When the Artists Met the Algorist

Roman Verostko’s importance to the history of generative art is unquestionable. A recipient of SIGGRAPH’s Distinguished Artist Award for Digital Art back in 2009, his legacy is evident from the artist’s online archive, which elaborates his remarkable journey from Benedictine monk to a leading figure among The Algorists. Historical TV footage like this from 1989 is a reminder of Verostko’s vital importance in bringing the conversation around art and technology to a mass public audience.

At a moment when generative artists are renewing painting, Verostko’s brush plottings prove that the physical and the digital have always been in dialogue. His work also unites Western and Chinese traditions in a way that foreshadows the global conversation that has become a hallmark of digital art since the NFT. Today, RCS is pleased to bring together two generations of artists in a conversation for the ages.‍

Alex Estorick: What distinguishes The Algorists from other artists working with computation?‍

Roman Verostko: The term “Algorist” applies to those artists who write code to generate art. I was a professional artist, exhibiting a retrospective in 1965, and brought certain mature ideas with me in my approach to writing code. My goal was to bring my art-generating ideas into software itself, so that it becomes a kind of artificial intelligence with the knowledge to execute visuals and concepts that I had in mind. 

From the very beginning, my entire process was a confrontation with personal experiences that I translated into visuals. I always worked with visual oppositions and came to understand interactive color from studying Josef Albers. For example, why is it that if I place a red and a green of the same value and density of hue side by side, they produce a vibration and complement? In philosophy and theology there are extremes and experiences that are in conflict with each other that I have tried to resolve. In my art, I created contradicting structures in the same field, trying to bring them to a resolution. I even gave my code the name, Hodos, meaning “way” or “passage” in Greek. 

When Jean-Pierre Hébert [who wrote the Algorist Manifesto] and I spoke, one of the things that was clear to us was that the coded procedures that embody an artist’s concept of art-making are unique. An artist like Manfred Mohr writes all his own software, and the code itself embodies a generative concept that is unique. That is an Algorist. I looked to artists like Harold Cohen and Vera Molnar, who understood that. ‍

AE: Are you saying that an Algorist has their own identifiable style?‍

RV: You could call it style, but I consider it to be a concept — an art-generating idea. ‍

AE: A show of Harold Cohen’s work with the program AARON has just opened in London. Do you feel an affinity with him as a painter-turned-programmer? Where, for you, does Cohen sit in the history of art?

‍RV: Harold Cohen’s program, AARON, was his alter [alternative personality]. Harold was there when I presented my paper on “Epigenetic Painting: Software as Genotype” in 1988 in Utrecht. We were even on the same panel and became good friends. When he received his big award at SIGGRAPH, I made sure I was there in Vancouver with him. That was pretty much the last time I saw him, and we had lunch. 

He had a whole lifetime of art and his struggle with the code was always to get it to that final touch where it was him. But you never get there. 

I remember at the very end of his talk, he picked up a pencil or brush and said, “Well, it didn’t quite get there. But I’ll just take this pen or brush…” and he made a couple of strokes towards the screen. That was it, and then he sat down. What he was really trying to say was, “in the end, it’s your hand and it’s never going to be something else in code.” I don’t necessarily share that perspective. I think several of my projects have been quite successful in joining technology with art. At the University of St. Thomas in Minneapolis, I have an 11-panel mural that embodies every aspect of my thinking up to that time in 1997. I think it’s the largest pen plotter drawing project in the world. It took over a year and was extremely difficult to achieve at that scale with steel-tipped pens, which often make sharp cuts into the paper. 

One of the most important projects in my life is the limited edition with illustrations of George Boole’s Derivation of the Laws [Chapter 3 of his book, An Investigation of the Laws of Thought, 1854] published by St. Sebastian Press in Minneapolis in 1990. That was intended to demonstrate the arguments in my paper, “Epigenetic Painting: Software as Genotype.” 

I still believe that code at its very best is a family of forms generated by the same parent code, with the only difference being that it is seeded with a different number.  

An advisor from the University of Minnesota, who worked with probability theory, taught me that there’s no such thing as a random act by a computer — rather, it is a pseudo-randomizer. It was my choice to celebrate the important contributors to the evolution of the computer: George Boole, Alan Turing, and Norbert Wiener. I paid homage to Wiener with my Decision Machines, which have circuits where you push a button and they make a decision. It was Wiener who anticipated that humans’ inability to control machines would eventually become a problem. 

Aleksandra Jovanic: In your tribute to George Boole, you wrote the following: 

‍Now, in the summer of 1990 we recognize that electronic art is still in its infancy. But, as in the early 15th century, there are artists now wrestling with the problems of the transformation and creation of art forms with computers. The new geometries and new technologies will surely bring a revolution as we approach the next century. ‍

32 years on, has the revolution happened and, if so, what are we wrestling with now? ‍

RV: It has happened, and it is happening, no question. I’m a historian and I taught art history. When I was a monk, I also edited the 1967 edition of The New Catholic Encyclopedia. In my lectures, I always cited the great moments in history. I grew up at the peak of the Industrial Revolution, which fed technologies into and after World War Two. 

Following the Industrial Revolution, we had the beginning of the information revolution in the 1960s, which changed everything — it is unbelievable how far we’ve come. And if you listen to the lectures of Yuval Noah Harari or read Sapiens (2011) and Homo Deus (2015), he’s right about the evolution of human consciousness and language. We’re going through an evolution now, where there is an intelligence that is extended from humans. Humans are using machines to enhance their ability to confront problems and to create art. Norbert Wiener was right as well. His 1948 book, Cybernetics: Or Control and Communication in the Animal and the Machine, showed an understanding of that. 

I remember discussing my Universal Turing Machine at a conference. I asked a mathematician friend, who was also a sculptor, what he thought about it and he said: 

“Roman, I’m going to write you a Universal Turing Machine right here, right now,” and he took a pencil and wrote down a one and a zero. That was it.  

And it unfolded in my mind, [recalling] something I had learned when I was studying theology. We had read Aristotle’s Metaphysics together with George Boole. I came across a passage in Boole’s third chapter and I almost passed out — Boole’s entire thesis hangs on a well-known quotation from Aristotle: the law of non-contradiction, which says that “a thing cannot be and not be at the same time in the same circumstances.” That is the whole ball of wax. When I [realized] that, it just unfolded. This is pretty much what we all work with — the switch is on or it’s off.  ‍

‍Alex Estorick: Can you take us back to the earliest origins of the algorithm, to the medieval period and specifically the world of ninth-century Persian mathematician Muhammad al-Khwârizmî. How did the Algorist movement evolve from those initial principles and was your involvement shaped by your own experience as a monk?‍

RV: In my theological training we all had to learn medieval philosophy. We studied scholastic philosophy that led up to Thomas Aquinas’s Summa Theologiae (c. 1265-74), where logic dominates [and] you have to have sufficient enumeration. There are about four or five sufficiencies you have to meet in any argument or discussion. That medieval argument is close to mathematics as well. There’s a famous image (see below) of the Algorist versus the Abacist.

I can assure you, if you’re an Algorist, you’re going to beat the Abacist.‍  

KRANKARTA: When does an algorithm become art?‍

RV: An algorithm is always an algorithm. An algorithm is a procedure, it’s not art. But one could speak about an art of creating a logical procedure, which would be an algorithm. I date the earliest algorithms back to the development of procedures for basket and cloth weaving. Weaving a basket from cloth is an algorithm, to do it as a work of art is pretty common. I have a woven basket hanging in my family room which came from an African process with a deep-rooted history. That’s interesting to me. I’m always looking for an interesting algorithm.‍

Marcelo Soria-Rodriguez: You have drawn a parallel between epigenesis and algorithmic art. Do you regard your algorithmic works as a way to expand the question of existence?‍

RV: There are so many possibilities. It’s beyond anything you could imagine. I don’t think I could regard algorithmic works as a way to expand the question of existence. I believe experiencing my Flowers of Learning (2006) at Spalding University contributes to a better life, but it doesn’t expand life. It’s not measurable, of course, but I think all kinds of art and experience are healthy for a human.  

Generative art has been around since computing, since the 1960s and even since Christopher Strachey’s Love Letters (1952). As I understand it, they were posted on a bulletin board at the university, signed M.U.C (“Manchester University Computer”). For whom were they intended? Well, that’s interesting.‍

Erick Calderon: What’s been your reaction to the recent spike in appreciation of generative art?

RV: I’m not aware of that, I’m afraid. But I am transforming some of my work into NFTs. 

‍Iskra Velitchkova: In long-form projects, where the artist releases a series of outputs from a single generative system, which is the artwork: the individual outputs, or the series as a whole? 

RV: It depends on the format. If you look at a work like The Magic Hand of Chance (1982-85), unfortunately, there’s no way for me to show it as a forever piece that will generate except with something like DOSBox. I think that is the first true generative art ever created with an IBM PC (model 5150) which was issued in 1981. It’s a little embarrassing for me to say that because the code is written in old fashioned GW-BASIC, but it works. It’s also some of the first generative art made with full color. I managed the trick with three colors at a time, but I would shift to two different screens, and I had all kinds of tricks to get six colors. But it works.‍

AE: Do you regard the code as the artwork? Because you’ve been using a plotter to produce physical outputs for a long time. 

RV: A drawing is a work of art. But is my code a work of art? I wouldn’t think so. If you were able to read it, I think you would discover that it’s pretty lean and pretty good for what we’re able to do with it. You could look at it as the art of coding.  

‍I’m going to have all of my software preserved at the Verostko Center. I don’t know that anyone will ever be interested but that’s another matter. I always considered the drawing arm of a plotter to be an extension of my hand to some extent. But they are not the same. If you are writing software for a pen-plotter drawing, you become concerned about the quality of the steel or tungsten-tipped pen. The pen point, the surface of the paper, the ink, and the speed with which the machine is allowed to draw — getting all of those things correct is an art in itself. But it is not the same as drawing by hand. ‍

AE: You’ve also used a plotter with a brush. How important is Chinese calligraphy to your work?‍

RV: I went to China in 1985 when I was invited to teach a course on modern Western art. Through that, I met Wang Dong Ling, who was a master of shufa [Chinese calligraphy]. He was later invited to the University of Minnesota to demonstrate shufa, and he lived with us for a while. He had a brush the size of a broom, which he kept in a bucket of water in our bathroom. He could manipulate that brush and make huge marks. I would do a pen plot and he would make brushstrokes to produce a collaborative work. 

When I came back [from China], I brought some brushes with me but I didn’t know what to do with them. Ultimately, I thought, “why not use this with the pen plotter?” So I built a special program to start and stop the pen and control its speed. Once I got control of it, I realized that I could make a brushstroke and remember that information. I could then repeat it and do twists and turns with it, so it became another tool. I think that it brought another quality to the work. ‍

Anna Carreras: How important is it for a generative artist to craft her algorithms from scratch instead of using libraries or existing software?

RV: I did it from scratch, when there was no software. I’ve never used software off the shelf. I’m pretty much self-taught in everything I’ve ever done. But I’m sure there’s a lot of software I would like to get. 

‍Looking at my Upside Down Book or my Upside Down Mural drawings from 2008, if someone could show me how to generate these characters into a three-dimensional existence, I would be interested. But I’m at that stage of life where I’m not up to a whole new project. I have hand drawings, made during COVID, that are really interesting but I haven’t shown them. I think I’m going to go back to them. I’m doing what I call “mergings” where I allow myself to do whatever I want, with no rules. So if I take a pen plotter, I might take one of my old drawings and make some hand drawings and mix things up — take parts and play with whatever I want. There are no rules. This is a new phase. ‍

‍Nat Sarkissian: Are there any ideas or projects that you haven’t been able to get to yet that you’ve always wanted to? ‍

RV: Oh yes, that’s the book I haven’t written. ‍

NS: Having seen such an increase in computing power over the course of your career, are there any techniques that you were excited about as hardware allowed them over time? ‍

RV: I think it is an exciting time now. We’re in a transition — goodness knows where it’s going to go. I didn’t see it so clearly 30 or 40 years ago, but computing is in evolution and we’re connected to it. Humans wouldn’t survive without it. So I think the problem going into the future is whether we will be able to sustain the human population we have. How will we do it? We can’t continue as we are right now. The current course is a dead end because the pollution will destroy us. Will we get through it? Can this technology be helpful in the end? Let’s hope so.‍

AE: A lot of generative artists see natural and machinic systems as closely related. What is your perspective?‍

RV: Oh, yes. There’s no question. We wouldn’t survive without the information systems we have today. Those systems are connected to us. Everybody who writes software uses expressions that relate to human experience. My 1988 paper said that a generative piece of art that can generate a family of forms is indeed an epigenetic process, there’s no question. With an acorn that grows into a tree, that’s an epigenetic process. How a mother carries a child in her womb, that’s an epigenetic process. 

Generative code is an epigenetic process that relates directly to a biological process. So we cannot separate human thinking from so-called artificial intelligence thinking anymore. We need it to survive. But it can also become criminal just as well as it can become something of great service. 

Where will it go? You want the good people to hang on to control because those in politics use the same systems. We don’t want it to become corrupted. I think it’s a biological extension. It’s separate, but it isn’t separate. It’s a unique time in evolution. ‍

IV: In the process of building a generative system, what is the balance between the artist and the machine?‍

RV: Ultimately, I think the artist has to remain in control. You don’t want the generator you make to turn on you. You don’t want it to be used for the wrong things. To end up doing generative work that’s destructive is always a problem. ‍

AE: It seems to me that you do not distinguish between generative systems as aesthetic and political mechanisms. Would that be fair?‍

RV: I would say that generative systems can be used for anything. Let’s say you want to generate propaganda or create some brilliant way to get people to experience something that changes their view. That’s powerful stuff. When I was a Benedictine monk in the 1950s, I made a Veronica’s Veil. Shortly after George Floyd was killed, within about 13 blocks from where I live, I looked at a photograph of that veil, and I thought, “my goodness, that could be George Floyd.” This led to a new work. What is art for? Is it to teach moral lessons? I think that, at its best, art is just a beautiful experience. ‍‍

‍Anna Lucia: At the moment, as generative artists, we can constantly talk with our peers through online networks and get instant feedback, exchange techniques, and examine each other’s work. We can also find online tutorials, blogs, and YouTube videos about different techniques. How did your community of artists exchange ideas and techniques?‍

RV: I didn’t communicate with that many. But, in the late 1980s and into the ’90s, I spent days at a time with my friend Jean-Pierre in California. I also attended conferences, which were the most important thing for me. I was director for the Third International Symposium on Electronic Art. We also had a small group at Minneapolis College of Art and Design, although there was a tremendous amount of resistance to the computer art that I did. I was a humanities professor and some of the faculty didn’t want me to introduce computer art in my classes. But there was a lot of communication among those who were interested.‍

AE: Recently, “generative art” has gained prominence in the collective conversation, but different terms hold sway at different moments, from “computer art” to “Algorist.” What’s the right language that we should be using right now, in your opinion?

RV: Generative art is algorithmic art. It’s not new. We’ve been doing generative art from the beginning. 

‍[...] Of course, “computer art” was the term that was used in the 1980s and ’90s. I spoke of my work as “algorithmic art” because I understand that it’s not the computer but the algorithm that generates the work. But, of course, the computer is the tool that drives it. As I’ve said many times: the plan of the architect, the score of the composer — those are algorithms. There are all kinds of algorithms for different things. 

Language is a code, and learning code is a wonderful thing. There are many ways to code and signal meanings and I think we have to appreciate the people who create the codes that we use in everyday life.‍

AE: Are Algorists and generative artists one and the same?‍

RV: Yes, of course. The thing is, if you are a generative artist, I assume you’re writing software. [...] I think it depends on what the algorithms do.‍

AE: A lot of people are using other people’s code today. Does that undermine generative art in your view? Is originality fundamental to generative art or algorithmic art?‍

RV: Yes. You’d have to recognize the composer if you played a piece of music that was written by them, so you would have to credit the source of the software for generating a work. [However] one can structure software in such a way that you give options for so many different factors that, pretty soon, the person who learns how to manipulate those relationships becomes an expert user. Then that person becomes an artist in another way. It is just a matter of whose art is involved. 

Whose art dominates? That would be my question, if you want to hand out credit. 

You could get hold of my software, and, because there are options for some of my routines, you might discover things that the creator of that software didn’t know. That would be interesting. My wife was very good at running my software. She understood it very well. And when she did a drawing with my code, it was always a little better than mine.