| Table (1) | ||
| AAVE | Standard English | gloss |
 |
 |
'watching' |
 |
 |
'doing' |
 |
 |
'riding' |
 |
 |
'running' |
 |
 |
'nothing' |
 |
 |
'something' |
If you considered only the data in Table (1), a generalization you might make is that AAVE has no velar nasal consonant, and that it always uses an alveolar nasal instead. However, this generalization is inaccurate. Consider the next set of words in Table (2).
| Table (2) | ||
| AAVE | Standard English | gloss |
 |
|
'think' |
 |
|
'pink' |
 |
|
'punk' |
 |
|
'thank' |
 |
|
'sank' |
 |
 |
'singer' |
 |
 |
'finger' |
Question 1. Is the distribution of the velar nasal (where it can occur) phonologically conditioned in AAVE? What is the evidence for your answer to this question?
A generalization you could make so far is that AAVE does have a velar nasal, which can appear before other velar stops, as in [] and [], or between vowels, as in []. Hence, its distribution is phonologically conditioned: it apparently cannot appear at the end of a word, but can appear in the middles of words.
| Table (3) | ||
| AAVE | Standard English | gloss |
 |
|
'thing' |
 |
|
'sing' |
 |
|
'song' |
 |
|
'long' |
Question 2. In your answer for Question 1, you may have decided that AAVE allows velar nasals in a restricted phonological context. Do the forms in Table (3) respect these same restrictions?
Briefly, no. The AAVE words in Table (3), in fact, do not seem to be subject to the same restriction as those in Table (1). In this case, velar nasals can occur at the ends of words. At first glance, there is a contradiction: [] is apparently not possible in AAVE, but [] is. Moreover, [sIn] is not possible for sing. We will see below, however, that this is not a contradiction, and that a generalization can still be made about the distribution of the velar nasal in AAVE.
If your answer to Question 2 was 'no', then there must be some other reason why velar nasals are not found in the words in Table (1), but are in the words in Tables (2) and (3). Do you think it's a random difference? Or is there something else about the words in (1) that distinguishes them from the others?
The words in Table (1) share other properties besides having velar nasals in Standard English and alveolar nasals in AAVE. Most of them are verbal participles; they end with the suffix we spell as -ing. Thus, in these words, the velar nasal occurs within a grammatical morpheme. The remaining words in Table (1), something and nothing, are quantifiers, which means that they too have a kind of grammatical function.
In contrast, the words in Table (3) are nouns, verbs, and adjectives. The velar nasal here does not occur in a grammatical morpheme. Therefore, we can say that the velar nasal in AAVE is also morphologically conditioned: it can only occur in certain grammatical categories (i.e., parts of speech) of words.
The basic generalization is thus as follows: AAVE does not allow velar nasals word finally in function morphemes. Thus, it is not the case that AAVE speakers cannot produce velar nasals; instead, the velar nasal does not always occur in the same contexts as it can in Standard English.
An additional point to be made is that this particular AAVE feature is actually found in many other vernacular or informal varieties of English. Furthermore, in each variety, including AAVE, the pattern of turning a velar nasal into an alveolar one is variable. That is, a speaker may produce either [] or [] for doing, depending on other circumstances like formality of the speech setting. No vernacular variety, however, would ever have forms like [sIn] for sing, even variably. We associate the pattern with AAVE most of all, since the tendency to use the alveolar form is higher in AAVE than it is in other dialects.
The idea that other dialects have a similar pattern brings us to a final implication to be drawn from this exercise. The phonological differences between dialects are always a function of the language, and not of its speakers. That is, we can only attribute the difference between two dialects to the fact that, when isolated geographically or socially from each other, they may evolve in different directions.
This is an important point, because it is often too easy (and ultimately incorrrect) to find some quick explanation for certain features of dialects. Since AAVE is associated with the African-American population, it may tempting for the novice linguist to claim that differences like the one illustrated in this exercise are a function of ethnicity. Another way of expressing this is to claim that AAVE speakers cannot produce velar nasals because their ethnicity or physiology makes it impossible for them to do so. Clearly, this claim is false, as the tables of data above show; as does the variable use of standard forms by AAVE speakers. Further, not all African Americans speak AAVE, nor are all AAVE speakers African American.
The reason for the emergence of a particular feature is often a matter of speculation. Scholars like Smitherman (1977) attribute the source of some grammatical features of AAVE to the grammatical features of some West African languages. This is not to say that AAVE speakers themselves have any knowledge of the grammar of any African language. Rather, the first African Americans learned English as a second language, and the English they acquired would surely have been influenced by their various first languages. This is not the only possible reason for the source of differences between AAVE and Standard English, but it remains a leading theory.
What we can say for sure is that the persistence of differences between AAVE and Standard English is a result of the fact that AAVE has been used by a large and coherent social group that historically has been somewhat isolated from other components of American society. Given that isolation, AAVE has been able to diverge without any strong restraint from the influence of Standard English, in the same manner that geographically distant dialects diverge.
Back to AAVE phonology exercise
© 2001 The Language Samples Project