PAVLE IVIĆ
BALKAN SLAVIC MIGRATIONS IN THE LIGHT OF SOUTH SLAVIC DIALECTOLOGY
Selective bibiliography ( General Works, Works on Dialects Transplanted by Migrations )
The topic of this brief report embraces two vast complexes of events : the Slavic migrations to the Balkan Peninsula, and the later migrations within the Peninsula (or away from it). Our objective, too, will be twofold: we shall use, whenever possible, dialectal data in order to determine the directions of migratory movements and the degree of ethnic coherence of the migrating population, and we shall try to establish the consequences of the migrations for the linguistic picture of the South Slavic area.
As to the first period (migrations TO the Peninsula, 6th-7th c. A.D.), the central question is: what phenomena, among those now differentiating South Slavic dialects, were developed before the migrations of Slavs to the Balkans? Only the isoglosses of such phenomena can indicate the directions of the movement to the Balkans. In this domain, South Slavic dialectology gives us a number of valuable clues, but no clear-cut answers. In trying to establish whether a dialectal difference is older than the settlement in the Balkans, we can use several criteria, such as:
(1) Is the given isogloss continued in the North Slavic area?
(2) Is the phenomenon attested in the oldest texts?
(3) Are there relics in the regions between present-day South and North Slavic areas (e.g., in toponymy or in oldest Slavic loanwords in Hungarian or Rumanian)?
(4) Does the isogloss belong to a bundle whose position cannot be explained by geographic or historical factors? If it does, this fact would increase the probability that the bundle reflects the boundary between two ancient migratory currents (i.e., between two already differentiated groups of Slavic population moving to their new habitat).
67
(5) Internal factors concerning the very nature of the given phe-
None of these criteria is fully reliable. In most instances, facts are ambiguous, and our arguments must be indirect, due to the lack of firsthand evidence. (We do not possess Slavic texts written before the migrations.)
The largest bundle of ancient isoglosses in the Slavic South is found along a line extending from the mouth of the Timok river-Berkovica-Breznik-Radomir-Osogov-northern Ovče Polje-Skopje-Tetovo-Šar Planina (with fairly insignificant deviations of particular isoglosses in the northern half of the bundle, and a considerable number of deviations in the south). Isoglosses in that bundle separate the two major groups of South Slavic dialects, western and eastern, and include features such as the merger of the two jers (West) versus the preservation of the distinction between them (East) [1]; rounded (W) versus unrounded (E) vowel as a reflex of ǫ; preservation (W) or loss (E) of epenthetic l; ć and đ or similar consonants (W) versus št, žd (E) as reflexes of *t’, *d’; -ga (W) versus -go (E) in the Gsg masculine and neuter of the pronominal declension; -mo (W) versus -m, -me (E) in the 1st person plural of verbal forms; loss (W) or preservation (E) of -t in the 3rd person plural of the present tense, etc. This is the most important bundle of ancient isoglosses in the Slavic linguistic world, except for those separating West and East Slavic. It is probable that some of the differences enumerated arose prior to the Slavic migration to the Peninsula. Before that migration, the Western South Slavic situation was characteristic of the dialects spoken in the Pannonian plain, and the Eastern situation characteristic of those in the Dacian plains, which were separated from the former by the Carpathian mountains north of the Iron Gate (Đerdap) and, probably, by the mountain range of the Munţi Apuseni, so that Slavic dialects in most of Transylvania belonged to the Eastern group. It is well-known that Slavs, originally farmers and inhabitants of plains, had no experience with the mountain type of pasturage normal in the Carpathian and Balkan mountains. Therefore we have no reason to suppose a lively communication between the two groups of South Slavs. The general direction of their movement was south or southwestwards, to the regions that belonged to the Byzantine empire, and it is probable that they followed the shortest paths to their new habitat, i.e. that their movement was
1. In the East the back jer merged as a rule with the reflex of the back nasal vowel (in Bulgarian dialects) or with o (in most of the Macedonian dialects).
68
basically perpendicular to the Danube-Sava line, so that the earliest isoglosses south of the Iron Gate simply reflected that movement.
However, in addition to the isoglosses already enumerated, the Timok—Osogov—Šar bundle comprises isoglosses of some evidently later innovations, such as vъ̯- > u- (W), čr- > cr- (W), devoicing of final voiced obstruents (E), metathesis vs- > sv- (W), -e (W) versus -i (E) in Npl of the a-declension, neutralization of gender distinction in adjectival plural (E), -ste (W) versus -hte (E) in 2nd person plural of the aorist and imperfect tenses, and -hǫ (E) in 3rd person plural of the aorist. The presence of such isoglosses can be explained by van Wijk's well-known hypothesis, according to which a close contact between the two major groups of South Slavs was established only several centuries after their settlement in the Balkans, due to the fact that they did not immediately penetrate into the mountainous area along the Serbian-Bulgarian border which remained inhabited by a population of Rumanian (and perhaps also Albanian) herdsmen for a rather long period following the Slavic invasion. [2] Only later, secondary Slavic migrations filled the gap. This hypothesis is supported to a certain extent by the presence of fairly numerous toponyms of Rumanian origin in the given area (Bukurovəc, Korbevəc, Svrļig, Džepa, Barbatovci, Surdulica, Bučumet, Maržini etc. in East Serbia, Banišora, Bazait, Bov, Bojana, Gavnos, Ursul, Vakarel, Pasarel etc. in West Bulgaria). Although toponyms with Rumanian etymologies are not infrequent in many other South Slavic regions, their number in the area near the Serbian-Bulgarian boundary appears to be higher than elsewhere. [3] It is very likely that Slavic migrations which eventually filled the gap between the two branches of South Slavs came chiefly from the West. This can explain the curved shape of the isogloss bundle : Western
2. Medieval sources record the existence of a Rumanian-speaking stock-breeding population in many mountainous areas of the Peninsula, and linguistic arguments such as phonological features of certain place names and the lack of indigenous maritime terminology in Albanian favor the hypothesis ascribing the origin of the Albanians to the interior of the Peninsula. In the epoch of the Slavic invasion the remnants of the autochthonous population of the Peninsula managed to survive mainly in two cases: when they were sheltered by fortress walls (only along the seashore; all Byzantine inland fortresses were taken by invaders after they succeeded in cutting their supply lines), or when they were protected by natural conditions in the mountain ambiences (and by the conquerors' lack of interest in such regions). This caused a population influx into the coastal towns, and probably also a switch from agriculture or from an urban life to a usually semi-nomadic kind of sheep-breeding by many inhabitants of the continental regions. Among the non-Slavic Balkan languages, Greek and Dalmatian Romance were spoken in coastal ureas, and Rumanian and Albanian in highlands.
3. However, there is no historical evidence to the effect that Rumanians lived in the region under consideration in any greater number, or longer, than in some other areas.
69
features extend across the perpendicular intersecting the Danube river at the Iron Gate, and their areas form a bulge, whereas the areas of Eastern features are concave. The causes that determined such a development are not clear. True, a southeastward movement of parts of the Serbian population took place during the expansion of the Serbian state in the 13th and the 14th centuries, but this does not suffice to explain the actual situation. The territory of the Timok—Lužnica dialectal type lies west of the isogloss bundle, and yet never belonged to the medieval Serbian state.
The linguistic gap between the two groups of South Slavs was so deep that even certain more recent phenomena, such as the elimination of h, did not spread across the already existing borderline. Different circumstances in the two patterns determined divergent paths of further evolution; perhaps a role was also played by the awareness of a linguistic non-identity creating a psychological barrier against the adoption of innovations coming from the other side of the line.
The relative importance of the factors enumerated so far cannot be easily ascertained. It is even possible, although not very likely, that none of the isoglosses mentioned continues a dialectal division older than the Slavic invasion of the Balkans; even the most ancient differences may have arisen in connection with the territorial gap after the settlement south of the Danube. Likewise, it is not quite certain that such a gap ever existed. Some of the differences may have appeared prior to the crossing of the Danube, and the rest might reflect a divergent evolution of already differentiated dialects. We might conclude that the hypotheses concerning the role of the gap (a) BEFORE and (b) AFTER the migration stand in an 'and/or' relation : at least one of them should be true, or else the concentration of isoglosses along the line Timok—Osogov—Šar would be inexplainable.
The hypotheses (a) and (b) have a bearing upon certain much-debated problems of the ancient ethnic history of South Slavs and other Balkan nutions. If (a) is true, it is more probable that central Pannonia never wus inhabited by Eastern South Slavs (those with št, žd < *t’, *d’), and that Slavic loanwords in Hungarian with Eastern South Slavic phonetic structure were borrowed in Dacia, where the Hungarians lived for a certain time before invading Pannonia. In such a case toponyms such as Pest should be regarded as based on appellative nouns of Slavic origin ulrcady present in the Hungarian language. And if hypothesis (b) is true, it increases the probability of the well-known theory that the original habitat of Rumanians was south of the Danube, and also the probability
70
of the view that Albanians originally lived in Dardania and only later migrated west and southwestwards. However, the mutual interchangeability of the two hypotheses weakens the value of arguments derived from them. And in addition, there is a third possibility to explain the present concentration of isoglosses separating the two major South Slavic linguistic types.
The isoglosses now concentrated in the bundle along the line Timok— Osogov—Šar may originally have been dispersed, but later pushed together by secondary migrations bringing large masses of new population into the region, so that the original local dialects, sharing both western and eastern features, were overwhelmed by the dialects of new comers. True, there are no reliable traces of early transitional dialects in the present dialectal picture of the region, covered mainly by Serbo-croatian Torlak dialects (particular words or morphemes with a as a reflex of ǫ, or with o, e as reflexes of the two jers in some limited areas northwest of the line are obviously secondary, introduced by later migrations or by lexical borrowing). Nonetheless, in the toponymy there are some instances indicative of an earlier wider distribution of št, žd as reflexes of *t’, *d’: Ljuberažda south of Pirot, Dragobužde, Tibužde, Roždace in the district of Vranje, Pobužje near Skopje, Graždenik, Obražda, Ljubižda, Toražda, Dobrušta, Nebregošte, Selogražde in the environs of Prizren. [4] Since some of these toponyms contain u as the reflex of ǫ (always in -bužd-, the possessive adjective form of the second component of compound personal names in *-bǫdъ), Macedonian scholars B. Koneski and B. Vidoeski recently suggested that a dialect with u < ǫ but št, žd < *t’, *d’ must have existed in the Middle Ages. This appears to be the most plausible explanation of the toponyms containing the element -bužd-. But such toponyms are limited to the region of Vranje and south of it, whereas Nebregošte seems to show that in the environs of Prizren not only the isogloss of *tj, but also that of ǫ > u was formerly placed somewhat more to the north than it is today. And in general, toponyms with št, žd occur only in areas fairly close to the present-day isogloss bundle; they do not cover the entire area which was probably embraced by the eastwards migrations of ć, đ speakers. The number of these toponyms is limited, most of them occur in clusters, and it is possible that they reflect a sporadic immigration from the east or south rather than the former general dialectal picture of the whole area. As to the toponyms in -bužd-, one cannot exclude the possibility of a secondary
4. Several of these place names were recorded with žd or št already in medieval charters.
71
mixture of dialects, or of an expansion of the possessive formation in žd, or of personal names in -bud (perhaps under the influence of the stem bud- of buditi etc.). In other words, the presence of toponyms containing št, žd < *t’, *d’ in the area northwest of the isogloss bundle does not allow apodictic conclusions regarding the linguistic past of the region, but suffices to render less conclusive the evidence of other facts.
Another important bundle of ancient isoglosses in the Slavic South is that separating the Kajkavian (and Slovenian) dialects from the Štokavian and Čakavian ones. These isoglosses include phenomena such as: a much broader occurrence of the so-called neo-circumflex, the lengthening of the short neo-acute (at least in the majority of instances), rj rather than r as the reflex of prevocalic *r’, -ǫ rather than -ojǫ in the Isg of the a-declension (all this occurs northwest of the bundle, but in some cases also partially in NW Čakavian). [5] Since there was no specific political link between Slovenia and the Kajkavian area in northern Croatia before the 16th century, and since the geographic conditions in the present habitats of Slovenians and Kajkavians would not favor their common linguistic development, distinct from that of their eastern and southern neighbors, it seems likely that their common linguistic features stem from the propinquity of their ancestors in the period preceding their settlement in what is now Yugoslavia. It should be added that the present strong concentration of isoglosses in the bundle along the eastern and the southern border of the Kajkavian area might be a secondary phenomenon. It is possible that a number of transitional dialects originally existed east of that line, but that they were later obliterated by the massive migrations in the period of the Turkish invasion.
The sound change g > γ and the use of the prefix vi- (< vy-) rather than *jьz- in a number of instances, both characteristic of northwestern Slovenian and the northwesternmost Čakavian dialects, probably belong to the features which differentiated (although possibly realized in a less clear-cut way) Slavic dialects already before the Slavic migration to the Balkans and the Eastern Alps. Basically the same applies to the preservation of the cluster dl in some northern Slovenian dialects. It is obvious that the northwesternmost South Slavic dialects once constituted a kind of transition between the South and the West Slavic linguistic groups.
5. It is possible that a less narrow pronunciation of the ancient back nasal vowel, yielding o (or ə, or u) rather than u as a result of denasalisation, should be added to the above list of features. The vowel u appears as a reflex of ǫ only in two peripheral parts of the Croatian Kajkavian territory: in the east, where a strong Štokavian influence is obvious, and along the western border, where a numerous Čakavian population was colonized in the 16th century.
72
The subsequent rupture of the geographic contact with West Slavic determined the decidedly South Slavic orientation of the later development of those dialects.
In Slovenian and in most of the western Serbocroatian dialects *d’ became j, whereas in other parts of the Serbocroatian linguistic territory this change did not take place. In those more eastern dialects of western South Slavic, the ancient clusters *st’, *sk’, *zd’, *zg’ underwent, in turn, a simplification to št, žd, embracing also all of the eastern South Slavic area. In both cases, we cannot exclude the possibility of a very early change, in which case the two isoglosses would be indicative of the course taken by the southward movement of two branches of western South Slavs.
The general direction of the oldest isoglosses in the South Slavic area is N—S or NNE—SSW; only in the northwest of the area is this direction sometimes NE—SW. All this is in perfect agreement with our assumptions, deriving from other reasons, regarding the course of the movement of South Slavic settlers arriving to their new habitat. Obviously, this increases the probability that at least some of those isoglosses are not younger than the 6th or the 7th century A.D. Only one isogloss of an undoubtedly very early origin has a deviating direction. The otherwise North Slavic ending *-ъmь in the Isg of masculine and neuter nouns appears in the dialects of Krašovani and Svinica in the Rumanian Banat, both deriving from the northeasternmost branch of the Torlak dialect group. In other words, this is a link between a small area located at about the center of the northern edge of the South Slavic world, and the Slavic North.
The best known migrations of South Slavs since they settled in the Balkans are those which occurred in the epoch of the Turkish invasion and domination (15th-19th c.). The bulk of the literature on the South Slavic migrations deals with that period. However, we have many reasons to believe that in the preceding centuries, too, there was much movement among the South Slavic population. In all probability, the economic causes of the later migrations were already present in those times; nor were wars and devastations lacking. To be sure, our data concerning the South Slavic migrations of that epoch are rather scarce. Most of the information at our disposal about later movements stems from two sources: folk tradition and dialects. Tradition fades with centuries, and transplanted dialects tend to assimilate to the surrounding ones, especially if the divergence is not far-reaching (and this was the case in the early epoch of the development ol'the Balkan Slavic languages).
73
Even the written historical sources are much less rich than those concerning more recent centuries. However, Byzantine sources testify that very early large groups of Slavs (defeated rebels or prisoners) were colonized by Byzantine emperors in Asia Minor. It is also known that in the 13th century a group of Bulgarians was forced to emigrate to Transylvania (and subsisted for a number of centuries in the village of Cserged), and that some of the Slavic settlements in southern Italy (Vasto, Molise, [6] Terra d'Otranto, environs of Foggia) were in existence already about 1300. It is not surprising that in all three cases our information concerns Slavs who settled in a non-Slavic environment. Slavic colonies surrounded by other Slavs had less chance to be noted by history. [7] Nevertheless, historical records show that the Serbian penetration into Macedonia in the 13th and 14th centuries was followed by a colonization of a considerable part of the Serbian nobility (probably accompanied by some of their servants or subordinates) in various portions of Macedonia. The impact of these events on Macedonian dialects was very significant. In addition, it is clear that the bulged shape of the isogloss bundle along the line Timok—Osogov—Šar can be explained only by an early eastward movement of a western South Slavic population — regardless of whether we assume that the territory covered by that migration previously was mostly unpopulated, inhabited by non-Slavs, or by Slavs speaking dialects different of those of the newcomers. Significantly enough, a number of important movements of non-Slavic Balkan peoples took place in the same period. Massive migrations of Arumanians brought them as far south as Thessaly and the Grammos mountains. Large groups of Albanians were transferred to southern and central Greece, and others to southern Italy. It is also likely that parts of the Slavic population were involved in similar movements. The question deserves further, more systematic study.
The momentous migrations of Balkan Slavs that started in the 15th and lasted until the 19th century were brought about by two types of causes, political and economic. Political causes include the ravaging wars in the epoch of the Turkish
6. Early Slavic settlers in Molise, whose presence was attested in 1294, subsequently became italianized. Existing Slavic colonies in Molise (Acquaviva, Montemitro, San Felice Slavo) date from the first half of the 16th century.
7. It is also natural that in all three instances descendants of the settlers eventually abandoned their Slavic dialects. This suggests that if other similar early colonies had existed in neighboring coountries, unrecorded by historical sources, they would also be ethnically assimilated to their non-Slavic surroundings.
74
invasion, pillaging expeditions of Turkish military commanders, adverse circumstances of life in the regions already dominated by Turks, enticement of the population by the neighboring Christian states trying to organize a defense of the borderline and to colonize devastated land alongside it, efforts of Turkish governors and commanders to colonize the regions on the Turkish side of the boundary, and fear of retaliation by Christians who had cooperated with Christian armies during their incursions into the Turkish territory after a subsequent retreat of those same armies. The general direction of migrations triggered by most of these causes was away from the Turkish territory towards that under Christian rule. In the western part of the South Slavic area this usually meant a movement to the Venetian, Austrian, or Hungarian territory (thus E—W, SE—NW, or S—N), and in the Bulgarian area to the Austrian-dominated province of Banat, to Rumania, or to Russia (thus SE—NW, S—N, or SW—NE). An exceptional case is that of reverse movements of Moslems trying to avoid occupation by Christian armies, or living under Christian rule after the wars which led to an expansion of Christian states.
Economic causes of migrations include infertility of the soil in the highlands, overpopulation of those regions, [8] and famine as a consequence of drought, especially in the barren mountains of the Dinaric mountain range. An additional impetus came from the fact that the population of plains and valleys was often decimated by wars, plagues, emigration, etc. The general direction of population movements brought about by such causes was from the highland to the lowland. In the western half of the Peninsula, this most frequently meant from the Dinaric mountains, or from those along the southern border of Serbia, to Pannonia or to the valleys of the Sava, Drina, Kolubara, Morava rivers, etc., thus SW—NE or S—N, whereas in Bulgaria the Balkan mountain range acted as the prime source of economic migrations, in the direction of the Moesian as well as the Thracian plains, i.e., both S—N and N—S.
On the basis of dialect facts, supported sometimes by written documents or preserved oral tradition, it is possible to reconstruct the following picture of the principal migratory movements. [9]
8. The mobility of highlanders was enhanced by their traditional semi-nomadic way of life. The classical pattern of the Balkan pasturage includes spending summers on high plateaus and winters in valleys.
9. In the present paper, the 20th century colonization, mainly organized by Yugoslav or Bulgarian authorities, is not taken into account.
75
Speakers of the Čakavian ikavsko-ekavski dialect, coming from Croatia south of the Kupa river, settled in the Austrian-dominated portions of Istria, in the southwestern parts of the island of Krk (in both cases they were accompanied by a Rumanian-speaking population), in the region of Grobnik near Rijeka, [10] in some places in Slovenia (especially around Kostanjevica north of the Gorjanci mountains), in Kajkavian northern Croatia, in northern Burgenland and the adjacent part of Hungary, in eastern Lower Austria, southern Moravia and southwestern Slovakia.
Speakers of a Čakavian i-dialect, with some Štokavian (and even some Kajkavian) elements, originating probably from the lower part of the Una basin, migrated to southern Burgenland and to the neighboring region of Szombathely on the Hungarian side of the present border. [11] A somewhat different but still closely related dialect was transplanted to the region between the Sutla, Sava, and Krapina rivers west of Zagreb.
Štokavian-Čakavian i-dialects (with various shades, and with varying shares of Štokavian and Čakavian features) were transplanted from Dalmatia to the Venetian-dominated parts of Istria and to the province of Molise in southern Italy (numerous other Croatian colonies in various provinces of the former Kingdom of Naples are now extinct, so that we cannot judge about their dialects). [12]
Štokavian i-speakers from western Hercegovina and adjacent parts of Bosnia moved to Dalmatia (remaining mostly on the mainland, and only in small number crossing the straits separating the Dalmatian islands from the mainland), to Lika (and thence to Bačka [13] and along the Danube to the regions south and north of Budapest), and to various portions of Bosnia.
The Štokavian i-dialect of southeastern Slavonia was transplanted to various places in western Bačka and eastern Baranja. A Slavonian dialect which preserved the phonemic individuality of ě is spoken in two villages near Kalocsa in Hungary.
10. Features of the dialect of Grobnik show clearly that the population is not autochthonous; it come probably from a portion of the Croatian littoral more to the south.
11. A striking similarity can be observed between the dialect of Vlahija in southern Hurgcnland described by Brabec and the dialect of Narda near Szombathely described by Ivić (Prilog rekonstrukciji).
12. Recent works by Rohlfs and Hraste (1963) have drawn the attention of scholars to traces of the former presence of a Serbocroatian-speaking population in Slavic lounwords in the Italian dialects of the Peninsula of Gargano and in the family names in that region.
13. In Bačka the i-speakers, called Bunjevci, occupied in the 17th century the places abandoned in 1598 by Serbians who migrated from there to Austrian territory in present-day Slovakia (and who subsequently disappeared as an ethnic group).
76
Štokavian je-speakers with šć < *st’, *sk’, originating from northeastern Bosnia, settled around Pécs in Baranja, and sporadically in some places in northern Croatia. The dialect of Kukinj near Pécs has ē < long ě and je < short ě, like the dialect of Maglaj and Tešanj in northern Bosnia.
Štokavian je-speakers of East Hercegovinan origin were the most numerous among the migrants. They colonized large portions of western Serbia, parts of eastern Bosnia, most of western Bosnia, considerable portions of the mainland of northern Dalmatia and of Croatia south of the Kupa river, a small number of places in Gorski Kotar, several enclaves north of Kupa, near or even beyond the Slovenian border (e.g., Bojanci, Marindol, parts of Žumberak), the region along the western border of Slavonia including some enclaves in the eastern Kajkavian territory such as the stretch around Bjelovar, various parts of Slavonia, among others Osječko Polje and the slopes of the Papuk mountain, southeastern Baranja, parts of the Tolna district in Hungary (where the Serbocroatian language is now almost extinct), and the region farther to the north (to Budapest and even further, where the je-speakers were mixed with those of the e-dialect, the features of which prevail in the surviving colonies).
Štokavian je-speakers with unshifted accents from eastern and southern Crna Gora settled in the Istrian village of Peroja and appeared dispersed in many places in Serbia. Some of their settlements were founded only in the 19th century, e.g., Petrovo Selo near the Iron Gate, or certain villages in Toplica.
The e-dialect with the Neoštokavian accent (šumadijskovojvodjanski) expanded its territory in present-day Vojvodina and beyond its boundaries into various parts of historical Hungary, among others to the region of the Moriš (Maros) river around Arad and west of that town, to the city of Budapest and towns and villages north and south of it, to some other towns such as Székesfehérvár (where the Serbocroatian language is now virtually extinct), Komárom (completely extinct), Eger (also extinct), etc. Speakers of this dialect emigrated from northern Vojvodina and the region of Moris to what is now Ukraine where they founded two groups of colonies: Novaja Serbija south of the Dnepr river (between the Dnepr, Sinjuxa, Bug and Tjasmin), and Slavjanoserbija south of the Donets, from Baxmut to Lugan. In both areas the Serbians underwent a complete ethnic and linguistic assimilation.
Štokavian dialects with preserved ě and with predominantly unshifted accents were dispersal from returns which later became ekavian in
77
northern Serbia (and perhaps also in southern Banat) to parts of eastern Banat, where they survive in Rekaš and in Banatska Crna Gora. Speakers of basically the same dialect were transplanted to the environs of Gallipoli in Turkey.
The Štokavian e-dialect with a mainly conservative accentuation spread from the regions of Metohija, NW Kosovo and Kopaonik to the valleys of Western and Great Morava, NE Serbia, southern Banat (including Banatska Klisura beyond the present Rumanian border), to parts of Šumadija and even to the region north of Budapest where a dialect of this type is still spoken in the village of Čobanac.
The dialect of the region of Prizren and SE Kosovo expanded into the valley of the Southern Morava, pushing the former dialect of the valley (now called svrljiško-zaplanjski) onto the hills east of the valley.
The Torlak dialect of the highlands along the Serbian-Bulgarian border was transferred to a number of places in eastern Šumadija and to the environs of Belgrade (villages Banjica, Rakovica, Jajince, Mali Mokri Lug, Veliki Mokri Lug, and Višnjica), to Novo Selo near Vidin, and to various other places in NW Bulgaria, including a group of twelve villages near the mouth of the Cibǎr river. The so-called Krašovani near Reşiţa in the Rumanian part of Banat, as well as the inhabitants of Svinica near the Iron Gate, are early colonists, probably from the Timok region. Their dialects display significant common features with that of Novo Selo near Vidin.
Speakers of the northwesternmost Macedonian dialects settled in various places farther to the northeast and east. Most of those colonists belonged to the Islamized group called Torbeši who settled in sporadic groups of villages in the districts of Tetovo, Skopje, Veles, Kičevo, Prilep etc. The movement of the Torbeši was paralleled by Albanian migrations basically in the same direction.
In both western and eastern Bulgaria, highlanders from the Balkan mountains descended in the direction of the plains situated along the Danube (north of the Balkan mountain range) and the Marica (south of the Balkan mountains), and also along the Black Sea coast. The dialects of their settlements show their origin from various parts of the Balkan region. One of the strongest movements was that from the central part of the Balkan mountain range northwards, to the regions of Tărnovo, Gorna Orjahovica, Svištov and Nikopol. Another group of settlers, coming from the 'sub-Balkan' (podbalkanski) area around Stara Zagora, Sliven and Jambol, crossed the Balkan ridge and settled in the easternmost parts of northeastern Bulgaria, mainly near Varna, Balčik and
78
Tolbuhin. A specially instructive case is that of Erkeč and Golica, villages in the eastern part of the Balkan area, whose inhabitants colonized a number of places in the districts of Varna, Provadija, Novi Pazar, Balčik, and Silistra.
The region of Teteven, with a specific kind of East Bulgarian dialect, was the source of a migration westwards, to the region of Kula in NW Bulgaria and Zaječar in eastern Serbia.
Highlanders from the Rhodope mountain region in South Bulgaria settled in the Marica valley, in the Strandža hills and in the coastal plain along the Aegean Sea.
One part of the Roman Catholic Bulgarians called Pavlikijani, who live in the Thracian plain near Plovdiv and whose dialect shows their origin from Rhodope, migrated to the district of Svištov in North Bulgaria, whereas others colonized several villages in Banat where they mixed with another group of Catholic Bulgarians, coming from Čiprovci in NW Bulgaria and speaking originally a Torlak dialect. Descendants of the latter subsequently adopted the Pavlikijani dialect, still alive in several villages in Banat, predominantly on the Rumanian side of the border between that country and Yugoslavia. [14]
Groups of Bulgarians settled in various places in southern Rumania.
Emigrants from Bulgaria were colonized in Bessarabia (mostly in its southern part) and the southern Ukraine (chiefly in the region of Odessa). Their extremely variegated dialects include many northeastern Bulgarian as well as some southeastern and (only in two villages) western Bulgarian types. About 1861, after the Crimean War in which Russia had lost southern Bessarabia, numerous Bulgarians emigrated further to the region north of the Azov Sea. A characteristic feature of Bulgarian colonies in (former) Russia is the frequent phenomenon of coexistence or mixture of two (or even three or four) dialects in the same village.
The consequences of the migrations in the Turkish period were manifold:
(i) The distribution of dialectal types, especially in the Serbocroatian territory, underwććent far-reaching changes. Areas of certain dialects were substantially enlarged (e.g., those of the je-dialect of East Hercegovina, the i-dialect of West Hercegovina, the Šumadija-Vojvodina e-dialect, the Kosovo-Resava dialect, the Prizren-Southern Morava dialect, certain
14. The far-reaching influence of denominational differences on migrations is clearly shown by the facts that the only Serbocroatian-speaking groups which settled in Russia were Eastern Orthodox, and the only Bulgarians colonized in Austrian provinces, Roman Catholics. And of course, all Serbocroatian and Bulgarian-speaking emigrants who settled in Turkey in the late 19th and in the 20th centuries were Moslems.
79
westernmost Macedonian, and the Balkan Bulgarian dialects). Certain other dialectal areas shrank, for instance, those of the Čakavian and Kajkavian dialects, as well as of most among the western Štokavian dialectal types with šć < *st’, *sk’. Some dialects disappeared from their original places (e.g., the i-dialects spoken now in southern Burgenland and in the area west of Zagreb), or were even eradicated completely (e.g., the majority of medieval dialects of northeastern Serbia). Boundaries between South Slavic languages and their non-Slavic neighbors were shifted at several places. In Vojvodina and further to the north, the Serbocroatian language expanded strongly; however, the more recent colonization of Germans, Hungarians, Slovaks, Ukrainians and Czechs reduced again the Serbocroatian language territory in the area. In southern Serbia and western (and even northern) Macedonia, the Albanian language expanded at the expense of Serbocroatian and Macedonian. In Bulgaria, numerous Turkish immigrants occupied entire districts, so that Bulgarian dialects virtually disappeared from Deliorman, Tuzluk, large parts of the eastern Balkan mountains, the Eastern Thracian plain, and the eastern Rhodope. Speakers of Serbocroatian and Bulgarian settled in various places in Italy, Austria, Moravia, Slovakia, Hungary, Rumania, southern Russia, and Turkey. A very large portion of those emigrants subsequently abandoned their native idioms, undergoing assimilation to their new environment.
(ii) Certain dialectal types were affected by a strong influence of another dialect whose speakers had settled in their areas or around them. E.g., some western Štokavian dialects in Slavonia and Bosnia, which originally had a number of peripheral peculiarities forming a transition, as it were, to Čakavian, became much closer to average Štokavian after the colonization of East Hercegovina je-speakers in Bosnia and Slavonia, and the rupture of territorial contact with Čakavian due to that colonization.
(iii) In various regions, the typological characteristics of linguistic differentiation were deeply altered. In most of the Serbocroatian language area the organically developed picture of intersecting, mutually dependent isoglosses, without clear-cut dialect boundaries and with gradual transitions everywhere, was replaced by a situation where isoglosses as a rule come in bundles, where territorial dialects are well-defined entities with sharp boundaries and relatively little internal differentiation. Another striking feature is the irregular shape and the discontinuity of the areas of many dialects due to migrations. The phenomenon of a dialectal enclave has become widespread. In a number
80
of instances, a 'pulverization' of dialects took place. Settlers from distant regions and with dissimilar dialects formed colonies in small groups, making the dialect picture extremely complex. At various places new dialect types arose from a mixture of heterogeneous dialects. Some of those dialects still are in a process of crystallization, characterized by parallel occurrence of numerous alternative forms.
(iv) The tremendous expansion of central Štokavian dialects created a broad dialectal basis for the modern standard language of Serbians and paved the way to the adoption of that language by Croatians in the second half of the 19th century, and thus to the creation of a common Serbocroatian standard language with basically unified phonological and grammatical patterns.
(v) In a number of areas where the population is mixed as to religion, and especially in western Bosnia and the adjacent parts of Croatia, dialectal features became associated with faith. E.g., (i)je < ě is now a peculiarity of the Eastern Orthodox population, and i < ě of the Roman Catholics (in most of West Bosnia also of the Moslems). This is chiefly a consequence of the fact that adherents of different denominations usually immigrated from different regions.
The linguistic consequences of the migrations provide an important insight into the historical study of certain South Slavic languages. Dialects of the diaspora are in a sense living documents concerning the language of the epoch of their transplantation. In turn, the profound changes caused by migrations impose specific tasks on the diachronic investigation of the same languages. It becomes necessary to determine the dialectal picture in the period prior to the migrations. This can be achieved by utilizing the data of ancient texts (wherever they exist) and of emigrant dialects, possible traces of the dialectal substratum in areas now covered by imported dialects, and in some cases also the information conveyed by toponymy. However, the results almost inevitably remain fragmentary. As an illustration, we shall mention some of the unclarified problems :
1) Since most of the present population of the region Gorski Kotar in western Croatia consists of relatively recent settlers, it is not clear where the original boundary between Slovenian and Čakavian dialects in that area was. It is possible that some stretches of land, especially those densely wooded ones immediately south of the Kupa river, were unpopulated for a very long time and subsequently colonized from various (and not yet all identified) parts of Slovenia, so that dialects of neighboring villages often differ considerably.
81
2) We lack information about the original southern and eastern border of the Kajkavian area in Croatia, as well as about possible ancient transitional Kajkavian-Štokavian dialectal types.
3) The present dialects of northwestern Bosnia, especially those between the Una and Vrbas rivers, were imported with migrations. Most of the ancient population has disappeared, probably in those hard days when the region ("Jajačka banovina") was the first defense line of Christians against the Turks. Now we cannot even tell where the original Čakavian-Štokavian boundary was (or, more exactly, where the isoglosses of the features differentiating Štokavian dialects from the Čakavian ones were).
4) The original distribution of šć and of št as reflexes of *sk’, *st’ in eastern Bosnia remains unclear, as well as the origin of the št-speaking group of Catholics in southeastern Slavonia.
5) We know little about the ancient isogloss of the ikavian and jekavian reflexes of jat in northern Bosnia, and also about the eastern border of the jekavian area before the powerful expansion of je-dialects due to migrations.
6) The medieval linguistic situation in what is now Vojvodina has not yet been completely determined. We know that both Hungarian and Serbocroatian were spoken in that region, and that the population of certain districts probably was mixed. We also have reasons to believe that Serbocroatian was more widespread in Srem and the southern part of Banat than in Bačka. However, we lack a clear picture about the ratio of the two populations, and about their geographic distribution. (Later, after the Turkish occupation, Hungarians disappeared almost completely from Vojvodina; the present Hungarian population was colonized in the 18th and 19th c).
7) Likewise, we do not possess full information on circumstances in the Serbocroatian-Rumanian contact zone in northeastern Serbia, and along the Serbian-Albanian and the Macedonian-Albanian borders. It is known that a large part of the Rumanians in northeastern Serbia settled in the 18th and 19th centuries from Banat (so-called Ungurjani) and from Oltenia ("Carani"), and that during the same period the Albanian population in southern Serbia and western Macedonia was greatly strengthened by immigration. Nonetheless, we know little about the possible share of Rumanians and Albanians in the population of the given regions in the epoch preceding the migrations.
8) The problem of the ancient boundary between Štokavian and Torlak dialects in central and northeastern Serbia is far from elucidated.
82
We are even unable to tell whether there were old transitional types.
Some of the questions enumerated will probably be at least partially clarified by future investigations, but we do not have the right to hope for an all-embracing success. It is impossible to reconstruct completely a world of dialects that has been erased by history.
a) General Works
Bălgarski dialekten atlas. I. Jugoiztočna Bălgarija, Săstaven pod răkovodstvoto na St. Stojkov i S. B. Bernštejn (Sofija, 1964); II. Severoiztočna Bălgarija, Săstaven pod răkovodstvoto na St. Stojkov (Sonja, 1966).
Barić, H., Lingvističke studije (Sarajevo, 1954).
Belić, A., Dialektologičeskaja karta serbskago jazyka, Stat'ji po slavjanovedeniju II (S.-Peterburg, 1906).
____, Dijalekti istočne ijužne Srbije (= SDZ 1), (Beograd, 1905).
____, Galički dijalekat (= SDZ 7), (Beograd, Sremski Karlovci, 1935).
____, "Les rapports mutuels du serbo-croate et du Slovène", RES 1 (1921), 20-27.
____, "O srpskim ili hrvatskim dijalektima", Glas SKA 78 (1908), 60-164.
Bernštejn, S. B., et al, Karpatskij dialektologičeskij atlas (Moskva, 1966).
Birnbaum, H., "Balkanslavisch und Südslavisch: Zur Reichweite der Balkanismen im südslavischen Sprachraum", Zeitschrift für Balkanologie 3 (1965), 12-63.
____, "The Dialects of Common Slavic", Ancient Indo-European Dialects, ed. by H. Birnbaum and J. Puhvel (Los Angeles, 1966), pp. 153-197.
Bratanić, B., "Uz problem doseljenja južnih Slavena", ZRFFZ 1 (1951), 221-250.
Brozović, D., "O problemu ijekavskošcakavskog (istočnobosanskog) dijalekta", HDZ 2 (1966), 119-208.
____, "O rekonstrukciji predmigracionog mozaika hrvatskosrpskih dijalekata", Filologija 4 (1963), 45-55.
Conev, B., Istorija na bălgarskija ezik I-III (Sofija, 1919-1937).
Cvijić, J., La Péninsule balkanique, Géographie humaine (Paris, 1918).
Densuşianu, O., Histoire de la langue roumaine, I (Bucarest, 1929), II (Paris, 1938).
Duridanov, I., "Novi danni ot toponimijata za izčeznalo rumänsko naselenie v Sofijsko", Ezikovedsko-etnografski izsledvanija v pamet na akademik Stojan Romanski (Sofija, 1960), pp. 469-478.
Georgiev, V., "Naj-starite slavjanski imena na balkanskija poluostrov i tjahnoto značenie za našija ezik i našata istorija", Bălgarski ezik 8 (1958), 321-342.
____, Trakijskijat ezik (Sofija, 1957).
Grafenauer, B., Die ethnische Gliederung und geschichtliche Rolle der westlichen Südslaven im Mittelalter (Ljubljana, 1966).
____, "Nekaj vprašanj iz dobe naseljevanja južnih Slovanov", Zgodovinski časopis 4 (1950), 23-126.
____, "Prilog kritici izvještaja Konstantina Porfirogeneta o doseljenju Hrvata", Historijski zbornik 5 (1952), 1-56.
____, "Slovanski naselitveni valovi ba Balkanski poluotok”, Zgodovinski časopis 18 (1964), 219-227.
83
Grégoire, H., "L'origine et le nom des Croates et Serbes", Byzantion 17 (1944-1945), 88-118.
Gyóni, M., "La transhumance des Vlaques balkaniques au Moyen âge", Byzantino-slavica 12 (1951), 29-42.
Hauptmann, Lj., "Seoba Hrvata i Srba", Jugoslovenski istoriski časopis 3 (1937), 30-71.
Istorija na Bălgarija, ed. BAN. I (Sofija, 1954).
Istorija naroda Jugoslavije I–II (Beograd, 1953, 1960).
Ivić, A., Istorija Srba u Vojvodini (Novi Sad, 1929).
Ivić, P., Die serbokroatischen Dialekte. I: Allgemeines und Die štokavische Dialektgruppe (‘s-Gravenhage, 1958).
____, Dijalektologija srpskohrvatskog jezika. Uvod i štokavsko narečje. (Novi Sad 1956).
____, "O nekim problemima naše istoriske dijalektologije", JF 21 (1955-1956), 97-129.
____, "Über den spezifischen Charakter der mundartlichen Ausgliederung des serbo kroatischen Sprachgebietes", Orbis (Louvain) VII/1 (1958), 134-140.
____, "Značaj lingvističke geografije za uporedno i istorisko proučavanje južnoslovenskih jezika i njihovog odnosa prema ostalim slovenskim jezicima", JF 22 (1958), 179-206.
Ivšić, S., "Hrvatska dijaspora u XVI. vijeku", Ljetopis JAZU 50 (1938), 99-102.
____, Jezik Hrvata kajkavaca. Ljetopis JAZU 48 (1936), 47-88.
Izvori za bălgarskata istorija, ed. BAN. I-XIV (Sofija, 1957-1968).
Jagić, V., "Ein Kapitel aus der Geschichte der südslavischen Sprachen", AfslPh 17 (1895), 47-87.
Jankulov, B., Pregled kolonizacije Vojvodine u XVIII i XIX veku (Novi Sad, 1961).
Kniezsa, L, Ungarns Völkerschaften im XI. Jahrhundert (Budapest, 1938).
Kodov, X., and S. Mladenov, Bit iezik na trakijskite i maloazijskite bălgari. II. Ezik. (Sofija, 1936).
Koneski, B., Istorija na makedonskiot jazik (Skopje-Beograd, 1965).
Lemerle, P., "Invasions et migrations dans les Balkans depuis la fin de l'époque romaine jusqu'au VIIIe siècle", Revue historique 211 (1954), 265-308.
Manojlović, S., "Glavne étape u razvitku hrvatskosrpskog jezika. Radovi Filozofskog fakulteta u Zadru. god. 7, sv. 7", Razdio lingvističko-filološki 4 (1968), 46-69.
Marguliés, A., "Historische Grundlagen der südslavischen Sprachgliederung", AfslPh 40 (1926), 197-222.
Melich, J., A honfoglaláskori Magyarország (Budapest, 1929).
Miklosich, F., "Über die Wanderung der Rumunen in den Dalmatinischen Alpen und den Karpaten", Denkschriften der Kais. Akad. der Wissenschaften. Philosophisch-historische Classe 30 (Wien 1880), 1-66.
Miletič, Lj., Das Ostbulgarische (Wien, 1903).
Miletič, L., Staroto bălgarsko naselenie v severoiztočna Bălgarija (Sofija, 1902).
Moór, E., "Zur Geschichte südslavischer Völkerschaften im Karpatenbecken", SS 8 (1962), 267-312.
Naselja i poreklo stanovništva. 1-40, [mostly] (Belgrade, 1902-1967).
Niederle, L., Slovanské starožitnosti II/1-2, Původ a počátky Slovanů jižních (Praha, 1906, 1910).
Nlkolić, B., "Sremski govor", SDZ 14 (1964), 201-412.
Pavičić, S., Podrijetlo hrvatskih i srpskih naselja i govora u Slavoniji (Zagreb, 1953).
Pavlović, M., Govor Sretetke Župe (= SDZ 8) (Beograd, 1938).
____, "O stanovništvu i govoru Jajca i okoline", SDZ 3 (1927), 97-112.
____, "Perspektive i zone balkanističkih jezičkih procesa", JF 22 (1957-1958), 207-239.
Popović, D. J., Srbi u Vojvodini I-III. (Novi Sad 1957-1963).
Popović, D. J. and Ž. Sečanski, Građa za istoriju naselja u Vojvodini (Novi Sad, 1936).
Popović, I. Geschichte der serbokroatischen Sprache (Wiesbaden, 1960).
84
Ramovš, F., Historična gramatika slovenskega jezika. VII. Dijalekt . (Ljubljana 1935).
____, Kratka zgodovina slovenskega jezika (Ljubljana, 1936).
____, "Une isoglosse čakavo-kaïkavienne", RES 3 (1923), 48-58.
Rešetar, M., Der štokavische Dialekt (Wien, 1907).
____, "Die čakavština und deren einstige und jetzige Grenzen", AfslPh 13 (1890-1891), 93-109, 161-199, 361-388.
Roesler, R., Romänische Studien (Leipzig, 1871).
Rosetti, A., Istoria limbii române4. I-III, Bucureşti 1964.
Sandfeld, K., Linguistique balkanique, problèmes et résultats (Paris, 1930).
Seliščev, A. M., Dialektologičeskoe značenie makedonskoj toponimii. Sbornik Miletič, (Soflja 1933), 29-46.
____, Polog i ego bolgarskoe naselenie (Sofija, 1929).
____, Slavjanskoe naselenie Albanii (Sofija, 1931).
Skok, P., Dolazak Slovena na Mediteran (Split, 1934).
Sławski, F., Zarys dialektologii południowosłowianskiej (Warszawa, 1962).
Stanislav, J., Slovensky juh v stredoveku I-II. (Turčianský Sv. Martin 1948).
Stojkov, S., Bălgarska dialektologija2 (Sofija, 1968).
Tapkova-Zaimova, V., Našestvija i etničeski promeni na Balkanite (Sofija, 1966).
Todorov, C, "Govorni krăstosvanija v krajnata severozapadna bălgarska oblast", Bălgarski pregled 1 (1929), 230-243.
____, "Naselenieto meždu Timok, Iskăr i Stara-planina", Spisanie na BAN 60 (1937), 237-287.
____, Severozapadnite bălgarski govori (= SNU 41), (Sofija, 1936).
Tomić, J., Naselje u mletačkoj Dalmaciji 1409-1797. Prvi deo, 1409-1645. Niš 1915.
Vasmer, M., Die Slaven in Griechenland (Berlin, 1941).
Vidoeski, B., "Kičevskiot govor", MJ VI1I/1 (1957), 31-90.
____, "Makedonskite dialekti vo svetlinata na lingvističeskata geografija", MJ XIII-XIV/1-2 (1961-1962), 87-107.
Vizantijski izvori za istoriju naroda Jugoslavije I-III (Beograd, 1955-1966).
Wijk, N. van, Les langues slaves. De l'unité à la pluralité2. (‘s-Gravenhage, 1956).
____, "Taalkundige en historische gegevens betreffende de oudste betrekkingen tussen Serven en Bulgaren", Mededelingen der Koninklijke akademie van wetenschappen, afdeling letterkunde, deel 55, série A (1923), 55-76.
Zaimov, J., "Die bulgarischen Ortsnamen auf -išt aus -itj und ihre Bedeutung für die Siedlungsgeschichte der Bulgaren in den Balkanländern", Linguistique balkanique IX/2 (1965), 5-80.
b) Works on Dialects Transplanted by Migrations
Bernštejn, S. B., and E. V. Češko, "Opyt klassifikacii bolgarskix govorov SSSR", Učenye zapiski Instituta slavjanovedenija IV (1951), 327-343.
Bernštejn, S. B., et al., Atlas bolgarskix govorov v SSSR (Moskva, 1958).
Bidwell, Ch., "Neke beleške o bugarskom narečju banatskog sela Belo Blato", ZFL 4-5 (1961-1962), 29-33.
Bošković, R., "Refleksi grupa TJ, DJ, TbJ, DbJ, STJ, ZDJ, SKJ, ZGJ (SK’, ZG’) u dijalektima južne i jugozapadne Istre", JF 27/1-2 (1966-1967), 85-142.
Brabec, I., "Govor podunavskih Hrvata u Austriji", HDZ 2 (1966), 29-119.
____, "Vlahijski govor", Ljetopis JAZU 67 (1963), 277-286.
Bulaxovskij, L. A., "Udarenie starokrymskogo bolgarskogo govora", Sbornik .A Teodorov-Balan, (Sofija 1955), 131-143.
Bunina, I. K., "Iz istorii vajsalskix govorov", Učenye zapiski Instituta slavjanovedenija 2 (1950), 242-259.
85
Deržavin, N. S., Bolgarskie kolonii v Rossii. T. II, Jazyk. (Petrograd 1915).
Erdeljanović, J., O poreklu Bunjevaca (Beograd, 1930).
Filipović, M., Galipoljski Srbi (Beograd, 1946).
Gucu, O., "Bolgarskij govor sela Čoplja", Revue roumaine de linguistique 10 (1965), 527-536; 11 (1966), 491-502.
Hadrovics, L., "Adverbien als Verbalpräfixe in der Schriftsprache der burgenländischen Kroaten", SS 4 (1958), 211-249.
Hraste, M., "Govori jugozapadne Istre", HDZ 2 (1966), 5-28.
____, "Nepoznate slavenske kolonije na obalama Gargana", Kolo Matice hrvatske 1963, 612-617.
____, "O štokavskim govorima na Hvaru i Braču", ZRFFZ 1 (1951), 379-395.
Ivić, P., "Jedna dosad nepoznata grupa štokavskih govora: govori s nezamenjenim jatom", Godišnjak Filozofskog fakulteta u Novom Sadu 1 (1956), 146-160.
____, O govoru galipoljskih Srba (= SDZ 12), (Beograd, 1957).
____, "O srpskom govoru u selu Lovri", SS 12 (1966), 191-201.
____, "Prilog rekonstrukciji predraigracione dijalekatske slike srpskohrvatske jezičke oblasti", ZFL 4-5 (1961-1962), 117-130.
Jagić, A., "Hrvatske naseobine u Banatu", Letopis Matice srpske 319 (1929), 33-39.
Jovanović. V., "Gavrilo Stefanović Venclović", SDZ 2 (1911), 105-306.
Kašić, J., "O jekavskom govoru Velikog Grdevca, sela jugoistočno od Bjelovara", ZFL 6 (1963), 149-158.
Kurelac, F., Jačke ili narodne pešme prostoga i neprostoga puka hrvatskoga po župah šopronskoj, mošonjskoj i železnoj na Ugrih (Zagreb, 1871).
Majnarić, M., "Jedno rovtarsko narječje u Gorskom Kotaru", JF 17 (1938-1939), 135-150.
Małecki, M., Przegląd słowiahskich gwar Istrji (= Prace Komisji Językowej PAU 17), (Krakow, 1930).
Miklosich, F., "Die Sprache der Bulgaren in Siebenbürgen", Denkschriften der Kais. Ak. der Wiss., Philosophisch-historische Klasse 7 (1856), 105-146.
Milčetić, I., Hrvatske naseobine u Moravskoj, Donjoj Austriji i zapadnoj Ugarskoj (Zagreb, 1898).
Miletič, L., "Knižninata i ezik na banatskite bălgari", SNU 16-7 (1900), 339-482.
____, "Našite pavlikjani", SNU 19 (1903), 1-369.
____, "Sedmogradskite bălgari i téhnijat ezik", (= Spisanie na BAN 26). (Sofija, 1926).
____, "Ueber die Sprache und die Herkunft der sog. Krašovaner in Süd-Ungarn", AfslPh 25 (1903), 161-181.
____, "Zaselenieto na katoliškite bălgari v Sedmigradsko i Banat", SNU 14 (1897), 284-543.
Muzyčenko, A. T., Istorija poselenija i fonetičeskie osobenosti govora krymskix bolgar (S. Peterburg, 1907).
Newcklowsky, G., "Zur kroatischen Mundart von Weingraben in Burgenland", Wiener slawistisches Jahrbuch 14 (1967-1968), 94-127.
Pavlović, M., "Govor Torbeša u okolini Skoplja u evolutivnoj perspektivi", Zbornik u cast A. Belića, (Beograd 1937), 439-449.
Peco, A., i B. Milanović, "Resavski govor", SDZ 17 (1968), 245-366.
Petrovici, F., Graiul Caraşovenilor (Bucureşti, 1935).
Petrovič, E, i E. Vrabie,"Bălgarskijat govor v s. Popešt-Leorden (Bukureštska oblast)", Bălgarski ezik 13 (1963), 110-122; 15 (1965), 110-126.
Popović, I., "O bačkim bunjevačkim govorima", Zbornik Matice srpske za knjiievnost I jezik I (1954). 123-146.
Popović, M., Žumberački dijalekat (Zagreb, 1938).
Reichenkron, G., "Serbokroatisches aus Süditalien", Zeitschrift für slavische Philologie 12 (1934), 325-339.
86
Rešetar, M., Die serbokroatischen Kolonien Süditaliens (Wien, 1911).
Ribarić, J., "Razmještaj južnoslovenskih dijalekata na poluotoku Istri", SDZ 9 (1940), 1-207.
Rohlfs, G., "Ignote colonie slave sulle coste del Gargano", Cercetări de lingvistică 3 (1958), Supliment (Mélanges Petrovici), 409-413.
Skok, P., "Mundartliches aus Žumberak (Sichelburg)", AfslPh 32 (1911), 363-383; 33 (1912), 338-375.
____, "Novi prilozi proučavanju govora žumberačkih čakavaca (prvi dio)", HDZ 1 (1956), 215-278.
Stat'ji i materialy po bolgarskoj dialektologii SSSR, 2-10. Institut slavjanovedenija AN SSSR, Moskva 1952-1962.
Stojkov, S., Banatskijat govor (Sofija, 1967).
____, Leksikata na banatskija govor (Sofija, 1968).
____, "Novi proučvanija na bălgarskite govori v Săvetskija săjuz", Izvestija na Instituta za bălgarski ezik 4 (1955), 430-452.
Šiškov, St. N., "Beležki po govora na s. Pišmank'oj, Malgarskata kaza, Odrinsko", Rodopski napredăk 7 (1910), 193-206, 241-250.
Tomljenović, G. B., Bunjevački dijalekat zaleda senjskoga s osobitim obzirom na naglas (Zagreb, 1911).
Ujević, M., Gradišćanski Hrvati (Zagreb, 1934).
Vasilev, Chr., "Die heutige čakavische Schriftsprache der Burgenland-Kroaten", Frankfurter Abhandlungen zur Slavistik 8 (1966), 189-237.
Vážný, V., "Čakavské nářečí v slovenském Podunají", Sbornik Filosofické fakulty University Komenského v Bratislavé V/47 (1927), 121-336.
____, "O chorvatském kajkavském nářečí Horvatského Gróbu", in: A. Václavík, Podunajská dedina v Československu (Bratislava, 1925), 111-176.
Vuković, J., "Govorne osobine Imljana. Glasnik zemaljskog muzeja u Sarajevu", Etnologija 17 (1962), 27-49.
Abbreviations
AfslPh — Archiv für slavische Philologie (Berlin)
BE — Bălgarski ezik (Sofija)
HDZ — Hrvatski dijalektološki zbornik (Zagreb)
JF — Južnoslovenski filolog (Beograd)
MJ — Makedonski jazik (Skopje)
RES — Revue des études slaves (Paris)
SDZ — Srpski dijalektološki zbornik (Beograd)
SNU — Sbornik za narodni umotvorenija (Sofija)
SS — Studio Slavica (Budapest)
ZFL — Zbornik Matice srpske za filologiju i lingvistiku (Novi Sad)
ZRFFZ — Zbornik radova Filozofskog fakulteta u Zagrebu (Zagreb)