
Kevin L. Ferguson, summed frame visualization of ‘The Searchers,’ directed by John Ford (1956) (courtesy the author)
It’s early enough in my career that I still get a thrill whenever I see my work shared online. As a scholar who works in both film studies and digital humanities, I use public domain scientific image analysis software to create “sums” of films, adding together the frames of a film to make one single abstract image. Since 2013, I have shared hundreds of these on Tumblr, Twitter, and Flickr, and written about my process in both popular and scholarly venues. Last year, the circulation of my work on social media brought me some attention, and I was interviewed by the Huffington Post and invited to contribute to The Best American Infographics 2016. I did some custom work, gave some talks, and wrote some book chapters, and when the venerable British Film Institute posted a link to my work on Facebook, I pretty much thought I had arrived.
So, last month, when I happened to see a British gallery tweeting about my work, I was curious and excited to see the context. I wondered how they found me? What other artists would they mention as a frame of reference? I wondered if they would want to display any of my work?
But no. What I was seeing was an announcement for a show by Jason Shulman at Cob Gallery called Photographs of Films. The press and interviews collected on the gallery’s website lauded a conceptual beauty and rigor in his work, but the only thing I could see was a rip-off. “Email for price list.” These images were unmistakably similar to the distinctive work I had been producing for years, and it was not long before friends started writing to let me know.

At left, the flier for Jason Shulman’s exhibition at Cob Gallery with a summed image of ‘Fantasia’ (screenshot by the author); at right, Kevin L. Ferguson’s summed frames of ‘Fantasia,’ directed by James Algar et al. (1940) (courtesy the author) (click to enlarge)
I am embarrassed now to recall how unseemly it felt to write the gallery before the opening to let them know about the situation. I was imagining the great expense of running a posh gallery in London, the effort of publicity, the time spent producing and mounting work, and here I was asking them to reconsider running the show in its current state. But I thought there was no way that the gallery would have been aware of my work beforehand, and so I naively expected them to be properly horrified at the obvious similarities and to respond appropriately.
Instead, two days later, I received 575 words about copyright in 10 contradictory bullet points. I started to see more press and more publicity. Photographs from the gallery’s opening, with well-known actors in attendance, appeared. Filmmakers praised the work. “My” work appeared all over social media and, maddeningly, on a growing number of websites that I admired.
In the message she sent, the gallerist was arguing that regardless of appearance, if Shulman and I used a supposedly different process to create the same image, then by definition his work had to be substantially different from mine. The rhetorical move here was helpfully explained to me by a lawyer, who also explained that it would be a very expensive case without any real financial reward, and that the gallery likely knew that and would just do whatever they wanted to anyway. Although I never mentioned copyright in my email, I could easily see the line that was being drawn: if Shulman’s work was created in “a different way” than mine, then it would not violate copyright, and if it did not violate copyright, then the gallerist could see no reason not to continue the exhibition.
It was hard to respond to this. In interviews, Shulman is very coy about his process, which was the first thing that bothered me about the similarities between our work. In a write-up in Wired, we learn that “Shulman won’t share many details about his process, but says he photographs the films ‘off a very, very high resolution monitor with a very big camera.’” Unlike Shulman’s strange secrecy (his process actually sounds pretty simple), for me it is important to share both the intellectual context as well as the specific nuts-and-bolts of my process, so that others can make use of my work. In this way, sharing my method not only satisfies the most basic requirement of ethical participation in humanistic culture — citation and the acknowledgment of other artists — but also reflects an awareness of a scientific tradition of reproducibility and an intellectual commitment to sharing knowledge freely.

Kevin L. Ferguson research on the “style space” of 176 Disney, western, Japanese, and gialli films plotted by entropy (a measure of randomness, on the x-axis) and mean intensity (i.e., brightness, on the y-axis) (courtesy the author)
So, in discussing my work, I acknowledge visual artists whose work I have used as a conceptual base to expand my thinking about film, the digital, and duration, including Hiroshi Sugimoto’s Theaters from the mid 1970s onward, Jim Campbell’s Illuminated Averages series (2000), and Jason Salavon’s Portrait (2009/10). Independently, other bloggers confirmed my sense of this mini-canon, remarking on similarities between these figures and Shulman. For instance, responding to the press around Shulman’s show, Michael J. Wilson criticizes the particular erasure of Sugimoto, a well-known Japanese photographer often exhibited in London. Ron Kretsch more directly questioned Shulman’s Wizard of Oz by placing it side-by-side with a much earlier version by Campbell. Austin Kleon compared Shulman’s work to mine in the context of captions. Because this work is so conceptually specific, its forebears are all the more easily recognized, and there is more consequence in failing to acknowledge these predecessors.
Beyond the imperative to acknowledge previous work, I share my work and process in a scientific tradition of reproducibility. If these are ever going to be anything more than just pretty pictures — whether that means a scholarly use for media studies research or as transformative artistic work in a particular aesthetic tradition — then understanding what these images do and how they work is critical. This tradition of reproducibility does not mean that art must be like science: reductionist, experimental, or empirical. Rather, it means that processes underlying the work should be made available to others, so that others may likewise experiment, expand, and engage with my practice. This is a core value of the digital humanities, particularly the tradition of deformance (Jerome McGann and Lisa Samuels), screwmeneutics (Stephen Ramsay), and weird DH (Mark Sample), all of which are centered on play and the use and reuse of other digital materials. While the art world has its own tradition of theft and attribution, as part of the broader humanities, the art world today confronts a crisis of intellectual property, commerce, and new forms of dissemination. In this new internet ecology, a gallerist staking a body of work on technicalities of copyright alone will no doubt find her endeavor doomed.

Kevin L. Ferguson, montage of the summed frames of 54 films produced by Walt Disney Animation Studios, 1937–2014 (courtesy the author) (click to enlarge)
While I find these abstract images beautiful and engaging on their own, I also use them as a form of media historical research that I call “digital surrealism,” treating the abstract images in their own right in order to study media history. For example, by comparing larger groups of summed films we can ask new research questions about film genres, historical periods, or national cinemas. Have animated films gotten darker over time? How distinct is the color palette of western films? Does Asian cinema frame its characters differently than Hollywood cinema?
But because Shulman is ignorant of this tradition that examines cinema and duration, his work fails to contextualize motion pictures in a meaningful way or to engage with cinema itself beyond clichés, such as his claim that these images show how “lots of Bergman films are kind of moody and psychological, much more so than other films,” or his misguided metaphor that “each of these photographs is the genetic code of a film — its visual DNA,” a lazy description that makes me uncomfortable as both a film scholar and a movie lover. Having thought about these kinds of images very hard for years, I see in Shulman a distinct lack of that quality of reproducibility that allows for deeper engagement, a failure to recognize prior work or to place his work in any tradition other than the superficial.
All of which made the gallery’s copyright-centric response even more frustrating. The gallerist’s email to me prioritized copyright details over common sense, placing my work in a frame of legal niceties that diminished my contributions in order to argue for the singular nature of Shulman’s work: well, he doesn’t use the credits but you do; his colors are somehow truer to the original than yours; in fact, his process is quite different and clearly produces a different, superior result. Unable to even acknowledge any of the obvious similarities between our work, the gallerist closed with a curiously sincere request: “I would be grateful if you would confirm that, in view of the above, you have no complaint with regard to the exhibition and exploitation of Jason’s work.”
“The exhibition and exploitation of Jason’s work.” What strange words to describe the function and value of art.
As far as “exhibition” is concerned, social media, search engines, and online databases have transformed the way knowledge is shared, and numerous disciplines have worked to respond to this change. For example, the New York Public Library has focused recently on expanding its digital collections, encouraging unprecedented access to, engagement with, and remixing of its holdings. The Modern Language Association just released a new version of its style handbook that departs from its predecessor’s fetishism of citation minutiae in order to show how documenting sources is a crucial way to publicly record meaningful conversations in a changing digital world. The Brooklyn Museum’s new ASK app creates a “dynamic and responsive museum” where visitors can pose geolocated questions in real time to a team of art historians and educators. Institutions like these, curators of humanistic values and culture, are shifting to embrace open exhibition practices as a way to facilitate productive connections between individuals past and present. Isn’t it reasonable to expect an art gallery to do so as well?
As for “exploitation,” this runs counter to my intellectual commitment to sharing knowledge freely. I do not mean, in this case, that art should be given away, but rather that artists and their representation should be more conscious about exactly what they are exploiting and should question the durability of a model of art centered on exploiting artists (and, as a consequence, artists exploiting other artists). “Exploitation” here requires the Cob Gallery to see me and Shulman as mutually exclusive. Exploitation requires a definition of his work as singular, sui generis, allowing no comparisons. Exploitation demands silencing any suspicion that his work is derivative in any way.
I sent one more email to the gallerist, asking to be acknowledged and credited in their literature, appealing to their sense of artistic community. This, after all, is a space that describes itself as being “founded on the principle of creative collaboration, adhering to a culture of collaboration between artist and gallery.” My email read, in part:
Last, and beyond what our lawyers might say, I appeal to your sensibilities as gallerists. Jason’s work is clearly similar to mine, and I’ve been doing it much longer than him. I can appreciate your lawyer’s advice to focus on the legal distinctions, but you must recognize that this is dubiously similar work, without much distinguishing conceptual consideration.
So, I’m asking to be fairly recognized as being associated with similar work. I would appreciate being mentioned in your literature in an appropriate way that gives me my due. My main interest is in getting credit and I’d like to work with both of you on the best way to provide attribution.
I received no response to my request. How naïve was I to expect one?
In my unsuccessful efforts to negotiate my concerns of plagiarism with the Cob Gallery before the opening, I discovered that the aims and principles behind my scholarly aesthetic practice are at odds with the reality of the commercial art world. Whereas citation, context, and conversation are important to my work and teaching, the gallerist responded to my complaint with narrow, technical details of copyright law in order to protect her interest in running the show. In doing so, the gallery sacrificed an opportunity to engage with my work or to contextualize its artist’s work in a larger field, missing precisely the values and practices an art gallery should perform in the face of digitally-produced and disseminated work like my own. Rather than “exhibition and exploitation,” I had hoped for conversation and collaboration. But for the gallery, that kind of conversation is incompatible with a privileging of exploitation as a mode for producing and disseminating art.
As a countermove that encourages others to better engage with the kind of work I make, I conclude with a description of how readers can easily create their own summed images.
- Download the free, public domain software ImageJ.
- Use Quicktime Player 7 or similar to create a folder of frames from a digital copy of a film (File / Export … and select “Movie to Image Sequence”). You can get good results using even only a frame every 2–3 seconds (try .15 fps to start). You may also want to size your window to 600 pixels to save processor speed.
- In ImageJ: Select Import / Image Sequence and navigate to the proper folder. For this purpose, you can select “Use virtual stack” to speed up the process.

Directions for creating summed images of films
- Once the folder of images is imported, select Image / Stacks / Z-Project. For “Projection type” select “sum slices” and hit “OK.”

Directions for creating summed images of films

Directions for creating summed images of films
- Share, engage, contextualize, respond, collaborate, remix, extend.

Kevin L. Ferguson, the summed image of the 55 films produced by Walt Disney Animation Studios (courtesy the author)
Thank you for writing up this article. First, let me congratulate you both on your idea, your work process, and your work itself; second, for calling out these filth.
I’ve also done work very similar to this, and was alerted by a British friend about the show you mention. I was similarly nauseated by the exceptionally poor description of the work by the artist (seriously, a film’s DNA???) and the effort to make a very simple process (using a large format camera to do a long exposure of a computer screen, something any Photo 2 student can do) sound mystical and labor intensive.
I do wonder though, about the claim that the work is necessarily derivative. What if he didn’t know of your work at all (he is, as you state, pretty oblivious to the world of film and cinema studies). I started my work in 2012, would this make your work derivative of mine?
Without mining Shulman’s web browser history, it’s probably impossible to definitively establish that he “stole” this idea from Ferguson. He’ll probably always have plausible deniability that he came up the idea totally independently. That said, it seems unlikely that a digital artist in the information age didn’t run across a virtually identical body of work that predated his. Either way, the gallery and artist would be remiss not to mention Ferguson’s work. Failing to do so indicates to me that they don’t want their collectors to draw the same conclusion that the author has, namely that this work is a fairly shameless rip-off.
As for your work, it seems quite different in intellectual interest, subject matter, process, and aesthetic result (at least judging from the image above). I wouldn’t immediately associate your work with that of Shulman or Fergson, who’s works are basically identical (legal technicalities aside).
This is an interesting situation, but unless I misunderstand, what has been appropriated is the process, not the result per se. The obvious similarity of the result is perhaps due to an infatuation with the (nonetheless impressive) technology itself, as opposed to what can be said with it. This is akin to saying that Perugino should have protested Michelangelo’s use of buon fresco, setting aside the different intentions, symbolism, themes, etc. Compare that to my situation, in which an entire online book of over 144,000 words was published, word for word, as a hardcover book in India under an invented name, with no acknowledgment of any kind. My publisher said, “Ain’t worth the trouble.”
In this case, the idea of applying this process to films *is* the work. The exact output of the process is largely incidental (though not uninteresting). In Michelangelo’s case, the “different intentions, symbolism, themes” are of paramount importance to his work.
Also, people like to casually throw around the word “appropriation” to cover all manners of theft. To me, appropriation implies a explicit or implicit acknowledgement of the work that is being borrow. We know Andy Warhol didn’t design the Campbell’s Soup can. We know Richard Prince didn’t originally shoot the Marlboro ads. Neither made any claim to the contrary. If Ferguson’s work was wildly famous and viewers of Schulman’s work would be expected to view it through the lens of Ferguson’s work, you might be able to make the case that he was appropriating Ferguson’s work. I think that’s pretty obviously not the case here. Schulman and Cob Gallery would clearly prefer that Ferguson’s work vanish.
Probably so, Monsoonking. Yes.
There’s an artist named Garry Neill Kennedy, from what is now NSCAD University in Halifax, Canada, who made a kind of low-tech progenitor to this called “Canadian Contemporary Collection – Average Size – Average Colour” (1978). In it, he took 32 contemporary paintings from the Art Gallery of Ontario and averaged their size and color, producing a 33rd canvas that sort of achieved a similar effect to that described by Ferguson, but without computation, of course.
I’m utterly confused. If you had written the gallery to inform them how similar the work was to Jim Campbell’s, I would have understood. But that you claimed Shulman was infringing on your own work, when you are already infringing on Campbell’s – I don’t understand. Jim deserves all credit for this idea. You do not. What are you adding? You use words like ‘scientific’ and ‘scholarly’ to describe your approach. How is it? You’re making this technique sound more than it is. The idea is everything – the execution is trivial.
Put another way, you are borrowing someone else’s idea to average together movies someone else made. How do you think you have any claim to ownership?
I just saw your plot where you put ‘entropy’ on one axis. Entropy actually has a specific meaning. Its most important attribute is time – it increases over time. Averages are ignorant of time. It doesn’t matter which frame came first or last. If you change the order of the frames, the average is still the same. How is your variable entropy? If it’s about randomness, call it randomness. Although I don’t see how it could be this either. Variance, maybe? Without your math, it’s tough to tell. Either way, how does this help analyze a film? This would be as valuable as counting up the spaces in novels and plotting the sum as ‘Dark Matter’, against, say, the average letter height. It seems like you stumbled on a research tool that you could load movies into and made the mistake of assuming everything you did with it would be research.
Interesting article. I see you exchanged email with a gallerist. Did you email or otherwise contact Jason Shulman directly to ask him about his work, the gallery show, etc?