My book, How Music Empowers: Listening to Modern Rap and Metal is published today by Routledge (Taylor & Francis). This book was written before the Digital Flows project began, and will be of interest to hip-hop listeners and creatives, as well as popular music academics more generally.
Here’s the blurb:
How Music Empowers argues that empowerment is the key to unlocking the long-standing mystery of how music moves us. Drawing upon cutting-edge research in embodied cognitive science, psychology, and cultural studies, the book provides a new way of understanding how music affects listeners. The argument develops from our latest conceptions of what it is to be human, investigating experiences of listening to popular music in everyday life. Through listening, individuals have the potential to redefine themselves, gain resilience, connect with other people, and make a difference in society.
Applying a groundbreaking theoretical framework to postmillennial rap and metal, the book uncovers why vast numbers of listeners engage with music typically regarded as ‘social problems’ or dismissed as ‘extreme’. In the first ever comparative analytical treatment of rap and metal music, twenty songs are analysed as case studies that reveal the empowering potential of listening. The book details how individuals interact with rap and metal communities in a self-perpetuating process which keeps these thriving music cultures – and the listeners themselves – alive and well. Can music really change the world? How Music Empowers answers: yes, because it changes us.
How Music Empowers will interest scholars and researchers of popular music, ethnomusicology, music psychology, music therapy, and music education.
For the first post on lofi hip hop before and during COVID-19, please see here.
Lofi as a genre, and its antecedents
To what extent can lofi hip hop be considered a genre, and to what extent do its producers, curators, and listeners participate in a virtual community? At present, there is a broad understanding of ‘internet lofi’ as a standalone paradigm – a distinct music style, at least – in hip hop producer communities. Emma Winston and Laurence Saywood characterise lofi hip hop as a reasonably large and stable genre (not merely a ‘microgenre’, so large is the audience) mediated online1Emma Winston and Laurence Saywood, ‘Beats to Relax/Study To: Contradiction and Paradox in Lofi Hip Hop’, IASPM Journal 9, no. 2 (24 December 2019): 40–54, 41.. Of course, genre systems are by necessity messy, a “blunt instrument” of sorts2Adam Krims, Rap Music and the Poetics of Identity (Cambridge: Cambridge University Press, 2000), 55.. To take one approach, hip hop could be situated as a genre and lofi a subgenre thereof, although the global manifestation of all sorts of musical styles and practices under the banner of hip hop may require higher-level categorisation. It is perhaps neater to see hip hop as a metagenre3Roy Shuker, Popular Music: The Key Concepts, 4th ed. (London and New York: Routledge, 2017), 148–151, a broad-strokes designation that recognises the breadth of musical activity which still bears some resemblance to a specific origin, stylistic identity, and international diffusion. Seeing it this way, lofi hip hop can be soundly understood as a standalone genre, although it is worth mentioning that the music and its accompanying practices are disconnected from most popular music industries.
The genre has gained many listeners outside of hip hop’s usual listener base: it is, by one account, a “YouTube phenomenon [… with …] millions of devoted followers”4Luke Winkie, ‘How “Lofi Hip Hop Radio to Relax/Study to” Became a YouTube Phenomenon’, VICE, 13 July 2018, https://www.vice.com/en/article/594b3z/how-lofi-hip-hop-radio-to-relaxstudy-to-became-a-youtube-phenomenon.. YouTube is unambiguously the contemporary home of the genre, with limited spread beyond: aside from a handful of journalistic accounts and appearances on streaming services, most online traffic is directed towards YouTube as the predominant platform for consumption. Furthermore, in the context of comments on lofi mixes which discuss ‘the comment section’ as a place, notions of virtual community come to the fore. The chat boxes of live lofi channels and the comments of mixes have been described as atypically convivial and supportive spaces, in contrast to YouTube comments’ tendency towards void-yelling or endless ad hominems5Kemi Alemoru, ‘Inside YouTube’s Calming “Lofi Hip Hop Radio to Relax/Study to” Community’, Dazed, 14 June 2018, https://www.dazeddigital.com/music/article/40366/1/youtube-lo-fi-hip-hop-study-relax-24-7-livestream-scene.. What happens to this space when listeners find themselves under the quarantine conditions of COVID-19? My focus here, on changes in listener commentary, offers some useful insights regarding the state of lofi as a genre and as a community, with manifest effects on how listeners self-identify in relation to the music.
Though a full genre history or even a folksonomy is far beyond the scope of this post, there are evidently complex links between earlier internet genres, other music described as lo-fi (with hyphen), and lofi hip hop. This modern manifestation of lofi, though musically quite distinct, can be traced back to certain genre conventions developed through hypnagogic pop, chillwave, and vaporwave. These are all, in a manner of speaking, genres enabled by YouTube’s quick content sharing and discovery facilities, and their aesthetic tendencies have emerged and dissipated in complex patterns. Hypnagogic pop is received as evoking “the state between sleep and wakefulness”6Georgina Born and Christopher Haworth, ‘From Microsound to Vaporwave: Internet-Mediated Musics, Online Methods, and Genre’, Music and Letters 98, no. 4 (2017): 601–47, https://doi.org/10.1093/ml/gcx095, 629., drawing on a romanticised and highly imaginary past characterised by ‘80s U.S. popular culture, especially consumerist rhetoric epitomised by cassette tapes. Aside from the amateur tape labels at the heart of hypnagogic pop, however, the genre in large part gave way to chillwave and vaporwave, both of which became better-known in mainstream digital culture and gained larger audiences, especially on YouTube.
Georgina Born and Christopher Haworth describe chillwave (the digital-native genre par excellence) essentially as hypnagogic pop with less experimentation and a sharper focus on conventional pop song structure. Its identity on YouTube is formed around retrofuturistic references to ’80s electronic dance music and video games, especially arcade racing games like Pole Position (borrowing the visual imagery and rhythmic consistency of driving). Vaporwave is often taken to be more actively ironic and politically engaged, serving as a critique of consumer capitalism and the failures of economic models emphasising constant growth. As Laura Glitsos has explored7Laura Glitsos, ‘Vaporwave, or Music Optimised for Abandoned Malls’, Popular Music 37, no. 1 (January 2018): 100–118, https://doi.org/10.1017/S0261143017000599., the disused shopping mall is an emblematic image for vaporwave aesthetics, tying together disdain for – and interest in the strangeness of – 1990s corporate marketing, the computerisation of office labour, muzak®, and the false future once promised by rapidly evolving technology markets like Japan.
To an extent, each of these genres has something lofi about them, not least their embrace of audio imperfections and disintermediation from the mainstream music industry. Indeed, Adam Harper’s 2014 doctorate traces ‘lo(-)fi’ as a term and an aesthetics of popular music production all the way up to the sounds and styles of lofi hip hop8Adam Harper, ‘Lo-Fi Aesthetics in Popular Music Discourse’ (DPhil Thesis in Musicology, Oxford, University of Oxford, 2014).. Although now only loosely connected to other meanings in the history of ‘low-fidelity’, it is striking how well-known lofi hip-hop, the contemporary flag-bearer for the term, has become across the social web and in youth culture. Consider, for one example, its appropriation in the rhythm game ‘lofi ping pong’, as Harper recently highlighted9Adam Harper, ‘What Is Lo-Fi? A Genealogy of a Recurrent Term in the Aesthetics of Popular Music Production’ (Research Seminar, City University London, 9 December 2020).. This transferral to other media indicates that the genre is associated with a reasonably stable set of aesthetic conventions (such as the sad/pensive young woman, pastel colouring, and visual artefacts – paint chips? – in the game artwork). But what exactly are the themes that characterise lofi hip hop as a genre, and how are these manifested in YouTube commentary – the genre’s primary discursive arena – before and during the COVID-19 pandemic?
Witches and ghosts
There is much to be said about the expression of personal identity in the lofi community, although little demographic data exists besides the informed impressions of Winston and Saywood10Winston and Saywood, ‘Beats to Relax/Study To’, 41., who assume the typical commenter to be a school or university student. I would add that the audience is likely to be in large part North American or European, given the concentration of producers and curators in these areas. Of those whose gender can be assumed by commenter name, my data shows a roughly even split between male and female participants, although this is not by any means a firm finding given the methodological issues here (as I outline in commentary on ‘WAP’). In any case, besides traditional social demographics, there are interesting new associations and personal identifications which have emerged during the pandemic.
The most statistically significant terms in the entire sample are “witch” and “ghost”. Related terms include “witches”, “witchcraft”, “wizard”, “spell”, “warlock”, and “wiccan”. First, a simple explanation: two of the most commented-on lofi mixes of 2020 are Homework Radio’s “Lo-fi for Ghosts (Only)”, uploaded in August 2019, and “Lo-fi for Witches (Only) [lofi / calm / chill beats]”, which went live on 27 March 2020. Naturally, there are few references to witches and ghosts prior to 2020, and the many participants responding to the video titles make use of these terms significantly more often in the during-pandemic sample11Even though the mix mentioning “ghost” was live for nine months before the cutoff date, the word appears at least twice as often since March. Furthermore, since both videos are on the same channel, the success of “Lo-fi for Witches” may have … See full note. This alone makes for a relatively uninteresting finding: there’s a new video and commenters mention its title. However, the mystical imagery and fantastical identities invoked by these mixes during the COVID-19 pandemic – and picked up on by the audience – are worth some consideration.
What does it mean to feel like a ghost or a witch? Certainly, lofi listeners identify more closely with these figures of late. Loneliness and social isolation are the topics of numerous research articles which have have examined the impacts of quarantines and lockdown measures12Yuval Palgi et al., ‘The Loneliness Pandemic: Loneliness and Other Concomitants of Depression, Anxiety and Their Comorbidity during the COVID-19 Outbreak’, Journal of Affective Disorders 275 (1 October 2020): 109–11, … See full note. Indeed, one technical term used often in public health advice is ‘self-isolation’, which encourages staying at home (sometimes ‘shelter in place’) as responsible behaviour for most citizens33This is not lost on Dazed journalist Sophie Atkinson, who describes lofi girl as “the poster girl for responsible coronavirus behaviour for Gen Z and millennials alike”. See Sophie Atkinson, ‘The “24/7 Lo-Fi Hip Hop Beats” Girl Is Our … See full note. The practical consequences of self-isolation align closely with the emotionally isolated imagery of ghosts and witches. For example, experiences of staying indoors (perhaps against one’s will), seeing less sunlight, feeling invisible or unable to engage with others, and having more time to dedicate to (potentially esoteric) hobbies may cause positive identification with these supernatural figures, who are excluded from a regularly functioning society. As for the other terms which appear in much higher proportions, such as “wizard” and “warlock”, these are also responses to the video titles, mostly asking questions to the effect of “what about lofi for wizards?”.
Somewhat more speculatively, the witch figure and its accompanying “witchcraft”, “spell”, and “wiccan” may have struck a chord with the lofi audience due to teenage-oriented associations between studying and magic (as popularised by the Harry Potter universe, for one example). The significant rise in studying to lofi while spending more time inside and socially disconnected makes the witch an engaging figure, especially playing on pop cultural tropes of the witch as a teen hero. Studiousness and natural magical power have gained close associations through characters like Hermione and Willow from Buffy the Vampire Slayer, which the video of “Lo-fi for Witches” draws upon through visual references. It replaces ChilledCow’s ‘lofi girl’ figure with a cartoon witch leant over a desk, complete with pointed hat, open spellbooks, wooden wand, black cat (itself boasting demonic wings and tail), vials, inks, a quill, and – perhaps anachronistically – skull-shaped pink headphones resting on her fantasy-/anime-inspired pointy ears. Naturally, as students keep going under lockdown conditions, confined to bedrooms and dorm rooms in self-isolation, the image of a lonely witch working hard on their arcane craft feels increasingly relatable. A handful of commenters asked the community for a spell to help them pass upcoming exams.
The references to witches also appear to have invited commentary from those interested in neopaganism, hence “wiccan” emerging as a term used in this space. Popular interest in mystical practices appears to have increased over the last few years, as evidenced by participants in communities like WitchTok13Lauren McCarthy, ‘Inside #WitchTok, Where Witches of TikTok Go Viral’, Nylon, accessed 11 January 2021, https://www.nylon.com/life/witchtok-witches-of-tiktok.. There may be overlap between contemporary witchcraft and a reemergence of anti-science rhetoric associated with, for instance, COVID-19 conspiracy theories. The connotations of alternative medicine and New Age thought more broadly are troubling given the infodemic precipitated by the disease14Matteo Cinelli et al., ‘The COVID-19 Social Media Infodemic’, Scientific Reports 10, no. 1 (6 October 2020): 16598, https://doi.org/10.1038/s41598-020-73510-5., although most comments here are pretty innocuous. On the whole, witches (and ghosts) appear to capture something distinct about the lofi audience’s experience of social isolation, circumstances undoubtedly intensified by the pandemic. Almost all of the discussion in this section comes from the “witches” video, so there is also the general observation to be made that mix titles and visuals can strongly sway the topic of commentary, even when the music is relatively indistinct from other lofi mixes. Homework Radio’s comparable attempts at producing lofi for “vampires”, “space cowboys”, “demons”, “mermaids”, “angels”, “grim reapers”, “cats”, and “insomniacs” have not (yet) proved quite as successful. The pop-cultural imagery of witches and ghosts evidently fit a particular niche that aligns with lofi’s aesthetic themes and the specific circumstances of COVID-19.
There are broader aesthetic tendencies that form part of the genre’s identity: a prominent example is ｔｈｉｓ ｋｉｎｄ ｏｆ ｔｙｐｉｎｇ , which has been recognisable as a feature of lofi discourse since at least 2015. Known as ‘aesthetic’ text or more reductively ‘the vaporwave font’, typing with fullwidth characters (or simply placing a space between each letter) is a hallmark of lofi hip hop comment sections. This use of vaporwave’s earlier stylistic signifiers – perhaps by way of capital letters and hanzi/kanji – is inherited fairly directly, and so the text finds its way into many lofi mix titles and comments15For example, tracks now considered to be vaporwave classics, from albums like フローラルの専門店 (‘Floral Shoppe’) by Macintosh Plus and FINAL TEARS by INTERNET CLUB, are commonly rendered in capitals and include Chinese and Japanese … See full note. The common letters “c”, “d”, “e”, “g”, “h”, “k”, “m”, “n”, “o”, “p”, “r”, “s”, “t” (which are processed by Mozdeh as independent words, and thus revealed in the sample) appear frequently within such texts. However, the statistical significance shows them in a much greater proportion of comments written prior to March than those posted during the pandemic. Put simply, individually spaced letters have been used less often recently: this kind of typing, once a key lofi signifier, appears to be going out of style.
There are many reasons why this could be the case. A straightforward explanation is that the COVID-19 pandemic and the genre’s rising popularity over time has brought many new listeners to the YouTube mixes16Julia Alexander provides various pieces of evidence for an increase in viewership correlated with COVID-19 lockdown measures. See Julia Alexander, ‘Lo-Fi Beats to Quarantine to Are Booming on YouTube’, The Verge, 20 April 2020, … See full note, who have posted using normal typography. While the aesthetic text tradition carries on today (and is not necessarily less meaningful for the genre’s participants), it was simply present in a greater proportion of comments before March 2020, as the arrival of new users has skewed the sample towards conventional text. Whether newcomers observed the typing trend and chose not to emulate it or are yet to encounter it altogether is unclear. However, the style might be seen as somewhat tired by now, mostly dissolved into meme fodder (as museumised by the Know Your Meme page). Across the web, the typing style might be interpreted pejoratively as a symbol for a subculture of over-sentimental digital natives with ｎｏｓｔａｌｇｉｃ ｆｅｅｌｉｎｇｓ17I write this in reference to the single-word titles of two popular lofi mixes: “ＮＯＳＴＡＬＧＩＣ” (15 million views), uploaded by Neotic, and “Ｆｅｅｌｉｎｇｓ” (11 million views) by Cabvno. Note that both were published … See full note.
Indeed, nostalgia is a central theme of Winston and Saywood’s study of lofi hip hop, as they observe the genre activating “hyper-specific memories of popular media which may have been consumed during, or at least associated with, a listener’s childhood”18Winston and Saywood, ‘Beats to Relax/Study To’, 41.. However, it appears that a recent emphasis on new topics – not only witches and ghosts but also COVID-19 measures, studying, etc. – has led to a significant decrease in pop-cultural references typical of the genre. Mentions of the animated TV series Cowboy Bebop (“bebop”) and Rick and Morty (“morty”) appear much more often prior to March 2020. On the one hand, this might be surprising, as viewership of popular TV shows surged during lockdown conditions19BBC News, ‘TV Watching and Online Streaming Surge during Lockdown’, 5 August 2020, sec. Entertainment & Arts, https://www.bbc.com/news/entertainment-arts-53637305., and we could expect to see an associated spike in commentary on them. On the other hand, this indicates the declining popularity of mixes using visual imagery and audio samples from cartoons. For instance, the 2016 mix “Chill Study Beats 2 • Instrumental & Jazz Hip Hop Music ” opens with spoken audio extracted from Cowboy Bebop, which listeners frequently identify. In 2017, Cabvno posted Ｆｅｅｌｉｎｇｓ, featuring a short video loop of Morty (plus superimposed VHS tape effects) which also appears as the video thumbnail.
Fewer visual references appear in recently uploaded lofi mixes. Evidently, the more generic imagery of witches have drawn a lot of attention away from specific TV shows, but there might also be a stylistic change present in lofi hip hop, with fewer samples of popular media used across the board.
Indeed, a 2017–18 series of ‘Volumes’ by the bootleg boy has an aesthetic identity strongly informed by specific cartoon characters. A still image or short animated loop of the character (usually crying) is accompanied by audio samples from the associated series, especially extracts of emotional dialogue. For instance, in the thirty seconds opening “ＢＡＤ ＦＥＥＬＩＮＧＳ”, a video of Bart Simpson crying is complemented by audio from The Simpsons drenched in reverb (ostensibly to seem more dramatic and significant). Marge asks Lisa to repress her emotions: “take all your bad feelings, and push them down…”. After a pause, Lisa responds “but I don’t feel like smiling”, and this resolution to embrace her negative emotional state continues echoing as the beat drops with the entry of drums and bass. Bart carries on spilling a tear every second in the video. Instances like this exemplify both referential and nostalgic qualities of lofi, which appear less common in recent months (for instance, “simpson” appears at least twice as often prior to March 2020). The cultural spread – at least of the meme status of the lofi mix title format – became evident when the BoJack Horseman episode “Sunk Cost and All That” mentioned lofi. the bootleg boy uploaded a BoJack Horseman ‘volume’ in August 2018, but this is the first time a TV show referenced lofi back so explicitly20Prior to this, The Steven Universe character Connie has been depicted emulating the ‘lofi girl’ artwork. See this image hosted on KnowYourMeme..
However, a couple of new connections have appeared. No significant lofi ping pong mentions in my sample of comments (yet?), but many listeners have revealed that they were linked to a specific lofi mix by the 2020 interactive detective game “Duskwood”. As for other games, older videos using generic and thematic imagery can gain relevance down the line. For instance, several commenters took to the 2018 mix “Samurai ☯ Japanese Lofi HipHop Mix” to mention the game Ghost of Tsushima (with a significant proportion of recent “tsushima” references), which was released in July 2020. Leaving the broadly stereotypical and Orientalist framing to one side, the game and the ‘Japanese’-inspired mix appear to be mutually enhancing for players and listeners. One commenter (paraphrased, but all significant words retained) reflects that:
“This mix feels like when I’m playing Ghost of Tsushima just riding through the sunny fields past all the flowers”
Mentions of the mix visuals have increased recently, with the terms “balcony”, “cat”, and “typing” referencing specific imagery used in the video content (“1 A.M Study Session” for the first two, “aesthetic song” for the typing animation). The increase in uses of “she” partially owes to the widespread commenting on the visuals of ‘lofi girl’, as in artwork by Juan Pablo Machado or Margaux Peltat, but also appears in a lot of posts about interpersonal relationships with female people (e.g., “I miss my girlfriend, she used to cheer me up”). However, references to “clock” appear significantly more often in the earlier sample, usually noting the static time reading displayed on the digital clock in “3:30am”. This 30-minute mix displays (you guessed it) ‘03:30’ for the entire duration, and the flashing pink light illuminating the clock face is the only animated part of the video. It surely draws attention to a sense of frozen time, although users commonly used to poke fun at the clock (either its disparity with the analogue star-shaped clock also depicted in the video or the YouTube scrub bar, conspicuously keeping track of seconds as the mix progresses). Maybe this video is simply less popular, or the joke might just be tired.
References to music have reduced substantially during COVID-19, save for one term which I’ll reveal in a moment (the suspense!). In the years prior to the pandemic, hundreds of commenters recognised a popular lofi beat, “Waves” by Freddie Joachim, and pointed out that it was used by Joey Bada$$ (“Waves”) and later J. Cole (“False Prophets”). This is much less common now, likely indicating a natural decline over time rather than any significant relation to COVID-19: a few years have passed since the rap tracks were released. Music is discussed more generally using the terms “mix”, “genre”, “vaporwave”, “chillhop”, “music”, “beat”, “where”, and “track”. All of these show statistically significant decreases since March 2020, as people are generally commenting on the music of the videos less often. What occupies their thoughts instead, one would assume, are the other terms discussed throughout this blog post (especially part one’s focus on COVID-19 and studying).
The term “mix”, once a staple for referring to any video featuring twenty or so lofi tracks, has dramatically fallen out of use (present in 1.99% and 0.98% of comments before and after March 2020 respectively). Use of this term is generally pretty simple, like “this is my favourite mix”. “Music”, “beat”, “where”, and “track” have similar applications, as listeners either discuss the music or attempt to locate pieces in different contexts. “Genre” is a more interesting decrease, because a few years ago it was much more common for people to ask questions like “what is this genre called?”. And 161 commenters gave the (erroneous?) answer “vaporwave”, which has dropped to only 10 mentions since March 2020. One user suggests that the term ‘vaporwave’ may have originated from “a group of guys who vape a lot”. Those inaccuracies aside, a couple of the lofi mixes do feature some tracks that might be better labelled vaporwave (e.g. “ＮＯＳＴＡＬＧＩＣ”), especially before the genres were more separately codified. Further pointing to the complexities of the genre network active in this space, posters were much more likely to use the term “chillhop” between 2016 and early 2020 than in recent months. This decline might also give some indication that the uploaded mixes (but not necessarily the live-streams) of the channel Chillhop Music have decreased in popularity.
The largest upswing in music-related terminology – here’s the big reveal – appears for the word (the one, the only) “lofi”. Clearly, the genre has coalesced around this label, being the only referential word that has become much more frequent in the sample of texts since major COVID-19 pandemic measures were put in place. Like many of the terms in this section, I doubt this change has any significant relationship to the coronavirus itself. Much more likely is that we are observing changes in the life of lofi hip hop as a genre, as discourse shifts in response to codifications and crystallisations of the music and the digital spaces it occupies. Slightly further down the list, still highly significant (p≤0.001) although beyond my cutoff for extreme significance, the terms “hop” and “hip” appear (in that order) as more popular in the earlier subset of the sample. So without the additional clarifications of “hip hop” and “beat(s)”, as well as “chillhop” and “mix”, the standalone genre label “lofi” is in and, presumably, here to stay. Users confidently use the term to refer categorically to ‘curated lofi hip hop mixes on YouTube’:
“Lofi is awesome. I listen to it everyday from 1pm to 10pm”
“I’m super high right now and I’m telling you this is the best lofi to listen to when you’re faded”
“The world would be ten times better if everyone listened to lofi”
“I feel so chill when lofi is on”
“Thank you for helping me study with the relaxing lofi”
“I love laying in bed with some lofi and just thinking about everything”
It therefore appears that, at present, lofi can be confidently pinned as a standalone genre. This is not an effect of COVID-19 per se, although shifts in the online audience precipitated by the pandemic conditions might indicate clearer definitions of its aesthetics, boundaries, and conventions (even as they remain in flux).
Topics of conversation: appraisal, conversation, memes
There has been less use of a number of related terms, which I have thematically clustered around appraisal of the music: “love”, “dope”, “good”, “first”, and “beautiful”. Why might these appear less often during COVID-19? The vast increase in terms like “safe” offer one explanation, as commenters are focusing on more fundamental needs, like ideas of security, and are therefore distracted from aesthetic appreciation. To read a little further, the less frequent appearances of “dope” are understandable given the similar drop-off in references to “hip hop”. With its origin in hip-hop slang, the term (like reference to the metagenre and related tracks by Joey Bada$$ and J. Cole) tails off over time into the contemporary context of (just) “lofi”. This development may suggest an influx of listeners from outside the typical cultural spaces of hip-hop, as the genre is further distanced from its origins in instrumental hip-hop production (Nujabes, J Dilla, Madlib) and establishes new aesthetics inspired by anime21This observation does not mean to downplay the solid ties between other forms of hip-hop and anime. and other popular animated TV shows, studying, creative hobbies, and cosy bedroom comforts.
Alongside the decrease of inquiries about the genre name, fewer listeners are asking about or evaluating the “first” track of a mix, which is typically how this word is deployed. The reduction in the frequency of “love” is more surprising (to me, at least), given how prominently it appears in general: it is used in 8.4% of comments prior to March 2020, and 6.8% after. The lofi audience is highly expressive about their appreciation for the mixes, other commenters, and the wider community itself. Not only is lofi more widely understood as a genre on its own terms, but listeners are increasingly recognising the presence of a lofi community, complete with specific values, priorities, and normative practices. It may come as no surprise that uses of the term “community” have significantly increased in recent months. Julia Alexander reports that
YouTube’s lo-fi hip-hop community has for years offered a place to virtually gather, do homework, and find comfort in the random messages of strangers that populate in live chats. Now, as we’re all stuck inside due to the pandemic, those streams have become more popular than ever — not just as background music, but as ways to find community in a difficult time.22Alexander, ‘Lo-Fi Beats to Quarantine to Are Booming on YouTube’.
As it grows, this virtual community has become increasingly conversational, as indicated by the terms “hi” and “comment”. The common greeting appears in many comments that are direct replies to other individuals, but also quite commonly to an assumed audience (“hi, everyone”). Commenters greet other listeners as (a substitute for) social interaction and, I might suggest, a momentary statement of ontological security. This hailing of a virtual community or imagined other, sometimes posed as “whoever ends up reading this comment”, may help users gain feelings of “security in the sense of self and confidence in the continuity of one’s being-in-the-world”23Lynn Jamieson, ‘Personal Relationships, Intimacy and the Self in a Mediated and Global Digital Age’, in Digital Sociology: Critical Perspectives, ed. Kate Orton-Johnson and Nick Prior (London: Palgrave Macmillan UK, 2013), 13–33, … See full note. Indeed, lofi mix comment sections appear to have become important outlets for coping with loneliness experienced during COVID-19. Frequently, commenters will offer advice to the crowd:
“hi everybody. make sure to drink enough water and regular meals today”
“Hi to everyone studying for exams, we’ll get through this, keep going!”.
Despite the increase in introductory terms like “hi”, the lengthy personal reflections typical of lofi commentary seem to be dwindling somewhat. There is a clear reduction in the use of the common sentence words “I”, “of”, “what”, “up”, “these”, “in”, “it”, “the”, “about”, and “this”. It seems likely that, instead of quite so many lengthy comments diarising individual feelings and events, users are more often greeting each other and partaking in newer trends of the comment section. The term “vibing” is extremely significant here, which I see as a conversational term due to the notable presence of comments like “who’s still vibing to this in 2020?”. This desire to share an activity with others further affirms a sense of communality united by common behaviours. One user gained two thousand likes on a post detailing an apparently relatable scenario:
“anyone else vibing to this in bed in a hoodie waiting for the coast to be clear at 2am to go grab some cereal?”
There is a vivid image here – a young person who lives in their parental home and with a private bedroom (or perhaps a dorm in a shared halls of residence) listening to lofi in the early hours using a personal laptop or smartphone and awaiting a quiet time when others in the house are asleep so that they can sneak to a shared kitchen for a late night snack – with which thousands of lofi listeners connect.
Not only are certain listening habits shared by many users, but lofi also boasts a series of popular in-jokes. In this context, by ‘memes’, I am basically referring to text-based humour. I borrow the term from web culture because this is how users would likely address these highly repetitive and formulaic comedic posts. Notably, the popularity of these memes is on the rise. The terms “zooted”, “homer”, “plot”, “pov”, “twist”, “copying”, and “copied” are much more frequently used after March 2020. There is one video where these memes run rampant, although variations on them do appear elsewhere: this GEMN video, featuring an oversaturated video loop of Homer Simpson (yes, another appearance of The Simpsons: the cartoon nostalgia runs deep) with heavy eyelids, nodding his head from side to side while driving as clouds pass him by. This animation has ‘stoned’ affordances of spaced-out relaxation and an altered sense of time, as though under the psychoactive effects of cannabis. The visuals have provoked users (and, likely, bots) to repeatedly post the phrase:
This spam/copypasta meme appears so many times that repetition and ubiquity become part of the joke itself. Out-of-the-loop visitors to the comment section can be struck by the uncanny omnipresence of this text string as they scroll down. Later, commenters vary on the theme:
“POV: You’re looking for a comment that says “Homer is so zooted he’s not even driving. The clouds are moving not him.””
This repeats at increasingly meta levels:
“POV: you’re looking for a comment that doesn’t say “POV: you’re looking […] “Homer is so zooted […]”””.
The “plot twist” format works similarly to “POV”, asking the reader to imagine the scenario which follows, typically as a form of subversive humour (as in “plot twist: she [lofi girl] is actually right handed and has just been practicing writing with her left all this time”). These simple comedic expressions manifest as small exercises in community recognition and bonding, as commenters come up with their own “plot twists” and engagements with the ongoing in-jokes. The increasing popularity of such memes may again indicate the arrival of a larger audience less familiar with lofi conventions. This would account for the “major bump in viewership” witnessed by lofi channels during March and April 202024Alexander, ‘Lo-Fi Beats to Quarantine to Are Booming on YouTube’.. As mixes have gained larger audiences, users’ tendencies have, on average, shifted towards generic web-culture meme spam and away from older staples of lofi commentary like personal reflection.
Emotionally sincere commentary has been a hallmark of lofi mixes since their earliest days on YouTube, with connections to Tumblr’s conventional mode of discourse. Winston and Saywood write of the “emotional narrative microfiction” that characterises much lofi commentary, which they observe in relation to “affective labour”25Winston and Saywood, ‘Beats to Relax/Study To’, 47–48.. While this was evidently a mainstay just a few years ago (Winston and Saywood mostly quote posts from 2017), numerous words relating to personal feelings have appeared less commonly since around the time that COVID-19 measures were enacted. These include, in rough order by statistical significance, “life”, “feel” and “make” (often in the phrase “makes me feel”), “alone”, “depressed”, “feeling”, “nostalgic”, “out” (e.g., “get all my feelings out”), “sad”, “childhood”, “everything” (usually to mean overwhelming feelings), “mind”, “nostalgia”, “depression”, and “lonely”. The only emotional term that has significantly increased in use is “safe”, which is understandable given the near-global public health messaging around safety (see also part one)26Note that “safe” frequently appears in the idiom “stay safe”, which may not be as emotionally charged as, say, “depressed”..
What should we make of the reduction in the use of so many emotional terms? Like many other words discussed here, there is undoubtedly a sense that more diverse commentary from outsiders and newcomers to the lofi comment sections has increased the general ‘noise’, enabling us to see a reduction in marked terms. It might also be the case that the emotional expression that used to accompany lofi mixes so popularly has runs its course, as users are finding different outlets for discussing their feelings. More speculatively, I wonder whether COVID-19 messaging has resulted in inhibited commentary on emotions – a sort of ‘stiff upper lip’ attitude – where listeners are attempting to demonstrate resilience in the face of the grave circumstances, and therefore less often detail how they feel in the comments. In any case, this is an interesting finding given that one would expect the pandemic to have prompted discussions of loneliness, even where commenters report positive connections, like
“im in quarantine and yall make me feel less alone”.
In line with the reduction in emotional language, there has been less recourse to ‘explicit’ language since March 2020: “shit”, “fuck”, and “fucking” all see a significant drop in usage. While “fuck” is usually employed as an expletive (rarely with sexual connotations), “shit” does not always have an intensifying function. “I’m such a piece of shit” certainly carries an intense, negative emphasis, but the term is more commonly used as a generic sentence object, as in “stuff”, e.g.:
“listening to this shit makes me feel alright”
It may come as a surprise that lofi commenters use less emphatic language in recent months, but it perhaps suggests endeavours to stay calm during extraordinary events. I am reminded of popular discourse accompanying the first lockdown in the UK, where there was an impression of an emotional shift over time as early anxieties turned to humdrum boredom: one study of Google Trends data (albeit pre-peer-review) over this period shows a substantial and sustained increase in web searches relating to boredom27Abel Brodeur et al., ‘Assessing the Impact of the Coronavirus Lockdown on Unhappiness, Loneliness, and Boredom Using Google Trends’, ArXiv:2004.12129 [Physics], 25 April 2020, http://arxiv.org/abs/2004.12129.. However, it is also possible that lofi “to relax to” plays an increasingly important part in the reduction of emotionally extreme words. The more moderate language used by lofi commenters since the start of the pandemic may suggest that the “supportive elements of the chatroom and the calming effects of the music” are working28Alemoru, ‘Inside YouTube’s Calming “Lofi Hip Hop Radio to Relax/Study to” Community’.!
The findings discussed in part one reveal a variety of responses to COVID-19 in lofi hip-hop commentary. Users described their current circumstances, with two particularly common scenarios: listening in lockdown or quarantine and having lofi music on while completing study or homework. Such contexts were capitalised upon by new mixes promoted to an audience of “witches” and “ghosts”, figures that appeared particularly relatable given the social isolation of lofi listeners during the pandemic. The other themes addressed in this blog post, however, have much less to do with direct impacts of COVID-19 and much more about developments in the genre’s norms and its changing audience between 2016 and 2020. These are more ‘natural’ shifts over time, I would argue, and may be comparable to the chronological evolutions of other genres.
The genre itself has thoroughly stabilised around the term “lofi” (with other terms falling out of favour) at the same time that conventional practices have diversified (for instance, less frequent use of signifiers like ａｅｓｔｈｅｔｉｃ text). New notions of identity have been prompted by video titles and artwork, and memes specific to the comment sections have become more proportionally represented. Popular media (especially cartoons) which evoke nostalgia in listeners are referenced less often, whereas the video imagery of lofi mixes themselves are discussed more frequently. On the whole, we see less frequent emotional expression, as fewer users are taking to the comment section as a place to share and process their feelings (with the notable exception of “safe”, no doubt influenced by COVID-19 public health messaging). In general, people wrote fewer full sentences than before the pandemic, and employed less ‘explicit’ language implying intense feelings. There is less negative emotional reflection as lofi’s large student listener base is, it seems, productively diving into their studies (that is, they appear to feel better now they’re working harder: how’s that for a neoliberal takeaway?).
Taken together, these developments suggest an influx of first-time viewers adjusting to (and thereby altering) the conventions of lofi commentary. Notions of community are strengthened as a new audience arrives during the strange, troubling, and often lonely circumstances of COVID-19. Individuals in this expanding listener base often introduce themselves in comments, and appear to be welcomed by others. In many cases, this introduction takes the form of participating in copypasta threads, a relatively novel feature of lofi commentary (inherited from other meme activity across the social web). The introduction of new elements might appear to be a threat to “one of the kindest communities on the internet”29Alemoru, ‘Inside YouTube’s Calming “Lofi Hip Hop Radio to Relax/Study to” Community’., but there does not seem to be much outgroup resentment, hipsterism, or gatekeeping, as far as my sample shows.
There is much more to say here, not least because some of the most popular lofi resources are now live-streamed ‘radio channels’ with a real-time chat function. Mix videos and comments may be revealing earlier trends in discourse that do not fully represent lofi in its most recent manifestation, as the community increasingly embraces live videos with accompanying chatrooms. The critical scholar in me is most troubled by commercial adoptions of the trend (a death knell for an internet subculture if ever there was one): in part one, I mentioned Will Smith’s merchandise sales mix, but there is also The AMP Channel (developer community)’s “code-fi”, Joma Tech (a programming course provider)’s “chill lofi beats to code/relax to” and Code Pioneers’ Amazon affiliate promotion of a smart mug:
I should point out that ChilledCow, a long-standing and fan-favourite lofi channel, also began selling its own merchandise in 2018, although this can be perceived as a more organic form of growth compared to the sudden appearance of coding businesses using lofi for marketing purposes in 202030For a more in-depth examination of how artists and uploaders generate income from lofi hip hop radio stations, see Cherie Hu, ‘The Economics of 24/7 Lo-Fi Hip-Hop YouTube Livestreams’, Hot Pod News, 28 January 2020, … See full note. Moreover, such product advertisements using the aesthetic veneer of lofi sit alongside a variety of political takes on the lofi ‘to [verb] to’ trope.
The Conservative Party (UK) evidently saw the potential for digital communications and youth outreach in their video lo fi boriswave beats to relax/get brexit done to, which mostly features beats by Audio Network’s George Georgia overdubbed with lengthy samples of Boris Johnson speeches31Thanks/complaints to Adam Harper for the link. I assume that the tracks are licensed correctly, though the video uses a Creative Commons Attribution license without any attribution of artist or track names. This subverts a lofi convention to credit … See full note. Similarly, lo fi merkelwave beats to relax/get nothing done to has racked up 1.7 million views. Warping the format further, the entire speech (not occasional audio snippets) and vdeo recording of an 8½ hour filibuster by Bernie Sanders (and Mary Landrieu) is accompanied by lofi-style beats in Bernie Sanders 8 1/2 hour Filibuster but it’s Lofi. The title here evokes another YouTube trend, such as “The Bee Movie but…”, featuring dramatic edits to humorous effect. Perhaps reductively, lofi is borrowed here as a widely known, stable entity approaching meme status.
Indeed, as lofi has established certain conventions (independent from associations with vaporwave and chillwave, fullwidth text, kind comment sections, cartoon nostalgia, etc.) it has become ripe for appropriation, with a transferable set of aesthetics recognisable in other contexts (e.g., lofi ping pong, advertising, propaganda, comedy). Such appropriations have proliferated as the lofi audience expanded, fuelled by an increase in working and studying from home. Businesses who are aware of this (widely reported) growth and seeking online avenues for promotion have employed lofi mixes and radio stations for commercial benefit. To be fair, the music has always been described as functional music32Amanda Petrusich, ‘Against Chill: Apathetic Music to Make Spreadsheets To’, The New Yorker, 10 April 2019, https://www.newyorker.com/culture/cultural-comment/against-chill-apathetic-music-to-make-spreadsheets-to. – to study, or relax, or chill, or code to – but “to be advertised to” represents a clear intrusion of corporate activity into this cultural space.
This has certainly affected the established community and pushed past traditions aside, although it is unclear where such developments will lead. Since the start of the pandemic, lofi’s music and comment sections have evidently been significant for many people as a social space: for expressing concern about COVID-19; for reassuring others about personal safety; for relaxing while studying; for invoking new identities, feeling like a witch or ghost working away late at night; for moving past sentimental recollections of cartoons; and for blowing off steam by jumping on meme trains, true to form for digital-native culture.
Georgina Born and Christopher Haworth, ‘From Microsound to Vaporwave: Internet-Mediated Musics, Online Methods, and Genre’, Music and Letters 98, no. 4 (2017): 601–47, https://doi.org/10.1093/ml/gcx095, 629.
Adam Harper, ‘Lo-Fi Aesthetics in Popular Music Discourse’ (DPhil Thesis in Musicology, Oxford, University of Oxford, 2014).
Adam Harper, ‘What Is Lo-Fi? A Genealogy of a Recurrent Term in the Aesthetics of Popular Music Production’ (Research Seminar, City University London, 9 December 2020).
Winston and Saywood, ‘Beats to Relax/Study To’, 41.
Even though the mix mentioning “ghost” was live for nine months before the cutoff date, the word appears at least twice as often since March. Furthermore, since both videos are on the same channel, the success of “Lo-fi for Witches” may have helped push more traffic towards the “Ghosts” mix.
Yuval Palgi et al., ‘The Loneliness Pandemic: Loneliness and Other Concomitants of Depression, Anxiety and Their Comorbidity during the COVID-19 Outbreak’, Journal of Affective Disorders 275 (1 October 2020): 109–11, https://doi.org/10.1016/j.jad.2020.06.036; Ben J. Smith and Michelle H. Lim, ‘How the COVID-19 Pandemic Is Focusing Attention on Loneliness and Social Isolation’, Public Health Research & Practice 30, no. 2 (30 June 2020), https://doi.org/10.17061/phrp3022008; Tzung-Jeng Hwang et al., ‘Loneliness and Social Isolation during the COVID-19 Pandemic’, International Psychogeriatrics 32, no. 10 (October 2020): 1217–20, https://doi.org/10.1017/S1041610220000988; Jing Xuan Koh and Tau Ming Liew, ‘How Loneliness Is Talked about in Social Media during COVID-19 Pandemic: Text Mining of 4,492 Twitter Feeds’, Journal of Psychiatric Research, 7 November 2020, https://doi.org/10.1016/j.jpsychires.2020.11.015.
For example, tracks now considered to be vaporwave classics, from albums like フローラルの専門店 (‘Floral Shoppe’) by Macintosh Plus and FINAL TEARS by INTERNET CLUB, are commonly rendered in capitals and include Chinese and Japanese characters. The artist t e l e p a t h テレパシー能力者, first active under that title around 2013, solidifies the trend in vaporwave. These borrowings can be misleading given that ‘lofi’ has different connotations, more akin to ‘DIY’ or ‘non-commercial’, in Tokyo’s contemporary underground jazz and hip hop scenes.
I write this in reference to the single-word titles of two popular lofi mixes: “ＮＯＳＴＡＬＧＩＣ” (15 million views), uploaded by Neotic, and “Ｆｅｅｌｉｎｇｓ” (11 million views) by Cabvno. Note that both were published in 2017, again implying that the emphasis on this stylised typography is an earlier feature of lofi compared to norms in 2020.
Prior to this, The Steven Universe character Connie has been depicted emulating the ‘lofi girl’ artwork. See this image hosted on KnowYourMeme.
This observation does not mean to downplay the solid ties between other forms of hip-hop and anime.
Alexander, ‘Lo-Fi Beats to Quarantine to Are Booming on YouTube’.
Lynn Jamieson, ‘Personal Relationships, Intimacy and the Self in a Mediated and Global Digital Age’, in Digital Sociology: Critical Perspectives, ed. Kate Orton-Johnson and Nick Prior (London: Palgrave Macmillan UK, 2013), 13–33, https://doi.org/10.1057/9781137297792_2, 15.
Winston and Saywood, ‘Beats to Relax/Study To’, 47–48.
Note that “safe” frequently appears in the idiom “stay safe”, which may not be as emotionally charged as, say, “depressed”.
Abel Brodeur et al., ‘Assessing the Impact of the Coronavirus Lockdown on Unhappiness, Loneliness, and Boredom Using Google Trends’, ArXiv:2004.12129 [Physics], 25 April 2020, http://arxiv.org/abs/2004.12129.
Alemoru, ‘Inside YouTube’s Calming “Lofi Hip Hop Radio to Relax/Study to” Community’.
Thanks/complaints to Adam Harper for the link. I assume that the tracks are licensed correctly, though the video uses a Creative Commons Attribution license without any attribution of artist or track names. This subverts a lofi convention to credit artists and song names appropriately (in the vein of hip-hop’s borrowing culture) even where tracks have been curated without permission. For more on lofi as pirate radio, see Jonah Engel Bromwich, ‘Pirate Radio Stations Explode on YouTube’, The New York Times, 3 May 2018, sec. Arts, https://www.nytimes.com/2018/05/03/arts/music/youtube-streaming-radio.html.
In the past, I have joked that I do not trust anyone whose YouTube suggestions do not include lofi hip hop radio – beats to relax/study to. Since the start of the COVID-19 global pandemic, as we (privileged sheltered) have receded to our homes to continue day-to-day life, leisure, and work, we have adopted video calls as a primary means of communication1Simon Kemp, ‘Report: Most Important Data on Digital Audiences during Coronavirus’, Growth Quarters | The Next Web, 24 April 2020, … See full note. With the screen share function coming in handy across all popular video meeting platforms, we have all seen more of each others’ web activity recently. And so, whether heading to YouTube to play an educational video for a class or simply sharing music socially, there has been more opportunity than ever to see who does and who does not boast that iconic image on their YouTube ‘home page’; who has or hasn’t spent time tuned in to the most popular channel playing a kind of music labelled lofi hip hop (thereby determining the ingroup and outgroup of the YouTube lofi hip hop camp). There’s more to be said here about YouTube’s recommendation algorithm, but I’ll put that aside for another day.
Lofi hip hop is by now an established genre taking pride of place in YouTube’s live radio and ‘mix’ categories, the latter term describing a curated playlist of songs uploaded as a single video (usually around an hour or longer in duration). The music has also been referred to as chillhop or referred to as lofi beats, with several variations on spelling like lo-fi and lo fi beats, though lofi is far and away the most common name, sometimes as a standalone label. It’s an ambiguous term, considering the genre relies entirely on digital technologies for both its production (almost always using DAWs) and consumption (streaming video online from YouTube). After all, the inverse term hifi – high-fidelity – was first used to describe technological advances in audio quality2Keir Keightley, ‘“Turn It down!” She Shrieked: Gender, Domestic Space, and High Fidelity, 1948-59’, Popular Music 15, no. 2 (1996): 149–77.. Nonetheless, the people have spoken, and this style of hip hop sounds sufficiently low-fidelity, especially due to analogue-inspired audio imperfections such as white noise, vinyl hiss and crackle, and cassette tape pitch-wavering (most often produced through ‘high-tech’ – which is to say digital – means)3Adam Harper, ‘Lo-Fi Aesthetics in Popular Music Discourse’ (DPhil Thesis in Musicology, Oxford, University of Oxford, 2014), 371..
Method and Scope
This blog post presents some in-progress research into lofi hip hop as a digital-native phenomenon (that is, something born – and continues to thrive – in the digital domain). A fantastic introduction to the genre has been published by Emma Winston and Laurence Saywood4Emma Winston and Laurence Saywood, ‘Beats to Relax/Study To: Contradiction and Paradox in Lofi Hip Hop’, IASPM Journal 9, no. 2 (24 December 2019): 40–54., which I would highly recommend to any uninitiated readers. Here, I focus on the most recent chapter in the life of lofi, examining changes in discourse during the novel coronavirus pandemic. My precise research question is: (how) has discussion on lofi hip hop changed since the start of the COVID-19 pandemic?
This study forms part of a series of posts testing out a word frequency method using the software Mozdeh. For more on methodology, please see my previous post here – note that this time I am comparing two subsets of texts by date rather than texts associated with commenter gender. By studying the words audiences use in commentary on the music in question, I aim to determine the values they hold. The text association comparison follows a standard thematic analysis procedure, which I support with more conventional cultural readings. This dual approach forms part of Digital Flows’ methodological innovation, combining data-scientific methods with close critical analysis. This post is the first of two I’ll write on lofi during COVID-19, since the study revealed several important findings, and I want to keep the post length manageable. Here’s the second part!
The representative sample of lofi hip hop selected for this research comprises 30 YouTube mixes published between 2016 and 2020. Combined, they have half a billion views5482.1 million views as of 17 December 2020, with some rounding. The videos individually range from 3.2 to 51.1 million views, with extreme disparities in the ratios of views to comments. There are approximately 367k combined comments.. Uploaders include ChilledCow, the bootleg boy, Homework Radio and Chillhop Music. Although I borrow its title for this post and it has a considerable number of comments, I excluded “chill beats to quarantine to”, uploaded by actor and musician Will Smith to promote Bel-Air Athletics merchandise. My reasoning was that a video making such specific mention of COVID-19 measures – and posted by a well-known figure – would generate anomalous commentary and add unnecessary noise to the sample. Note that no radio-style, live-streamed YouTube videos were used, because this format disables static comments preserved on the page in favour of a live chat function where comments cross the screen in realtime. As such, text is not as easily accessible or archivable by viewers of the public webpage (instead stored privately, one would expect, by Google).
The mixes were selected for having a high quantity of comments (at least 4k, up to 34k) and a range of musical and semantic signifiers that are associated with lofi hip hop. Given the memes that circulate around the genre title, here are some quick insights on the mix titles: two thirds (20) of them used the word “lofi”, half (14) used “hip hop” and/or “chill”, and around a third (9) made some mention of time (e.g., “2 A.M.”, “late night”). The term “study” was mentioned in a third of titles (9), as was “beats” (8) and “jazz” (8). Five video titles used ａｅｓｔｈｅｔｉｃ text (more on this in the next blog post). As an indicative lofi mix title, Feardog Music captured most of the common keywords with the video “3:30 a.m. ~ lofi hip hop / jazzhop / chillhop mix [study/sleep/homework music]”. These terms have gained cultural popularity to the extent that Australian clothing company Cool Shirtz has taken to printing (and in December 2020 had sold out of) t-shirts with a pastiche of Universal Music Group’s Now That’s What I Call Music compilations.
I collected 97k comments posted prior to March 2020 and 85k comments posted in the subsequent nine months (between 1 March 2020 and 27 November 2020), a period which included government-ordered lockdowns affecting half of the global population6Gerry F. Killeen and Samson S. Kiware, ‘Why Lockdown? Why National Unity? Why Global Solidarity? Simplified Arithmetic Tools for Decision-Makers, Health Professionals, Journalists and the General Public to Explore Containment Options for the 2019 … See full note. Although provinces in mainland China announced quarantine measures in January and the WHO did not declare COVID-19 a pandemic until 11th March, many major countries launched lockdowns throughout March. 1st March was chosen as the cutoff point as a neat division between pre-pandemic conditions and the current period, which includes a number of ongoing public health measures. Given that YouTube is an American platform, lofi hip hop is predominantly communicated and commented on in English, and it is fair to assume that the audiences generally reside in English-speaking countries and other territories with high levels of internet connectivity and streaming platform use (USA, India, UK, Germany, France).
The prominence of the proposed function – beats to study/relax to – is worth some attention. There has been widespread commentary on the implications of this elision, as though studying and relaxing are interchangeable activities (or at least benefit from similar accompaniment)7Adam Harper, ‘What Is Lo-Fi? A Genealogy of a Recurrent Term in the Aesthetics of Popular Music Production’ (Research Seminar, City University London, 9 December 2020); Michael Wu, ‘What Are Lofi Hip Hop Streams, and Why Are They So … See full note. Moreover, the conversations about homework and academic expectation that play out in the live chat indicate a large student audience attempting to relax while studying, playing the music as a means of enhancing periods of high-pressure and anxiety-provoking academic work. Rather than high-octane music used for ‘motivation’, lofi purports to offer the opposite. It resolves to calm the listener, as the pairing of study/relax alludes to seeking a more relaxed form of studying. Lofi hip hop is clearly a tool employed effectively by students while working, a practice which may have proliferated when COVID-19 caused many educational institutions, especially colleges and universities, to transition to virtual forms of engagement8Olasile Babatunde Adedoyin and Emrah Soykan, ‘Covid-19 Pandemic and Online Learning: The Challenges and Opportunities’, Interactive Learning Environments (2 September 2020): 1–13, https://doi.org/10.1080/10494820.2020.1813180.. It is worth examining how lofi’s student audience has been affected.
Essentially, I was interested to see whether discourse on lofi hip hop has changed since the pandemic became a fact of everyday life, so I compared the texts posted before March 2020 with those appearing after in order to determine statistically significant differences in the proportions of words used in comments. Reading the terms that had substantially increased and declined in popularity revealed a variety of changes in the contexts of listening, conventions of lofi discourse, and emotional expressions. This post focuses on the first of these: how listeners discussed the circumstances of their engagement with the videos. In a Dazed article profiling the genre, Kemi Alemoru points out that the collective sentiments of the YouTube comments on lofi mixes are (by YouTube’s standards) “uncharacteristically friendly” and that commenters form a “therapeutic community”9Kemi Alemoru, ‘Inside YouTube’s Calming “Lofi Hip Hop Radio to Relax/Study to” Community’, Dazed, 14 June 2018, https://www.dazeddigital.com/music/article/40366/1/youtube-lo-fi-hip-hop-study-relax-24-7-livestream-scene.. Similarly and more specifically, Winston and Saywood note the prevalence of listeners sharing “confessions and worries”, as well as the “relative anonymity and consequence-free interactions between strangers” that facilitate acts of “interpersonal care”10Winston and Saywood, ‘Beats to Relax/Study To’, 50.. Given that the comment sections are typically emotionally expressive, interactive, and mutually supportive, and that the pandemic has prompted significantly more use of online spaces for interpersonal communication, I was keen to find out whether such conventions would be altered by the pandemic conditions, and if so, in what ways. Here’s what I discovered.
Themes: Listening Context
It should come as no surprise that there were many more discussions related to COVID-19 during the pandemic than there were prior to its onset. The terms “quarantine”, “corona”, “covid”, “virus”, “pandemic”, “coronavirus”, “lockdown”, and “covid-19” appeared in the sample with extreme statistical significance (in order from most to least significant: “covid-19” appears about a tenth as commonly as “quarantine”). There were hardly any mentions of each of these terms before 1 March 2020 – “covid”, “pandemic”, and “lockdown” do not appear even once – compared to hundreds of uses in the subsequent period. The most popular term, “quarantine”, emphasises the social effects and personal implications of the pandemic, generating at least 500 comments in which listeners report their specific circumstances11All example texts are paraphrased, or cut up and rearranged, to preserve poster anonymity. I rarely introduce entirely new words so that a faithful portrait of the sentiments expressed in lofi commentary is maintained.:
‘This made me think of better times before the quarantine’
‘Listening to this in quarantine’
‘I’m so lonely now my city’s gone into quarantine’
The melancholy tone is characteristic of lofi discourse, with a notable presence of nostalgia. A number of comments express gratitude for the music bringing comfort or improving relaxation during lockdown periods. There were relatively few comments on the reasons for quarantine, such as preventative measures or discussion of the virus itself, which suggests the audience to be accepting the orders as a fact of contemporary life (though the occasional “fuck quarantine” does appear). More technical terms like government, state, and transmission did not appear in statistically significant proportions of comments before or after March.
The word “stay” saw a large increase in usage (more significant even than “covid-19”)12When I read this back, it seemed erroneous, so I checked the stats. “Covid-19” previously appeared in 0% of texts (0 mentions) and appears in 0.0686% of texts after 1 March (58 mentions). “Stay” jumped from 0.9457% (921) to 1.3821% (1168), … See full note, likely because it carries several meanings relevant to the pandemic. It has long been a word in common use across lofi comment sections, however: the phrase “stay awake” was often posted by tired students and “stay alive” was advocated by emotionally supportive participants. During the pandemic, the phrases “stay safe” – either as a genial expression or personal desire – and “stay inside” – as a recommendation or received instruction – have also become prevalent. Notably, these uses “stay” are fairly conversational expressions, implying commenters responding to one another. In the following blog post, I’ll make some more in-depth inferences about how interpersonal discussions have changed during the pandemic. In any case, I should note that “stay” has a very broad range of uses beyond the few common ones I’ve highlighted, frequently appearing when listeners report their current circumstances or state of mind more generally, as in “still staying at home”, “I just want to stay in bed”, or “trying to stay chill”.
Studying and working
As well as positioning oneself in the specific circumstances of the pandemic, lofi listeners increasingly report periods of study accompanied by the YouTube mixes. A range of associated terms speak to the pervasiveness of studying while listening: “online”, “doing”, “module”, “assignment”, “study”, “finished”, “homework”, “studying”, “session”, and “classes”. Although lofi videos have customarily advertised the benefits of ‘beats to study to’, this appears to have become much more common during the pandemic, as listeners increasingly discuss putting music to this very use. While “doing” would seem too broad to consistently indicate education, it almost always occurs in relation to another word in this theme, especially “online school” or “online class”. This makes sense. Many students are working from home during lockdowns and have unprecedented control over their background music: they might enjoy a lofi mix while on a video call for a lesson or lecture, for instance, a privilege likely denied in a physical classroom. The term “school” has only become ever so slightly more popular, and is not even close to the benchmark for statistical significance13All terms quoted as significant in this article use p≤0.001., evidencing the novelty and ubiquity of “online” learning in particular.
With education primarily taking place remotely, there is a lot of “homework” done (or at least reported) by lofi’s listener base. Granted, this was conventional for lofi pre-pandemic, though the public health measures find commenters making more mention of it than ever before. I was struck by the personal organisation and time management implied by some of the terminology, as people frequently write about a “module” or “session” (supposedly commenting after study to mark their completion of a study period – high levels of discipline, too!). The use of such formal time divisions invites further analysis. Presently, there might be a popular conception that remote school and university learning enables students to pay less attention and get by without putting in much effort. For instance, The Center for Research on Education Outcomes at Stanford University released a report on “Estimates of Learning Loss” and USA Today recently reported “alarming numbers of kids falling behind, failing classes or not showing up at all”14Erin Richards, ‘Students Are Falling behind in Online School. Where’s the COVID-19 “disaster Plan” to Catch Them Up?’, USA Today, 14 December 2020, … See full note. Putting aside how these institutions undoubtedly amplify the pressure placed on students to meet arbitrary benchmarks, it is worth noting that concerns about cyber-slacking long precede the widespread rush to distance education15Li Li and Scott Titsworth, ‘Student Misbehaviors in Online Classrooms: Scale Development and Validation’, American Journal of Distance Education 29, no. 1 (2 January 2015): 41–55, https://doi.org/10.1080/08923647.2015.994360; Aakash Taneja, … See full note.
However, instead of learning less, the students using lofi hip hop are more expressive than ever about the work they are getting done in this period: the term “module” had 3 total references prior to March compared to 152 mentions after. This might simply emphasise the novel structure of virtual learning environments, but it is still a striking increase indicating where commenters’ attention is directed. At least twice as many commenters took to the comments to report being “finished” with a chunk of work during the first 9 months of the pandemic compared to the three years prior. For what it’s worth, I don’t have any evidence to suggest a causal link between the use of lofi and actual student productivity, although it seems intuitive that students would benefit from the ability to customise their environment to suit their learning preferences. This applies at least to those with the privilege of home computer access and who submit their thoughts to the public forum to celebrate (for example) “another online module finished!”.
The lofi audience, at least, appears to be taking more responsibility for their own learning, perhaps due to anxieties about ‘keeping up’ with class. There is a clear precedent for this commitment to productivity and schoolwork (or other educational labour) in the community, which Winston and Saywood associate with the social frame of Jonathan Crary’s “24/7 capitalism”16Jonathan Crary, 24/7: Late Capitalism and the Ends of Sleep (London: Verso Books, 2014).. They describe lofi hip hop as a fitting soundtrack to a (principally U.S.) context where the time spent in school and doing homework has continuously increased since the 1980s, and where periods of time formerly used for leisure and relaxation (especially, I would note, evenings) have been filled with more work17Winston and Saywood, ‘Beats to Relax/Study To’, 49.. As distance learning boasts various capacities for surveillance – consider Zoom’s controversial attention monitoring feature18Zoom removed this feature in April, although competitors like GoToWebinar retain it., or virtual learning environments’ activity completion tracking – it is clear to see how greater demands for work are regulated, a convention likely intensified by the pandemic, as manifested in the lofi comment sections with its abundant references to “doing”, “homework”, “studying”, and “classes”. That said, the organised implications of a study “session” or “module” that I raised above imply that students are successfully putting boundaries around periods of study accompanied, of course, by lofi beats.
Time and place
If quarantine and studying have largely been occupying the minds of commenters in the last 9 months, it follows that there are fewer references to other distinct times and places. Indeed, the terms “night”, “Monday”, and “summer” were much more frequent prior to the pandemic. The declining use of the term “night” might be surprising in the face of cultural commentary pointing out how uncanny the passing of time has felt and how poor sleep practices have been exacerbated since March, not to mention ongoing research into sleep quality as a public health concern. Taken together, these perspectives imply that more individuals are awake at nighttime (even if they are looking for ‘beats to sleep to’). However, lofi’s audience has significantly reduced reports of listening late at night (e.g., ‘It’s 1am here, a perfect night to forget about everything’), and on Mondays, in line with the declining popularity of the 2017 video titled “early monday mornings”.
Memes exemplifying the general mood of timelessness and screen use that imply declining sleep hygiene.
There has also been less reflection on “summer”, despite this word appearing in many nostalgic texts both before and during 2020, (e.g., posts to the effect of ‘this takes me back to the summer of 2015’). It would be fair to assume that listeners’ memories of simpler and more innocent times might be more common now, but this is not reflected in the commentary. A simple explanation is that attention is so focused on COVID-19 and its associated terms that the pandemic takes precedence in commenters’ thoughts: for instance, a user might write “I’m still awake listening in lockdown” rather than “I’m still awake listening late at night”. Interestingly, the word “during” appears much more frequently since March 2020, and it works in context with many of the terms that have appeared in this section: ‘listening during quarantine’, ‘on in the background during my classes on zoom’, ‘this kept me going during hours of homework’. Such comments appear to serve a purpose of recording current feelings and/or diarising, with sentiments frequently shared by the wider community. These range from the trivial – essentially, ‘who else is using these “beats to study to” to study to?’ – to the strikingly specific, where one finds microfiction like:
‘these comments are giving me loads of comfort during a really shit time. You are all lighting up this dark place I feel like I’m in. I’m supposed to be studying for my math exam but I couldn’t stop crying. Now it’s 2am and all I can hear is this calming music, and it’s putting me in a much better mood’.
Jotting down thoughts like this can be considered a form of digital lifelogging19Stefan Selke, ed., Lifelogging: Digital Self-Tracking and Lifelogging – between Disruptive Technology and Cultural Transformation (Wiesbaden: Springer Fachmedien, 2016), https://doi.org/10.1007/978-3-658-13137-1_3., a particularly significant practice during the exceptional circumstances introduced by COVID-19. Comments that position individual expressions in time and place act as a kind of chronicling, accruing to a record of cultural responses to the pandemic. Individuals in the lofi audience seem eager to acknowledge their environments and publicly attest to their living through this notable period, often in the form of questions (‘who else is listening in quarantine?’). Perhaps this kind of expression acts as self-affirmation or aims to stave off loneliness. More on this in the subsequent post!
This blog post analysed a sample of around 182k comments on lofi hip hop YouTube mixes to determine changes in commentary during the first 9 months of the COVID-19 pandemic. Between March and December 2020, commenters more often described the current contexts of their lofi listening (compared to before the pandemic), emphasising quarantine/lockdown circumstances and the use of lofi mixes while attending online school and doing homework . References to other specific times, such as “night” and “summer”, reduced in this period as the social effects of COVID-19 took centre stage. The second half of this post presents other themes, analysing lofi as a genre and community – as revealed by differences in commentary prior to and during the pandemic – in addition to considering the changing role of emotional expression among the lofi audience.
Keir Keightley, ‘“Turn It down!” She Shrieked: Gender, Domestic Space, and High Fidelity, 1948-59’, Popular Music 15, no. 2 (1996): 149–77.
Adam Harper, ‘Lo-Fi Aesthetics in Popular Music Discourse’ (DPhil Thesis in Musicology, Oxford, University of Oxford, 2014), 371.
Emma Winston and Laurence Saywood, ‘Beats to Relax/Study To: Contradiction and Paradox in Lofi Hip Hop’, IASPM Journal 9, no. 2 (24 December 2019): 40–54.
482.1 million views as of 17 December 2020, with some rounding. The videos individually range from 3.2 to 51.1 million views, with extreme disparities in the ratios of views to comments. There are approximately 367k combined comments.
Gerry F. Killeen and Samson S. Kiware, ‘Why Lockdown? Why National Unity? Why Global Solidarity? Simplified Arithmetic Tools for Decision-Makers, Health Professionals, Journalists and the General Public to Explore Containment Options for the 2019 Novel Coronavirus’, Infectious Disease Modelling 5 (1 January 2020): 442–58, https://doi.org/10.1016/j.idm.2020.06.006.
Adam Harper, ‘What Is Lo-Fi? A Genealogy of a Recurrent Term in the Aesthetics of Popular Music Production’ (Research Seminar, City University London, 9 December 2020); Michael Wu, ‘What Are Lofi Hip Hop Streams, and Why Are They So Popular?’, Study Breaks, 2 December 2018, https://studybreaks.com/culture/music/lofi-hip-hop-streams-popular/; Winston and Saywood, ‘Beats to Relax/Study To’, 49.
Olasile Babatunde Adedoyin and Emrah Soykan, ‘Covid-19 Pandemic and Online Learning: The Challenges and Opportunities’, Interactive Learning Environments (2 September 2020): 1–13, https://doi.org/10.1080/10494820.2020.1813180.
Winston and Saywood, ‘Beats to Relax/Study To’, 50.
All example texts are paraphrased, or cut up and rearranged, to preserve poster anonymity. I rarely introduce entirely new words so that a faithful portrait of the sentiments expressed in lofi commentary is maintained.
When I read this back, it seemed erroneous, so I checked the stats. “Covid-19” previously appeared in 0% of texts (0 mentions) and appears in 0.0686% of texts after 1 March (58 mentions). “Stay” jumped from 0.9457% (921) to 1.3821% (1168), which is a more statistically significant jump according to the Bejamini-Hochberg procedure.
All terms quoted as significant in this article use p≤0.001.
Li Li and Scott Titsworth, ‘Student Misbehaviors in Online Classrooms: Scale Development and Validation’, American Journal of Distance Education 29, no. 1 (2 January 2015): 41–55, https://doi.org/10.1080/08923647.2015.994360; Aakash Taneja, Vincent Fiore, and Briana Fischer, ‘Cyber-Slacking in the Classroom: Potential for Digital Distraction in the New Age’, Computers & Education 82 (1 March 2015): 141–51, https://doi.org/10.1016/j.compedu.2014.11.009.
Jonathan Crary, 24/7: Late Capitalism and the Ends of Sleep (London: Verso Books, 2014).
Winston and Saywood, ‘Beats to Relax/Study To’, 49.
This blog post examines YouTube comments on ‘WAP’, performed by Cardi B and featuring Megan Thee Stallion. The aim is to demonstrate some of the data collection and thematic analysis methods that inform the Digital Flows research project, so it can be considered a small pilot study of sorts. In what follows, I’ll introduce the track with some brief context, then discuss the methods I used, explaining their benefits as well as various limitations. The second half of this piece describes seven major themes that emerged in my analysis of comments by male and female users, pointing out gendered differences in listener/viewer responses. These comments indicate the different values that men and women express in their appraisal of ‘WAP’. As this is a blog post, I freely combine analysis and discussion, although I encourage you to consider other conclusions to the ones I make. This work might form part of future publications, and the post will persist here akin to a pre-print (in the spirit of Open Science). I’ll share my dataset too.
‘WAP’ is an interesting choice for analysis because it is one of the most prevalent hit songs of 2020 in mainstream rap as well as broader pop culture. Over the last couple of months, it is perhaps the most discussed song in what I think of as the ‘online rap mainstream’, a virtual public space for discourse on rap music and culture. Its chart success speaks to its popularity: it debuted at #1 on the Billboard Top 100 and it spent 3 weeks at #1 on the UK official charts. Its official music video has 289 million views as of 20th November 2020. These stats aside, it has broader significance and spread on social media, such as having sparked a TikTok dance trend and various memes. Although the song received substantial critical acclaim – let’s face it, it’s a bop – it also drew comments from conservative media personalities, including Candace Owens and Ben Shapiro, in the run-up to the 2020 U.S. presidential election. It warrants examination of gendered differences in commentary as it is an extremely popular song performed by two women in a historically male-dominated field (Lafrance et al. 2011; Strong and Raine 2019), and it has been received as overtly sexual (e.g., McClinton 2020; Holt 2020). ‘WAP’ has prompted discussions about expressions of female sexuality, sex-positive feminism, and empowerment, the latter being a term which my forthcoming monograph How Music Empowers (Routledge, 2021) tackles directly. Here’s the YouTube link:
These notes on method take up a substantial portion of the post, because I’m still early on in the project, and part of what Digital Flows offers is methodological innovation in the use of digital ethnography for studies of popular music and digital culture. Hence, a lot of what I’m working out here is quite crucial to the research practices going forward, as I try to develop robust methods that take advantage of digital tools for the study of online data. If you’re not interested and just want to read the discussion of results, you can click here to skip ahead.
To undertake data collection and analysis, I used the free program Mozdeh, developed by Professor Mike Thelwall at the University of Wolverhampton. I also followed the WATA (Word Association Thematic Analysis) method described in his forthcoming book on using Mozdeh for such purposes. (Thanks to Mike for the pre-publication copy). This is a two-part method which proceeds according to an initial stage of quantitative statistical analysis followed by a stage of qualitative thematic analysis: an objective, automatic calculation of the proportions of given comments and their statistical analysis, then subjective and context-sensitive interpretation of the quantitative findings.
As for the sample, I collected 126,345 comments from three YouTube pages: the official video, the official audio, and a third-party lyric video of the song. Although the primary focus is the single and its accompanying music video, using the additional videos ensured enough comments were collected in total to achieve a decently sized set of texts from which to draw statistically significant findings: Thelwall recommends 100,000 as a baseline. Naturally, some comments only relate to the audio and some to the video, although the pilot data collection showed that generally similar topics emerged in each comment section. The majority of texts in the sample were drawn from comments on the official video. Only one comment per user was collected in order to cut down on spam and avoid skewing the sample towards the words used by prolific commenters.
The identification of commenter gender is perhaps the largest limitation of the study, but practical steps are taken to reduce the difficulties such analysis presents. Mozdeh infers participant gender by matching a commenter’s first name as used in their YouTube username to a comprehensive list of male- and female-associated names. This is not entirely adequate, but nor is it worthless given that a substantial proportion of YouTube users do indeed use a name on their public YouTube profile that associates accurately with the gender with which they identify: most likely, their preferred real name. (No erasure of non-binary and other-gendered folks is intended by this method, and it is worth noting that Mozdeh can identify non-binary authors on Twitter by reading gender-neutral pronouns, e.g. they/them, in their Twitter profile description.) The large size of the dataset slightly reduces the problematic nature of gendering by username. From the sample of 126k comments, 12.3% can be associated with male commenters, and 16.6% can be associated with female commenters. This doesn’t necessarily mean more women than men commented on the video. After all, the gender of 71.1% of the commenters (from Max Smith to CardiBFan123, approximating some usernames) could not be reasonably assumed, and estimates of the gender ratio in that majority would be extremely speculative.
In my discussion, I will use the term ‘men’ and ‘women’ to refer to YouTube commenters using commonly recognisable male or female first names as their usernames, to avoid harmful associations with ‘females’ when used as a noun. However, ‘male’ and female’ is the more precise assumption being made: there is no age information collected (thus technically distinguishing, e.g. a ‘boy’ from a ‘man’) and, although YouTube has minimum age policies, I do not verify for age. At the time of writing, the video is not marked as explicit and does not require login or age confirmation on YouTube (though the music video uses the ‘clean’ audio version). I use the word ‘song’ to encompass the multimediated artefact that is the recorded track and its accompanying music video, in line with commenters’ common use of the word ‘song’ to refer to both the audio and visual content. (The word ‘song’ is used around three times more often than ‘video’, even in reference to visual elements). A final reiteration of the methodological assumptions about gender in the discussion that follows: a phrase like ‘men tend to X’ stands in for ‘people who have posted YouTube comments using male-associated first names in their username tend to X’. From here onwards, I’ll elide that distinction. Let’s get into it!
Writing And Posting: Discussion on ‘WAP’
There are seven main themes I will draw out, presented roughly in order from the most to the least significantly gendered differences.
This label refers to personal opinions posted on the song, aligning with the general purpose of YouTube comments (alongside assisting Google’s data harvesting & surveillance practices). Compared to women, men overwhelmingly share negative views of the song, using the terms ‘trash’, ‘garbage’, and ‘worst’. ‘Trash’ and ‘garbage’ are two of the most statistically significant differences between men and women commenting on the song. Note that these words are not as associated with personal feeling as, for example, ‘bad’, ‘boring’, or ‘annoying’, but demean ‘WAP’ in terms of objective value. They attack the worth of the song. It can be inferred that, by comparison, women less often undermine the value of the song in itself, and are more likely to take issue with specific elements, or interpret negative responses as their individual opinion. In other words, women are less likely than men to dismiss ‘WAP’ as a piece of work altogether.
Men also make broader (false) claims about the song as representative of cultural decline, using the terms ‘society’, ‘humanity’, and ‘porn’. To paraphrase for context (preserving anonymity), texts like ‘this is what is wrong with our society’ or ‘this song makes me lose my faith in humanity’ appear much more commonly in the sample of male-associated comments. This trend is a ‘slippery slope’-style expansion of the commentary in the previous paragraph, whereby men are more likely to describe their dislike of the track in terms of an objective social ill rather than their personal view: ‘WAP’ both is worthless and represents a real social problem. Perhaps unsurprisingly given the surge in publicly broadcast conservative criticism in the last decade or so, the words most closely associated with ‘society’ are ‘our’ (whose exactly?), ‘decay’, and ‘western’. I should clarify that this is an extremely small proportion (0.2%) of all comments. In any case, at least twice as many men mention ‘society’ than do women.
Women express significantly more enjoyment of the song, especially using the word ‘love’. They also describe their surprise more often, disproportionately employing the term ‘omg’. Such a finding aligns with media reportage on the song, where journalists have described ‘WAP’ as “truly provocative” (Haider 2020) and “among the filthiest things I’ve ever seen in mainstream American popular culture” (Rosenberg 2020). Thelwall (2018) has found ‘omg’ to be strongly female-associated in other studies. It may therefore not be the case that women are more surprised by the video than men, but that they are more likely to express their surprise using that specific term. Women are also more likely to consider other peoples’ views more, especially that of their ‘mom’, and make more frequent reference to ‘man’ (usually as in ‘my man’). The reference to ‘my man’ in relation to the song makes sense given that ‘WAP’ spotlights sentiments of female (cishetero)sexual desire. That said, ‘man’ also has frequent usage by all commenters as an exclamation or for generic emphasis, like ‘gee!’.
While most comments are written in English, there is a considerable portion of comments written in Portuguese, and a smaller body of texts in Spanish. Common words in Portuguese and Spanish sentences including ‘que’, ‘eu’, ‘la’, ‘de’, ‘essa’, ‘esta’, ‘o’, ‘q’ (que), and ‘en’ are largely used by commenters with female-associated first names. The gendered difference here appears to evidence Cardi B’s large female Brazilian fanbase, which seems plausible considering the artist’s well-known collaborations with Latin American artists like Anitta and Bad Bunny. By contrast, men write more sentences using common words in English (‘this’, ‘is’, ‘of’, ‘to’, ‘an’, ‘a’, ‘the’, and ‘with’). The exceptions to a male preponderance of English sentence terms are ‘my’ and ‘I’, both of which women write more in English. From this, we can infer women describe their personal responses more frequently in comments (and this maps on to the earlier discussion of women appraising in terms of opinion, by contrast to men appraising in terms of objective quality). Indeed, the second most statistically significant gap overall (of all words) is for the term ‘this’, suggesting men implicate themselves far less in their comments about the song (writing, for instance, ‘this is great’ rather than ‘I love this’).
There are pronounced gender differences in references to specific public figures. In general, men write more names of recognisable people. However, this trend swings significantly in the other direction for Kylie Jenner, whom women reference considerably more. Women mentioned ‘Kylie’ 749 times compared to men at 281 times, with a similar though narrower gap for ‘Jenner’. There is support for two obvious conclusions here, which are that (a) Kylie Jenner has a larger female audience and is recognised more frequently by women and (b) that Kylie Jenner’s performance in the ‘WAP’ video is especially worth commentary among women. The gap for ‘Jenner’ closes somewhat because men tend to write full names of public figures (i.e., ‘Cardi B’ rather than ‘Cardi’). Though slightly speculative, this observation implies that women (who are more likely to write only ‘Cardi’) feel more personally connected to the artist, which suggests a stronger parasocial relationship with female consumers.
Of the performing artists mentioned in the comment section, Cardi B is unsurprisingly mentioned most (with no statistically significant difference by commenter gender), followed by Kylie Jenner (whom significantly more women cite), then Megan Thee Stallion, Nicki Minaj, and Anitta. Only the first three of these feature in the video, and other featured performers, including well-known artists Rosalía and Normani, are lower down this list. Their less remarked-upon presence, despite critical acclaim and vast audiences, might speak to the stranglehold and cultural dominance of Kylie Jenner as a popular figure. However, she appears in the video for a full 25 seconds, interrupting the song altogether, whereas the others share around 10 seconds of screen time as the beat plays out at the end, so the video emphatically spotlights her presence. None of the featured performers except for Cardi B and Megan Thee Stallion are listed in the YouTube credits, though this is reasonably conventional for pop music videos.
Mentions of Nicki Minaj are particularly interesting given she has no involvement in the song or video, but is a major ‘competitor’ of Cardi B and Megan Thee Stallion, with a passionate fanbase who appear to express loyalty by insulting artists who are deemed rivals. The director of the ‘WAP’ video, Colin Tilley, also directed Nicki Minaj’s 2015 video for ‘Anaconda’, a song which bears a number of similarities: it also featured prominently in popular culture, was received as ‘sexually explicit’ and, like ‘WAP’, its beat is developed around a male voice sample from early 1990s hip-hop/dance music. Anitta most often appears in comments mentioning her song featuring Cardi B (and Myke Towers), ‘Me Gusta’, which was released in the same period as ‘WAP’. Given how many of these posts instruct readers to listen to ‘Me Gusta’ and announce the release date and performers with no other commentary, I suspect a large presence of bots for this particular trend, or perhaps a fan campaign to synergistically create awareness of the ‘Me Gusta’ video.
Men write more male names (and more names in general) than women do. For instance, Tory Lanez is exclusively referenced by men in the sample. Around the time of the release of ‘WAP’, the artist was reported to have shot Megan Thee Stallion and, in reference to this event, ‘shot’ also appears statistically significant, at 32 male to 5 female mentions. Men also reference more conservative commentators: ‘Ben’ appears with a wide gap in commenter gender (as does ‘Shapiro’), with a narrower but still significant gap for ‘Owens’ (usually accompanied by ‘Candace’, but occasionally misspelled ‘Candice’). These two figures made public statements in response to ‘WAP’, and evidently became part of the cultural conversation surrounding the track (especially in men’s subsequent commentary).
4. Audience and listening context
Women disproportionately mention TikTok in their comments on the video. There appears to be little consensus on whether the platform name is one or two words, with both ‘Tik’ and ‘Tok’ more frequent among female comments, as well as the combined ‘TikTok’. Many such comments also use ‘dance’ to reference TikTok dance trends, especially the routine inspired by ‘WAP’. For instance, several users linked the timestamp for ‘the WAP TikTok dance’, pointing to a section of the music video that has become popularly recreated on the short-form video-sharing platform. It can be inferred that this is a phenomenon generally more associated with female participants, at least in their public postings about the track.
There is also much more frequent reference to ‘brasiliero’ (Brazilians) by women, often to point out the prevalence of this audience either listening to the song or virtually present ‘in the comments’. By contrast, men more commonly used the word ‘people’, perhaps (with reference to my earlier discussion about men’s tendency towards objectification) generalising broad responses to the song as a way of expressing their own views. Many such comments suggest disbelief, though they are not always negative: (paraphrased) examples like ‘why do people like this?’, ‘I can’t believe people get upset about this’, and ‘remember when people said women couldn’t rap?’ abound.
5. Censorship and comparison
There is a statistically significant gender difference in the use of the terms ‘cold’ and ‘outside’. The context for these words is the song ‘Baby It’s Cold Outside’, which became the subject of intense social media discussion in the run-up to Christmas in 2018, after a handful of radio stations removed it from their seasonal playlists citing concerns about the lyrics in the context of the MeToo movement (which have since returned it to regular play). The backlash was substantially more heated than the initial decision to question the lyrics that evoke male sexual predation (and, perhaps more ambiguously, the original context of the line ‘what’s in this drink?’, either implying date rape drugs or ‘merely’ joking about alcohol intoxication). The charge of political correctness and opposition to censorship that emerged in late 2018 are reinvoked in responses to ‘WAP’, where overwhelmingly male commenters question the decision to censor ‘Baby It’s Cold Outside’ yet ‘allow’ the sexual content of ‘WAP’. This trend clearly engages with discourses of women’s sexual expression and the feminist critique of rape culture. Essentially, it appears to be more about ‘Baby…’ than ‘WAP’ per se, with a continual spilling over of outrage claiming the innocence of the popular Christmas song. It can be assumed that the ‘explicit content’ of ‘WAP’ feels less permissible than the casual misogyny of ‘Baby…’ for many male users.
Besides contrasting ‘WAP’ and ‘Baby…’, men generally use more comparative terms in their comments. One tendency is to compare the performers directly, such as in the comment ‘Megan’s verse is better than Cardi B’s’. As well as ‘better’, there is a greater prevalence of the terms ‘YouTube’, ‘has’ and ‘than’ in men’s comments, which often draw connections between the song itself and broader changes, in a similar vein to the ‘cultural decline’ comments discussed above. These are typically used as a means of interpreting the song’s value, with a variety of expressions including outrage (‘how has this become popular?’), commentary on genre (‘hip hop has really changed’), and further censorial views (‘why has YouTube not removed this?’, ‘this is dirtier than porn’). Men also make more generic comparisons about ‘WAP’, such as ‘this is better than Cardi B’s last song’.
This category comprises the ugliest terms used with statistically significant gendered differences. Men write more misogynistic comments, though it should be noted that the overall proportion of explicitly misogynistic comments is thankfully very small: certainly less than 1% of the entire sample, and possibly as low as 0.5%, i.e. 1 in 200 comments. This says little about the implicit misogyny of many of the comments on the video, however, and it is perhaps worth mentioning that no male humans feature in the video, so all comments are directed purely at audiovisual content featuring women (notwithstanding the Frank Ski voice sample). The misogynistic terms men used disproportionately in comments were ‘smell’, ‘std’, and ‘plastic’. Men were 15 times more likely than women to write ‘smell’, referring to vaginal odour (and women’s body odour more broadly). This is notable given that the lyrics rarely mention smell, focusing instead on sensory aspects of touch and later taste, though Cardi B’s verses offer the occasional olfactory suggestion: ‘swipe your nose like a credit card’ and ‘a weed smoker’. (‘Macaroni in a pot’ evokes sound rather than smell, at least in its reference to a popular meme.)
Furthermore, men more frequently suggested that the performers have sexually transmitted diseases. It is reasonable to conclude that men explicitly mention sex much more often (so the song’s affordances of cisheterosexual desire are evidently landing) although women may allude to sex more commonly, as in comments describing ‘my man’, discussed above. For a song that takes sex quite unambiguously as its focus, relatively few comments (estimated less than 5%) are explicitly sexual. Although men were still much more likely than women to use the word, the gap closes slightly for mentions of the term ‘plastic’, most often referring to (posited) plastic surgery or the otherwise fake appearance of the performers.
A short observation to close this discussion out: men slightly more often make jokes about the song title. The gist of comedic comments is that users stumbled upon the ‘WAP’ video while searching for instructions about a ‘Wireless Access Point (or Protocol)’ and were surprised by the song instead. The ambiguity of the acronym evidently invites such attempts at humour, with which men are more likely to engage, using the terms ‘wireless’ and ‘protocol’ (although ‘point’, a more multivalent word, does not appear with statistically significant gender difference). Note also that, because the word ‘WAP’ in itself does not invite censorship, the song does not require an alternate title for the ‘clean’ version (even though the ‘clean’ chorus lyric, one would assume, calls for ‘WAG’). Moreover, I might speculate that such comments indicate slight discomfort with ‘the real’ meaning of ‘WAP’, in that they make the user’s sensibilities the subject of attention (as in, ‘I just wanted to know how to use a wireless access protocol…’), although they are just as likely to highlight the inherent ‘shock factor’ of the song, given cultural constraints on women’s sexual expression.
Comments collected on Cardi B’s ‘WAP (feat. Megan Thee Stallion’ demonstrated statistically significant gender differences between male and female commenters. Across seven main themes, men were much more likely to attack the song’s objective worth while women far more often wrote ‘love’, women tended to write more Portuguese and Spanish sentences while men wrote more common sentence words in English (save for a female preponderance of the terms ‘my’ and ‘I’), men named specific public figures more often (except for Kylie Jenner whom appeared disproportionately in women’s comments), women were more likely to mention TikTok dance trends, men more frequently decried the censorship of ‘Baby It’s Cold Outside’ given the ‘explicit’ nature of ‘WAP’, and more overtly misogynistic comments were written by men, as were jokes about the song title being an acronym for something else.
In the spirit of making research more collaborative, iterative, and responsible, I’d be grateful for comments on any aspect of this blog post, whether about the form or methods or discussion. The research methodology is still in its formative stages, and gender analysis is only one part of Mozdeh’s capabilities, not to mention other tools for data analytics that might be used down the line. Would you be interested in reading more things like this? What’s lacking (or overrepresented) in the analysis? Please feel free to get in touch with any thoughts. Thanks for reading!
This 32-minute video is a recorded talk given in the UCC Department of Music (opening the academic year’s Speaker Series) on 22nd October 2020.
It serves as an introduction to the project for an academic audience in music studies. The talk covers three things. First, it reviews how music scholars have approached the Internet in the past. Next, it defines ‘the online rap mainstream’, describing what the social web looks like for rap music in the present day. Finally, it explains how Digital Flows plans to analyse digital-native hip-hop culture, with its groundbreaking methodology combining data-driven analysis with culturally-informed digital ethnography.
Welcome to the project website. Throughout the lifespan of the research programme, I am keen to provide updates on progress, describe the methodology, and mark particular milestones. This site will be used for such blog posts. I also plan to explain how the research unfolds and where data can be accessed, in the spirit of open science.
Publications emerging from this project will be described and publicised here along with any other outputs, such as conference papers and online materials.
There is also a plan to host a conference (likely in early 2022) on the broader theme of popular music and digital culture.