African American Vernacular English in Advertising: a Sociolinguistic Study

ABSTRACT - In our diverse society, language variation is an important cultural identifier. To understand and target important consumer segments, one must be aware of the dialect they speak. In examining the language used by African Americans in advertising, this sociolinguistic study finds that only 14% of a sample of current television ads with black actors use grammatical features from African American Vernacular English (AAVE), while only 34% used AAVE phonological features. In order to increase our understanding of language as it relates to cultural diversity, a framework based on perceived fit is developed.

Citation:

Jennifer Edson Escalas (1994) ,"African American Vernacular English in Advertising: a Sociolinguistic Study", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 304-309.

Advances in Consumer Research Volume 21, 1994      Pages 304-309

AFRICAN AMERICAN VERNACULAR ENGLISH IN ADVERTISING: A SOCIOLINGUISTIC STUDY

Jennifer Edson Escalas, Duke University

ABSTRACT -

In our diverse society, language variation is an important cultural identifier. To understand and target important consumer segments, one must be aware of the dialect they speak. In examining the language used by African Americans in advertising, this sociolinguistic study finds that only 14% of a sample of current television ads with black actors use grammatical features from African American Vernacular English (AAVE), while only 34% used AAVE phonological features. In order to increase our understanding of language as it relates to cultural diversity, a framework based on perceived fit is developed.

CULTURAL DIVERSITY AND LANGUAGE

Cultural diversity is on the rise in America. The most recent census projects minority groups to continue growing, given the higher fertility rates of Asians, Hispanics, and African Americans compared to the rate of whites (Harper 1992). Past consumer research has pointed out that these growth patterns in ethnic subcultures have significant impact for the consumption aspects of American life (Deshpande, Hoyer & Donthu 1986).

One important aspect of culture is language. The Whorfian hypothesis asserts that the structure of the vocabulary and grammar of an individual's language actually shapes that person's view of the world (Hunt & Agnoli 1991). Others have argued that language constitutes the most important instrument of socialization; that reality is filtered, apprehended, encoded, codified and conveyed via some linguistic shape (Smitherman 1991). Regardless of whether one accepts such strong hypotheses, the notion that language is a critical aspect of culture cannot be rejected (Fasold 1991, Gumperz & Hymes 1964).

In relating language to diversity, part of the definition of a subculture may be manifested in linguistic differences (Wolfram 1991). Language evolves and changes continually, with people tending to speak most similarly to those around them. Furthermore, language can serve a unifying function for sociocultural groups (Fasold 1984). Language is a means by which individuals locate themselves in social space. Speech is an act of identity: when we speak, we identify ourselves as belonging to a particular group, be it gender, social class or race (Coates 1986).

THE AFRICAN AMERICAN MARKET

African Americans constitute an important subculture in the U.S. This market segment totals 29.5 million consumers earning over $200 billion, 80% of which is disposable income (Stith 1989). Many subcultures, regions, and social groups in the U.S. have their own unique dialects. [The term dialect is used in sociolinguistics as a neutral label to refer to any variety of language which is shared by a group of speakers. Dialects are not deviant forms of language, rather they are simply different systems with distinct sets of language patterns (Wolfram 1991).] This is true for African Americans as well. The dialect spoken by African Americans was called Black Vernacular English by sociolinguists in the '70s and '80s, but has more recently been given the name African American Vernacular English (AAVE) (e.g., Baugh 1991, Smitherman 1991). Vernacular refers to dialects that incorporate nonstandard language forms. Unfortunately, because the phonetic or grammatical features found in vernacular dialects are not taught by formal grammarians to be "proper," they are often stigmatized by American society.

AAVE

It has been estimated that a large majority of the African American population uses AAVE. People often quote Dillard, as an expert in this area, who asserted that 80% of African Americans use AAVE to some extent (e.g., Dillard 1972 in Williams & Qualls 1989). Many years ago, people believed AAVE was nothing more than distorted Standard English, spoken by unintelligent or uneducated people. Sociolinguistic research has shown that AAVE is a complex, highly consistent dialect that follows grammatical rules to the same extent as do the more socially accepted forms of English (Wolfram 1991). This is truly a dialect of English, not just slang. Sociolinguistic studies conducted all over the U.S., from major cities to rural areas, from the East Coast to the West, provide strong evidence that there is clearly a common core of linguistic features and structures that define AAVE (Wolfram 1991, Fasold 1991). There are some regional differences, but there exist overriding similarities everywhere AAVE is spoken.

One example comparing AAVE to Standard English (SE) illustrates that widely accepted language forms need not be more consistent than those labeled vernacular: reflexive possessive pronouns. Compared to SE, the AAVE forms are actually more consistent, as follows:

TABLE

The SE forms are a mixture of possessive pronouns and objective pronouns, while in AAVE they are completely consistentCthey are always the possessive pronoun. Myself is equivalent to my book, as is yourself (your book), herself (her book), ourselves (our books) and yourselves (your books). However, the SE third person masculine forms are inconsistent: himself (compared to his book) and themselves (their books). AAVE's forms (hisself and theirselves) are actually more consistent than those of SE.

It turns out that language change is guided by logical, evolutionary principles (Wolfram 1991). There are a variety of language influences acting on English, adapting it to carry out communication needs under ever changing physical and social conditions. Some groups adopt changes while others do not, leading to linguistic variety. Vernacular dialects are often the result of these natural processes and thus initiate more internal consistency in language than the rigid forms prescribed by elementary school grammar teachers. And it cannot be said that one form of language has evolved from the other. In some cases dialects cling to older forms than those in prescriptive grammars, while in other cases the dialect form is an evolutionary step forward.

Language Attitudes

In American English, standardness is defined and prestige is achieved by a lack of stigmatized grammatical structures (Wolfram 1991). Vernacular dialects, on the other hand, exhibit the presence of socially obtrusive structures. One example of a frowned upon dialectal feature is the use of double negatives in AAVE, and many other American dialects, (e.g., They don't want none), even though this is a widely used form in languages other than English. Since non-standard forms are socially disfavored and of low status, and AAVE contains nearly entirely socially stigmatized variants as its set of unique features, general population attitudes towards AAVE are surely unfavorable. There may, however, be covert prestige in the use of AAVE, especially among African Americans who strongly identify with black culture. In this way, use of AAVE provides status and solidarity. The problem, however, is that covert prestige is relevant only to subculture members as it is unlikely to be understood or valued by the general population.

Fasold (1984) summarized research on how people in general tend to rate speakers of the high and low forms of a language. It has been found that high language forms are considered to be high on status and intelligence, while low forms are found to be high on affect, trustworthiness, and friendliness. Other studies have found Standard English speakers to be judged as more ambitious, competent and self confident, while vernacular speakers were evaluated as higher on personal attractiveness (i.e., good natured, talkative and good sense of humor) (Coates 1987).

Sociolinguistic studies have found an interaction between language attitudes and other preexisting stereotypes. For example, Williams (1974 in Fasold 1984) found that teachers, listening to the same voice track, rated the speaker differently depending on whether a black, white, or Hispanic child was pictured speaking. The teachers' ratings were anchored by their preexisting stereotypes on dimensions that have been shown to correlate with school performance. Extending these findings, people with strong negative stereotypes may find people who use a moderate degree of dialect features to be speaking with a very strong dialect and to possess the negative characteristics that they associate with the stereotyped group.

RELEVANT MARKETING THEORY

The consensus in marketing is that both blacks and whites are offended by the use of vernacular speech. Advertisers who use "black slang" are accused of promoting negative stereotypes. For example, in the 1930s and '40s, advertisers used offensive stereotypes that were perceived as being racist, as in a toothpaste ad that showed a black boy eating a watermelon and declared "Go right ahead, Sambo! Sink those ivories in that luscious watermelon." In another ad from the 1930s, a black "mammy" exclaimed: "Lawsee! Folks sho' whoops with joy oyer Aunt Jemima pancakes" (Alsop 1984).

One of the problems with the current use of AAVE in advertising is that many ads are written by whites and are thus unnatural for the black actors. Blacks are told to speak in a "street style" even if they normally would not speak in that fashion. This has been referred to as the "right-on" school of advertising (Alsop 1984). Obviously such lack of naturalism will be detected by African Americans watching the commercial, leading to unfavorable reactions. It may be that people associate AAVE with negative images because of the correlation between AAVE use and stereotypical presentations.

Actually, much of recent consumer behavior research into black advertising has found that whites are not averse to seeing blacks in ads, while African Americans respond more favorably to ads when black actors are present (Whittler 1989, Pitts et. al. 1989). However, these studies used print ads and television commercials with singing and jingles rather than spoken words. The print ad study found that strong race identifiers, both blacks and whites, used heuristics to process ads with African American actors (Whittler 1989). Specifically, the heuristic was to base one's response on a similar-agree, dissimilar-disagree standard (i.e., strong black identifiers would agree with black actors but disagree with white actors and vice-versa for strong white identifiers), rather than to evaluate the ad's message content. Furthermore, the author states that video and audio presentations of ads may make an actor's race more salient, thus increasing the use of these heuristics.

The television ad study asserted that previous studies using print media lost rich cultural messages that video and audio components of commercials provide (Pitts et. al. 1989). The authors found that TV ads produced by black advertising agencies targeting an African American audience conveyed value themes to black viewers that white viewers failed to perceive. This occurred due to the use of symbols and icons understood only by the distinct black subculture, which possesses a unique complex of behaviors, tradition, language and values. While black respondents displayed more positive affect towards these commercial messages than did comparable whites, the latter group did not respond negatively to the ads.

While marketing studies have often examined the realistic portrayals of blacks in advertising from the viewpoint of it being marketers' social responsibility to erase undesirable stereotypes (e.g., Kassarjian 1969), in today's diverse society, believable incorporation of cultural ad elements provides a benefit for marketers as well: it improves their ability to successfully target important subcultures. AAVE, as an important linguistic cultural element, should create connections to this growing market segment.

SOCIOLINGUISTIC DATA ANALYSIS

In order to assess the degree to which AAVE is used in television advertisements, an exploratory sociolinguistic study was conducted.

Methodology

Selection of Advertisements

Two sources of commercials were used. The first was a collection of 77 commercials produced by black advertising agencies, targeting the African American consumer. These ads, however, date back to the mid 1980s. For the second set, I taped over 100 hours of prime time and sports television from the three major networks, various local and cable stations, including MTV and BET, in the Fall of 1992. To this was added over 20 hours of daytime and BET television from Los Angeles to include geographic diversity. The ads videotaped from television were reduced to a set that included only those ads that prominently featured black actors. These ads were selected for analysis because they featured black protagonists or contained African Americans with speaking parts.

Categorization

In the real world, AAVE features are not categorical. AAVE is made up of a constellation of phonological and grammatical structures. However, because commercials are "standardized" in their language, I took any example to be categorical for this study. Each ad has been placed in just one category, based on the highest feature found in the continuum described below. For example, a commercial featuring a black African American who uses both AAVE vocabulary and mild AAVE phonology would be classified as "mild phonology." The classification system is highlighted in Table 1.

The final three classifications deserve further elaboration. Phonology refers to the sound patterns of language. Here, phonological features were divided into two groups: mild and strong. Mild phonology was essentially a standard Southern dialect, which has undergone a vowel shift moving short front vowels upward (e.g., bit becoming like beet), long front vowels backward (e.g., bait becoming like bet), and back vowels forward (e.g., suit becoming like soot) (Wolfram 1991). This categorization also included AAVE intonation and vocabulary. G-dropping (e.g., fishin', runnin') was also labeled a mild phonological feature.

TABLE 1

TABLE 2

Strong phonology included final stop cluster reduction (e.g., bes for best); /d/ (the initial sound in that) and /q/ (the initial sound in thing) becoming [d] or [t], respectively, at the beginning of a word (e.g., dat for that), or [f] or [v] respectively, at the end of a word (e.g., toof for tooth); /d/, /z/, and /v/ becoming stops mid word (e.g., sebm for seven); and other specific word examples, such as ask becoming aks.

AAVE grammatical structures consist of the following forms: present tense third person /s/ absence (e.g., He play basketball), plural /s/ absence (e.g., Those five kid over there), double negatives, remote time been (e.g., I been known her meaning I have known her a long time), possessive /s/ absence (e.g., John book), copula absence (e.g., She nice), and habitual "be" forms (e.g., He be looking good, which implies a continuing, habitual form not present in Standard English but found in many other languages).

Results

As mentioned above, the study was conducted on two commercial sets. Because the two data sets are not scientifically comparable, the results are discussed separately.

Television Ads Targeted to African Americans

The first ad set consisted of 77 commercials produced by black advertising agencies in 1987. The product categories for these ads included: deodorant, laundry detergent, cooking oil, beer (and malt liquor), coffee, soft drinks, diapers, toothpaste, shampoo, fast food restaurants, hair relaxant, amusement parks, public service announcements, tires, automobiles, telephone services, auto products, film, and baby food.

The first column in Table 2 displays the tabulation of the degree of AAVE used by the African Americans in this set of advertisements. The percent indicates the proportion of the ads that were coded as using a particular level of AAVE (e.g., 10.4% of the 77 ads from 1987 had AAVE intonation patterns, but no AAVE vocabulary, phonology or grammar). Each ad was only categorized once, depending on the strongest form of non-standard language used.

Obviously, the singing only ads (19.5%) did not feature any spoken language, and so an assessment of AAVE use could not be made. In those ads classified as "Can't tell Black" (16.9%), the unseen announcers had voices that were so standard that the race of the speaker could not be determined. Just over 20% of the time, the voice of the announcer was identifiably black, but the language spoken was standard. The voice was usually male and essentially the classification was based on voice quality alone. [Studies have shown that people are surprisingly good at identifying an unseen Standard English speaker's race, although it is difficult to specify on what basis the judgments are being made (Wolfram 1991).]

Next, 10.4% of the ads were categorized as "Intonation." In these cases the rhythm of the speech and the fluctuation in pitch were identifiably African American. Some examples include: I just l!o!v!e our privacy, and Looks this g!o!o!d. Of the ads, 5.2% went beyond simple intonation and included black vocabulary, often stereotypical words. Examples from this ad set include looks just so, kind of sweet on you, Mama, Baby, and Child. Furthermore, I included such expressions as Ooowie, Mmm mm, and Ooow as vocabulary classifiers.

Mild phonology was present in 11.7% of the ads. Examples of mild phonology include the speech of Lena Horne and Debbie Allen, two entertainer celebrity endorsers featured in these ads. Just over 14% of the ads were categorized as "Strong Phonology." Some examples of the AAVE phonological features found in this set of ads consist of: Don't let the smoov taste fool ya (King Cobra Malt Liquor), Da new Ford Taurus, Crest toofpaste, and Wid every can ah Crisco.

Finally, it was noticeable that essentially no ads used AAVE grammatical structures. I was liberal in my classification of one ad as having AAVE grammar. This particular ad presented an interview with an employee and was filled with other features of informal English, including repeated use of the phrase you know. The employee said things such as Look atcha, How am I cusomer service if I can't help you when you aks for ma help? and I stand there and go diggin' through the merchandise wid 'um. These are border line, non-prescriptive English grammatical structures, but do not really constitute a full constellation of AAVE. However, this was by far the most realistic portrayal of AAVE speech found in this sampling of advertisements.

General Viewing Television Ads

In the set of ads from 1992, 42 distinct ads had blacks prominently displayed. The product categories for these ads included: beer, diapers, automobiles, laundry detergent, cereal, education and public service announcements, digestive products, credit cards and financial services, orange juice, athletic shoes, department stores, fast food restaurants, telephone services, auto products, toothpaste, games and toys, undergarments, soft drinks, and insurance companies. Although 42 ads may seem to be a small number, ads that were aired repeatedly only count as one ad for this study. Additionally, strict criteria were used to narrow down the selection of ads. Many ads have black actors in a large group of racially diverse individuals but the blacks do not have speaking roles. These ads would not have been included in the sample. [A study conducted in 1984 showed that 9% of all ads featuring live actors had African Americans in them, slightly less than the black population of roughly 12% of America (Alsop 1984), although this figure has likely increased since that time.]

The second column in Table 2 displays the tabulation of the degree of AAVE used by the African Americans in this second set of advertisements. The coding scheme for this sample of commercials followed the same categorization criteria described in Table 1 above. Much of the same vocabulary found in the 1987 sample was present in the 1992 ads. Vocabulary features present uniquely in this collection of ads included Man and Amen. Words exhibiting strong (AAVE) phonological features are found in the following phrases: Dinnah basket, I love dis place, Dat's the wildest thing, Dey oughta, o evah hopeto, and dis cup.

Grammatical features included in these ads consisted of one double negative, several irregular uses of the verb to get, and one fairly complete constellation of AAVE features. At the end of a K-mart commercial, where many different customers are shown walking through the department store and the viewer is allowed to "eavesdrop" on their conversations, a black woman uses a double negative, subtly and as the ad fades away.

The irregular use of the verb get, in the place of have, is found in the popular Diet Pepsi campaign, featuring Ray Charles, with the now familiar slogan, You got the right one baby, Un Huh. The use of you got is either a case of auxiliary verb deletion as in you've got, or a correctly formed past tense. However, the prescriptively correct grammatical structure is you have. Nevertheless, use of the got form is, in terms of descriptive grammar, becoming more and more widespread and accepted across all varieties of English.

In a commercial for Gain detergent, a man is shown to be an entrepreneur, running a landscape business. He says such things as Lucky for me, I got this green thumb. Here what is missing is the contracted form of have, because his more correct wife responds: Lucky for me, I've got this little scoop. The husband also says specially for especially and plantin' with g-dropping. Another commercial, for Sunny Delite, has a group of young boys raiding the refrigerator. They also use the got rather than the have form, as in: We got soda.

There is one case of present tense third person /s/ absence. In a commercial for Budweiser, a group of black men are playing basketball. Curly Neal approaches them, asking to play ball. They respond, Say what? Wid chu? To which he says, No, wid you mama. Later, after the former Harlem Globetrotter has turned out to be an excellent dribbler and is able to slam dunk, one of the basketball players says, Man, he deserve one a deese, referring to the beer, of course.

Finally, a McDonald's ad features three young black men talking about a friend who has recently done well in his job at McDonald's. They greet each other with Waz up, a greeting of solidarity, and reply Nothin' much. When the young employee calls his mother to tell her about a promotion, he calls her Mama and she calls him Baby. The friends admit the money he earns allows him to buy fresh clothes. When they start teasing him, one says Don't rip my boy too hard now. These aren't really AAVE grammatical structures, but I rank this ad high in terms of overall AAVE flavor and use of a symbolic interaction routine, neither of which is captured in my categorization criteria. Again, it is noticeable that plural /s/ absence, remote time been, possessive /s/ absence, copula absence, and habitual "be" forms were not included in a single ad, despite the fact that many of these forms, especially habitual "be," are recognized AAVE structures. The grammatical features highlighted above were very "mild" examples of AAVE.

DISCUSSION AND FUTURE RESEARCH

This paper reviewed the role of AAVE in television's portrayal of African Americans in advertisements. The sociolinguistic analysis showed that the use of AAVE was quite low, with a some phonological features and stereotypical vocabulary items, but very few AAVE grammatical structures. The explanation for the lack of AAVE found in television advertisements is threefold. First, the language attitudes that permeate our society greatly stigmatize vernacular grammatical structures. Second, given the historical portrayal of blacks in commercials, advertisers are especially careful to avoid any ad variable that could be misconstrued as racially insulting. And third, for a majority of products, the individuals who dominate the marketing decision making process are primarily non-African American. Thus advertising that includes black actors is developed based on non-African American conceptions of what it should look and sound like.

This study is not without limitations. First, the coding taxonomy will be difficult to replicate. Second, only a single judge was used, which precludes the reporting of reliability measures. In all, the study suggests that dialectal language variation is under utilized as a cultural targeting mechanism. This is primarily due to a lack of understanding on the part of consumer researchers and marketing practitioners. In order to help remedy this situation, I next propose a framework to guide future research in this area.

Directions for Future Research

In the increasingly diverse American society, marketers must be able to target one subculture without offending another or the general population. An important aspect of reaching diverse groups is to include cultural elements in ad campaigns. Language variation is an important cultural symbol for many potential market segments. Grammar, phonology, and vocabulary vary across ethnic groups, regional groups, class structure, gender and age. In fact, many of the AAVE findings reported above could be replicated for Hispanic advertising or commercials targeting teens. But language variation can evoke strong reactions, from covert prestige to negative stereotypes. In order to understand dialectal differences, a variety of issues must be addressed by consumer behavior research. Figure 1 highlights a framework for the study of the effectiveness of language variation in advertising.

FIGURE 1

To target a particular subculture, one must relate to its members through realistic cultural messages. In order to create realism, there must be a perceived fit within the cultural aspects of the ad and between these aspects and non-cultural ad features. Thus, reaction to dialect features in ads will be mediated by the perceived fit between linguistic form and other aspects of the advertisement. Language variation should not be examined independently of a wide range of ad elements that impact whether viewers respond positively or negatively to the use of vernacular language features. These ad elements include, but are not limited to, ad setting, communication style, product category, type of persuasion (e.g., informational vs. transformational), media selection, degree of targeting, program context, actors, music, and use of other cultural symbols and values. Perceived fit issues relating to these execution variables will be briefly discussed.

As mentioned above, language variation is not categorical in nature. There are degrees of non-standardness; dialect usage spans a continuum. In normal daily speech, nearly everyone uses some language features that could be described as vernacular. The question is to what extent, what percent of the time are these features used. Therefore fit implies the correct level of vernacular features given other goals for the ad campaign.

Setting is an important variable for perceived fit. Language use changes depending on the environment. In more formal settings, language becomes more formal, more "standard." As familiarity with others increases and the setting becomes more informal, people tend to use more variation in their speech. Thus, when an advertisement's setting is formal, especially in a professional environment, more standard English forms should be used. In a more informal setting, use of vernacular forms can increase, properly reflecting the true usage of language in our society. The degree of formality issue extends to other ad elements as well, including communication style and product category. For example, an ad for insurance using a fear appeal will not be as conducive to dialect use as an ad for a sports-related product in an action driven setting.

Written language is, by its very nature, more formal than spoken language (Stubbs 1983). Therefore, language variation effectiveness depends on the proper choice of media, be it print, radio, or television. Non-standard linguistic forms, which are more informal in nature, are better suited to be heard than read. A content analysis of advertisements in three well known magazines targeting African Americans (Jet, Ebony, and Upscale, April 1993 issues) found no written forms of AAVE phonology or grammar, supporting the notion that written language is more formal than oral speech acts. There were two cases of limited dialect differences in vocabulary: written reference to McDonald's restaurant as "Mickey D's" (Jet) and a Nike ad which encourages fathers not to "sweat" their sons (Ebony).

The more specifically marketers can target their messages, the more they can use idiosyncratic cultural elements, because fear of offending other groups is reduced. The degree to which advertisers can target their messages depends in part on the product category and media selection. For example, written AAVE vocabulary, phonology and grammar were found in the magazine Rap Pages (April 1993), targeted at young Rap fans, primarily black and Hispanic, although only in ads for Rap music (e.g., "Dead enz kidz doin' lifetime bidz"). Obviously, the surrounding text, which is also written almost exclusively in AAVE, plays a major role in what degree of non-standard lect can be used. The same extends to program context in television advertising. For example, given the increasing use of AAVE forms in television programming (e.g. Fresh Prince of Bel Air, A Different World), advertising that includes AAVE shown during these programs should be more effective.

Despite the fact that the majority of African Americans speak AAVE, it would be a mistake to assume the black subculture is entirely homogeneous. One subsegmentation strategy is based on the intensity of ethnic affiliation. Deshpande et.al. (1986) found that strength of cultural identification influenced Hispanics' attitudes towards Spanish media. Williams and Qualls (1989) found that strong black identifiers had more favorable attitudes towards black celebrity advertising than weak black identifiers and whites. The hypothesis is that language variation should be more effective for highly intense ethnic affiliators than for those who are weak cultural identifiers.

Finally, marketers must be careful to combine language variation with other cultural ad elements appropriately. The actors must be clear members of the subculture. The cultural symbols and values must be carefully researched and understood. Choices in music selection must also be made sensitively. Each of these cultural elements can backfire when used inappropriately. Alternatively, portrayal of situations that are strongly correlated with ethnic culture need to include dialect features in order to be considered believable. The Budweiser commercial set in the inner city with black actors playing basketball is a case in point. How credible would this ad be, to any viewer, if it used Standard English?

In conclusion, who we are is reflected in how we speak. In order to communicate with and relate to the growing and diverse subcultures in American society, we must understand how language is used across these groups. Use of language variation to target subcultures must be done with sensitivity, because improper use of vernacular English will be rejected by the targeted subculture and may be offensive to other groups who view the ad. An important dimension to the successful implementation of targeted language strategy is the perceived fit between the level of dialect and other ad elements.

REFERENCES

Alsop, Ronald (1984), "Firms Still Struggle to Devise Best Approach to Black Buyers," Wall Street Journal, October 25, p. 35.

Baugh, John (1991), "The Politicization of Changing Terms of Self References among African American Slave Descendents," American Speech, Summer, v. 66, pp. 133-146.

Coates, Jennifer (1987), Women, Men and Language: A Sociolinguistic Account of Sex Differences in Language, New York, NY: Longmen Inc.

Deshpande, Rohit, Wayne D. Hoyer and Naveen Donthu (1986), "The Intensity of Ethnic Affiliation: A Study of the Sociology of Hispanic Consumption," Journal of Consumer Research, v. 13, September, pp. 214-220.

Fasold, Ralph (1991), The Sociolinguistics of Language, Cambridge, MA: Basil Blackwell, Inc.

Fasold, Ralph (1984), "Language Attitudes," Chapter 6 in The Sociolinguistics of Society, Cambridge, MA: Basil Blackwell, Inc.

Gumperz, John J. and Dell Hymes (1964), The Ethnography of Communication, American Anthropologist Special Publication, December, v. 66, n. 6, part 2.

Harper, Lucinda (1992), "Census Bureau Lifts Population Forecast, Citing Fertility, Immigration, Longevity," Wall Street Journal, December 4, s. B, p. 1.

Hunt, Earl and Franca Agnoli (1991), "The Whorfian Hypothesis: A Cognitive Psychology Perspective," Psychological Review, v. 98, n. 3, pp. 377-389.

Kassarjian, Harold H. (1969), "The Negro and American Advertising, 1946-1965," Journal of Marketing Research, v. 6, pp. 29-39.

Pitts, Robert E., D. Joel Whalen, Robert O'Keefe, and Vernon Murray (1989), "Black and White Response to Culturally Targeted Television Commercials: A Values Based Approach," Psychology and Marketing, v. 6, n. 4, pp. 311-328.

Smitherman, Geneva (1991), "'What is African to Me?': Language Ideology and African American," American Speech, Summer, v. 66, pp. 115-132.

Stith, Melvin T. (1989), "Introduction," Psychology & Marketing, v. 6, n. 4, pp. iii-iv.

Stubbs, Michael (1983), Discourse Analysis: The Sociolinguistic Analysis of Natural Language, Chicago, IL: University of Chicago Press.

Whittler, Tommy E. (1989), "Viewers' Processing of Actor's Race and Message Claims in Advertising Stimuli," Psychology & Marketing, v. 6, n. 4, pp. 287-310.

Williams, Jerome D. and William J. Qualls (1989), "Middle-Class Black Consumers and Intensity of Ethnic Identification," Psychology & Marketing, v. 6, n. 4, pp. 263-286.

Wolfram, Walt (1991), Dialects and American English, Englewood Cliffs, NJ: Prentice-Hall, Inc.

----------------------------------------