Mark Fisher and Anti-Hauntology

In my previous article I claimed that recently departed artist/producer SOPHIE marked a significant shift in moving contemporary popular music away from Mark Fisher’s idea of hauntology. I argued that progressive pop producers such as SOPHIE and Arca were moving the whole idea of popular music in a direction it a new direction and that the combination of sounds they use represent a significant break from the hauntological past. I termed this anti-hauntology.

The article sparked an interesting discussion online in which a few commenters rose up to defend Fisher against claims of anti-hauntology in pop music. As one redditor claimed: “Fishers point is precisely that 20th century genres weren’t just a combination of current sound. They were radical, qualitative steps forward that didn’t resemble anything current.” His argument was that, much as Fisher claimed, a genre such as jungle (and it’s various incarnations) represented a truly qualitative shift that hasn’t been replicated since. The twenty-first century had been left to repeating old tropes and recombining them in various ways which has left us stuck with the illusion of creativity and novelty, when all we were left with was the spectres of the past. The qualitative shifts in genre that had defined the twentieth century were still haunting a twenty-first that could no longer create anything truly new for itself. In other words, hauntology still persists, even in the most progressive music today.

However, this was precisely where I departed with Fisher. His defenders might claim that jungle and the other genres which, to him, represented qualitative changes in musical timbre also represented a radical break from what came before. But is this really true? Can we not trace a clear progression through almost all musical styles? From blues to rock? From rock to metal? From early electro to techno, house and their subsequent incarnations? Even jungle’s roots can be traced quite clearly through a certain combination hardcore breakbeat, techno, ragga, reggae, and a variety of other styles. Yes, perhaps certain technical innovations made the leap between those genres and jungle larger than many we have seen in recent years, but no musical innovation appears out of a vacuum. Every cultural form is created through the difference of virtual potentialities and their interaction with other potentialities. It is precisely this interaction that leads to the actualisation of new forms of expression.

I believe that Fisher’s harking back to jungle as the last bastion of creativity in the twentieth century is a result of his own hauntological tendencies. The fact that Fisher’s intellectual maturation happened to coincide with an era of music that had produced something as progressive as jungle had clearly influenced his ideas in a radical way. Despite it’s radical inventiveness, jungle was, as much as any other genre, an assemblage of ideas that had been developed before it. The whole CCRU scene was so concerned with inventing the future that any future which arose would inevitably be seen as a disappointment. I feel that many of Fisher’s comments on hauntology in Ghosts of My Life, despite their intrigue as reference points for cultural theorists, are a reflection of that. I’m not convinced hauntology can be applied to the more progressive elements of modern music as well as Fisher thought.

Here Xenogothic makes an interesting comparison between anti-hauntology to accelerationism. He claims that:

The “anti-hauntological” turn that Bluemink christens, no matter how valid his appraisal, is arguably already well-established. In fact, I’d like to argue that “anti-hauntology” already has a name, and it’s a name worth remembering (or at least reminding ourselves of) because, without a better appreciation of its purpose, we run the risk of mourning SOPHIE only to retread old arguments and take the shine off of what she was doing. … We should remember that anti-hauntology’s first name was accelerationism.

However, I would argue that hauntology is essentially the innevitable post-90s comedown from the buzz of CCRU accelerationist movements. The relationship between anti-hauntology and accelerationism is an interesting one, but it I think it requires it’s own post which I will respond to in the coming week.

Sonic Deterritorialisation

In a 2018 interview Sophie herself commented on the state of popular music as she saw it. She argued that:

there’s a huge amount of work to be done socially and culturally. The gap between where we are now and where I’d imagined where we could be and the places our imaginations could take us are so far away from what we’re presented with a lot of the time. So I can’t get too excited about anything happening now. I’m really excited about what should be happening in the future.

Sophie’s muted excitement about the current state of music and her forward looking mentality which sought to bring her ideas about the future into the present is precisely why I described her music as anti-hauntological. She accepts, much as Fisher does, that many aspects of the current cultural climate are hauntological, but she wants to move beyond these constraits and she sees hope in what could be produced in the years to come.

However, when thinking about anti-hauntology, I believe there are ‘popular’ artists out there who are bringing the creative futurism into the present in a way that perhaps even Sophie wasn’t able to. Indeed, when considering the music of the future it’s difficult to omit the work of the experimental Venezuelan producer Arca.

I was first made aware of Arca’s work on FKA Twigs’ EP2 (2013) which they wrote and produced together, and on Twigs’ LP1 which was released the following year, the influence of Arca’s unique sound had infiltrated every aspect of Twigs’ music. A little while later, I was reminded of how interesting I found Arca’s work when I heard her production on Kelela’s ‘A Message’ in 2015.

The genius of this track is that the instrumental consists almost entirely of warped samples of Kelela’s own voice, pitch bent to produce bass, treble, and harmonies that arrive organically in stark contrast to the lead vocals which soar over the top of the track. To me, ‘A Message’ wasn’t only one of the most interesting and innovative instrumentals for a pop or RnB track I’d heard before, it seemed to really be moving popular music production into new territory. Indeed we might say that it was deterritorialising the standard modes of music production in novel new ways.

Deleuze and Guattari saw the tendency towards deterritorialisation as one of the unique properties of a capitalist system. This concept is used in a variety of ways throughout the Capitalism and Schizophrenia books, but if we use Manuel DeLanda’s definition of deterritorialisation as “any process that takes the subject back to the state it had prior to the creation of fixed associations between ideas” (Assemblage Theory, 27) we can understand that capitalism has the inherent tendency to strip back the established frameworks that govern the functioning of society (deterritorialisation) and to re-establish them within capitalism itself (reterritorialisation). In other words, it has the capacity to decentralise the ‘flows’ of desire and reorganise them in novel new ways. This process of deterritorialisation therefore “constitutes the most characteristic and the most important tendency of capitalism” (Anti-Oedipus, 48). Here were can certainly see the Deleuzo-Guattarian influence in Fisher’s work. Essentially Fisher’s hauntology is a reticent acceptance of capitalism’s ability to reterritorialise culture within it’s own boundaries.

However, Deleuze and Guattari make an interesting observation regarding how this tendency functions. They claim that the decentralisation of the flows of desire, and the tendency to upset previously established orders, draws capitalism ‘near to its limit’: “Capitalism therefore liberates the flows of desire, but under the social conditions that define its limit and the possibility of its own dissolution, so that it is constantly opposing with all its exasperated strength the movement that drives it toward this limit” (Anti-Oedipus, 164). I would argue that anti-hauntology is exactly what happens when cultural forms, especially in relation to music, are pushed to their limits. Perhaps that is exactly the point that Xenogothic is making when comparing anti-hauntology to acceleartionism. However, I wouldn’t consider wouldn’t consider the artists I’m discussing to be inherantly accelerationist (although there are surely parallels). What producers like Arca are doing is taking, morphing, shifting, and reconnecting disparate aesthetic elements together in creative new ways that escape the kind of hauntological spectres of the past that Fisher was wary of.

Auditory Assemblages

Although Arca has been producing her own material since 2013, she released her most critically acclaimed album, KiCk i, in 2020. This groundbreaking fusion of styles as disparate as IDM and reggaeton, with features from artists like Sophie and Bjork, was sure to stand out amongst the crowd. As The Guardian‘s review of the album describes:

It’s an album with a liquid heart, Arca shapeshifting through a look book of sonic identities. “We don’t have to be any one fixed thing in a nonstop way in order for us to be normal,” she said earlier this year. “When are we ever not in flux? We’re literally, like, fucking fluid.” … She pirouettes through reggaetón, techno, power ballads and bubblegum electro, absorbing each element into a chaotic buzz of high-definition sound design and glossy, post-PC Music sonic detritus. Haunted, gender-fluxing rap is twisted with wobbling melodies and the clattering rhythms of shell casings hitting the floor.

Despite the talk of haunted vocals and shell casings hitting the floor (a reference to samples used by Burial) there’s no doubt that Arca’s record is something of an advancement in popular music. The emphasis on fluidity and flux is something inherently creative; something anti-hauntological. But, as a reddit user commented: “is that a qualitative advancement? Isn’t the fact that people are now mashing together any old sonic trope and genre just proof that we are no longer able to create radically new genres? … Ok, you’ve mashed together everything that already exists — where do you go from there?”

To this question I might reply, as John Twells from The Guardian does: “The danger of this level of fusion is part of its appeal: when something doesn’t work, the failure acts as a reminder of the complexity of existence.” This was exactly the point I was trying to raise when coining anti-hauntology. Never before heard sonic aesthetics can be produced through a variety of means, through technological advancements, or radical exotic combinations of pre-existing sounds. Artists like Sophie and Arca are essentially creating assemblages with sound. As Deleuze and Guattari state in A Thousand Pleateaus: “An assemblage is precisely this increase in the dimensions of a multiplicity that necessarily changes in nature as it expands its connections” (A Thousand Plateaus, 8). In other words, it is precisely the novel combination of sounds that allows a sonic assemblage to be classified as ‘new’. Every new relation inherently changes it’s nature and revitalises possibilities for new connections with the listener. Arca’s music offers, Twells claims, “a red pill to a more hopeful future.”

AI: Riquiquí;Bronze-Instances

The question of whether combinations of genres can produce something qualitatively new is a fascinating one. However, to conclude this article I’ll leave my readers to ponder something truly innovative; something that certainly would produce a jolt of ‘future shock’ in listeners from 20 years ago.

Arca released an album of 100 tracks. 100 remixes of a single cut on her LP. 100 remixes that were created by an AI.

As she described in a statement to pitchfork:

Did you know that up until now I had never allowed anyone to remix an Arca song? There existed 0 official remixes to an Arca track until today—“Riquiquí” has gotten 100 remixes by an intelligent sentience, created and trained by the genius minds at Bronze. I’ve worked with Bronze once before; in 2019 I gave musical language to Echo, a musical being birthed into the Museum of Modern Art’s lobby, and then Echo began to speak of its own volition. There you will never hear the same thing twice—for two years it is an evermorphing stream thanks to Bronze’s trained yet unpredictable musical AI. I recognized the textures and melodies, but never the song—for a composer such as myself it remains something truly new which I had never experienced before, a moment of unforgettable experience in virtue of the mystery and wonder Bronze makes possible.

The fact that something as groundbreaking as this is even possible should be proof of Arca’s insistence on utilising innovative technologies to break free of the constraints of hauntology. To my mind, there’s no doubt that utilising AI to remix music is anti-hauntological. However, it raises a whole new host of questions concerning the nature of human creativity and the role of art in society. Can machines really create art for us? Is using an AI to remix a piece of music any different than using a computer plugin to imitate an analogue instrument? In the future, is AI going to become a tool in the creative process or will it subsume the creative process altogether? Yes, the initial input and the sonic textures used are produced by the human being, but if they can be reshuffled, reordered, and reorganised in such a way as to create a new song 100 times over, how much of this is the input of the producer and how much is the input of the AI?

I will have to save those thoughts for another article, but for now I will leave my readers with a quote from philosopher and neuroscientist Andy Clark that certainly embodies the direction Arca wants the creative process to move in:

Pretty soon, and still without the need for wires, surgery, or bodily alterations … we shall be cyborgs not in the merely superficial sense of combining flesh and wires but in the more profound sense of being human-technology symbionts: thinking and reasoning systems whose minds and selves are spread across biological brain and nonbiological circuitry (Natural Born Cyborgs, 3).


10 thoughts on “Anti-Hauntology: Arca, AI, and the Future of Innovation

  1. Pingback: Anti-Hauntology: Where are the New Forms of New? – xenogothic

  2. the whole discussion is great. To add to your argument because Arca is mentioned and he is one of my fav artists. I believe reggaeton is key in Arca’s work and he somehow paved the way for latin-american producers who test the limits of where the sound can go – people like Kelman Duran or Kamixlo, the whole neoperreo scene, etc. I don’t see where hauntology fits into their music – maybe somebody can explain. In this case I believe we see the progression of electronic music genres that were born into the 90s but never went one step further until now – and of course technology and identity played a big part here. To add also to a previous comment I made Fisher couldn’t also have predicted that english was not going to be the dominant language of pop music in 2021 when he wrote Ghosts of my Life. When I first came across this fact I started thinking that maybe hauntology is not relevant anymore.

    Liked by 1 person

    • Thanks! I think that’s a very interesting point about the language of pop music. I agree that it would have been impossible to predict that Kpop and Latin music would have become so prevelant now. I will have to think about how that fits into the anti-hauntology narrative.


  3. It’s funny that your argument of Arca’s “anti-hauntological” novelty rests on her utilizing AI / machine learning to create supposedly ‘new’ music.

    For people who actually have an insight into how various ‘generative’ machine learning algorithms work (most used would be the so-called Recurrent Neural Network, LSTM’S or GAN’s), there is actually minimum of novelty in the material that is being output by such an algorithm. I make this point well aware that one would need first to define what actually is novelty, but bear with me –

    The algorithm ‘learns’ from a dataset that is provided to it and in the end reaches a point where it can generate similar material. Actually the way how that algorithm is usually ‘trained’ is that during ‘training’ the intermittently generated material is compared to the original dataset and the neural network’s parameters are tuned so as to generate material that more closely resembles the input dataset. (I’m using the words ‘learn’ and ‘train’ in quotes because they convey the idea of what’s happening, and are used as technical terms, even though the actual process is quite different from what is usually understood by learning or training).

    From this it should be obvious that the generated material is hardly new. It is just a close approximation on the original material, twisted and reshuffled in various ways, de- and re-contextualised within the structure / composition of the whole piece / track, but there are no qualitatively new elements present. Many find a certain weirdness and surprise in the new combinations, and sometimes by chance they can create interesting and unexpected results, but in the end so far most of the generative machine learning methods are not much better as a more or less random cut-ups of the input dataset. And this was already done – for example in literature (Burroughs’s and Gysin’s cut-ups, Queneau’s A Hundred Thousand Billion Poems), and i’m sure you can think of many musical counterparts.

    In a way it leads back to the Reddit comment you are quoting, asking where is the qualitative change in the whole ? You answer, it lays in using AI, but in my opinion using an algorithm (a simple random cut-up, or orders of magnitude more sophisticated deep-learning architecture – complex and thus endowed with a mythical quality – just see Arca’s comments about “intelligent sentience”) to generate output from input doesnt make the output more novel, and hardly groundbreaking.

    Btw I’m replying only because its tiring to stumble across this overvaluation of AI as a field and machine learning algorithms in particular on every step, being used to justify/support any sort of conclusions and discourses.


    • Hey thanks for the info about how the system works, but I would say that’s all fairly evident to anyone who’s listened to the record. I don’t think that’s some kind of secret, you quite literally hear these things being done track to track.

      My point here is twofold. First Arca’s music was anti-hauntological before the AI. Second that the utilisation of AI could push aspects of music production in new directions that would further move away from hauntology. It’s not the AI in itself that anti-hauntological. It’s merely a tool. But as I said at the end of the article, it raises a whole host of philosophical questions on the nature of creativity which is central to the hauntological problem.

      I can’t help but think you may have missed the point slightly.


  4. Pingback: The Anti-Hauntology Debate Continues – U+2205

  5. Pingback: Anti-hauntology: SOPHIE, Stiegler and the Ruins of Accelerationism | Blue Labyrinths

  6. mmmmm …

    It seems to me that some concepts are already quite primitive for the moment. In Argentina we have an artist known as WOS champion of red bull free style and precisely what he said is this, not with the term hauntology but that to continue calling artists to reuse what has already been is to go against musical evolution. It is also true that for those of us who come from the same continent we can detect how in some of its tunes Arca uses native melodies that many could relate to the past, but the real question is the past? or is it what remains in time for being a cultural icon?


  7. Pingback: Anti-Hauntology: Iglooghost, Mark Fisher, and the New Digital Modernism | Blue Labyrinths

  8. Pingback: 反幽靈學 2 Arca,人工智慧,以及創新的未來 - 金妖獸

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s