Volume 48, No.4 - Winter 2002
Editors of this issue: Violeta Kelertas
ISSN 0024-5089
Copyright © 2002 LITUANUS Foundation, Inc.


University of Southern California


Numerous studies have documented changes that languages undergo in emigration. Researchers have studied Norwegian spoken in the US, Spanish in Los Angeles, and German in Pennsylvania, just to mention a few. They all noted that languages in emigration undergo reduction in grammatical complexity as compared to the languages spoken in their countries of origin (Seliger and Vago 1991, Campbell and Muntzel 1989, Silva-Corvalan 1994). The fascinating question remains: is it all simplification and loss of linguistic system as a result of decrease in language use? Can it be explained as language acquisition in reverse, or is it more complex than that? In this paper, I look at the Lithuanian language spoken in the United States and argue that language attrition cannot be explained by simplification and reduction alone. More interesting changes are occurring as a result of the new circumstances the users of those languages find themselves in: using the language in much more restricted circumstances (primarily at home and with a limited number of friends and relatives) and living alongside another more dominant language, in this case English.

Data and methodology

The data used in this study consists of a corpus of 20,000 words of interviews conducted by me with speakers of Full Lithuanian (FL) spoken in the old country and second-generation American Lithuanians (AL) on similar topics of local news, events and activities. The interviews with speakers of FL were recorded in the summer of 1998 in Vilnius, and the interviews with AL speakers were recorded in St. Casimir's Lithuanian Saturday school in Los Angeles in the spring of 1998.

For purposes of this study, I examined demonstrative pronouns as used in the two language varieties, Full Lithuanian and American Lithuanian, and focused on the discourse level effect of their usage. The number of demonstrative pronouns in FL is much larger than represented here. I discuss only those that showed more than one instance of occurrence in American Lithuanian. The demonstrative pronouns that were found in Full Lithuanian but not in AL are not discussed here for lack of comparison. I first examine the pronominal system in FL and then compare it to what I found in the spoken data of AL.


Ambrazas (1997:95) lists three functions of demonstrative pronouns in Lithuanian:

(1) to refer to a definite thing (person, phenomenon): tas 'that,' tas pats 'the same.'
(2) to refer to a definite property of a thing (of a person or phenomenon): toks 'such.'
(3) to refer to a situation: tai 'it,' viskas 'everything,' tas pats 'the same.'

In Full Lithuanian, demonstrative pronouns used in functions 1 and 2 usually have anaphoric reference to an antecedent noun. For example1:

Iš tėvo jis gavo šiek tiek pinigų, tie jam labai pravertė.
'He got some money from his father, it came in very handy.'

In this case, the demonstrative pronoun tie 'it' refers to its antecedent pinigų 'money.'

Demonstrative pronouns in functions 1 and 2 can be used in adjectival position, i.e., before the noun, as well. In this manner, they contribute definite status to that noun. For example:

tėvo jis gavo šiek tiek pinigų, tie pinigai jam labai pravertė.
'He got some money from his father, that money came in very handy.'

Here the demonstrative pronoun tie 'that' modifies the noun meaning money.

Only situational demonstratives, which are used in function 3 to refer to a situation, often have a clause, a sentence, or a larger segment of discourse as their antecedents. Whereas demonstrative pronouns used in functions 1 and 2 can be used in nominal and adjectival positions, as we have just seen from the above examples, situational demonstratives occur only in the nominal positions. For example:

Nežinau, ar gali žmogus savo laimę atspėti. Negalvojau apie tai.
'I don't know if a person can guess his lot. I haven't thought about it.'

The demonstrative pronoun tai 'it' refers to the whole sentence preceding it, i.e., the fact that the person had not thought whether one can guess his lot.

Demonstrative pronouns in nominal position (i.e., used by themselves) perform a discourse function and facilitate cohesion within a text. They help to avoid excessive repetition of a noun phrase: a demonstrative pronoun in an utterance links what is being said at the moment to previous discourse.

In contrast to Full Lithuanian, where only situational demonstratives refer to larger pieces of text, American Lithuanian uses other demonstrative pronouns in this function. As a rule, demonstrative pronouns used in the American Lithuanian text do not have a concrete single noun as antecedent, but rather refer to what has been said before, whether it is a phrase or a longer utterance larger than a sentence. In some instances, the demonstrative pronouns refer to the discourse topic, which may not even have an explicitly mentioned referent. With such practice, speakers of American Lithuanian create a sort of "false" cohesion and, instead of being specific in their references, look for acceptance of the entire proposition on the part of their interlocutor, perhaps to establish solidarity with that interlocutor. As a result, the overall cohesiveness of spoken American Lithuanian texts differs from that of spoken FL. For example:

(1)    1. vaikai gauna namų darbus ir taip toliau,
        2. turi prižiūrėti, kad vaikai atliktų tas pamokas ir ką nors išmoktų,
        3. tai nuo tėvų priklauso tas, kad vaikai atliktų ,
        4. tai tėvų komitetas skatina
        5. ir bando tėvus laikyti sujungusius tuo,
        6. kad žinai būtų tas susidomėjimas ir atsakomybė,
        7. kad jie atliktų.

        1. 'children get homework and so on,
        2. you have to supervise that children do that homework and learn something,
        3. so that depends on the parents that children do that,
        4. so the parents' committee encourages that
         5. and tries to keep parents united by means of that,
         6. that you know there would be that interest and responsibility
        7. that they would do that.'

In example 1 I have underlined all instances of the demonstrative pronoun tas 'that' found in an interview of AL. Tas is used in FL to refer to a definite thing (Function 1).2 The frequency of demonstrative 'that' in example 1 is impressive and deserves an explanation. Let's look at cases where 'that' occurs in the nominal position (i.e., by itself); such cases are marked in bold. The first instance occurs in Line 3 and is repeated in Line 4. The referent of the first demonstrative pronoun in Line 3 is in Line 2, where 'that' refers to 'children do[ing] that homework and learn[ing] something [from the Lithuanian Saturday school].' The referent of tą 'that' in line 4, however, is ambiguous. It is not clear whether the speaker is referring to 'the children doing the homework' (information provided in Line 2) or to 'parents supervising the children' (information added in Line 3). Again, in Line 7, it is ambiguous whether 'that' refers to "the parents' role in children's performance at school" or to "the children performing their school responsibilities." In addition, the subject 'they' in Line 7 is ambiguous with respect to its referent: we do not know whether the speaker is referring to the parents or to the children; the question remains, "it is important that who does what?"

In other words, by using demonstrative pronouns to refer to larger portions of text or even concepts that have never been explicitly articulated in the discourse, American Lithuanian speakers do not create explicit references, but rather rely on confirmation from their listeners that they accept the speaker's entire proposition. In this manner, the speaker checks with the hearer and relies on the hearer's cooperation in interpreting what has been said previously (or even what was meant, but never articulated) without actually going back and reiterating or making specific references to specific points in question. To illustrate this, let us look at one instance of such evasive reference in the sociolinguistic interview in AL:

(2)     ...tai mums tie žodynai labai brangūs [...]
        Pavyzdžiui kur mes gavom vieną,
        nėra anglų kalba tik visas lietuviškas,

        bet gauni tą žodį ir išaiškintas jis yra,
        ir yra visokie formatai viską,
        kaip tą žodį vartoti, kaip sakyti, kaip tarti,
        tai tas yra labai gerai,
        ir... tokį tokį mes labai vertinam.

' for us those dictionaries are very valuable [...]
For example, when we received one,
there is no English, only all Lithuanian,
but you get a word and it is explained what it is,
and there are various formats everything,
how to use that word, how to say it, how to pronounce,
so that is very good,
and... such such we value a lot.'

Let's look at the demonstrative pronoun tokį 'such.' Toks is generally used to refer to a definite property of a thing in FL (Function 2). In 2 it is used twice in the last line, but it does not refer to the concrete noun introduced in Line l, žodyną, 'dictionary,' (because žodyną is masculine and the demonstrative is used in the feminine form). Rather, it refers to the entire passage above it. In fact, tokių here translates better as 'of such kind,' referring to a larger discourse topic, Lithuanian teaching material (the noun 'material' being assigned feminine gender in Lithuanian) that speaker is describing, that is, 'the kind of teaching material that the Lithuanian Catholic school is looking for.' From the perspective of a Full Lithuanian listener, such avoidance of explicit references in American Lithuanian can create a sense of evasiveness, inexplicitness and reliance on the speaker's cooperation in accepting "all that has been said" before.

To illustrate the contrast between such inexplicit reference links in American Lithuanian and the reference links created by means of the same demonstratives in Full Lithuanian, let us look at a Full Lithuanian spoken text:

(3)    1. kažkokios ten avangardistinės ten tipo pjesės,
        2. kur nežinau
        3. ten užsieny turbūt kiečiausi avangardistai tokių nežino...

        1. There are some sort of avant-garde type plays,
        2. where I don't know,
        3. probably the toughest avant-gardists abroad do not know such.'

In Example 3 the same demonstrative pronoun toks 'such' is used as in Example 2. However, in this case, the demonstrative pronoun tokių GEN:PL 'such' in Line 3 refers to what has been said in Line 1 and has a concrete antecedent noun phrase meaning 'plays'; one could paraphrase Line 3 into 'the toughest avant-gardists do not know such plays.' In this manner, this demonstrative pronoun is different from the demonstrative 'such' used in Example 2, where no concrete referent can be established as its antecedent. In Example 3 the demonstrative tokių 'such,' third person plural form of toks 'that,' is used to refer to a definite property of things and has an antecedent noun pjesės 'plays.' It can also be used in adjectival position, as in tokios pjesės 'such plays.'

Spoken American Lithuanian does not make these referential distinctions and uses any of the demonstrative pronouns irrespective of their different functions to refer to a unit larger than a noun phrase, as was shown in Examples 1 and 2. Because, in most cases, these demonstrative pronouns do not have a specific noun phrase as their antecedent, they can occur only in nominal position. In this manner, they are similar to situational demonstrative pronouns in FL. Instead of referring to a situation, however, demonstrative pronouns in AL refer to a larger discourse topic and in this manner select 'all that has been said' as a point of reference.

These demonstrative pronouns in AL discourse serve a function similar to the 'You know?' tag in colloquial spoken American English,3 and it can be interpreted as 'Will you accept all that?' What American Lithuanian speakers are doing here is seeking confirmation from their interlocutors, 'Do you share the same presupposition?' 'Are we on the same wave length?'


The data show that demonstrative pronouns in spoken AL are developing a new discourse function, i.e., to avoid the use of explicit references and to rely on the interlocutor's cooperation to make inferences about what has been said and, by doing so, to check with the interlocutor for approval and acceptance of the entire proposition, similar to English 'and all that' or 'You know what I mean.'

Recent studies on language attrition have documented cases of linguistic creativity in terms of innovative word formation (Gal 1989, Dal Negro 1998). Interestingly enough, both Gal and Dal Negro note that the observed changes take a solidarity or in-group-defining function. The development of a new discourse function of demonstrative pronouns in American Lithuanian provides further evidence of linguistic creativity at a discourse level. The demonstrative pronouns used in the nominal position serve to establish an in-group solidarity with the listener. These findings underscore the complexity of language change. Furthermore, the evidence of linguistic creativity speaks against the notion of describing language attrition solely as an acquisition process in reverse and provides a new dimension for the study of attrition in addition to the universally noted phenomena of reduction, simplification, and loss.


1 Examples used to illustrate the different functions of demonstrative pronouns in FL are taken from Ambrazas (1997: 195-197).
2 The pronoun tas 'that' takes different forms, depending on its grammatical case.
3 I owe this observation to Prof. John Rohsenow at the University of Illinois at Chicago (personal communication).


Ambrazas, Vytautas (ed.) 1997. Lithuanian grammar. Vilnius: Baltos lankos.
Campbell, Lyle and Muntzel, Martha C. 1989. "The Structural Consequences of Language Death." Investigating Obsolescence: Studies in Language Contact and Death, ed. by Nancy Dorian, 181-97. Cambridge: Cambridge University Press.
Dal Negro, Silvia. 1998. Language Shift and Change in a Walser Dialect in Italy. Paper presented at Third International Conference on Maintenance and Loss of Minority Languages. Veldhoven, The Netherlands.
Gal, Susan. 1989. "Lexical Innovation and Loss: the Use and Value of Restricted Hungarian." Investigating Obsolescence.: Studies in Language Contact and Death, ed. by Nancy Dorian, 313-31. Cambridge: Cambridge University Press.
Macevichius, Jolanta. 2001. "The Beginnings of Language Loss in Discourse: A Case Study of Referentiality in American Lithuanian." Sociolinguistic and Psycholinguistic Perspectives on Maintenance and Loss of Minority Languages, ed. by Ammerlaan, T.; Hulsen, M.;
Strating, H; and Yagmur, K., 235-249. Waxman: Munster/New York.
---------. 2000. The Beginnings of Language Loss: A Discourse Study of American Lithuanian. USC dissertation.
Seliger, Herbert W., and Vago, Robert M. 1991. "The Study of First Language Attrition: an Overview." First Language Attrition, ed. by H.W. Seliger, and R.M. Vago, 3-17. Cambridge; Cambridge University Press.
Silva-Corvalan, Carmen. 1994. Language Contact and Change: Spanish in Los Angeles. New York: Oxford University Press.


Artwork of Vytautas O. Virkau