CC BY
40
Towards a Systemization of Common Balkan Lexical Evidential Markers
Maxim Makartsev
Institute for Slavic Studies
of the Russian Academy of Sciences,
Moscow
К систематизации
общебалканских
лексических
эвиденциальных
маркеров
Максим Максимович Макарцев
Институт славяноведения РАН, Москва
Abstract
A simple listing of lexical evidential markers in Bulgarian, Macedonian and Albanian uncovers unusual problems, because a significant part of the markers are common, both due to genetic relations between the languages (e.g. Bulg. and Maced. spored) and to areal factors (e.g. Turk. word güya/güya was loaned into Bulg. dialectal gyoa, Maced. goa and Alb. gjoja; this marker also exists in Serb.). But these common markers with the same etymology do not necessarily have similar meanings, which is both a theoretical problem for the description of the language data and a practical issue for translation between the languages. As Bulgarian, Macedonian and Albanian have grammatical evidential systems as well, there is a question how the lexical evidential markers interact with evidential forms. Here the distinction between analytic and holistic reading can be quite helpful, as it clarifies the role of each of the components in constructions. In the article it is analysed on the basis of translations between the Balkan languages. The definition of evidentiality I employ in this article is the one stated by A. Aikhenvald: "evidentiality is a linguistic category whose primary meaning is [the] source of information. [...] [T] his covers the way in which the information was acquired, without necessarily relating to the degree of [the] speaker's certainty concerning
2012 №1 Slovene
28 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
the statement or whether it is true or not" [Aikhenvald 2004: 3]. It is well known that there is a certain variety of domains for expressing evidentiality; first and foremost there is a distinction between lexical and grammatical markers.1 In the following article I will concentrate on the common lexical evidential markers in Albanian, Bulgarian, and Macedonian, with a short introduction to grammatical evidentiality in these languages.
Keywords
evidentiality, lexical evidential markers, Bulgarian grammar, Macedonian grammar, Albanian grammar, Balkan linguistics, translation
The problem of lexical evidential markers, first dealt with on the basis of Bulgarian and Slavic data in Ivan Kutsarov’s works in the late 1970s [Куцаров 1978a; 1978b; 1978c; 1978d], was not analysed systematically until the middle of the 1990s.1 2 Studies of lexical evidentiality became more numerous in the late 2000s with the works of the Mainz Evidentiality Circle — see the collections [Wiemer, Plungjan 2008] and also [Wiemer, Vrdoljak 2011; Wiemer, Vrdoljak (in press)]; and [Wiemer, Stathi 2010; Kampf, Wiemer 2011]. The latter work is also a good account of up-to-date research on Bulgarian lexical markers and [Wiemer, Vrdoljak 2011; Wiemer, Vrdoljak (in press)] — on Bosnian, Croatian, Serbian, and Slovene markers. The survey [Гајдова, Лаброска 2008] is the first attempt to provide a complex study of the interactions between grammatical and lexical evidential markers (verbs of speech and conclusion) in Macedonian. In their study [Гајдова, Лаброска 2010] they give a short list of lexical evidential markers in Macedonian. [Тофоска 2008] treats only verbs of speech and [Bužarovska 2006] only one marker — kako da. [Петроска, Тофоска 2011] continues these works and compares certain lexical evidential markers in Macedonian and Polish. [Petroska 2012] gives an account of one Macedonian marker navodno and its interactions with grammatical evidential markers. For Albanian there are no published studies on lexical evidentiality.3
1 As for the grammatical evidential markers in the Balkan languages, there is a tradition of their analysis dating back to the second half of the 19th century for Albanian and Bulgarian. Comparative analysis of evidential forms in the Balkan languages conducted by Victor Friedman [Friedman 1982, 1986, 1999], Grace Fielder [Fielder 1994, 1996, 1997, 1998, 1999], and several other linguists, has answered some questions about what is common among Balkan languages and what is unique for each of them in the domain of grammaticalised evidentiality. There is a great deal of literature on grammatical evidentiality in general and it is scarcely possible to cite it all within a single article, but a recent study [Aikhenvald 2004] contains the main bibliography on the issue.
2 See [Rakhilina 1996, Ramat 1996], published in the best-known book, edited by Z. Guentcheva — [Guentcheva 1996], and also the large study by E. Padučeva — [Падучева 1996].
3 I give an overview of Albanian, along with Bulgarian and Macedonian, lexical markers in my Ph.D. thesis (see the abstract — [Макарцев 2010]).
Slovene 2012 №1
Maxim Makartsev
I 29
The data for the present study was taken from three digital collections of Bulgarian, Macedonian, and Albanian literary texts I created in 2007-2009 for my study of evidentiality in the Balkan text space (defended in 2010 as a Ph.D. at the Institute of Slavic Studies — [Макарцев 2010]). The databases are of unequal size, due to the inequalities in the existing digital libraries for these languages: the Bulgarian database is 803 750 words; Macedonian, 224 750 words; and Albanian, 398 700 words.4
General Problems
Bulgarian, Macedonian, and Albanian are members of the Balkan Sprach-bund, sharing many features at all language levels. From an areal point of view, a simple listing of lexical evidential markers in each of the three languages uncovers unusual problems, because a significant number of the markers happen to be common due to areal factors or common linguistic genealogy. But these common markers with the same etymology do not necessarily have similar meanings, which is both a theoretical problem for the description of the language data and a practical issue for translation between the languages.
Among the common features the languages under analysis share are also special forms with evidential meaning.5 Their existence poses a challenge to the researcher: while the use of lexical means for expressing evidential meanings may be considered universal, in these three languages the system of lexical markers interacts with the grammaticalised evidentiality. This results in unusual research problems, such as describing what kinds of markers (lexical vs. grammatical vs. both lexical and grammatical) are preferred in different situations. This problem, valid for every language with grammaticalised evidentiality, had not received much attention until recently in Peter Kehayov’s works (see [Kehayov 2008]).
These problems overlap, because for a proper description of a marker an inventory of possible contexts is essential. This is why the analysis of the common markers is preceded by a list of items and an overview of grammatical evidential systems in Bulgarian, Albanian, and Macedonian.
4 Bulgarian has also several corpora, among them the largest is the Български национален корпус, created at the Institute of Bulgarian Language at the Bulgarian Academy of Sciences, and sometimes I use this corpus data as well. These examples are marked ‘BNK.’
5 To be strict, we need to emphasise that, whereas Bulgarian and Macedonian have grammaticalised evidentiality (because it is obligatory in some contexts and has special forms), Albanian has grammaticalised evidential strategy (non-obligatory marking with forms of another category — admirative).
2012 №1 Slovene
30 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
The List of Common Lexical Evidential Markers in Albanian, Bulgarian, and Macedonian
By common markers we mean markers that have the same source and generally the same form in at least two of the languages. As I will show, it is quite possible that certain markers have evidential meaning in only one language, which is the case with Bulg. gyoa, Maced. goa and Alb. gjoja. In this case its counterpart may have a general epistemic meaning or may be a discourse marker.
To date, I have found only four markers which are used in at least two of the languages under analysis and which belong to neutral literary speech at least in one of the languages:
božem ‘allegedly, seemingly, under the pretence, as if’ (Bulg. and
Maced.), according to [БЕР I 1971] derived from Old Slavic bo + že with the final prothetic m under the analogy with adverbs such as ръГет, denem, noštem, hodom etc.
demek ‘it means, so to say; as they say’ (Bulg. dial. and popular,
Maced. and Alb.), loaned from Turkish demek ‘say, speak; mean’ [Boretzky 1976; БЕР I 1971].
gyoa/goa/gjoja ‘allegedly; sort of’ (Bulg. dial., Maced. dial. and colloq. and Alb.), loaned from Turkish güya/güya ‘as if; perhaps, maybe’, actually being a loan in Turkish itself, where it was taken from Persian gü ‘say, speak’.6
spored ‘according to’ (Bulg. and Maced.), according to [БЕР VII 2010] derived from *гфъ ‘line, row’ by the use of two prefixes зъ + po. Its primary idea is that of accordance of the object within the scope of spored to other similar objects in the row. All the other meanings of this preposition are secondary, that is, the evidential meaning is based on the concept that the fact conveyed by the speaker corresponds to what someone else has said.
There are several reasons for the presence of common markers across several Balkan languages: the markers can either be of the same origin in genetically closely related languages (spored and božem in Bulg. and Maced.) or they can be loans from another language from the same area (such as demek and
6 According to [БЕР I 1971], gyoa/goa/gjoja has many variants within the Balkan Slavic dialectal space, among them gjöa, gjoe (mi), gjöemi-ti (without a proper territorial specification); gjove (south-western Bulgarian dialects); gjoj, gjojkim (Smolyan region in Bulgaria), gjüem (Bansko region in Bulgaria); gjuva (Debar region in Macedonia), but little can be said about its meaning in the respective dialects. In Bulgarian gyoa is a dialectal word; it is not included in [РБЕ]. The search in BNK showed 126 uses of this item, all of them in folklore publications between 1945 and 2010. This is why I am not going to discuss it in this article.
Slovene 2012 №1
Maxim Makartsev
I 31
gyoa/goa/gjoja, which were loaned from Turkish), in which case the genetic ties do not apply.
In this second case a direct loan is not the only method of borrowing. It is worth noting that in the Balkan Slavic dialects, spored is not the only preposition with an evidential meaning; cf. porddf ‘according to’ (< po ‘as, on’ + radi ‘for’, it is possible, that both prepositions originally had the meaning of aim) in Kajnas, a Slavic dialect of Boboshtica in the Korga region in south-east Albania (my field research, 2011). I suppose that this meaning of spored, which is primarily a preposition of correspondence (compare Macedonian поради ‘because’), could have emerged through matching it part-bypart to the Albanian preposition sipas/simbas ‘according to’, which is made up of two parts: si ‘as’ + pas ‘after’.8 If this is true, it shows yet another type of correspondence between the markers of two languages: not a loan from a common source, but a structural calque.
Typological probability is also among the reasons why similar markers emerge in different languages. For example, the verbs of speech are likely to become petrified lexical evidential markers without any connection to genetic origin or areal proximity: compare Bulg. kaj (< kde (dialect.) < 3sg of kdja (Banat dialect) ‘say’)9; Maced. veli, Serb. kaže, Greek Aeel, Russian colloquial grit (< govorit) with the common etymology ‘s/he says/tells’; Alb. thote (‘id.’), tha (‘s / he said/told’); Romanian cicä (< zice cä ‘s/he says /tells, that’), Spanish10 dizque (< ‘id.’), etc.11 The Bulgarian kaj may also be an interesting example of an evidential marker having infiltrated colloquial speech from an isolated, non-Bulgarian language area, that of the Banat Bulgarian dialect surrounded by Romanian, German, Hungarian, and Serbian (if it is not just a shortened form of the verb kaža ‘say’ which may have emerged independently in colloquial speech or in other less remote dialects).
The number of languages under comparison is the only limit for the list of the markers — if we extend the list of languages, the list of evidential markers common to at least a pair of languages will also be broadened (cf. Serb. and Maced. navodno, kao (da)12 and kako da; Romanian and Bulg. maj, etc.), and the same thing will happen if we add dialectal (local and social) data.
One may add grammatical evidential markers as well. First, in both Bulgarian and Macedonian the evidential semantics can be conveyed by l-forms * 11
7 The markers are in bold at their first appearance.
8 Unfortunately, [Qabej 1976; Meyer 1891], and [Orel 1998] do not provide any etymology for sipas, perhaps presuming it to be obvious.
9 See [БЕР II 1979: 135, 144, 303].
10 In several Latin American varieties.
11 See [Aikhenvald 2004: 140-142] for evidential markers derived from verbs of speech in other languages.
12 Which also has formal equivalents in Greek: oav va [Bužarovska 2006].
2012 №1 Slovene
32 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
which have the same etymology (even though there are dialects in which those forms have only admirative meaning, for example in Boboshtica, or in Banat, where they have no modal or evidential meaning at all and stand for a general past — [Стойков 1967]). We may also consider the Aromanian Frasheriote dialect of Beala di Sus in Macedonia, whose admirative suffix -ka is a loan from Albanian [Friedman 1994]. In another Balkan dialect, Sliven Romani, which is strongly influenced by Bulgarian, the suffix -li is used in various contexts (among them renarrative), and thus it can be called “Annäherung an das Bulgarische” [Igla 2006: 59].
Of course the Balkan Sprachbund has shaped itself on the dialect level through language contacts between the dialects, so the most proper way to study lexical evidential markers is through the dialectal data, taken both from contact areas and from areas outside the direct contact with other languages. But even when only standard languages (or their non-dialectal varieties) are taken into consideration, the main tendencies and oppositions can be revealed. That is why this study can be taken only as a preliminary step before embarking on a more complex description of the situation at the dialectal level.
The English (as well as any other) translations give only a rough picture of the general meaning of these items, whereas my research aims at a more precise description of their semantics. To provide this description, I use some elements of the outline of the database of evidential markers in European languages [Wiemer, Stathi 2011]. The basic idea of the database is that a unified structure is created to describe all evidential markers of the respective languages as entries in a database. The names of the fields in the database will contain the possible features of evidential markers and the fields will include information valid for the respective markers. This database will unify the approaches to different markers in different languages, thus providing the opportunity to compare them on various parameters. This database provides a perfect tool for comparing evidential markers in several languages. The database is not launched yet, and I use only the principle elements of its structure to describe the markers.
In this article I will describe the following three markers: Bulg., Maced., and Alb. demek; Maced. and Alb. goa/gjoja; and Bulg. and Maced. spored. Other cases, primarily Bulg. and Maced. božem; Serb. and Maced. navodno, kao (da) and kako da; Bulg. and Romanian maj; Bulg. gyoa; kaj in Bulg. dialects; Kajnas poradi and Alb. sipas/simbas, and many others seem to deserve special attention.
An Overview of the Unwitnessed Evidential Forms
Here I am going to dwell briefly upon the forms I refer to as unwitnessed evidentials in the Balkan languages. In Bulgarian and Macedonian they are
Slovene 2012 №1
Maxim Makartsev
I 33
based on the so-called /-participle and are formed according to the following model: the verb “to be” + /-participle (some of the more complex verb forms such as future or pluperfect can have a slightly different structure; however, there are no unwitnessed evidentials without an /-participle). In Bulgarian the third person of the verb “to be” (the copula) is often dropped; in Macedonian it is always dropped. It is obvious that the origin of the Bulgarian and Macedonian /-forms should be sought in the Slavic perfect. In Bulgarian and Macedonian there is still a strong tie between grammatical evidentials and the perfect. Technically, in the majority of cases the perfect and the unwitnessed eviden-tials are identical in form and the meaning of the particular construction can be understood only from the context. In Macedonian, due to the obligatory drop of the copula in the perfect, there is no difference between perfect (1) and unwitnessed evidentials for aorist/imperfect (3) (in (2) an example of a witnessed evidential form is given):
Macedonian
(1) Vo.toa.vreme toj bid13 vo Skopje (Reported Imperfect)
At.that.time he be-LPARTicipLE.MAsc.SG in Skopje
‘[They say that] at that time he was in Skopje.’
(2) Vo.toa.vreme toj be-še vo Skopje (Witnessed Imperfect)
At.that.time he be-lMpF.WiTN.Ev.2-3sG in Skopje
‘At that time he was in Skopje [and I vouch for it].’
(3) Toj bi-l vo Skopje nekolku pati (Present Perfect)
He be- LParticiple-Masc.Sg in Skopje several times
‘He has been in Skopje several times.’
It is often stated in the Bulgarian linguistic tradition that the feature distinguishing perfect from grammatical evidentials is the copula in the third person (perfect retains it, while in unwitnessed evidentials it is dropped). This is insisted on by the Academic Grammar:
“The Reported Aorist forms are shaped in the same way as the forms of Present Perfect, but in the third person (Singular and Plural) they are different: in Present Perfect (4)14 the copula e/sa is restored and in the reported tense (5) it is omitted” [Gramatika 1983: 354]:
Bulgarian
(4) Toj e_ zastana-l pred vhod-a (Present Perfect)
He be.PRAES.3sg stand- LParticiple.Masc.S at entrance-DEF
‘He stood at the entrance.’
13 The illustrated verb forms in the examples are underlined, the lexical evidential markers are in bold.
14 The numbering is lacking in the original. The original examples were transliterated and grammatical notation was added (4-5).
2012 №1 Slovene
34 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
(5) Toj zastana-l pred vhod-a (Reported Aorist)
He stand-LParticiple.Masc.Sg at entrance-DEF
‘[Somebody says that] he stood at the entrance.’
Compare this with a witnessed aorist form:
(6) Toj zastana pred vhod-a (Witnessed Aorist)
He stand-Aor.Witn.Ev.2-3Sg at entrance-DEF
‘He stood at the entrance.’
However, many linguists have shown that a “vice versa” situation is also possible (see [Friedman 2001] and [Levin-Steimann 2004] for bibliography and a general discussion of the topic): there are forms with omitted copula conveying some sort of Resultative meaning (Anke Levin-Steinmann called this meaning Zustandskonstatierung, following Ljubomir Andreychin’s term констатация на състояние ‘a statement of a state’ — [Levin-Steinmann 2004]):
Bulgarian
(7) Pole-to pusto. Pat kalen i bezkraen. (Z-konstatierung)
Field-DEF empty way muddy and without.end
Nebe-to se shlupi-l-o nad zemja-ta
sky-DEF REFL hang-LParticiple-N.Sg over earth-DEF
‘The field is empty. The way is muddy and has no end. The sky hangs over the earth.’ (Elin Pelin15)
The example in (7) contradicts the viewpoint of the Bulgarian Academic Grammar: it does not refer to any source of knowledge, but the copula is omitted. At the same time, it is also possible that a form with the copula present conveys some unwitnessed information based on the words of other people, as in (8):
Bulgarian
(8) Kazva-t, če car Boris III Hitler go
They.say that tsar B. III H. Cl. Acc. Masc.Sg
e otrovi-l-0,
be.PRAES.3sg poison-LParticiple-Masc.Sg
kači-l-i sa go na edin samolet i
put.up-LParticiple-Pl be.PRAES.3pl Cl. Acc. Masc.Sg on one plane and
sa
be.PRAES.3pl
kadeto
where
Car-jat
Tsar-DEF
go
Cl. Acc. Masc.Sg
vazduh-a
air-DEF
e_
be.PRAES.3sg
vdigna-l-i
raise.-LParticiple-Pl
e
be.PRAES.3sg
mnogo visoko,
very high
razreden.
rarefied
ima-l-0
have-LParticiple-Masc.Sg
15 In the brackets the sources of the examples are shown. In the majority of cases we do not need a specific indication of the source, just the genre or, if it is taken from fiction, the author.
Slovene 2012 №1
Maxim Makartsev
I 35
slabo sarce i se e
weak heart and REFL be.PRAES.3sg
počuvstva-l-0 lošo,
feel-LParticiple-Masc.Sg bad
i togava sa mu da-l-i otrova-ta.
and then be.PRAES.3pl Cl. Dat. Masc.Sg give-LParticiple-Pl poison-DЕF
‘They say that tsar Boris III was poisoned by Hitler: they put him on an airplane and raised him to a high altitude, where the pressure was low. The tsar had a weak heart, he felt sick, and then he was given the poison.’ (from a conversation)
The drop of the copula in the third person can signal an opposition between the reportive vs. conclusive evidential meanings (9), or between disbelief/low belief vs. neutral evaluation of the reported information (10) and so on.
Bulgarian
(9) Palto-to i e na zakačalka-ta.
coat-DEF Cl.DAT.3sg.Fem be.Praes.3sg on hanger-DEF
Znač-i, tja e doš-l-a.
it.means-Praes.3sg she be.Praes.3sg come-LPart-Fem.Sg
‘The mother’s coat is on the hanger. So she must have returned home.’
(10) Saddam nesamneno ima-l-0 oraži-ja
S. undoubtedly have-LParticiple-Masc.Sg weapon-Pl
za masovo porazjavane.
for mass destruction
Okaz-a se, če ne e ima-l-0,
Turn.out-Aor.3sg Refl that Neg be.Praes.3sg have-LParticiple-Masc.Sg
ama kakvo ot tova.
but what from that
‘[A lie was the excuse to start the intervention in Iraq.] Saddam, they said, undoubtedly had weapons of mass destruction. It turned out that he didn’t, but it would have changed nothing.’ [Kehayov 2008: 175]
While what we are dealing with in (9) is still evidential (in this case it is the conclusive meaning of the perfect), in (10) we have a clear case of the epi-stemic modal meaning: the forms имал and не е имал are opposed not only as positive and negative, but also as conducting disbelief and neutral evaluation of the respective propositions.
So, it seems that dropping the copula in the third person has little or nothing to do with the opposition of witnessed and unwitnessed forms, as opposed to the point of view advocated by the Academic Grammar.
That is why in Bulgarian and Macedonian a complicated situation arises: forms shaped in the same way conduct meanings from fairly different semantic domains (such as resultative and unwitnessed evidentials, for example). I have not yet found any convincing account of this situation in
2012 №1 Slovene
36 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
the literature (these forms cannot be treated as homonymous forms of different categories, nor can they be considered to belong to a single category). The most interesting idea was proposed by Ronelle Alexander [Alexander 2001], who called the Bulgarian /-forms ‘generalized past’, which should be treated as a maximally neutral category, highly dependent on the context. Still, as it is not my aim in this article to come up with a label for these forms, I will refer to them simply as ‘/-forms’. The Bulgarian system is discussed in greater detail in [Макарцев 2008], and an inventory and overview of the meanings of other evidential forms in Macedonian can be found in [Усикова 2003].
In Albanian, unwitnessed evidentials (they are usually called “admira-tive” in the literature, due to their primary context meaning) follow a different pattern, though again evidential forms are based on the participle. The ending of the participle is removed and replaced by the finite form of the copula (the verb kam ‘have’). The tie with the perfect can still be seen, as the perfect consists of the same elements in an inverted order: copula + participle (which retains its ending in this case). Compare the following two sentences, with the ordinary perfect (11) and with the admirative/eviden-tial present (12):
Albanian
(11) Ai ka qenö nö Tiranö. (Present Perfect)
he have.3sg be.Participle in Tirana
‘He has been to Tirana.’
(12) Sipas njö neokomunist-i serb Kosov-a na (Present Admirative)
according to a neocommunist-Masc.Gen/Dat Serb Kosovo-Fem.DEF 1Pl.Dat
qen-ka “pjes-a mö e sigurt e Serbisö.”
be.Participle-have.3sg piece-Fem.DEF most Cl.Nom secure Cl.Nom of-Serbia
‘According to a Serbian neocommunist, Kosovo is “the most
secure part of Serbia.”’16
The same principle is valid for all the other Albanian admirative/eviden-tial forms (with differences due to the use of different auxiliaries); for details see [Buchholz, Fiedler 1987: 154-160].
Bulgarian and Macedonian /-forms and Albanian evidentials have a complex system of meanings, which is shown in Figure 1:
16 The example is taken from [Friedman 2000: 343], with reference to the Kosovo Information Center, Informatori ditor, nr. 1167, 26. IX.1995. The primary meaning of this example is of course admirative-dubitative, but based on reported information, “sarcastic disbelief of the statement being reported”, as Friedman puts it. The marker na (lit. ‘to-us’) is a Dativus Ethicus form [Цивьян 1999: 91-103]. In Albanian this form, when combined with admirative, adds an unambiguous dubitative meaning to the statement [Hubbard 1980].
Slovene 2012 №1
Maxim Makartsev
I 37
Figure i. The meanings of Bulgarian and Macedonian /-forms and Albanian evidentials
Grammatical subdomain Meaning Bulgarian Macedonian Albanian
Perfect Resultative + + ( - ) *
Zustandskonstatierung + + ( - ) *
Evidential Reportive + + +
Conclusive + + +
Emotive Mirative + + +
Other emotions - - +
Modal Epistemic + + +
Optative + + -
*In Northern Gheg Albanian dialects the forms of the so-called inverted perfect are preserved, which have resultative meaning and can also have the function of Zustandskonstatierung.
At the same time, this list of meanings has a different sequence for Bulgarian and Macedonian, on the one hand, and for Albanian, on the other. For Albanian unwitnessed evidentials, the most common meaning is the emotive, which can often have an additional epistemic flavour of disbelief. The pure evidential meaning is quite rare, but is still present in some cases. Therefore, Albanian unwitnessed forms are often called “admirative” (Alb. menyre habitore ‘mood of surprise’). For Bulgarian and Macedonian, the primary meaning is the unwitnessed evidential (with an epistemic flavour of disbelief),17 and they may also have an additional mirative meaning. In some short cliches, the optative meaning can also be found.
Evidential Subfunctions of the Markers
Examples of demek in Bulgarian literary speech are rare, and within my collection of Bulgarian literary prose I was able to find them only in 19th-century novels (I. Vazov, Z. Stoyanov). In [РБЕ III 1981] demek is marked as popular (‘простонар.’). In colloquial speech, however, it is still used. In BNK I was able to find 257 examples18 with demek (with the exception of folklore texts); in the majority of cases demek means ‘so’ and thus has no evidential function. A particularly interesting use of demek is as a ‘translation marker’, when it introduces a translation following a phrase in some other language, e.g.:
17 Victor Friedman proposes that the invariant meaning of Bulgarian and Macedonian /-forms should be not evidential, but non-confirmative (see [Friedman 1986]), thus proposing that they are forms of an epistemic category. Because my primary interest in the present article is the evidential spectrum of meanings, I prefer to use the more traditional term ‘evidential’ and speak about evidential and epistemic semantic components within the evidential forms.
18 In periodicals, translations, and literature. In the Bulgarian Brown Corpus, now an integrated part of BNK, there were also eight examples.
2012 №1 Slovene
Towards a Systemization of Common 38 I Balkan Lexical Evidential Markers
Bulgarian
(13) Alone (demek “Sam”)
[Engl.:] alone demek alone
‘<The English word> “Alone”, meaning “Alone”’ (from a blog)
This usage is a variant of the meaning ‘so to say’. Though it sometimes accompanies quoted or renarrated words, it cannot be called an evidential marker, because its primary function is to explain or to provide an interpretation of the words (whether renarrated or translated).
Among the meanings of the Bulgarian demek there still is a clearly evidential one — conclusive:
Bulgarian
(14) [...той изучаваше подозрително моите тесни беневреци от панагюрски шаяк и влашкия ми калпак, а на думите малко внимание обръщаше.]
Demek ti si ot onija aratlic-i, ot komic-i-te —
demek you be.2sg from those friend-Pl from comician-Pl-DEF
zabelež-i Dočo uvereno i metn-a pogled naokolo.
remark-Aor.3sg D. with.confidence and throw-Aor.3sg look around
‘[He studied suspiciously my narrow trousers made of cheviot from Panagyurishte and my Vlach cap, but to my words he paid little notice.] So you are from those friends, from the comicians,19 said he with confidence, and looked around.’ (Zahari Stoyanov)
This usage is not very common in Bulgarian. Currently, the open access database of BNK provides no more than 30 random items in a search session (from the total of 257 for this marker), but after several sessions I estimate the number of examples with this meaning to be very small in comparison with other meanings. Evidentials are also possible within the scope of demek:
(15) [Мине, не мине време, и в прословутото политическо пространство се появява поредното упражнение върху темата: “В партиите няма личности за продан.”]
Demek, kadri-te se bi-l-i iznosi-l-i i ako
demek personnel-DEF REFL be-LParticiple-Pl outdate-LParticiple-Pl and if
ne bi-l-i proporcionaln-i-te list-i, v bъdešti-ja
not.for be-LParticiple-Pl proportional-Pl-DEF list-Pl in future-DEF
parlament njama-l-o da vleze nikoe ot poznat-i-te
Parliament have.not- LParticiple-3sg Sub enter none of known-Pl-DEF
lic-a.
face-Pl
‘[When some time has passed and in the notorious political space emerges yet another exercise on the topic “There are no people
19 Ironically used instead of komitite ‘rebels’.
Slovene 2012 №1
Maxim Makartsev
I 39
for sale in the parties.”] So, the personnel is outdated and if not for the proportional lists, no known people would enter the future Parliament.’ (BNK)
I assume that /-forms within the scope of demek can emerge in cases in which they introduce the speaker’s interpretation of another person’s statement; their use in such situations is epistemic, transmitting the sense of disbelief.
In dictionaries of standard Macedonian only a conclusive meaning for demek is attested; however, demek is quite often used as a reportive with an additional epistemic meaning of disbelief (16):
Macedonian
(16) Raste-l-a, demek, poddrška-ta za stav-ot
grow.LParticiple-Fem.Sg demek support-DEF for position-DEF
na makedonska-ta vlada vo sporot okolu ime-to
of Macedonian-DEF authority in argument about name-DEF
so Grcija.
with Greece
‘[They say that] demek the support of the position of the Macedonian government in their argument with Greece about the official name is growing [which is not true, taking into consideration the official messages from Brussels and Washington].’ (from a magazine)
Whenever there is a proposition within the scope of the reportive demek, two ways of using evidentials are possible: for the past, /-forms are used; for the non-past, evidentials (cf. 16) as well as neutral forms can be used. This means that at least after demek in the non-past, the opposition between /-forms and other forms is both epistemic and evidential, while for the past, /-forms are the most common possible choice, but in cases of marked confirmative or an epistemic meaning of belief, forms in -v-/-sh-/-j- are also possible.
In dictionaries of standard Albanian demek is described as a conclusive, epistemic, or mirative marker. At the same time, it can also be used as a reportive marker, but again, with a primary epistemic meaning (in this example the epistemic meaning is strengthened by the epistemic marker sikur):
Albanian
(17)
[Ai] vete jep-0 intervist-a, del-0 nö televizor
he oneself take.Aor-3sg interview-Acc.Pl go.Aor-3sg to TV
e u thotö brogkull-a sikur demek, Amerik-a
and Refl say.Praes.3sg nonsense-Acc.Pl as.if demek America-DEF
i ka sy-tö te parti e tyre!
Cl. Acc. Pl have.Praes.3Sg eye-Acc at party Cl.Agr their
‘He went on TV and said a lot of nonsense there, saying that America was spying on their party!’ (Pellumb Kulla)
2012 №1 Slovene
40 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
The use of evidentials after demek in Albanian is not obligatory and they can be used any time the speaker wants to emphasise his or her personal attitude towards the information.
In Macedonian dictionaries goa is described as an epistemic and a report-ive marker. However, a search in written texts did not provide clear examples with a reportive meaning. Apparently, in Macedonian the basic semantics of this marker is epistemic, and its use as a reportive marker (if there is any — we found it neither in written texts nor on the Internet) is only secondary. The evidential meaning that is really associated with this marker is visuality, but it is always contrafactive (‘seemingly it wasp, but in fact it was not’):
Macedonian
(18) Gospodin-ot Dimitar ја napušt-i svoja-ta uloga,
Mister-DEF D. Cl. Acc. Fem leave-Aor.3sg his-DEF role
sedn-a i se zadlaboč-i, goa, i toj
sit-Aor.3sg and Refl concentrate-Aor.3Sg goa too he
vo slušanje-to.
in listening-DEF
‘Mr. Dimitri left his role, sat, and it looked like he concentrated on listening.’ (Slobodan Mickovik')
At the same time, some very clear epistemic uses of this marker are quite common as well:
Macedonian
(19) Se e izmislen-o goa od delikatnost, obdzir-i
Everything be.Praes.3sg imagine-Neut goa from delicacy carefulness-Pl
moraln-i pričin-i [a samo Maza ja izlaga na sram i potsmev pod
moral-Pl cause-Pl
vistinskoto ime]
Everything is made up, as if from delicacy, carefulness, and moral causes, [and only Maza is exposed for shame and laughter, uncovering her real name]. (Slobodan Mickovik )
The Albanian gjoja is also defined in the dictionaries as an evidential and epistemic marker, but its evidential meaning is limited to reportive:
Albanian
(20) Bexhet-i nis-i tö bö-nte njohje-t e mua
B.-DEF go-Aor.3sg Subj.Mark make-Subj.3Sg information-Pl Cl.Agr me
mö paraqit-i si kryetar tö degö-s sö
with introduce-Acc.Sg as head Cl.Gen branch-Gen.Sg Cl.Gen
parti-sö sö fij nö NjuJork, qe Э'Ф
party-Gen.Sg Cl.Gen his in New York that gj°ia
nömöro-nte gjer tashi katörqind antare.
number-Impf.3sg right now 400 members
Slovene 2012 №1
Maxim Makartsev
I 41
‘Bexhet went to inform me, introducing himself as the head of his
party branch in New York, which allegedly had 400 members at
that moment.’ (Pellumb Kulla)
In the dictionaries of Standard Macedonian goa is described as dialectal, colloquial, or archaic, but the Albanian gjoja does not have this association.
Spored ‘according to’ is attested in Bulgarian and Macedonian and seems to have the same meaning in both languages. In both, the most common meanings of spored are accordance (e.g. Bulg. Spored nuždite i sredstvata/Maced. Sporedpotrebite i sretstvata/According to the needs and resources) and stance (Bulg. and Maced. Според мене.../I think...). At the same time, there are at least two meanings connected to the domain of evidentiality: conclusive and reportive. The conclusive meaning can be seen in (21) and it perfectly matches the English preposition according (to):
Bulgarian
(21) Amerkanka-ta ni sčita-l-a spored
American.woman-DEF we.Acc consider-LParticiple-Sg.Fem spored
izlož-en-i-te predmet-i, za edno južnoamerikansko pleme.
exhibit-Participle-Pl-DEF object-Pl for one South.American tribe
‘The American woman, based on the objects at the exhibition, considered us to be a South American tribe.’ (Aleko Konstantinov)
In Macedonian the picture is largely the same. The syntactic structure of sentences with spored is the same as in Bulgarian, and its meanings are the same, too. The only difference is that there are many more examples with conclusive meaning in Macedonian than in Bulgarian.
As we saw, evidential and conclusive meanings are usually combined in the semantics of the lexical markers under analysis, and the epistemic value is usually intertwined with them. They can also convey conclusive meaning (which is less represented in texts for Bulg. spored and Alb. gjoja). Macedonian goa can also convey visual meaning. The epistemic component in demek and goa/gjoja is conventionalised (but sometimes the epistemic degree can be lowered because of the context). It is quite interesting that spored is very often associated with the category of stance (the category indicating the attitude of the speaker towards the information s/he conveys). Thus, the evidential values of this marker might have emerged from the stance meaning.
Scope
The distribution of these markers can vary. Example (22) shows us that it is not necessary to have a proposition within the scope of demek. An adjective phrase is possible in this position as well.
2012 №1 Slovene
42 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
Macedonian Ne e ni
Neg be.Praes.3sg neither
(demek stvarno, а
demek really but
se rešeta-at
Refl riddle-Pl
vo videoigri-te kaj koi vo
in videogame-Pl at which in
vsušnost stvarno) so
in.fact really with
3D
3D
rafal-i
automatic.fire-Pl
vojnici-te na neprijatel-ot.
soldiers-DEF of enemy-DEF
‘Neither does it take place in the video games, where in 3D format (as if it were real and which feels very real) the enemy’s soldiers are riddled by machine-gun fire.’ (Venko Andonovski)
Modern Bulgarian provides multiple examples with NPs within the scope of demek, though in the majority of these examples, demek is used not in an evidential subfunction, but in order to express identity between the left and the right context, cf.:
Bulgarian
(23) TX časovnici (demek Timex) ...
TX.watches demek T.
‘.TX, or Timex watches.’ (from a blog)
As can be seen from (23), not only propositions can fall within the scope of goa, and the following example with an NP in its scope shows this even
more clearly:
Macedo nian
(24) od ovoj goa-disident-ov, g’zo-ližač na Enver.Hoxha
from this goa dissident-DEF ass-licker of E. H.
‘from this seeming dissident, Enver Hoxha’s ass-licker’ (from a forum)
In Albanian any constituent can fall within the scope of both gjoja and demek, but some speakers of literary Albanian from Tirana reject the notion that there can be an NP (which is normal in the texts written by authors from both southern Albania and northern Albania and Kosovo).
Almost any consituent can appear within the semantic scope of the entire PP with spored as a head; cf. (25), where, in English, there is an AP in this position:
Bulgarian
(25) Nakraja vse.pak reši-hme da otid-em v
Finally still decide-Aor.3Pl Subj.Mark go-Praes.lpl to
absolutely natural, spored dum-i-te na Kejt, taverna
absolutely natural according.to word-pl-DEF of Kate tavern
‘Finally we decided to go to an absolutely natural, as Kate said, tavern.’ (Alek Popov)
edna
one
Slovene 2012 №1
Maxim Makartsev
I 43
Syntactic Class
In spite of the clear morphological form and morphotactic status of the markers (free, one-word items), there is little or no correspondence between the various dictionaries as to how to treat them in terms of syntactic class (the clearest is the case of spored, which is treated as a preposition in the majority of dictionaries;20 demek and goa/gjoja are sometimes called adverbs, conjunctions, or particles). The question of syntactic class goes far beyond the boundaries of this article, but at least I can try to point in the direction of some possible solutions.
It is clear that demek and goa/gjoja, as well as many other lexical evidential markers (božem, kaj, navodno, and others), should belong to the same syntactic group or class, because the rules of their use and their government do not differ significantly. There is still a wide range of variation in grammatical terminology in different language traditions. For example, in the English tradition the words modifying the entire preposition (frankly, personally, apparently, luckily, etc.) are usually called sentential adverbs. However, the words modifying the entire preposition (though the list differs in different languages) in Russian, Bulgarian, and Albanian are included into частицы / частицы (Rus./Bulg.) or pjeseza (Alb.) ‘particles’, the term наречие (Rus., Bulg.) and ndajfolje (Alb.) ‘adverb’ being generally reserved for words modifying VPs or APs. The Macedonian tradition introduces a special group of модални (начински) зборови ‘modal (mood) words’ [Конески 1967: 543], which includes discourse and stance21 markers, so evidential markers such as demek and goa would fit perfectly into this category.
A good way out of this problem of classification would be the introduction of a special group of words (evidential words), as is done in [Бояджиев, Куцаров, Пенчев 1999: 361; Куцаров 2007: 169] with the term частици ‘particles’ — they introduce преизказни частици ‘renarrative particles’ and конклузивни (умозаключителни и предположителни) ‘conclusive (inferential and assumptive)’, along with дубитативни ‘dubitative’. I tried to do the same thing in my reference book of Bulgarian grammar [Макарцев, Жерновенкова 2010: 184-185].22 It is quite clear though that the taxonomy of the “evidential words” would depend on the author’s idea of the syntactic classes.
20 Though [Мургоски 2005] defines it as an adverb.
21 See [Biber, Finnegan 1988] for discussion of this topic.
22 The three cited books are manuals and reference books, so it is inevitable that they sometimes simplify and skip the less important points to give a clear main idea. At the same time, I think that A. Wierzbicka’s comments about a semantic metalanguage can be applicable here, as well as in many other complicated areas of linguistics: “The semantic metalanguage will only be genuinely «explanatory» if it is so clear and immediately comprehensible as not to require «explanation» in its turn... A semantic language which purports to explain must make the complex simple, the confused transparently plain, the obscure self-explanatory” [Wierzbicka 1972: 1-2].
2012 №1 Slovene
44 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
The Macedonian classification provides the most reasonable solution. It appears that the class of модални зборови ‘modal words’ has a great deal in common with the evidential words. It would be justified to combine them under the term “propositional modifiers” or “stance words”.
Balkan Evidential Markers and Translation
The comparative analysis of Balkan evidential markers in translation from one Balkan language into another enables us to elucidate both of the general problems we indicated at the beginning of the article. The many phenomena from Macedonian electronic media are particularly fruitful in this respect because recently (after the Ohrid agreements in 2001) Albanian has started to play a very important role in the multinational community of this country. Since for some Albanian politicians the question of the language they use in communication is of high importance, many of them speak only Albanian, even though they are fluent in Macedonian as well. Some media present their speeches with subtitles, which gives us the opportunity to compare the original with the translation.
In this context, I would like to refer to an interview Ali Ahmeti (a member of The Democratic Union for the Integration of Albanians in Macedonia) gave in 2008 to Radio Free Europe. The radio made up a short film and uploaded it onto YouTube. Ali Ahmeti spoke Albanian (26) and the film was provided with Macedonian subitles (26a):
Albanian
(26) Me sa unö e kam vözhguar situatön nga mbrenda Parlamentit, ka qenö nji situatö jashtzakonisht e rendö, qö Parlamenti i Maqedo-nisö ka qenö i rrethuar prej njisive tö Alfa, tö pauniformuar, por me kallashnikov, me arm tjera pör tö paisur me armö pör antiter-rorizöm, köshtu qö ajo qö thuhet (1) se... gjoja (2) se simpatizantö apo anötare i Bashkimit Demokratik pör Integrim kanö qenö (Perf) tö paisur apo kanö ardhö (Perf) (3) armatosur fare nuk qöndron.
‘As far as I could follow the situation from the Parliament building, it was an extraordinarily difficult situation: The Macedonian Parliament was surrounded by Alfa troops, without uniforms, but armed with Kalashnikovs and other guns, to fight against terrorism, so the thing that has been said (1) that. gjoja (2) that supporters or members of the Democratic Union for Integration were armed or came armed (3), are not right.’
Macedonian
(26a) Onaka kako što možev da sledam vnatre od sobranieto, toa beše navistina teška situacija. Sobranieto na Makedonija beše
Slovene 2012 №1
Maxim Makartsev
I 45
opkruženo od specijalni edinici Alfi, koi bea bez uniformi, so kalašnikovi i so drugi oružja za protivterorističko dejstvuvanje, taka što onie što velat (1) deka navodno (2) simpatizeri ili členovi na DUI bile dojdeni (PERF. ш EviD.) (3) vooruženi, toa voopšto ne drži.
In (26) Ali Ahmeti uses two evidential markers with a perfect form after them. The first marker, thuhet ‘say.Reflexive.Praes.1pl’ (1), underscores the evidential component (as it is an ordinary verb of speech); the second marker, gjoja (2), underscores the epistemic component; and the forms kane qene (PERF) tepaisur apo kane ardhe{PERF) ‘have been armed or have come armed’ (3) are neutral both evidentially and epistemically. In the Macedonian translation (26a) the situation differs: velat ‘say.Praes.3pl’ (1) underlines the evidential component (as does the Albanian thuan) as well, but navodno (2), which translates gjoja (2), also underlines the evidential component, and only the form within the scope of navodno, namely bile do/deni(PERFjII EviD.) (3) combines evidential and epistemic meanings in its semantics. So both the epistemic and the evidential components of the original sentence emerge in the translation, but different means are used to accomplish this. This is schematically shown in Fig. 2.
Figure 2. Evidential and epistemic components of the meaning of (26) and (26a)
1 2 3
English been said that<...> allegedly <...> were <...> or had come <...>
Albanian (26) thuhet se <...> gjoja se <...> kane. qene ... apo kane ardhe <...>
meaning Ev Ev+Ep 0
Macedonian (26a) velat deka navodno <...> bile doideni
meaning Ev Ev Ev+Ep
This example illustrates the issues I am dealing with in this article. Both in Albanian and Macedonian the meaning of the utterance is shown as a sum of the meanings of different markers, some of them evidential and some evidential+epistemic. It is very important to emphasise that the translator did not use the Macedonian word goa, formally corresponding to the Albanian gjoja, because those interlinguistic homonyms have different stylistic associations and meanings (it can as well be considered dialectal in Macedonian).
The other example of ‘inter-Balkan’ translation is a very characteristic fragment from Aleko Konstantinov’s “Бай Ганьо” (“Baj Ganjo”), which has a complicated narrative structure. It is a passage from Baj Ganjo’s direct speech in which he describes what happened to a friend of his at a university. I have
2012 №1 Slovene
46 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
discussed elsewhere the narrative structure of this fragment and the interplay between the lexical and grammatical markers [Макарцев 2008]. The translations into Macedonian and Albanian are quoted according to V. Friedman [Friedman 1998], and the English translation is also his. While Friedman focuses on the verb forms, expressing different degrees of distance and confirmation (comparing these translations with Romanian and Turkish), what interests us here is the use of lexical evidential markers:
Bulgarian
(27) Dosega da e stanal veke doktor, ama inat hora tukašnite. Rektora, kaj (1), ne go ostava. Ne može, kaj (2), kazal mu, za tri meseca da staneš doktor, kaj (3). A be kak da ne može, kogato momčeto znae?
Macedonian
(27a) Dosega vek'e trebaše da stane doktor, ama inaetčii se ovdešnive. Rek-torot, veli (1), ne go ostaval. Ne može, veli (2), mu rekol za tri meseca doktor da stanes, veli (3). A, be, kako ne može, koga znae momčeto?
Albanian
(27b) Do tö qe böre doktor gjer tani, po köta tö kötushmit janö kokö-fortö. Rektori, gjoja (1), nuk e lejuaka! Nuk qenka e mundur, gjoja (2), t’u böka doktor, gjoja (3), njeriu, vetöm pör tre muaj!
E po pse s’mundet, ore t’i thuash, kur guni i di pör bukuri?
‘He should have been a doctor already by now, but the people here are spiteful. The rector, he says (1), won’t let him. “It is impossible” — he says (2) [that] he said to him — “for you to become a doctor in three months” — he says (3). Well, but how can it be impossible when the fellow knows [everything]’ 23
As we have seen, technically it would have been possible to use a single marker in all of these sentences, for example demek or even gyoa/goa/gjoja. But the only translation which actually uses one of these common markers is the Albanian. The situation becomes even more interesting if we note that the Turkish translation does not use either of these markers, preferring evidential mij-forms instead.24 Here again, the common items in the list are somehow rejected due to their system of meaning and semantic aura in their respective languages.
23 The Bulgarian original is cited here according to [Константинов 1980: 162], the Macedonian and Albanian translation according to [Friedman 1998], with the source, respectively, [Константинов 1967] and [Konstantinov 1975]. The translation into English is V. Friedman’s.
24 Here is the Turkish translation: “Bu akilla artik doktor olmaliydi, ama buradakiler inat! Rektör razi gelmezmi§ “Üg ayda doktor olunmaz,” demi§ Rektör. A be, oglan biliyor ya; bildikten sonra nasil olunmazmi§” [Konstantinov 1972], cited according to [Friedman 1998]).
Slovene 2012 №1
Maxim Makartsev
I 47
Conclusion
As we expand the list of Balkan languages under consideration, the list of phenomena overlapping at least two of the languages also expands. This is the case with the evidential markers as well, because many of them have parallels in other Balkan languages. That is why, in my opinion, in a project for a database for evidential markers in European languages (as in [Wiemer, Stathi 2010]), it would be essential to add a special field for links between lexical evidential markers common to several languages. For example, the Macedonian and Albanian goja/gjoja would be linked both together and to the Turkish marker, as well as to the Serbian (ko)đoja, which has the same etymology.
References
Aikhenvald 2004
Aikhenvald A., Evidentiality, Oxford.
Alexander 2001
Alexander R., “Bridging the Descriptive Chasm: The Bulgarian «Generalized Past»”, in:
V. A. Friedman, D. L. Dyer (eds.), Of all the Slavs my Favorites: In Honor of Howard I. Aronson (= Indiana Slavic Studies, 12), 1-29.
Biber, Finnegan 1988
Biber D., Finnegan E., “Adverbial Stance Types in English”, Discourse Processes, 11, 1-34. Boretzky 1976
Boretzky N., Der Türkische Einfluss auf das Albanische, 2: Wörterbuch der albanischen Turzismen (= Albanische Forshungen, 12), Wiesbaden.
Buchholz, Fiedler 1987
Buchholz O., Fiedler W., Albanische Grammatik, Leipzig.
Bužarovska 2006
Bužarovska E., “Pathways of Semantic Change: from Similarity Marker to Sensory Evidential”, Slavia Meridionalis. Studia Linguistica Slavica et Balcanica, 6, 185-208.
£abej 1976
Qabej E., Studimegjuhesore, I: Studime etimologjike ne fushe te shqipes, A-O, Prishtine.
Fielder 1996
Fielder G.E., “The Relationship Between the ima (‘have’) Perfect, the sum (‘be’) Perfect and Auxiliary Loss in Macedonian and Bulgarian”, in: Studies in Macedonian Language, Literature and Culture, Skopje, 177-185.
Fielder 1997
Fielder G.E., “The Discourse Properties of Verbal Categories in Bulgarian and Implications for Balkan Verbal Categories”, Balkanistica, 10, 162-183.
Fielder 1998
Fielder G.E., “Discourse Function of Past Tenses in Pre-Modern Balkan Slavic Prose”, in:
A. Timberlake, M. Flier (eds.), Proceedings of the 12th International Congress of Slavists, Columbus, 344 - 361.
2012 №1 Slovene
48 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
Fielder 1999
Fielder G. E., “The Origin of Evidentiality in the Balkans: Linguistic Convergence or Conceptual Convergence?”, Mediterranean Language Review, 11, 60-89.
Friedman 1982
Friedman V. A., “Admirativity in Bulgarian Compared with Albanian and Turkish”, in: D. Kosev (ed.), Bulgaria: Past and Present, 2, Sofia, 63- 67.
Friedman 1986
Friedman V.A., “Evidentiality in the Balkans: Bulgarian, Macedonian and Albanian”, in:
W. Chafe, J. Nichols (eds.), Evidentiality: the Lnguistic Coding of Epistemology, Norwood,
168- 187.
Friedman 1994
Friedman V. A., “Surprise! Surprise! Arumanian Has Had an Admirative!”, Indiana Slavic Studies, 7, 79-89.
Friedman 1998
Friedman V.A., “The Grammatical Expression of Presumption and Related Concepts in Balkan Slavic and Balkan Romance”, in: M. Flier, A. Timberlake (eds.), American Contributions to the 12th International Congress of Slavists, Bloomington, 390-405
Friedman 1999
Friedman V. A., “Evidentiality in the Balkans”, in: U. Hinrichs (Hrsg.), Handbuch der Südosteuropa-Linguistik, Wiesbaden, 519-544.
Friedman 2000
Friedman V. A., “Confirmative/Nonconfirmative in Balkan Slavic, Balkan Romance, and Albanian with Additional Observations on Turkish, Romani, Georgian, and Lak”, in: B. Utas , L. Johanson (eds.), Evidentials in Turkic, Iranian and Neighbouring Languages, Berlin, 329-366.
Friedman 2001
Friedman V. A., “Hunting the Elusive Evidential: The Third-Person Auxiliary as a Boojum in Bulgarian”, in: V.A. Friedman, D.L. Dyer (eds.), Of all the Slavs my Favorites: In Honor of Howard I. Aronson (= Indiana Slavic Studies, 12), 203-230.
Gramatika 1983
Ст. Стоянов (ред.), Граматика на съвременния български книжовен език, 2: Морфология, София.
Guentcheva 1996
Guentcheva Z. (ed.). L’enonciation mediatisee, Louvain, Paris.
Hubbard 1980
Hubbard Ph., The Syntax of the Albanian Verb Complex, San Diego (Ph. D. dissertation, University of California, San Diego).
Igla 2006
Igla B., “Zum Renarrativ im Slivener Romani”, Балканско езикознание, 45/1, 55-61.
Kampf, Wiemer 2011
Kampf V., Wiemer B., “Inventarisierung und Analyse lexikalischer Evidenzialitätsmarker des Bulgarischen: Adverbien, Partikeln und Prädikative” (in press).
Kehayov 2008
Kehayov P., “Interactions between Grammatical Evidentials and Lexical Markers of Epistemicity and Evidentiality: a Case Study of Bulgarian and Estonian”, in: B. Wiemer, V. A. Plungjan (Hrsg.), Lexikalische Evidenzialitäts-Marker in Slavischen Sprachen (= Wiener Slawistischer Almanach, Sonderband 72), München, Wien, 165-201.
Slovene 2012 №1
Maxim Makartsev
I 49
Konstantinov 1972
Konstantinov A., Bay Ganü, transl. I. Aglagül, Istanbul.
Konstantinov 1975
Konstantinov A., Baj Gano, perkth. Dh. Xaxa, Prishtine.
Levin-Steinmann 2004
Levin-Steinmann A., Die Legende vom Bulgarischen Renarrativ. Bedeutung und Funktionen der kopulalosen l-Periphrase (= Slavistische Beiträge, 437), München.
Meyer 1891
Meyer G., Etymologisches Wörterbuch der albanesischen Sprache, Strassburg.
Orel 1998
Orel V., Albanian Etymological Dictionary, Leiden, Boston, Köln.
Petroska 2012
Petroska, E., “Grammatical and Lexical Markers of Evidentiality in Macedonian”, Balkanistica, 25/2, 139-152.
Rakhilina 1996
Rakhilina E. V., “Jakoby comme precede de mediatisation en russe”, in: Z. Guentcheva (ed.), L’enonciation mediatisee, Louvain, Paris, 299-304.
Ramat 1996
Ramat P., “«Allegedly, John is ill again»: Strategies pour le mediatif”, in: Z. Guentcheva (ed.), L’enonciation mediatisee, Louvain, Paris, 287-298.
Wiemer, Plüngjan 2008
B. Wiemer, V. A. Plüngjan (Hrsg.), Lexikalische Evidenzialitäts-Marker in Slavischen Sprachen (= Wiener Slawistischer Almanach, Sonderband 72), München, Wien, 239-284.
Wiemer, Stathi 2010
Wiemer B., Stathi K., “The Database of Evidential Markers in European Languages. A Bird’s Eye View of the Conception of the Database (the Template and Problems Hidden beneath it)”, Language Typology and Universals, 63/4, 275-289.
Wiemer, Vrdoljak 2011
Wiemer B., Vrdoljak I., “Evidenzielle Partikeln vs. Satzadverbien im serbisch-kroatischen und slovenischen. Teil I. Ein Forschungsbericht”, Die Welt der Slaven, 56, 100-130.
Wiemer, Vrdoljak (in press)
Wiemer B., Vrdoljak I., “Evidenzielle Partikeln vs. Satzadverbien im serbisch-kroatischen und slovenischen, Teil II: Auswertung eines Forschungsberichts” (in press).
Wierzbicka 1972
Wierzbicka A., Semantic Primitives, Frankfurt a. M.
БЕР I
Български етимологиченречник, София, 1971-.
Бояджиев, Куцаров, Пенчев 1999
Бояджиев Т., Куцаров И., Пенчев Й., Съвременен български език (фонетика, лексикология, морфология, синтаксис), София.
Гајдова, Лаброска 2008
Гајдова У., Лаброска В., “Категоријата дистанца во македонскиот, во бугарскиот и во српскиот јазик (граматички и лексички показатели)”, в: Реферати на македонските слависти за XIV-от Мегународен славистички конгрес во Охрид, 10-16 септември 2008 година, Скопје, 79-93.
2012 №1 Slovene
50 I
Towards a Systemization of Common Balkan Lexical Evidential Markers
Гајдова, Лаброска 2010
Гајдова У., Лаброска В., “За евиденцијалноста во македонскиот јазик”, Македонски јазик, 61, 53-58.
Конески 1967
Конески Б., Граматика на македонскиот литературен јазик, Скопје.
Константинов 1967
Константинов А., Бај Гањо,прев. Г. Цаца, Скопје.
Константинов 1980
Константинов А., “Бай Ганю”, в: Константинов А., Събрани съчинения, I, София. Куцаров 1978а
Куцаров И., “Преизказни модификатори в южните славянски езици”, Съпоставително езикознание, 4, 41-54.
Куцаров 1978в
Куцаров И., “Преизказни модификатори в източните славянски езици”, Съпоставително езикознание, 5, 40-48.
Куцаров 1978c
Куцаров И., “Преизказни модификатори в западните славянски езици”, Съпоставително езикознание, 6, 47-54.
Куцаров 1978d
Куцаров И., “Функционално-семантически полета на категорията преизказност в славянските езици”, Славянска филология, 15, София, 391-401.
Куцаров 2007
Куцаров И., Теоретична граматика на българския език. Морфология, Пловдив. Макарцев 2008
Макарцев М., “К вопросу о связи лексических и грамматических показателей эвиденциальности в болгарском языке”, в: B. Wiemer, V. A. Plungjan (Hrsg.). Lexikalische Evidenzialitäts-Marker in Slavischen Sprachen (= Wiener Slawistischer Almanach, Sonderband 72), München, Wien, 239-284.
Макарцев 2010
Макарцев М. М., Категория эвиденциальности в пространстве балканского текста (на материале болгарского, македонского и албанского языков), Автореферат дисс. ... к.ф.н., Москва.
Макарцев, Жерновенкова 2010
Макарцев М. М., Жерновенкова Т. Г., Болгарский язык. Грамматика, Москва. Младенов 1941
Младенов С., Етимологически и правописенречникъ на българския книжовен език, София.
Мургоски 2005
Мургоски З., Македонско-англискиречник, Скопје.
Николаева 2008
Николаева Т. М., Непарадигматическая лингвистика (История “блуждающих” частиц), Москва.
Падучева1996
Падучева Е. В., Семантические исследования (Семантика времени и вида в русском языке. Семантика нарратива), Москва.
Slovene 2012 №1
Maxim Makartsev
I 51
Петроска, Тофоска 2011
Петроска Е, Тофоска С.-С., “Лексички показатели за прекажана евиденцијалност во македонскиот и во полскиот јазик”, во: Зборник од Осмата македонско-полска научна конференција, Скопје.
РБЕ I-
Речник на българския език, София, 1977-.
Стойков1967
Стойков С., Банатският говор (= Трудове по българска диалектология, 3), София. Тофоска 2008
Тофоска C.-С., “Глаголски предикати-носители на информацијата за евиденциалност во македонскиот и полскиот азик”, во: XXXV Научна конференција — лингвистика (Охрид, 11. VIII. — 28. VIII.2008), Скопје, 303-310.
Усикова 2003
Усикова Р. П., Грамматика македонского литературного языка, Москва.
Цивьян 1999
Цивьян Т. В., “Языковой жест, направленный на себя (dat. eth. и его трансформы)”, в: Цивьян Т. В., Движение и путь в балканской модели мира, Москва.
Максим Максимович Макарцев, к.ф.н.
Институт славяноведения РАН,
отдел типологии и сравнительного языкознания
119334 Москва, Ленинский пр-т, 32а
Russia / Россия
makartsev@bk.ru
