (v) the common European culture of mod- ler 1993). ern times, from the to the The main argument against the substratum Enlightenment. view is that the SAE features seem to be gain- ing ground too late for a pre-Indo-European The fifth possibility must be rejected because substratum to have caused them. Some SAE Ϫ a time depth of 300 500 years is not suffi- features appear only in the first millenium cient to account for grammatical common- CE, but also the earlier features usually come alities of the kind discussed above. If lexical fairly late, so that the earliest records of Indo- similarities between the European languages European-languages in Europe still show are discussed Ϫ for instance neoclassical traces of the Proto-Indo-European patterns compounding (socio-/paleo-/ortho-/demo-, (e. g. causatives, relative clauses, locative com- -graphy/-logy/-cracy, etc.) or idiomatic struc- parative, “A B-and” conjunction). If these ture (e. g. ivory tower/torre d’avorio/Elfenbein- SAE features were caused by a substratum, turm, as poor as a church mouse/pauvre comme then we should have much more evidence of un rat d’e´glise/arm wie eine Kirchenmaus) Ϫ the population speaking this substratum lan- then the last several centuries are the appro- guage. Moreover, a Vasconic substratum can priate time frame for explaining the historical hardly account for the SAE features because links, but the basic syntactic structures com- modern Basque is in most relevant ways very mon to SAE languages must be older. much unlike the SAE languages.

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1507

Of the remaining two possibilities, we can 6. Abbreviations of language names probably exclude option (iv) (the influence of Latin in the Middle Ages), because most SAE Alb Albanian features were absent in Latin and developed Arm Armenian only in the Romance languages. There are Blg Bulgarian only two features for which Latin influence Brt Breton is a likely factor: negation and relative pro- Bsq Basque nouns. In the case of these two features, the Cz Czech standard languages sometimes show devia- Dut Dutch tions from the , so at least Eng English the written standard languages may have Est Estonian been influenced by Latin, the European writ- Fin Finnish ten language par excellence for many centu- Fr French ries. Thus, non-standard English has con- Gae Scots Gaelic structions like I won’t do nothing (‘I won’t do Grg Georgian anything’), and similarly in non-standard Grk Greek German and French (cf. Haspelmath 1997: Grm German 205). Analogously, Latin-type relative pro- Hng Hungarian nouns occur widely in the standard languages Ice Icelandic of Europe, but vernacular speech often pre- Ir Irish fers relative particles (Lehmann 1984: 88, It Italian 109). However, Latin probably only helped Kom Komi to reinforce these structures in those lan- Lat Latin guages where they existed already indepen- Laz Laz dently as variants. Lit Lithunian Thus, we are left with option (iii), the time Ltv Latvian of the great migrations at the transition be- Lzg Lezgian tween antiquity and the Middle Ages. This Mar Mari seems to be the appropriate time frame at Mlt Maltese least for articles, the ‘have’-perfect, the par- Mrd Mordvin ticipial passive, anticausatives, negative in- Nnts Nenets definites, nominative experiencers and verb Nor Norwegian fronting. The rise of these constructions can Pol Polish be observed only with difficulty because they Prt Portuguese were by and large absent in the written classi- Rom Romanian cal languages but seem to be well in place Rus Russian once the vernacular languages appear in the SAE Standard Average European written record toward the end of the first Sam Saami millennium CE (cf. also Fehling 1980). This SCr Serbian/Croatian hypothesis derives some further plausibility Sln Slovene from the fact that must Spn Spanish have been particularly intensive and effective Srd Sardinian during the great migrations, and in the case Swd Swedish of French and northern Italian we have am- Tat Tatar ple records of the lexical effects of these con- Trk Turkish tacts. However, it is not so easy to fit features Uby Ubykh such as particle comparatives, „A and-B” Udm Udmurt conjunction and relative pronouns into this Ukr Ukrainian picture, because these features seem to have Wel Welsh developed around the middle of the first mil- lenium BC or even earlier (cf. Haspelmath 1998). Of course, we must always reckon 7. References with the possibility (or even likelihood) that Abraham, Werner. 1999. “Preterite decay as a different SAE features are due to different European areal phenomenon”. Folia Linguistica historical circumstances, and the correct pic- 33.1: 11Ϫ18. ture is likely to be much more complicated Ameka, Felix. 1996. “Body parts in Ewe gram- than we can imagine at the moment, let alone mar”. In: Chappell, Hilary & McGregor, William discuss in this article. (eds.). The grammar of inalienability: A typological

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1508 XIV. Typological characterization of language families and linguistic areas perspective on body part terms and the part-whole Fehling, Detlev. 1980. “The origins of European relation. Berlin: Mouton de Gruyter, 783Ϫ840. ”. Folia Linguistica Historica,1:353Ϫ387. Bechert, Johannes & Bernini, Giuliano & Buridant, Feuillet, Jack (ed.). 1998. Actance et valence dans Claude (eds.). 1990. Toward a typology of European les langues d’Europe. (Empirical Approaches to languages. (Empirical Approaches to Language Ty- Language Typology-EUROTYP, 20Ϫ2.) Berlin: pology, 8.) Berlin: Mouton de Gruyter. Mouton de Gruyter. Beckman, Natanael. 1934. Västeuropeisk syntax: Garvin, Paul. 1949. “Standard Average European Na˚gra nybildningar i nordiska och andra västeuro- and Czech”. Studia Linguistica 3: 65Ϫ85. peiska spra˚k. [West European syntax: Some inno- Gensler, Orin D. 1993. A typological evaluation of vative constructions in the Nordic and other West Celtic/Hamito-Semitic syntactic parallels. Ph. D. European languages.] Göteborg: Göteborgs hög- dissertation, University of California at Berkeley. skolas a˚rsskrift. ´ ˆ Gilligan, Gary M. 1987. A cross-linguistic approach Benveniste, Emile. 1960. “ ‘Etre’ et ‘avoir’ dans to the pro-drop parameter. Ph. D. dissertation, Bulletin de la Societe´ leurs fonctions linguistiques”. USC. de linguistique de Paris 55: 113Ϫ134. (Reprinted in: Benveniste, E´ mile. 1966. Proble`mes de linguistique Haarmann, Harald. 1976. Aspekte der Arealtypolo- ge´ne´rale. Paris: Gallimard, 187Ϫ207.) gie: Die Problematik der europäischen Sprachbünde. Tübingen: Narr. Bernini, Giuliano & Ramat, Paolo. 1996. Negative sentences in the languages of Europe: A typological Haspelmath, Martin. 1990. “The grammaticization approach. (Empirical Approaches to Language Ty- of passive morphology”. Studies in Language, 14.1: pology, 16.) Berlin: Mouton de Gruyter. 25Ϫ71. Bossong, Georg. 1998. “Le marquage de l’expe´- Haspelmath, Martin. 1993. “More on the typol- rient dans les langues d’Europe”. In: Feuillet (ed.), ogy of inchoative/causative verb alternations”. In: 259Ϫ294. Comrie, Bernard & Polinsky, Maria (eds.). Caus- Ϫ Comrie, Bernard. 1998. “Rethinking the typology atives and transitivy. Amsterdam: Benjamins, 87 of relative clauses”. Language Design 1: 59Ϫ86. 120. Cristofaro, Sonia & Putzu, Ignazio (eds.) 2000. Haspelmath, Martin. 1997. Indefinite pronouns. Languages in the Mediterranean area: typology and Oxford: Oxford University Press. convergence. Milan: Franco Angeli. Haspelmath, Martin. 1998. “How young is Stan- Dahl, Östen. 1990. “Standard Average European as dard Average European?” Language Sciences 20: an exotic language”. In: Bechert et al. (eds.), 3Ϫ8. 271Ϫ87. Dahl, Östen. 1995. “Areal tendencies in tense- Haspelmath, Martin. 1999. “External possession in aspect systems.” In: Bertinetto, Pier Marco & Bi- an European areal perspective”. In: Payne, Doris anchi, Valentina & Dahl, Östen & Squartini, Mario L. & Barshi, Immanuel (eds.) External possession. (eds.) Temporal reference, aspect and actionality, (Typological Studies in Language, 39.) Amster- vol. 2: Typological perspectives. Turin: Rosen- dam: Benjamins, 109Ϫ135. berg & Sellier, 11Ϫ28. Haspelmath, Martin. 2001. “Non-canonical mark- Dahl, Östen. 1996. “Das Tempussystem des Deut- ing of core arguments in European languages”. In: schen im typologischen Vergleich.” In: Lang, Aikhenvald, Alexandra & Dixon, R. M. W. (eds.) Ewald & Zifonun, Gisela (eds.). Deutsch Ϫ typo- Non-canonical subjects and objects. (Typological logisch, Berlin: de Gruyter, 359Ϫ368. Studies in Language) Amsterdam: Benjamins. Dahl, Östen (ed.). 2000. Tense and aspect in the lan- Haspelmath, Martin. To appear. “Coordination”. guages of Europe. (Empirical Approaches to Lan- In: Shopen, Timothy (ed.) Language typology and guage Typology-EUROTYP, 20Ϫ6.) Berlin: Mou- linguistic description. 2nd ed. Cambridge: Cam- ton de Gruyter. bridge University Press. Dahl, Östen & Koptjevskaja-Tamm, Maria (eds.) Haspelmath, Martin & Buchholz, Oda. 1998. (2001) Circum-. Vol 1Ϫ2. Amster- “Equative and similative constructions in the lan- dam: Benjamins. guages of Europe”. In: van der Auwera (ed.), De´csy, Gyula. 1973. Die linguistische Struktur 277Ϫ334. Europas. Wiesbaden: Harrassowitz. Heine, Bernd. 1994. “Areal influence on grammati- Dryer, Matthew. 1989a. “Large linguistic areas and calization”. In: Pütz, Martin (ed.) Language con- language sampling”. Studies in Language 13: tact and language conflict. Amsterdam: Benjamins, 257Ϫ92. 55Ϫ68. Dryer, Matthew. 1989b. “Article-noun order”. Chi- Heine, Bernd. 1997. Possession: Cognitive sources, cago Linguistic Society 25: 83Ϫ97. forces, and grammaticalization. Cambridge: Cam- Dryer, Matthew. 1998. “Aspects of word order in bridge University Press. the languages of Europe”. In: Siewierska, Anna Jakobson, Roman. 1931. “K xarakteristike evrazij- (ed.) Constituent order in the languages of Europe. skogo jazykovogo sojuza”. [Characterizing the Berlin: Mouton de Gruyter, 283Ϫ319. Eurasian linguistic area.] Reprinted in: Jakobson,

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 107. The European linguistic area: Standard Average European 1509

Roman. 1971. Selected writings, vol. I. The Hague: Ramat, Paolo & Bernini, Giuliano. 1990. “Area in- Mouton, 144Ϫ201. fluence versus typological drift in : Kahrel, Peter. 1996. Aspects of negation. Ph. D. dis- the case of negation”. In: Bechert et al. (eds.), Ϫ sertation: University of Amsterdam. 25 46. König, Ekkehard & Haspelmath, Martin. 1998. Siewierska, Anna. 1999. “From anaphoric pronoun “Les constructions a` possesseur externe dans les to grammatical agreement marker: Why objects Ϫ Ϫ langues d’Europe”. In: Feuillet (ed.), 525Ϫ606. don’t make it”. Folia Linguistica 33.1 2: 225 51. König, Ekkehard & Haspelmath, Martin. 1999. Stassen, Leon. 1985. Comparison and universal “Der europäische Sprachbund”. In: Reiter, Norbert grammar. Oxford: Blackwell. (ed.). Eurolinguistik. Wiesbaden: Harrassowitz, Stassen, Leon. 2000. “and-languages and with- 111Ϫ127. languages”. 4.1: 1Ϫ54. König, Ekkehard & Siemund, Peter. 1999. “Inten- Stolz, Thomas. 1991. Sprachbund im Baltikum? sifiers and reflexives: a typological perspective.” In: Estnisch und Lettisch im Zentrum einer sprachlichen Frajzyngier, Zygmunt & Curl, Traci S. (eds.) Re- Konvergenzlandschaft. Bochum: Brockmeyer. flexives: Forms and functions (Typological studies Ϫ Stolz, Thomas. 1996. “Some instruments are really in language, 40.). Amsterdam: Benjamins, 41 74. good companions Ϫ some are not: On syncretism Kortmann, Bernd. 1997. Adverbial subordination: A and the typology of instrumentals and comita- typology and history of adverbial subordinators tives”. Theoretical 23.1Ϫ2: 113Ϫ200. based on European languages. Berlin: Mouton de Stolz, Thomas. 2001a. “No Sprachbund beyond Gruyter. this line! On the age-old discussion on how to de- Lazard, Gilbert. 1990. “Caracte´ristique actancielles fine a linguistic area”. Sprachtypologie und Univer- de l’‘europe´en moyen type’.” In: Bechert et al. salienforschung 54. Ϫ (eds.), 241 53. Stolz, Thomas. 2001b. “Ordinalia Ϫ Linguistisches Lazard, Gilbert. 1998. „De´finition des actants dans Neuland: Ein Typologenblick auf die Beziehung les langues europe´ennes”. In: Feuillet (ed.), 11Ϫ zwischen Kardinalia und Ordinalia und die Sonder- 146. stellung von eins und erster”. In: Igla, Birgit & Lehmann, Christian. 1984. Der Relativsatz. Tübin- Stolz, Thomas (eds.) Was ich noch sagen wollte … gen: Narr. Festschrift für Norbert Boretzky zu seinem 65. Ge- burtstag. Berlin: Akademie-Verlag. Lehmann, Winfred P. 1974. Proto-Indo-European syntax. Austin: University of Texas Press. Ternes, Elmar. 1998. “Lauttypologie der Sprachen Europas”. In: Boeder, Winfried & Schroeder, Lewy, Ernst. 1942. Der Bau der europäischen Christoph & Wagner, Karl Heinz & Wildgen, Wolf- Sprachen. (Proceedings of the Royal Irish Acad- gang (eds.) Sprache in Raum und Zeit: In memo- emy, Vol. 48). Dublin: Hodges & Figgis. riam Johannes Bechert. Band 2. Tübingen: Narr, Masica, Colin P. 1976. Defining a linguistic area. 139Ϫ152. Chicago: University of Chicago Press. Thieroff, Rolf. 2000. “On the areal distribution of Nedjalkov, Igor’ V. 1998. “Converbs in the lan- tense-aspect categories in Europe”. In: Dahl (ed.), guages of Europe”. In: van der Auwera (ed.), 265Ϫ305. Ϫ 421 455. Ultan, Russell. 1978. “Some general characteristics Nichols, Johanna. 1992. Linguistic diversity in of interrogative systems”. In: Greenberg, Joseph space and time. Chicago: The University of Chi- H. (ed.) Universals of human language, vol. 4. Stan- cago Press. ford: Stanford University Press, 211Ϫ248. Nichols, Johanna & Peterson, David A. & Barnes, Ureland, P. Sture. 1985. “Sprachkontakt und Glot- Jonathan (to appear). “Causativizing and decau- togenese in Europa”. In: P. Sture Ureland (ed.) sativizing languages”. Ms., University of Cali- Entstehung von Sprachen und Völkern: Glotto- und fornia, Berkeley, ethnogenetische Aspekte europäischer Sprachen.Tü- Ϫ Nikolaeva, Irina. 1999. Ostyak. Munich: Lincom bingen: Niemeyer, 7 43. Europa. van der Auwera, Johan. 1998a. “Conclusion”. In: Ϫ Nocentini, Alberto. 1996. “Tipologia e genesi del- van der Auwera (ed.), 813 36. l’articolo nelle lingue europee”. Archivio Glottolo- van der Auwera, Johan. 1998b. “Phasal adverbials gico Italiano 81.1: 3Ϫ44. in the languages of Europe”. In: van der Auwera Ϫ Pisani, Vittore. 1969. “Tipi d’accento nelle lingue (ed.), 25 145. dell’Europa”. In: Actes du Xe Congres International van der Auwera, Johan (ed.). 1998. Adverbial con- des Linguistes, Bucuresti, vol. II, 57Ϫ60. Bucuresti. structions in the languages of Europe. (Empirical Pokorny, Julius. 1927Ϫ30. “Das nicht-indogerma- Approaches to Language Typology-EUROTYP, Ϫ nische Substrat im Irischen”. Zeitschrift für celti- 20 3.) Berlin: Mouton de Gruyter. sche Philologie 16: 95Ϫ144, 231Ϫ266, 363Ϫ394; 17: van der Hulst, Harry & Hendriks, Bernadet & van 373Ϫ388; 18: 233Ϫ248. de Weijer, Jeroen. 1999. “A survey of word pro-

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM 1510 XIV. Typological characterization of language families and linguistic areas sodic systems of European languages”. In: van der says in memory of . Menasha, Wis.: Hulst, Harry (ed.) Word prosodic systems in the lan- Sapir Memorial Publication Fund, 75Ϫ93. [Re- guages of Europe. (Empirical Approaches to Lan- printed in Whorf (1956), 134Ϫ159.] Ϫ guage Typology-EUROTYP, 20 4.) Berlin: Mou- Whorf, Benjamin Lee. 1956. Language, thought, Ϫ ton de Gruyter, 425 475. and reality: Selected writings of Benjamin Lee Vennemann, Theo. 1994. “Linguistic reconstruction Whorf. Edited by John B. Carroll, Cambridge/MA: in the context of European prehistory”. Transac- MIT Press. Ϫ tions of the Philological Society 92: 213 282. Wintschalek, Walter. 1993. Die Areallinguistik am Wagner, Heinrich. 1959. Das Verbum in den Spra- Beispiel syntaktischer Übereinstimmungen im Wolga- chen der britischen Inseln: ein Beitrag zur geographi- Kama-Areal. (Studia Uralica, 7.) Wiesbaden: Har- schen Typologie des Verbums. Tübingen: Niemeyer. rassowitz. Whorf, Benjamin Lee. 1941. “The relation of habit- ual thought and behavior to language”. In: Spier, Martin Haspelmath, MPI Leipzig Leslie (ed.) Language, culture, and personality: Es- (Deutschland)

108. Aire linguistique balkanique

1. Ge´ne´ralite´s effet, les traits communs sont trop nombreux 2. Phonologie pour qu’ils soient le fruit du hasard. Il est 3. Syste`me verbal vrai que les spe´cialistes discutent encore de 4. Syste`me nominal la notion de « balkanisme », que l’on de´finira 5. Autres unite´s 6. Relations phrastiques ici comme un trait typologique propre a` au 7. Subordination moins trois langues de l’union. Ce trait n’a 8. Re´fe´rences pas besoin d’eˆtre unique en son genre (ainsi, l’article de´fini postpose´ existe dans les lan- 1. Ge´ne´ralite´s gues scandinaves, le « redoublement » de l’- jet se retrouve dans les langues romanes); il La linguistique balkanique est une discipline doit eˆtre le re´sultat d’une convergence qui relativement re´cente, bien que la de´couverte aboutit a` un re´sultat identique ou quasi iden- de traits communs entre les langues balkani- tique, alors qu’il n’existait pas a` des stades ques remonte a` la premie`re moitie´ du XIXe plus anciens. sie`cle. Les spe´cialistes (Asenova 1979: 5Ϫ45; Les taˆches de la linguistique balkanique Schaller 1975: 37Ϫ45) s’accordent a` diviser sont consigne´es dans l’histoire de la disci- l’histoire de la discipline en trois pe´riodes: pline. Elles ont un triple aspect: synchronique une pe´riode pre´liminaire,ou` l’on cherche a` (description) panchronique (extension) et expliquer les traits communs par l’influence diachronique (formation et e´volution). Bien du substrat, une pe´riode classique ou` la lin- que l’essentiel du travail descriptif semble guistique balkanique acquiert ses lettres de avoir e´te´ acompli (la monographie de Sand- noblesse graˆce a` la publication en 1930 de feld a e´t´e comple´te´e, souvent ame´liore´e, par Linguistique balkanique. Proble`mes et re´sul- des centaines d’articles et d’e´tudes de de´tail tats de Sandfeld, qui repre´sente la premie`re qui ont permis d’accroıˆtre et d’approfondir synthe`se comple`te, et une pe´riode moderne, les donne´es), il reste toujours beaucoup a` marque´e par le polycentrisme et l’internatio- faire. L’e´tude de l’extension des balkanismes nalisation des recherches (nombreuses revues ne´cessite le recours a` la ge´ographie linguisti- spe´cifiques et organisation de congre`s). que (ou linguistique are´ale) pour de´terminer La linguistique balkanique ne consiste pas avec exactitude le lieu d’apparition de chaque a` juxtaposer des descriptions de langues di- balkanisme et son extension re´elle sur le ter- verses dont le seul lien serait la contiguı¨te´ rain. Enfin, la perspective diachronique n’est ge´ographique: il faut que ces langues for- jamais perdue de vue par les balkanologues, ment une « union linguistique » (Sprachbund). malgre´ les nombreuses difficulte´s auxquelles Meˆme si certaines voix s’e´le`vent encore pour ils sont confronte´s, faute de documents e´crits. nier la re´alite´ de l’union balkanique (Andrio- Trois aspects sont a` prendre en conside´ra- tis & Kourmoulis 1968), la plupart des lin- tion: 1) La gene`se de l’union linguistique guistes sont convaincus de son existence. En balkanique; 2) La gene`se des balkanismes; 3)

Brought to you by | Max-Planck-Gesellschaft - WIB6417 Authenticated | 194.94.96.194 Download Date | 2/18/13 9:31 AM