Exploring Ebonics: Linking Origins, Phonology, and Social Issues

Colin Widdifield

Without a doubt, the controversy surrounding Black English (BE), or Ebonics, "dwarfs" (Wolfram, 1991, p. 106) that of other varieties of English. In fact, there is so much information on Ebonics that one search on the Internet found "2,214 pages" containing the word "Ebonics". But, definitions of Black English, or more recently Ebonics, have not all been "compatible" (Bailey, 1993, p. 289) and there is really "no clear definition" (Schneider, 1989, p. 4). One description states that Black English, Black Dialect and Black Idiom are just some of the many terms used when discussing the speech used by eighty to ninety percent of American Blacks (Smitherman, 1977, p. 1-2). Another view is that Black English can be understood as a dialect spoken by members of the black population in the "socioeconomically lower class" (Schneider, 1989, p. 5). A simpler, but still vague, idea is that it is the casual speech of "lower-class black Americans" (Schneider, 1989, p. 8).
Equally slippery is the debate on the origin of Ebonics. Is this variety of English uniquely different from other varieties of English due to its purported ties to West Africa, or is Ebonics strictly a product of British and American English, or is BE possibly a product of both positions? In keeping with the theme of the course, this essay will explore the origin question in terms of phonology. Scholarly works on the phonological features of Black English will be analyzed to try to find out which of these positions has the greatest support. Links will be drawn from this modest analysis to the pertinent social issues of linguistic convergence/divergence of Black and White English Vernaculars, and Ebonics in education. These links will become clearer as the discussion progresses.

In this essay, as in the literature, some of these like terms for Ebonics will be used interchangeably in order to avoid the monotonous use of one single term. Teaching American English Pronunciation (Avery & Ehrlich, 1996) will serve as a model for standard English, and as a guide for the transcriptions.

Specifically, the purpose of this essay is to discuss Ebonics in terms of its origins, phonology and these two pertinent social issues, all of which are linked.


The term "origins" is used in the plural because of its dual references, one being the origin of the term "Ebonics", and the second being the origin or beginnings of "Ebonics" itself.
Black English, Negro Non-Standard English and Merican (Dillard, 1972, p. ix) were just some of the terms commonly used before the word "Ebonics" was coined. After a broad search in such sources as academic journals, books, magazines, and on the Internet, there does not seem to be a clear picture as to when or by whom this term "Ebonics" was coined. A source on the Internet called Leslie Frieden`s Ebonics Page claims that the term "Ebonics" was first used in the title of a 1975 book called Ebonics: The Language of Black Folk. Similarly, a publication of the Linguistic Society of America cites a 1975 book by Robert L. Williams titled Ebonics: The True Language of Black Folks (LSA, 1997, p. 3). These two works could be the same, due to the same dates and similar titles. Another source explains that Ebonics was coined in the early 1970s from the two words "ebony" and "phonics" (McMillen, 1997, p. A16). A further source points to a number of "self-coined labels" (Schneider, 1989, p. 9) for BE, with "Ebonics" first appearing in a 1975 dissertation by E.A. Smith called The Evolution and Continuing Presence of the African Oral Tradition in Black America. A rough assumption seems to place the origin of the term "Ebonics" somewhere in the early to mid-1970s, with Williams or Smith as the inventor of this term.

Although Ebonics also has lexical, morphological and syntactic features that differ from standard and nonstandard English, this paper will focus on the phonology of Ebonics as much as possible. In order to provide the reader with some insight into the historical arguments, a brief outline dealing with the beginnings of Black English will precede the modest phonological analysis of the scholarly literature.

The origin of Ebonics has not been "completely resolved" (Bailey, 1993, p. 287) and remains "indeterminate" (Fasold, 1981, p. 165). However, there are two main hypotheses about the origin of Ebonics: the creolist hypothesis and the anglicist or dialectologist hypothesis.

The "most complete argument" (Wolfram, 1991, p. 112) for the first hypothesis is a book titled Black English: Its History and Usage in the United States. This work fervently claims that the differences between Black English and other English dialects are traceable "specifically" (Dillard, 1972, p. ix) to influences from West Africa. Dillard believes that Black English originates from a creolized variety of English, which was based upon a pidgin spoken by slaves from the West Coast of Africa (Dillard, 1972, p. 6). He further claims that both the culture and the language of most Blacks remain "distinct from those of any large group of whites" (Dillard, 1972, p. 5). This statement is made in the absence of tables, references or any sort of hard evidence, but in his words "the evidence of one's senses", which seems to sum up the approach of this work. He attacks the "anglicist" point of view, and accuses proponents of taking the "most blatantly racist position" (Dillard, 1972, p. 10).

In the chapter "A Sketch of the History of Black English" (Dillard, 1972, p. 73-138), about one-quarter of his book, slave speech is analyzed in order to justify the creole-origin hypothesis. These samples of slave speech were written by people who had contact with slaves, such as captains of slave ships. Dillard states that by the end of the 1700s, there is a clear picture of the Black language due to "sufficient accounts" (Dillard, 1972, p. 87) of slave speech. However, this view is seen as being difficult to test, due to the "paucity" (Montgomery, 1991, p. 179-180) of early documentary material on speech patterns. Further, there is still "no unambiguous evidence" (Holm, 1991, p. 244) that the English of Blacks in North America was ever totally creolized. One Nigerian scholar called the idea of Ebonics having a West African origin a thesis of "nonsense" (Abati, 1997, p. 1). Butters observes that social agendas associated with the creole-origin hypothesis have mainly served to "muddy the linguistic waters" (Butters, 1989, p. 4). Dillard`s accusatory views throughout his work such as "The prejudice, as usual, has been against the American Negro" (Dillard, 1972, p. 116), certainly do not seem to clarify these waters.

The "most extensive rationale" (Wolfram, 1991, p. 116) for the anglicist, or dialectologist, hypothesis is a book titled American Earlier Black English. This book, like Butters`, also acknowledges that the history of Black English has been dominated by "social rather than linguistic issues" (Schneider, 1989, p. 1).

The anglicist hypothesis maintains that Black English is probably based on British and American English, and not on West African languages, as the creolists maintain. For example, in some words, the merging of /_/ and /_/ before nasals was found to be "borrowed from England" (Butters, 1989, p. 152-153). Schneider refutes the creole-origin hypothesis on a number of grounds (Schneider, 1989, p. 33-34). Historical information does not support the belief that slaves transported to America already spoke a uniform pidgin English, but more likely started learning English after reaching the North American continent. Furthermore, slaveholders commonly separated slaves of the same speech group in order in avoid insurrection. (One wonders, though, how these slaveholders separated slaves of the same speech group; a task that one imagines to be easier said than done). Schneider goes on to say that English was the only means by which to converse, and African languages were practically never spoken. Ultimately, slaves either spoke a pidgin, later a creole, or on the other hand, completely adapted to the English spoken on the plantation, with both scenarios depending upon degree of language contact. Another weakness with the creolist view is that "reliable source material" (Schneider, 1989, p. 2) does not exist, and therefore one cannot substantiate that Ebonics arose from a creole stage. Current knowledge about early BE is "highly unreliable and anecdotal" (Schneider, 1989, p. 2). Also he describes the reliability of historical literary works by ship captains and others involved with slavery as being "more than questionable" (Schneider, 1989, p. 3).

Schneider summarizes his analysis by stating that most of the features of early Black English originated from British and American English of the colonial period, with Whites and Blacks retaining and losing linguistic features over time (Schneider, 1989, p. 276-278). He notes however that some Black slaves in America did possess creole-influenced structures, though only in very restricted and independent environments (Schneider, 1989, p. 278). These findings led him to conclude that the Ebonics of today is not the descendant of a creole (Schneider, 1989, p. 278). However, based on the data Schneider analyzed, some of his conclusions "have to be viewed with caution" (Wolfram, 1990, p. 126). Schneider also seems to spot this, as he hedges his conclusion by stating that it may not be "totally" correct.

The strongly opposing views of Dillard and Schneider have been criticized for the tendency to "restrict their attention" (Winford, 1992, p. 351) to only those features which support their respective positions. Similarly, Fasold states that such sharp disagreements are mainly due to differences in collecting and analyzing the data, i.e. differences in "methodology" (Fasold, 1981, p. 165).

Baugh`s view of the creolist hypothesis is that Black self-respect will be enhanced by the truth, and should not be based upon evidence which is highly questionable (Baugh, 1983, p. 17-18). He feels that combining the creolist and the anglicist hypotheses is the most accurate position when the best historical evidence is considered (Baugh, 1983, p. 18). Baugh sees that the most current studies point to a more logical, "combined hypothesis" (Baugh, 1983, p. 12). Winford also questions both sides of the debate, and favors a position whereby the creolist and anglicist views might be "reconciled" (Winford, 1992, p. 351). These "medial positions" will figure prominently in the phonological analysis.

Phonological analysis

In order to retain the specific examples and comments, the following sets of features are presented in detail, and not in chart form. Including specific examples and comments will provide the reader with a broader sense of what seems to be the "common core of features" (Wolfram, 1991, p. 111) of Ebonics. It is important to note that the following sets of phonological data have been modified to conform to the phonetic symbols on page 7 of the course textbook.
In the appendix of his book, Dillard discusses the pronunciation of Black English. He explains that these phonological features are generalizations, and may not be exclusively valid for any one speaker (Dillard, 1972, p. 308-309):

Diphthongs are monophthongized, e.g. allen = island. (Shared with many Southern White dialects)
Initial [D ] ® [d], in words such as the, then, that, those, though, there. (Shared with some Northern [White] dialects, but connection is extremely unlikely)
Final [q ] ® [f], in words such as with, both, birth. (This process sometimes occurs in Black English and has a "stylistic relationship" [not clearly defined] which is not present in the White dialects)
Medial [D ] ® [v], in words such as mother, other, brother. (This process never occurs in standard English, and also has this "stylistic" difference)
Deletion of final [r]. (This occurs in many English dialects, but is more widespread in the Black community and in some areas only occurs within the Black community)
Smitherman states that the following phonological features are used by large numbers of black speakers" (Smitherman, 1977, p. 17-18):
Initial [D ] =[d], e.g. them = dem and then = den
Final [q ] =[f], e.g. south = souf and mouth = mouf
Deletion of middle and final [l] and [r], e.g. help =hep and star = sta
Deletion of word-final stops, e.g. hood = hoo and bed = be
Final consonant cluster reduction, e.g. test = tes and wasp = was
[IN] = [ÊN], e.g. thing = thang, ring = rang, sing =sang
Diphthongs are monophthongized
Another set of features comes from Childs (Childs, 1996, March 4):
Diphthongs are often monophthongized [ay] ® [a], [aw] ® [a], [oy] ® [a], e.g. homophonous pairs find = fond, proud = prod, oil = all. (Also found in some White dialects)
Neutralization of vowel contrasts before nasals [I ] = [e ], e.g. homophonous
pairs of pin = pen, Lynne = Len, sinned = send, bin = Ben. (Also found in White dialects)
Final interdentals change to labiodentals, e.g. with = [wIf] or [wIv], bath = [bÊf], both = [bowf] (Also found Afrikaans and Cockney English)
Absence of post-vocalic [l] and [r], e.g. help = hep, guard = god(Also found in some White dialects)
Final consonant cluster reduction, e.g. six = sick and first = firs
Baugh found that the following features account for some of the "major differences" (Baugh, 1983, p. 93) between street speech and standard English:
Deletion of postvocalic [r], e.g. car = ca
Deletion of word-final stops [t] and [d] when there is no semantic value e.g. lost = los and bold = bol
In order to put these scholarly views into perspective, this analysis will make use of a threefold distinction, which is "often not made" (Fasold, 1981, p. 167) in the literature. These distinctions are as follows: unmarked features (i.e. features shared with standard English); nonstandard features (i.e. features shared with nonstandard White varieties; and unique features (i.e. features which seem to occur only in Black English). To keep track of the five authors and all phonological features, a bold, capital letter will follow each feature. This capital letter will correspond to the authors` views \ with F for Fasold, D for Dillard, S for Smitherman, C for Childs, B for Baugh.

Unmarked features

Neutralization of [I ] and_[e ] before nasals F
Deletion of postvocalic [r] and [l] F
Deletion of final "g", e.g. walking = walkin` F
Initial "th" ® [d] or [t], e.g. those = dose, throw = trow F
Non-initial "th" ® [f] or [v] F
Deletion of word-final stops, e.g. bad = ba` F
Final consonant cluster reduction, e.g. west side = [w_sayd] F, S, C

Nonstandard features

In final consonant clusters [k] ® [t], e.g. ask becomes ast in "I`ll go ast `er". (Could be a lexical variant and not phonological) F
Neutralization of [I ] and_[e ] before nasals S, C
Deletion of postvocalic [r] and/or [l] D, S, C, B
Initial "th" ® [d] or [t] D, S
Non-initial "th" ® [f] or [v] D, S, C
Deletion of word-final stops S, B
Diphthongs are monophthongized D, S, C

Unique BE features

In the devoicing of final stops in stressed syllables, there is a noticeable glottal coarticulation, e.g. bed = [__!_] F
To recapitulate, this modest analysis shows that most BE features are shared with standard English, or nonstandard White varieties, while only one feature was found to be unique to BE. This analysis favors the view that most of the phonological features of Ebonics are neither strictly a product of West African influences, nor British or American influences, but possibly a product of "both positions", as mentioned in the introduction. This finding contradicts the view of Dillard. It also runs counter to Smitherman's position as he holds that the major structures of Black English are "of course" (Smitherman, 1977, p. 9) based on adaptations of African language rules. These two views seem implausible as both standard English and some or many White varieties share the same phonological features with Ebonics. If grammatical features were considered, however, Dillard`s and Smitherman`s views would have had more support in the area of "unique BE features".

Tabulation errors could have resulted from the placing of all but one of Smitherman`s features into the nonstandard features distinction. This was done because she did not include any mention of standard English or White dialects. All features of final consonant cluster reduction were placed in the unmarked features distinction because this was the only feature that appeared in the course text used to model standard English (Avery & Ehrlich, 1996, p. 86).

The differences in the number of features within each set of data may be, in part, because Ebonics has not been standardized. Although English has been standardized, Wardhaugh and other linguists believe that Ebonics has not undergone this process (Wardhaugh, 1994, p. 30).

Fasold concludes that the differences between Black and White English are "few and subtle" (Fasold, 1981, p. 186). Similarly, Schneider states that most researchers would agree that the differences between Black and White speech are "rather superficial in nature" (Schneider, 1989, p. 2). Past studies have shown that some linguistic features are used more frequently among Blacks, however very few, if any, are used exclusively by Blacks (Schneider, 1989, p. 2). In addition, Bailey sees that significant numbers of African-Americans speak varieties of English identical to those spoken by comparable Whites (Bailey, 1993, p. 312).

Fasold states that the apparent differences between the speech of Blacks and Whites are of theoretical interest, but "by no means" (Fasold, 1981, p. 164) do they indicate widespread differences in phonology. However, including factors such as age, education, sex, and social position (Schneider, 1989, p. 256) could have improved the precision of the phonological data presented in this essay. Linguistic research into Black English "cannot be detached" (Schneider, 1989, p. 10) from these social contexts. For this reason, this discussion will now focus on two pertinent social issues surrounding Ebonics.

Pertinent social issues

Linguistic convergence/divergence

These two hypotheses focus on BEV (Black English Vernacular) and WEV (White English Vernacular), and do not directly involve "standard English" (Butters, 1989, p. 196). The debate over linguistic convergence/divergence carries great social importance, with convergence resulting in the death of BEV and divergence resulting in significant social division (Butters, 1989, p. 116-117).
In his extensive research, Butters found evidence for convergence in North Carolina. This evidence led him to consider that consonant cluster simplification was a "good indication" (Butters, 1989, p. 119) of convergence in BEV and WEV. He also found that the prenasal merger of /_/ and /_/ in BEV and WEV in the southern United States "certainly supports" (Butters, 1989, p. 155) the convergence hypothesis. An example of this merger is the widely recognized homophonous pair pin/pen. Bailey points out that convergent features are all older and divergent features are "almost all recent" (Bailey, 1993, p. 311). This recent finding seems to explain why the divergence hypothesis is much more widely discussed.

The divergence hypothesis is a "very controversial claim" (Childs, 1996, March 24). It is closely linked to important social and historical issues such as whether or not Blacks and Whites are becoming increasingly distant, causing socioeconomic and linguistic segregation (Butters, 1989, p. 1-2). This claim has been put forward by serious scholars and meticulous researchers such as Labov and Bailey, despite the lack of "direct linguistic evidence" (Butters, 1989, p. 63). There is really "no simple answer" (Myhill, 1988, p. 320) to this question. Compounding the difficulties is the fact that there are so many linguistic features which need to be given more "extensive treatment" (Butters, 1989, p. 143). Butters feels that a divergence hypothesis seems "uncalled for" (Butters, 1989, p. 181). He cites examples from American and English history which show that despite great lengths of time, most people and communities have remained mutually intelligible and have not "drifted dangerously" (Butters, 1989, p. 181).

The evidence that does exist on the convergence/divergence hypotheses is contradictory and too weak to predict the future (Butters, 1989, p. 117). It seems that for the "immediate future" (Wolfram, 1991, p. 115-116), no radical restructuring of Vernacular Black English will occur toward or away from other vernaculars. Furthermore, there is "no certain way" (Butters, 1989, p. 162) of knowing whether or not the absence of change in dialect B, or the presence of change in dialect A, is temporary or permanent.

Another view is that BEV and WEV are probably converging in some ways and diverging in others, and the extent of this process varies widely, depending upon who we consider to be typical speakers of BEV (Myhill, 1988, p. 320). Also, Butters states that it is not at all surprising we find both convergence and divergence, due to the historical patterns of partial isolation and partial contact between Blacks and Whites (Butters, 1989, p. 4). This view of language contact is similar to that of Wolfram and Johnson who speak of the adjustment and readjustment that occur in all phonological systems (Wolfram & Johnson, 1982, p. 163).

The convergence/divergence debate seems to have influenced language instruction in education, with the convergence side striving to maintain its ties to Black English, and the divergence side striving to maintain its ties to standard English. This is the focus of the next social issue.

Ebonics in education
Problems in education have been discussed from the "very beginning" (Schneider, 1989, p. 11) of the Black English debate. The controversial nature of this subject is discussed in many books, academic journals and public schools. Most notably, the school board in Oakland, California caused a great deal of debate by passing a resolution giving Ebonics the status of a second language (LSA, 1997, p. 1). A recent example focussed on Parker Elementary School in Oakland. Second graders were part of an experiment that used a storybook called Flossie and the Fox \ Flossie spoke Ebonics and the Fox spoke standard English (Leland & Joseph, 1997, p. 78). The result of using this pedagogical approach, called "contrastive analysis", was immediate improvements in oral language skills among students and outraged headlines in the media. Stanford linguist John Rickford approved of this type of teaching, and even went as far as calling it a shortcut process that works (Leland & Joseph, 1997, p. 79). These two positive views, however, are greatly outnumbered by the negative ones.
The contrastive analysis approach has been called a "complete failure" (Wardhaugh, 1994, p. 340) due to its unproven claims of effectiveness. In fact, numerous teachers of Black children are coping well without having to introduce "Niger-Congo idioms" (Cose, 1997, p. 80). These teachers are probably coping well because of the fact that even young children recognize that people speak differently, and these differences have "social consequences" (Wardhaugh, 1994, p. 342). Fasold sees that "precious little" (Fasold, 1981, p. 186) has been accomplished throughout the years of intensive attention paid to Ebonics for the purpose of educational advancement. In addition, Baugh points out that educational materials with spelling changes such as standard English "the", to the Black vernacular "duh" do Black students "very little good" (Baugh, 1983, p. 110). He sees that vernacular spellings in readers only serve to postpone exposure to standard English and therefore diminish the chances for success in the future, which is dominated by standard English. The professional working world demands competency in standard English, while street rappin` is seen as "foreign" (Baugh, 1983, p. 121) to most corporate boardrooms.


A great deal has been written about Ebonics, making it one of the most widely discussed varieties of English. Ebonics and Black English are just two of the many terms used to describe a variety of English spoken by many Blacks. This variety has neither been clearly defined, nor standardized. The coining of the term, a product of "ebony" and "phonics", is also unclear. Possibly, Williams or Smith coined "Ebonics" in the early to mid 1970s. Equally slippery is the question of how Black English originated. One side of the debate views BE as possessing unique features from West African languages, while the other side of the debate views Ebonics strictly as a product of British and American English. The "threefold distinction" analysis of BE phonological features showed a possibility for these opposing views to be reconciled. The evidence showed that Black English and White English share many phonological features, with only one of these features being unique to Black English.
This modest analysis reached the same conclusion that Fasold and others reached, that is, they saw no significant differences in the phonological features among standard and nonstandard English, and Black English.

Linguistic research into Black English is bound to its social contexts. For this reason, this discussion turned to two pertinent social issues surrounding Ebonics. The first issue discussed was the convergence/divergence of BEV and WEV. This issue was particularly significant because convergence results in the death of BEV, while divergence results in great social division. Bailey states that convergent features are all older than divergent features, which is probably the reason why the divergence hypothesis has been the more widely discussed of the two. Linguistic convergence and divergence were found to be very difficult to ascertain for reasons such as the ongoing adjustment and readjustment of phonological features. These two hypotheses seemed to be similar with the situation of Ebonics in education. It is possible that contrastive analysis was used to teach Ebonics in Oakland because of the convergence side trying to maintain its ties to Black English, and the divergence side trying to maintain its ties to standard English. This method of instruction was seen by many as ineffective, and even detrimental to future success in the working world.


  • Abati, R. (1997, January 26). To an African, 'Ebonics' or black English sounds foreign. The Detroit News, Outlook.
  • Avery, P. & Ehrlich, S. (1996). Teaching American English pronunciation. Oxford: Oxford University Press.
  • Bailey, G. (1993). A Perspective on African-American English. In D. R. Preston (Ed.), American dialect research (pp. 278-318). Amsterdam & Philadelphia: John Benjamins.
  • Baugh, J. (1983). Black street speech: Its History, structure, and survival. Austin: University of Texas Press.
  • Butters, R. R. (1989). The Death of Black English: Divergence and convergence in Black and White vernaculars. Frankfurt: Peter Lang.
  • Childs, T. (1996, March 4). Handout. JAL254 Sociolinguistics. Toronto: University of Toronto.
  • Childs, T. (1996, March 24). Lecture. JAL254 Sociolinguistics. Toronto: University of Toronto.
  • Cose, E. (1997, January 13). Why Ebonics is irrelevant. Newsweek, 80.
  • Dillard, J.L. (1972). Black English: Its History and usage in the United States. New York & Toronto: Random House.
  • Fasold, R. W. (1981). The Relationship between Black and White speech in the South. American Speech, 56 (3), 163-189.
  • Holm, J. (1991). The Atlantic creoles and the language of the ex-slave recordings. In G. Bailey, N. Maynor, & P. Cukor-Avila (Eds.). The Emergence of Black English: Text and commentary (pp. 231-248). Amsterdam & Philadelphia: John Benjamins.
  • Leland, J. & J. Nadine. (1997, January 13). Hooked on Ebonics. Newsweek, 78-79.
  • Linguistic Society of America. (1997, January). LSA resolution on the Oakland "Ebonics" issue. Chicago: LSA.
  • McMillen, L. (1997, January 17). Linguistics find the debate over 'Ebonics' uninformed. The Chronicle of Higher Education, A16-A17.
  • Montgomery, M. (1991). The Linguistic value of the ex-slave recordings. In G. Bailey, N. Maynor, & P. Cukor-Avila (Eds.). The Emergence of Black English: Text and commentary (pp. 173-189). Amsterdam & Philadelphia: John Benjamins.
  • Myhill, J. (1988). The Rise of be as an aspect marker in Black English Vernacular. American Speech, 63 (4), 304-326.
  • Schneider, E. W. (1989). American earlier Black English. Tuscaloosa: University of Alabama Press.
  • Smitherman, G. (1977). Talkin and testifyin: The Language of Black America. Detroit: Wayne State University Press.
  • Wardhaugh, R. (1994). An Introduction to sociolinguistics. (2nd ed.). Oxford: Blackwell Publishers.
  • Winford, D. (1992). Back to the past: The BEV / creole connection revisited. Language Variation and Change, 4 (3), 311-357.
  • Wolfram, W. (1991). Dialects and American English. New Jersey: Prentice-Hall and Center for Applied Linguistics.
  • Wolfram, W. (1990). Re-examining vernacular Black English. Language: Journal of The Linguistic Society of America, 66 (1), 121-133.
  • Wolfram, W. & Johnson, R. (1982). Phonological analysis: Focus on American English. Washington D.C.: Center for Applied Linguistics.