Bulgarian-English Code-switching in Internet Forum Communication : The BG-mamma Case

The aim of the paper is to introduce the author's observations on the influence that English has on the Bulgarian forum "Netspeak" for women living in the USA. The collected data for the investigation are excerpted from a Bulgarian forum platform. The emphasis is on code-switching, one of the main topics in sociolinguistics; as well as on the interlanguage, which is a result of the penetration of some elements from English into the first language of communicators. The orthography may also be influenced as in Bulgaria the Cyrillic alphabet is used. The informal situation is a prerequisite for the mixed style use, especially in vocabulary. Different topics have been observed and discussed by users in the forum: the choice of baby accessories; baggage loss in airports; child diseases, child physical and mental development, etc. The investigated forum discussion includes 52,020 lexical items. Excerpted English words are few only 1.5%. Most of them are nouns because of their easy grammatical adaptation. The verb group is not affected so noticeably. The Bulgarian verbal system is very complicated and English verbs used in Bulgarian context are grammatically adapted. There are some exceptions when people use ready-made constructions like idioms and phrasal verbs.


General remarks
The aim of the paper is to examine the influence of English on Bulgarian speech of Bulgarian women who live in the USA. They are 1 st generation immigrants, and they have lived in Bulgaria for a long time. Most of them have written their personal immigrant stories elsewhere in the forum platform. As these narratives are not part of the threads I have observed, they are not included here. Generally, these women come to the USA after finishing school, sometimes after university, i.e., their Bulgarian is excellent. Their English level varies but it is usually good enough to live in the USA without communication problems. Thus, I put the emphasis on code-switching, one of the main topics in sociolinguistics, as well as on the interlanguage, which in the case of Bulgarians results from the penetration of some elements from English, the more influential language (in fact the most influential language in the world), into the first language, the mother tongue of the communicators.
Such observations concerning Bulgarian speech of people living abroad are relatively few. Sotirov (2000) investigates the Bulgarian speech of Bulgarians, living in Hungary; Sivova-Tsankova (2006) is interested in the speech of Bulgarian students in Germany, Pachev (2005) conducts research into the speech of Bulgarian people in Slovakia, Krejcova (2015) -into the speech of Bulgarian expatriates in the Czech Republic, and Kocheva (2017) -into the speech of Bulgarians in Vienna, Austria.
The influence of English onto Bulgarian speech of people living in the USA is partly investigated by Mitsova (2013) who observes the oral speech of 5 Bulgarians -two women and three men. According to her, there are not many examples of code-switching between English and Bulgarian, most of which results from the process of accommodation in the life of the host country and its culture (Mitsova 2013:151). She gives some examples of such mixed uses. However, the elaborate investigation of the English forms and words in her corpus is not available.

Code-switching and interference -some terminological notes
According to Crystal (1987), code-switching, or language switching, occurs when an individual who is bilingual alternates between two languages during his/her speech with another bilingual person. From another perspective it is described as "the use of elements from more than one language within a conversation" (Deumert 2011: 268). However, a lot of languages adopt into their vocabularies a large number of foreign words and some monolinguals also use these borrowed words in specific situations without any intention of mixing codes, so Deumert's definition is partial. In Haugen's view, code-switching "occurs when a bilingual introduces a completely unassimilated word from another language into his speech" (Haugen 1956: 40). In fact, code-switching is a result of bilingualism or multilingualism. A lot of researchers argue that monolinguals are a minority around the world (Macha 1991, Franceschini 1998. For this reason, Deumert says that the phenomenon is ubiquitous among multilinguals (Deumert 2011: 268). The same is claimed by Eastman '[w]here people use a mixed language regularly, codeswitching [i.e. varietyalternation] represents the norm' (Eastman 1992: 1; see also Alvarez-Caccamo 1998: 42). Rüdiger recalls the Poplack and Sankoff's (1984) view that it can be problematic to distinguish between the use of loanwords, code-switching, and interference (Rüdiger 2018: 185). The authors (Poplack & Sankoff 1984: 103-104) present some criteria for the loanwords characterization, i.e. frequency of use, native-language synonym replacement, morphophonemic and/or syntactic integration, and acceptability. As the investigated bg-mamma forum users live in the USA and their language is not representative of the English influence onto Bulgarian, the English uses in their texts could not be treated as loanwords.
Because of its mixed character and its pragmatic origin code-switching (CS) is close to pidginisation. However, their linguistic characteristics are different. According to Franceschini: Common features of CS and pidgin languages are restricted to external, social factors; the internal systems of CS and pidgins are completely different. For instance, CS does not have a reduced number of vowels, restricted vocabulary and/or a smaller set of morphological possibilities as it is attributed to pidgin languages (cf. Mühlhäusler 1986; Arends, Muysken & Smith 1995) -quite the contrary: CS enormously multiplies the possibilities of expression. (Franceschini 1998: 67). Some researchers claim that code-switching requires strong proficiency in all languages being utilized and serves complex social functions (Flores and Lewis 2016: 100). At the same time, bilinguals are not always excellent speakers of both languages. The same is said by Meeuwis and Blommaert: "code-switching speakers are not necessarily speakers who would be capable of producing monolingual speech in the languages used in code-switching. From an empirical point of view, a simple connection between codeswitching and bilingualism rests on shaky grounds." (Meeuwis & Blommaert 1998: 77).
The terminological observations vary and different researchers present their specific viewpoints. Sometimes the discussed terms may be very close in their meaning. For example, interference may be viewed as the transference of elements of one language to another at various levels including phonological, grammatical, lexical and orthographical (Berthold et al. 1997). According to Poplack and Sankoff (1984: 102) "[c]ode-switching is simply the alternate use of the two languages in discourse, and even in the sentence without any necessary influence of one language on those stretches of discourse realized in the other." Haugen includes 'code-switching', 'interference' and 'integration' in his 'three stages in diffusion' (Haugen 1956: 40). The perspective accepted here is Deumert's (2011: 268) mentioned above: "the use of elements from more than one language within a conversation". However, as English is the source of newcoming loanwords in Bulgarian (Blagoeva 2006, Popova 2013, Yordanova 2013, Grozdanova 2017, the investigated English uses are treated as elements of code-switching process if they are not included in Bulgarian dictionaries.

Object of investigation
The object of the research is informal written communication in the largest Bulgarian maternity discussion forum -bg-mamma. As the interference may be seen in different levels and to different degrees, it is important to see the extent to which English, the official language of the country where users live, influences their first language, Bulgarian. Some investigations show that Spanish-English bilinguals "report more linguistic interference (code-switching) when they communicate in Spanish, their first-language, and little or no code-switching when they communicate in English, their second-language." (Heredia & Brown 2005: 215). Deumert imposes several facts for code-switching -it is typical of multilingual persons; it occurs within sentences, and at syntactic boundaries, it happens in spoken language, and in informal written language (italic is mine, B.T.), as well as in stylized linguistic performances (Deumert 2011: 269). As it could be seen, there are some possible reasons for the switching from one language to another -different levels of language proficiency, to express solidarity, to show attitude, language play, a topic of conversation. (Crystal 1987). Therefore, the maternity discussion forum communication of users who live in the USA and who write in Bulgarian is the place where code-switching is expected to be found. The hypothesis is that the influence of English on Bulgarian speech of users might be viewed at the orthographical, grammatical and more noticeably -at the lexical level. It is expected that the orthographical level may be influenced by the use of the Latin alphabet or by preferring the written form of an English word which is different from the officially accepted in Bulgarian.
The grammatical level is more conservative and the changes there are slightly visible. However, it will be investigated how different parts of speech are morphologically adapted to Bulgarian.

Corpus
As mentioned, the collected data for the investigation are from the biggest Bulgarian forum platformbg-mamma. In this platform, there are a couple of forums and sub-forums focused on different topics and problems. I have chosen to make the observation of a virtual place, called 'тема' (topic) that unifies women living in different parts of the USA. Every topic is, in fact, a thread of a limited number of posts. The whole topic consists of 50 pages. Every page contains 15 posts.
When the required pages are reached a user may start a new one. The excerpted words and expressions are collected by the topic № 150, named '150 Джетлагнати американски летни ваканции!' ['150 jetlagged American summer holidays']. It was active between August 7 th and September 21 st 2013, i.e., for a month and a half. The authors are 45 users living in different parts of the USA.
The topic of mothers living in the USA is a part of the forum 'Дом и семейство' ['Home and family'], sub-forum 'Родители в чужбина' ['Parents abroad'].
The following discussion topics may be identified in the corpus: -Physical and Mental Development of Baby; -Travel problems -jet-lag; luggage loss, etc.; -Where to live in the USA; -Baby accessories -bassinets, prams, cribs, etc.; -Help from parents when the baby was born; -Kindergartens and primary schools in the USA; -Ethnic diversity in American schools; -Documentations -insurances, visas, etc.; -Nutrition; -Public breastfeeding.
As one may see, topics are various and linked to everyday situations. Some of them include motherhood problems (nutrition, accessories, school problems, etc.), immigrant problems (visas, insurances, etc.), others are more general (travelling, lodging, etc.).

Methods
The analysis combines a quantitative approach with a qualitative and analytical one (see Rossman and Wilson 1985). The data include the frequency of the use of English words in Bulgarian informal written discussions as well as the parts of speech which are influenced by code-switching.
At the same time, the so-called discourse-functional approaches are used -they investigate why some linguistic choices are made; what are the functions of the grammatical and lexical alternations of a language (Cumming & Ono 1997: 114).
In the contribution it is accepted the view of discourse as a set of interrelated elements1 -topics discussed, participants in communication and the language used (Halliday and Hasan 1990: 24;Todorova 2010, etc.). In this view, the linguistic data are not independent of the 'context of situation' (the term of Malinowski, used from Halliday & Hasan 1985) -from the profile of users, as well as of their interests and topics of discussion.
The possible ways of investigating the language change and the influence of English according to Teubert and Čermakova (2004: 22) are: -"the change of frequency of words or other units of meaning (compounds, multi-word units, collocations, set phrases), which is often indicative of a change in meaning or a change in the domains in which words are used; -the occurrence of new words; -the occurrence of new larger units of meaning; -changing context profiles, i.e. changes in the frequencies of words occurring in the contexts of words or other units of meaning.". The immediate qualitative analysis consists of the identification and description of the English words. When they are identified, they are classified according to the part of speech they belong to. The definition of the English word will be the following: the word which comes in Bulgarian speech from English and which is not a part of active vocabulary of Bulgarian people, so it is not found in the Dictionary of Bulgarian language (RBE 2001(RBE -2015, or which is used in Bulgarian with other meanings, different from the use in the investigated discussion.
Therefore, the use of the new words as well as the extending of the meaning of some words because of the influence of English is the aim of this paper. The change of typical collocations will also be investigated.
The English words and phrases in the topic are excerpted manually and double-checked as there is no available software which may recognize the forms. The investigated discussion is copied in the MS Word file and the calculation of the word number in the investigated corpus is made by the help of MS Word tools. The percentage of English nouns, adjectives and other parts of speech as well as the proportion of Cyrillic and Latin forms is made by the help of Excel.

Code-switching in the bg-mamma "American" topic -scope and characteristics
The influence of one language on another is a result of a large number of reasons -behavioural, social, cultural, etc. The analysis of the reasons for the language transfer also includes the similarities and differences between the compared languages as well as the specifics of the code-switching and the interference.
Despite the expectations, the excerpted English items in the investigated topic are few -only 769 out of 52020 lexical items, or ≈1,5% (1,478)2. However, there are some explanations for this fact -the topic is written by people, whose first language is Bulgarian and this is their 'Bulgarian place'. The rules of the forum require they write in Cyrillic. Thus the English orthography use is hindered.
A lot of researchers argue that nowadays the English language influences Bulgarian. However, the observations show that in the majority of the bg-mamma topics only 3-4 English words in all 50 threads could be seen. In the investigated topic they are more frequent, but their number is not significant -1,5 %. Unfortunately, there is not a reference corpus in Bulgarian, where the typical number of English words may be seen. Therefore, the scope of the influence and the reasons for it will be more important in the research.
As previously emphasized, the extracted items are divided into parts of speech (see Table 1.), and I observe the specific use of every part. It is also important whether they are morphologically adapted to Bulgarian or are used without change. Thus the characteristics of English words and phrases used in the investigated part of the forum are presented successively. I have divided the words into word classes according to their grammatical characteristics.

Nouns
Nouns come first because their number in the corpus is the greatest. The excerpted nouns are 364 (≈74 % of the excerpted lexemes). Therefore, nominatives are very important parts of the texts as they mark objects and ideas and they form the thematic structure of the texts. In fact, in all languages, the lexical borrowings are usually nouns. Most of them are used in their original form, others are written in Cyrillic (Table 2.)

2
There are 765 excerpted items which form 492 lexemes including compound and phrasal words. I assume that the difference between compound words and collocations is a difference in degree. Compounds are linked together -they usually have their own meaning, different from the meaning of their components. They cannot be divided into morphological and syntactic elements. Collocations are phrases of words that usually go together. One of them is a modifier. "There is no formal means in distinguishing between compounds and collocation except that compounds are syntactically inseparable, whereas the items in a syntactic collocation are more loosely associated" (Macdonald & Darjowidjojo 1967: 61). So the challenge for the researcher is how to organize the investigated material; whether to list every word that can be found or the whole expression presents the unit. As it is known, sometimes the word combination represents the concept, or serves as a catchword, has a phatic or emotive function. The appropriate approach has to be mixed. When people live in a certain country, they usually 'catch' the vocabulary of the environment. I assume that they will use more English nominatives than verbs because they read or listen to and use these words all the time. If someone is expecting a baby in the USA, they will know that 'the bassinet' is 'the bassinet', and will use this word instead of its Bulgarian correspondence ('бебешка люлка'). In addition, objects are the first words we learn in our language learning process. It does not matter if we learn our mother tongue or second, third, etc. language. There is another important reason for the noun borrowing. When talking about the transfer from English to Bulgarian, the easiest words to be borrowed from a grammatical point of view are nouns. They do not need a grammatical transformation in many cases. They receive their gender according to the Bulgarian grammatical rules.
When Bulgarian people use English nouns in a native milieu, they form plural and definite forms using active language models. Most of the expressions in the corpus are likewise grammatically adapted to Bulgarian. In the investigated corpus all uses of noun definite forms (54 examples) are formed with the help of a Bulgarian definite article. In Bulgarian this article is a morpheme at the end of the word -'баунсър-а' (DEF), 'евалюейшън-а' (DEF), etc. When the plural forms are used, they are more often with English inflections -'carbs ', 'colard greens', 'карс'[cars], etc. (37 times). Bulgarian plural noun forms are used 20 times (see Table 3.). All uses of Bulgarian definite or plural forms are written in Cyrillic. English endings for plural are used both in Cyrillic and Latin examples.

B. Common nouns
It is interesting that almost 14 percent or 41 common nouns are written with the help of the Latin alphabet (i.e. in English). These are cases of digraphia -in the first example brahok chegn chram is in the Latin alphabet, the other words are in Cyrillic.
Although borrowings are not very frequent and digraphia is not preferred in the posts above,

mean doctors and insurances.'
An interesting phenomenon is the fact that some of the excerpted nouns are used in Bulgarian but with a different meaning. The "extending" of lexical meaning is not something unexpected, and the process depends on its relevance (Diez-Arroyo 2016: 616). In the following example, the word 'основа' instead of 'база' will be used by Bulgarians who are not influenced by English.
( Some of the excerpted nouns are loanwords in English as well -i.e. English is a mediator between the language of origin and Bulgarian. English is a language that borrows foreign words without oppression. For example, 'камбуча' (kombucha), 'алфредо сос' (alfredo sauce), 'brahok chegn chram' clearly are not originally "English" words. Because they come in the corpus from English, I have included them in my investigation (see López-Morales 1987: 303;Balteiro 2011: 27 As already mentioned, nouns are not very often used alone. For as much as they tend to describe concepts, they are often described by word combinations. In some cases, the result is a compound noun, in others -a collocation, noun phrase (NP). The boundary between them is not clear enough -it is claimed to be a semantic one -if the word combination has its own meaning and represents a new concept, it is a compound. Nouns are usually extended by another noun (as above), or an adjective. Sometimes the whole noun phrase is used -Noun + Noun (6 a., b., c., d.) or Adj.+ Noun (7) The examples above (7 a., b., c.) show the tendency of noun phrases with attributes without suffixes to be used. As it is known, nouns are often modified by other nouns or adjectives. A noun in the position of a modifier in postposition is an active model in Bulgarian. However, it has not been very popular until recently. These days, because of English, one can see a great number of new combinations, though the modifier appears in the pre-position. Sometimes the whole phrase is from English, in others -only the modifier ('Макларън количката' [Maclaren stroller], 'неорганик матрачета' [non-organic mattresses]). It is problematic when they are made of more than one word, or they are compound words or nominative phrases as it is also not clear where the boundary between lexicology and grammar stands. The language system is interconnected, and the change in vocabulary may affect the grammar as well. As it is known, the close relationship between nouns and adjectives was noticed a long time ago. Alfieri (2015) states that in the Greek-Latin parts of speech theory the difference between adjectives and nouns is not significant. According to Dionysius Thrax's Τέχνη γραμματική "[t]he noun is divided into three types: 1) ὄνομα κύριον 'proper noun', which refers to the individual substance (τὸ τὴν ἰδίαν οὐσίαν σημαῖνον); 2) ὄνομα κοινόν or προσηγορικόν 'common noun', which refers to the common substance (τὸ τὴν κοινὴν οὐσίαν σημαῖνον); and 3) the epithet as the third εἶδοs 'type' (Tekh. 12)" (Alfieri 2015: 363).
To sum up, the partial substitution of the adjectives with the nouns in Bulgarian is an important but not unpredictable phenomenon having in mind the closeness between nouns and adjectives, the existence of such a construction in Bulgarian, as well as the worldwide spread of English.

Adjectives
75 adjectives are excerpted from the corpora. More of them are modifiers of nouns, and some of them are a part of compounds. Some of them are a part of NP (10 a, b, c), others are a part of the predicate or VP (10 d, e). Some of them are grammatically adapted to Bulgarian; others are used in their original form. Gender and number are inflectional features of Bulgarian adjectives that mark agreement with nouns. Therefore, English and Bulgarian adjectives are grammatically different. It is demonstrated in example d.the participle 'обявена' is in its feminine form. However, the English word 'selfish' is gender-neutral. When English adjectives appear in Bulgarian texts morphologically adapted they sound natural. However, only 14 adjectives in the corpus (19%) have Bulgarian suffixes and inflexions at the end of the word (евалю-иран-а F 'evaluated'; джетлаг-нат-и PL 'jet-lagged', сензитив-ен 'sensitive', etc.). Others appear unchanged and in Bulgarian texts, they become grammatically close to adverbs or nouns. 62 out of 75 adjectives, or 83% are written in Cyrillic, and 13 out of 75 or 17% are in Latin (Table 4.). All Cyrillic forms are transcribed, which prove their persistence in the users' speech. None of the forms are transliterated.

Verbs
The verb in Bulgarian includes the full paradigm of person and number (in some cases -when participles are used in the verb forms, the category of gender is activated as well). The other verb categories expressed morphologically are aspect, tense, voice, and mood. There are also a great number of perfect-like forms expressing evidentiality, mirativity, etc. In English, one and the same word may function as a noun or as a verb, and the context is determinative for proper reception. In Bulgarian, the forms of verbs and nouns are completely different because of dissimilar inflexions. This is why it is difficult to use English verbs in Bulgarian speech without adaptation.
This might be the reason why there are few English verbs in the corpora. The extracted forms are 33. 27 of them (or 82%) are morphologically adapted to Bulgarian. As one may suppose, most of them are in their most neutral form -Imperfective aspect, 3 rd conjugation (10 a., c.), Indicative, Present tense. To conclude, the English verbs in the investigated corpus are not numerous. In most cases, they are inflected like Bulgarian verbs. The used suffixes and inflexions are in concordance with the tendency of the newlyformed verbs in Bulgarian to be of 3 rd Conjugation, Imperfective aspect. These verbs sometimes build a Perfective form of 1 st Conjugation.
Some verbs present actions connected to the computer communication and they are widely used by Bulgarians in their native milieu as well (especially younger generations), i.e. "пействам", "спамя", "хъртвам". Others are used accidentally for various reasons -because the verb is a part of a longer expression, because the combination sounds better to the user or because this is the word that comes to their mind first.

Other parts of speech
The adverbs in the corpus are very few -only five: авъридж (average), precisely, ълонг (along), eвър (ever), евентюали (eventually). All of them except for 'precisely' are in Cyrillic and transcribed.
At the end, I would like to mention that there are some other expressions from English. They are formed as inseparable phrases or sentences, and they express emotions or attitude. They are borrowed as ready-made clichés. Some of these phrases and abbreviations are very popular in all types of informal Internet communication: 'WTF!?!?'; 'Not!', 'тнкс' (thanks), WOW (of all nights!); 'Со уот!'(So what!); 'хепи фрайдей, лейдис!' (Happy Friday, ladies).
To sum up, the excerpted language uses represent the process of the interrelation between Bulgarian and English in Bulgarian speech of people who live in the USA. While several words and expressions could not be found in the speech of Bulgarian people living in Bulgaria, the investigated examples conform to the general process of active adoption of English words in the contemporary Bulgarian language, especially in the fields of computer communication, finance, jurisprudence, etc. Most of the borrowed words are nouns or noun phrases as they are easily adapted to other languages. Moreover, nouns designate things, and in everyday life, they could quickly become a part of some people's vocabulary. At the same time, sometimes English words are used to make the message more interesting and impressive.

Conclusion
The contact between languages is one of the reasons for language change. Some phenomena like emigration/ immigration, world globalization, Internet access, etc. 'catalyze' the language interaction and determine the expansion of English all over the world.
Millions of Bulgarians have emigrated since 1990. Some of them have been abroad for a short time. Others stay longer, and they see their future outside Bulgaria. Nevertheless, almost all of them need to communicate in Bulgarian from time to time. It is not surprising that some of the most active blog commentators as well as many forum members live abroad, but they prefer to use their native language on the Internet. However, they use more loanwords because of their everyday contact with other languages. As English words become popular in every language in the world, they enter the vocabulary of more and more people. Therefore, the English language influences other languages because it changes the individual language use of people who are in touch with it.
All these points could be found in the corpus. There are some examples of orthographical interference as digraphia is observed in several cases. However, since the forum rules require the use of the Cyrillic alphabet, the use of transcribed words from English is more frequent. Some words are used in Bulgarian context for the first time; other excerpted words have already been used in Bulgarian with fractions of their meaning and as a part of a limited number of collocations. Because of the influence of English, other meanings of these words are activated, and one may see different collocations as well as new word combinations.
Since the 1990s a large number of new words in Bulgarian have come from English. The loanwords are usually nouns because of their simple grammatical adaptation. The Bulgarian-English interaction process, however, leads to changes in the noun phrase. Nouns are usually used as modifiers in place of adjectives. This is a significant change -the grammatical system is conservative and difficult to change.
The verbal group is not so noticeably affected. The Bulgarian verbal system is very complicated and English verbs used in Bulgarian context are grammatically adapted. There are some exceptions when people use ready-made constructions such as idioms and phrasal verbs.
In the investigated corpora there are some adverbs, as well as some inseparable expressions such as abbreviations, phatic functioning phrases, etc.
In conclusion, as one might expect, the Bulgarian speech of people who live in the USA is more influenced by English than the speech of Bulgarians in their native milieu. Nevertheless, some of the characteristics mentioned above could be seen in the speech of Bulgarians in Bulgaria too. The interaction between English and Bulgarian changes the vocabulary of Bulgarian, but this is a global phenomenon. The continuing influence of English on Bulgarian is expected to keep the tendency of loaning new words and models from English active for a long time. Therefore, it will be very fruitful to observe the comparison between these results and those from some future investigations.