Interview with Adam Basanta, Canadian Visual and Sound Composer

Kinetic mixed-media installations and an (almost) autonomous AI art factory

by Michael Palumbo

On 22 January 2019, Michael Palumbo (Toronto) remotely interviewed Adam Basanta in his Montréal studio.

[Michael Palumbo] So amidst everything that I have read that is going on, how are you doing? How are you feeling? I suspect there are things you can’t talk about, but how are you as a person doing?

[Adam Basanta] Yeah, I’m pretty good, I’m keeping very busy. And I guess you’re referring to the lawsuits?

I mean, emotionally that’s where I went right away, yeah.

To be honest, it’s been going on for a while. That whole thing started around June as something that was occupying a little bit of mind space or whatnot. But I try not to let it occupy very much of my time. I mean, there are some practical matters that occasionally I have to attend to, and it’s going forward. I’m very confident in my position and that I’m right, legally. 1[1. In 2018, a lawsuit for copyright and trademark infringement was filed against Basanta because of an image generated by his AI “art factory” that resembled a work by another artist. See Chris Hannay, “Artist faces lawsuit over computer system that creates randomly generated images,” The Globe and Mail [Toronto], 3 October 2018.] So, it’s going forward, and I really don’t think about it too often, to be honest. Just on occasion, when I need to deal with something practical or there’s a next step coming, or something like that. It occasionally will be on my mind a bit. But as a person, I’m doing great.

That’s great to hear.

It’s a busy period and I’m working on other projects; this is something in the background.

What are some of the other projects you are doing? What’s keeping you excited?

In April I have a new solo exhibition that’s opening at Optica, which is an artist-run centre in Montréal, and I’m making two sound installations for that show — that’s the main thing on the agenda. And then there are some other projects towards the fall that I’m starting to work on. But the kind of intense period right now is towards that show in the spring.

Are you working with a particular curator? Are there certain parameters that you’re facing?

No, I have a room, I can do what I want with it. I had an idea that I proposed and they accepted it; I changed it a bit since and I’m going through with it in some way or another. But no, there are no parameters. Basically, I just need to show up and set up.

What can you divulge at the moment?

It’s two kinetic art pieces, kinetic sound pieces or sculptures. They involve these objects that I’ve used before, specifically microphones; I’m not really using that many speakers. But it’s kind of this language that relates to a slightly older work of mine, and so I’m revisiting it with a lot of very large-scale material approaches. I’m using a lot of materials inspired by construction — there is construction all around me! — and so [it involves] a lot of cement and gravel. There are going to be two works that are quite related. They’re both kinetic, they use microphones in some way to create sound and to create other sorts of actions. But at the same time, they’re both very Sisyphean kinds of machines that are just working for no apparent reason. But sonically, I’m working with the idea that one of them is making very continuous sounds and the other is punctuating it. In terms of the sound design of the pieces and how they work together in a room, that’s the combination.

Right, I was going to ask, they’re in some way contingent upon each other?

They could be presented alone as well, but I think together they gain something. I think they’re in some ways conceptually related, in some ways conceptually opposed; there’s something that’s very continuous, something that’s very punctuated. One thing is going to be kind of a delicate, continuous sound, the other one’s going to be very loud, very aggressive action that leads to sound-making. So, they’re kind of very different characters, and together it’ll be a dialogue, which I think will take it to a very interesting place. But they could be independent in another context.

I’m so curious about this. Had you conceived of them together? Was one of them something that had been kicking around in your head for a while? How did these two find themselves together?

Yeah, for sure one of them was kicking around for a while, I’d say maybe a year and a half — the original sketches — or something like that. And actually, I thought it could be an independent piece that would work really nicely. But there was another kind of idea that I was playing with, I thought they could be a duo together. Because, just in terms of the gallery space, I thought there was enough room to have another piece. And then I decided that I didn’t really like that other piece [laughs], or I didn’t think it complimented this nicely. And this new idea was sparked out of… honestly, I can’t remember how, out of nothing, basically. And I thought, this idea has a lot of contrast to the other idea. And I could start thinking how they’re related and how they’re different, and develop all the details of the piece that way. They’re still developing, in some way, until I will be ready to show it, I guess.

And I suspect, as is often the case, they will continue to be developed in your mind, perhaps even afterwards.

Yeah, sure.

In the interest of the archive, can you share either titles or working titles for each of these pieces?

Oh, yeah. There’s one piece right now called A Large Inscription that I’m imagining will be continuous and more delicate, in some way. And the other one that’s punctuating it is called A Great Noise. So the title of the show is [a combination of the two titles]: A Large Inscription, A Great Noise (Video 1). That’s my main occupation the last little while. Some other projects, but those are the main projects at the moment.

And am I right to assume that A Large Inscription was the one that you had thought about for a long time?

Yeah, that’s right. I’ve been working on it or a year and a half; it’s maybe a year and a half, two years ago that the idea came up — some central part of the piece, like the kinetic system, maybe. I knew I wanted to use gravel for some reason, that was very clear to me. And then, you know, you kind of kick it around in a notebook et cetera, et cetera, maybe write a little text. Then I started thinking about applying for some funding or a show, or something like that. And now I’m actually working on it daily, figuring out logistical problems and stuff like that. But a lot of the conceptual development — going back to the notebook, adding little ideas, figuring out can this be done, can that be done — that’s been going on for longer.

Video 1. Documentation of Adam Basanta’s mixed-media installation A Large Inscription, A Great Noise (2019). Vimeo video “A Large Inscription, A Great Noise” (4:13) posted by “Adam Basanta” on 29 April 2019.

This is in your own studio space in Montréal?

Yeah, exactly.

Is it your first time working with gravel or with stone?

I worked with cement before once. I made a sculpture, a cast with cement. And actually, I recently did a piece with some cinder blocks, and I think I may use cinder blocks again, I’m not sure. But those are the only two times, so they’re fairly recent, I think 2016, 2017, 2018. I worked with it a few times, but never in this way and never on this scale. I’m going to use a lot of gravel, like bags and bags of gravel — a truck of gravel.

In the iteration / ideation phase, or in the finished piece?

In the finished piece.

So, I just caught something, when you brought up the cinder blocks, you sort of chuckled a little bit there. I wonder, is there something that you struggled with? Or, what delighted you when you brought up the cinder block?

Oh, I guess it’s just funny, because it’s so cheap. You just go to Home Depot and pick a couple up for $3 or something. I was working on this piece and I just wanted to elevate something a little bit vertically, but I didn’t want to build a stand or something. It needed to look rough. And I just had the cinder blocks kicking around. I don’t know, they just have such a look — I mean, they’re so iconic in some ways. And also, Home Depot is just around the corner. It’s one of those decisions where you don’t know if it’s convenient or if you like it anymore. Somehow the two have just come together and it just becomes something you draw on — as a private joke.

I haven’t seen that piece of yours. What’s that one called, the one with the cinder blocks?

There’s a video online called Persistent Teenage Gestures (Video 2). But the cinder blocks are not central elements, [laughs] they’re just cinder blocks kicking around. But they looked really good in that project, and then I still had them lying around, so I was playing around with them.

They’re getting some due course right now, I guess.

Video 2. Documentation of Adam Basanta’s mixed-media installation Persistent Teenage Gestures (2018). Vimeo video “Persistent Teenage Gestures” (2:29) posted by “Adam Basanta” on 29 October 2018.

Yeah! One thing I will say about this show, it’s really trying to mix ideas that are, I suppose, more philosophical or conceptual with elements that are very personal to me and to my life over the last year or so that I’m working on these pieces. And as you know, as somebody who has lived in Montréal, the construction here, I mean, I really feel like it’s the worst I’ve ever seen it. It’s right by my studio, it’s right by my house. You just have these materials around you all the time — these huge tractors and pile drivers, and gigantic sewer pipes and cement blocks and cinder blocks, and all this stuff is just around. And gravel… and you just see it everywhere.

It shows up out of nowhere, and halfway down the street — like it doesn’t seem to have a pattern!

Yeah, yeah. And there’s something kind of amazing about this scale of construction. And, I mean, Montréal, it’s not that big in terms of the scale, if you think about other places in the world. But you know, they just, like, open up two blocks or something. And it’s, on the one hand, inconvenient and annoying and whatever; on the other hand, it’s kind of incredible. It’s kind of, the greatest art installation in the neighborhood is the construction site! It’s just this insane scale of subterranean architecture that’s opened up. I don’t know, just for some reason this year, it struck me more than others. And so a lot of that has kind of filtered into the decisions that I’m making in this work.

It’s funny you mentioned that, because I’m also thinking about how the street still flows somehow — it is striking. I don’t know that I ever reached a level of frustration with it, but it can be an eyesore. If you’re a cyclist, it can certainly be something that pushes you further into danger.

Sure, or a driver. I mean, my studio’s right near Plaza St. Hubert and that whole street was closed. And [when] you have to bring something to your studio you have to go around, on another street, there’s a lot of traffic… And yeah, there are two sides to it. So a lot of kind of “construction” themes, and I guess the cinder blocks are one of them.

That’s fun. And gravel is as well. So am I right — just trying to paint a picture in my mind — that the cinder blocks are making an appearance in the solo exhibition that’s coming up as well? Or this was in the previous work?

I’m not sure if they will or not, they might. There’ll be stuff of a similar kind of look. But definitely, there’ll be these cement blocks — that’s going to be a big part of it. And gravel is going to be a big part of it. Gravel is also a [construction] product, so there’s a relationship between the two. But of course, also the sonic associations are going to be very different, because gravel is something that’s particulate; it has this kind of granular kind of quality to it in the sounds that it makes. A block of cement either doesn’t make sound or blocks sound, or it’s hitting something and it’s going to make a big loud sound, basically. So, they have very different kinds of sonic characters that are developed, both in terms of the sonic characteristics per se and also in terms of the (maybe we can be charitable and say) “musical phrasing”; each installation is going to be quite a different vocabulary. I think the two vocabularies will actually overlap in the same sonic niche, in the same room, in a really, really nice way. Because they are really so, so different.

And when will you be able to access the space and determine where you’re going to place them?

I’ve been there a few times recently, just to get the sense of it. Every time there’s a new show there, I try to go — it’s pretty close to where I live. But in terms of the actual setup, [that takes place] four or five days before the show. I like to be very well planned; I don’t think I’ll set up the whole thing exactly [prior to the show], but I’ll do all the tests on a small scale to know that the plan will work.

Does computation figure into the performance and presentation of either of these two pieces?

To be honest, I’m not sure right now if I’m going to use just Teensys 2[2. USB-based microcontroller development boards.] and analogue electronics or these old Mac minis that I have from way back… I’m just in the middle [of the project] in terms of logistics. But there is some sort of computational control because there are motors which are changing patterns in some way. There’s going to be some minimal, very, very subtle sound processing, but some of that will be analogue. I’m using guitar amplifiers, so they will add a tone, and then I’m going to use multiple amplifiers, possibly, to get different kinds of sound characters out of them. And there probably will be some very subtle kind of DSP.

Which Microphone Works Best in Gravel?

When you say motors and changing patterns, are you referring to patterns within the gravel itself?

Yeah, basically there’s gonna be a microphone that is being dragged through the gravel in a circle. It’s not going to be all over the place, but it’s going to create a circle which will get deeper every time it goes around (Fig. 1). But there are some basic motor controls for starting, calibrating, stuff like that. All of that will be [controlled] either with a microprocessor or a simple Max patch. In these two pieces, specifically, the computation is not really that important, conceptually. It doesn’t really matter if I’m using Max or if I’m just using guitar pedals — it’s more about the effect in these two pieces. At the same time, in a weird way, there’s this sense that the pieces themselves are kind of these useless machines. And it does speak to some sort of question about automization in various fields. But it’s not about computation per se, it’s more about automization in the larger sense.

Certainly, but I think [in] asking about computation, I wonder if it might elicit to some extent what it is that you’re playing with, what it is when you are encountering the gravel and you have microphones. What systems do you envision designing here, or encountering? Whether it’s a Mac mini or a Teensy or anything like that. And I suppose it’s also a question of whether that’s going to be in full view or behind the curtain. Which, again, is a choice, but also perhaps not one that you necessarily want to focus on, and I totally respect that. What I am particularly curious about now is have you tried different microphones? I’m curious, are you using, say, a large diaphragm microphone…

No!

… as a kind of contact mic? I mean, not contact in the sense of the way that a Piezo works, but the grill, I presume, is being dragged. What are the characteristics of certain grills, you elicited certain textures? You know what I mean? I could go in a couple of directions here, but I’m curious what you found.

No, that’s an interesting question. I would say, actually, that I totally get the thinking behind your question. But it’s not exactly how I’m thinking about it in this sense, and I’ll explain it. That kind of approach of, for instance, testing different microphones and some grills and getting the resonance of this grill as opposed to that grill is a super, super interesting kind of approach. And especially if I’m doing a sound object piece or something like that, that’s how I’m gonna get started: this one has this characteristic and that one has that characteristic. The way that I’m thinking about it, in some ways it’s not so much about the exact sound of the microphone. It’s about the microphone as an object that manifests in many ways. And sound is one of them, but [that is] not the only way that it operates. I usually use SM58 microphones [laughs] for one reason only…

It’s the most droppable microphone? [Both laugh]

Dropping on the gravel, that’s an added advantage.

Yeah, that’s right! [Laughs]

But if you think about it, what’s the first microphone that comes to my mind, like in the Google Images of your brain, right? What’s the image of a microphone that comes up? For me — and I think for a lot of people — it’s the SM58, you know, it’s this kind of classic microphone. You know it’s for a vocalist or speaker…

Karaoke.

It could be [for] karaoke. So it’s a cheap consumer product, relatively speaking; it’s not a specialist kind of product. It’s a product that sort of allows us to “express ourselves”, you know? That’s how these microphones are sold. You want to start a band so you can express yourself, or you can sing and people can hear you, right? These are kind of the metaphors that work in the background there. And although, of course, you can use an SM58 or 57 on another instrument, specifically for a person to give a speech or an address: world leaders use the SM58, and so do rappers, you know what I mean? And we are in a time where a world leader and a rapper or a celebrity of some sort, those two ideas are becoming closer and closer together. So for me, all of that is in this microphone. And that gives me a lot more than the sound of the grill. That’s how I’m choosing these elements. And then it’s my responsibility to work with the sound of that grill and make it the best fucking grill-dragged-on-gravel [mic] that I can make it!

The Microphone as Material

But it’s always starting from that kind of approach. And there are kind of themes in the two pieces. In addition to construction, there is this play on these objects which are used for communication, they’re used for self-expression, they’re used for speeches, they’re used by a dictator… There are a lot of these kinds of metaphors that, for me, are kind of oozing throughout the works. And I’m trying to make decisions based on those instinctual feelings about the materials. So, that’s how I choose the microphone. It starts with the idea of the microphone, and then it goes down to very, very specific details. For instance, if you need a lot of SM58s and you’re going to treat them badly, you can also buy a cheap copy of an SM58 for a quarter of the price, and you get four of them.

When you asked me, “What happens when you Google SM58?” I was gonna say an SM58 will appear and then right next to it the Behringer copy, right?

Right. Yeah, exactly! So there are definitely practical decisions relating to the mic and the dimensions and all this kind of stuff. But they are always preceded by these decisions, these more conceptual approaches, where I’m thinking, “What do I want to say with this piece?” The sound has to be part of a larger idea that I’m trying to birth out there into the world. [Pause] That was a really long way of saying that I didn’t try different microphone grills. [Both laugh]

Yeah, well, it’s lovely, because your response still centred around the grill, right?

Yeah, because I care, there’s a relationship of care towards these materials. Even though I’m actually doing really, really nasty things to them with cement and gravel, and misusing them, right (Fig. 2)? But for me, this show is kind of like a… I don’t want to say “culmination” or something like that. But I’ve made pieces with microphones and speakers because that was my language, those are things that were around me since I was 14, playing in bands. Those are the materials that I grew up with and that I learned how to make art with. And then when I was making more visual art, it was my personal language already from another life. It was a very comfortable way to start, but I was trying to say new things in this new sphere of making or creating, in some way. And now that I’ve done that enough times, it’s actually becoming a personal language. With these pieces that I’m making, there are decisions I’m making because I feel that it relates to a piece that I made four or five years ago. And that becomes kind of like a mini personal history that I feel quite strongly about. So as to whether I care about the mic: I care, really. [Laughs]

That’s evident when you talk about what a microphone does for people. Surely, the world leaders use an SM58. But as you were alluding to, so does a band in a garage when it’s their first jam, a couple of young people in their first band ever, right?

Totally. And the situation is so complicated, because you could have a world leader saying something really terrible at an SM58, you can have a world leader saying something really inspiring — it’s not all bad, you know? But at the same time the kid in the garage is expressing themself. They were also sold an idea that you deserve to express yourself, which then creates a need for certain items. And then once you get this item, well, you also need that one — and that one is actually better. So the next time you buy a more expensive thing, and it’s branding. So there are good and dark sides to every facet here, and it’s very tangled up. That’s why I think it’s an interesting material to work with.

And, you know, I think that regardless of what art form you think of that makes use of microphones, chances are that dragging a microphone through gravel is not the only way to make that microphone dirty. I mean, if you think about any of the open mics out there, they’re probably filled with spit, right?

Sure, sure.

And dropped, probably dropped on beer-soaked floors. And there’s something about particularly that kind of microphone, where you can hand it off to somebody else. It doesn’t need to sit in a shock mount [like a condenser mic], right? Where even if I give an articulatory plosive with my lips, I could potentially risk damaging the very, very sensitive membrane on the inside.

Right. Yeah, this is a working man’s mic, for sure! [Both laugh]

That’s right.

No, this is the mic, you can drive a nail into the wall with it and you can still use it.

This is the “Local Traffic Only” microphone.

But there is another layer. [For me] as a “sound person”, there is an ethos and a mythology to all of these items that is particular to only those of us who’ve been in that scene. So, yeah, the SM58 means something to us. And there may be other microphones that we have associations with, but I think [for] the general public, that is not necessarily so intimately familiar [with various models], I think the SM58 is “the one” — it’s the one that has some sort of meaning to them, because it really is so ubiquitous.

Right.

So, that was a great discussion about SM58s, I really enjoyed that. [Laughs]

Yeah me too, Adam!

On Authorship and AI Creativity

There was something that I had wondered about. Is there a precedent for an artist being sued for a system that they created? I don’t know what you and your legal team are presenting, so that may colour this question differently…

No, but I would say that there is no precedent in Canada, at least — that doesn’t mean that somebody was not sued. It just means that that suit was never actually filed in court. It was just a cease and desist, or there was a deal, a compromise.

Let alone challenged, as you were saying.

Yeah. So, an actual legal precedent, no, there isn’t. One reason that I think it’s particularly interesting to people right now is because AI (and machine learning) is such a buzz term — in art, but also outside of it. There’s also no precedent on any sort of copyright or trademarking or whatever of content created with AI. There are just copyright cases of a code that’s written in AI but that’s a very different kind of thing. So both things don’t have a precedent: on the one hand because they’re new and on the other because it just so rarely actually [gets] filed in a Provincial Court and goes to court, goes to trial.

Do you think that if your system was aware it was breaking the law, it would continue to break the law?

[Laughs] If it was aware of breaking the law, would it continue to break the law? You know, I didn’t code any ethics into the system at all, so I think it would continue. I mean in some ways, there’s a beautiful irony here. The piece is supposed to be an art factory that doesn’t need humans, basically. It doesn’t need us to make any decisions, it doesn’t need us to look at the art. It’s just producing art for its own sake, through this kind of ecosystem-like design of technologies (Fig. 3), which would, in theory, go on forever, until one piece of the ecosystem breaks and then nothing will happen. So, in that sense, if it knew it was breaking the law, then it probably wouldn’t care because it didn’t need us anyways. [Laughs] It’s all about this kind of self-sustaining movement — it’s actually designed not to care about us, so it probably still wouldn’t care.

Well, I’m just curious, because in the validation stage (stage two)… And I suppose just to set this up for the listener, or the reader, stage one is creation, as I understand… oh, I’m gonna explain your piece to you, would you like to take it? [Laughs]

Sure. I mean, I would say step one is a random system. It’s actually several overlapping random systems, which have no idea what the [others are] doing, or what state the [others are] in. So, you have computers that are controlling the scripts for the scanners, there’s some weighed randomness in the decision-making, some uniform randomness when it randomizes the parameters of the scanner, and then it presses “scan”. But it doesn’t know where the other computer is in its phase cycle: Did it start a scan, did it not start a scan? Is it randomizing this parameter or that parameter? And there’s a light above the two computers, and that just has random patterns. Again, some kind of behaviours that I’ve coded: it could be a pulse, and that pulse would last for a certain amount of time and then it’ll turn off for a while, and it’ll ramp up… These are all also kind of random decisions, which are made with complete disregard to whether or not the scanner is scanning or not, at this moment in time. And so it’s three entirely random systems that don’t know anything about one another, and that’s how the images are created. And then in the validation phase, the created image is tested against this database of a million artworks or something like that, through machine learning techniques and some other machine vision scripts. Then it gives it a score and if it is close enough, above a certain threshold in the score that it gets from the analyzer, it posts it online, and it can sometimes print it in the gallery, etc. (Video 3). So that’s the difference between the validation and the creation phases.

I think I have the answer now, but I had been wondering, from the validation stage, whether there was any feedback into the creation stage?

None whatsoever.

Gotcha. So the validation is like the tastemaker, in a sense?

That’s exactly right. The two scanners, though, that’s the artist going into the studio every day and mixing the paint a little bit differently, just randomly mixing pigments and then saying, “Okay, we’re gonna produce a canvas.” And the analyzing computer, that is the validation, that is really the “curator” or the gallerist or the consumer… I wouldn’t say consumer, because it’s just choosing what should be separated from the rest and put on a pedestal. So that’s more about [its role as] a curator, maybe, or something. Definitely two very different roles in the piece.

Video 3. Documentation of Adam Basanta’s mixed-media installation All We’d Ever Need Is One Another (2018). Vimeo video “All We’d Ever Need Is One Another” (3:12) posted by “Adam Basanta” on 9 May 2018.

And see, I think that’s why I had wondered about whether or not ethics have been programmed in, if it had known that it was potentially creating something that was close enough to someone else’s work that they would feel uncomfortable.

Yeah, I think this is actually really relevant, in the sense that this piece is, I think, quite different from the way that AI is used. And most of what you see, if you just look for GAM experiments or AI art or something like that, a lot of it is really teaching this software by looking at a database and trying to get [a result that is] closer to something that has already been done. To me, that’s a less interesting question, or a less interesting way of working, at the moment. The idea of the database is super, super interesting to me. The database is something that has been valued by society, right? Like, we said this needs to be in a museum — not “we”, but somebody did and now it is in the museum. But at the same time, the machine is quite unintelligent about what it’s doing. It’s just creating… less from the idea that it needs to be inspired or it needs to know something. It needs to work with the idea of pastiche, basically: “Okay, that was done, so I’m going to try to get closer to that. And every time I get trained, I get a bit closer and I get a bit better at realizing this kind of ideal.” The idea of creativity in this piece is not that. The idea is very much that there are many, many accidents that occur. And that, in fact, those are the interesting artistic results — those that are unpredictable, that are occurring without complete control by a single author. There’s a sense that the material itself is in play here. And then, of course, you have this selection process where it’s deciding what should be chosen, what shouldn’t be chosen, and that actually has nothing to do with creativity, whatsoever. It has to do with what the market — whether that’s the commercial art market or the institutional art market — has said is important. So, it’s actually a very different type of project than many of the other AI kind of paintings that have been popping up in the last couple of years.

Well, it’s so lovely, because it’s like a schema for the artist and art manager, which very often is one person, but the tension within that is, “When do you put which hat on?” And when is which hat allowed to be put on?

Right.

And I suppose in either space, or with either hat, the accidents that you’re referring to are valued perhaps differently, or can be.

Yeah, I think there’s something there, definitely. I mean, for me personally, I feel more and more that the accidents are what I find interesting, both as a process and as a way of working. And also in terms of the results. Just a funny little anecdote is that when I was making the piece — this is kind of similar to the mic grill conversation — I was trying different scanners to see what they look like, what the software’s like. I kind of settled on the scanner that I thought would work best, and I started experimenting. I just had two computers, two scanners, and I had a mouse. I would just randomize the parameters of computer A and then I would randomize the parameters of computer B and then I would press “scan.” And I would just see what the results are like — this is maybe the first month of working on the piece. Just to get a basic proof of concept, like, you know, this is going to work. And the more I did it, I realized that I’m actually not randomizing the parameters; I’m looking at the preview window of the scanning software and I’m adjusting based on what I see. So, okay, more contrast, yeah less red; oh, it’s too green now, I’m going to pull the green back, you know? And I was producing what I thought at the time were pretty decent images. And then I automated that whole process.

So I automated all the randomization of the scanning parameters, so that, first of all, I wouldn’t do it, and also it could be done all day, and I left it running in my studio on two computers for 24 hours. I came back and it had created about a couple thousand images. And I was looking through the images, not in this precious way like I was looking at my first images, where [I was] doing it myself, so I had created 30, 40 semi-random images. I was just looking at these hundreds of images, flying by on the screen. And I’m telling you, they were better images, they were more interesting — visually — than the ones that I had made myself, where I was guiding the parameter changes. Because by virtue of the randomization and running it continuously for 24 hours, it had come up with possibilities that I wouldn’t have; I wouldn’t have gone there, you know? There would be some sort of conservatism in my choices that would be like, “No, I don’t know if that looks so good. I’m going to pull this green slider back,” or you know, whatever. It took removing myself from the system, aside from the general stochastic design, to actually create images which I thought were really, really interesting. Much more varied, much more vibrant colours and things that I just wouldn’t have gotten to myself. There’s a sense that the materials and the approach to them, and kind of removing yourself, has resulted in something that I am so much happier with, that I find so much more interesting.

There really is something to that phrase of “seeing how the sausage was made.” When you removed yourself from that process, you were able to come across it… not in the sense of the beginner’s mind, because the beginner’s mind is something that you intentionally step into, with that condition placed on how you perceive or how you react to stimuli. But I think it’s more along the lines of the privilege that somebody [has] who is paying for the album of their favourite artist and gets to listen to it, whereas the artist was there during the process. And presumably they’re proud of their work afterwards, but being able to hear it without being privy to the process has a certain power to it.

Sure.

Not necessarily something that is better or worse…

Well, it’s almost like the “better” or “worse” is some part of the decision-making, it’s really where the authorship is. I mean, there is this kind of sense of, “Okay, so you put these random things together, and they make cool images,” but at the same time, it’s the validation, saying: “Yes, these are the best ten images.”

Yeah, right.

So as I said, I made a couple of thousand images. I didn’t produce a couple thousand images, I just let the computer do its thing and then I selected [the better images] along with the algorithm. But even after the matching algorithm, there are still too many images. So, I still have to go through them and choose them and [decide on] colour combinations. You know, that part I think only a human could do still, because I am a human artist and I just happened to make a machine that makes the art for me, but that’s still from the perspective of the human artist. I think that’s a super, super important part of it. At the end of the day, it’s the decision to say, “Yes, this. This goes on a canvas in a gallery, or in an art space” (Fig. 4). That is why a readymade can be there, that’s why anything can be there. You know, there’s that impulse that is still the most important one in terms of authorship.

I can’t help myself, I have to ask this, because I’m really struck. You said the art factory doesn’t need humans, it doesn’t need us to look at it. But now, from what you just said there, unless I misunderstood, it seems as though the validation stage still needs you, is that not the case?

Yeah, I mean, I’d say that the entire piece still needs me — the author — because it wouldn’t be without me. And I mean that not just in the sense of putting it together, but [also] in the sense that I could not show it — I could make it, I could not show it, and it would not be actually complete. So, it does need me as a kind of force behind it. And I would say that the pieces that I make on canvases, they operate as works within a work. So those are [individual] pieces that somebody could have in their living room. I selected those works out of a preselection that the computer had made for me. So, those are more in line with the idea of myself as an artist asserting myself as the author and, by that, taking the risk of being sued, but also collecting a cheque if somebody buys one. But the piece itself, when it’s actually running in the gallery, is not a painting; it’s an art factory that is performing 24 hours a day. And in that I have removed myself. In its performance form I removed myself entirely, in the sense that all the choices that I have made, or the ranges of possibilities that I felt were appropriate, and that made sense, those decisions have already been made. And now it’s playing, without me, for a month, or for a day, or for three months, in a gallery space that has no windows. The images will be different in a gallery space that has lots of windows, they will be different in each show; in each hour of every show, it’s making different images. And that’s entirely without me as somebody that’s watching over it, or giving it feedback or anything like that. It just operates continually with its own kind of circular logic: making images, determining if they’re similar enough and posting them, if that’s the case.

Well, that’s really lovely. And again, I’m thinking about the contingency of stepping into a studio at a particular time. And whatever happens in the space, happens, right?

Yeah, exactly.

Great. Well, this was fun.

Yeah, thanks, Michael! I enjoyed that.

January 2019, April 2022