The Impact of Violent Music on Youth

Barbara Wilson & Nicole Martins. Handbook of Children, Culture, and Violence. Editor: Nancy E Dowd, Dorothy G Singer, Robin Fretwell Wilson. Sage Publication. 2008.

In 1999 in Littleton, Colorado, parents of the Columbine shooting victims linked gothic punk rocker Marilyn Manson to the killings at Columbine High School. They argued that Manson’s (1996) Antichrist Superstar album, along with other forms of violent media, inspired Dylan Klebold and Eric Harris to walk into their school heavily armed and shoot their classmates (“Forget God,” 1999). In 2004, inConfession Fart II, Rapper Joe Budden sang about the beating of a pregnant woman who is unwilling to abort her baby: “Pray that she abort that, if she’s talkin’ ‘bout keepin— it/One hit to the stomach, she’s leakin’ it” (“Island/Def Jam Rap Song,” 2004). The song was pulled from radio stations all across the United States after pregnancy crisis centers claimed the song advocated violence against women. In the same year another rapper, Jay-Z, produced a music video that featured a dramatization of the singer being shot and killed. Controversy over the video resulted in MTV restricting its airtime and including a disclaimer stating that the network does not promote gun violence (Rotter, 2004).

These are just a few examples of popular music that have come under public scrutiny in recent years because of violent content. But they are not isolated cases. Gangsta rap, which emerged as a genre of hip-hop in the late 1980s, is known for its raw and explicit lyrics detailing the street violence, drugs, and sexual experiences of urban gangs. For example, in his 2003 song “What Up Gangsta,” controversial rapper 50 Cent chants the following:

They say I walk around like got an “S” on my chest
Naw, that’s a semi-auto, and a vest on my chest
I try not to say nothing, the DA might want to play in court
But I’ll hunt or duck a nigga down like it’s sport
Front on me, I’ll cut ya, gun-butt ya or bump ya
You getting money? I can’t get none with ya then fuck ya

The album featuring this song, Get Rich or Die Tryin’, sold 872,000 copies in its first week, and went on to become the number-one selling CD in 2003 in the United States (“50 Cent Tops Album Charts,” 2004).

Indeed, rap is the fastest-growing music genre in the United States (Recording Industry Association of America, 2003a). The success of this genre is due primarily to youth listeners. In one nationally representative study of U.S. teens, 53% of 12- to 18-year-olds reported that they had listened to rap music on the previous day (Roberts & Foehr, 2004). The popularity of such music raises alarms for many parents and critics. One well-known sports commentator for the British Broadcasting Corporation recently called gangsta rap “a deadly virus” that is “killing some of our children” (Denny & Suchet, 2004). According to one study, nearly 50% of mothers with children in junior high school believe that violent rap music contributes “a great deal” to school violence (Kandakai, Price, Telljohann, & Wilson, 1999).

This chapter explores the impact of violent music on youth. We begin by describing the music listening habits of children and adolescents, looking at total time spent with music compared to other media, preferences for different genres, and motivations for listening. We also examine how these patterns differ according to age, sex, and race and ethnicity. We then explore the content of music that youth find most appealing. In the third section, we examine young people’s comprehension of music lyrics and videos. This section will be framed in terms of cognitive development and the requisite skills for processing musical lyrics and videos. The chapter then turns to the effects of violent music on children and adolescents. Most research to date has assessed the impact of exposure to violent music on aggressive attitudes and behaviors, but some studies have looked at other outcomes such as depression, risk taking, and even racial stereotyping. The chapter concludes with a discussion of various policy-making efforts launched over the years to deal with violent music.

Music Habits and Preferences

Children in the United States are immersed in music. The typical child in this country lives in a home with three radios, three cassette players, and two CD players (Roberts & Foehr, 2004). Moreover, a full 53% of children under age 8 and 92% of 8- to 18-year-olds have a music medium of some kind in their bedroom (Roberts & Foehr, 2004). Higher levels of household income and parent education increase the chances that a younger child will have a CD player or radio, but possession of music equipment is so high among preteens and teens that it does not vary with income or education. Personal CD players, MP3 players, and iPods are becoming standard gadgets for adolescents.

As might be expected with all this equipment, young people devote a great deal of time to listening to music (see Roberts & Christenson, 2000). Children under 6 years of age spend an average of an hour a day listening to some type of music (Rideout, Vandewater, & Wartella, 2003). Exposure to audio media nearly doubles between the preteen and teen years, so that by age 15 the average adolescent spends 2Vihours a day listening to music (Roberts & Foehr, 2004). During the teenage years, young people also begin showing a preference for listening to music over watching television (Roberts & Foehr, 2004). In fact, in one study, teens were asked to select which of various media they would take if they were stranded on a desert island (Roberts & Henriksen, 1990, as cited in Christenson & Roberts, 1998). Junior and senior high students selected music media (radio/recordings) as their first choice substantially more often than they did any other medium, including television. Based on data like these, many scholars have argued that music is the most important medium for adolescents (Christenson & Roberts, 1998).

In addition to age, gender predicts use of audio media. Girls and boys in early elementary school, do not differ in their music listening, but toward the end of grade school, gender-based differences emerge. In a nationally representative sample of over 3,000 U.S. youth, Roberts and Foehr (2004) found that girls between the ages of 8 and 14, the “tween” years, spend more time listening to music than boys of the same age do. Although previous studies found that this sex difference continues through high school (e.g., Lyle & Hoffman, 1972), recent research suggests it may disappear by late adolescence (Roberts & Foehr, 2004).

The large-scale study by Roberts and Foehr (2004) revealed no substantial differences in time spent with music by race and ethnicity, although earlier studies employing less representative samples found greater music listening by African American youth (Brown, Childers, Bauman, & Koch, 1990) and by Hispanic youth (Lyle & Hoffman, 1972). Based on their findings, Roberts and Foehr (2004) concluded that, “one of the more striking characteristics of audio media is how very democratic they seem to be, at least in terms of how much they appeal to young people from all backgrounds” (p. 90).

Young people not only listen to music, but they can also “watch” it on television with the advent of MTV and other music video channels. The vast majority of préadolescents and adolescents watch music videos at least occasionally (Christenson, 1992b). Yet most studies indicate that listening to music is a far more frequent activity for youth than is watching music videos on television (see Christenson & Roberts, 1998). In the national study by Roberts and Foehr (2004), for example, 96% of 15- to 18-year-olds had listened to some kind of music on the previous day, whereas only 9% of these same teens had watched a music video on television on that same day.

Content Preferences

Sorting music into genres is challenging because the industry is constantly changing. Not only do we have rock, hip-hop, jazz, Christian, country western, new age, and rhythm and blues, but we also have a proliferation of subgenres for many of these categories. For instance, Latin music can be divided into salsa, flamenco, samba, tango, mariachi, Latin pop, tejano, Caribbean, and Andean folk music, to name a few. Despite the ever-changing music scene, young people show strong preferences for certain types of music and these preferences vary considerably across individuals. Music styles are powerful social markers for youth (Christenson & Roberts, 1998), as evidenced by terms such as “punker” and “metalheads.” Musical genres are also important because they differ widely in terms of violent content, as we will explore in a subsequent section.

Children under the age of 8 mostly listen to music created specifically for child audiences (Roberts & Foehr, 2004), such as Jim Gill, Raffi, and Disney tunes. Nevertheless, even very young children can be exposed to music that is violent, presumably because they are in the room with an older sibling or parent. In the Roberts and Foehr (2004) national survey, 7% of children between the ages of 2 and 7 had listened to rap music on the previous day; only 1% had heard hard rock or heavy metal.

By third or fourth grade, children clearly migrate toward popular music (Christenson, 1994). Two genres dominate the preferences of preteens and teens: rap/hip-hop and rock. In the national study by Roberts and Foehr (2004), 52% of the seventh through twelfth graders had listened to rap/hip-hop on the previous day, and 42% had listened to alternative or “modern” rock. The third most popular genre was heavy metal/hard rock, with 19%.

There are some sex differences in musical tastes. Boys are more likely to listen to heavy metal and “harder” forms of rock than girls are; girls are more likely to listen to soft rock, country western, and top 40 (Christenson & Peterson, 1988). However, musical tastes generally vary more strongly by race and ethnicity than by sex. African American youth tend to listen predominately to rap/hip-hop, but also like rhythm and blues/soul, and to a lesser extent, gospel music (Christenson & Roberts, 1998), all of which are linked to Black performers and Black culture. In one study, a full 84% of African American teens had listened to rap/hip-hop on the previous day (Roberts & Foehr, 2004). Hispanic youth also select rap/hip-hop as their favorite, but they listen to alternative rock, rhythm and blues, Latin, and heavy metal as well (Roberts & Foehr, 2004). Caucasian teens spread their listening over a wider range of genres. The two most popular are alternative rock and rap/hip-hop (Roberts & Foehr, 2004). Caucasian youth also report listening to heavy metal, country western, classic rock, and punk in greater numbers than do African American and Hispanic teens. Two generalizations can be drawn here. First, the only genre that is popular with all racial/ethnic groups is rap/hip-hop. Second, African American youth listen primarily to what has been labeled “Black” music in the industry; Caucasian youth listen to both “Black” and “White” music; and Hispanic youth listen to “Black,” “White,” and “Latin” music (Roberts & Foehr, 2004).

Uses and Gratifications

Preadolescence and adolescence are developmental periods often characterized as challenging and sometimes even turbulent (Roth & Brooks-Gunn, 2000). As early as age 9 or 10, youth are beginning to experience physical and hormonal changes (Archibald, Graber, & Brooks-Gunn, 2003), an increased interest in sexuality (Crockett, Raffaelli, & Moilanen, 2003), questions about their identity (Kroger, 2003), and a growing independence from parents (Zimmer-Gembeck & Collins, 2003; see also the Prologue in this volume dealing with developmental differences). During this time, music is a form of communication that plays a central role in the lives of young people (Lull, 1992). Particular songs and even genres of music are often a source of comfort and a way to connect to the peer group, which takes on increasing importance during this time (Berndt, 1996).

When asked why they listen to music, young people list several reasons. Among the most prominent is to manage their moods (Christenson, 1994). In one study of junior high and college students (Gantz, Gartenberg, Pearson, & Schiller, 1978), high proportions of youth said they frequently listen to music to relieve tension or troubles (83%), to get into a certain mood (79%), and to relieve loneliness (67%). Interestingly, males are more likely to use music to get energized or “pumped up,” whereas females are more likely to report using music when they feel lonely or troubled (Larson, Kubey, & Colletti, 1989). Such findings correspond with the different genres that boys and girl prefer. In one study, almost half of heavy metal fans, which are mostly male, reported that they were most likely to listen to the music when they are feeling angry (Arnett, 1991a).

A second gratification sought from music is to relax or relieve boredom. In the Gantz et al. study (1978), a full 91% of young people reported that music helps them pass the time while engaging in other activities such as homework, driving, and even trying to sleep. A third use of music is to establish and maintain social identities (Lull, 1992). Music helps young people make friends, connect to social groups, and even socialize at gatherings. In one study, over half of the teens reported that musical knowledge was an important factor in judging the status of their classmates (Brown & O’Leary, 1971); only school performance and clothing were ranked higher.

Finally, children and teens listen to music for informational purposes. Rock music introduces preteens and teens to explicit themes involving violence and sex that are often not contained on television or in popular films (Lull, 1985). Although young people seldom acknowledge the socialization function of music (Christenson & Roberts, 1998), there is ample evidence that such learning occurs. High schoolers in one study were asked to compare music against other possible sources of guidance on moral and social issues, including church, parents, and friends (Rouner, 1990). A full 25% of the teens ranked music in the top three sources for obtaining information about social interaction, and 16% ranked it in the top three for moral guidance. In another study, 66% of 11- to 15-year-olds indicated that music has “some” or “a lot” of influence on how they deal with problems in their lives, and nearly 50% reported that a particular song had influenced the way they thought about an important topic (Leming, 1987).

Part of the way in which young people learn from music is by listening to the lyrics. There is actually some debate about whether children pay much attention to the words in songs. Certainly the melody and the performers are key aspects of what makes a song popular. But words matter, too. Boyle and colleagues asked nearly 400 students in 5th, 7th, 9th, and 11th grades and in college what their reasons were for liking favorite songs (Boyle, Hosterman, & Ramsey, 1981). The lyrics of the song were ranked as the fourth most important reason, after melody, mood, and rhythm. The singer or group, the dance-ability of a song, and whether friends liked it were three other features that followed lyrics in order of importance. Christenson (1994) found that first through sixth graders also rated lyrics as less important than aspects of the sound and the mood a song encouraged, but the significance of lyrics increased with age. Still, certain genres of music are defined by the very nature of the messages they convey. Heavy metal fans report that they like such music because the lyrics resonate with the way they see the world (Arnett, 1991a), and avid rap listeners cite the lyrics as the most important feature of the music for them, more important than the rhythm or the danceability of a song (Kuwahara, 1992). In other words, the lyrics that are most explicit and oppositional may get the most attention from young people.

Amount of Violence in Music and Music Videos

Although many would agree that music has become more explicitly violent in nature, few studies have systematically analyzed the content of songs. In fact, only one published study could be found that assesses song lyrics for violence. Armstrong (2001) analyzed the content of 490 gangsta rap songs produced between 1987 and 1993. He found that 22% of the songs contained violent and misogynistic (i.e., expressing hatred of women) lyrics. Of those songs with violence, assault was the most frequently occurring criminal offense (50%), followed by murder (31%), rape (11%), and rape and murder combined (8%). Unfortunately, the study did not assess the frequency of violence within particular songs. Also, the sample represents an early period in the genre’s development when lyrics were tamer than they have since become, according to Armstrong. As evidence of this trend, Armstrong analyzed a single top-selling rap album released in 2000, Marshall Mathers LP by Eminem, and found that 11 of the album’s 14 songs (78%) contained violent lyrics. Thus, the figure of 22% is likely to considerably underestimate the amount of violence in gangsta rap today.

One possible reason that lyrics have received so little research attention is because of the advent of music videos. Since MTV’s inception in 1981, critics have charged that violent and sexually explicit videos can have negative effects on adolescent viewers (Duff, 1995). Now, with as many as four music networks offering a constant stream of videos, social scientists have turned away from lyrics to focus more on the visual images that accompany popular music. In one early study, Baxter and colleagues analyzed a random sample of 62 videos that aired on MTV in 1984 (Baxter, DeRiemer, Landini, Leslie, & Singletary, 1985). They found that 53% contained violence. Most of the antisocial behavior consisted of physical aggression against another person. Weapons were seldom used. In a similar study around that same time, Sherman and Dominick (1986) analyzed 166 music videos featured on three different channels, including MTV. The researchers only coded “concept” videos or those that told a dramatized story rather than just showing a group performing. They found that 57% of the concept videos featured violence, a figure very close to that found by Baxter et al. (1985). On average, the videos contained nearly three separate acts of violence. Most of the violence was perpetrated by White male characters between the ages of 18 and 34. Like in the Baxter et al. study, hand-to-hand combat was the most common form of aggression. Only 3% of the violent videos portrayed death of the victim.

More recent content analyses suggest that violence may be a bit less prevalent than the above research suggests. Tapper, Thorson, and Black (1994) analyzed 161 videos that aired on four music channels in 1992. Fifteen percent contained violence. Rich, Woods, Goodman, Emans, and DuRant (1998) also found that 15% of the 518 videos they analyzed from 1994 contained violence. In the most comprehensive analysis to date, Smith and Boyson (2002) analyzed a sample of 1,962 videos drawn randomly to create a composite week of music video programming across three channels. They also found that 15% of the videos featured violence. At first glance, it appears that the amount of violence in music videos has declined since the 1980s. However, several of the earlier studies looked only at concept videos (e.g., Sherman & Dominick, 1986), which inflates the percentages because performance videos typically are nonviolent. Moreover, earlier studies did not assess the range of channels that have been examined more recently. Rich et al. (1998), for example, included the Country Music Television network and Tapper et al. (1994) included the Nashville Network in their samples. When a broader range of networks is examined, the study encompasses more genres, some of which are less violent.

Despite the fact that only about 15% of music videos features physical aggression, the variation by genre is substantial. Rap music videos are consistently more likely to feature violence than other types of videos are (DuRant et al, 1997; Tapper et al, 1994). Smith and Boyson (2002) found that nearly 30% of rap videos contained violence compared to only 12% of rock videos and 9% of rhythm and blues videos. The second-most violent genre is heavy metal (Tapper etal, 1994). Smith and Boyson (2002) found that a full 27% of heavy metal videos contained violence. Music channels also vary in violent content. MTV consistently features more violence than other music channels do (DuRant et al, 1997; Smith & Boyson, 2002). Black Entertainment Television (BET) is also more violent than channels such as VH-1 (Smith & Boyson, 2002). These patterns are primarily due to genre differences in programming. MTV features rap, heavy metal, and alternative rock in almost equal proportions, whereas BET concentrates on rap and soul/rhythm and blues (Tapper et al., 1994). In general, the more a television network carries rap and heavy metal, the more violent are its videos.

What is the nature of the violence featured in music videos, particularly rap videos? As part of their content analysis, Smith and Boyson (2002) assessed a number of contextual factors such as whether the violence is rewarded or punished, who the perpetrator is, and what the motive for violence is, all of which affect the likelihood that a viewer will learn aggressive behaviors from violent content (see Wilson et al., 1997). Smith and Boyson (2002) found that the majority of perpetrators and victims in rap videos are Black males. In contrast, the majority of characters involved in violence in rock videos are White males. The violence in rap videos is also more likely to be presented as justified than in other videos, is more likely to involved repeated acts of aggression against the same target, and is less likely to be punished. Unfortunately, the researchers could not assess the contextual features in heavy metal videos, the second-most violent genre, because too few were contained in their sample.

To summarize, we know much more about the violent content in music videos than in songs themselves. In the mid-1990s, roughly one in five rap songs contained violent lyrics (Armstrong, 2001). Some speculate that rap music is becoming more violent, but we have no clear data to substantiate that. Future research should assess the nature of lyrics more systematically, with particular attention to rap and heavy metal music. With regard to music videos, about one in six contain some violence. Most of this physical aggression is found in rap and heavy metal videos, two highly popular genres among youth. Nevertheless, content analyses of music videos have concentrated on the visual images and have yet to simultaneously analyze the lyrics for violence. In most cases the visuals reinforce the lyrics. But sometimes the lyrics contain additional information. Guns, for example, are displayed in only 17% of violent rap videos (Smith & Boyson, 2002), but guns are talked about in nearly 50% of rap videos (Jones, 1997). Presumably, the combination of violent words and violent images sends a powerful message regarding antisocial behavior.

Young People’s Comprehension of Music Lyrics

The impact of violent music on children depends to a great extent on what sense young listeners make of the content. We have already noted that the words of a song may not be the most important factor in determining what becomes popular, but youth do pay attention to lyrics. Moreover, young people who are attracted to songs that are controversial or oppositional seem to pay closer attention to the lyrics than do fans of more conventional music (Wass et al., 1989). But how well do young listeners truly understand the meaning of popular songs, especially those that are violent?

One of the challenges in comprehending music is that the lyrics often contain oblique references to risqué or illicit topics. Innuendos about violence and sex are common, as is the use of metaphors to express ideas. For example, Insane Clown Posse, a heavy metal/ rap band known for its violent lyrics and carnival-like live shows featuring fires and chainsaws, released a song in 2004 titled “Bowling Balls.” Parts of the lyrics are as follows:

We take a quick ride, homicide, then I confide in you
And I can love you and technically even though you’re dead
You’ll always be around cause I’m keeping your head
I keep heads on shelves everywhere in my cellar….
Your head would mean so much to me
Sometimes I put ‘em in my bowling bag and bring ‘em to work
Play with their hair under my desk, with my bare feet

The song’s title sounds innocent enough but in fact refers to a murderer collecting the heads of his victims.

Understanding lyrics such as these is likely to be a challenge for youth. In one early study, Rosenbaum and Prinsky (1987) asked a sample of 266 12- to 18-year-olds to name their three favorite songs and then to describe the songs in a few sentences. Students were unable to explain 37% of the songs listed as their favorites. Many teens reported that they had no idea what their selected song meant but that they liked the rhythm or melody. Although no precise percentages were given, the researchers found that teens frequently misinterpreted the songs they liked. A number of them thought, for example, that the song “Stairway to Heaven” by Led Zeppelin, which focuses on a young woman’s quest for meaning in life, was about going to heaven on a stairway. The researchers pointed out that students, often unable to grasp complex metaphors, provided very literal descriptions of the songs.

Leming (1987) played three “Top 40” songs to a sample of 11- through 15-year-olds and asked them to describe in their own words what the songs meant. Roughly 25% of the youth reported that they had no idea what each of the songs meant, even though nearly all had heard the music before and knew the artists. Misinterpretations were also common. For example, 36% of the youth thought Olivia Newton-John’s song “Physical” was a plea for physical exercise. Only one third of the sample correctly interpreted the song as advocating sexual relations. Likewise, only 26% of the youth understood that “I Want a New Drug” by Huey Lewis and the News was a song advocating drug use. A higher proportion of teens understood Madonna’s “Material Girl” (67%), but comprehension was still far from universal.

In one of the few studies to assess age differences in comprehension, Greenfield and colleagues (1987) presented two popular rock songs, Bruce Springsteen’s “Born in the USA” and Madonna’s “Like a Virgin,” to a sample of fourth, eighth, and twelfth graders, and college students. Participants received a copy of the lyrics as they listened to the songs and then were asked a series of questions about the lyrics’ meaning. Strong differences in understanding among different age levels were observed for both songs. Only 20% of the fourth graders understood the term “hometown jam” in the Springsteen song, whereas 60% of the eighth graders and nearly all of the twelfth graders and college students understood it. None of the fourth graders comprehended that Springsteen was singing about his despair and disillusionment over living in this country, compared to 30% of the eighth graders, 40% of the twelfth graders, and 50% of the college students.

Although participants showed a slightly better understanding of Madonna’s song, age differences still existed. Only 10% of the fourth graders correctly understood Madonna’s feelings in the song, whereas 50% of the eighth graders, 60% of the twelfth graders, and 80% of the college students did. Overall, younger children were more literal and concrete in their interpretations of the songs, whereas high school and college students were more abstract and metaphorical.

Age differences have also been found in the comprehension of music videos. Christenson (1992b) exposed 54 fourth-through sixth graders to Billy Ocean’s music video “Get Outta My Dreams, Get Into My Car” and then asked the children to describe what the story was about. The video features Billy Ocean pulling into a carwash and asking the female attendant to get into his car. The couple then goes through the carwash and later to a drive-in movie. Roughly half of the children knew the video was about a romantic relationship, but the remaining half did not. Next the children were presented with three forced-choice alternatives about the meaning of the song: an abstract one that focused on the budding relationship between the characters (“a guy who likes a girl and wants to know her better”) and two that were more concrete (“a guy who wants to take a girl for a ride in his car,” “a guy who is getting his car washed”). Sixth graders were more likely to pick the abstract, more correct interpretation of the song (88%) than were fourth graders (51%).

We can expect children’s comprehension of music to improve with development, in part because of the preponderance of symbolism and figurative language in popular songs. Other media-related research supports this idea. In one study, second, fourth, and sixth graders were exposed to print ads that featured either a literal or a metaphorical message about the same product (Pawlowski, Badzinski, & Mitchell, 1998). Sixth graders were better able to interpret the metaphors than were the younger two age groups. In another study, children between the ages of 12 and 16 were exposed to a series of television clips that contained sexual innuendos about intercourse, rape, and prostitution (Silverman-Watkins & Sprafkin, 1983). Children over the age of 13 generally understood the televised innuendos better than did the 12-year-olds.

As it turns out, the age differences in comprehension of the media are quite consistent with research on children’s understanding of symbolic language such as metaphors (Wilson, 2001). In general, children aged 6 and under have difficulty interpreting even common metaphors (Johnson & Pascual-Leone, 1989). By age 7 or 8, children can understand those metaphors that bare strong physical resemblance to their referents. Not until 11 or 12 can children reliably interpret most types of metaphors. These patterns are congruent with broad shifts in cognitive development in which children move from concrete thought processes closely tied to how things appear (Flavell, Miller, & Miller, 2002) to more abstract and flexible thinking during the adolescent years (Byrnes, 2003).

Age is not the only factor that can affect comprehension. Being a committed fan seems to encourage a differential response to music (Kuwahara, 1992), as suggested above. In one study, Wass et al. (1989) surveyed 694 middle and high school students about their music preferences and favorite songs. Nearly 20% of the students qualified as heavy metal fans and most were White males. The researchers found that 41% of the heavy metal fans knew all the lyrics to their favorite songs as compared to only 25% of students who liked other types of music. Unfortunately, the researchers did not test actual lyric comprehension, but it stands to reason that those who pay more attention to the words will show increased comprehension of them. Fans will have more experience with a genre’s themes and conventions, and this experience should result in the development of “scripts” or routinized mental expectations for the sequence of actions that constitute a typical story or event (Schank & Abelson, 1977). According to script theory, songs that are scripted should be easier to understand and memorize than songs that are novel or unfamiliar (Desmond, 1987).

To test this idea indirectly, Hansen and Hansen (1991) conducted an experiment in which undergraduates were either given a written copy of the lyrics (low cognitive load) or not (high cognitive load) while they listened to four heavy metal songs. Providing the lyrics was designed to reduce the mental effort or load required to make sense of the songs. The researchers reasoned that their low cognitive load condition is parallel to an avid fan’s experience in that repeated exposure to a song should also reduce the mental resources needed to process its meaning. Across three experiments, those in the low cognitive load condition showed better recall, song comprehension, and extraction of detailed content than did those in the high cognitive load condition. However, even when the lyrics were provided, most participants showed far from perfect comprehension of the songs.

Several conclusions can be drawn from this research. First, comprehension of music is particularly challenging because of the symbolic language and complex metaphors involved. Second, comprehension varies greatly with age; the more obscure the lyrics are, the more likely that a child under the age of 11 or 12 will not understand their meaning. Some have argued that this lack of comprehension might in fact protect younger children from possible harmful effects of explicit or violent songs (Greenfield et al., 1987).

Third, fans of musical genres are more likely to pay attention to and understand song lyrics than are casual listeners. With regard to violence, we should be most concerned, then, about fans of rap and heavy metal music. Notably, these are the very genres that place a strong emphasis on the words as central to the music. Finally, it is not at all clear that a “correct” or absolute understanding even exists with regard to music, given the multiple meanings inherent in this artistic form (Hansen & Hansen, 1991). Furthermore, we do not have any evidence to suggest that a fuller understanding leads to more learning or imitation. It may be enough to simply understand the themes of violence in order to learn from such content. We now turn to the research on effects.

Effects of Violent Music on Aggressive Attitudes and Behaviors

Most research on the impact of violent music has focused on whether such content encourages aggressive attitudes and behaviors in listeners. There are literally hundreds of studies showing that exposure to violent television programming increases aggression (see Bushman & Anderson, 2001). By contrast, there are far fewer studies of the effects of violent music and music videos. However, several prominent theories suggest that music would have the same effects as television programs do. Social learning theory posits that children can learn new behaviors by observing others in their social environment (Bandura, 1977, 1994). According to this theory, children are more likely to imitate models that are rewarded than those that are punished. Children are also more likely to imitate a model that is perceived as attractive or similar to the self (Bandura, 1986). Therefore, highly popular singers and bands can serve as potent role models for young people. When Arnett (1991a) asked a group of high school students to list people they admired, 60% of heavy metal fans named a favorite musician. Indeed, a casual observer can find numerous examples of preteen and teen fans trying to emulate the gothic style of punk rocker Marilyn Manson or the pop style of Britney Spears.

Social learning theory is a useful framework for understanding how children can learn new attitudes and behaviors from role models in the media. Once such behaviors are acquired, cognitive priming theory explains how the media can prompt their enactment in certain situations. According to priming theory (Berkowitz, 1990; Jo & Berkowitz, 1994), violent stimuli in the media can activate aggressive thoughts, feelings, and even motor tendencies stored in a person’s memory. For a short time after exposure, then, a person is in a “primed” state and several conditions can encourage these thoughts and feelings to unfold into aggressive behavior. One such condition is the person’s own emotional state. Individuals who are angry or hostile are more likely to be primed to act aggressively by the media because they are in a state of readiness to respond (Berkowitz, 1990). Another condition that encourages priming is justified violence. If media violence is portrayed as morally proper, it can help to temporarily reduce a person’s inhibitions against behaving in an antisocial way (Jo & Berkowitz, 1994). Finally, reminders in the immediate environment of violence that a person has just witnessed can trigger aggressive behavior (Jo & Berkowitz, 1994).

Both social learning and priming explain the short-term effects of exposure to media violence. Huesmann (1998) has developed an information-processing model that helps to account for the long-term effects of media violence. The model focuses on the learning and reinforcement of scripts, or mental routines stored in memory. A script typically includes information about what events are likely to happen, how a person should respond, and what the likely outcome of these responses will be. According to Huesmann, a child who has been exposed to a great deal of violence, either in real life or through the media, is likely to develop scripts that encourage aggression as a way of dealing with problems. Consistent and repeated exposure to violent messages helps to reinforce these scripts and make them easy to retrieve in memory. In other words, a preteen or teen who listens to a great deal of gangsta rap should have a stable and enduring set of cognitive scripts that accentuate aggression, particularly aggression against women, as an appropriate response to social situations. How well does the research on violent music support these theories? We turn now to the evidence.

Correlational Studies

Several studies have found a relationship between preference for violent music and aggressive behavior. In one study, Rubin et al. (2001) surveyed 243 college students about their listening habits, emotions, aggressive attitudes, and attitudes toward women. Rap and heavy metal listeners exhibited significantly more aggressive attitudes than did fans of classic rock, rhythm and blues, country, and alternative rock. Heavy metal listeners also expressed more negative attitudes toward women and rap listeners exhibited significantly more distrust of other people than did listeners of the other genres.

Took and Weiss (1994) found that adolescents who like rap and heavy metal music were more likely to have been suspended or expelled from school for behavior problems and to have arrest records. Other studies have found that heavy metal fans in particular are more likely than other teens to engage in delinquent behaviors (Arnett, 1991b; Martin, Clarke, & Pearce, 1993) and to experience conflict in family relationships (Christenson & van Nouhuys, 1995, as reported in Christenson & Roberts, 1998). The findings are consonant with social learning, priming, and Huesmann’s model in that exposure to the most violent music is associated with increased aggression. The data, however, are correlational and therefore impossible to untangle in terms of causality. Do violent lyrics cause adolescent aggression or do troubled, violent youth seek out such content? The only way to resolve this issue is to use more controlled studies.

Experiments Assessing Violent Lyrics

Several experiments have assessed the impact of listening to violent music. One important caveat here is that the vast majority of these laboratory studies have used college-aged students rather than children and teens. In such cases, the best we can do is extrapolate from what is known about older teens and young adults to what is likely to occur with younger listeners.

Wannamaker and Reznikoff (1989) assessed feelings of hostility after exposure to a single heavy metal song that had either aggressive lyrics and aggressive music (i.e., “hard, driving beat”) or just aggressive music but nonaggressive lyrics. Compared to a group who heard a completely nonaggressive song, no differences in hostility were found in the two heavy metal conditions, although the researchers acknowledged that the vast majority of participants did not understand the theme in either of the two heavy metal songs used.

In another experiment, male college students listened to four rap songs that were either misogynous (sexually violent) or neutral (no sex or violence) in nature (Barongan & Hall, 1995). Participants then viewed a violent, a sexually violent, or a neutral vignette from the movie / Spit on Your Grave and chose one of the vignettes for a female confederate to watch. Males who had heard the misogynous rap music were significantly more likely to select the violent vignette for the female confederate than were males who had heard the neutral rap. The researchers argued that selecting a violent movie clip to show another person is a proxy for aggressive behavior, although others have questioned the validity of such a measure (Anderson, Carnagey, & Eubanks, 2003).

In the most extensive research to date, Anderson and colleagues reported five experiments on the impact of violent lyrics on aggression (Anderson et al., 2003). In the first experiment, college students listened to a violent song or a nonviolent song, both performed by the same band, and then rated their feelings on a standard hostility scale (e.g., “I feel furious,” “I feel like yelling at somebody”). Results revealed that the violent song produced higher levels of hostility than did the nonviolent song. Experiment 2 was identical except that it assessed aggressive cognitions rather than feelings. After listening, participants were presented with a series of ambiguous words and asked to rate how similar they were to other words. As predicted, the violent song led participants to interpret ambiguous words such as “rock” and “stick” in an aggressive way. The researchers argued that the violent song made aggressive thoughts more accessible, as priming theory suggests.

In Experiment 3, the researchers added a no-song control condition and increased the number of songs tested in the violent and nonviolent conditions. Again, participants who heard a violent song reported a higher level of hostility, but only when measured immediately after exposure, suggesting that the impact of music on aggressive feelings may be fleeting. In addition, those who heard a violent song reacted more quickly when reading aggressive words than did those in the nonviolent and control conditions. In Experiment 4, participants listened to a violent humorous song, a nonviolent humorous song, or no song. As predicted, the violent humorous group showed the same level of hostility as the control group did, supporting the idea that humor can cancel out the impact of violent lyrics. Nevertheless, those who listened to the violent humorous song were still more likely than those who heard the nonviolent song to create aggressive words out of word fragments (e.g., make “h_t” into “hit” rather than “hat”).

Experiment 5 tested song violence and humor independently in a 2 (violent vs. nonviolent lyrics) X 2 (humorous vs. nonhumorous lyrics) design. Irrespective of humor, those who heard the violent song had higher hostility scores and produced more aggressive word completions afterward than did those who heard a nonviolent song. To summarize, songs with violent lyrics increased feelings of hostility in four of the five experiments, and this effect occurred across the range of humorous and nonhumorous songs. In addition, violent songs led to more aggressive cognitions in four of the experiments.

Two studies have assessed whether violent songs enhance aggression against women in particular. St. Lawrence and Joyner (1991) exposed 75 male undergraduates to 17 minutes of sexually violent heavy metal rock, Christian heavy metal rock, or classical music and then measured attitudes toward women, acceptance of violence against women, and self-reported arousal. Males in both heavy metal song conditions were significantly more likely to endorse sex-role stereotypes than were males who had listened to classical music. Furthermore, both the sexually violent and the Christian heavy metal music increased acceptance of interpersonal violence against women and of rape myths (e.g., “Women who get raped hitchhiking get what they deserve”), although the increase was statistically significant only for the Christian heavy metal listeners.

Wester, Crown, Quatman, and Heesacker (1997) looked at gangsta rap and its impact on attitudes toward women. Male participants with little exposure to gangsta rap were randomly assigned to one of four conditions: listened to five gangsta rap songs (music only), listened to the same songs while reading the lyrics (music plus lyrics), read the lyrics but did not hear the songs (lyrics only), or neither heard the songs nor read the lyrics (control). Men exposed to the gangsta lyrics, regardless of whether music accompanied them, expressed more adversarial views about sexual relations with women than did those who were not exposed to the lyrics. Music alone, the researchers concluded, might not suffice to affect naïve listeners who have no prior scripts in memory for such content. Instead, consistent with social learning, more explicit exposure to a song’s words may be necessary to foster the creation of new scripts.

Experiments Assessing Violent Music Videos

Several scholars have argued that music videos are likely to have more impact than songs alone (Strasburger & Hendren, 1995). Adding a visual portrayal of violence to reinforce violent words certainly enhances the number of aggressive cues in the message. Consistent with this idea, college students evaluate music videos more favorably and rate them to be more “potent” than audio-only versions of the same music (Rubin et al., 1986). One of the main reasons young people like music videos is because the visual portrayals help in interpreting the song’s meaning (Sun & Lull, 1986). Thus, a video may make complex or ambiguous lyrics more accessible and meaningful, especially to a younger child.

In one of the earliest experiments on music videos, Greeson and Williams (1986) had seventh and tenth graders watch either a randomly compiled group of videos from MTV or a pre-selected group of high-impact videos containing sex, violence, and anti-establishment overtones. The high-impact videos increased tenth graders’ self-reported acceptance of the use of interpersonal violence. The same videos unexpectedly tended to decrease seventh graders’ acceptance of violence, although this difference was not statistically significant. Unfortunately, because the high-impact videos contained numerous themes and because some low-impact videos actually contained violence, this study does not provide a clean test of the effects of violent music videos.

In a more tightly controlled experiment, Hansen and Hansen (1990) had 56 undergraduates watch three rock videos portraying antisocial behavior (e.g., destroying a home during a wild party) or three neutral rock videos containing no antisocial acts. Afterward, participants watched a videotape of a confederate who was ostensibly on a job interview and who made either an obscene hand gesture toward the experimenter or no such gesture during the interview. Overall, those who had seen the neutral videos rated the confederate as less likeable, more threatening, and more irrational when he made the gesture than when he did not. In contrast, those who had seen the antisocial videos rated the confederate the same regardless of the obscene gesture, suggesting an increased tolerance for antisocial behavior.

Johnson, Jackson, and Gatto (1995) conducted an experiment on the impact of rap music videos in particular. African American males between the ages of 11 and 16 were exposed to eight violent rap videos, eight nonviolent rap videos, or no music videos. Next, the teens read two vignettes ostensibly as part of another study, one of which involved a jealous man who assaulted his girlfriend as well as another man who vied for her attention. Participants were then asked about their attitudes regarding the use of violence in the vignette and about the likelihood that they themselves would respond similarly. Teens in the violent video group expressed greater acceptance of the use of violence against the man than did teens in the nonviolent and control groups. Compared to the control group, participants in the violent video group also reported a greater acceptance of the use of violence against the girlfriend and a greater likelihood of personally engaging in similar violence. In a subsequent study, Johnson and colleagues found that even nonviolent rap videos could increase African American adolescents’ acceptance of teen dating violence, but only among girls and not among boys who were already high in acceptance (Johnson, Adams, Ashburn, & Reed, 1995). The researchers argued that nonviolent videos still contain images of women as sexually subordinate, which could lead to increased tolerance for violence against females.

In a study conducted outside the laboratory, researchers assessed the level of violence in a maximum-security forensic hospital before and after the removal of MTV (Waite, Hillbrand, & Foster, 1992). The study was prompted by clinical staff observations that MTV was continuously on in at least one of two television rooms and that patients often experienced behavioral problems after prolonged exposure to such content. The hospital staff measured verbal and physical aggression of 222 patients over a 55-week period, 33 weeks before and 22 weeks after MTV was banned from the viewing options. Following MTV’s removal, aggression dropped from an average of 44 incidents per week to 28 incidents per week, with significant decreases in verbal as well as physical aggression. Although the study lacks a true control group, it is the only evidence we are aware of that links exposure to violent music with changes in aggressive behavior observed in a naturalistic environment.

To summarize, a modest amount of evidence links exposure to violent music with aggression. Even brief periods of listening to or watching violent material can increase aggression. Certain violent music, particularly that which is misogynistic in nature, can also increase aggression toward women. Most research has focused on rap and heavy metal music, the two most violent genres. All of the studies have looked at short-term effects only, predominantly with college students.

Future research can take several directions. First, we simply need more studies. Compared to the database on violent television programming, the literature on violent music is scant and unsystematic. Second, researchers need to pay more attention to the music and the aggression measures they select so that stronger conclusions can be drawn across studies. If violent rap music is tested, for example, the experimenter needs to specify whether the lyrics are explicit or vague, whether the songs employed also contain sexual content, and what type of violence the song endorses. As mentioned previously, justified violence, rewarded violence, and attractive perpetrators all enhance the impact of television violence on aggression, warranting their assessment in music as well. Third, more studies should assess actual aggressive behavior as an outcome variable rather than self-reports of hostile feelings and attitudes. Fourth, we need to study younger age groups, especially preteen and teens, who spend a great deal of time with music and are still developing scripts for social problem solving and for dating relationships. We also need to focus specifically on teens that are ardent fans of violent music. Finally, we need to conduct longitudinal research to ascertain the long-term impact of exposure to violent music. Tracking youth over time will help determine whether rebellious teens are simply attracted to violent music or whether such music actually contributes to aggression.

Other Antisocial Effects of Violent Music

Although the vast majority of studies have concentrated on the impact of violent music on aggression, there may be other negative outcomes of listening to violent songs. We review three here: increased depression, risk taking, and racial stereotyping.

Violent Music and Depression/Suicide

Several studies have explored whether violent music is linked to mental health problems in youth. In an experiment, Ballard and Coates (1995) had college students listen to one of six songs that varied by genre (rap vs. heavy metal) and by content (homicidal, suicidal, or nonviolent). Rap music generally made students feel angrier than heavy metal did, but none of the songs increased depression or suicidal thoughts. This is not surprising given that changes in mental health are not likely to occur after short-term exposure to a single song.

Correlational studies suggest more of a link. One survey of 247 Australian teens found that a preference for rock/heavy metal music was positively correlated with suicidal thoughts, acts of deliberate self-harm, and depression, particularly among females (Martin, Clarke, & Pearce, 1993). For example, 62% of female teens who liked rock/heavy metal reported that they had deliberately tried to harm themselves within the last 6 months, compared to only 14% of females who preferred pop music. Unfortunately, the study combined those who listened to rock with those who listened to heavy metal, so it is difficult to isolate the effect of violent music.

Scheel and Westefeld (1999) surveyed 121 U.S. high schoolers and found that 40% liked or strongly liked heavy metal music. They also found a positive correlation between heavy metal fanship and suicidal ideation or thinking about killing oneself. The relationship was especially prominent among females: 74% of female heavy metal fans reported that they had occasionally or seriously thought about killing themselves, whereas only 35% of nonfans did. Heavy metal listening was also associated with significantly lower scores on the Reasons for Living Inventory (RFL), which taps reasons to live and reasons not to commit suicide. In contrast, being a rap or rock fan was associated with higher scores on the RFL. Like most teens in the study, heavy metal fans reported that listening to their favorite music put them in a positive rather than negative mood (only 9% said the music made them feel angrier). This self-reported mood elevation has led some scholars to argue that the pessimistic lyrics of heavy metal may actually be therapeutic for teens coping with stress (Arnett, 1991a). Disentangling whether heavy metal music attracts or creates troubled youth is an important challenge for future research.

Violent Music and Reckless Behavior

Several studies have found a link between violent music and risk taking. Arnett (1991b) surveyed tenth and twelfth graders about a variety of reckless activities they might have engaged in over the past year. For males, liking heavy metal music was positively associated with drunken driving and with marijuana use. These relationships held even after controlling for individual differences in sensation seeking and in satisfaction with family relationships. For females, liking heavy metal was positively associated with having sex without contraception, marijuana use, and shoplifting.

Looking at a sample of teens treated for problems at clinics and psychiatric units, Took and Weiss (1994) found that those who liked heavy metal and rap music were more likely to have used illegal drugs and alcohol and to have engaged in sexual activity than were those who did not like such music. Many of these teens had a history of problems dating back to elementary school, leading the researchers to speculate that troubled teens may gravitate to violent music rather than that violent music causes such turmoil.

To resolve the directionality issue, two longitudinal studies have been conducted. Robinson, Chen, and Killen (1998) surveyed 1,533 ninth graders at a baseline point and again 18 months later about their alcohol use. Results showed that after controlling for age, sex, ethnicity, and other media use, music video viewing was significantly associated with the onset of drinking. In fact, each increase of 1 hour per day of music video exposure was associated with a 31% increased risk of starting to drink during the next 18 months. Unfortunately, the study did not assess the types of videos viewed so it is impossible to know whether violence, or some other content feature such as drinking portrayals, contributed to this effect. The study also did not control for parental influence.

Wingood, DiClemente, and Bernhardt (2003) surveyed 522 African American girls between the ages of 14 and 18 at two points in time concerning their exposure to rap music videos and their sexual activity, drug use, and aggressive behavior. After controlling for parental monitoring as well as teen employment, age, involvement in extracurricular activities, and participation in religious events, early exposure to rap music videos was found to be a significant predictor of subsequent risk-taking behaviors 12 months after exposure. In particular, those who reported greater exposure to rap at baseline were twice as likely to have had multiple sex partners, more than 1.5 times as likely to have contracted a new sexually transmitted disease, and more than 1.5 times as likely to have used drugs and alcohol by the follow-up.

Longitudinal studies begin to address whether at-risk teens are merely attracted to violent music or whether such music contributes to teen recklessness. In the end, it may be a cyclical process whereby violent music and risky behavior are mutually reinforcing over time. At a minimum, we can conclude that a strong preference for violent music is one of many factors that can signal adolescent turbulence.

Violent Music and Racial Stereotyping

Fictional television shows and even television news often portray Blacks as violent and as engaged in disproportionate amounts of criminal behavior (Dixon & Linz, 2000; Oliver, 1994). Rap music has also been criticized for stereotyping (Morano, 2004).

Only two published studies exist, to our knowledge, on the impact of rap music on racial stereotyping. In one experiment (Johnson, Trawalter, & Dovido, 2000), 180 Caucasian and African American undergraduates were randomly assigned to listen to 4 minutes of violent rap music, 4 minutes of nonviolent rap, or no music. Students then participated in an ostensibly unrelated study in which they read three stories, one about a young man who threatened his fiancée with violence, one about a job applicant interested in a management position that required “demonstrated intelligence,” and one about an ROTC graduate seeking a position in a helicopter training school that required “excellent spatial skills.” Participants were either told that the man in each story was Black or that he was White. When the target in the story was Black, participants exposed to the violent rap were significantly more likely to make negative attributions about the Black man’s violent personality than were those exposed to either nonviolent rap or no rap. When the target was White, rap music had no impact on attributions about the violence. This stereotyping effect occurred regardless of the participants’ race. Similarly, compared to the nonviolent rap and the no-music conditions, exposure to violent rap music led participants to rate the Black man as less qualified for a job requiring intelligence. Yet violent rap music had no impact on judgments regarding the Black ROTC applicant, consistent with the idea that spatial skills are not part of people’s stereotypes about race.

In another experiment, Rudman and Lee (2002) exposed mostly Caucasian undergraduates to either six violent and misogynistic rap songs or six pop/rap songs with no violent or sexist content. Students who listened to the violent and misogynistic rap rated a Black target that behaved ambiguously as more violent, more sexist, and less intelligent than controls did. Violent rap music did not affect judgments when the ambiguous target was White. These findings are consistent with research indicating that other types of violent media content can also prime racial stereotypes (see Greenberg, Mastro, & Brand, 2002).

Policy Implications and Conclusions Regarding Violent Music

In the last two decades, violent music has provoked spirited debate between members of the music industry, parents, and policy makers. As early as 1985, the wives of several members of the United States Congress formed the Parents Music Resource Center (PMRC). Spearheaded by Tipper Gore, the PMRC’s mission was to educate parents about the “alarming trends” in popular music, particularly the increasingly violent and sexual lyrics as well as the graphic album covers (PMRC, 2004). As a way to combat this issue, the PMRC asked the Recording Industry Association of America (RIAA) to provide parental advisory labels on recordings that contained explicit content. Contemporaneously, the U.S. Senate Commerce, Technology, and Transportation Committee, responding to pressure from the PMRC, began an investigation into the content of rock music. Several famous rock musicians were called to testify before the committee, including Frank Zappa who called Tipper Gore a “cultural terrorist.” Before the hearing ended, the RIAA agreed to label explicit audio recordings. Today, the labels are applied by individual record companies and artists, and are in the form of a black and white logo with the words “parental advisory, explicit content” placed on the packaging. As of 2000, one third of the 100 top-selling CDs contained a parental advisory label (FTC, 2000).

Some retail chains like Wal-Mart have refused to stock any music containing such advisories (“Wal-Mart Cleans Up,” 1997). Record companies have responded by altering certain songs and creating “clean” copies to meet retailing standards. Notably, these edited versions often outsell the original releases (FTC, 2000). Despite the efforts by the RIAA and some retail chains to help parents, the advisory labels may make little difference as more and more music becomes available on the Internet. In one recent poll, 70% of preteens and teens (ages 12–18) reported that they are more likely to download music off the Web than go to a record store (Holland, 2002).

The effectiveness of the labels may also be hampered by the parents themselves. Although 75% of parents report being concerned about the music their children listen to (Woodard, 2000), many do not seem to use the advisory labels. In one national survey, 84% of parents reported that they had used the MPAA film rating system to make media choices for their children, but only 50% had used the parental advisory labels on music (Kaiser Family Foundation, 2001).

Another problem with advisory labels is that they may actually attract young consumers to the very content that parents want their children to avoid. Bushman and Cantor (2003) tested this “forbidden fruit” idea in a meta-analysis of 18 studies conducted on the impact of ratings and advisories on attraction to different types of media content. They found that ratings and advisories generally increased rather than decreased attraction to media, particularly among males. But the effects varied with age. Ratings served as a repellent for children under age 8, but as an enticement for older children and adolescents.

Other approaches for dealing with violent music are more direct. Some schools have forbidden young people to dress in any style that represents punk or heavy metal music (Garza, 2004), and some communities have devised workshops for parents to “depunk” or “demetal” their children (Rosenbaum & Prinsky, 1991). The American Academy of Pediatrics (1996) issued a statement that, “It is in children’s best interest to listen to lyrics that are not violent, sexist, drug-oriented, or antisocial,” (p.1219) and it recommends that pediatricians encourage parents to monitor the music that their children listen to and purchase. And in one national poll, over 70% of parents supported the outright prohibition of the sale of gangsta rap music to children under the age of 18 (Joint Center for Political and Economic Studies, 1997).

The music industry and avid fans have charged that any efforts at music censorship are a violation of the First Amendment (RIAA, 2003b). Yet governmental pressure continues. In 2000, the Federal Trade Commission issued a report stating that the marketing of explicit-labeled music to children under 17 is pervasive. The report encourages the music industry and independent retailers to require parental permission before selling labeled music to minors. At the same time, rap music is becoming increasingly popular and, some argue, more explicit. In addition, heavy metal music attracts a smaller but just as avid group of fans who are mostly young White males.

The research to date indicates that exposure to violent music can lead to short-term increases in aggressive attitudes and cognitions, and that some forms of rap music can prime racial stereotypes. We need far more evidence about the long-term effects of violent lyrics on aggressive behavior, risk taking, depression, and even desensitization to violence. As with all forms of violent media, we should be most concerned about those young people who are frustrated, isolated from social support, surrounded by aggressive cues in their environment, and strongly attracted to musicians who glorify violence in their songs.