Bilingual verb constructions in southwestern Spanish.
The Spanish of the southwestern United States employs a unique bilingual verb paradigm containing the auxiliary hacer (do) with an English bare infinitive. The discourse motivators for this construction follow what has been reported by other scholars on language contact (e.g. Zentella 1997 and Silva-Corvalan 1983), but the syntactic structure belies any sense of "typical" borrowing or code-switching. While the geographic distribution of this phenomenon appears to be limited to sections of the southwest United States (unlike other similar bilingual phenomena) and its explanation as a product of English or Spanish syntax proves problematic, parallel constructs occur in other language-contact situations throughout the world. The hacer verb is a morphosyntactic dummy auxiliary in these constructions, yet its repeated presence in different bilingual periphrases appears to be a function of its lexical nature.
Discussions on verb usage in Spanish/English bilingual settings generally focus on three types of bilingual strategies: calquing, verbal borrowing, and code-switching. Calquing involves the use of an existing word in a new semantic environment, such as "correr para presidente," taking the English meaning of the verb "to run (for office)" and extending it to the Spanish verb correr, which in monolingual Spanish does not carry such meaning.
Borrowing generally occurs when the English verb is phonologically and morphologically adapted to Spanish in cases like taipear (to type), huachar (to watch/see), liquear (to leak), mapear (to mop), etc. Valdes (1988) notes that verbal borrowing can also occur without adaptation in cases where the English verb conveys a more direct meaning, as is the case with weld, since soldar can mean "to solder" as well as "to weld":
1. Necesito weld esos fierros. (Valdes 1988:123) (I need to weld those pieces of metal.)
While on the one hand this example illustrates a borrowed verb from English, the fact that the verb is unincorporated makes this sentence an example of code-switching, i.e. switching between Spanish and English, in its simplest form. While code-switches can occur at the word, phrase, or sentence level for a variety of reasons (see Zentella 1997, Silva-Corvalan 1983), when they occur as a single, word-level switch it is often because the verb is unknown to the speaker or because it conveys a special meaning not expressed in Spanish, as is the case with weld, or specialized English-language terms such as hanging out:
2. No tengo tiempo ni para estar hanging out con los amigos. (Valdes 1988:125)
(I don't even have time to be hanging out with friends.)
While cases of bare English verbs are not uncommon in Spanish/English bilingual communities, this paper will focus on a structurally unique form that also appears to be relatively limited in its distribution. In the Spanish of the southwest United States, hacer (do) is used as an auxiliary verb with an English bare infinitive to form a bilingual periphrastic construction:
3. Tengo otro hermano que hizo RETIRE.
(I have another brother who did RETIRE.
(I have another brother who retired.)
In this bilingual compound verb, the infinitive serves only to carry lexical value, whereas all of the verbal morphological markers (tense, mood, aspect, subject) are found in the auxiliary hacer. This type of construction does not fit into the general definition of borrowing since the English infinitive is not phonologically adapted. Additionally, labeling this as code-switching does not paint the whole picture since the morphological markers are carried by the Spanish auxiliary hacer and the "switch" does not happen before or after the verb, but rather within the verb structure. Finally, given the fact that hacer does not function as a non-causative auxiliary in monolingual Spanish, this innovative verbal paradigm is truly unique.
The examples in this study come from interviews with fifteen bilinguals from northern New Mexico and southern Colorado as part of the New Mexico and Southern Colorado Linguistic Atlas and Survey (Bills and Vigil 1994). The informants who produced the verb structure in question range in age from 25-51, though they represent a much larger group of interviewees aged 18-96, perhaps suggesting that the oldest generation of bilinguals in this region do not employ this construction. This would not be all that surprising given that education in New Mexico was often provided in Spanish in the earlier decades of the twentieth century (Grosjean 1982). The upper age range of those who employ the periphrastic construction appears to correspond to the first generation of speakers who were punished in school for speaking Spanish. As we examine the lexical categories of the verbs in question below, we will see that the scholastic context is relevant.
It appears that the hacer + infinitive construction is not present in other Spanish bilingual communities, but rather is exclusive to the Southwest region. It has been greeted with surprise or initial rejection by bilinguals, including linguists, from eastern U.S. Spanish-speaking groups, including Puerto Ricans, Dominicans, and Cubans and may even be limited in its geographic scope within the Southwest. Reyes (1982) notes that this structure is acceptable in the speech of bilingual Chicanos in Oakland, California; Pfaff (1982) also mentions California and indicates that "a number of University of Texas students report that they find the construction grammatical but would not use it themselves" (1982:276).
The apparent limited geographical/dialectal distribution of this phenomenon stands in sharp contrast to many Spanish/English bilingual constructions, e.g. pa(ra) atras as a translation of the English particle back, as in llamar patras (call back) or venir patras (come back), which is present in virtually every Spanish/English bilingual community in the world, including linguistically isolated communities such as Gibraltar and that of the Islenos in rural Louisiana (Lipski 1985).
The motivations behind the use of this structure are not new to the study of English/ Spanish language mixing. We find in the data of the present study that these bilingual verbs occur after repair phenomena, in English-language domains, in "new" verbs, and in low-frequency lexical items. Silva-Corvalan (1983) mentions code-switching to fill in memory lapses (indicated by pauses) and to compensate for lack of competence in English. Zentella (1997) similarly indicates a large percentage in her own data of speakers switching to English because they did not know or could not recall a term in Spanish.
In some, but not all, cases in the present data, the switch occurred after repair phenomena. In example 4, the speaker attempted to utter the Spanish verb for correct, only to doubt the formation, pause, and rephrase the verb with the hacer periphrastic construction. Example 5 occurred following a pause and repetition of the clitic object pronoun.
4. Me, me corre ... me hacia CORRECT a mi.
Me, me (she) corre ... me she did CORRECT me.
(She, she, corre ... she corrected me.)
5. Si hablabamos espanol nos ... nos hacian PENALIZE.
If (we) spoke Spanish us ... us they did PENALIZE.
(If we spoke Spanish they ... they penalized us.)
The above examples, as well as the two that follow, all occur within English-language domains for the speaker. Bilinguals generally have compartmentalized linguistic repertoires based on the language of their experience. In the case of Spanish/English bilinguals in the United States, these speakers often find that in their home and community settings they function quite well in Spanish while their occupational and educational environment is often monolingual in English. This dichotomy contributes to the incorrect assumption by many second- or third-generation bilinguals that their own Spanish is inadequate, incorrect, or even slang, given that their Spanish does not serve them in educational or professional settings. In fact, in each of the first four examples in this section (4-7) the speakers discuss some aspect of their (often negative) school experience in an era when Spanish was strictly prohibited in school, in a sense "footing" in English to discuss this uni-dimensionally monolingual aspect of their fives (see Zentella 1985, 1997).
6. Me estaban haciendo PUNISH.
Me they were doing PUNISH.
(They were punishing me.)
7. No nos hacian ENCOURAGE.
Not us they did ENCOURAGE.
(They did not encourage us.)
Along the fines of English language domains are the following utterances that contain, in effect, "new" verbs to the particular speech community in this study, as illustrated in examples 8 and 9.
8. Tengo que hacer VACUUM.
(I) have to do VACUUM.
(I have to vacuum.)
9. Le digo que haga RECORD las oraciones de estos viejitos.
(I) tell him that do RECORD the sentences of these old folks.
(I tell him to record the sentences of these old folks.)
Traditional Spanish speakers in New Mexico never had items like vacuum cleaners or recording devices until after English permeated the community; the lexical repertoire of the speakers is effectively halved with regard to these lexemes. The only way that the Spanish notions of pasar la aspiradora or grabar would have entered into the speech of these informants would be through some type of educational experience in Spanish, the likes of which we have already seen to have been rather limited in their formative years.
In addition to those verbs that come under the purview of English-only experiences, the hacer periphrasis also occurs with low-frequency verbs. These items may or may not be present in the Spanish lexical inventory of the informants, but given their relatively low frequency in quotidian speech the Spanish counterparts of these verbs are simply unavailable to the speakers at the time of the utterance. This is frequently cited as one of the main reasons for a switch: "The large percentage of shifts that occur after some editing phenomena appears to indicate that the speakers are trying to avoid the shift. The fact that the shift still occurs shows that certain lexical or syntactic constructions are not available to the speakers in Spanish, so they resort to English" (Silva-Corvalan 1983:82). There is no functional need for a bilingual to shift to English for a frequently used verb. Reyes notes the following:
Certain English items are never borrowed in this way. For example, we cannot say *hacer eat 'to eat', *hacer go 'to go', *hacer sit down 'to sit down' ,*hacer see 'to see', *hacer walk 'to walk', *hacer drink 'to drink' It seems that the Spanish equivalents of most of the borrowed forms of this type belong to a more formal style typical of Chicano speech. Thus, terms such as mejorar 'to improve', dibujar 'to draw', retirar 'to withdraw,' dirigir una peticion 'to petition', have a lower frequency in Chicano speech than comer 'to eat,' ir 'to go,' sentarse 'to sit down,' ver 'to see,' andar 'to walk,' and tomar 'to drink.' (1982:162)
Hacer + Infinitive: Noun or Verb?
Whether this construction suggests a nominalization strategy or an integration of the English infinitive into a bilingual compound verb is a topic for discussion. While in example 8, given the absence of any direct (or indirect) object, one could make a case for nominalization, our remaining examples thus far suggest a compound bilingual verb. In some cases, a patent direct object is present, thus leaving no doubt as to the transitive verbal position of the infinitive, such as in example 9 (as well as examples 4 through 6). In examples 10 and 11, the presence of the reflexive and direct object pronouns preclude placing a nominalized infinitive in that position.
10. Se hacian SABOTAGE each other.
Each other (they) did SABOTAGE each other.
(They sabotaged each other.)
11. Cuando mi abuela se murio, la hizo INHERIT mi mama.
When my grandmother died, it did INHERIT my mom.
(When my grandmother died, my mum inherited it.)
Some cases are more ambiguous. In example 12, rhyme could be construed as a (nonverbal) noun or a verb, since hacer rima is a viable form in Spanish. However, it is equally possible that rhyme is a verb, as it follows the same pattern as the rest of the examples in this study.
12. Tenia como ... algo que hacia RHYME con su nombre.
(She) had like ... something that did/made RHYME with her name.
(She had like ... something that rhymed with her name.)
There are certainly clear-cut examples of nominal use in the present data. In examples 13 and 14, hacer is followed by a noun, but with a corresponding verbal meaning.
13. Me hice un fracture en la nariz.
(I) myself did a fracture in the nose.
(I fractured my nose.)
14. Yo no hacia ... interruption.
I not did ... interruption.
(I didn't interrupt.)
These two examples are included here to show a distinction, yet at the same time they represent more than a simple case of noun borrowing, since the switch occurs at the point of the verb and the corresponding meaning is verbal. Neither in Spanish nor in English does one hacer una fractura (do/make a fracture) or hacer una interrupcion (do/make an interruption). Garrett (1998:310) suggests that "as objects of lexical do, bare action nouns are semantically already verbs." This appears to be the case here. In example 14, it appears that the speaker was searching for an appropriate verb phrase, then settled on interruption to complete it.
Such constructions are not unique to this community nor to bilingual communities, for that matter. In monolingual Spanish communities, it is not uncommon for one to hacer autostop (hitchhike), hacer camping (camp) or, more recently, hacer clic (click), where the nominal counterparts have been phonologically adapted to form a resultant verbal compound.
There are, of course, cases of simple noun switches following hacer. The following examples illustrate a simple switch from Spanish to English and are not verbal in nature. These cases serve as a comparison with the preceding examples in the sense that there is nothing grammatically or semantically awkward about "doing construction" or "doing penance." Here the verb hacer serves not as an auxiliary but rather as a lexical verb, carrying semantic weight (minimal as it may be) as well as syntactic value.
15. Consiguio trabajo en Colorado haciendo construction.
(He got work in Colorado doing construction.)
16. En su religion hizo penance; lena la Biblia.
(In her religion she did penance; she read the Bible.)
If we consider similar constructions where do is not present (e.g. tengo que vacuum), no debate is necessary. These cases are clearly switches at the point of the verb, with no hint of the verb functioning as a direct object. By adding the hacer verb, the infinitive does not shift to an object position, but rather is incorporated into the bilingual compound, providing the semantic counterpart to the syntactic auxiliary.
Similar Structures in Other Languages
While we have already seen that the hacer periphrasis seems limited in its geographic scope with regard to Spanish/English bilingual communities, do as a bilingual auxiliary appears to be widespread in other language contact situations.
Japanese employs a compound "light verb" in which the verb suru (do) is coupled with a verbal noun, generally of Chinese origin (Matsumoto 1996:109). Korean apparently employs a parallel construction using a Sino-Korean predicate with the Korean verb ha ta (do) (Choi and Wechsler 2001: 103). While these verbs employ a Chinese predicate, they are not restricted to use by bilinguals but rather are common in monolingual speech.
The do periphrasis appears to be common in several English-dominant bilingual communities. Japanese/English bilinguals utilize the light verb construction between English and Japanese in phrases such as "operate suru" (Muysken 2000:196). Moravcsik (1975) shows a similar pattern in American Greek in which the "do" verb kani is employed with an English bare infinitive:
17. O Petros kani RETIRE.
The Petros do retire.
(Petros is retiring.) (Moravcsik 1975:5)
A similar example is found in American Portuguese with the auxiliary fazer (do), though "sometimes the embedded verb is nominalized with the Portuguese masculine singular definite article o":
18. fazer o save "save"
19. fazer o find out "find out" (Muysken 2000:195)
As we attempt to determine the basis of the hacer + infinitive construction, there are several available options to explore. The first is to examine the language (i.e. Spanish) in which the formation occurs and the second to examine the other language of the bilingual community (in this case English).
Although hacer does not function as a noncausative auxiliary verb in monolingual Spanish, it does occur in causatiye periphrastic constructions. Reyes suggests that the bilingual formation may find its origin in the monolingual causative form:
The Chicano hacer + English bare infinitive seems to be based on a syntactic frame of the Spanish language. Chicano phrases like hacer improve, hacer reenlist, etc., are probably based on Spanish phrases like hacer llorar ... lograr mejorar ... etc., in which the governing verb always takes an infinitive complement. The function of the Chicano frame, of course, is different, since hacer in the Spanish phrase hacer llorar, for example, conveys the lexical meaning 'to induce,' whereas in the Chicano phrase hacer improve it merely expresses the inflectional endings that are normally expressed by the Spanish verb. (Reyes 1982:157-158)
This syntactic structure may look like the causative structure with hacer + infinitive, yet the examples in this study display no causative meaning whatsoever. Additionally, the causative construction and the bilingual periphrasis have nothing in common semantically. Given the relatively narrow dialectal distribution of the bilingual construct, it is unclear how such a semantic shift would have occurred; this is especially curious in light of the structural fact that the causative generally employs a personal dative object pronoun as the affected party, whereas the bilingual periphrasis does not necessarily, and in some cases cannot, use any such critic. Additionally, it is important to note that the informants for this study do use the causative construction in cases in which it would be expected, regardless of whether or not it involves code-switching, as in example 20:
20. Nos hacian escribir "I will not speak Spanish in class."
(They made us write "I will not speak Spanish in class.")
Given the dual linguistic repertoires of the speakers, another logical point of departure for considering the origin of the hacer + infinitive formation is English. Pfaff notes that "Spanish and English verbal grammar is identical in a significant way: both employ periphrastic verbal constructions in which unconjugated verb forms follow conjugated auxiliaries or complement-taking verbs" (Pfaff 1982:274).
English does indeed employ do as an auxiliary, though not in exactly the same way as the bilingual constructions in this study. In Modern English the auxiliary do is found in at least three declarative situations. First, it is employed in emphatic affirmations or as a substitute verb in responses to questions (which also use the auxiliary do):
21. "I do like green eggs and ham!"
22. "Do you like them?" "Yes, I do."
Secondly, and not unrelated to the preceding examples, the auxiliary do is used in sentential negation:
23. "I do not like them."
24. "I don't know."
Finally, we find do as a generic verb coupled with the gerund (i.e., the nominal form of the verb) to express ideas like "He does the gardening" or "I do some collecting." This third, generic use of the verb seems to most closely parallel the bilingual periphrastic structure by applying the lexical content of do in answering the question "What does one do?" One difference, however, is the inability of the English example to allow use of an object after the gerund, as in "*I do some collecting paintings." On the other hand, modification of the gerund through compounding is permissible in English: "I do some stamp collecting." Thus, this final use of do as an auxiliary is certainly closest to the bilingual periphrasis, although the syntactic question of the direct object goes unanswered.
It is interesting to note that examples of what appears to be a parallel development of do as a syntactic marker were common in Early Modern English:
Do as a meaningless element, the only structural function of which is that of a tense-carrier, developed in early Middle English, especially in the Southern dialects.... By [Early Modern English], do was used extensively in all types of writing; some authors use it more than others, but all use it. Most grammarians of the time fail to mention it, perhaps because it was redundant, or because it was so common that it was felt not to merit discussion. What occasional mentions there are of it fully support the contention that there was a meaningless do. For example, in Palsgrave (1530) we find the comment: "I do is a verbe moche comenly used in our tonge to be put byfore other verbes; as it is all one to say 'I do speke ...' and 'I speake....'" (Traugott 1972: 138)
In making such an observation, one need not imply a suggestion that the southwest Spanish construction is derived from the Early Modern English formation, since such a hypothesis would be problematic at best. The remnants of this earlier do + verb construction are evident in Modern English interrogatives and affirmations; however, the semantic difference is such that direct transfer is unlikely. Additionally, Traugott speculates that this Early Modern English dummy do was likely a derivative of the affirming do, a role not paralleled in Spanish by the verb hacer (or any other verb for that matter).
Historical use of the verb do as a syntactic marker also accounts for current past tense markers in Modern English. "Apparently, [the English past tense suffix -ed] was formed before the Old English period from the past tense of the verb do, which in Old English was dyde. In Old English, both the ablaut formation and suffixation existed with the same meaning" (Bybee, et al. 1994). Suffix formation cannot be the future of bilingual constructions given morphological language-mixing constraints (see Poplack 1982), but it is noteworthy that do played the role of syntactic marker in English from very early on.
While do does show to be a syntactic marker coupled with an infinitive lexifier from very early English, the case at hand is probably not one of simple evolution or derivation from that early construction.
What makes the parallel between Early English and Modern southwest Spanish relevant is the fact that do as a syntactic marker not only occurs in these two settings, but in other languages and language-contact situations. We have already seen that, in the case of bilingual verb structures, do as a syntactic marking auxiliary has been documented in other contact situations of Romance and non-Romance varieties. To some degree, this fact would appear to have theoretical foundations in a discussion on universal and/or bilingual grammars.
Despite the fact that the lexical value of the hacer + infinitive compound is contained in the infinitive, the motivation for the use of hacer in this structure may be lexical as well as auxiliary in nature. While hacer clearly functions as an auxiliary in the sense that it carries the syntactic value of the bilingual construction, a view of it as a lexical verb is not without merit. If we consider the fact that, in the bilingual examples given, the bare infinitive answers the question "?que hace/hizo/hacia/etc.?" ("What does/did/etc, one do?"), this may well be the explanation as to why the do verb is almost universally applied in bilingual verb constructs. With regard to the verb do, the Oxford English Dictionary explains that "since every kind of action may be viewed as a particular form of doing, the uses of the verb are as numerous as the classes of objects which it may govern" (OED, as quoted by Garrett 1998:301). In other words, do is the default verb. Since the semantic function of any non-stative verb relates to doing something, this verb becomes the logical choice for any such description. Its role as an auxiliary verb is a function of its lexical nature.
The discourse motivators for the hacer + infinitive verb structure are similar to those in other borrowing and code-switching contexts, yet this bilingual southwestern verb paradigm is distinctive for several reasons. First and foremost, while most borrowing strategies entail adaptation of the loan verb, the present case employs a bare form from English and a conjugated Spanish verb to create a bilingual compound. The structure in question is not a typical case of code-switching given the bilingual syntactic paradigm and its lack of equivalency in either language.
Why this construction is limited in its geographic scope remains a mystery. Other U.S. Spanish dialects appear to have chosen different ways to deal with English verbs, most notably morphophonological adaptation (which also happens in the Southwest), but why Southwestern Spanish employs the hacer periphrasis while others do not can only be speculated upon at this point.
While limited in its Spanish distribution, do has historically been employed in several languages, including English, as a dummy auxiliary. Not withstanding the fact that the lexical content of the paradigm is found in the English infinitive, it is the lexical nature of the verb hacer/do that encourages the use of this specific verb in these constructions. The fact that parallel structures occur in other language-contact situations suggests that this phenomenon has a place in discussions on bilingual syntax and universal grammar.
Bills, G., and N. Vigil. 1994. "New Mexico and Southern Colorado Linguistic Atlas and Survey." University of New Mexico.
Bybee, J., R. Perkins, and W. Pagliuca. 1994. The Evolution of Grammar: Tense, Aspect and Modality in the Languages of the World. Chicago: Univ. of Chicago Press.
Choi, I., and S. Wechsler. 2001. Mixed Categories and Argument Transfer in Korean Light Verb Construction. In Proceedings of the 8th International Conference on Head-Driven Phrase Structure Grammar, 103-120. Stanford, CA: CSLI Publications.
Garrett, Andrew. 1998. On the Origin of Auxiliary Do. English Language and Linguistics 2 (2):283-330.
Grosjean, F. 1982. Life with Two Languages. Cambridge: Harvard Univ. Press.
Lipski, John. 1985. The Construction pa(ra) atras in Bilingual Spanish-English Communities. Revista/Review Interamericana 15:91-102.
Matsumoto, Yo. 1996. A Syntactic Account of Light Verb Phenomena in Japanese. Journal of East Asian Linguistics 5:107-149.
Moravscik, Edith. 1975. Verb Borrowing. Wiener Linguistische Gazette 8:3-31.
Muysken, Peter. 2000. Bilingual Speech: A Typology of Code-Mixing. Cambridge: Cambridge University Press.
Pfaff, Carol. 1982. Constraints on Language Mixing: Intrasentential Code-switching and Borrowing in Spanish/English. In Spanish in the United States: Sociolinguistic Aspects, ed. J. Amastae and L. Elias-Olivares, 264-98. New York: Cambridge Univ. Press.
Poplack, S. 1982. Sometimes I'll Start a Sentence in Spanish y termino en espanol: Toward a Typology of Code-Switching. In Spanish in the United States: Sociolinguistic Aspects, ed. J. Amastae and L. Elias-Olivares, 230-63. New York: Cambridge Univ. Press.
Reyes, Rogelio. 1982. Language Mixing in Chicano Spanish. In Spanish in the United States: Sociolinguistic Aspects, ed. J. Amastae and L. Elias-Olivares, 154-65. New York: Cambridge Univ. Press.
Silva-Corvalan, Carmen. 1983. Code Shifting Patterns in Chicano Spanish. In Spanish in the U.S. Setting: Beyond the Southwest, ed. L. Elias-Olivares, 69-87. Rosslyn, VA: National Clearinghouse for Bilingual Education.
Traugott, Elizabeth Closs. 1972. A History of English Syntax. New York: Holt, Rinehart and Winston, Inc.
Valdes, Guadalupe. 1988. The Language Situation of Mexican Americans. In Language Diversity: Problem or Resource?, ed. Sandra McKay and Cynthia Sau-Ling Wong, 111-139. New York: Newbury House.
Zentella, Ana Celia. 1985. The Fate of Spanish in the United States: The Puerto Rican Experience. In Language of Inequality, ed. N. Wolfson and J. Manes, 41-59. Berlin: Mouton.
Zentella, Ana Celia. 1997. Growing up Bilingual Malden, MA: Blackwell.
Devin L. Jenkins
University of Colorado at Denver
|Printer friendly Cite/link Email Feedback|
|Author:||Jenkins, Devin L.|
|Date:||Sep 1, 2003|
|Previous Article:||The selected bibliography.|
|Next Article:||Gathering voices: storytelling and collective identity in Judith Ortiz Cofer's Silent Dancing: A Partial Remembrance of a Puerto Rican Childhood.|