Geoffrey K. Pullum
It is unusual for a policy announcement at a city school-board meeting to trigger a worldwide media frenzy, but one California school-board meeting in December 1996 did exactly that. Within days of the announcement, school-board members could not leave their homes without being besieged by journalists. They were vilified, ridiculed, and attacked in newspapers and magazines around the entire world. What had happened?
The board had issued a statement to the effect that it was changing its educational policies with regard to one aspect of the local linguistic situation. They would pay more serious attention to the language spoken at home by most
of the district’s school students. Its status would be recognized, teachers would be trained to look at it objectively and appreciate its merits, and it would be used in the classroom as appropriate. This much was reported by the
New York Times quite accurately and fairly. Yet opinion writers proceed to fall upon the topic like starving dogs attacking a bone. They ridiculed, they sneered, they frothed, they flamed, they raged, they lived off the
story for weeks. The talk-radio switchboards lit up, and the intemperate opinions flared. What was going on?
The answer lies in the fact that the language being recognized by the school board was not Spanish or Polish or Russian or any such relatively uncontroversial language. The city was Oakland, a poor city on the east side of San Francisco Bay where half the population is African American, and the language was one that linguists usually call African American Vernacular English (AAVE).
What makes AAVE so dramatically different as a political issue from, say, Spanish (also spoken in Oakland, by up to a quarter of the population) is its close relation to another language of much higher prestige. Most speakers of Standard English think that AAVE is just a badly spoken version of their language, marred by a lot of ignorant mistakes in grammar and pronunciation, or worse than that, an unimportant and mostly abusive repertoire of street slang used by an ignorant urban underclass. An editorial in the New York Times a few days after the first news report said that the Oakland school board had “declared that black slang is a distinct language.”
Let me begin by getting that myth out of the way. The Times’s statement about slang was completely untrue, and the writers should be ashamed of themselves. So should all the newspapers and magazines that followed them. The governing board of the Oakland Unified School District never mentioned slang and never intended to imply anything approving about it.
We call an expression slang when it represents a vivid, colloquial word or phrase associated with some subculture not yet incorporated as part of the mainstream language. No subculture’s slang could constitute a separate language.
The mistake is like confusing a sprinkle of hot sauce with a dinner. Slang is by definition parasitic on some larger and more encompassing host language. It has no grammar of its own; it is a small array of words and phrases used
under the aegis of some ordinary language and in accordance with its grammar. The majority of slang words and phrases are in the language already and are merely assigned new slang meanings by some subpopulation.
The Oakland school board was not endorsing the nonsensical idea that black slang should be recognized as a new language. The statement it released may have been wordy, diffuse, and filled with bad bureaucratese and pompous-sounding
references to “African Language Systems,” but the intent was clear enough: the board wanted to acknowledge that AAVE was distinct in certain respects from Standard English, and it proposed to be responsive to the educational implications.
Buried among the jargon of the announcements was a mention of a name for AAVE, suggested by a Black scholar in 1975 but never adopted by linguists: Ebonics. That word, concocted from ebony (a color term for the name of a dark-colored
wood) and phonics (the name of a method for teaching reading), was destined to attach to the board as if chiseled into a block of granite and hung round their necks. They would never hear the end of it.
One problem with the name was that it lent itself irresistibly to stupid puns and jokes. The Economist picked it up and printed a brief story headed “The Ebonics virus,” a tasteless reference to the then-recent outbreak of
the horrible Ebola fever in Zaire (the subliminal link: nasty things out of Africa). People rapidly invented other -onics words to mock the idea of letting African Americans have their own claim to a language. Would Jewish
people propose that their way of speaking English should be designated Hebonics? Could stupid people complain that they were victims of their native language, Moronics? Cartoonists seemed to find such possibilities endlessly amusing,
and the jokes kept coming for more than a year.
But I will not be primarily concerned here with analysis of the politics and the rhetoric that the “Ebonics” story provoked. I will be concerned with AAVE itself, the everyday speech of millions of people in largely segregated African
American districts. The majority of English speakers think that AAVE is just English with two added factors: some special slang terms and a lot of grammatical mistakes. They are simply wrong about this.
Let me begin by pointing out that there is obviously a difference between being an incorrect utterance of one language and being a correct utterance in another (perhaps only slightly different). This is obvious when the two languages
are thoroughly different, like English and French. When a French speaker refers to the capital of the United Kingdom as Londres, it isn’t a mistake; that’s the correct French name for London. But the same is true when we are talking
about two very closely related languages. There is a strong temptation, especially when one of the two has higher prestige, to take one to be the correct way to speak and the other to be incorrect. But it is not necessarily so.
I will consider a fairly subtle example involving a grammatical difference between two varieties of Standard English. Some speakers, but not all, use the word whom instead of who in some contexts. Those who do use
whom always use it after a preposition (“a man in whom I have complete confidence”) and may use it after a verb (“And after that you visited whom?”), and they may use it at the beginning of a sentence
(though, “Whom did you visit?” sounds rather stiff, and many would avoid it). But most people, even expert writers of English, will confess to scratching their heads a little about the following two cases:
Which version is right? Dimly we may remember something from the grammar books about using who for subjects and whom for nonsubjects. Can’t we just apply that? No, we can’t. The rule “use who for subjects
and whom for nonsubjects” is insufficiently explicit. These examples involve a relative clause that begins after the word man. The next word (who or whom) introduces the relative clause (everyone seems to think __
will one day be king). There are two things that “subject” might mean here: “subject of its clause,” that is, subject of the clause that it logically belongs to, or “subject of the relative clause.” The word who is logically the subject in a clause that has will one day be king as its predicate; if that allows it to count as a subject, then version (1) is correct. But the subject of the whole relative clause is not who but rather everyone. The word who is not the logical subject of that, but just a piece of it. If that’s what we mean by being a subject, then we should pick sentence (2).
Where do we turn to decide this point? We look in good manuals of English usage. And we immediately find something very interesting: there are clear examples of both types in literary works by the best authors. The who group—those whose writing suggests that they would plump for (1)—includes Arnold Bennett, Charles Dickens, Henry Fielding, and William Safire (the New York Times’s language pundit). Good company. But the whom group, whose usage shows they would select (2), includes early writers like William Caxton and Izaak Walton, famous novelists like Charles Kingsley and Rudyard Kipling, romantic poets like John Keats and Percy Bysshe Shelley, and at least some New York Times and Publishers Weekly writers, together with Charles Darwin and William Shakespeare. That, too, is a dream team.
What is going on? The answer is that there are two different rules involved. Some writers follow one, while others follow the other. The members of the who group tacitly assume that who has to be the subject of its clause; the members of the whom group tacitly assume that who has to be the subject of the relative clause.
What we have discovered is a division between two syntactic dialects—two sectors of the speech community that have very slightly different grammars. By assuming that speakers fall into these two dialect groups, you can explain
both the consistencies of practice and the disagreements that occur. The disagreements do not become evident very often: only on the rare occasions when who occurs as the understood subject of a clause contained in a relative
clause without being the subject of the relative clause as a whole.
It is crucially important to notice that people do not just flounder, as if they did not know who from whom any more. To describe either the who group or the whom group as people who are ignorant
of the grammar of the language in which they are writing would be absurd. More specifically, it would predict quite falsely that you will find them blundering into sentences like the follow (the asterisk prefix means that the string
is not a sentence):
Nobody writes things like this. So it is not true that “anything goes” regarding the use of who and whom. There are rules. But it is also not true that there is a single correct rule governing everything about who and whom and some people have failed to learn it. That does not predict the facts correctly. There is a better theory, and it is set out informally in (5):
Rules (i) and (ii) both agree that you should use who in “anyone who wants to leave” because there it is both the subject of its own clause and the subject of the relative clause (which are one and the same). But on other
sentences the two rules decide things differently. The who group (Dickens, Safire, and company) assume (i) and thus regard example (1) as correct; the whom group (Shakespeare, Darwin and company) assume rule (ii)
and thus favor example (2). What this shows is that even within Standard English there is a difference between making mistakes and speaking a different variety correctly—a difference between marching out of time and marching to
the beat of a different drum.
With these preliminaries out of the way, I now turn to AAVE, which turns out to be the most interestingly divergent dialect of modern English.
The Oakland school board suggested that AAVE had its own rules and structure. Opinionmongers displayed anger and contempt at this. They mocked, lampooned, and attacked the very idea. But a large amount of the ink they splattered was
wasted on a pointless side issue: whether AAVE was really “a separate language” from English or whether it was “just a dialect.” The Oakland school board made the mistake of insisting explicitly that it was a separate language.
That is another red herring (like the slang issue).
Dialects and languages are in fact the same kinds of things. “Dialect” does not mean a marginal, archaic, rustic, or degraded mode of speech. Linguists never say things like “That is just a dialect, not a language.” Rather, they refer
to one language as a dialect of another. The claim that Tosk is a dialect of Albanian is a classificatory claim, like saying that the white-tailed deer is a kind of deer. It is not some kind of put-down of Tosk speakers.
Of course, in practice, political considerations do tend to enter decisions about when one language should be called a dialect of another. Albania is currently one country, and Tosk and Gheg are treated as dialects of one Albanian
language (even though it is hard to understand one if you only know the other); but in the wrecked ex-country of Yugoslavia, the language people used to call Serbo-Croat began during the 1990s to be deliberately split up into three
separate languages, Serbian, Croatian, and Bosnian, because the speakers had fallen out politically. No one would have said before that there were three languages here, even though the Croatian Yugoslavs talked a bit differently
from the Serbian and Bosnian Yugoslavs. There was assumed to be one Serbo-Croat language (there is a book called Teach Yourself Serbo-Croat). This single language would have been described as having several different dialects—regional
or ethnic variants of the same basic linguistic system. But once Serbia, Croatia, and Bosnia separated, the differences among the three trivially differing languages had to be emphasized and accentuated. The Serbs, stressing their
Eastern Orthodox religious roots, revived Old Church Slavonic words and mandated use of the Cyrillic (Russian) alphabet; the Croats stressed their Catholic roots, borrowed more words from Latin, and stuck with the Roman alphabet;
and the Bosnian Muslims borrowed words from Turkish to emphasize their connections with the Islamic world. Linguistic boundary lines began to emerge to reinforce the ethnic boundary lines.
Reference works on the languages of the world take different views of how many linguistics boundaries to recognize. Reasonable scholars can differ on how coarse grained or fine grained a classification is appropriate (and on how much
politics should be taken into account). So it is important that in classifying AAVE there is no dispute. For example, I keep near my desk two reference books on the languages of the world; Grimes’s Ethnologue and the Voegelins’
Classification and Index of the World’s Languages. The Ethnologue recognizes more than fifteen separate romance languages in Italy, while Classification and Index recognizes just three (Italian, Friulian,
and Ladin). But both list AAVE as a dialect of English.
This is undoubtedly the right classification. Virtually all the words used in AAVE can be clearly identified in Standard English too, and most of AAVE grammar is the same as that of Standard English. The bits that are not are mostly
paralleled in certain other dialects of English that are never mistaken for other languages. As linguist John McWhorter has pointed out, there certainly are examples, in some parts of the world, of African languages that have kept
their African syntax and simply replaced the words of the dictionary by English words (they are known as creole languages), and AAVE just is not like that.
This is not an insulting or demeaning thing to say about AAVE. It merely places it in a linguistic classification that unites the language of Shakespeare and the language of the Oakland ghetto very closely compared to languages in
other families or in other parts of the Germanic family. It is no more insulting to call AAVE a variety (or dialect) of English than it is to say that African Americans are U.S. citizens.
It was not merely a matter of an incorrect classificatory claim when the Oakland school board made its wordy statements about how the district’s black students used “African Language Systems” that did not constitute a dialect of English.
It was also a major tactical error. It deflected virtually all the discussion into negative channels; journalists vied with each other to insist that AAVE should not be recognized as a language. What got lost was a much more sensible
and reasonable point: AAVE as a dialect of English still deserves respect and acceptance. It has a degree of regularity and stability attributable to a set of rules of grammar and pronunciation, as with any language. It differs
strikingly from Standard English, but there is no more reason for calling it bad Standard English than there is for dismissing Minnesota English as bad Virginia speech, or the reverse. Journalists did not get this; nearly everything
they wrote about the purported errors that characterize AAVE was factually incorrect. It is worth devoting a little time to close analysis of some examples that demonstrate this.
There is a technical term for the auxiliary verb that takes the forms be, been, being, am, are, is, was, and were. It is called the copula. The most popular
myth about AAVE is that it involves misuse of the copula: that is carelessly omitted or is used in incorrect forms like be out of ignorance. Let me try to untangle the jumble of falsehoods that are commonly bruited about.
The AAVE copula can be omitted, but there are strict rules—surprisingly detailed and specific ones—about how and where. In the following summary, AAVE examples are italicized and translations into Standard English are in double quotation
marks.
Only when none of these conditions obtains—when the copula is present tense, not first person, not accented, not negative, and not expressing the habitual or the remote present perfect—can it be omitted in AAVE speech. Most of this
can be easily confirmed from the smattering of AAVE one can pick up from popular culture—snatches of AAVE dialogue in movies, African American popular songs, and so on. Almost every American knows that the Standard English greeting
How are you doing? can be reducing to How you doin’? in AAVE, but the AAVE greeting What is it? is never given in the form *What it?; they know that tin the line I ain’t lying (in “I
Put a Spell on You”) you cannot leave out the ain’t; that Otis Redding sings I was born by a river (in “It’s Been a Long Time Coming”), and it could not be reduced to *I born by a river; that song lines
like I’m your puppet and I’m a hog for you, baby couldn’t be reduced to *I your puppet or *I a hog for you, baby; and so on. Any general claim that AAVE speakers leave out the copula is clearly
false.
Few English speakers are aware that Russian, Hungarian, Arabic, Swahili, and many other languages have rules for omitting the copula, under conditions that are of a quite similar sort though differing in details. In Arabic, for example,
the copula may be omitted in the nonemphatic affirmative present tense. In Hungarian, the copula must be omitted in the affirmative third-person present-tense (i.e., there is no such thing as an affirmative third-person
present-tense copula form.) In Russian the copula is optionally omitted in the nonemphatic affirmative present indicative when it is not expressing existence (as in There is a god).
Some languages, for example, Standard French, differ very clearly in that they always have a pronounced form of the copula in relevant sentence types and show no superficially verbless sentences. Standard English is a little closer
to AAVE than to Standard French: it does not entirely omit the copula, but almost omits it, reducing it to a single consonant when unstressed and in the affirmative present tense (I’m certainly interested, You’re so kind, He’s my brother,
and so on, but not *I certainly’m, where am needs to be accented, or *He’sn’t my brother, where is has been negated). This is a pattern also seen in, for example, Turkish (an entirely unrelated
language). These reductions are not random carelessness; they are the results of rules, and so is the omission of the copula in AAVE.
Notice that in many languages another part of the sentence is regularly omitted: languages like Russian, Swahili, Spanish, Italian, Japanese, and Chinese omit nonemphatic subject pronouns nearly all the time when the context permits
the meaning to be identified without difficulty. Te amo in Spanish means “I love you” despite the nonoccurrence of the word for I. (There is such a word—it is pronounced yo—but it is only used emphatically.)
But AAVE does not permit omission of the subject pronoun. Even in the most careful Spanish you can say ¿Dónde estás? for “Where are you?” (omitting the word for “you”) or Llegará esta noche for “She will arrive
tonight” (omitting the word for “she”), but you cannot say *Where are? or *Will arrive tonight in AAVE. Subject pronouns are never omitted except in imperatives and in very casual telegraphic usages that other
dialects share (we can say ‘S a mystery to me instead of It’s a mystery to me in very informal speech, or write Will write more soon to save words in a telegram or on a postcard). AAVE has a grammar that
determines such things, just as Standard English does, and the grammar of the two dialects agree on this point: subjects are obligatory in nearly all sentence types. AAVE speakers do not carelessly leave out subjects any more than
Standard English speakers do.
AAVE critics talk about something called “double negation” and treat it as an illogicality. The critics generally do not know what they are talking about. AAVE negates the same way other languages do: You ain’t ugly is true
if and only if You ugly is false, and conversely. What does differentiate AAVE from Standard English is that negation can be multiply marked: Standard English I am not an ugly fellow translates into AAVE as I ain’t no ugly dude;
Standard English I haven’t ever seen anything like it corresponds to AAVE I ain’t never seen nothin’ like it; standard English He did not see anything is AAVE He di’n’t see nothin’. The critics’
claim about this is that logic tells us that two negatives make a positive: if he did not see nothing, that means he did see something, and it is illogical to use that form of words to mean the
opposite.
I will only make a brief detour into logic (it is a technical business, and there is much more that could be said). It is certainly true that in formal logical languages such as the propositional calculus, under classical interpretations
of negation, two negatives are equivalent to an affirmative (what a logician writes as ~~P means “it is not so that it is not so that P,” where P is any statement you like; and it is
equivalent to saying P). But there is little similarity between this mechanical logical system and the grammar of any natural language. Under the classical interpretation, any even number of consecutive negations is equivalent
to none, and any odd number is equivalent to one. So three negatives are equivalent to a single negative in logic, which means that the critics have some explaining to do: what is supposed to be objectionable about I ain’t never seen nothin’ like it?
It has three negative words (ain’t, never, nothin’) and a negative meaning, and that ought to be fine if all that is important is for negative sentences to have an odd number of negative words.
The truth is that many of the people who grumble about AAVE being illogical do not know enough logic to even explain what they mean by their critique. They stress something completely irrelevant (the way interference is formalized
in the invented languages of logic) while missing major complications concerning the interaction of negation with quantifiers (a topic that is beyond the scope of this chapter) and failing to see the crucially relevant fact about
natural languages: that the grammar of negation is not the same in all languages.
Negation in Standard English, German, or Arabic is different from negation in, for example, Spanish or Russian or Italian. In Standard Italian, the way to say “nobody telephoned” is Non ha telefonato nessuno, literally “not
has telephoned no one.” The non at the beginning and the additional negativity of nessuno ‘no one’ are both required. Italian demands that a sentence like this be negated in a particular way that demands both
a non and a nessuno.
To be slightly more precise about this, in those positions where indefinite words (words like anybody) would appear in a standard English negative clause, Italian requires their negative counterparts (words with meanings like
nobody). This a rule of the grammar, usually known to linguists as negative concord. Negative-concord languages require use of negative words instead of indefinite words. This is not an error; it is a
demand of the grammar, rather like an agreement rule.
AAVE turns out to be like Italian with regard to negative conrod, not like Standard English. The AAVE sentence Ain’t nobody called shows exactly the same negative concord as the Italian Non ha telefonato nessuno:
the negative element ain’t requires that nobody be chosen just as the negative element non requires that nessuno be chosen. Neither AAVE nor Italian is illogical; it is just that their grammatical
rules for expression of indefinites in neglected clauses differ from the rules for Standard English.
AAVE is not alone as a dialect of English with negative concord. Cockney (a working-class dialect spoken in the East End of London, England) and numerous other working-class dialects in England (and America) also have it. For example,
you can hear Pink Floyd sing “We don’t need no education; we don’t need no thought control” on their album The Wall; they are a white British band, and they are singing in working-class British English, not AAVE.
It is merely an accident that the negative-concord dialects of English today have a low-prestige class background. In Italy nothing of the sort is the case: people of noble birth and newspapers of the highest quality use negative concord.
Pope John Paul II uses negative concord when he speaks in Italian. (Actually, he uses it when he speaks Polish, too, because Polish is like AAVE and Spanish and Italian, not like English and German and Arabic.)
Another key feature of AAVE (not shared by Cockney and other dialects) involves repositioning a negative auxiliary verb at the beginning of the sentence when the subject is indefinite. Thus we find Ain’t nobody gonna find out,
meaning “Nobody is going to find out.”
In the Standard English version, negation is expressed purely on the subject. If you want to make a sentence that is true just in those circumstances where “Somebody is going to find out” is false, you simply replace somebody by nobody, and you get “Nobody is going to find out.” In AAVE the clause has to be marked as negative by its auxiliary verb, so you use ain’t, and in addition (remember, AAVE is a negative-concord language), all
words like somebody must be replaced by negative forms like nobody throughout the clause. Complying with these demands would yield Nobody ain’t gonna find out. But that is not necessarily the most acceptable
way of saying things; it is quite common for AAVE speakers to switch the order of the subject (nobody) and the auxiliary verb (ain’t), yielding Ain’t nobody gonna find out. In Standard English the auxiliary
occurs before the subject in interrogatives (Isn’t anyone going to find out?), but not in declaratives (except where, in rather formal style, a negative adverb occurs at the beginning of the sentence: Never have I seen such a thing).
The AAVE Ain’t nobody gonna find out is a declarative sentence (used for making a statement), not an interrogative sentence (used for asking a question), yet it has the auxiliary as the first word. This is another example
of a syntactic difference between AAVE and Standard English
There is a complication to this. Negative inversion in AAVE is not found with every type of subject noun phrase. For example, when the subject is a simple name like Mary, it is impossible, or at least extremely unlikely. We
do not hear *Ain’t Mary gonna find out. Similarly, when the subject has the definite article the or a possessive article like your, we do not get negative inversion: *Ain’t the teacher gonna find out;
*Ain’t your mother gonna find out.
So there is more to be said about the various conditions that encourage or discourage the use of negative inversion in particular sentences, but my point has been illustrated enough: negative inversion is a construction type that the
standard dialect does not have. Again, AAVE shows that it has certain regular syntactic principles of its own.
AAVE also has special principles of pronunciation. One fairly clear one concerns consonant clusters at the end of words. AAVE speakrs say res’ for ‘rest’, lef’ for ‘left’, respec' for ‘respect’, han’ for ‘hand’, and so on. A superficial glance might suggest that they leave consonants off the ends of words. But it is not that simple.
We will need a small amount of phonetic terminology at this point. The stop consonants of English are the ones normally represented by the letters p, t, k, b, d, and g, as found on the ends of the words up, out, oak, rob, cod, and
log. The voiceless consonants, quieter because they are pronounced with no vibration of the vocal cords, are the sounds for which p, t, and k are normally used. The other voiceless sounds
in English include the ones heard at the ends of the words off, miss, and fish, which are called fricatives. The voiced stops, which are produced with vocal-cord vibration are b, d, and g. (Some
consonant letters get doubled in English spellings, but this is not relevant to the pronunciation.) Other voiced sounds in English include the consonants heard at the end of words of, Oz, ridge, and
fill. (Standard English also has interdental fricatives like those in the words breath and breathe, but these are replaced by stops in several dialects, of which AAVE is one: they is
pronounced dey.)
With this terminology we can describe some rather intricate restrictions concerning which consonants can be left off and which have to be retained in AAVE. The basic principle involved in the one shown in (6).
Note first that consonants are always pronounced when they follow a vowel. In words like up, out, oak, rob, rod, and log, no AAVE speaker leaves off the final consnant. In words
like dump, sink, and belt, again all the consonants are pronounced. The consonants p, t, and k are voiceless stops, but the sounds represented by m, n and l are voiced, so nothing gets dropped. Similarly,
in words like raps, rats, or racks, the s is always pronounced, because although it is a consonant of the same voicing as t, it is not a stop, so it is retained. In words like Dobbs, Dodds, or
dogs, we hear a voiced fricative z sound at the end, which is not dropped, because although it has a consonant of the same voicing before it, that consonant is not a stop. We therefore get consonants dropped only
in cases like these.
| test: | voiceless stop t dropped after voiceless s | tes’ |
| desk: | voiceless stop k dropped after voiceless s | des’ |
| left: | voiceless stop t dropped after voiceless f | lef’ |
| respect: | voiceless stop t dropped after voiceless k (spelled c) | respec’ |
| stopped: |
voiceless stop t (spelled -ed) dropped after voiceless p | stop’ |
| hand: |
voiced stop d dropped after voiceless n |
han’ |
| old: |
voiced stop d dropped after voiced l |
ol’ |
One might be tempted by the spelling to think that AAVE also leaves g off the end of words like nothin’, somethin’, and singin’, but this is not true either. In Standard English these words end in
the ng sound of sing (phoneticians call this sound the velar nasal), but today in many dialects the -ing ending is frequently replaced by -in, often represented in writing as -in’.
Nothing has been dropped, despite that apostrophe; rather, one sound has been replaced by another.
Or, at least, nothing has been dropped in recent times. But there is an interesting little historical wrinkle to this. Further back in the history of English a consonant was in fact dropped from words like sing, but the careless
culprits were the speakers of what we now flatter with the term “Standard” English. Dialects in the north of England have undergone fewer changes than the standard dialect, and those dialects still have the sound g (as
in egg) on the other end of word roots like sing and in derivatives of it like singer. In other words, singer rhymes with finger and linger in Northern England. Some centuries
ago the southern England dialects began to lose the g sound after a velar nasal at the end of a word root, so now finger and singer no longer rhyme in southern England. The g sound is heard in
finger, a simple word containing a g following a velar nasal, but not in singer, a complex word in which the root sing ends in a velar nasal and the suffix -er follows.
No dialects in southern Britain (or America, which was settled from there) have a g after a velar nasal at the end of any word anymore. So it is true that in the speech of Queen Elizabeth II, the g sound that used
to end sing has been lost. But no one calls the queen sloppy or mistaken in her speech. Why? Because there is a double standard here. When Standard Southern British English introduces a simplifying change in the rules
of punctuation (like “do not pronounce the g sound after a velar nasal except in the middle of a word”), it is respected as the standard way to speak, but when AAVE introduces such a change (like “do not pronounce a stop
at the end of a word after another consonant with the same voicing”), it is unfairly regarded as sloppiness.
At this point I should make it clear that the features of AAVE I have discussed are characteristic of a rather “pure” AAVE, a version only minimally influenced by Standard English. But people are flexible, and even African Americans
who have hardly any social contact with whites know an enormous amount of Standard English by the time they are adults. (It is 5-year-olds who have trouble on first confronting Standard English, perhaps in school, where they may
take a while to latch onto the new way of speaking.) It is quite typical for speakers of AAVE to be able to switch back and forth between their dialect and one much closer to Standard English. The different features mentioned earlier—copula
omission, negative concord, negative inversion, and others that could be cited—are options that can be called upon in one utterance and then not used in the next; a speaker can in effect switch between dialects at high speed.
This phenomenon is well known, of course. If two African Americans are chatting together privately in AAVE behind the counter in a store when a white customer enters and asks a question, the next utterance heard may have a dramatically
different grammar from the last as the person who responds switches dialects.
Sometimes speakers may use both more AAVE-like and less AAVE-like ways of phrasing a sentence, for example, using Nobody likes him and Don’t nobody like him interchangeably. But this happens in Standard English too.
For example, although it was not mentioned in the discussion of who and whom earlier, there are writers who use either form of the pronoun in the contexts we considered (examples like We are talking about a man who[m] everyone seems to think will one day be king).
Fine writers of English like James Boswell (the biographer of dictionary maker Samuel Johnson), Benjamin Franklin, John Galsworthy, and Oliver Goldsmith have all been found to use the when-it’s-the-subject-of-its-clause rule or
the when-it’s-the-subject-of-the-relative-clause rule interchangeably. But they are not making mistakes; they are not blundering about confusing their who with their whom. We know that, because we know they never
write *Whom are you? or *To who was it addressed? Languages have many rules and regularities of sentence structure, and speakers select from among the possibilities in ways that are highly complex. But that is
a sign of having a sophisticated and flexible grasp of the possibilities in a rule system; it is not a sign of ignorance
The grammar of AAVE—negative concord, copula omission, dropping of final consonants, and the rest of it—has to be learned by anyone who wants to speak AAVE and be accepted linguistically in the AAVE-speaking community. Knowing AAVE
does not come free with either knowing American English or having African American ethnicity. This point is beautifully (though unintentionally) illustrated by the mocking column on Ebonics that was published in the Washington Post on December 26, 1996, by the distinguished African American columnist William Raspberry.
Raspberry’s column reported a fictional conversation between an imaginary Washington, D.C., cab driver and an imaginary alter ego of Raspberry himself. The cab driver speaks in AAVE: We get a total of just thirty-two words of Raspberry’s
made-up AAVE dialog in the piece. But, Raspberry has a problem. He just does not know AAVE. The dialog he invents has grammatical errors.
First, he has the imaginary cabbie saying, “What you be talkin’ ‘bout, my man?” But as we saw earlier, the uninflected be of AAVE marks habitual aspect. The context makes clear that the cabbie means “What are you talking about
right now?” not “What do you habitually talk about?” So we can be quite sure that the normal AAEV for that would be “What you talkin’ ‘bout” with the zero copula (it is present tense, unemphasized, affirmative, and not first-person
singular); it would not contain the uninflected be. (Strike one!)
The second error is in a highly contrived interchange where the fictional cabbie, who is eating fish and chips when we come upon him, says “Sup?” meaning “What’s up?” and the fictional Raspberry misunderstands this as short for “Do
you wish to sup?” and replies that he has already dined (I did say it was contrived). The cabbie protests the misunderstanding and says, “I don’t be offerin’ you my grub.” He clearly means “I am not offering you my food,” so this
should be I ain’t offerin’ you my grub. Again we need the present progressive, not the habitual. (Note the line from “I Put a Spell on You” mentioned earlier: we get I ain’t lyin’, not *I don’t be lyin’, because the meaning is “I am not lying right now” rather than “I do not habitually lie.”) For a more distinctively AAVE utterance Raspberry could have had the cabbie say “I ain’t offerin’ you no grub,” with the multiple negation
marking that is such a distinctive feature of AAVE, but does not occur at all in Raspberry’s AAVE dialog. (Strike two!)
The third error Raspberry makes is also in the cabbie’s response to the misunderstanding about this monosyllabic greeting. He explains what he meant by saying “Sup?” thus: “I be sayin’ hello.” He means “I am saying hello to you (right
now),” not “I habitually say hello.” I be sayin’ is not the way to express this. (There is an utterance containing they be sayin’ quoted from the speaker called Larry in William Labov’s article “The Logic of Nonstandard
English,” and it is quite clearly habitual in meaning. But I have not encountered one in which the meaning is progressive.) Here, notice, the copula cannot be omitted: *I be sayin’ hello would be ungrammatical, as we noted
earlier. So the most likely form we would get for this meaning would be I’m sayin’ hello. (Strike three!)
With three strikes against him, Raspberry is already out. But there are more errors to report. The fourth comes when the Raspberry alter ego switches into AAVE for the punchline, as he realizes that he could augment his columnist’s
salary by giving AAVE language lessons if the Ebonics catches on. Raspberry should not give up his job at the Post, because he does not know AAVE well enough to teach it. He says, “Maybe you be onto somethin’ dere, my
bruvah.” But once more it is the immediate present he is referring to: he does not mean “Maybe you are habitually onto something there,” but rather “Maybe you are onto something there (right now).” The usual way of saying this
in AAVE would be Maybe you onto somethin’ dere (second person, non emphatic, not negated, present tense, so this is another of the situations where we can get the zero copula).
There are further errors in the things Raspberry has his characters say about AAVE rather than in it. The cabbie cites his brother-in-law as claiming that in order to speak AAVE you have to “leave off final consonants.” But from the
cabbie’s first word, ‘Sup, there isn’t a single final consonant missing in any of Raspberry’s AAVE dialog. (As noted earlier, words like somethin’ are not missing a final consonant.) The dropping of consonants
described earlier is not illustrated in Raspberry’s dialogue at all. His imaginary cabbie’s brother-in-law would have us believe that in general, or at random, the last consonant of an AAVE word is (or may be) dropped. That simply
is not true, as I explained earlier.
What does Raspberry’s effort at humor teach us? That a Washington columnist can sometimes come up with rather lame stuff around Christmas time. But also this: AAVE should not be thought of as the language of Black people in America.
Many African Americans neither speak it nor know much about it, as Raspberry demonstrates. But those who do speak AAVE are not just blundering; they have learned a complex set of rules that happens not to be the same as the complex
set of rules that defines Standard English. They know a language that the highly educated Mr. Raspberry has not learned.
Facts about the grammar of AAVE like the ones I have reviewed in this chapter have been known to American linguists for decades largely because of the pioneering work of people like William Labov at the University of Pennsylvania and
a significant number of AAVE specialists who originally trained with him like John Rickford of Stanford University. Knowing what they knew, the members of the American linguistics profession were dismayed at the ignorance betrayed
by the media commentators’ angry and offensive attacks on AAVE and the Oakland school board.
Confusing lexicon with syntax, accent with dialect, difference with deficiency, and grammar with morality, the commentators clarified little except the deep hostility and contempt whites feel for the way Blacks speak (the patois of
America’s meanest streets, columnist George Will called it, as if AAVE could only be spoken in slums) and the deep shame felt by Americans of African descent for speaking that way (former Black Panther party official Eldridge Cleaver
published an article in the Los Angeles Times in which he compared acknowledging AAVE with condoning cannibalism). The saddest thing is that in their scramble to find words to evince their fury and contempt at AAVE, columnists
both black and white ignored the genuine issues of educational policy that had motivated the Oakland school board.
One persistent confusion was perhaps stimulated by an ambiguity in English. The phrase “instruction in French” can be understood in two ways: it could mean instruction on how to speak French, or it could mean instructions given via
the medium of French. The press never figured out the difference between these two. The Oakland board members talked about “imparting instruction to African-American students in their primary language”; but this was discussed in
editorials as if the proposal had been to make AAVE a school subject. The school board’s statement did not suggest to me that they even considered adding AAVE to the curriculum. The plan was to use AAVE as a medium of instruction,
but not to hold classes on how to speak AAVE. (The whole point, after all, is that no such classes are necessary since the children arrive speaking it.) In addition, it was explicitly stated that in part this was “to facilitate…
acquisition and mastery of English Language skills.”
Lost from any of the press coverage that I saw was the fact that Oakland faced an issue very similar to the dilemma of Norwegian schools in which children from rural areas arrive speaking rural dialects of Norwegian very different
from the standard language of Oslo. The difference is that the Norwegians love their nonstandard dialects and treat them as valued symbols of a traditional Norwegian identity. They have experimented with both total immersion, where
the child is plunged into a Standard Norwegian-only environment from day one, and techniques more reminiscent of bilingual education, where at first the rural dialect is used in the classroom and gradually the children are taught
to move toward the standard language for public and official interaction and formal writing. There is some evidence that the latter works better. Teaching children to read first in their own dialects and then gradually introducing
the standard language can speed and improve the acquisition of reading skills.
There is good evidence of a similar sort pertaining to African American students in the United States. College students in chicago who received instruction concerning the contrasts between AAVE and Standard English grammar showed improved
Standard English writing skills as compared to a control group. In Oakland itself, nearly twenty-five years before the great furor, it was found that teachers who condemned AAVE pronunciations and interpreted them as reading errors
got the worst results in teaching AAVE-speaking children to read, while teachers who used AAVE creatively in the classroom got the best results. These and other results in the education literature suggest that the Oakland school
board’s policy decision had some clear motivation and scientific support.
But instead of a sympathetic consideration of the strategy they were suggesting and the evidence that supported it, the members of the Oakland school board suffered months of unrelenting ridicule, neelding, and abuse from politicians,
poets, and pundits in editorials, articles, talk shows, news programs, speeches, and government statements (the U.S. Department of Education pointedly announced that no federal bilingual education funds would be spent on AAVE).
It was a sad spectacle.
The horror with which Americans react to the idea of using AAVE in the classroom has something to teach us about the prejudice still targeted on America’s Black citizens, whose variety of English is decried as if it were some repellent
disease (recall the Economist with its jokey headline about the Ebonics virus). Educational conservatives often deny that prejudice is involved, dismissing linguists’ objective attitude toward nonstandard dialects as if
it were just left-wing propaganda. But it is not. Even conservative linguists acknowledge the facts mentioned earlier. The Linguistic Society of America’s vote on a January 1997 resolution in support of the Oakland school board
was actually unanimous.
Linguists will agree, of course, that Standard English has prestige and AAVE does not, and they are not trying to suggest that Time and Newsweek should start publishing articles in AAVE. They merely note that grammar
in and of itself does not establish social distinctions or justify morally tinged condemnation of nonstandard dialects. AAVE has nothing inherently wrong with it as a language; it is only an accident of history that it is not the
standard language of the United States. In the United States, the standard language lacks multiple negation marking, but has syllable-final consonant clusters and interdental fricative consonants. In Italy the facts are the reverse:
the standard language of that country lacks syllable-final consonant clusters and interdental fricative consonants and does exhibit multiple negation marking, exactly the same combination (coincidentally) as AAVE happens to have.
The horror among the readers who witnessed the feeding frenzy in the press over the Oakland resolution got through to the parents in the city. They were alarmed enough that, fearing linguistic ghettoization of their children, they
pressed the school board to reconsider. On January 15, 1997, the board did revise its statement, dropping the reference to imparting instruction in the primary language of many of its schoolchildren (while retaining, sadly, some
of its misguided references to alleged African origins of AAVE). This seemed to me to be throwing out the baby of a rather sensible change in language policy while carefully saving the bathwater of Afrocentric nonsense about AAVE
going back to West Africa.
Behind the statements, however, lay more than just rhetoric for public consumption. Behind the scenes, the Oakland school system was steadfast in its intent to use AAVE wherever it could assist the learning of Standard English or other
subjects by its students. The board’s rhetorical backdown did not mean that the policy had changed. This should not be surprising. Several other school districts in California and elsewhere have long operated programs involving
AAVE in the classroom. Of course, teachers have some discretion, and many Oakland teachers speak AAVE natively; it would be impossible in practice to police classrooms tightly enough to prevent them from using their native tongue
when speaking all day to children who share it.
There is thus little doubt that African American English will continue to be heard in Oakland’s classrooms. It may improve the rapport in the classroom in ways that will be beneficial to learning. It is worth a try. But whatever the
success of the new policy, I hope that one thing is clear: using AAVE in the classroom does not necessarily mean lowering standards of teaching things that are wrong. AAVE speakers use a different grammar, clearly and sharply distinguished
from Standard English at a number of points (though massively similar to that of Standard English overall). AAVE speakers need to learn Standard English. But anyone who thinks that AAVE users are merely speaking Standard English
but making mistakes is wrong. They can try to make the case that speakers of AAVE are bad or stupid or nasty or racially inferior if they want to, but they will need arguments that do not depend on language, because linguistic
study of AAVE makes one thing quite clear: AAVE is not Standard English with mistakes.