Roots of English in the African American diaspora

Links & Letters 5, 1998 147-165 Abstract In this paper we use the world-wide variability in the past tense paradigm of the verb be (e.g. I/you/we/they was/were) to examine the similarities and differences across four geographically separated and ethnically diverse dialects of English spoken in North Preston, Guysborough Enclave and Guysborough Village in Nova Scotia, Canada, and Buckie in northern Scotland. Through comparative linguistic analysis of the distribution of forms across the verbal paradigm, we demonstrate unexpected parallelisms across three of these varieties. We conclude that these are the result of longitudinal continuity of the verb to be. The critical factor in explaining the similarities across dialects is their relative isolation from ongoing linguistic change in the English language. Key words: was/were variation, African Nova Scotian English, Buckie English, Continuity. Table of Contents 1. Introduction The verb to be is an intriguing item of the English language. A hybrid by any measure, its origins may be traced to two separate verbs, one for the present tense, Old English beon, ‘to exist’, and one for the past, wesan, from IndoEuropean vasati ‘to dwell’ (Lass, 1987). Today, it continues to be ‘a badly mixed up’ verb (Pyles and Algeo, 1993), and is widely-attested as exhibiting (*) This research was made possible through the generous support of the Social Science and Humanities Research Council of Canada (SSHRCC). We thank Salikoko Mufwene and Lesley Milroy for their helpful comments. 1. Introduction 2. Historical precursors of was/were variability 3. Contemporary research on was/were variability 4. Data and method 5. Results 6. Discussion References Roots of English in the African American diaspora* Sali Tagliamonte Jennifer Smith University of York Department of Linguistics September 1997 148 Links & Letters 5, 1998 Sali Tagliamonte; Jennifer Smith non-standard forms, not just in Britain (Cheshire, 1982; Trudgill, 1990; Henry, 1995), but in all the major countries where English is spoken: Australia (Eisikovits, 1991), the United States (Feagin, 1979; Christian et al., 1988; Schilling-Estes and Wolfram, 1994; Hazen, 1996), and Canada (Meechan and Foley, 1994). Consider the examples in (1): (1) 2nd person singular: a.You wasn't allowed to use their toilets. (NP:p:367.51)1 b.You were home (NP:p:512.53) 1st person plural: c. And we was the only colour family. We werewe were just surrounded. (GE:l:1591.7) d.We were all thegither...I think we was all thegither. (BE:h:72.44) Full NP plural: e. The kids was all here. (GE:l:124.20) f. My kids were all men and women before I moved here to Lincolnville. (GE:l:1006.8) Existential there: g. And there was nine years between me and my brother. (GV:¢:51.17) h. There were three girlsno, there were four girls. (GV:¢:16.12) Are these utterances instances of non-standard dialect, slang, or simply mistakes? On the other hand, might this represent yet another robust example of systematic linguistic variation typical of all languages? At least part of the answer resides in the more compelling question: how did varieties of English from locations all over the world get this way? The standard answer is that was/were variation is the result of ‘regularisation’ processes in language (e.g. Fries, 1940). This explanation is based on the idea that the verb to be is gradually becoming more like the other more regular verbs in English in having the same form, i.e. was, throughout the verbal paradigm, rather than the more complex distinction between 1st and 3rd person singular was, and 2nd person singular and 1st, 2nd and 3rd person plural were, as in (2). (2) a. See that the house was evenly halfed. (BE:g:408) b. We never did that in our days. The doors were open. (BE:b:265.51) But then where did the variation come from in the first place? One hypothesis is the standard diffusionist explanation, in which was/were was carried to these locations by people speaking varieties which contained the same featu1. The letters in brackets signal the name of the community — NP: North Preston, GE: Guysborough Enclave, GV: Guysborough Village, BE: Buckie English, and the speaker code. The numbers signal co-ordinates in the transcription. Roots of English in the African American Diaspora Links & Letters 5, 1998 149 res (cf. Weinreich et al., 1968). Another hypothesis argues against diffusion and suggests that this is the result of a more general tendency in all non-standard varieties of English to gravitate toward more primitive (i.e. not learned) linguistic patterns (Chambers, 1995). What type of linguistic analysis might shed light on the possible explanations for this variation? First, it is necessary to employ a method which can disentangle the many different influences on a linguistic variable through systematic comparison of its behaviour. In order to determine the status of a form, it is not its current existence in a variety which is decisive, nor even its rates of occurrence, but its distribution, i.e. precisely where it occurs in the language, as determined by the relative frequency of the feature across its contexts of use. Such evidence is particularly critical in order to assess the claims that was/were variation represents diffusion and/or ‘regularisation’ of the verbal paradigm. For example, precisely how is a given dialect regularised and to what degree? Crucially, it is necessary to situate the variation through comparison with the same features in related varieties (Rickford, 1977: 195). In other words, is the variation patterned in the same way across varieties? Is it embedded in the social strata in the same way? And perhaps most importantly, how can the variability be contextualized within the social history of the community (Mufwene, 1996)? In an attempt to shed light on these questions, this paper provides a consistent cross-variety comparison of was/were variability in four varieties of English. These have been selected because they represent polar extremes on a number of different extra-linguistic dimensions. They may be differentiated on three major criteria: their geographic location, their relative degree of isolation from mainstream developments in the English language and their ethnicity. Three are spoken in Nova Scotia, one of the maritime provinces on the east coast of Canada (see Map 1) —North Preston, Guysborough Enclave, Guysborough Village. The fourth is spoken in Buckie, on the north-east coast of Scotland (see Map 2). Although it is obvious that many linguistic changes have taken place in North American and British varieties of English over the last few centuries, the effects of different contact situations to such changes are not well known. This article contributes to ongoing comparisons between varieties of English in contrastive dialect situations. Because our data come from communities whose cultural heritage cross-cuts their social history, we are also able to address competing hypotheses about the origins of varieties of English spoken by people who do not trace their lineage back to the British Isles.2 2. This is a highly controversial topic in contemporary sociolinguistics. The extent to which varieties of English spoken in various locations in the world are ‘legitimate’ or ‘illegitimate’ offspring of English (Mufwene, to appear) or may be more appropriately viewed within a creole-origins framework (Rickford, to appear) is a matter of much debate in the literature. 150 Links & Letters 5, 1998 Sali Tagliamonte; Jennifer Smith Map 1. Map showing location of Nova Scotian communities. Map 2. Map showing location of Buckie. Guysborough North Preston HalifaxDartmouth NEW BRUNSWICK P.E.I. NOVA SCOTIA Buckie London Edinburgh Roots of English in the African American Diaspora Links & Letters 5, 1998 151 However, any attempt to explain the current state of was/were variation in English is enhanced by a time-depth perspective of the past, as the contemporary standard forms of the verb be have gone through many changes since the time of Old English. Therefore, we first consider the historical record. 2. Historical precursors of was/were variability Despite the dual etymological origins of the verb to be, the past tense paradigm derives from a single source, which through various phonological processes gave rise to the two contemporary variants — was and were. Standard English requires strict subject-verb agreement with 1st and 3rd person singular was and 2nd person singular and 1st, 2nd and 3rd person plural were. However, data from the historical record confirms that this has, in fact, always been variable (Jespersen, 1942; Forsström, 1948; Pyles, 1964; Visser, 1970; Curme, 1977), particularly the occurrence of non-standard was, as in (3). (3) a Thyrtty knyghtes ... Forsothe was in that companye. (c. 1300 Rich. Coeur de Lion (in Weber, Metr. Romances 1810) 95) b. This womans wordes was well harde. (c. 1523-5 Ld Berners, Froiss III 78) c. Thousands and thousands at that banquet was spent. (c. 1569 Preston, Cambises (in Manley, Spec. II) 199, 949) d. There was many Dukes, Erles and Barons. (Ld. Berners, Huon 2, 22, 1533) e. And there was in that tyme many gode holy men and holy heremytes (c. 1400 Mandev. 30, 30) [examples a-e from Visser, 1970] f. Indeed, when you was in the irreligious way, I should not have been pleased with you (James Boswell, London Journal, 1762-63) [example from Pyles and Algeo, 1993]. Despite a lack of consensus as to the explanation for was/were variability, the most salient factor identified in the literature is that the occurrence of socalled ‘non-standard’ was is not consistent across the verbal paradigm, but varies in frequency according to the person, number and even type of the subject.3 Moreover, careful reading of the extensive literature on was/were variability suggests a relatively consistent pattern differentiating these contexts. In Middle English with personal pronouns, the pre-eminent context for non-standard was was 2nd person singular you, but apparently only rarely with the plural personal pronouns you, we and they (Forsström, 1948; Mossé, 1952; Brunner, 1970). NP plural contexts also exhibited non-standard was in Middle English, while existential contexts, in which the NP follows the verb, are attested as far back as the Old English period (Quirk and Wrenn, 3. These and other constraints are dealt with elsewhere (Tagliamonte, 1997; Tagliamonte and Smith, 1998). 152 Links & Letters 5, 1998 Sali Tagliamonte; Jennifer Smith 1960; Visser, 1970). All uses of non-standard was during this time period are identified as a feature of northern England and Scotland in particular. ‘You was’ is reported in more southern regions from the 17th century onwards (Pyles and Algeo, 1993). While historians are not in total agreement as to the explanation for was/ were variability, not only is it clear that it existed in earlier varieties of English, it is also evident that it had specific patterns across the verbal paradigm (Forsström, 1948; Mossé, 1952). 3. Contemporary research on was/were variability One might expect that this variation has by now disappeared from contemporary varieties of English, particularly with dramatically increasing literacy (Chambers, 1997). Nothing could be further from the truth. In the last couple of decades there has been a plethora of studies on was/were variation (e.g. Feagin, 1979; Cheshire, 1982; Christian et al., 1988; Meechan and Foley, 1994; Schilling-Estes and Wolfram, 1994; Henry, 1995), all attesting to the continuing, and robust, variation between was and were. The foremost pattern has been the occurrence of was in contexts of were, providing support for the regularisation hypothesis. However, a number of studies demonstrate that were may also be the pivot for regularisation (Cheshire, 1982; SchillingEstes and Wolfram, 1994; Wolfram, 1997), suggesting that ‘regularisation’ processes may actually be far more complex than otherwise expected.4 Nonetheless, a prominent linguistic feature involved in the patterning of was and were is the person and number of the subject. Although different patterns are reported for different varieties of English, the personal pronoun you is consistently singled out as having a high degree of non-standard was, and first person plural we is also mentioned (Labov et al., 1968; Feagin, 1979). However, contemporary research focuses far more on 3rd person plural contexts, which exhibit differential frequencies of nonstandard was, depending on whether they are encoded with a personal pronoun, (e.g. they were), a lexical noun (e.g. the boys were), or an existential subject, (e.g. there were boys). There are actually striking cross-variety similarities with respect to the appearance of was or were across these contexts (e.g. Christian et al., 1988; Montgomery, 1989). Third person personal pronoun they is identified as the context which is least likely to exhibit non-standard was (Feagin, 1979; Eisikovits, 1991). In the case of third person NPs however, there is a clear distinction between contexts in which the NP is postposed to the verb, i.e. existential constructions, versus those in which the full NP occurs in subject position. Non-standard was is reported as more fre4. Researchers investigating were regularisation phenomena have suggested a remorphologisation of was/were across affirmative vs. negative constructions respectively (Cheshire, 1982; Schilling-Estes and Wolfram, 1994) Roots of English in the African American Diaspora Links & Letters 5, 1998 153 quent in the former and less frequent in the latter (Wolfram and Christian, 1976; Feagin, 1979; Christian et al., 1988; Eisikovits, 1991; Schilling-Estes and Wolfram, 1994). In sum, the cumulative findings from both the historical and contemporary dialectal record demonstrate the pervasiveness of was/were variation. However, this research does not entirely support the universality of regularised was, because some varieties demonstrate the regularisation of were. Moreover, the question of precisely which person and number was is regularised to, and in which linguistic circumstances, is clearly an issue to be grappled with. As we have demonstrated, the verbal paradigm is not equal in this regard. We have been able to identify specific and consistent patterns of variation, not just in the historical record, but also reflected in contemporary varieties of English. We now address a number of questions which arise from this research in our own data: 1. How frequently does was/were variation occur in the specific contexts across the verbal paradigm? 2. What are the similarities and/or differences in terms of where and how often levelling takes place? 3. How do contemporary dialects from widely-separated geographic locations compare with respect to this patterning? 4. What do patterns of was/were variation reveal about the underlying mechanism which produced the forms? 5. What can patterns of was/were variation reveal about the socio-cultural history and origins of different varieties of English? 4. Data and method 4.1. The communities 4.1.1. Nova Scotia, Canada Nova Scotia, Canada, recognises four founding populations: indigenous Indian and Inuit tribes, as well as early British and African American immigrants. The three Nova Scotian communities we examine here are descendants of the latter two groups. People of African descent came to Nova Scotia in two waves of migration: Black Loyalists after the American Revolutionary War (between 1783 and 1785) and Refugee Slave immigration following the War of 1812. Today, some communities in Nova Scotia still remain which are almost entirely populated by the descendants of these migrations (Poplack and Tagliamonte, 1991; 1994). The two small hamlets in our sample —North Preston and Guysborough— were each populated by one of these migrations. The input settlers to Guysborough were almost entirely Black Loyalists, mainly freedmen and house slaves with service-related skills, while North Preston was populated largely by field hands from the southern United States. 154 Links & Letters 5, 1998 Sali Tagliamonte; Jennifer Smith Since settlement, both communities have functioned as enclaves, at least in terms of the amount of direct contact they have had with the surrounding whites. In addition to the remoteness of these settlements, poor socio-economic conditions, lack of opportunity, and dissociation from the predominantly British-origin mainstream populations all conspired to create circumstances in which African Nova Scotians were maximally separate, thereby providing optimal conditions for the maintenance of a distinct in-group vernacular.5 In the case of North Preston, this is clearly due to geographical isolation. The speakers live in a remote area, separated from the relatively proximate urban centre of Dartmouth/Halifax by wilderness and one limited-access road. In Guysborough, although the African community has coexisted in proximity to their British-origin neighbours, they live in relative isolation in small hamlets outside the main village of Guysborough. The speakers who constitute our sample of British-origin Guysborough speakers, on the other hand, have lived most of their lives within the village of Guysborough. These people trace their lineage to migrations of (white) British American Loyalists who came into the area at approximately the same time as the Africans. 4.1.2. Buckie, Northern Scotland Buckie is a small fishing town on the far north-east coast of Scotland. As a traditional fishing industry community, it has had a long history of cultural cohesiveness with entire families directly or indirectly taking part in the fishing trade. The population (approximately 8000) has changed little in the last thirty years. Unlike similar rural areas in Scotland, which have suffered from de-population, Buckie has a relatively stable economy, so there is no reason for residents to leave. Therefore, a tradition of endogamy has been effectively maintained up to the present-day (Smith and Tagliamonte, 1997). Although contact with mainstream culture exists through local government agencies, media, work on the oil rigs in the North Sea, and particularly education, sustained contact with outsiders is best characterised as minimal. While Buckie is not a linguistic enclave in the strict sense of the term, the majority of residents have existed through to the present in a situation of relative isolation 4.2. The data The goal of our fieldwork in each of these locations was to collect a representative sample of the vernacular norms of each community. Therefore, the speakers all have relatively homogeneous socio-economic characteristics. They were born and raised in the community in question and in each case 5. See Poplack and Tagliamonte (1991) for a detailed description of the corpora and justification for their categorisation as linguistic enclaves. Roots of English in the African American Diaspora Links & Letters 5, 1998 155 they represented the oldest living generation at the time of collection. The data consist of tape-recorded conversations ranging from 1-3 hours.6 They include discussions about local traditions, narratives of personal experience, group interactions and even local gossip. With the exception of Guysborough Village, all were conducted by in-group community members. Our samples, depicted in Table 1, consist of 30 speakers, ranging from 70 to 87 years of age and divided between men and women.7 The occupations of all the speakers were related to traditional or service industries of the community. Most of the women had been housewives, although in Buckie some had worked in the fishing industry and in the Nova Scotian enclaves some had been domestics. In Buckie, all the men were fishermen. In North Preston and Guysborough, they had mostly worked at jobs as labourers, largely in the area of caretaking and cleaning (North Preston) or in the woods, mines, or fisheries (Guysborough Enclave); or as farmers and mechanics (Guysborough Village). Level of education among the informants ranges from almost none to 12 years, with most speakers falling into the lower range. The sociological profile of all the speakers are similar in that they are members of 'dense' networks (Milroy, 1980) and their social circles were generally confined to the community in question. Each community differs with respect to the relative degree of exposure of residents to mainstream culture and language. While all areas can be described as rural, the African Nova Scotian enclaves and Buckie have relatively limited contact with mainstream culture and outsiders.8 In each, the speakers 6. Each corpus is entirely machine readable, with transcription and correction protocols which follow the same methods used in other large corpus construction projects (e.g. Poplack, 1989; 1991; Tagliamonte, 1996). 7. This is actually a small sub set of much larger bodies of materials (see Poplack and Tagliamonte 1991 and Smith, (in preparation). See Tagliamonte and Smith (1998) for an examination of was/were variability across age groups and other communities. 8. In today’s media-enriched and mobile society, isolation is, of course, a relative term. We note however that linguists are relatively unanimous in claiming that while the media may serve to spread vocabulary, new idioms and fashionable pronunciations, it cannot precipitate change in a linguistic system (Milroy and Milroy, 1985; Trudgill, 1986). Table 1. Distribution of speaker sample. Buckie North Preston Guysborough Enclave Guysborough Village Age male female male female male female male female 70-79 2 2 1 1 1 80+ 4 4 1 1 2 2 3 6 Total 8 6 6 10 162 Links & Letters 5, 1998 Sali Tagliamonte; Jennifer Smith However, we are still left with the question of why African Nova Scotian English is so similar to Buckie English, since it is clearly not the case that African Americans in rural Nova Scotia actually came in contact with people from an insular fishing village in northern Scotland. Undoubtedly the Buckie speakers, because of the characteristics of their community, have retained patterns of was/were from the history of English in Britain (see Smith and Tagliamonte, 1997), but the fascinating question is where did the African Nova Scotians get it? The fact that the descendants of the two African enclaves originated in the United States, coupled with the fact that neither of them resembles the patterns found in the Guysborough Village, suggests that they must have got the patterns from a source that predates their migration to Nova Scotia. Although we cannot unambiguously reconstruct the precise origins of the English varieties that were spoken in the United States during the late 1700’s, the fact that non-standard varieties of English from the British Isles were spoken all over the United States during the same time period is not in question. The fact that the African Nova Scotians, who had migrated from both the northern and southern US, exhibit such dramatic similarities with Buckie suggests that they must have acquired the patterns from the varieties of British English spoken in the US at that time. So why is Guysborough Village so different? We believe that the only possible explanation for the patterns we have discovered here is that the varieties under investigation represent different stages in a longitudinal process of linguistic change in the verb to be. The extent of the parallels between Buckie and the African Nova Scotian English enclaves becomes interpretable if we think of them as retaining features from an earlier stage in the history of English. All the varieties which have evolved in enclave or enclave-like situations pattern the same. Only Guysborough Village, the least separate community, does not. In fact, it resembles the high frequency of non-standard was recently reported for standard, urban varieties of English in Canada (Meechan and Foley, 1994) as well as in the UK (Tagliamonte, 1997) and undoubtedly elsewhere, i.e. non-standard was in existentials, but much less frequently elsewhere11. Thus, the differences we observe between Guysborough Village and the other varieties is due to its contact and participation with mainstream developments, not only in Nova Scotia, but in English more generally. The more isolated varieties have simply lagged behind ongoing linguistic developments due to their isolation as well as the in-group solidarity found in the communities. 11. The high rate of non-standard was (33%) in NP plural contexts in Guysborough Village is worthy of note. Here we may be viewing the conflation of two different patterns, one which contemporary varieties appear to be moving towards, i.e. high rates of nonstandard was in existential vs. non-existential contexts, and an older pattern in which NPs had higher rates of non-standard was than pronouns (see Tagliamonte and Smith, 1998). Roots of English in the African American Diaspora Links & Letters 5, 1998 163 The results of this study taken in conjunction with earlier findings on African Nova Scotian English (Poplack & Tagliamonte, 1991; 1994; 1995; 1996a) provide further confirmation that these varieties are a reflection of the state of the English language at the time the ancestors of our informants were acquiring it. Though it perhaps goes without saying, these findings highlight the fact that differences across varieties of English have far less to do with the ethnic heritage of their speakers (e.g. African, Caucasian or otherwise), than with the origins and development of the communities in which they are spoken. Finally, and perhaps most importantly, this study has provided empirical confirmation of the benefits of a more ecologically-focused approach to language contact. References BRUNNER, K. (1970). An outline of Middle English grammar. Oxford: Basil Blackwell. CHAMBERS, J. (1997). Mass media and literacy in language change. Paper presented at Invited Lecture Series. University of York. April 1977. CHAMBERS, J.K. (1995). Sociolinguistic theory: linguistic variation and its social significance. Oxford: Blackwell. CHESHIRE, J. (1982). Variation in an English dialect: a sociolinguistic study. Cambridge: Cambridge University Press. CHRISTIAN, D.; WOLFRAM, W.; DUBE, N. (1988). Variation and change in geographically isolated communities: Appalachian English and Ozark English. Vol. 74. Tuscaloosa, Alabama: American Dialect Society. CURME, G.O. (1977). A grammar of the English language. 1 & 2. Essex, Connecticut: Verbatim. EISIKOVITS, E. (1991). «Variation in subject-verb agreement in Inner Sydney English». In CHESHIRE, J. (ed.), English around the world: sociolinguistic perspectives, 235-256. Cambridge: Cambridge University Press. FEAGIN, C. (1979). Variation and change in Alabama English: a sociolinguistic study of the White community. Washington, D.C.: Georgetown University Press. FORSSTRÖM, G. (1948). The verb ‘to be’ in Middle English: a survey of the forms. Lund: C.W. K. Gleerup. FRIES, C.C. (1940). American English grammar. New York: Appleton, Century, Crofts. HAZEN, K. (1996). «Dialect affinity and subject-verb concord: the Appalachian Outer Banks». SECOL Review. 20(1): 25-53. HENRY, A. (1995). Belfast English and Standard English: dialect variation and parameter setting. New York/Oxford: Oxford University Press. JESPERSEN, O. (1942). A modern English grammar on historical principles. Part VI. Morphology. Copenhagen: Ejnar Muksgaard. London: Allen and Unwin. LABOV, W.; COHEN, P.; ROBINS, C.; LEWIS, J. (1968). A study of the non-standard English of Negro and Puerto Rican speakers in New York City. Comparative Research Report. U.S. Regional Survey, Philadelphia. LASS, R. (1987). The shape of English: structure and history. London: Dent. 164 Links & Letters 5, 1998 Sali Tagliamonte; Jennifer Smith MEECHAN, M.; FOLEY, M. (1994). «On resolving disagreement: linguistic theory and variation. There’s bridges». Language Variation and Change. 6(1): 63-85. MILROY, L. (1980). Language and social networks. Baltimore: University Park Press. MILROY, J.; MILROY, L. (1985). Authority in language. London: Routledge. MONTGOMERY, M. (1989). «Exploring the roots of Appalachian English». English World-Wide. 10: 227-278. MOSSÉ, F. (1952). A handbook of Middle English. Baltimore: John Hopkins University Press. MUFWENE, S.S. (1996). «The founder principle in creole genesis». Diachronica. 13(1): 83-134. — (to appear). The legitimate and illegitimate offspring of English. In FORMAN, M.L.; SMITH, L.E. (eds.). World Englishes 2000. College of Languages, Linguistics and Literature, University of Hawaii and the East-West Center. O’LEARY, C.F. (1997). «Ethnolinguistic vitality and survey research in India». Paper presented at International Language Assessment Conference. Hosleys Green, England. POPLACK, S. (1989). «The care and handling of a megacorpus: the Ottawa-Hull French project». In FASOLD, R.; SCHIFFRIN, D. (eds.). Language change and variation, 411-451. Amsterdam: John Benjamins. POPLACK, S.; TAGLIAMONTE, S. (1989). «There’s no tense like the present: verbal -s inflection in early Black English». Language Variation and Change. 1(1): 47-84. — (1991). «African American English in the diaspora: the case of old-line Nova Scotians». Language Variation and Change. 3(3): 301-339. — (1994). «-S or nothing: marking the plural in the African American diaspora». American Speech. 69(3): 227-259. — (1995). «It’s black and white: the future of English in rural Nova Scotia». Paper presented at NWAVE 24. University of Pennsylvania, Philadelphia. — (1996a). «The grammaticization of gonna in six varieties of English: a cross-linguistic comparison». Paper presented at NWAVE 25. University of Nevada, Las Vegas. — (1996b). «Nothing in context: variation, grammaticization and past time marking in Nigerian Pidgin English». In BAKER, P. (ed.). Changing meanings, changing functions. Papers relating to grammaticalization in contact languages, 7194. Westminster, UK: University of Westminster Press. POUSADA, A.; POPLACK, S. (1982). «No case for convergence: the Puerto Rican Spanish verb system in a language-contact situation». In FISHMAN, J.A.; KELLER, G.D. (eds.). Bilingual education for Hispanic students in the United States, 207237. New York: Teachers College, Columbia University. PYLES, T. (1964). The origins and development of the English language. New York: Harcourt, Brace and World, Inc. PYLES, T.; ALGEO, J. (1993). The origins and development of the English language. Orlando: Harcourt Brace and Company. QUIRK, R.; WRENN, C.L. (1960). An Old English grammar. London: Methuen and Company Ltd. RAND, D.; PATERA, T. (1992). Concorder. Computer program. Montréal: Centre de recherches mathématiques. 1.1S. Roots of English in the African American Diaspora Links & Letters 5, 1998 165 RICKFORD, J. (1977). «The question of prior creolization of Black English». In VALDMAN, A. (ed.). Pidgin and creole Linguistics, 190-221. Bloomington: Indiana University Press. RICKFORD, J.R. (1998). «The development of African American Vernacular English: focussing on the creole origins issue». In MUFWENE, S.S.M.; RICKFORD, J.R.; BAILEY, G.; BAUGH, J. (eds.). African American Vernacular English. London: Routledge. SCHILLING-ESTES, N.; WOLFRAM, W. (1994). «Convergent explanation and alternative regularization patterns: were/weren’t levelling in a vernacular English variety». Language Variation and Change. 6(3): 273-302. SMITH, J. (in preparation). Synchrony and diachrony in the evolution of English: evidence from the far reaches of Scotland. DPhil Dissertation, University of York. SMITH, J.; TAGLIAMONTE, S. (1997). «We was all thegither, I think we were all thegither: was variability in Buckie English». Submitted to World Englishes. TAGLIAMONTE, S. (1996). Roots of identity: variation and grammaticization in contemporary British English. ESRC Grant. University of York. September 1996-August 1998. — (1997). «Was/were levelling across the generations: English in the city of York». Paper presented at Variation Workshop. Reading, UK. April 3-5, 1997. TAGLIAMONTE, S.; POPLACK, S. (1988). «How Black English past got to the present: evidence from Samaná». Language in Society, 17(4): 513-533. TAGLIAMONTE, S.; SMITH, J. (1998). «Was/were variation in English: primitive, remnant, syntactic restructuring or act of identity? Evidence from inter-variety comparison». Paper presented at American Dialect Society Meeting, January 8-11, 1998. New York. THOMASON, S.G.; KAUFMAN, T. (1988). Language contact, creolization and genetic linguistics. Berkeley and Los Angeles, California: University of California Press. TRUDGILL, P. (1986). Dialects in contact. Oxford: Blackwell. — (1990). The dialects of England. Oxford: Blackwell. VISSER, F.T. (1970). An historical syntax of the English language. Leiden: E. J. Brill. WEINREICH, U.; LABOV, W.; HERZOG, M. (1968). «Empirical foundations for a theory of language change». In LEHMANN, W.P.; MALKIEL, Y. (ed.). Directions for historical linguistics. Austin: University of Texas Press. WOLFRAM, W. (1997). «Alternative regularization patterning and ethnic marking in a tri-ethnic rural southern community». Paper presented at SECOL 56. Charlotte, NC. April 1977. WOLFRAM, W.; CHRISTIAN, D. (1976). Appalachian speech. Arlington, Va.: Center for Applied Linguistics.