An Army, A Navy, and Ebonics

Terence Odlin

Ohio State University

This paper is the written version of presentations in Helsinki and Joensuu, Finland in 1997. A summary appears in the final section. A print version is also available. Comments may be sent to odlin.1@osu.edu.

    1. Introduction
    2. The Ebonics controversy
    3. Afro-American languages and dialects
    4. Language, identity, and politics
    5. General summary and conclusion

    REFERENCES
    APPENDIXES

    1. Introduction

    Many people who try to get acquainted with linguistics find the terminology frustrating. There are, for sure, a lot of terms, and they are not always consistently used. Moreover, many of the terms come into use quickly and go out of use just as quickly. To some extent, these problems are hard or even impossible to avoid. When so much about human language remains only partially understood, it should come as no surprise that new attempts to understand will succeed only partially--if at all. Even basic terms such as language and dialect are subject to innumerable controversies. The common textbook definition of dialects as mutually intelligible varieties of the same language is useful, but linguists often point out problems with the definition. For example, just how intelligible do dialects have to be? On the other hand, critics of the definition have not succeeded in providing a widely accepted replacement. Another problem with the language/dialect distinction is that history shows that today's dialect can become tomorrow's language. One such case is Macedonian (which will be briefly discussed in section 4 of this paper). Facts such as this make it easy to see why linguists have facetiously defined the term language as a dialect with an army and a navy.

    The recent controversy in the United States over Ebonics also illustrates some of the problems that can arise in attempts to distinguish languages and dialects, and it provides a useful opportunity to consider just how it is that a dialect might become a language. The following discussion first considers some of the history of the Ebonics issue and then looks at the history of what is called Black English by some and Ebonics by others (not to mention still other terms, as will be seen). After these historical considerations, the paper will turn to the issue of how a dialect can become a language.


    2. The Ebonics controversy

    Before December 18, 1996 few people had heard the term Ebonics, although it has been around for over twenty years, as seen in a collection of articles edited by Robert Williams (1975) with the title Ebonics: The True Language of Black Folks. By New Year's Day of 1997, however, the term had became known to many people. The coverage in the Washington, D.C. newspapers, the media that I myself had access to in late December, emphasized three claims: 1) that on December 18 the School Board of Oakland, California had declared Ebonics to be a separate language; 2) that the School Board would soon submit an application for funding a bilingual education program for speakers of Ebonics to learn English; 3) that actual classroom teaching was to be done in Ebonics.

    If news can be defined as breathtaking information, this story was certainly news. The School Board's alleged actions seemed to challenge widely shared assumptions, and so, not surprisingly, the reports in the media led to a furor. Condemnations came quickly, crossing political and racial lines. The Washington Post, a liberal newspaper, carried many articles and letters to the editor condemning the School Board's decision, as did the conservative Washington Times. For a short period, the black liberal Jesse Jackson and the black conservative Thomas Sowell agreed about something (both excoriating the School Board), although Jackson later changed his position. Moreover, the U.S. Secretary of Education, who is a white Democrat, took very little time to declare that no bilingual education funds would be available for an Ebonics program.

    Not all of the early reactions were unfavorable, however. Some of the columns and letters in the December issues of the Washington Post expressed sympathy for the School Board's approach, and the coverage in the Post changed somewhat after a resolution passed on January 3, 1997 by the Linguistic Society of America at their annual meeting. In that resolution, the LSA stated that

    Moreover, the Society praised the School Board's efforts:

    While the entire LSA resolution is given in Appendix 1 of this paper, a number of points in the above passages should be noted: 1) the warning about terms such as "slang"; 2) the support for the teaching of Standard English; 3) the support for the School Board. Anyone familiar with linguistics in the twentieth century will not be surprised that the first two of these points are evident in the LSA resolution. On the other hand, the discussion to follow will show that the LSA's unalloyed praise for the School Board may come back to haunt the profession since the School Board's declarations show a stunningly poor understanding of linguistics.

    In both the condemnations and the support that the Oakland School Board received in the early weeks of the controversy, the actual text of the resolution was rarely quoted. Despite all the coverage in the media, it was not clear exactly what the School Board had said. Moreover, in the LSA resolution there is no part which actually quotes the School Board's resolution. Thanks to the World Wide Web, however, the text of the original resolution is easily available and is given in full in Appendix 2 along with the revised resolution of January 15, 1997 in Appendix 3. The text of the original and revised resolution will be considered with regard to the three points emphasized in the early media coverage: 1) that the Oakland School Board had declared Ebonics to be a separate language; 2) that the School Board would soon submit an application for funding a bilingual education program for speakers of Ebonics to learn English; 3) that actual classroom teaching was to be done in Ebonics.

    2.1 Declaration of a separate language

    The text of the December resolution, which passed unanimously, leaves no doubt that the Oakland School Board chose to consider Ebonics as a language distinct from English. The first paragraph of the December resolution defines Ebonics thus:

    The mention of "numerous validated scholarly studies" is peculiar, since the terms Ebonics and African Language Systems are rarely found in linguistic analyses, the most common terms in recent times being Black English and African-American Vernacular English (AAVE). Because the former term is more transparent than AAVE and is shorter than African-American Vernacular English, it will be used in the following discussion. The School Board's use of Ebonics instead of any term with English in it is no accident, as the next paragraph of the resolution makes clear:

    The phrase "genetically-based" shows a very garbled conception of historical linguistics, where the term "genetic" denotes a type of language change very different from the kind of historical account the Board probably has in mind (Thomason and Kaufmann 1988). Aside from its inaccurate use as a linguistic term in the Board's December resolution, the phrase probably led some members of the public to assume that it had something to do with genetics and perhaps with the biology of race. In any case, it is a phrase that does not reappear in the Board's January resolution.

    Other phrases in the December resolution also make it clear the School Board considered Ebonics a separate language. In paragraphs 8 and 9 appears the phrase "instructing African American children both in their primary language and in English," although once again these phrases were deleted from the January resolution. Although some people have seen the January resolution as a retreat from the declaration of a separate language, some of the wording shows the School Board continuing to stick by its original claim. Thus in the second paragraph of the revised resolution, the following assertion appears:

    Although the red flag raised by "genetically based" has been lowered, the School Board has not backed away from its claim that Ebonics is a separate language. Moreover, the supposed rationale for why Ebonics should be considered a separate language is laid out in the policy statement attached to the December resolution, and the January resolution endorses this same policy statement (as discussed further in section 3.4).

    2.2 Bilingual education funding

    The sixth paragraph of December resolution cites U.S. Federal law:

    This phrase "limited English proficiency," which is used to designate pupils served by bilingual education programs, appears elsewhere in the resolution as well as in the policy statement. Thus paragraph 8 of that statement clearly shows that speakers of Ebonics are viewed as having limited English proficiency:

    In this paragraph the School Board follows through consistently on its claim that Ebonics is a separate language, not a dialect of English, and sees Ebonics speakers as comparable to speakers of other languages who have children eligible for bilingual education. With regard to the issue of non-standard English, it might seem at first glance that this paragraph is an open-minded concession to parents who disagree with the School Board. However, the language is in fact patronizing and coercive. The notion of non-standard English is equated with speech disorders, which is certainly not an equation that linguists would make. Unfortunately it was not unusual earlier in this century for speakers of non-standard varieties of English to be deemed to have speech disorders or "cognitive deficits." In this context, the language of paragraph 8 can be read as a threat: either allow us to put your children in Ebonics classes, or we will assign them to special education classes along with children having speech disorders. The School Board may never have intended to make their position sound so threatening, but if so this paragraph is nevertheless one of several parts of the resolution that are poorly written.

    The mention of Federal law on bilingual education in the December resolution can hardly be accidental. Immediately after the paragraph that mentions the Bilingual Education Act Is the convoluted phrasing of paragraph 7:

    No part of the resolution actually states that the School Board intended to apply for bilingual education funding. As the controversy developed, the School Board issued a synopsis of its positions, a document intended to dispel misconceptions that have arisen. One supposed misconception is that "The [School] District is trying to classify Ebonics (i.e., 'Black English') speakers as Bilingual." In view of the paragraphs that I have cited in this section, that particular denial of the School Board is disingenuous. Ebonics is declared to be a separate language, and speakers of that language are considered to be "limited English proficient." The language of the December resolution clearly indicates that the School Board was maneuvering to get bilingual education funding for its Ebonics program.

    2.3 Classroom teaching and Ebonics

    There is very little in either the December or the January resolution that specifies how the actual teaching is to be done. Nevertheless, the basic principle endorsed by the Board is clear in paragraph 12 of the original resolution:

    In the School Board's synopsis, it insists that the goal of its program is to teach Standard English:

    This claim of the School Board rings true, in contrast to its denials about its position on bilingual education (section 2.2). The approach taken in the Oakland schools seems to be some form of contrastive analysis, i.e., systematic comparison, which has been part of educational systems for literally thousands of years (Odlin 1989). John Rickford, a linguist who has written extensively on the Ebonics issue and who has defended the School Board, asserts that "it is possible to do contrastive analysis between Ebonics and Standard English, and shortcut the process of helping Ebonics speakers master Standard English" (Rickford 1996). Whether the speech of schoolchildren is called Ebonics or Black English, it is possible to compare it with Standard English, as Rickford argues. To use a contrastive approach does not mean that two languages are being taught simultaneously. In classes on Finnish as a second language, for example, teachers often make systematic comparisons between the target language and English: in such cases, the goal is to teach Finnish and not English (even though some incidental learning about English might take place). If the Oakland School Board is using a contrastive approach, as it seems to be, it can legitimately claim that it is not teaching Ebonics.

    Rickford and others have argued that if contrastive analysis works then it deserves to be promoted, and he has cited a number of studies on the use of contrastive approaches in helping, for example, speakers of non-standard Swedish (Rickford 1997). Moreover, the Oakland School Board cites successes in its system with Ebonics (although the School Board's credibility is indeed a real problem). Pragmatic arguments such as Rickford's may eventually prevail, but the anxiety over contrastive analysis has surfaced before in Ann Arbor, Michigan, for example, where legal battles were fought over a program far less ambitious than the Ebonics initiative.

    Linguists who remember cases such as Ann Arbor have to some extent viewed the Ebonics controversy as yet one more example of the same conflicts. In one sense this is true, but the Oakland School Board has raised the stakes considerably by its declaration of linguistic independence. The controversy also poses a difficult problem for linguists who are aware of the unsupportable technical arguments invoked by the School Board (as will be seen in the next section). Moreover, the language of both the December and January resolutions of the School Board is, to say the least, extremely embarrassing for any of the Board's supporters. The December resolution is riddled with problems of usage, not to mention the patronizing diction discussed in section 2.2. In some cases, the usage problems are so serious that it is impossible to know what the School Board had in mind. Paragraph 10 of the December resolution reads

    Whatever "each language" was supposed to mean, this phrasing was dropped in the January resolution as were some other howlers (e.g., "prejudicially and unconstitutionally vetoed" in paragraph 4). Nevertheless, much of the incompetent usage remains in the revised resolution, including the first sentence of paragraph 8:

    Among the problems in this paragraph is the lack of a verb to go with "teachers and instructional assistants," and attentive readers will no doubt spot other problems here and elsewhere in the resolution.

    The incompetent writing does not mean that the School Board intended to formulate its resolutions in Ebonics or Black English; the problems are simply further examples in a long tradition of bad bureaucratic writing in (supposedly) Standard English. Indeed, teachers in the Oakland School District could use these instances as negative examples in writing classes. As noted before, little in the early press coverage that I saw cited the actual language of the December resolution, and this is a case of undeserved good fortune for the School Board. The Board's credibility is open to question not only because of its duplicitous position on bilingual education but also because its own use of language might well leave anyone to wonder if School Board members themselves can actually write in Standard English.

    2.4 Summary and comment

    The chief claims in the early media coverage of the Ebonics story were: 1) that the Oakland School Board had declared Ebonics to be a separate language; 2) that the School Board would apply for the funding of a bilingual education program for speakers of Ebonics to learn English; 3) that actual classroom teaching would be done in Ebonics. Now that the Board's resolutions are available on the Internet, it is possible to determine just how accurate the three claims were. The first is certainly true: both the December and January resolutions make it clear that the School Board considers Ebonics to be a separate language. The second claim in the media is also true: even though the School Board has denied that it ever intended to seek bilingual funding, the references to Federal law and the use of "limited English proficient" make the Board's original stratagem clear. As for the claim that classroom teaching would be done in Ebonics, the Board's resolutions say little specific about actual pedagogical practices. Other evidence, however, indicates that the Oakland School District was using contrastive analysis, i.e., systematic linguistic comparisons.

    Even though contrastive analysis offers a promising approach to help with the teaching of Standard English, the School Board has made this approach much more controversial than it has been in earlier conflicts such as one in Ann Arbor, Michigan. The Board's smoke-screen about its intentions on bilingual funding has hurt its credibility, as has the incompetent writing in both the December and January resolutions. Moreover, its declaration of linguistic independence is based on arguments that no linguist could take seriously, as will become evident in the next section.

    3. Afro-American languages and dialects

    The Oakland School Board appropriates ideas from historical linguistics to justify its declaration of linguistic independence. The Board's appropriation is amateurish, but there are linguistic arguments can be used to support the claim that Ebonics is a separate language. Even so, these arguments are not in themselves sufficient to decide the question.

    Both the December and January resolutions invoke the same policy statement, the first paragraph of which provides the following rationale for the proclamation of Ebonics as a separate language:

    Once again, the first sentence is syntactically ill-formed, but it manages to state that the linguistic variety in question has a structure, something that no linguist would challenge even though some members of the public believe otherwise. As for the claims of linguistic independence, the next sentence provides the rationale. This rationale will be scrutinized in more detail later in this section, but the key claim should be noted now: that Ebonics shows the influence of African languages and therefore is a language independent of English. The very claim of African influence led the conservative columnist Thomas Sowell to raise the following objection:

    Neither Sowell nor the Oakland School Board has offered sound linguistic arguments, but the claims are nevertheless significant for two reasons. First, they show that historical linguistics, which is not usually a political hotbed, can be invoked to serve political ends. This is not first time historical studies have been politicized: a far more sinister example is when the Nazis appropriated the term Aryan (which is in fact a valid term to designate the Indo-Aryan branch of Indo-European languages) to distinguish the Master Race from peoples who could be enslaved or exterminated. Aside from the political purposes of Sowell and the School Board, their claims are also noteworthy in that they reflect, however, unfaithfully, the two main approaches to understanding the history of Black English: the creolist position and the dialectologist position.

    3.1 Black English: the creolist position

    The creolist position holds that Black English has its origins in a creole created by Africans on plantations in the American South, with this creole in turn related to pidgins that developed in the trans-Atlantic slave trade. As with the terms language and dialect, the terms pidgin and creole are difficult to define well. The following definitions are typical of those found in textbooks and reference works:

    Most creolists are ready to point out that such definitions greatly simplify and even distort the circumstances under which pidgins and creoles develop. While it is indeed possible to formulate a better definition, there is none that most specialists routinely cite. Some other commonly found criteria in the alternatives are nevertheless important to note: 1) pidgins are usually simpler in structure than creoles; 2) even though varieties such as Melanesian Pidgin English and Jamaican Creole owe most of their vocabulary to English, their very different structures make them unintelligible to native speakers of English; 3) some of these varieties now serve as national languages, as in the case of Tok Pisin in New Guinea and Bislama in Vanuatu.

    To explain the history of Black English, linguists look to the creoles in former Caribbean colonies such as Barbados and Jamaica. When Europeans from the Caribbean brought their slaves to establish new colonies such as South Carolina, they created the conditions for creoles in what was to become the United States. In fact, one such creole survives on the Sea Islands of South Carolina and Georgia: Gullah. Virtually all linguists who have compared the structure of Gullah with Jamaican and other Caribbean creoles concur that it constitutes a language apart from English. By many creolist accounts of Black English, then, it started out as Gullah (and perhaps related creoles) but changed over centuries as slavery ended and as some of the barriers to education and to contact with white speakers of English fell. By this account, Black English is a "decreolized" vestige of Gullah, with the original creole remaining intact only on the Sea Islands, where there has been some change although the extent of that change remains controversial (Mufwene 1991).

    It may well be that the status of Gullah as a distinct language led the Oakland School Board to adopt an extreme creolist position: that very little change has occurred in working-class Afro-American speech, and therefore this variety is a really creole and not a dialect of English. Very few linguists would be willing to make this claim, however, especially since there are clear structural differences between Gullah and Black English (e.g., Edwards and Winford 1991, Weldon 1998). Even if all Afro-American working-class speech comes from Gullah, as some creolists would maintain, the fact must still be acknowledged that significant departures from Gullah structure have occurred.

    3.2 Black English: the dialectologist position

    In contrast to the creolist position, the dialectologist position emphasizes the similarities between vernacular Black English and vernaculars in the British Isles. Several of the examples mentioned by Sowell (quoted in section 3.1) are indeed real evidence for the dialectologist position, though Sowell's use of "Englishmen" is naïve: the use of dis and dat, for example, is most common in Irish, not British, English. In any case, linguists have long noted similarities between some structures in Black English and ones in the vernaculars of England, Scotland, and Ireland (with considerable variation occurring within each of those areas). The similarities were so numerous that they led early dialectologists to assume that everything in Black English can be explained by invoking some antecedent in the British Isles.

    3.3 Toward a synthesis

    In their strongest forms, the creolist and dialectologist positions offer radically different explanations for the history of Black English. Yet despite these differences and despite occasional bickering, more and more linguists recognize that an adequate account must incorporate both points of view. On the one hand, Black English does show many parallels with vernacular structures found in the British Isles. On the other hand, Black English does have structures which differ considerably from anything in British or Irish vernaculars and which indeed have parallels with Gullah and other creoles. These linguistic facts thus reflect Afro-American history: black slaves often had encounters with working class British and Irish immigrants in some areas, but the barrier of slavery itself no doubt reduced the possibilities for contact between blacks and whites. Whatever contact there was would be conducive to the learning of British or Irish vernacular structures, while social separation would be conducive to the creation or maintenance of creole structures.

    The history of Barbados, one of the first British Caribbean colonies, is especially relevant here. Research by Niles (1980), Rickford (1986), and others shows that in its earliest days, Barbados had a very diverse labor force, some plantation workers being political prisoners from Southwest England, others "transported" Irish, and still others slaves brought from Africa. In 1645, the population of Barbados was about 85% European, but later in the same century the demographic balance greatly shifted, and today blacks constitute some 90% of the population (Lawton 1983). Despite the fact that Bajan, the creole of modern Barbados, is the predominant language on the island today, creolists see its origins in the period before 1650 when European vernaculars were the most commonly heard speech. The slaves in Barbados in that early period no doubt acquired English as a second language, albeit a nonstandard variety reflecting the British and Irish vernaculars which constituted the target language. As more slaves were brought to Barbados, the social distance between blacks and whites grew, according to Rickford (1986), and slaves who had already acquired some English increasingly became the speakers who would model the target language for the new arrivals. As has happened in other cases of creolization, the target eventually drifted from the original nonstandard vernaculars to what was increasingly a language reshaped by slaves (Odlin 1991a).

    As the earliest British colony in the Caribbean, Barbados played a key role in the creation of new colonies, and creolists have recognized the significance of Bajan in the formation of new creoles (even though they disagree on a number of the details). The fact that British and Irish folk speech played an important role in the early development of Bajan itself makes it unwise for creolists to ignore the observations of dialectologists. On the other hand, Bajan is not simply Irish or Southwestern British English, and so many of the vernaculars in the Caribbean or North America influenced by Bajan may well have a creole as well European folk speech as part of their heritage. Just what the similarities are between Bajan, Gullah, Jamaican and other creoles remains a question only partially answered by linguists (e.g., Winford 1993), and to that extent the exact historical connections between Bajan, Gullah, and Black English remain incompletely understood.

    In a recent paper, Winford (1997-1998) has attempted to untangle several of the historical and linguistic problems concerning the formation of Black English (or, in his terminology, African American Vernacular English). He first considers the different settlement patterns seen in the formation of the earliest British colonies in the American South. In the case of Virginia, the most populous colony for many years, whites constituted the majority in the period between 1685 and 1790, although the proportion of blacks increased considerably during the second half of that period. In contrast, blacks outnumbered whites in South Carolina during the same period for about sixty years (from 1715 to 1775). While Winford takes account of the regional variation within these colonies (e.g., the preponderantly white population in the western regions of South Carolina), he sees the rise of Gullah in South Carolina as especially likely given the fact that the coastal areas were heavily populated by slaves. On the other hand, many areas of Virginia, especially those far from the coast, had much smaller black populations and in such areas the influence of British vernaculars would be likely be strongest. Winford also discusses two other colonies: North Carolina, where the demographic patterns resemble those of Virginia, and Georgia, where they resemble those of South Carolina.

    After his consideration of historical and demographic factors, Winford focuses on two types of structures: those which Black English has in common with Southern White Vernacular English, and those found in Black English but not in SWVE. The relative number of structures of each type is revealing: Winford considers fifteen examples of the former type and seven of the latter. Among the structures shared with Southern White Vernacular English are ones fairly well known to most Americans such as nonstandard past tense forms (e.g., I seen it) and ones not well known or widely used such as verb phrases using double modals (e.g., I might could do it), a structure also common in Scotland and Ulster. It is worth noting that along with the fifteen structures common to Black English and Southern White Vernacular English, there are many others that these two varieties share with most other dialects of English, standard and nonstandard alike. Thus, for example, article usage does not diverge considerably from Standard English even though many speakers of English as a second language have considerable problems in mastering the English article as seen in errors such as Picture is very dark and This plan of the Kissinger (Odlin 1989). The similarities in article usage in Black English, Southern White Vernacular English, and most other native speaker varieties constitute just one of the areas where the overlap of dialects is considerable-and often taken for granted.

    The seven structures that appear in Black English but not in SWVE include several that are highly similar to ones found in Caribbean creoles. For example, is and are are frequently omitted in verb phrases as in He David, too (Whatley 1981). While such structures are clearly comparable to ones in Caribbean creoles, there remains the problem of explaining how they have become so pervasive in Black English. If a strict creolist interpretation were correct, the explanation would be straightforward: all varieties of Black English would have their origins exclusively in a Caribbean creole. One problem with this explanation is, of course, why there are such strong resemblances to Southern White Vernacular English. In some cases it could be argued that the speech of whites was influenced by the language of slaves, especially in areas where slaves were the majority population. Although this explanation is plausible in some instances, it is certainly not in others. Double modals, for example, are found not only in Black English and SWVE but also in Scotland and Ireland: it is farfetched to argue that double modals started in Caribbean creoles and found their way eventually to the British Isles. Recognizing such problems, Winford suggests that in some areas (e.g., much of Virginia) speakers of African languages acquired white settler dialects (which evolved into SWVE) as a second language without ever learning a creole, while in other areas (e.g., parts of South Carolina) speakers of Gullah and perhaps other creoles acquired SWVE. This explanation seems very likely: it accounts for the identity of some widespread features in the speech of blacks and whites (as in I seen it), and it provides a simple explanation for the retention of several creole structures in Black English.

    The account just outlined thus suggests that both creoles and vernaculars of the British Isles played an important role in the formation of Black English. Much of Winford's analysis has of necessity been simplified, and Winford acknowledges that there are many unanswered questions, including ones about structures which may owe something to creoles or British vernaculars but which remain to be investigated. In any case, it is clear that anyone interested in the history of Black English needs to consider the complexity of its sources.

    3.4 On the issue of African influence

    As noted above, the December and January resolutions endorse a policy statement that provides the theoretical justification for the declaration of Ebonics as a separate language. The key section of that statement is the claim that "African-Americans (1) have retained a West and Niger-Congo African linguistic structure in the substratum of their speech and (2) by this criteria [sic] are not native speakers of black dialect or any other dialect of English." According to this claim, early influences from African languages (the substratum) are evident in twentieth century speech, and these influences alone make Ebonics a separate language. This section will consider the dubious logic of that argument.

    The most serious defect of the School Board's position arises from the fact that substratum influence is neither a necessary nor a sufficient condition for considering some form of speech to be a separate language. This defect will be analyzed below, but some other problems with the policy statement should also be noted. The very phrasing of the statement suggests the School Board's unfamiliarity with the issue of substratum influence. An obvious question to ask the School Board's theoreticians is what particular African language or (languages) can account for the supposed influence. The most typical case of such influence is when a single language constitutes the substratum. For example, vernacular Irish English shows considerable influences from Irish (also known as Gaelic), and Ecuadorean Spanish shows the influence of Quechua, the language of the Incas (Muysken 1984, Filppula 1986, Odlin 1991b, 1997). It is also possible for a form of speech to show the influence of two or more languages: one likely case involves the influence of Kwaio and other Oceanic languages of the South Pacific on Melanesian Pidgin English (Keesing 1988). The School Board seems to claim that more than one African language contributed to the substratum. While such influence is certainly possible, the School Board's use of the term "West and Niger-Congo African linguistic structure" is vague, to say the least. There is no family of "West African" languages comparable to Oceanic languages, for example. "West African" is simply a geographic designation which does not say very much about the languages, certainly no more than the term "East Asian language" says about Chinese, Japanese, and Vietnamese, three different languages from three different language families. The term "Niger-Congo" is more informative, but it encompasses a family of very different languages extending, by some classifications, all the way from Mauritania to South Africa. Making claims about Niger-Congo influence says nothing about what particular languages contributed the most to the alleged substratum.

    While the School Board's claims are amateurish, this does not mean that substratum influence is impossible. The question of Africanisms in Gullah got serious scholarly attention in a study published nearly fifty years ago by Lorenzo Turner (1949). Not only Gullah but other Afro-American varieties have been the subject of several investigations, some of the most recent appearing in a volume from a conference held at the University of Georgia in 1988 (Mufwene 1993). Although Turner's study makes a plausible case for some kinds of influence, as in the words he attributes to the Kongo language, not all of his claims are credible. Likewise, the quality of substratum research on other creoles and pidgins varies considerably, and creolists continue to debate both the specifics and the generalities concerning substratum influence. Even so, there have now appeared some very credible accounts of Africanisms. Among the best studies are three that deal with various kinds of grammatical influences in Liberian Pidgin English (Singler 1988), in Berbice Dutch, a creole spoken in Guyana (Smith et al. 1987), and in Saramaccan, a creole spoken in Surinam (McWhorter 1992). The details of these studies make them fairly technical, and adequate summaries would require a long discussion. Even so, it is worth noting that the three articles share certain strengths, especially where each provides a detailed examination of specific structures in specific African languages and where each makes a serious attempt to show that speakers of these languages were present during the formative period of the pidgin or creole.

    McWhorter's study of Saramaccan probably has important implications of for Gullah. These two creoles share a structure known as serial verbs, as in a Gullah sentence Komin iista ol a wi hav egz, yu no, wi kya iit, which Mufwene (1990: 93) translates as "Coming on Easter, all of us have eggs, you know, we carry [and] eat them," where kya iit (carry-eat) is the serial verb (cf. Winford 1993). While these constructions need not arise from substratum influence (and are often found in languages that are not pidgins or creoles), McWhorter and others have made strong arguments that some creole serial verbs reflect highly particular properties of certain African languages. Similarly, Winford notes a structure found in Black English that may reflect an earlier serial verb: He just took and hit me. As he observes however, the research has yet to be done on the way this structure has been used or its possible relation to similar structures in Caribbean creoles.

    Yet even though the question of African influences in Gullah and in Black English has yet to be thoroughly investigated, the substratum argument is unnecessary to make for anyone who wants to have Ebonics considered as a separate language. After all, there are the seven structures discussed by Winford that make Black English distinct from other forms of English. It is not clear how these seven characteristics came to be in Gullah as well as in Black English: creole structures can arise from a variety of sources besides substratum influence (Odlin 1989). Yet whatever the best historical explanation may be for the seven creole structures, the fact remains that they make Afro-American varieties of speech distinct. Since structural differences are a necessary condition for arguing that two forms of speech are separate languages, there is a possible structural basis for declaring Ebonics to be a separate language.

    African substratum influence is thus not a necessary condition for any declaration of linguistic independence. Moreover, it is not a sufficient condition. That is, even if better evidence is found (as is likely) for the African sources of Gullah and Black English, finding such evidence will not really strengthen the claim that Ebonics should be considered a separate language. In Irish English, there are easily a dozen grammatical structures that show some influence from the Irish substratum even though British vernaculars played a role in the development of some of the structures (e.g., Harris 1986, Odlin 1992, 1997). Despite the fact of Celtic influence, however, it would certainly be mistaken to consider the vernacular to be a Celtic language: Irish English has much more in common with other forms of English than it does with Irish. By the same token, Black English has far more in common with other forms of English than it does with any Niger-Congo language. This fact may not sit well with the School Board or its supporters, but no linguist would seriously argue that black children in Oakland could learn languages such as Twi, Yoruba, or Kongo as easily as they could learn Standard English (even though there certainly are difficulties in learning to speak or write a standard dialect). The School Board's claims about " Pan African Communication Behaviors or African Language Systems" no doubt arise in part from a desire to find a noble pedigree that will make children in the Oakland School District unashamed of the way they speak. Such an intention is commendable, but the bad linguistics endorsed by the School Board is not.

    3.5 Summary and comment

    The Oakland School Board invokes historical arguments to support its claim that Ebonics is a separate language, while some opponents of the Board have marshaled other historical arguments. Although the arguments made by both sides are amateurish and highly politicized, they do reflect two distinct positions that linguists have taken on the origins of Black English. The creolist position holds that Black English has its origins in a creole created by Africans on plantations in the American South, with this creole in turn related to pidgins that developed in the trans-Atlantic slave trade. The dialectologist position emphasizes the similarities between vernacular Black English and vernaculars in the British Isles.

    Although these two positions diverge radically in their extreme versions, more and more linguists recognize that an adequate account must incorporate elements of both the dialectologist and the creolist explanations. On the one hand, Black English does show important parallels with vernacular structures found in the British Isles. On the other hand, Black English does have structures which differ considerably from anything in British and Irish vernaculars and which have parallels with Afro-American creoles. Winford (1997-1998) has attempted to synthesize the dialectologist and creolist approaches, partly by looking at which parts of the American South would have been conducive to creolization and which would have brought Africans into fairly close contact with speakers from the British Isles. He also focuses on two types of structures: those which Black English shares with Southern White Vernacular English, and those found in Black English but not in SWVE. Winford identifies fifteen examples of the former type and seven of the latter.

    The School Board has argued that early influences from Niger-Congo languages (the substratum) are evident in twentieth century speech and that these influences alone make Ebonics a separate language. However, there are serious flaws in this argument, even though there may indeed be some substratum influence. One problem is that the School Board fails to specify what particular Niger-Congo language (or languages) can account for the alleged substratum. An even greater problem is that substratum influence is neither a necessary nor a sufficient condition for considering any kind of speech to be a separate language. The seven structures identified by Winford may or may not reflect African influence, but in either case they do provide a possible basis for declaring Ebonics to be a separate language. Nevertheless, Black English has far more in common with other forms of English than it does with any Niger-Congo language, and so designations such as "African Language Systems" border on quackery.

    The structural descriptions of Black English are not thorough--nor for that matter are descriptions of any other dialect, including Standard English. Even so, enough is known to allow for reliable comparisons between Black English and other dialects as well as other languages. As stated above, the seven structures identified by Winford could be used to justify considering Ebonics a separate language (even though no one can seriously argue that Ebonics is a Niger-Congo language). However, structural arguments are not always sufficient to decide the question of where a dialect ends and a language begins, and certainly not when only a handful of structural features distinguish Black English from other dialects. In such cases, other factors besides linguistic structure become especially important, and these will be considered in the next section.

    4. Language, identity, and politics

    The preceding section described how there are indeed structural differences between Black English and other dialects, with these differences making a possible case for declaring Ebonics a separate language-even though the specific arguments used by the Oakland School Board fail to make the case well. Yet while the structural differences are thus significant, the question remains as to how many structures constitute sufficient evidence for a separate language. Common sense might suggest that seven structures are not enough. After all, the differences between languages are usually vast: English and Chinese, for example, show countless contrasts in their vocabulary, grammar, and writing systems, and even languages with a fairly close historical relationship to English, such as German, diverge on many points. In general the more contrasting points shown by two varieties, the easier it is to claim that they are in fact separate languages. Nevertheless, the boundary between languages and dialects is not always so clear-cut. On the one hand, two varieties can be rather different but still considered dialects of the same language. Writing about Estonian, Comrie (1981: 99) makes the following claim:

    On the other hand, two forms of speech can show relatively few contrasts yet still be considered two different languages. Such is the case with Danish and Swedish. Variation between these languages is no greater than variation within each of them. Writing about two dialects of Danish, Joergensen and Kristensen (1995: 153) stress that "the difference between a Jutland Danish and a Sealand Danish dialect is not linguistically distinct from the difference between Danish and Swedish." The fact that Danish and Swedish are deemed separate languages does have something to do with structural differences, but clearly there are other factors, especially political ones, which make the differences seem more significant than the contrasts between Jutland Danish and Sealand Danish. The importance of political factors is also evident in the history of some dialects of southern Sweden, which, according to Chambers and Trudgill (1980), were once thought of as varieties of Danish, though when Sweden acquired the region, these same varieties came to be considered dialects of Swedish.

    Another example of the political dimensions of the language/dialect issue is seen in the history of Macedonian, a South Slavic language (Herson-Finn 1996). About two hundred years ago Macedonian scholars started to analyze their own language, but they called it Bolgarski jazik (Bulgarian language). This judgment is not surprising because structurally, Macedonian resembles Bulgarian much more than it does any other Slavic language. On the other hand, the Yugoslavian government fostered a separate identity for Macedonia; in the second half of the twentieth century, the state encouraged linguists and educators to take some useful steps (e.g., writing grammars) for Macedonian to be considered a distinct language. With the break-up of Yugoslavia, Macedonia became a separate nation, and now the political independence of Macedonia complements its linguistic independence. Supporters of Ebonics might take heart from this example since it shows that yesterday's dialect can become today's language.

    4.1 Obtaining linguistic recognition

    While history shows that dialects can become languages, the Ebonics issue raises the question of just what is necessary for the change of status to occur. A school district's resolution is clearly not enough-not even a resolution with better linguistics and better writing than what the Oakland School Board served up. No declaration of independence, political or linguistic, has any force unless a significant number of people concur with the declaration. In the case of nations, a decisive number of people within the country must support the declaration, and they must act as people in independent nations act. Also important is other governments' recognition of a new state as part of the community of nations. For one legal scholar (Franck 1990), recognition is a "symbolic" act and, not surprisingly, is accompanied by old and elaborate rituals of protocol. As a symbolic act, recognition may not always correspond to the reality, as in the following example:

    Franck goes on to note that such symbolic acts "may have some constitutive effect, erecting a bridge from a present reality to a preferred tomorrow" (1990: 126), and he cites the early recognition of Israel by the United States as a case of where such symbolic action proved to be just such a bridge. However,

    One example cited by Franck of where the validation thus failed is the earlier American policy of recognizing the government on Taiwan as the legitimate government of mainland China up until the mid-1970's.

    Just as declaring and recognizing the independence of a nation are symbolic acts, so are proclamations of the existence of a language. As with the symbolic acts involving nationhood, those involving language may not always succeed in their aim. The question, then, is what can induce people to concur with a declaration of linguistic independence. Sympathy is clearly important, but in the case of the Ebonics issue, the extent of such sympathy is not clear, despite the letters and columns in the press mentioned earlier, or the resolution of the Linguistic Society of America (a resolution which actually equivocates on recognizing Ebonics as a separate language). If only a minority of Americans or other interested onlookers are convinced that there is some reality corresponding to the School Board's declaration, the use of the term Ebonics will be no more than a failed symbolic act. Ebonics supporters have a variety of options, but three seem to be especially helpful to convince large numbers of people: 1) creating a written language used for many purposes; 2) identifying distinctive traits through linguistic analysis and description; 3) developing distinctive norms based on the description that are adhered to by a large number of people. These options together can do much to give the world a sense that a distinct language exists, conferring what Stewart (1968) terms "autonomy." Accordingly, each option will be discussed in some detail.

    Linguists frequently note, with good reason, that writing is not essential for a language to exist. It is no exaggeration to say that for most of the languages of the world today, writing is a recent development, as in the case of Nivkh, a language long spoken in Siberia though it had no writing system until the 1930's (Comrie 1981). Yet even while writing is not a precondition for having a distinct language, literacy does help to convince people who may be skeptical about the existence of a language. It also helps enormously if authors create a wide variety of texts. In the Renaissance, writers of English felt it necessary to justify their desire to use English in essays and other literature that had traditionally been written in Latin (Baugh and Cable 1993). Enough readers found the justifications satisfactory, and nowadays virtually no one questions the use of English for any purpose in the English-speaking world.

    For a dialect to obtain linguistic recognition, literacy alone will not be enough. It is not unusual for some kinds of writing to represent vernacular speech: for centuries there have been representations of Irish English as well as Black English. Many of these representations were little more than attempts to poke fun at vernacular speakers-indeed, humor (or attempts at humor) may be the most common purpose found in so-called dialect writing. On the other hand, not all such writing seeks to parody speakers. Since the 19th century many authors including Walter Scott and George Bernard Shaw have sought to create literary characters who seem admirable despite-or actually because of-their use of vernacular forms. Yet even with sympathetic dialect writing, vernaculars are unlikely to be regarded as "real" languages unless authors use them for additional purposes. While Black English has had novelists such as James Wright and Zora Neale Hurston willing to champion characters who speak the dialect, there are not yet many editorials, news stories, science textbooks, sermons, or computer manuals written with structures such as the seven distinctive ones identified by Winford (Section 3.3). It is also significant that the Oakland School Board did not attempt to prepare parallel texts for their December resolution, with one version in Ebonics and the other in English. Until such practices become common-and that will be a long time in coming, if it ever happens-many members of the public will find it hard to see how Black English could ever be considered a "real" language.

    The second option that can help to achieve linguistic recognition is to identify distinctive traits through linguistic analysis and description. For over thirty years Black English has gotten considerable attention from linguists, and now there are detailed descriptions of many structures, including the seven that seem to be, as Winford argues, distinctive in terms of American speech. While Winford and others have noted structural areas that still remain only partly described, there is a solid base of descriptive work that could help the advocates. of linguistic independence.

    Closely related to the second option is the third: developing norms based on the description that are adhered to by a large number of people. With relatively small differences between Black English and other varieties, only a few norms can serve to demonstrate the existence of a separate language. If large numbers of authors begin to use the vernacular for the many kinds of writing discussed above, they will no doubt feel a need to show how their editorials, for example, are written in a separate language. In effect, a tradition of linguistic purism would likely develop, bringing with it the problems inherent in purism. In the future there may be calls for clear boundaries to distinguish a text in Ebonics from one in English, in effect, calls for what Uriel Weinreich termed "stability of form" in his classic study of language contact (1953/1968: 105). However, the purists issuing such calls would surely run afoul of critics ready to note that everyday speakers frequently mix what would be considered English with what would be considered Ebonics. Anti-purists would probably oppose attempts to have school children conform to the norms of a purist Ebonics, just as there are now many people ready to point out the burden placed upon children who speak a non-standard variety when they arrive in classrooms. Along with the problems of setting up norms that users would agree to follow, there are other difficulties, especially in the creation of reliable dictionaries, grammars, and other reference books. Even though much descriptive work has been done, the exact divergences between the independent language and English would have to be documented in areas such as word usage and style. Such challenges could no doubt be overcome, but doing so presupposes that there is strong public support for pursuing all three options. Such support seems unlikely any time soon.

    The three options suggested here may not be only ones that Ebonics advocates could pursue. Yet unless they try to develop Ebonics into a distinct written language, supporters will encounter much skepticism, in part because there are so few distinctive characteristics in speech. Skeptics could argue that Ebonics is no more different from other forms English than is, for example, vernacular Irish English or Appalachian English. While linguistic arguments might be devised to counter such objections, the strategy of creating a widely used written language would have a greater effect on the general public who, after all, have little background or interest in linguistics.

    4.2 Prospects

    The preceding discussion in this section noted the similarity between obtaining linguistic recognition and obtaining diplomatic recognition. Most importantly, people have to be convinced that a separate language or a separate nation exists. Despite the similarity, however, linguistic recognition does not depend on diplomatic recognition: there are many cases of distinct speech communities that have never had any kind of political independence. For example, Yiddish is widely recognized as a distinct language even though it is structurally similar to German in many ways, and even though the people who speak it have never lived in a single independent region where Yiddish was the national language. Nevertheless, Yiddish is generally associated with Jews who have traditionally lived in Central and Eastern Europe. This identification is aided by a number of other factors, including the distinctive writing system (similar to that of Hebrew) that Yiddish authors adopted. What Yiddish shows, then, is that neither geography nor political independence is a necessary condition for linguistic recognition. By the same logic, Ebonics could come to be considered a separate language.

    Possible ways for Ebonics to get linguistic recognition would be through the three options discussed in section 4.1: creating a written language used for many purposes, identifying distinctive traits through linguistic analysis and description, and developing norms based on the description that are adhered to by a large number of people. However, the first and third of these options will probably not be pursued any time in the near future, and for that reason it seems unlikely that Ebonics will achieve widespread linguistic recognition any time soon. The fact that few writers use vernacular forms such as He David, too in editorials or news stories, for example, suggests that most authors still consider Black English to be a nonstandard dialect unsuitable for written texts apart from fiction, drama, poetry, or advertising (where its use is also restricted). The prejudices that attach to other non-standard dialects are no doubt especially strong in the case of Black English since racial bias often reinforces linguistic bias. Under such circumstances, then, it is no surprise to find even extremely militant texts of black nationalists written in Standard English.

    If writers ever do change their stance and begin to write a wide range of texts in Ebonics (or whatever else they might call it), that would be an important first step toward linguistic independence. One group that might eventually adopt such a course is the Muslim population within the black community. While not all black Muslims view themselves as divorced from white society, many do, and linguistic separatism would follow naturally from the social separatism. Yet even if Muslims or any other authors pursue this option, there is still the problem of norms for the new written language. If the norms were seen as "Islamic," they would not be widely accepted by non-Muslims.

    Whether or not they developed within a Muslim or non- Muslim community, effective norms would likely take a lot of time and money to develop. Such difficulties should not be underestimated, especially since the American public would probably never agree to pay taxes to fund the creation of norms for anything with the name Ebonics. No matter how carefully thought out any norms might be, moreover, there would still be the problem of getting people to accept them and conform to them. Such acceptance seems possible only if writers first choose to use vernacular forms to create a new written standard, and the low prestige that Black English vernacular labors under will certainly work against the creation of such a standard.

    A less radical alternative to the declaration of the Oakland School Board would be to create a standardized version of Black English, one which has distinct norms but which is considered to be a dialect of English. If this strategy is ever pursued, the three steps already discussed will be helpful. The history of American English shows that it is indeed possible to create a new standard dialect. No standardized version existed before Noah Webster. An intense nationalist in the early years of American independence from Britain, Webster saw spelling as a way to help foster a sense of nationhood:

    The rhetoric here is clearly similar to what the Oakland School Board employs in its resolutions, though it should be noted that Webster's use of " national language" is equivocal. On the one hand, the tone is certainly separatist, yet on the other hand, the spellings he advocated were published in his Grammatical Institute of the English Language, the title of which indicates that American speech did not, for Webster, amount to a new language. A similar equivocation is seen in H.L. Mencken's American Language (1936), which, despite its title, considers American speech to be a dialect of English. Webster and Mencken were thus more moderate than what some of their own words suggest. If Webster's ambition was limited, he was nevertheless successful: many of the spellings he recommended in the Grammatical Institute did catch on. Even though there are still some people in and outside the United States who consider British English to be a superior variety, almost no one questions the notion that there is a standard American English. As such, American speech gets a certain recognition and respect not accorded to varieties such as Irish English and Black English.

    While creating a standardized form of Black English distinct from American English might lead to greater respect for black vernacular speech, there are clearly obstacles to such a project, obstacles only a little less formidable than those confronting Ebonics. The low status of Black English makes unlikely the prospect of any widespread public support for standardization. Moreover, it is not clear that there would be a critical mass within the black community for such a project. Finally, it is an open question as to whether supporters of Ebonics would agree to anything less drastic than the declaration of independence seen in the December and January resolutions.

    In the case of national identity, there is clearly a continuum from areas where political independence is not a goal to areas that are recognized as full-fledged nations. Between those extremes, some regions have a limited autonomy, as in the Aland Islands, a part of Finland but with their own flag and postage stamps, while other places aspire to getting recognition as completely independent states. In some cases, the aspirations remain unachieved. A break-away region of Moldavia has declared itself to be the Dnestr Moldavian Republic but so far has had almost no recognition from the rest of the world. A similar case is evident in the self-proclaimed Turkish Republic of Northern Cyprus, which no nation apart from Turkey has accorded diplomatic recognition. The ultimate fate of Ebonics remains unclear. For the moment, it resembles the Turkish Republic of Northern Cyprus in that few believe that black people speak a separate language even though some individuals have shown sympathy for the Oakland School Board's declaration of independence. It may be that Ebonics will achieve wider recognition some day, or perhaps the less ambitious strategy of creating a new standardized dialect of English will prevail. On the other hand, the problems of achieving either independence or autonomy remain formidable--and probably insurmountable.

    4.3 Summary and comment

    Because only a handful of structural features distinguish Black English from other dialects, other factors become especially important for those arguing that Ebonics should be considered a separate language. Any declaration of independence, political or linguistic, is ineffective unless a significant number of people recognize the new nation or language. For proponents of Ebonics three options together could help to achieve widespread linguistic recognition: creating a written language used for many purposes, identifying distinctive traits through linguistic description, and developing norms based on the description that are adhered to by a large number of people. Although languages can and do exist without writing systems, the practice of writing can do much to enhance the prestige of a language. Even so, the first and third of these options will probably not be pursued any time in the near future. The use of Black English in writing is largely restricted to literary representations, with few other genres such as editorials or school board resolutions appearing in the vernacular. Until writers become convinced that Ebonics is a suitable language for many purposes, the dominant perception will probably continue to be that the vernacular is not a "real" language.

    This paper has used both the terms Black English and Ebonics, and some readers may believe that it is impossible to decide which is the more accurate term to use. Some might be tempted to say that it is all in the eye of the beholder as to whether something is best considered a language or a dialect. However, such a position ignores the importance of structural factors. Even if many people agreed that Chinese and English were simply dialects of the same language, such a mistaken judgment would trivialize the language/dialect distinction: such a judgment would be tantamount to saying that there are no languages but only dialects of the same language. If the eye of the beholder is not enough to confirm something as a language or a dialect, the distinction rests on structural and functional principles. The structural principle can be stated as follows: the less alike two varieties are, the stronger the claim is that they are different languages, and the more alike two varieties are, the stronger the claim is that they are dialects of the same language. The functional principle complements the structural one: the more similar two varieties are, the weaker the claim is that they are two distinct languages unless both are used for largely the same purposes. As noted before, Black English is rarely used in written genres such as newspaper stories and editorials, textbooks, or school board resolutions. Unless it is used for the same purposes associated with Standard English, the claim that Ebonics is a separate language will remain weak. At present the term is more an expression of wishful thinking than a valid linguistic designator, and however much supporters of Ebonics dislike the term Black English, it more accurately reflects the structural and functional facts.

    5. General summary and conclusion

    Early media coverage of the Ebonics story suggested that the Oakland School Board had declared Ebonics to be a separate language and that the School Board sought to fund a bilingual education program for speakers of Ebonics to learn English. Moreover, the coverage suggested that actual classroom teaching would be done in Ebonics. The first of these claims is true: the Board's resolutions clearly deem Ebonics to be a separate language. The second claim is also true: despite denials by the Board, the language of the resolutions invokes a Federal law on bilingual education to justify funding for Ebonics programs. As for the third claim, the Board's resolutions say little about actual classroom teaching, though other evidence indicates that the Oakland School District was using contrastive analysis, i.e., systematic linguistic comparisons. Although contrastive analysis offers a promising approach to help with the teaching of Standard English, the credibility of the School Board may be questioned for a number of reasons, including the poor writing in the resolutions.

    Both the Oakland School Board and some of its critics have invoked historical arguments to support or deny the claim that Ebonics is a separate language. Although arguments on both sides have been poorly stated, they do marshal facts that must be acknowledged in any sound history of Black English, which has structures found in vernaculars of the British Isles but which also has structures showing parallels with Afro-American creoles, parallels not to be found in the British Isles. The School Board has argued that early influences from African languages are evident in twentieth century speech and that these influences alone make Ebonics a separate language. While there may indeed be African survivals, Black English has much more in common with other forms of English than it does with any Niger-Congo language, and so designations such as the School Board's "African Language Systems" are ludicrous.

    Even though there is no credible defense of the School Board's linguistic arguments, there do exist at least seven structures which could be used justify considering Ebonics a separate language. Even so, structural arguments are not always sufficient to decide the question of where a dialect ends and a language begins, especially not when only a handful of structural features can be invoked. In such cases, other factors assume extra importance. There are certain parallels between obtaining linguistic and diplomatic recognition since in either case it is necessary to have a significant number of people concur with the declaration of linguistic or political independence. For proponents of Ebonics, three options together could help to achieve widespread linguistic recognition: creating a written language used for many purposes, identifying distinctive traits through linguistic description, and developing norms based on the description that are adhered to by a large number of people. At present, the use of Black English in writing is largely restricted to literary representations, with few other genres such as editorials or school board resolutions appearing in the vernacular. Until writers become convinced that Ebonics is a suitable language for many purposes, the dominant perception will continue to be that the vernacular is not a "real" language.

    One of the most significant aspects of the Ebonics controversy is that it raises the question about the extent to which a language is in the eye of the beholder. If that eye is not sufficient, as argued in this paper, the distinction between a language and a dialect rests on structural and functional principles. The structural principle can be stated as follows: the less alike two varieties are, the stronger the claim is that they are different languages, and the more alike two varieties are, the stronger the claim is that they are dialects of the same language. The functional principle complements the structural one: the more similar two varieties are, the weaker the claim is that they are two distinct languages unless both are used for largely the same purposes. Until Ebonics is used for the same purposes associated with Standard English, the claim that it is a separate language will remain weak. Although changes in patterns of use could strengthen this claim in the future, the term Ebonics is at present more an expression of wishful thinking than a valid linguistic designator, and however much Ebonics supporters dislike the term Black English, it more accurately reflects the current linguistic facts.

    Acknowledgments I would like to thank Don Winford and an anonymous referee for their suggestions on revising this paper. As always, the responsibility for any shortcomings is mine alone.

    References

    Baugh, Albert, and Thomas Cable (1993) A History of the English Language . Englewood Cliffs, New Jersey: Prentice Hall.

    Chambers, J.K., and Trudgill, Peter (1980) Dialectology. Cambridge: Cambridge University Press.

    Comrie, Bernard (1981) The Languages of the Soviet Union. Cambridge: Cambridge University Press.

    Crystal, David (1987) The Cambridge Encyclopedia of Language. Cambridge: Cambridge University Press.

    Edwards, Walter, and Donald Winford (eds.) (1991) Verb Phrase Patterns in Black English and Creoles. Detroit: Wayne State University Press.

    Filppula, Markku (1986) Some Aspects of Hiberno-English in a Functional Sentence Perspective. Joensuu: Joensuu Publications in the Humanities.

    Franck, Thomas (1990) The Power of Legitimacy Among Nations. New York: Oxford: Oxford University Press.

    Harris, John (1986) Expanding the superstrate: habitual aspect markers in Atlantic Englishes, English World-Wide 7: 171-199.

    Herson-Finn, Viktoria (1996) What is nash: Toward a theory of ethnolect in the South Slavic dialect continuum. Unpublished Ph.D. dissertation, Ohio State University.

    Joergensen, J. Normann, and Kjeld Kristensen. (1995) On boundaries in linguistic continua. Language Variation and Change 7: 153-168.

    Keesing, Roger (1988) Melanesian Pidgin and the Oceanic Substrate. Stanford, California: Stanford University Press.

    Lawton, David (1983) English in the Caribbean. In English as a World Language, ed. by Richard Bailey and Manfred Görlach. Ann Arbor: University of Michigan Press.

    McWhorter, John (1992) Substratal influence in Saramaccan serial verb constructions. Journal of Pidgin and Creole Languages 7:1-55.

    Mencken, Henry Louis (1936) The American Language. New York: Knopf.

    Mufwene, Salikoko (1990) Serialization and subordination in Gullah: Toward a definition of serial verbs. In When Verbs Collide: Papers from the 1990 Ohio State Mini-Conference on Serial Verbs., ed. by Brian Joseph and Arnold Zwicky. Columbus, Ohio: Department of Linguistics, Ohio State University.

    ---.(1991). Is Gullah decreolizing? In The Emergence of Black English, ed. by Guy Bailey, Natalie Maynor, and Patricia Cukor-Avila. Amsterdam: John Benjamins.

    ---.(ed.) (1993) Africanisms in Afro-American Language Varieties. Athens, Georgia: University of Georgia.

    Muysken, Pieter. (1984) The Spanish that Quechua speakers learn: L2 learning as norm-governed behavior. In Second Languages, ed. by Roger Andersen. Rowley, Massachusetts: Newbury House.

    Niles, Norma Anita (1980) Provincial English dialects and Barbadian English. Unpublished Ph.D. dissertation, University of Michigan.

    Oakland School Board Resolutions. Available on the LINGUIST List Vol. 8-53, January 18, 1997. Also available on the Internet: http://jan.UCC.nau.edu/~jmw22/1stOaklandRes.html.

    ---.(1997) Synopsis of the Adopted Policy on Standard American English Language Development. Available on the Internet: http://www.ousd.k12.ca.us/oakland.standard.html

    Odlin, Terence (1989) Language Transfer. Cambridge: Cambridge University Press.

    ---.(1991a): Toward a definition of "target language." Presented at ALMS: Applied Linguistics at Michigan State, Michigan State University, East Lansing, Michigan.

    ---.(1991b) Irish English idioms and language transfer. English World-Wide 12:175-193.

    ---.(1992) Transferablility and linguistic substrates, Second Language Research 8: 171-202.

    ---.(1997) Hiberno-English: pidgin, creole, or neither? CLCS Occasional Paper No. 49. Centre for Language and Communication Studies, Trinity College Dublin.

    Rickford, John (1986) Social contact and linguistic diffusion. Language 62:245-289.

    ---.(1996) The Oakland Ebonics decision: Commendable attack on the problem. San Jose Mercury News, December 26, 1996.

    ---.(1997) The evolution of the Ebonics issue. Paper on the Internet. http://www.leland.stanford.edu/~rickford/ebonics/NYTOpEd/html

    Singler, John (1988) The homogeneity of the substrate as a factor in pidgin/creole genesis. Language 64:27-51.

    Smith, Norval, Ian Robertson, and Kay Williamson (1987) The Ijo element in Berbice Dutch. Language in Society 16: 49-90.

    Stewart, William (1968) A sociolinguistic typology for describing national multilingualism. In Readings in the Sociology of Language, ed. by Joshua Fishman. The Hague: Mouton

    Thomason, Sarah, and Terrence Kaufman (1988) Language Contact, Creolization, and Genetic Linguistics. Berkeley: University of California Press.

    Turner, Lorenzo (1949) Africanisms in the Gullah Dialect. Chicago: University of Chicago Press.

    Webster, Noah (1789/1966) An essay on spelling. In The English Language, ed. by W.F. Bolton. Cambridge: Cambridge University Press.

    Weinreich, Uriel. 1953/1968. Languages in Contact. The Hague: Mouton.

    Weldon, Tracey (1998) Examining the AAVE-Gullah relationship: A study of copula variability. Unpublished Ph.D. dissertation, Ohio State University.

    Whatley, Elizabeth (1981) Language among Black Americans. In Language in the U.S.A., ed. by Charles Ferguson and Shirley Brice Heath. Cambridge: Cambridge University Press.

    Williams, Robert (ed.) (1975) Ebonics: The True Language of Black Folks. St. Louis, Missouri: Institute of Black Studies.

    Winford, Donald (1993) Predication in Caribbean English Creoles. Amsterdam: John Benjamins

    ---.(1997) On the origins of African American Vernacular English: A creolist perspective. I: The Sociohistorical Background. Diachronica 14: 305-344.

    ---. (1998) On the origins of African American Vernacular English: A creolist perspective. II Diachronica.



    APPENDIX 1

    Resolution On The Oakland "Ebonics" Issue by the Linguistic Society of America on January 3, l997. The bibliography attached to the resolution has been omitted.

    Whereas there has been a great deal of discussion in the media and among the American public about the l8 December l996 decision of the Oakland School Board to recognize the language variety spoken by many African American students and to take it into account in teaching Standard English, the Linguistic Society of America, as a society of scholars engaged in the scientific study of language, hereby resolves to make it known that:

    1. The variety known as "Ebonics," "African American Vernacular English" (AAVE), and "Vernacular Black English" and by other names is systematic and rule-governed like all natural speech varieties. In fact, all human linguistic systems -- spoken, signed, and written -- are fundamentally regular. The systematic and expressive nature of the grammar and pronunciation patterns of the African American vernacular has been established by numerous scientific studies over the past thirty years. Characterizations of Ebonics as "slang," "mutant," "lazy," "defective," "ungrammatical," or "broken English" are incorrect and demeaning.

    2. The distinction between "languages" and "dialects" is usually made more on social and political grounds than on purely linguistic ones. For example, different varieties of Chinese are popularly regarded as "dialects," though their speakers cannot understand each other, but speakers of Swedish and Norwegian, which are regarded as separate "languages," generally understand each other. What is important from a linguistic and educational point of view is not whether AAVE is called a "language" or a "dialect" but rather that its systematicity be recognized.

    3. As affirmed in the LSA Statement of Language Rights (June l996), there are individual and group benefits to maintaining vernacular speech varieties and there are scientific and human advantages to linguistic diversity. For those living in the United States there are also benefits in acquiring Standard English and resources should be made available to all who aspire to mastery of Standard English. The Oakland School Board's commitment to helping students master Standard English is commendable.

    4. There is evidence from Sweden, the US, and other countries that speakers of other varieties can be aided in their learning of the standard variety by pedagogical approaches which recognize the legitimacy of the other varieties of a language. From this perspective, the Oakland School Board's decision to recognize the vernacular of African American students in teaching them Standard English is linguistically and pedagogically sound.

    APPENDIX 2

    Complete text of the Ebonics resolution and policy statement adopted by the Oakland School Board on December 18, 1996. [Paragraph numbers and emphases have been added by me; bold italics (e.g., in paragraph 9) indicate some of the parts deleted in the revised resolution of January 18, 1997.]

    RESOLUTION OF THE BOARD OF EDUCATION ADOPTING THE REPORT AND RECOMMENDATIONS OF THE AFRICAN-AMERICAN TASK FORCE; A POLICY STATEMENT AND DIRECTING THE SUPERINTENDENT OF SCHOOLS TO DEVISE A PROGRAM TO IMPROVE THE ENGLISH LANGUAGE ACQUISITION AND APPLICATION SKILLS OF AFRICAN-AMERICAN STUDENTS.

    No. $597-0063

    [1] Whereas, numerous validated scholarly studies demonstrate that African American students as part of their culture and history as African people possess and utilize a language described in various scholarly approaches as ``Ebonics'' (literally Black sounds) or Pan African Communication Behaviors or African Language Systems; and

    [2] Whereas, these studies have also demonstrated that African Language Systems are genetically-based and not a dialect of English; and

    [3] Whereas, these studies demonstrate that such West and Niger-Congo African languages have been officially recognized and addressed in the mainstream public educational community as worth of study, understanding or application of its principles, laws and structures for the benefit of African American students both in terms of positive appreciation of the language and these students' acquisition and mastery of English language skills; and

    [4] Whereas, such recognition by scholars has given rise over the past 15 years to legislation passed by the State of California recognizing the unique language stature of descendants of slaves, with such legislation being prejudicially and unconstitutionally vetoed repeatedly by various California state governors; and

    [5] Whereas, judicial cases in states other than California have recognized the unique language stature of African American pupils, and such recognition by courts has resulted in court-mandated educational programs which have substantially benefitted African American children in the interest of vindicating their equal protection of the law rights under the 14th Amendment to the United States Constitution; and

    [6] Whereas, the Federal Bilingual Education Act (20 USC 1402 et seq.) mandates that local educational agencies ``build their capacities to establish, implement and sustain programs of instruction for children and youth of limited English proficiency,'' and

    [7] Whereas, the interests of the Oakland Unified School District in providing equal opportunities for all of its students dictate limited English proficient educational programs recognizing the English language acquisition and improvement skills of African American students are as fundamental as is application of bilingual education principles for others whose primary languages are other than English; and

    [8] Whereas, the standardized tests and grade scores of African American students in reading and language art skills measuring their application of English skills are substantially below state and national norms and that such deficiencies will be remedied by application of a program featuring African Language Systems principles in instructing African American children both in their primary language and in English, and

    [9] Whereas, standardized tests and grade scores will be remedied by application of a program with teachers and aides who are certified in the methodology of featuring African Language Systems principles in instructing African American children both in their primary language and in English. The certified teachers of these students will be provided incentives including, but not limited to salary differentials,

    [10] Now, therefore, be it resolved that the Board of Education officially recognizes the existence and the cultural and historic bases of West and Niger-Congo African Language Systems, and each language as the predominantly primary language of African American students; and

    [11] Be it further resolved that the Board of Education hereby adopts the report recommendations and attached Policy Statement of the District's African American Task Force on language stature of African American speech; and

    [12] Be it further resolved that the Superintendent in conjunction with her staff shall immediately devise and implement the best possible academic program for imparting instruction to African American students in their primary language for the combined purposes of maintaining the legitimacy and richness of such language whether it is known as ``Ebonics,'' ``African Language Systems,'' ``Pan African Communication Behaviors'' or other description, and to facilitate their acquisition and mastery of English language skills; and

    [13] Be it further resolved that the Board of Education hereby commits to earmark District general and special funding as is reasonably necessary and appropriate to enable the Superintendent and her staff to accomplish the foregoing; and

    [14] Be it further resolved that the Superintendent and her staff shall utilize the input of the entire Oakland educational community as well as state and federal scholarly and educational input in devising such a program; and

    [15] Be it further resolved, that periodic reports on the progress of the creation and implementation of such an educational program shall be made to Board of Education at least once per month commencing at the Board meeting of December 18, 1996.

    POLICY STATEMENT

    [PS 1] There is persuasive empirical evidence that, predicated on analysis of the phonology, morphology and syntax that currently exists as systematic, rule governed and predictable patterns exist in the grammar of African-American speech. The validated and persuasive linguistic evidence is that African-Americans (1) have retained a West and Niger-Congo African linguistic structure in the substratum of their speech and (2) by this criteria are not native speakers of black dialect or any other dialect of English.

    [PS 2] Moreover, there is persuasive empirical evidence that, owing to their history as United States slave descendants of West and Niger-Congo African origin, to the extent that African-Americans have been born into, reared in, and continue to live in linguistic environments that are different from the Euro-American English speaking population, African-American people and their children, are from home environments in which a language other than English language is dominant within the meaning of "environment where a Language other than English is dominant" as defined in Public Law 1-13-382 (20 U.S.C. 7402, et seq.).

    [PS 3] The policy of the Oakland Unified School District (OUSD) is that all pupils are equal and are to be treated equally. Hence, all pupils who have difficulty speaking, reading, writing or understanding the English language and whose difficulties may deny to them the opportunity to learn successfully in classrooms where the language of instruction is English or to participate fully in classrooms where the language of instruction is English or to participate fully in our society are to be treated equally regardless of their race or national origin.

    [PS 4] As in the case of Asian-American, Latino-American, Native American and all other pupils in this District who come from backgrounds or environments where a language other than English is dominant, African-American pupils shall not, because of their race, be subtly dehumanized, stigmatized, discriminated against or denied. Asian-American, Latino-American, Native American and all other language different children are provided general funds for bilingual education, English as Second Language (ESL) and State and Federal (Title VIII) Bilingual education programs to address their limited and non-English proficient (LEP/NEP) needs. African-American pupils are equally entitled to be tested and, where appropriate, shall be provided general funds and State and Federal (Title VIII) bilingual education and ESL programs to specifically address their LEP/NEP needs.

    [PS 5] All classroom teachers and aids who are bilingual in Nigritian Ebonics (African-American Language) and English shall be given the same salary differentials and merit increases that are provided to teachers of the non-African American LEP pupils in the OUSD.

    [PS 6] With a view toward assuring that parent of African-American pupils are given the knowledge base necessary to make informed choices, it shall be the policy of the Oakland Unified School District that all parents of LEP (Limited English Proficient) pupils are to be provided the opportunity to partake of any and all language and culture specific teacher education and training classes designed to address their child's LEP needs.

    [PS 7] On all home language surveys given to parents of pupils requesting home language identification or designations, a description of the District's programmatic consequences of their choices will be contained.

    [PS 8] Nothing in this Policy shall preclude or prevent African-American parents who view their child's limited English proficiency as being non-standard English, as opposed to being West and Niger-Congo African Language based, from exercising their right to choose and to have their child's speech disorders and English Language deficits addressed by special education and/or other District programs.

    APPENDIX 3

    The revised resolution passed by the Oakland School Board on January 15 1997:

    AMENDED RESOLUTION OF THE BOARD OF EDUCATION ADOPTING THE REPORT AND RECOMMENDATIONS OF THE AFRICAN-AMERICAN TASK FORCE;

    A POLICY STATEMENT AND DIRECTING THE SUPERINTENDENT OF SCHOOLS TO DEVISE A PROGRAM TO IMPROVE THE ENGLISH LANGUAGE ACQUISITION AND APPLICATION SKILLS OF AFRICAN-AMERICAN STUDENTS

    No. 9697-0063

    [1] Whereas, numerous validated scholarly studies demonstrate that African-American students as a part of their culture and history as African people possess and utilize a language described in various scholarly approaches as "Ebonics" (literally "Black sounds") or "Pan African Communication Behaviors" or "African Language Systems"; and

    [2] Whereas, these studies have also demonstrated that African Language Systems have origins in West and Niger-Congo languages and are not merely dialects of English; and

    [3] Whereas, these studies demonstrate that such West and Niger-Congo African languages have been recognized and addressed in the educational community as worthy of study, understanding and application of their principles, laws and structures for the benefit of African-American students both in terms of positive appreciation of the language and these students' acquisition and mastery of English language skills; and

    [4] Whereas, such recognition by scholars has given rise over the past fifteen years to legislation passed by the State of California recognizing the unique language stature of descendants of slaves, with such legislation being vetoed repeatedly by various California state governors; and

    [5] Whereas, judicial cases in states other than California have recognized the unique language stature of African American pupils, and such recognition by courts has resulted in court-mandated educational programs which have substantially benefited African-American children in the interest of vindicating their equal protection of the law rights under the Fourteenth Amendment to the United States Constitution; and

    [6] Whereas, the Federal Bilingual Education Act (20 U.S.C. 1402 et seq.) mandates that local educational agencies "build their capacities to establish, implement and sustain programs of instruction for children and youth of limited English proficiency; and

    [7] Whereas, the interest of the Oakland Unified School District in providing equal opportunities for all of its students dictate limited English proficient educational programs recognizing the English language acquisition and improvement skills of African-American students are as fundamental as is application of bilingual or second language learner principles for others whose primary languages are other than English. Primary languages are the language patterns children bring to school; and

    [8] Whereas, the standardized tests and grade scores of African-American students in reading and language arts skills measuring their application of English skills are substantially below state and national norms and that such deficiencies shall be remedied by application of a program featuring African Language Systems principles to move students from the language patterns they bring to school to English proficiency; and

    [9] Whereas, standardized tests and grade scores will be remedied by application of a program that teachers and instructional assistants, who are certified in the methodology of African Language Systems principles used to transition students from the language patterns they bring to school to English. The certified teachers of these students will be provided incentives including, but not limited to salary differentials;

    [10] Now, therefore, be it resolved that the Board of Education officially recognizes the existence, and the cultural and historic bases of West and Niger-Congo African Language Systems, and these are the language patterns that many African-American students bring to school; and

    [11] Be it further resolved that the Board of Education hereby adopts the report, recommendations and attached Policy Statement of the District's African-American Task Force on the language stature of African-American speech; and

    [12] Be it further resolved that the Superintendent in conjunction with her staff shall immediately devise and implement the best possible academic program for the combined purposes of facilitating the acquisition and mastery of English language skills, while respecting and embracing the legitimacy and richness of the language patterns whether they are known as "Ebonics", "African Language Systems", "Pan African Communication Behaviors", or other description; and

    [13] Be it further resolved that the Board of Education hereby commits to earmark District general and special funding as is reasonably necessary and appropriate to enable the Superintendent and her staff to accomplish the foregoing; and

    [14] Be it further resolved that the Superintendent and her staff shall utilize the input of the entire Oakland educational community as well as state and federal scholarly and educational input in devising such a program; and

    [15] Be it further resolved that periodic reports on the progress of the creation and implementation of such an educational program shall be made to the Board of Education at least once per month commencing at the Board meeting of December 18, 1996.