BULGARIAN DIALECTS

Bulgarian dialects (Bulgarian: български диалекти, balgarski dialekti, also български говори, balgarski govori or български наречия, balgarski narechiya) are the regional spoken varieties of the Bulgarian language, a South Slavic language. Bulgarian dialectology dates to the 1830s and the pioneering work of Neofit Rilski, Bolgarska gramatika (published 1835 in Kragujevac). Other notable researchers in this field include Marin Drinov, Konstantin Josef Jireček, Lyubomir Miletich, Aleksandar Teodorov-Balan, Stoyko Stoykov.

Bulgarian dialects are part of the South Slavic dialect continuum, linked with Serbian and Macedonian to the west and bordering Albanian, Greek and Turkish to the south, and Romanian to the north. Bulgarian linguistics and dialectology[1][2] and many foreign researchers[3][4][5][6][7][8] consider all Slavic dialects spoken in the geographical regions of Macedonia (region) and Thrace to be dialects of Bulgarian.[9][10][11] Macedonian authors tend to treat all dialects spoken in the geographical region of Macedonia as Macedonian, including those spoken in the westernmost part of Bulgaria (so-called Pirin Macedonia)[12].

The main isogloss separating the Bulgarian dialects into Eastern and Western is the Yat border, marking the different mutations of the Old Bulgarian yat form (ѣ), pronounced as either /ʲa/ or /ɛ/ to the east (byal, but plural beli, "white") and strictly as /ɛ/ to the west of it (bel, plural beli). In order to avoid political issues, many linguists use interchangeably Western Bulgarian and Macedonian in national and geographical contexts, respectively; however, this is not precise because Western Bulgarian dialects include also non-Macedonian dialects while some dialects in the region of Macedonia (Drama-Ser, Solun, and Korca dialects) are classified as Eastern Bulgarian on the basis of the Yat vowel.[1][13]

Dialect area

Bulgarian language area is located in the Eastern part of the Balkan Peninsula. To the north, it borders Romanian language, from the Roman language family. The language border goes along Danube from the Timok Estuary to the town of Silistra, then it crosses Dobrudja and ends at the Black Sea coast. In the past, a numerous Bulgarian population lived in Romanian (Northern) Dobrudja but in 1941 according to an agreement between the Bulgarian and Romanian governments, these people were moved to the Bulgarian (Southern) Dobrudja in the place of re-settled Romanian population. Therefore, the northern border of Bulgarian language is clearly delineated as it separates two different languages: Bulgarian and Romanian. The eastern border is the Black Sea. The southern border of Bulgarian is not well determined because the number and location of Bulgarians in Greece and Turkey is not known.[46]

To the west, Bulgarian borders Serbian language. This border, however, is not clearly defined. Bulgarian and Serbian are very similar Slavic languages, and Bulgarians and Serbs have a lot in common in their languages and customs. Because of the specific historical circumstances on the lands around the Bulgarian-Serbian border, the population there lived for many centuries in a single economical, political, and cultural community. Thus, on a dialect basis, the languages are not easily distinguished. Until 15th century, these lands were alternately under Bulgarian and Serbian rule, and then for 5 centuries they were ruled by the Ottomans. The state border was established only in 1878; until then the Serb-Ottoman border went much further to the west.[47]

The dialects along the western Bulgarian border, so-called 'transitive dialects', became an object of the Greater Bulgarian and Greater Serbian jingoism. Bulgarian and Serbian politicians tried through dialectology to prove that the dialects in the border area are pure Bulgarian or pure Serbian. Bulgarian linguists drew the border of Bulgarian language far to the west — from the Timok Estuary through Zaecar, Bolevac, Stalac, Pristina to Prizren. Serbian linguists placed the eastern border of Serbian language at Iskar River or even at the Yat border.[27]

In fact, not only along the Bulgarian western border but everywhere, especially among Slavs, in the border areas between close languages there are always transitional dialects and the change from one language to another is very gradual. The transitional dialects can be explained with the instability of political borders between the peoples during their national formation. The population in the border area usually had been ruled alternately by one or another Middle Age state or Empire, and had lived together with close economical, cultural, and political ties. Such transition is seen for Czech and Polish dialects, Polish and Byelo-Russian, Russian and Ukrainian, etc.

Macedonian dialects which possess all the characteristics of the Bulgarian language system and are very similar in grammar and vocabulary had been described as Bulgarian dialects in the large majority of publications before WWII. The similarity of Bulgarian and Macedonian dialects is a result of their common origin and identical development for more than 12 centuries in the Bulgarian national and cultural area.[48] Bulgarian and Macedonian are part of a language continuum which is different from the Serbo-Croatian language continuum. After the codification of Standard Macedonian language in the Republic of Macedonia on the basis of two southwestern Bulgarian dialects (Prilep-Mariovo dialect and Bitola dialect) in 1944-45, the majority of linguists recognised the new standard as a separate language,[citation needed] although Bulgarian (including some members of the codification committee) and many non-Bulgarian linguists do not accept the codification, describing it as a political decision without a solid linguistic basis.[2][49][50][51][52][53][54][55][56][57][58][59][60][61]

Classification

Bulgarian language developed in historical circumstances that contributed to its dialect segmentation and crossover. Therefore, today it is among the the most dialectically segmented Slavic languages. Modern Bulgarian dialects carry remnants from old tribal divisions of the Bulgarian ethnos during its historical development from the First Bulgarian State until the end of the Ottoman rule.

When classifying Bulgarian dialects, the Bulgarian dialectology lays stress on two kinds of traits: traits that distinguish individual dialects, and, on the other hand, traits that are common to two or more dialects and unite them in larger dialect groups. Distinguishing individual dialects in the present state of Bulgarian dialectology is a very difficult, almost impossible, task. Bulgarian dialects are not systematically studied by the methods of linguistic geography to show the territorial distribution of linguistic phenomena. Furthermore, the specific historic fate of Bulgarians resulted in a complicated dialect segmentation of the Bulgarian dialect area which spans at present several countries. [63][64][65][66][67][68][69][70][71][72][73][74][75][76][77][78][79][80][81][82]

Classification by the Yat mutation

The oldest and most widely accepted Bulgarian dialect isogloss is the Yat border. It was established more than 150 years ago and has received strong support throughout. According to the Yat classification (the mutation of the Old Bulgarian vowel ѣ (Yat)), Bulgarian dialects are divided in 2 large groups: Yakavian (Eastern) and Ekavian (Western). Ekavian dialects pronounce ѣ solely as /ɛ/ (/bɛl/, /bɛli/) while Yakavian pronounce it /ʲa/ or /e/ (/bʲal/, /beli/) or solely /ʲa/ (/bʲal/, /bʲali/).

The Yat border was first defined by Hristodul K. Sichan Nikolov[99] and the Russian Slavist Viktor Grigorovich[100]. It was next described by Petko Slaveykov and Konstantin Jireček and accurately traced by towns and villages by Prof. Benyo Tsonev and Prof. Lyubomir Miletich.[101][102][103][104][105]

The Yat border (see map opposite) begins from the Vit Estuary and goes to the south in the direction Pirdop — Panagyurishte — Razlog — Gotse Delchev (Nevrokop) — Solun (Thessaloniki), making a sharp turn around Pazardzhik. The border goes through the regions of Nikopol, Pleven, Lukovit, Lovech, Teteven, Pirdop, Panagyurishte, Ihtiman, Peshtera, Chepino, Razlog, Gotse Delchev, Melnik, Petrich, Demir Hissar, Kukush (Kilkis), Ser, and Solun. Thus, it divides in two regions the Bulgarian (including Macedonian) dialects.[101]

In the past, Bulgarian dialectology distinguished two groups of Yakavian dialects, also on the basis of the Yat mutation: North-Eastern dialects pronouncing Yat only as я /ʲa/ (/bʲal/, /bʲali/) and South-Western dialects pronouncing it only as /ɛ/ (/bɛl/, /bɛli/). The border between North-Eastern and South-Western dialects started at Pazardzhik and went along the right bank of Maritsa River, or, more accurately, along the northern slopes of the Rhodopes until the village of Skobelevo (Parvomay Region) where it crossed Maritsa and with small turns went to Burgas. This border was also traced by villages by Tsonev. [101]

Therefore, Bulgarian dialects were divided in 3 groups on the basis of Yat mutation:

  1. Western dialects (mutational Yat) which instead of Yat use only e /ɛ/ (/ɡɔlˈɛm/, /ɡɔlˈɛmi/, /ˡlɛtɔ/, /ˡlɛtɛn/, /ˈmlɛkɔ/, /mlɛkˈar/). Such were the dialects on the Danubian Plain west of the Yat border, plains to the north and south of the Balkan Mountain: Botevgrad, Sofia, Ihtiman, Samokov, Radomir, Dupnitsa, Kyustendil, the Western Borderlands, dialects in Central and Western Macedonia and transitional dialects.
  2. Northeastern dialects (semi-mutational Yat) which instead of Yat use я /ʲa/ or e depending on stress and on the next syllable (/ɡɔlˈʲam/, /ɡɔlˈemi/, /ˡlʲatɔ/, /ˡleten/, /ˈmlʲakɔ/, /mlekˈar/). Such were the dialects on the Danubian Plain east of the Yat border, all Balkan and Middle Mountains Region and to the south in the Thracian Plain to the slopes of the Rhodopes.
  3. Southeastern dialects (non-mutational Yat) which instead of Yat use either я /ʲa/ (/ɡɔlˈʲam/, /ɡɔlˈʲami/, /ˡlʲatɔ/, /ˡlʲaten/) or e /ʲe/(/ɡɔlˈʲem/, /ɡɔlˈʲemi/, /ˡlʲetɔ/, /ˡlʲeten/). Such were the dialects in the southern part of Burgas, Elhovo, Topolovgrad, Harmanli, Dimitrovgrad, Haskovo, south of Plovdiv, Asenovgrad, Peshtera, Smolyan, Devin, Chepino, etc.

By 1964, however, after the compilation of the first volume of the Bulgarian Dialect Atlas which encompasses dialects in the South-Eastern Bulgaria to the east of the 25th meridian and south of the Balkan Mountain ridge, it was found that a second Yat border Pazardzhik-Burgas didn't exist.[106] There is a relatively small group of non-mutational Yat dialects (/bʲal/−/bʲali/, /mlʲaku/−/mlʲatʃen/) along Maritsa River while the rest of the southeastern dialect area consists of semi-mutational Yat dialects (/bʲal/−/bʲeli/, /mlʲaku/−/mlʲetʃen/) to the east and non-mutational Yat dialects (/bʲel/−/bʲeli/, /mlʲeku/−/mlʲetʃen/) to the west. It is unclear if the present state of the southeastern Bulgarian dialects is due to changes that happened after 1903 — the time when Prof. Tsonev carried out his field studies — or this is an earlier situation which he did not detect. It is more likely that in his trip, Tsonev determined only partially the northern border of non-mutational Yat dialects (/bʲal/−/bʲali/, /mlʲaku/−/mlʲatʃen/) and incorrectly stated that south of this border, all such non-mutational Yat dialects were located. For this reason, the idea of a second Yat border as a distinguishing isogloss was abandoned and the earlier idea of a single Yat border dividing all Bulgarian dialects in Yakavian (eastern) and Ekavian (western) held sway.[105][107][108][106][109][110]

The Yakavian-Ekavian classification of Bulgarian dialects is clear-cut but it has serious flaws. First of all, there are no other isoglosses that coincide with the Yat border. Usually, at both sides of the Yat border there are completely identical dialects which differ only by the Yat pronunciation. Furthermore, a single linguistic trait, whether phonetic, morphological, or lexical, is not sufficient to characterize a dialect or dialect group.

More recent studies, however, showed that the Yat border is not so singular and isolated as thought before. Indeed, there are no other isoglosses that completely coincide with it. However, close to the Yat border, especially in its part to the north of Pazardzik, there are several other isoglosses running in parallel, mainly to the west of the Yat border. Such are some phonetic, stress, morphological and lexical isoglosses like the mutation /ˡʲa/-/ˡɛ/ (/polˈʲani/—/polˈɛni/, /pijˈani/—/piˈɛni/), stress in some verb forms (/tʃˈɛta/—/tʃɛtˈɤ/, /bˈɛri/—/bɛrˈi/, /bˈɛrete/—/berˈɛte/), the suffix for verb conjugation in first person plural present tense (/berˈɛme/—/berˈɛm/, /nˈɔsime/—/nˈɔsim/), some words (/ʲa/—/as/, /ɔn/—/tɔj/, /ʒeʒɔk/—/ɡoreʃt/, /krap/—/kɤs/, /razbɔj/—/stan/), etc.[111][106]

For this reason the Yat division of Bulgarian dialects is actual and important at present. One must have in mind that the Yat pronunciation is very characteristic for Bulgarian dialects — it can readily identify colonists from individual regions of the Bulgarian linguistic territory. Furthermore, there is no other linguistic trait that groups so clearly and regularly the Bulgarian dialects. This is evident by the attempts of Prof. Tsonev to classify Bulgarian dialects by other traits like the Big Yus (Big Nasal) substitutes, diphtong /ʃt/-/ʒd/ mutations, Yer substitutes, stress, etc.

Classification by Big Yus (Big Nasal) substitutes

The Big Yus (Big Nasal) ѫ in Old Bulgarian corresponded to the nasal vowel /*oⁿ/. In standard Bulgarian it existed until the spelling reform of 1945 when it was replaced with ъ as it had long before lost its original phonemic equivalent in most dialects. There is a variety of Big Yus mutations and substitutions (see map opposite) in Bulgarian dialects (shown below on the example of Old Bulgarian рѫкa /r*oⁿk'a/ hand, зѫбъ /ˡz*oⁿb/ tooth, зѫби /z*oⁿbˈi/ teeth):

  1. ъ-dialects (ръкà /rɘk'a/, зъп /ˡzɤp/, зъби /zɘbˈi/). These are the majority of Bulgarian dialects and the standard Bulgarian. These dialects are located mostly in the North-Eastern and North-Western Bulgaria, the eastern part of South-Eastern Bulgaria, large part of Macedonia (Drama, Ser, Kukush, Doiran, Gevgelia, northern Kostur, Dolna Prespa, southeastern Lerin, Ohrid, Resen, Struga, Gostivar, north of Debar), Gora[113], Thrace (Aegean Thrace, European part of Turkey), and small areas in Asia Minor.
  2. а-dialects (ракà /rаk'a/, зап /ˡzap/, заби /zˈabi/). These dialects are also widespread. They are located mostly in Western Bulgaria — Vratsa, Botevgrad, Eastern Sofia Region, Pernik, Radomir, Kyustendil, Ihtiman, Samokov, Dupnitsa, large part of Vardar Macedonia (Veles, Kichevo, Bitola). In this group are included also some Yakavian (Eastern) dialects: Pirdop, Koprivshtitsa, Klisura, and some parts of Rhodopes. Prof. Tsonev calls these dialects central dialects because with consideration of Macedonian dialects geographically they take a central position.[114]
  3. у-dialects (рука /rˈuka/, зуб /ˡzub/, зубе /zˈubɛ/). These dialects are the transitional dialects located around the Bulgarian-Serbian border (Belogradchik, Tsaribrod, Tran, Breznik, Bosilegrad) and northern Macedonia (north of Kumanovo, Kratovo, Skopje).
  4. о, ô-dialects (рòка /rˈɔka/, зоп /ˡzɔp/; рồка /rˈɐka/, зôп /ˡzɐp/). These dialects have isolated character and are found in the Central and Eastern Rhodopes and around Debar in Macedonia (ô-dialects also in Western Rhodopes).
  5. ê-dialects (рềка /rˈæka/, зêп /ˡzæp/). These are found only in Teteven Region, two villages in eastern Bulgaria (Kozichino (Erkech), Pomorie Municipality and Gulitsa, Varna Municipality), three villages in the southwestern corner of Vardar Macedonia around Struga (Radozhda, Vevchani and Mali Vlay), and the village Lin in Albania.

The above dialects have lost the old nasalism of ѫ. The next 3 groups have preserved the nasalism in a slightly modified form. Modifications include increased articulation of the nasal vowel to the extent of separating the nasalism in a consonant − н /ŋ/ in front of voiceless consonants or м /ɱ/ in front of voiced consonants − and dialect-specific substitution of the nasal vowel. This trait defines the nasal dialects as very ancient.

  1. ън/ъм-dialects (рънка /rɘŋkˈa/, зъмб /ˡzɤɱb/). These dialects existed in the area to the north of Solun (Thessaloniki), the northern Kostur Region, and most of Dolna Prespa.
  2. ан/ам-dialects (ранка /rɑŋkˈa/, замб /ˡzɑɱb/). These dialects existed in an area around Korca.
  3. ôн/ôм-dialects (рôнка /rɐŋkˈa/, зôмб /ˡzɐɱb/). These dialects existed in the Kostenariya and Nestram in the southernmost Kostur Region . Probably, ôн/ôм-dialects are the oldest because in them ѫ is closest to its original pronunciation.

A serious flaw of this classification is that it, too, is isolated with no other confirming isogloss. Moreover, unlike the Yat border, it does not divide Bulgarian dialects in a regular manner.[115]

Classification by *tj, *dj mutations

By the *tj, *dj mutations, there are 5 dialect groups: [116]

  1. шт/жд-dialects (нош(т) /nɔʃ(t)/, леща /leʃta/, среща /sreʃta/; прежда /preʒda/, вежди /veʒdi/, межда /meʒda/) take the major part of the Bulgarian dialect area being exclusive in all Eastern and Northwestern Bulgaria, prevalent in Samokov, Sofia, Ihtiman, Ohrid, Struga, pocket close to Lerin (Banitsa, Pətele, Ekshi Su, Zeleniche, Prekopana), Dolna Prespa, Solun,[117] Drama, Ser, Dupnitsa, Kyustendil, Petrich and mixed with other *tj, *dj mutations in the rest of the dialect area.
  2. ч/дж-dialects (ноч /nɔtʃ/, лeча /lɛtʃa/, среча /sretʃa/; преджа /prɛdʒa/, веджи /vɛdʒi/, меджа /medʒa/) in which the palatal affricates ћ /cç/ and ђ /ɟj/ that were borrowed from the Serbian language since the 16-17th century were hardened (de-palatinized) to become ч /tʃ/ and дж /dʒ/. These are the transitional dialects around the Bulgarian-Serbian border in the regions of Belogradchik, Tsaribrod, Tran, Breznik, Bosilegrad,[118] and some regions in Eastern Serbia (Pirot, Surdulica), Kosovo (Gora), and Northern Macedonia (north of Skopje and Kumanovo).
  3. к/г(ќ/ѓ)-dialects (нокь(ноќ) /nɔc/, лекя(леќа) /leca/, срекя(среќа) /sreca/; прегя(преѓа) /preɟа/, вегьи(веѓи) /veɟi/, мегя(меѓа) /meɟа/) are spoken to the west and north-west of Kyustendil (Kyustendil Kraishte and Kamenitsa), northern Macedonia (Gorni Polog, Skopska Crna Gora, Kumanovo, Ovče Pole, Kratovo, Kriva Palanka), Tetovo, Veles, Prilep, Mariovo, Bitola and north of it; and partly (mixed with other шт/жд mutations) in the regions of Debar, Prespa, Kastoria (Kostur) (very rare), Doyran, Florina (Lerin), Kilkis (Kukush), Edessa (Voden), Pianec. The plosive consonants /c/ and /ɟ/ are pronounced usually without frication with different degree of plosion (dorsal to frontal) and palatization. A variation of this mutation, шт → йкь(jќ) /ʲc/ and/or жд → йгь(jѓ) /ʲɟ/, occurs in some villages around Kukush, Voden, and Lerin. The area of mutation of the future tense forming particle (ще /ʃte/ → ке /kɛ/, ќе /ce/) is much wider and includes also southern Sofia region, Ihtiman, Samokov, Pazardzhik, and some Eastern Rup dialects (Thrace and Strandzha dialects, e.g., Elhovo, Svilengrad, etc.).
  4. шч/ждж-dialects (ношч /nɔʃtʃ/, лешча /leʃtʃa/, срешча /sreʃtʃa/; прежджа /preʒdʒа/, вежджи /veʒdʒi/, межджа /meʒdʒа/) are found in a relatively pure state in Korca, Kostur, Doyran, and Ohrid town; mixed with ќ/ѓ and/or шт/жд mutations in Maleševo, south of Bitola, Debar, Struga, Dolna Prespa, Lerin; mixed with ч/дж mutations in Samokov (Shishmanovo) and Breznik. Variants of the жд → ждж mutation occur in Kostur (жд → ж /ʒ/) and Doyran (жд → йдж /ʲdʒ/).
  5. шќ/жѓ-dialects (ношќ /nɔʃc/, лешќа /leʃca/, срешќа /sreʃca/; прежѓа /preʒɟа/, вежѓи /veʒɟi/, межѓа /meʒɟа/) are limited to very small area south of Bitola (also variant шьк /ʃʲk/) and Eastern Rup around Strandzha (variant жьгь /ʒʲɟ/ as in вежьгьи /veʒʲɟi/).

As repeatedly mentioned, *tj, *dj reflexes are an important phonological trait, characteristic for each individual Slavic language, and used for language differentiation. Bulgarian language is characterized with the шт(щ)/жд mutation on which the Standard Bulgarian is based. The presence of numerous *tj, *dj variants on the Bulgarian linguistic territory indicates outside linguistic influences and/or spontaneous modifications due to linguistic isolation of peripheral areas. Alternatively, this variability may be (and has been) explained with coexistence of fundamentally different dialect systems. This latter case would be indicated by an approximately simultaneous separation of two or more reflexes from the primitive Proto-Slavic *tj, *dj.

This issue became actual and received an added importance, with political overtones, in connection with the Bulgarian vs. Macedonian controversy. Historically, as in other border regions, this started as Bulgarian vs. Serbian controversy. Stojan Novaković, Serbian politician and diplomat, charged by the Serbian government with the mission to organize the assimilation campaign in Macedonia in accordance with the Serbian national (Garašanin) doctrine, propagated the idea that Slavs in Macedonia were Serbs because they spoke a dialect of Serbian language with the typical Serbian ћ,ҕ reflex of Proto-Slavic *tj, *dj. Another Serbian linguist, Aleksandar Belić, was more careful in making some distinction between Serbian affricates ћ,ҕ and Macedonian plosives ќ,ѓ; however, he considered the latter as a very close variant which is evident by the symbols that he used for those − ћК and ҕГ. According to Belić, Macedonian dialects were divided into Southern (Solun, Kostur, Korca to Bitola, Resen, Ohrid and Debar) and Northern ("true Macedonian dialect", Tetovo and Štip). In the Southern dialects *tj reflects in шт(шч), and *dj reflects in жд(ждж), while in the Northern the reflexes are ћ(ћК) and ҕ(ҕГ). [119] Bulgarian linguists, on the other side, denied or belittled the existence of *tj, *dj reflexes other than шт/жд, without factual support.

The most detailed and comprehensive study of the *tj, *dj reflexes in Macedonia and Albania was carried out by the Russian linguist Afanasy Selishchev.[120][121] He studied this phonetic trait in all aspects: not only the territorial distribution of dialect variants but also occurrence in different words, in different grammatical positions with a clear idea of all acustic and physiological specifics in relation to the neighboring Slav languages using data not only from written documents but also from toponymy. Selishchev stressed the fact that unlike the Serbian ћ and ҕ, the Macedonian ќ and ѓ are pronounced without frication. A fricative element was found very rarely to the north of Tetovo but even there, ќ and ѓ were not identical to the Serbian affricates. Selishchev also noted the reduced palatization of ќ and ѓ, which varies in different Macedonian dialects.

On the basis of his study of the territorial distribution of шт/жд, шч/ждж, and ќ/ѓ, Selishchev found that the basic reflex of the Proto-Slavic *tj, *dj in Macedonian dialects is шт/жд or its more ancient stage шч/ждж. Using a huge amount of dialectological material, he found that this pronunciation was fixed on a large part of the territory of Macedonia: Debar, Struga, Ohrid, Resen, Kostur, Lerin, Solun, Kukush, Doyran, Maleševo. The шч/ждж is archaic and gives way to шт/жд. For this conclusion, Selishchev used the studies of V. Oblak and B. Tsonev who noted that in some dialects the old people spoke шч/ждж while the young knew only шт/жд. It was notable that the archaic шч was more resistant to the newer шт than ждж to жд. Thus, in some Debar, Resen and Kostur dialects the archaic шч was found along with жд.

Palatal ќ, ѓ in a limited number of words and forms (e.g., куќа, ќе, веќе) are found also far to the east of Macedonia. Thus, the particle ќе is found in many Eastern Bulgarian dialects. Nevertheless, through deeper analysis, Selishchev came to the conclusion that ќ, ѓ are not innate to the Macedonian dialects but were calqued from the imported Serbian analogs together with the whole word. For example, the word куќа (house) is found throughout Macedonia. The process of borrowing is confirmed by the root vowel: here in the place of the dorsal nasal vowel one finds the vowel у /ʊ/ (compare bg:къща /kɤʃta/ with sr:кућа /kʊcça/). Even in Northern Macedonia, in Lešok, Selishchev heard both куќа and къща. [122]

Toponymy provided very valuable data for study of ќ and ѓ. Selishchev showed with toponyms that in the central Macedonian dialects шт, жд are more ancient than ќ and ѓ. For example, while in the vernacular from Prilep region the word for trousers is гаќи, diphthong шт is still preserved in the names of villages around Prilep − Кривогаштани /krivɔgaʃtani/ ("Crook-trousers").[120] Later, he used toponyms to solve very difficult and baffling problems from the history of Macedonian dialects. Supported by many sources, Selishchev showed that the process of replacement of шч (шт) - ждж (жд) by ќ, ѓ had been long and, by WWII, yet unfinished in many parts of Macedonia. Selishchev showed convincingly that this process goes from north to south. In the southeastern and southwestern corners, a very limited number of words were spoken with ќ, ѓ.

There was one serious flaw in this analysis. Since the time of Gilferding (1850-60) toponyms in Albania and Greece that end in -kaki were thought to originate from the Proto-Slavic *tj. If this was true, one had to agree not only that ќ is a specific Macedonian reflex of *tj but, what is more, that it existed in the language of those Slavs that in the 6-7th centuries flooded in great numbers not only in Epirus and Thessaly but also in Peloppones. Selishchev either had to explain the origin of these toponyms or disavow his views on the origin of ќ, ѓ in the modern Macedonian dialects. He solved this problem in a special chapter in his book Slavic population in Albania (1931).

The suffix -kaki interpreted as a reflex of the Proto-Slavic *ko,tja is a composition of heterogeneous elements that are never found in Macedonia. In Macedonian dialects occur either къшта (къшча) or куќа. There are no examples of кушта (кушча) or къќа (каќа). If such were found, they would contain both Bulgarian and Serbian reflexes of the Big Yus and *tj. The toponym Gardikaki, for example, is unique in this respect. Selishchev pointed out that these toponyms can end in -i but also without it. In Epirus and Albania they less often end in -u. These comparisons showed that the second element of the toponym was not related to the Proto-Slavic *ko,tja. The thorough research was successfully completed after taking into account the geologic and soil science publications on Albania and North Greece. From Albanian language, it was seen that the suffix -kuќ can be interpreted as "red"; it is often found on territory with Albanian settlements and specifically in places where the soil has reddish hue. In the region of Upper Shqumba the mountain peaks in Lurje are called Gurikuќ which means "red stone". North of Argirocastro one finds Hundokuќ or Hundёкуќ which means "red nose". In his book, Selishchev described many more examples. [121]

This comprehensive analysis of dialects and toponymic data in Macedonia and Albania showed convincingly that ќ, ѓ in place of the Proto-Slavic *tj, *dj is a late phenomenon arising under the influence of the Serbian phonetic system introduced (sometimes sporadically) by the Serbian administration, rulers, settlers, and teachers.[123] This analysis has wider significance beyond the narrow frame of classification of Slavic dialects in Macedonia and Albania. It bears direct relationship to the origin of Old Church Slavonic (Old Bulgarian) language. Before the publication of this analysis, in the Slavistic literature was current the hypothesis of N.S. Trubetskoy, N.N. Durnova and some other Slavists, according to which ќ, ѓ arose directly from the Proto-Slavic *tj, *dj and were present in the language of Cyril and Methodius. The studies of Selishchev disproved this hypothesis.[124] Most linguists accepted Selishchev's conclusions,[125] including most Bulgarian linguists.[126][127] Some of the latter (Blagoj Shklifov, V. Georgiev) disagreed on the issue of whether ќ, ѓ are internal or borrowed and in 1981-82 hypothesized that these reflexes arose as a result of specific processes, innate for the Macedonian dialects.[128][129] However, this new point of view was not supported by argumentation. [124]

Some leading linguists from Republic of Macedonia (e.g., Acad. Božidar Vidoeski) appraised the study of Selishchev and used his classification and data in their works;[130] others implicitly acknowledged Selishchev's conclusions. For example, Blaže Koneski in his A Grammar of the Macedonian Literary Language (book 1, 1952) in which he stipulated the norms of the Macedonian Standard, listed the ќ, ѓ reflex only as a second, supporting distinctive trait in Macedonian language after the ъ → o reflex; the latter is, indeed, a very early trait.

Classification by the ъ − о isogloss

The Dutch Slavist Nicolaas van Wijk proposed a classification of Bulgarian dialects by the reflex of OBg. ъ /*ɘ/ in o /ɔ/, that is, by the isogloss ъ − о: сън, дъш → сон, дош instead of OBg. сънъ, дъждь. [131]

This reflex is known for its antiquity and since the time of Old Bulgarian divided the Bulgarian language area in two main dialect groups — Eastern and Western. However, the isogloss о—ъ is not very clearly distinguished and exhibits a great variability. Thus, о instead of OBg. ъ is found in the Western dialects and Pirdop dialect but in some dialects it is only in prepositions and prefixes (воф, воздàхна, сос, собỳе), in others − in prefixes and suffixes (воф, сос, пèток, песòк), and in third − in all closed syllables (вос, сос, пèток, бòчва, дош). Only in some Macedonian dialects (Malashevo, Veles, Prilep-Mariovo, Debar, Kostur, Doyran, Kukush, Voden) vowel o completely substitutes ъ. Complete reflex ъ → о is found in some Rup dialects but there it is a result of a specific late process replacing a secondary Yer when two Yers/Nasals (Big or Small) occur in the same word: дош from дъждь (Big and Small Yers), мòгла from мьглѫ (Small Yer and Big Nasal), зоп from зѫбъ (Big Nasal and Big Yer), кльòтва from клѧтвѫ (Small and Big Nasals). In the Moesian dialects of Northern Bulgaria there is a reflex (vocalisation) of OBg. ъ in о in the article form for masculine (гърбò, нусò) and in the suffix -ък (добѝтọк, пèтọк). Therefore, this trait does not allow for a clear-cut classification of Bulgarian dialects. [132]

In order to improve this classification, Prof. Tsonev proposed to classify Bulgarian dialects by both ъ (Big Yer) and ь (Small Yer) reflexes,[133] thus dividing Bulgarian in 4 dialect groups:

This classification, too, has serious flaws in that it does not divide regularly the Bulgarian dialect area and the 4 groups do not include all variations. For instance, in the Moesian dialects of the North-Eastern group there is a reflex ъ > о, in the East Rup dialects of the Rup-Rhodopes group there is a reflex ь > е, etc. Taking into account only the reflexes of the OBg. ь, Bulgarian dialects have been classified in 3 groups: [134]

Dialects with soft reflexes on ь comprising Rup-Rhodopes with the soft reflexes е or йъ /ʲɘ/ Dialects with hard reflexes on ь represented by the transitional dialects: ь > ъ Dialects with hard and soft reflexes comprising North-Eastern, South-Western, and the Eastern half of the North-Western dialects:ь > ъ (а) and е. This classification is also not very clear-cut and does not divide the Bulgarian dialect area in approximately equal parts. Classification by morphological and lexical traits In the Bulgarian dialect area, several morphological and lexical boundaries can be drawn that approximately coincide and form a band of isoglosses. Such isoglosses are, e.g.: The suffix for plural in polysyllable masculine nouns that end in a consonant. In the North-Western Bulgaria (except part of Moesian dialects), this suffix is -и /ɪ/, in Western and South-Eastern Bulgaria it is -е /ɛ/: пръстени, ръкàви, гълъби — пръстене, ръкàве, гълъбе. The suffix for first person plural present tense for first and second conjugation verbs. In North-Eastern Bulgaria it is -м /m/, in Western and South-Eastern Bulgaria it is -ме /mɛ/: берèм, четèм, вървѝм — берèме, четèме, вървѝме. The words: аз/я, недей/немòй, крак/ногà, рѝза/кошỳля, стан/разбòй, горèщ/жèжък, къс/крап, etc. The first word of these pairs occurs in North-Eastern Bulgaria, the second — in Western and South-Eastern Bulgaria. Morphonological isoglosses in the stress of disyllabic neutral singular nouns and in the form for the imperative mood second person singular in the verbs of first and second conjugation. In North-Eastern Bulgaria the stress is on the suffix, in Western and South-Eastern Bulgaria — on the root: месò, кроснò — мèсо, крòсно; берѝ, носѝ, метѝ — бèри, нòси, мèти. Interestingly, the band of morphological and lexical isoglosses goes along the Yat borders: both the primary, accepted one (Vit estuary to Vardar delta) and the secondary, apparent one (Pazardzhik to Burgas). Thus, they define a central (middle) region comprising North-Eastern and Central Bulgaria, and a lateral (peripheral) region comprising North-Western, South-Western, and South-Eastern Bulgaria which envelops the central region.

Veroeros Etiam

1