In 1979, Rapper's Delight by The Sugarhill Gang became the first hip-hop song to receive national radio play in the U.S. For many outside of New York City, Rapper's Delight was a first exposure to hip-hop. Couched in all the novelty was a non-mainstream vocabulary, much of which was African-American English (AAE; also called African-American Vernacular English or Black English). AAE words used in Rapper's Delight, such as fly (“cool/attractive”) and bad (“cool”), subsequently enjoyed some prominence in Mainstream American English (MAE; also called “Standard” American English) throughout the 1980s. Could it be that non-African-American speakers learned these words through listening to hip-hop songs such as Rapper's Delight?

Possibly. A speaker's vocabulary grows dramatically during adolescence and young adulthood [2], although more research is severely needed to investigate how this occurs. During these stages, various media forms are used for socialization purposes [3]. Perhaps it is not surprising that, given enough visual and linguistic context, speakers appear to acquire vocabulary from watching movies or television shows [4].

It is less clear, however, that vocabulary acquisition can take place through listening to music, particularly hip-hop. Even outside of hip-hop, listeners often misunderstand lyrics: they were seven times more likely to incorrectly transcribe sung words than spoken words [5]. Thus we would expect music listeners to have only a thematic understanding and memory of lyrics; previous research suggests this is the case [6], [7]. Furthermore, several barriers make it difficult to adequately understand hip-hop lyrics in particular [8], [9]:

These factors make for “excruciatingly difficult” conditions for lyric comprehension and transcription [9], not to mention subsequent vocabulary acquisition. A well-researched example of vocabulary acquisition under suboptimal conditions concerns hearing-impaired populations, who have lower vocabulary acquisition rates than their non-hearing-impaired counterparts [12]. Similarly, it seems challenging, perhaps especially for non-African-Americans, to acquire vocabulary as a result of listening to hip-hop.

On the other hand, adolescents and young adults listen to quantitatively more music than previous generations [3]. For example, 79% of American teens and three-quarters of 18–24 year-olds have an mp3 player [13]. Due to the ubiquity of not only mp3 players but also smartphones and internet connectivity, these populations receive increased linguistic input from popular culture icons such as music artists. From its South Bronx roots and subculture origins, hip-hop music has evolved to be fully mainstream [10]; many adolescents and young adults are regularly exposed to it. In listening to the same songs more, listeners benefit from repeated learning, enabling them to better process details. Furthermore, the availability of videos online enables immediate video viewing, which could facilitate vocabulary acquisition by offering visual context accompanying unclear lyrics whenever listeners want to watch. Websites devoted to slang, such as Urban Dictionary (http://www.urbandictionary.com/http://www.urbandictionary.com/), and to hip-hop/AAE vocabulary in particular, such as Rap Dictionary (http://www.rapdict.org/http://www.rapdict.org/), allow for explicit querying of words with unclear meanings, and Rap Genius (http://rapgenius.com/http://rapgenius.com/) aims to explicitly decode hip-hop lyrics, with or without new vocabulary.

Widespread listening to particular artists using the same words could lead to large-scale vocabulary acquisition across social groups. In fact, given the increasing prevalence of the media in young adults' lives, it is surprising that few studies examine first-language vocabulary acquisition through the media. This is particularly important research since vocabulary acquisition represents long-term learning: unlike speed, memory, and reasoning skills, vocabulary skills improve with age [14].

The context of AAE in the U.S. is ideal for testing vocabulary acquisition through listening to hip-hop music. Although AAE and MAE are mutually intelligible, AAE has regular linguistic features, including vocabulary differences, that make it a legitimate, distinct variety (or dialect) from MAE. Due to continued segregation patterns for African-Americans [15], [16], many MAE speakers might rarely interact with speakers of AAE. When they do, it is possible that AAE speakers do not use full-fledged AAE [17]. On the other hand, as the prestigious linguistic variety amongst hip-hop artists, AAE is often used in hip-hop lyrics [18]. Hip-hop could thus represent a primary means of exposure to AAE vocabulary for many MAE speakers.

To see if speakers might be learning AAE vocabulary from hip-hop, I studied speakers' comprehension vocabulary of AAE by asking them to give definitions for a subset of AAE lexical items that could likely occur in hip-hop songs. Due to hip-hop's genre-specific themes of violence, however, the AAE vocabulary used in hip-hop should not be considered representative of AAE vocabulary in general. The term grip is an example of a stimulus of item in this study, occurring in the 2003 Jay-Z song “Dirt off your shoulder” at 1∶52–1∶56:

(1) I paid a grip for the jeans, plus the slippers is clean Anecdotally, some listeners report not understanding the above lyric at all because of the fast tempo, but the visual context of the video (http://www.youtube.com/watch?v=Oz_-VaTHpc8) could help speakers interpret a grip to mean “a lot”. Other vocabulary in hip-hop lyrics can be more or less easy to interpret. The following study provides evidence that comprehension vocabulary of stimulus items such as grip stems in part from listening to hip-hop.

Because these words are not (yet) well-known by many speakers of MAE, it is likely that familiarity with these words will have some link back to the AAE speech community, either from social ties or from exposure to popular culture including hip-hop music. I hypothesized that a preference for hip-hop music would be positively associated with participants' AAE comprehension vocabulary. If however the communication channel is too noisy for vocabulary acquisition, as previous research seems to suggest, then we would expect no effect of a preference for hip-hop music on AAE comprehension vocabulary.

Several control factors were also assessed to determine whether a preference for hip-hop was not masking any underlying variables actually responsible for AAE comprehension vocabulary. A series of demographic variables such as age, sex, ethnicity, and geographic area are important for other types of language use by adolescents and young adults (see, inter alia, [19]–[22]). Second, social network variables such as strong and weak ties (roughly, close friends/family vs. acquaintances) are also important factors in the diffusion of linguistic innovations [23], [24]. Finally, general popular-culture knowledge could be responsible for knowledge of AAE vocabulary. This possibility was assessed by asking participants questions about popular culture that college-aged students could know.

In a multiple linear regression analysis, results from the survey indicated that musical preferences, weak social ties to African-Americans, and knowledge of popular culture were significant in predicting a participant's AAE vocabulary score. Specifically, the more hip-hop artists participants listened to, the more they were familiar with the AAE vocabulary items tested. Additionally, participants' weak social ties to African-Americans and their knowledge of elements of African-American popular culture such as Charles Barkley, Mo'nique, and the comic strip and TV series The Boondocks were significantly associated with increased AAE vocabulary knowledge. In contrast, an increased preference for country music was negatively associated with knowledge of AAE vocabulary items. The model in Table 1 has a multiple of 0.39. These results are visualized in the added-variable plots of Figure 1, which plots the partial correlations between predictor variables on the x-axes and the dependent measure on the y-axes after partialing out all other predictor effects.

Demographic variables such as sex, age, and hometown population information (city and county populations, as well as the county's African-American population) were not significant predictors, nor were the strong-tie variables to any ethnicity.

To ensure that the effects were not due to multicollinearity, I inspected pairwise correlations between predictors; correlations were modest [25] (p. 138). For example, the numbers of hip-hop and country music artists participants listened to were only weakly correlated (). The highest correlation between predictors in the model (, between the number of hip-hop artists listed and the number of weak ties to African-Americans), was similarly low. The variables in the model also had low correlations with those not in the model, such that the former are not masking the explanatory power of the latter. The highest such correlation was , between the number of hip-hop artists a participant listed and his or her hometown county population.

To summarize, the number of hip-hop artists a participant listens to was predictive of his or her AAE comprehension vocabulary. This effect was still significant when other factors, such as participant demographics, knowledge of popular culture, and overall musical preferences, were taken into account.

These results support the hypothesis that non-African-American young adults learn African-American English (AAE) vocabulary through listening to hip-hop. A positive association was observed between AAE comprehension vocabulary and the number of hip-hop artists participants listened to. Additionally, use of a particular word by a participant's preferred artists was predictive of the participant's knowledge of that word.

In addition, the present results suggest that vocabulary acquisition is a multifaceted process in which personal identity, in the form of cultural knowledge, social ties, and other musical elements, plays a crucial role. First, knowledge of African-American popular-culture figures was also associated with increased AAE comprehension vocabulary. Furthermore, the number of weak ties to African-Americans is a better predictor of AAE vocabulary knowledge than the number of strong ties, a finding which replicates the work on the strength of weak ties in the diffusion of information [29]. Third, a preference for country music was negatively associated with AAE comprehension vocabulary, yet this predictor was not negatively correlated with a preference for hip-hop. This finding is in line with country music's connection to cultural Whiteness in the United States [30], [31] (p. 24–30). Crucially then, this personal identity appears to be self-constructed. Factors that speakers have some degree of freedom in choosing, such as social networks and cultural consumption preferences, were predictive of AAE comprehension vocabulary. In contrast, no structural factors influencing identity, such as hometown population size, had a significant effect. Future work could further examine this distinction by looking at a more ethnically diverse non-African-American population, for example.

When accounting for these other possible means of vocabulary acquisition, hip-hop listening preferences are highly associated with AAE comprehension vocabulary. Still, the variables probed in this study are not meant to be exhaustively representative of speakers' media exposure and social networks. Rather, they are indicative of the types of factors that could affect AAE comprehension vocabulary and can suggest future directions for controlled laboratory studies. For example, in a comprehensive attempt to find interactions between all effects (see supporting information S1), none were found, which suggests that hip-hop listening preferences might have an effect on AAE comprehension vocabulary that is independent of listeners' social networks.

Nevertheless, for four reasons, the data in the current study suggest that AAE vocabulary acquisition will likely be linked to hip-hop listening preferences if other variables such as peer group usage and media exposure were able to be exhaustively examined and controlled for. First, some participants made explicit references to hip-hop lyrics in which a particular stimulus item occurred when defining the word. An example is the definition “where you go when you die/from the Bone Thugz [sic] song” [“Crossroads”, Bone Thugs-n-Harmony, 1996] for the stimulus item crossroads (“heaven”); for the item make it rain (“throw money up in the air”) one participant simply gave “Lil Wayne” as a definition. Lil Wayne, a well-known hip-hop artist, was featured in the song “Make it Rain” by Fat Joe, a less popular artist. Second, the correlations between hip-hop listening preferences and all other social and cultural factors such as social networks and non-musical media influences were low; the highest was . The lack of interactions found between predictor variables suggests that hip-hop listening preferences do exhibit an independent effect on AAE comprehension vocabulary. Third, the stimuli in this study were selected because of their relatively low rates of diffusion in Minnesota at the time participants took the study (Fall 2010). In fact, fully 76% of responses to the stimuli indicated that participants were not familiar with the word. This lower rate of diffusion, throughout both social networks and various forms of media, goes hand in hand with a decreased probability of multiple sources of exposure to a word. Finally, the fact that artists' aggregate popularity was a better predictor than the number of artists using the word suggests that even less involved listening to hip-hop could be responsible for familiarity outside of listeners' preferred artists. Listeners can often hear a popular artist such as Lil Wayne on the radio and through their friends, but may not actively choose to listen to this artist.

Most work on vocabulary acquisition from the media examines infant or non-native speaker populations, often in a pedagogical setting. If explicit attention is directed to the media, second-language learners are able acquire vocabulary from that source [32]. However, infants do not learn vocabulary from baby videos well [33]. Neither of these types of studies examine vocabulary acquisition as a result of speakers' voluntary interactions with various media forms. Adolescents can use media in constructing their personal identities [3], and learning in this context is most likely very different from pedagogical learning. For one, learner motivation is different. Adolescents could learn AAE vocabulary (from hip-hop) to be “cool”, to show affiliation with a social group, and to establish themselves as linguistic early adopters. Second-language classroom learning involves motivation to master the material and get a good grade.

Other work on vocabulary acquisition from the media makes claims that appeal to common sense. Reference [4] observes that buzzwords are clearly propagated by the media and are integrated into the active and passive vocabularies of at least some speakers. Examples of this phenomenon in the U.S. include trendy memes such as truthiness, Stephen Colbert's 2005 coining. But the words examined here are not buzzwords or memes; rather, they stem from a non-mainstream variety of English. Whereas buzzwords generally have a clear instance of coining, the same is not necessarily true for AAE vocabulary. And without an empirical study also examining social networks, it is impossible to tell whether this vocabulary is learned directly from the media for all speakers, or if they pass first from the media to influential speakers and then to other speakers. Previous linguistic theories posit that the media only influences a small number of speakers, who then pass on linguistic innovations to the majority of speakers in a community [19] (p. 356–357). In contrast, in this study, 48% of non-African-American participants listened to at least one hip-hop artist. Coupling this information with the present finding that young adults can learn AAE vocabulary from listening to hip-hop, as well as previous findings that adolescents use media as a form a self-socialization [3], it is possible that a larger percentage of the population than previously thought could be directly influenced from linguistic input from the media. In sum, further studies are needed on first-language vocabulary acquisition from voluntary interactions with various media forms.

Recent work shows an association between Scottish speakers who watch the London-based TV series East Enders and the use of phonetic features of Southern English [34]. Similarly, since speakers' comprehension vocabularies appear to be influenced by hip-hop, it is possible that speakers will subsequently use AAE vocabulary they hear in hip-hop songs. Since administering this survey, I have heard and seen members of the demographic studied here use finna (“gonna”), chedda (“money”), and ballin' (“living the good life”), three of the stimulus items of the present study. AAE is not the prestigious linguistic variety throughout the U.S. and is perceived as indicative of lower socio-economic status than MAE [35] (p. 223). However, its covert prestige [36] in the AAE and MAE speech communities also means its non-mainstream nature is perceived as cool [37]. This could make MAE speakers more motivated not only to learn, but also to use, AAE vocabulary they hear when listening to hip-hop.

For musical, linguistic, and social reasons, hip-hop represents a particularly interesting medium through which vocabulary acquisition can occur. The language of hip-hop lyrics is poetic, emotionally charged, accompanied by music and a fast tempo, and perhaps from another variety of English than the listener's. For these reasons, and because hip-hop artists rarely include lyrics in their liner notes, proper comprehension and transcription of hip-hop lyrics is “excruciatingly difficult” [9]. Thus it is far from established that hip-hop lyrics are understandable, and that unfamiliar words are presented in informative contexts therein. To have an idea of the degree of difficulty of learning vocabulary from hip-hop lyrics, the reader is encouraged to listen to Jay-Z and Kanye West's “Otis” (2011). A stimulus item in this study is featured somewhere between 0∶43–0∶57:

The reader can note at what point, if at all, he or she understands the meaning of the stimulus item. When compared to other examples, in my estimation this item in this context is easy to moderately difficult to understand.

The social factors accompanying hip-hop make for an even more intriguing medium through which vocabulary is learned. AAE is a socially stigmatized variety of English, traditionally spoken by members of an economically disadvantaged ethnic group. Yet African-Americans have a rich musical tradition (blues, jazz, gospel, soul, R&B, hip-hop) that mainstream America is keen to embrace [38]. This disparity has been called into question by African-Americans such as Roger Guenveur Smith, who has asked, “Why does everyone love Black music but nobody loves Black people? ” [38] (p. 5). Hip-hop is the current dominant form of African-American music, and the de-facto language of hip-hop is AAE. Thus hip-hop represents a prime medium through which speakers of MAE are exposed to AAE; as a result they associate the latter with coolness.

Paradoxically, as soon as speakers of MAE learn and adopt AAE vocabulary, hip-hop artists turn away from it. The word bling-bling, or simply bling, has hip-hop origins and was added to the Oxford English Dictionary in 2003, and to the Merriam Webster Dictionary in 2006 (see http://news.bbc.co.uk/2/hi/uk_news/magazine/3192258.stmh and http://www.mtv.com/news/articles/1564621/crunk-makes-it-into-dictionary.jhtml, respectively). Yet hip-hop artists no longer use this term. In 2004, MTV ran a cartoon showing the diffusion of bling starting with hip-hop artists and being adopted by subsequently more mainstream speakers, ending with a shot of bling's lifespan, 1997–2003 (http://www.youtube.com/watch?v=JcMHC-W_Qmc). Street “cred”, the respect gained from familiarity with a tough urban lifestyle, is an essential part of a hip-hop artist's persona. Part of establishing street cred is using cutting-edge, in-group vocabulary not used by the middle-aged European-American woman at the end of the MTV clip. As a result, hip-hop artists are not motivated to make their lyrics accessible to their listeners. This is perhaps one of the reasons why hip-hop artists rarely include lyrics in their album liner notes.

The construction of a speaker's vocabulary as an adolescent and as an adult seems a vital component of his or her identity. Proper use and knowledge of slang, learned or technical terms, regional colloquialisms, and vocabulary from diverse languages and ethnic groups establishes a speaker as a credible member of a social group. Vocabulary use that a social group considers improper could result in mocking (friendly or unfriendly), ostracism, or even lawsuits and termination of employment (as an example, see http://articles.philly.com/2011-01-05/news/27010905_1_n-word-staff-meeting-federal-courts). Although crucial factors of a person's identity are developed during adolescence and young-adulthood [3], little research has been done on the myriad ways in which first-language vocabulary is acquired after puberty. Examining vocabulary acquisition through media sources such as hip-hop seems a promising first step in this line of research.