Prosecuting rap: what does the case law tell us?

Abstract This article explores the admissibility and use of rap music as evidence in English criminal trials. It presents findings from an analysis of over 30 appeal cases. As well as unpacking the link between rap, race and gangs that is prevalent in these cases, the article challenges the categorisation of rap as ‘bad character evidence’, and critiques the way in which questions of relevance and prejudicial effect have been addressed by the courts. In particular, when making admissibility decisions, the courts appear to give little consideration to the cultural context, artistic conventions or social influences within the rap music genre, or the racialised nature of rap evidence. It is argued that, if rap is to be admissible evidence, a much more rigorous and informed approach is required.


Introduction
The use of rap music as evidence in criminal trials has received less scrutiny in England and Wales than in the USA, where there is a longer history of putting 'rap on trial', and where serious concerns have been raised about the prejudicial and discriminatory nature of this practice (Dennis 2007;Kubrin and Nielson 2014;Lutes et al. 2019;Nielson and Dennis 2019;Lerner and Kubrin 2021). There is, however, a growing body of literature from this side of the Atlantic which surveys some of the causes and consequences of 'prosecuting rap', often from a sociological, criminological and cultural studies perspective (Quinn 2018;Fatsis 2019a, b;Ilan 2020;Fatsis 2021;Keenan and Paul 2021). Informed by these perspectives, this article seeks to expand knowledge of the way in which rap is prosecuted in England and Wales, by focusing on what happens in the courtroom. It presents findings from an exploratory study of reported appeal cases in which the creation of lyrics and participation in music videos were used as evidence at trial or treated as an aggravating factor at sentencing. Through an analysis of appeal cases, the article reveals and critiques the profile of 'rap cases' and scrutinises the way in which the law of evidence has been applied to rap music. Before embarking on this task, it is necessary to briefly frame the discussion within the broader context.

Background
Rap is a form of Black expressive youth culture, originating from the Bronx, New York, in the 1970s. It evolved from party music to include social and political commentary, and spread across the globe to become one of the most popular (and profitable) genres of music. In the UK, distinct subgenres have emerged, influenced not only by American rap and UK dance music, but also by music and culture directly from the Caribbean and West Africa (Bramwell 2015;White 2016;Hancox 2018a). The most recent subgenre to gain significant attention and enter the mainstream is 'drill'. Thapar explains that UK drill: [I]s characterised by a thumping, swinging bassline that sounds like the engine of a doubledecker bus; it is stitched together by tinny, marching-band snare drums and haunting piano or synth melodies. Since it took off in London, it has veered towards the fast-paced energy of grime, its wise old musical uncle, and away from its roots in Chicago drill and US trap. Unforgiving lyricism about the extremes of disfranchised, hypermasculine adolescent lifenihilistic references to knife violence (often in the form of provocation and bleak, detailed descriptions of drug dealing)fills nearly every song. (Thapar 2019) Despite the violent (and often misogynistic) nature of some rap music, including drill, participation in rap can be hugely beneficial. There is, of course, the prospect of commercial success and financial reward which influences much of the content of mainstream rap music, including violent themes (Dennis 2007;Charnas 2010;Stuart 2020). Rap music can also serve as an outlet for creativity and self-expression, facilitate identity development, support emotional intelligence and build self-esteem. Bramwell explains that rap can help young people develop a sense of themselves, including an awareness of their place in the world and the relations with others through which their life experiences are structured (2015, p. 4;White 2020). Through social and political commentary, rap enables individuals from marginalised and neglected sections of society to express dissatisfaction, resist oppression and seek change. Rap can also be used as a means of developing linguistic skills and as an educative tool, making traditional academic subjects more appealing and accessible (Virk 2020). The financial, social and personal gains derived from rap music can make it an attractive alternative to participation in the criminal exploits that some rap about.
Yet rap has long been treated with suspicion by the authorities, culminating in a history of attempts to suppress rap culture. In the USA, this dates back to the emergence of hip-hop and includes FBI investigations into rap groups such as N.W.A., as well as arrests of artists for lewd or profane performances (Nielson and Dennis 2019, ch. 1). The UK also has 'a legacy of criminalization' of Black cultures and music forms (Ilan 2020, p. 997;Fatsis 2019aFatsis , 2021. A recent example is the Metropolitan Police's live music risk-assessment form, Form 696, which was used as a means of racial profiling, to shut down Black music nights in the 2000s and 2010s (Bramwell 2015, p. 65;Hancox 2018a, ch. 8;Fatsis 2019bFatsis , pp. 1306. The British media, public figures and politicians have been persistent in their efforts to link Black music and youth cultures to crime (Hancox 2018a, ch. 6). For example, in 2003, the Labour Culture Minister, Kim Howells, professed that for years he had 'been very worried about these hateful lyrics that these boasting macho idiot rappers come out with', describing it as 'a big cultural problem' (Gibbons 2003). In 2006, former Prime Minister David Cameron accused BBC Radio 1 of playing (rap) music on Saturday nights that 'encourages people to carry guns and knives' (Morris 2011). Having gained prominence since the mid-2010s, drill is the most recent target of the media and the authorities. The first drill single to reach number one in the UK Official Charts was described in a Daily Mail headline as the 'soundtrack to murder' (Boyle 2021). In the absence of empirical evidence demonstrating a causal link, public opinion is divided as to whether drill is a cause of, or reaction to, violent crime (Lynes et al. 2021;Savage 2020). The authorities, on the other hand, have been quick to 'clamp down' on the genre (Fatsis 2018). Drill music videos have been removed from YouTube at the request of the Metropolitan Police (LBC 2019;Pritchard 2022), and, perhaps most controversially, gang injunctions and Criminal Behaviour Orders have been imposed, restricting, for example, participation in music videos, as well as what artists can rap about (Papamichael 2018;Clowes 2021). While it is not suggested that there has never been any link between violent lyrics and particular incidents of violence (Hancox 2018b), the nature and implications of such links are incredibly difficult to identify and unpack, and the criminal justice response to the perceived threat of rap has been disproportionate, damaging and discriminatory.
The British establishment's relationship with Black youth cultures is no doubt informed by, and contributes to, the criminalisation of Black people, evidenced by the overrepresentation of Black people throughout the criminal justice process (Ministry of Justice 2017; Lammy 2017; Joint Committee on Human Rights 2020). To cite just one recent statistic, between July and September 2020, 33% of children held in custody on remand in England and Wales were Black, compared with approximately 3% of the general population (Grierson 2020). Writing in the context of the USA, Nielson and Dennis see 'rap on trial' as 'both a window into the broader racial inequalities that play out in our criminal legal system and a causality of those inequalities' (2019, p. 25). As we will see, in England and Wales, Black people are most at risk of having their involvement in music used against them in court. Without question, then, race is central to the topic of prosecuting rap (Nielson and Dennis 2019, p. 21), and the use of rap music as evidence in court fits into a wider pattern of marginalising and criminalising Black youth and Black cultures (Ilan 2020;Fatsis 2021).

Methodology
The appeal cases relied on for this study were accessed through legal databases, namely Westlaw, using the following search terms in the 'free text' box: 'lyric'; 'rap'; 'rapper'; 'music video'; 'drill music'; 'drill video'; 'grime music'; 'grime video'. A number of other search terms which include the word 'lyric' and 'rap' were also used. The searches generated 107 cases from England and Wales that were categorised as 'crime' cases on the database, and reported up to January 2021. In 26 of these cases, lyrics or participation in music videos had been used as evidence against a defendant or treated as an aggravating factor at sentencing. Twenty-three of these cases concerned rap music. A further seven cases on Westlaw were identified as being relevant to this analysis, as they demonstrate a perceived link between rap and criminality or character. For example, rap has been used or attempted to be used as evidence of a victim's character, has led to a police investigation into possession of firearms and has formed part of a gang injunction. These 33 'Westlaw cases' were supplemented by a further four cases accessed through LexisNexis in which music was used as evidence against a defendant at trial or sentencing (three of which concerned rap music), and one such case that was not accessed through a legal database, bringing the total number of cases to 38. The cases are listed in the Annex.
Case searches on legal databases do not uncover every appeal case, and there are limits to what we can learn from appeal judgments. In particular, they tell us little about how often prosecutors seek to rely on rap music, or whether trial judges tend to admit or exclude rap evidence. Also, many judgments lack details about the exact nature or context of lyrics and videos. In many cases, this could be because admission of the evidence was not challenged on appeal. In fact, the admission, interpretation or use of lyrics or videos against a defendant was a subject of appeal in a minority of cases, and was usually not the sole ground of appeal. Notably, in only one case (R v Alimi [2014] EWCA Crim 2412) was the admission of lyrics or videos against a defendant successfully challenged.
The above limitations notwithstanding, where admission of music lyrics or videos is challenged on appeal, one would expect to see the greatest level of scrutiny of the issues that arise, including the legal basis for admission of the evidence, as well as questions of fairness and prejudicial effect. Where admission of the evidence is not challenged, the cases still provide insight into how rap is used in court. Accordingly, an analysis of over 30 appeal judgments can tell us much about the handling of rap music as evidence, as well as the kinds of cases in which this occurs.
What follows is an overview and qualitative analysis of key themes that emerge from the case law, namely: the 'nature of cases and profile of defendants'; the relationship between 'rap, race and gangs'; the 'relevance' of rap; the categorisation of rap as 'bad character evidence'; and 'prejudicial effect'. The cases show that, for the most part, a problematic and uninformed approach has been taken towards the use of rap as evidence. In particular, prosecutors have been permitted to rely on stereotypical narratives to construct case theories, inviting jurors to take rap literally in a way that is not done with other genres.

The nature of cases and profile of defendants
The cases relied on for this analysis were reported between March 2005 and January 2021. Notably, the most recent 16 cases were reported between 2018 and 2021. This could indicate an upward trend in the use of rap in criminal proceedings and challenges to the use of rap. It could also be a reflection of the media and criminal justice response to the rise of drill music.
Twenty-seven of the 38 cases concerned offences that were committed and tried in London. In a further four cases, the location of the offence was unspecified, although two were tried in London. Outside of London, four cases were from Birmingham, one from Nottingham, one from Maidenhead and one from Aberystwyth.
Thirty-two of the cases concerned offences involving weapons (mostly firearms) and/or violence, including nine murders. That most cases involved violence or firearms is unsurprising, given that this is common subject matter within some rap subgenres. While some people rap about their lived experiences or things they have observed, sometimes as a form of social commentary, much of rap is fictional entertainment and intentionally provocative. Rap is characterised by complex wordplay. It relies on symbolism, figurative language, hyperbole, braggadocio and humour. Violent lyrics are often metaphorical. Stoia et al. explain that, 'boasts about one's strength coupled with metaphorical threats against one's enemies have been a defining feature of rap lyrics since the genre's inception ' (2018, p. 352). Violent content can also be used to vent frustration or gain attention and, in turn, commercial success, as violence sells (Dennis 2007). Drill, in particular, is characterised by themes of violence and criminality. It draws from the earlier gangsta rap genre, in which rappers adopt a 'badman' persona and typically speak in the first person, often about criminal exploits involving violence, drug dealing and gangs (Quinn 2005). While references to criminality are often fictional, success may require the appearance of authenticity (Ilan 2020(Ilan , p. 1003. At his trial for a weapons offence, drill artist, Loski, is reported to have explained that: I have to sound as real as possible otherwise you don't get far. It's all about YouTube views. If you say something, everyone talks about you and you get more views. Rap doesn't always have to make sense, it has to rhyme. I don't look scary so I have to say something that looks more than I am. (Taylor 2020) All of this means that if the defendant writes or performs rap, particularly drill or gangsta rap, some of their lyrics are likely to reference weapons or acts of violence, and may do so in an intentionally convincing way. As argued below, because of the conventions of rap and the factors influencing its content, these references should not be taken at face value. Within the case law, however, the courts were largely dismissive of attempts by defendants to explain and contextualise their music. For example, in R v O [2010] EWCA Crim 2985, at a trial for possession of a firearm and ammunition with intent to endanger life, the prosecution relied on a YouTube video in which O appeared 'rapping with many others and using words which were said to relate to guns and gangs' (at [6]). There is nothing in the judgment to indicate that the video had any connection to the specific circumstances of the offences charged (beyond it being a firearms case), with O having been arrested in a taxi with a loaded handgun in the footwell of his seat six months after the video was posted online. Moreover, the lyrics 'involved no specific threat to anyone in particular on any particular occasion' (at [25]). Nonetheless, the trial judge and Court of Appeal paid little mind to O's explanation that the video was an attempt to 'gain attention in the commercial music market' and that 'references to guns and violence were metaphorical' (at [7]), allowing the video to be used as evidence of O's propensity 'as a gang member to use gun violence for the purposes of endangering life' (at [17]).
In terms of the profile of defendants in the cases, all were male and the vast majority were young, mostly teenagers. The race of defendants was usually not explicitly stated. However, based on information in the judgments, as well as names and the context of the cases, it seems that lyrics and participation in videos was almost exclusively used as evidence against Black people. This is consistent with Nielson and Dennis' finding from the USA where, based on approximately 500 cases involving 'rap on trial', an estimated 95% of defendants are Black or Latino (2019, p. 18).
Taken as a whole, the case law indicates that music lyrics and videos are most often used as evidence against young Black men and boys in London who are accused of serious crimes. This is done with little regard for the conventions of the genre. Before further exploration of how the courts view and review rap evidence, the significance of race warrants further analysis.

Rap, race and gangs
One might expect most defendants in cases involving rap evidence to be Black, given that rap is a form of Black expressive culture. However, this in itself cannot explain the absence of white defendants in the case law. Rap is a hugely popular genre of music (BBC 2021), with a large white fanbase and many white artists, both amateur and professional. Also, not all of the search terms used for this study include the word 'rap'. Some terms, such as 'lyric', were broad enough to capture a range of genres that commonly reference illicit activities, including pop, rock and heavy metal. Yet, in 16 of the 29 crime cases from England and Wales on Westlaw with the word 'lyric' in the judgment, rap lyrics had been used as evidence against a defendant at trial or sentencing. Conversely, in just one of these cases is it clear that lyrics from a genre other than rap had been admitted as evidence against a defendant. 1 On LexisNexis, of four relevant cases with the word 'lyric' in the text which did not also appear on Westlaw, three concern rap music. This suggests that rap is targeted by police and prosecutors in a way that other genres are not, and it is targeted for use against Black defendants. In fact, Crown Prosecution Service (CPS) guidance on gang-related offences includes a section which explicitly links drill music to gangs. 2 Drill is also singled out in CPS guidance on offensive weapons, which advises prosecutors that consideration be given to the use of drill music videos as 'bad character evidence'. 3 The use of rap as character evidence is explained below.
The targeting of rap for use against Black defendants is indicative of a deliberate tactic, whereby police and prosecutors can draw on stereotypical narratives about Black urban youth culture to construct (or support) case theories. Despite the fictional or exaggerated nature of much of rap, the genre has long been associated with criminality (Fried 1999;Dunbar et al. 2016). This may be partly attributable to the conscious exploitation of stereotypes in rap music and the appearance of authenticity which many artists strive to achieve. Above and beyond this, however, the notion that rappers are dangerous criminals reflects longstanding stereotypes about Black people (particularly boys and men) as criminals (Gilroy 1982(Gilroy , 1987(Gilroy , ch. 3, 2003Drakulich 2015;Nielson and Dennis 2019, p. 81). These stereotypes are reinforced and reproduced by the persistent over-policing of Black people and the overrepresentation of Black people, especially Black children, as suspects, defendants and prisoners (Institute of Race Relations 1987; Lammy 2017; Joint Committee on Human Rights 2020; JUSTICE 2021), alongside negative representations of Black people in the media (Cushion et al. 2011;Turnnidge 2021), and the way in which rap has been linked to crime by the establishment, including the mainstream press. Consequently, rap music can be used to help build a case in which Black boys and men represent, or fit into, what Angela Davis referred to as the 'racialized figure of the criminal' (1997, p. 270), without expressly stating as much. In other words, as Quinn puts it, 'Police and prosecutors who don't want to appear overtly discriminatory can . . . let the rap music do the racist signalling for them ' (2018).
We see the racialised use of rap evidence also in the link to 'joint enterprise' and gangs. The doctrine of joint enterprise enables individuals to be convicted of crimes without committing the criminal act, or even being at the scene of the crime. Applying the term as a synonym for 'secondary liability', guilt is based on intentionally assisting or encouraging someone else to commit the crime. Many of the cases analysed for this study involve offences committed prior to the decision in R v Jogee [2016] UKSC 8, when the concept of joint enterprise included 'parasitic accessory liability', whereby Defendant 1 could be liable for a crime committed by Defendant 2 if Defendant 1 foresaw the possibility of Defendant 2 committing that crime while they were engaged in a common purpose to commit some other crime(s).
Approximately half of the cases concerned offences committed by way of joint enterprise. In these cases, lyrics and videos were sometimes used to help show a criminal association between defendants and/or to establish a shared intention to commit a crime. In R v Lewis [2014] EWCA Crim 48, for example, the seven appellants were accused of being part of a group of 42 individuals involved in a violent incident during civil unrest in Birmingham in 2011, which included setting fire to a pub and shooting at the police. The appearance of five of the appellants in music videos, some of which featured two or more of the appellants together, was used as evidence at their trials for a number of offences, including riot and possession of a firearm with intent to endanger life. The videos were not about the offences but were said to reference guns, gangs and/or gang signage, demonstrating association, gang association, pro-firearm and anti-police tendencies. As such, they were used alongside other evidence to show that the appellants were present in the area of the offences with the common purpose of carrying out unlawful acts. In other words, that they were 'in it together' (at [37]) regardless of whether they personally engaged in acts of violence or carried a gun. In several other cases, lyrics or videos were used to help show that a defendant had a reason or willingness to be involved in a joint enterprise, usually by way of their association with a gang, as explained below.
Joint enterprise doctrine has been criticised on the basis that it casts too wide a net, including by: applying to individuals whose involvement or association with a crime is far removed from the core liability of the principal offender; creating a risk of wrongful convictions and unfair sentences; and being disproportionately applied to Black people, acting as 'an indiscriminate ethnic vacuum cleaner' (Crewe et al. 2015, p. 268).
In their research on joint enterprise prosecutions, Clarke and Williams found that rap videos and lyrics 'formed a resource for building criminalised associations against negatively racialised groups and individuals' (2020, p. 126). The overrepresentation of Black people in joint enterprise cases, in turn, has the potential to reproduce the enduring construction of 'the violent Black offender' (Clarke and Williams 2020, p. 120;Williams and Clarke 2018), particularly if one does not appreciate the racialised assumptions behind the application of accessorial liability to Black people. According to Clarke and Williams, central to prosecution strategy in joint enterprise cases, is 'a criminalising guilt-producing "gang" narrative, deeply dependent on a range of reproductive racialised constructs of the Black criminal Other' (2020, p. 121). In other words, prosecutors use 'racialised signifiers' to indicate and reinforce criminality, to link defendants to each other and to the crime. The case law confirms that rap can operate as a 'racialised signifier' to help build a gang narrative.
Twenty-three cases were said to be gang related. In most, lyrics or videos were used as evidence of gang membership, association or affiliation, which was then used to link the defendant to a joint enterprise and/or as evidence to prove issues in the case, such as the motive for a crime. Gang involvement was sometimes, but not always, disputed by the defendant, as was the gang context of the case. In R v Awoyemi [2016] EWCA Crim 668, for example, the three appellants unsuccessfully challenged the admission of handwritten lyrics and appearance in a video, which was adduced at their trial for attempted murder and possessing a firearm with intent to endanger life. The lyrics were said to relate to 'violence, drugs, guns, using guns to get drugs and the DAG gang' (at [9]). In the video, two of the appellants were said to have made 'threatening gestures with their fingers to indicate guns' and rapped about the DAG gang, shooting, and using violence, including the line 'don't fuck with my family. Why? Cos I'll be eager to let slug fly' (at [9]). This was used to help establish gang association and, in turn, motive, association with firearms and violence, and to negative innocent association with each other and innocent presence, despite their being little to support the claim that the offences were gang related. In particular, the identity of the intended victim was not known, and there was no evidence of hostility between the two gangs the prosecution claimed to be involved. Without a link between the crimes and the gangs, it is unclear how lyrics about a gang actually furthered the prosecution case (McKeown 2017). Rap has also been used to link defendants to gangs for the purpose of sentencing. In R v Vasilieou [2020] EWCA Crim 742, for example, the judge relied on participation in 'drill videos' to find that the defendant was a 'willing and rising member' of a gang, thereby aggravating his sentence for an offence of wounding with intent (at [14]).
The term 'gang' is vague and has been heavily racialised in much the same way as the concept of joint enterprise (Williams 2014;Williams and Clarke 2016). Black people are overrepresented on gang databases, including the Metropolitan Police's 'Gangs Violence Matrix', which was created as a risk-assessment tool following civil unrest in 2011. In 2017, there were over 3000 people on the matrix, many of whom had no record of involvement with serious crime and were at low risk of offending (Amnesty International 2018). Seventy-eight per cent of those on the matrix were Black, despite police figures indicating that 27% of those responsible for serious youth violence in the London area were Black (Amnesty International 2018, p. 3). Although 40% of individuals have been removed from the matrix since 2017, in February 2021, 80% were still from an African-Caribbean background (Dodd 2021). These figures are unsurprising given Amnesty International's finding that 'Many of the indicators used by the Metropolitan Police to identify "gang members" simply reflect elements of urban youth culture and identity that have nothing to do with serious crime' (2018, p. 3). Consequently, youth culture has become inextricably linked to perceptions of what a gang is, and aspects of Black culture (including participation in rap) have 'started to become viewed through the prism of gang activity' (JUSTICE 2021, p. 33).
By disproportionately applying the term 'gang' to Black men and boys, the term comes to evoke (and perpetuate) stereotypical images of Black criminality. Consequently, when a young Black man or boy is on trial for an alleged gang-related offence, straight away, they fit the image of a gang member, and the prosecution theory becomes plausible. This is especially so where there are multiple defendants in the dock, as there often were in the cases analysed. Rap can then be used to amplify these images of Black criminality, to further link Black men and boys to crime, and, in joint enterprise cases, to establish common purposes or shared intentions. As the defendant's appearance is 'consistent with the stock image of gang-members and rappers as criminals' (Dennis 2007, p. 15), lyrics can easily be (mis)construed as statements of fact.
The Court of Appeal has justified the use of circumstantial evidence (including rap) to prove gang membership on the basis that violent gangs 'are unlikely to issue membership cards, and so proof of membership will almost inevitably involve the prosecution putting forward evidence of a number of circumstances from which gang membership could be inferred' (R v Elliott [2010] EWCA Crim 2378 at [31]). Unfortunately, circumstantial evidence of gang membership is not always carefully scrutinised before it is used in court. In particular, when it comes to rap, the courts have been content to accept (literal) interpretations and contextualisation of rap from police officers acting as 'gang experts'. However, being an expert on gangs does not in itself make one an expert on rap, and 'intense crime-fighting motivations and institutional racism might discourage more circumspect readings' of rap by the police (Ilan 2020(Ilan , p. 1003. In a report on tackling racial injustice, JUSTICE took the view that, in the context of drill, 'the use of police officers as experts amounts to no more than the prosecution calling itself to give evidence. They have little understanding of the culture within which Drill is created, and how it is made' (2021, p. 41). Although Ward and Fouladvand do not take such a strong line, they question the reliability of police opinions on rap, noting that, 'there is nothing in recently reported cases to indicate that police gang expertise is being subjected to any kind of rigorous scrutiny' (2021, p. 449).
The courts' acceptance of 'street-illiterate' (Ilan 2020) police officers as rap experts is revealing of its attitude towards rap. According to Nielson and Dennis: [B]y allowing [officers] to testify to the meaning or significance of a highly complex art form they have little to no familiarity with, courts are refusing to acknowledge that rap music is an art form, or that its creators are artists, which in the process gives prosecutors yet another tool with which to incarcerate young men of colour. (2019 p. 139) The racialised nature of rap evidence was not mentioned or addressed in any of the cases analysed for this study. This indicates a lack of concern about the way in which 'prosecuting rap' disproportionately affects young Black people. However, the courts have recognised that rap evidence is prejudicial, as discussed below in the context of 'prejudicial effect', where we will further consider how rap can be used to reinforce biases. Before turning to prejudice, the legal basis for admission of rap is explored. Within the case law, lyrics and videos were most commonly said to be relevant to (i.e. help to prove) a defendant's state of mind (including their intention), the motive for committing a crime, or they were used to help rebut a defence, such as innocent presence at the scene of a crime. This was often done by using rap to link defendants to a gang, and sometimes by using rap as propensity evidence, to show a propensity towards, for example, violence or using firearms. While the courts have been ready and willing to find rap relevant based on literal interpretations of lyrics and videos, it is argued here that rap is rarely relevant evidence of a crime, and would need to be very specific before it could be said to have probative value. This is because rap cannot usually be taken at face value: 'Rap music lyrics are neither inherently truthful, accurate, self-referential depictions of events, nor necessarily representative of an individual's mindset' (Dennis 2007, p. 4). Rap can tell us little, if anything, about a person's actions or disposition. Nor can we use rap to establish a comparative propensity to commit crime. There is no evidence to suggest that those who rap about crime (even a specific kind of crime) are more likely to engage in that crime than those who do not.

Relevance
Unfortunately, the cases include limited consideration of how the complexities, conventions and commercial drivers of rap music affect relevance. Instead, the perspective seems to be that the conventions of rap go to the weight to be attached to the evidence by the jury, rather than its relevance and admissibility. In other words, if the court is satisfied that lyrics and videos, when taken at face value, can (at least potentially) help to prove guilt, it is for the jury to decide whether, and to what extent, they do so. In R v Soloman [2019] EWCA Crim 1356, for example, the jury was left to consider whether the title line of a song, 'sold guns to str8 killers', helped to prove the appellant's activities and state of mind in respect of charges for possession of a firearm and ammunition with intent to endanger life, with the Court of Appeal reasoning that it would be 'reasonably apparent to the jury that lyrics of a song do not necessarily or perhaps commonly bear a connection with actual real life events' (at [15]). This position overlooks research in which participants tended to take rap more literally than other genres of music and associate rap with criminality (Fischoff 1999;Fried 1999;Dunbar et al. 2016;Dunbar and Kubrin 2018), as discussed below. It also overlooks the fact that, even where lyrics contain statements of fact, the conventions of the genre can make it impossible to distinguish fact from fiction. Thus, ordinarily, the reliability of lyrics as statements of fact (and as statements of fact which can be attributed to the behaviour or character of the lyricist) cannot be easily or properly assessed by a jury.
The cases also demonstrate little scrutiny of various factors surrounding the creation and content of songs and videos that affect their relevance and probative value. For example, many of the cases lack basic information about when lyrics or videos were created (whether, and how long, before or after the offence). When mentioned, the age of the material appeared to be of little concern even though, generally, the older the evidence of previous conduct, the less likely it is that it will indicate a propensity towards certain behaviour or be indicative of mindset, and the more likely admission of the evidence will adversely affect the fairness of the proceedings (Redmayne 2015). Yet, in R v Sode [2017] EWCA Crim 705, for example, the fact that one of the appellants had appeared in a 'rap video' two years before an allegedly gang-related murder, at the age of 14, was said not to 'reduce its impact or diminish its relevance' (at [53]), with no explanation as to why. The video, in which the appellant made a gesture and remarks said to be consistent with support for a gang, was used to help prove gang membership and, in turn, motive for joint enterprise murder.
The case law is also unclear about how active a defendant must be in a music video before it can become relevant evidence of their character and/or behaviour. In R v Alimi [2014] EWCA Crim 2412, mere presence as an 'extra' was insufficient to link a defendant to the contents of a video (or to a gang featured in a video). However, taking on a minor or supporting role by, for example, making gun gestures and getting a 'shout out', may be enough for the video to be deemed relevant. 4 Thus, a defendant who, despite not having written or even performed violent lyrics, can be taken to have endorsed violence through participation in a video, and that endorsement becomes evidence of guilt (Nielson and Dennis 2019, p. 80).
Most surprisingly, the specificity of lyrics (or lack thereof) seemed to have little bearing on determinations of relevance. The connection between lyrics and the details of the offence is perhaps the most obvious factor that could affect the relevance of rap to a criminal charge. If there is a strong and irrefutable connection, the evidence may be probative of guilt. This might occur where the defendant wrote lyrics which refer accurately to the specific events which make up the subject matter of the charge (with, for example, reference to names, dates and locations), and include information that was not in the public domain, shared in the local community or otherwise easily accessible to the defendant. Here, the lyrics may be relevant because they demonstrate intimate knowledge of the offence, and someone with such knowledge might be more likely to have been involved in the offence than someone without such knowledge. However, even in this situation, one must remain alert to the risk of misinterpretation and unreliability, and exclusion of the lyrics may be warranted on the basis of undue prejudice, as explained below. Yet, notwithstanding the fact that lyrics are not cited in full in most of the judgments, it seems that, in most of the cases, prosecutors were permitted to adduce lyrics that lacked specificity and did not reference the crime that the defendant was accused of. Rather, lyrics tended to reference violence, firearms or 'gangs' in a broad or general way. To mount a successful challenge against the use of rap evidence and provide the court with an explanation of rap music and culture, some defence lawyers have turned to expert witnesses. Anecdotal evidence suggests that this can be effective, but it does not guarantee the exclusion of rap evidence and, in any event, it appears to be relatively rare for the defence to offer expert opinion on the irrelevance of rap.

Bad character evidence
Rap lyrics and videos were usually adduced at trial as bad character evidence. Section 98 of the Criminal Justice Act 2003 (CJA) defines bad character evidence as 'evidence of, or of a disposition towards, misconduct' other than evidence which 'has to do with the alleged facts of the offence with which the defendant is charged' or 'is evidence of misconduct in connection with the investigation or prosecution of that offence'. Evidence of misconduct 'to do with' the alleged facts of the offence can include, for example, preparatory acts or evidence of a motive. For example, in Sode, even though the video at issue did not refer to or include details about the offence charged, it was said to link the appellant to a gang. Gang rivalry was the purported motive for the killing and, therefore, the video was said to be 'to do with' the alleged facts of the offence (it was, however, admitted as bad character evidence through gateway d of the CJA 2003, explained below).
Where evidence of a defendant's 'misconduct' (or disposition towards misconduct) is not 'to do with' the alleged facts of the offence charged, it is only admissible if it falls within one of seven routes to admission of bad character evidence, or 'gateways', set out in sections 101(1)(a)-(g) of the CJA 2003. The routes to admission of bad character evidence can overlap, with evidence sometimes being admitted through more than one gateway (and sometimes also as 'to do with' the offence, particularly where it goes to motive). In the case law, rap was most often admitted through gateway d (section 101(1)(d)), under which bad character evidence is admissible if it is 'relevant to an important matter in issue between the defendant and the prosecution'. The 'important matter in issue' was most commonly: intention; motive; or a defence, such as innocent presence at the scene of a crime. Rap was also admitted through gateway d as going to the important matter of the defendant's propensity towards violence or familiarity with weapons. As noted above, these issues were often framed within a gang context. In R v Rashid [2019] EWCA Crim 2018, for example, music videos were admitted through gateway d, as 'capable of showing that the defendants were "members or associated with gangs which exhibited violence or hostility and links with firearms"' (at [37]), and as evidence of intent and rebuttal of innocent association and presence. The appellants had been tried for weapons offences, including conspiracy to possess a firearm with intent to endanger life.
A detailed exploration of the statutory criteria for admission of bad character evidence, and its application to rap, is beyond the scope of this article. Suffice to say that, in accordance with the arguments made above in respect of relevance, the courts have been too quick to find that rap is 'to do with' the alleged facts of the offence or is 'relevant to an important matter in issue'. It is important, however, to question the categorisation of rap lyrics and videos as 'bad character evidence', which is further revealing of the courts' attitude and approach to rap (Owusu-Bempah 2022).
As explained, bad character evidence is 'evidence of, or of a disposition towards, misconduct'. Section 112 of the CJA 2003 defines 'misconduct' as 'the commission of an offence or other reprehensible behaviour'. 'Reprehensible behaviour' has been given its ordinary meaning, carrying with it 'some element of culpability or blameworthiness' (R v Renda [2005] EWCA Crim 2826 at [24]). The cases did not usually specify which aspect of the definition applied to rap. However, since writing or performing rap does not usually amount to a crime, it seems that rapping (or participating in music videos) about gangs and criminal behaviour is treated as reprehensible or, alternatively, as showing a 'disposition towards' committing crime, even where lyrics do not include any specific threats to any particular person(s). Here, again, we can see rap treated differently to other genres of music or violent pastimes. If writing or performing violent or graphic rap amounts to 'misconduct', or is otherwise evidence of bad character, is it not also misconduct to write or perform violent plays, write violent novels, play violent video games, or perform violent lyrics from other genres, such as pop or opera? What sets rap apart from other forms of art and violent pastimes which are not commonly thought of as being 'reprehensible' or as showing a 'disposition towards' committing crime? Most obviously, rap is a form of Black expressive culture, performed primarily by young people who fit a pre-existing image of what a criminal looks or behaves like. At play may be a 'racist assumption that [Black arts] cannot reach the same levels of sophistication as their white counterparts' (Ilan 2020(Ilan , p. 1003, and as such, can be taken literally and attributed to one's character in a way which other genres are not. Likewise, rap may be wrongly seen as 'a source of dissonance and incivility', excluding it 'from the pantheon of art' (Fatsis 2021, p. 39).
In the analysed cases, the categorisation of rap as 'bad character evidence' was not challenged by the defence. This might be because, in theory, the 'gateways' in the CJA 2003 create a hurdle for prosecutors, potentially safeguarding the defendant against admission of some evidence. If evidence is not 'bad character evidence' then, subject to other admissibility rules, the prosecution need not apply to adduce it and admission is based on a simple test of relevance. However, in practice, whether the 'gateways' affect admissibility is open to question. As noted above, as bad character evidence, rap was most commonly admitted through gateway d, as being relevant to an important matter in issue. If rap is deemed to be relevant, it is likely to go to an 'important matter' in issue, satisfying gateway d. In other words, the 'bad character evidence' label does little, if anything, to prevent admission of rap, and whatever little protection it may offer would probably become redundant if courts were to take a more rigorous and informed approach to assessing the relevance and prejudicial effect of rap.

Prejudicial effect
Even if one disagrees that rap is rarely relevant evidence of a crime, not all relevant evidence is admissible. Relevant evidence can be excluded where, for example, it would unduly prejudice a jury against the defendant. In particular, under section 78 of the Police and Criminal Evidence Act 1984, prosecution evidence should be excluded where admission would have such an adverse effect on the fairness of the proceedings that it ought not be admitted. In a very broad sense, evidence is prejudicial where it is adverse to the defendant or damages their case. In this sense, most prosecution evidence will be prejudicial. However, it may be more appropriate to think of evidence as having a 'prejudicial effect' where it will attract illogical inferences or lead the jury to reach an improper conclusion (Dennis 2020, p. 788). Where this occurs, the evidence is unduly prejudicial. The potential for rap music to unduly prejudice the court or jury against a defendant cannot be overstated. When jurors are invited to take rap literally, there is a danger that they will attach more weight to the evidence than is warranted, known as 'reasoning prejudice' or the 'risk of cognitive error'. There is also a danger that jurors will act in an emotional or irrational way, perhaps concluding that a defendant is worthy of conviction based on their (perceived) character or prior conduct, rather than proof of the crime alleged. This is known as 'moral prejudice' or 'the risk of emotivism' (Law Commission 1996;Ho 2006, pp. 61-2). The risk of both reasoning and moral prejudice arises from negative perceptions of rap as a genre, and from 'the artistic conventions of rap music lyrics: the stereotypical image of rap artists as criminals, and the content of rap music as it typically portrays a life consumed by violence, drugs, and crime' (Shumejda 2014, p. 32). Relatedly, to treat rap as evidence of a crime risks creating or evoking racial prejudice.
Several American studies have found bias against rap music, rooted in racial stereotypes. In Fried's 1999study, which was replicated in 2016(Dunbar et al. 2016, two groups of participants were given identical violent lyrics. The group that believed the lyrics were from a rap song rated them as more objectionable, dangerous and in need of regulation than the group that believed the lyrics were from a country song (the lyrics were from a folk song). Fried found that the same results were produced when, instead of referring to a musical genre, participants were presented with photos of the singer, with more negative responses when the singer was thought to be Black than white (Fried 1996). More recently, in a study by Dunbar and Kubrin (2018), participants were told that the same violent lyrics used by Fried were from a country song, heavy metal song or rap song. Participants were 'more likely to assume that a rapper is in a gang, has a criminal record, and is involved in criminal activity than are artists from other music genres, and this is based merely on the genre of the lyrics' (p. 521). In the study, when participants were told the race of the songwriter, no significant difference was found between evaluations of white and Black songwriters. However, when participants who were not given information about race imagined the songwriter to be Black, he was judged significantly more negatively than when he was imagined to be white, especially concerning assumptions about his criminal propensity. While the reason for this is not clear, one possibility identified by the authors is that participants who were told the songwriter was Black may have provided lower 'bad character' scores so as to not appear racist, whereas when a more subtle cue was presented, such as the genre of the lyrics, participants were able to provide an arguably raceneutral reason for a racially biased decision (p. 522).
The empirical research indicates that use of rap music in court can reinforce biases and preconceived notions of the criminality of rappers, with racial stereotypes playing a role in perceptions of rap music and the people who create it. As such, the risk of undue prejudice must be considered in the light of the over-criminalisation of Black people and racist stereotypes that disproportionately link Black people to crime. As Nielson and Dennis warn, 'rap offers police and prosecutors a convenient way to talk about young men of colour while invoking racial stereotypes that would otherwise be unacceptable ' (2019, p. 23). This is all the more possible when prosecutors cherry-pick lyrics and take them out of context (Rymajdo 2020), allowing them 'to "paint a picture" of the defendant at the time of the crime that is consistent with the prosecution's evidence and that resonates with jurors' (Dennis 2007, p. 1).
It has already been noted that the racialised nature of rap evidence was not addressed in the case law. More generally, where prejudice was acknowledged, its extent, and the reasons why the evidence was not considered to be unduly prejudicial, were seldom explained. For example, in Awoyemi, the Court of Appeal took the view that rap lyrics and videos were prejudicial but 'not unduly so', merely stating that the evidence 'indicated the extent to which the individuals concerned had signed up to gang and gun culture' (at [33]). This conclusion seems to have been reached by taking lyrics at face value and without appreciating that the extent to which someone has signed up to 'gang or gun culture' is precisely the kind of thing that cannot easily be deduced from rap lyrics. Likewise, in R v R [2011] EWCA Crim 1067, the Court of Appeal found that, although the 'YouTube evidence' at issue 'undoubtedly had prejudicial content', the trial judge was within his discretion to conclude 'that safeguards existed which would prevent undue weight being given to the prejudicial aspects of the evidence' (at [20]- [22]). The judgment does not explain the nature of these 'safeguards'.
It might be thought that undue prejudice can be prevented through judicial directions; the judge must explain to the jury the legal issues and evidence in the case. 5 Where rap is admitted as bad character evidence, the directions should include 'assistance as to its relevance that is tailored to the facts of the individual case', and 'It is, of course, clearly highly desirable that the jury should be warned against attaching too much weight to bad character evidence let alone concluding that the defendant is guilty simply because of his bad character' (R v Campbell [2007] EWCA Crim 1472 at [24] and [43]). In Awoyemi, for example, the Court of Appeal was satisfied that the trial judge: 'directed the jury on how to use the evidence and he repeatedly warned the jury that, even if they found the individuals were members of the DAG gang, they should not assume they were guilty of an offence' (at [39]). Rather, the jury could use the fact of gang membership to help resolve other issues in the case.
While directions explain the use to which evidence can be put, there is no general requirement that the judge (or anyone else) explain to the jury the culture, artistic conventions or social influences of rap music, and how the broader context can affect its probative value. By and large, jurors seem to have been trusted to decide for themselves whether rap is 'part of art or part of life' (R v |O [2010] EWCA Crim 2985 at [24]). Consequently, at present, the courts are likely to be 'underestimating the prejudicial impact of the lyrics on jurors' and 'overestimating the ability of jurors to ferret out their biases and prevent those biases from impacting their decision-making' (Dennis 2007, p. 30). If the defence want the jury to take account of the broader context of rap, they must adduce evidence of it. This can be through the testimony of the defendant, a witness, or, ideally, an expert. Yet just one of the analysed cases mentions an expert giving evidence on behalf of the defence, meaning that, in terms of 'expert' opinion, many jurors have only that of police officers to assist them.

Conclusion
The aim of this article is to provide information, context and critical analysis as to how, when and why rap is used as evidence against defendants in criminal trials in England and Wales. As noted at the outset, appeal judgments cannot tell us all that is happening in first instance trials. In fact, anecdotal evidence suggests that trial judges are becoming more receptive to arguments against admission of rap evidence, particularly when the defence is assisted by an expert. Still, the case law paints a worrying picture in which relevance and prejudicial effect are not carefully assessed, rap is uncritically categorised as 'bad character evidence', and the racialised nature of rap evidence is ignored. These findings should inform future developments on the admission and use of rap evidence. They could, for example, support an argument in favour of an exclusionary rule. In the US context, Nielson and Dennis have proposed 'rap shield rules' (2019, p. 157), completely banning the use of rap lyrics, videos, or promotional material as evidence in criminal proceedings. This is proposed not because rap can never be relevant evidence, but because, 'As a group, judges haven't shown themselves capable of applying the rules of evidence thoughtfully when it comes to rap music' (2019, p. 157).
A blanket exclusionary rule would be the most effective means of preventing unfair trials, wrongful convictions and further criminalisation of Black people and Black cultures, all of which are risked by the admission of irrelevant, unreliable and/or highly prejudicial rap lyrics and videos. However, this proposal is unlikely to be welcomed by all, and could create practical difficulties, including disagreement as to whether something amounts to 'rap' (Dennis 2007, p. 31). In the absence of an exclusionary rule, the courts should adopt a more rigorous and informed approach to assessing the relevance and prejudicial effect of rap. Formulating a meaningful test of admissibility, that sets fair and appropriate boundaries, is no easy task. But there are some basic factors that affect relevance, reliability and prejudice, and which should be considered before rap is allowed in court (Dennis 2007, p. 33;Lerner and Kubrin 2021, p. 39). These include: the extent to which lyrics and videos conform to the conventions of rap; who wrote the lyrics or what role the defendant played in music videos; the age of the material and whether it was created before or after the offence; the specificity of lyrics, including accurate and reliable reference to the facts of the offence at issue; whether lyrics contain information about the offence which is not readily accessible; and also, if the prosecution seeks to rely on only certain lyrics from a song or parts of a video, how this fits into the broader context of the song or video (as well as the defendant's wider body of work) and whether it has been cherry-picked to fit a certain narrative. All of this will probably need to be considered with the assistance of a suitably qualified expert, such as a rapper, industry insider, or scholar of rap and popular culture. Where rap is presented as evidence, the jury should be informed about the conventions of rap and what drives or influences its content. This information should form part of the direction from the judge, but, ideally, the jury should also hear from the expert.
Until a more rigorous approach is developed, it is important that: prosecutors think carefully about the (ir)relevance of rap evidence to the issues in the case, the implications of its use in court, and the suitability and reliability of police opinions on rap; defence lawyers continue to push back against the use of rap evidence, with assistance from experts; and judges display a desire and willingness to be educated on rap music and culture, so that they can make informed decisions. Without even these basic steps, 'an artistic act that shows drive, determination and creativity' will continue to be presented as 'dangerous and criminal', negating 'positive aspects of a defendant's character, making a finding of guilt based on weak evidence more likely', and sending a message 'to Black boys and young men that their cultural activities will be policed and prosecuted' (JUSTICE 2021, p. 41