This story is over 5 years old.

Erin Gee Makes Feminist Robots That Respond to Human Emotion

The new media artist hooked some method actors up to biofeedback sensors, mapped the data to correspond to particular lights and sounds, and, in essence, made it so the robots reflect a live emotional, human performance.
December 16, 2014, 5:00pm

Broadly is a women's interest channel coming soon from VICE. Follow us on Twitter, Facebook, and Instagram.

I don't really know what I expected going into my conversation with Erin Gee, a new media artist and composer whose work, "Swarming Emotional Pianos," premiered in Montreal this weekend. Robots are a hot topic right now, but my discomfort with them—which is probably the same discomfort that most people have with them, obviously based on sci-fi but nevertheless not totally shakeable—means I generally avoid the conversation about a future that looks increasingly artificial. I'm also not that interested; I've always had the vague and "problematic" sense that robots are "for boys," not only because they're usually advertised as such but also because they're cold and emotionless, which has been my experience with boys thus far. I feel a certain amount of guilt about this—both for being on the wrong side of history and on the wrong side of feminism—but not enough to actually, you know, do anything to educate myself.


That is, until I heard about Erin's robots, which seem decidedly less robotic according to my misunderstanding of the word. Her work, which incorporates video, performance, audio, opera, and, yes, the robots, uses the human to examine the digital, and vice versa. Gee's newest piece is a group of mobile robots featuring light and instrumental components. These robots form a moving orchestra of light and sound, but their conductors are not traditional; rather, they're driven by emotions. Working with the Australian neurophysiologist Vaughan Macefield, Gee hooked some method actors up to biofeedback sensors that measure physiological responses to emotion (heart rate, breathing, sweating, blood pressure, and skin sensitivity/neural activity), mapped the data to correspond to particular lights and sounds, and, in essence, made it so the robots reflect a live emotional, human performance.

Despite all of this information, plus lots of cool videos of Erin singing along with a choir of herself in a series of MacBook monitors, I still had no real idea what was going on. Ultimately, I wanted to know: Are robots people, or are people robots?

VICE: There seem to be three fundamental elements in your work: robotics, sound/music, and "humanity," which we could say is the emotional/non-analytical stuff that drives us and/or gets in our way. To me—and to many—there's an implied struggle between humans and robots, ongoing until they take over the world. But how did you come to incorporate music?
Erin Gee: I have always been a very musical person, but I never felt drawn to making straightforward songs. Since I started going this route a few years ago—with musical sculptures—I came to appreciate new media practice as something that could become sonic, multiple, echoing and uncanny, rather than the visual, mirror-like aspect that is heavily theorized through such writers as Rosalind Kraus. If we can have an ocular-centric narcissism, I would like to experiment with how new media can be echoist and sonic.


OK, first: can you explain what "new media" is?
I think new media practice might just be digital art practice. I think it's a bit expanded; maybe I want to say "digital art practice." I have never been too precious about words that describe new practices like this. What I mostly mean, I think, is just that if you are working in technology as an artist [you're engaged in new media practice]—using technological languages, tools, and objects in your work. New media allows for something beyond the digital, but honestly this is something I try not to think very hard about.

So you see something inherently wrong with new media being "ocular-centric"—tending toward the visual—then? It seems like our culture/lives are oriented toward the visual generally. I mean, I just used the word "see" for "think."
I don't see anything wrong with an ocular-centric position, except that it imposes the eye's mode of relations in a normative way. Take, for example, the modes of reflection: A mirror is a visual reflection based on an object, which reflects back to the subject in relation to the position of the subject. It is a controlled, subject-centered experience based on the position of a subject in relation to an object. And Kraus says that narcissism allows the viewer to ignore media altogether and only see themselves, or their psychological projection of themselves, as opposed to the apparatus. This is a way of seeing the world; it's very popular.


No kidding. Can you explain "echoist" now?
I think echoist new media presents a problematic third body, an annoying environmental echo that can't be fully controlled because the echoes belong to, and emphasize, the psycho-spatial environment. Narcissus finds his reflection comforting, seductive—but repetition isn't all he wanted. If it was, what was wrong with Echo—why did he find her repetitions of his voice irritating?

I think there's something about consciousness there. A mirror is obviously an object, whereas Echo was a nymph, an in-between object/subject. So I think that having in-between objects play at actually being human is uncanny valley territory that reaches back into our most fundamental mythologies. I don't want to say one mode or another is more valid—just that narcissistic relations are obviously more comfortable and useful for capitalist technological production. Artists use both.

Do you see the robots as a means to the end of creating "emotional pianos"?
I guess from looking at robots not as opposing, physical, non-souled mirrors to our humanity, I like to consider the robots in this project through sonic metaphors: maybe they are amplifiers to our humanity, our emotions as expressed through physical actions in our bodies. Maybe they are pesky, uncanny echoes in our environment. But there is a part of the human in every robot made; there is certainly a part of me. Furthermore, this project is a lot more about humans behaving mechanistically than the other way around: by asking method actors to control and perform their emotional states in order to create physical changes in their bodies [by raising their blood pressure, heart rate, etc.], they are really using their emotions as a means to manipulate their bodies in very physical ways, like a cello player would manipulate strings on their instrument.


That is totally fascinating—especially if you consider the robots to be acting as "sonic metaphors." What about the movement aspect, though,? The "swarming"—where does that fit in? Why have the pianos move?
The movement research is still in progress. Ideally I wanted the robots to move because this makes them more echo-like, environmental. A moving target cannot be easily seen as a mirror, but something other. I want to push that otherness, but of course at a certain point the artwork becomes its own thing, beyond my intent.

I listened to a video in which you explain that your ultimate goal with the project is to "help people better understand and express their emotions." You mention how data-mapped emotions could be particularly beneficial to autism patients, which I can definitely understand, but I'm wondering about the broader implications of drawing emotions out of people.
Hmm. On further reflection I'm not sure if that is such a great ultimate goal, but generically I think it is a fine one for an artist to have.

It's just that to have a robot show you—or demonstrate, to use a less visual word—what you're feeling seems like a strange perversion of our current fears of a robot takeover. My initial reaction to this idea is something like, "Should people have to learn emotional intelligence themselves? Is this making things too easy?" Obviously—and you mention this—autism patients are something different than my ex-boyfriend who hasn't cried in ten years, but do you get what I'm saying here?
Some people will see this as utopic, others terrifying—I think if technological work is too comfortable, it becomes a technocratic pat on the back for a smug audience that wants to feel good about a comfortable, naturalized future where everything is the same except that it is Plexiglas and backlit and more convenient. I am not sure that I succeed in making this work creepy enough, actually. I hope to make it creepier in the future. But also more playful. Joy in the de-naturalization of the human body is something I like trying to put into my creepiest technological works. I feel like the best people working in new media are like psychological graffiti artists, creeping around scrawling "HUMANITY IS AN OLD MAN" on institutional walls.

Gee as chorus girls in Sydney

Do you see your work as mapping data that could later potentially answer whether there's an inherent structure in our emotions—like an algorithm? And how do you imagine a reality in which we can answer that question? It seems alternately amazing and horrifying, to be able to anticipate feelings like that.
If there is a structure to our emotions, I am curious to see if a language can emerge, but this will only expand a poetic vocabulary. This poetic vocabulary can be very useful to certain people and maybe allow them to "speak." At the moment I am working on developing other musical instruments for organized emotional events. I think that what counts as a voice changes what we define as a human with agency, and I am interested in giving agency to people who do not have voices like our own. So yes: if I can collaborate with scientists and medical professionals on projects that would benefit these populations it would be fantastic. I think that to most people the ideas behind this research can be terrifying, but it might be the same as that terror of having a prosthetic limb replace your natural one. If you aren't in need of this technology, it is unnatural and probably unnecessary. But it can also vastly improve some qualities of life.


Could you imagine research like this becoming dangerous?
So many technologies that we consider scary are going to get developed and sold to us in the next ten years whether we like it or not, and even though I may never have the resources of Yamaha or Google, I think that having a diversity of voices working in these fields will benefit the research immensely. You may be comforted by the knowledge that the ability to completely and accurately read the human body's physiological signals is still a long ways off—it turns out that most humans are special snowflakes in the way that their bodies process emotions. In fact, something as subtle as the "direction" of the emotion—whether you are angry at another person or at yourself—can change your body's response to it.

Rather than profess that I have discovered what happiness "sounds like" (I cannot), I am much more interested in working with the bodies of others, in instrumentalizing their bodies through emotion to see what the musical potential of the body and its rhythms might be. Of course, my musical algorithms bastardize the signals somewhat, but this struggle between honoring the language of the other and translating this language into something meaningful and unique in my art and music—this is something I'm always working through.

You're also explicitly engaged with feminist issues. Do you see feminism fitting into this work in particular?
As I said before, I think it is important for a diversity of people to be working in knowledge production and developing knowledge about our physical world, and that this is a feminist act. I also think that my work is often about trying to give space to otherness—whether through collaboration, through algorithmic processes, or moving through notions of performance, there is a political aspect to how I am trying to be a composer that makes a lot of room for otherness to speak. In this way I'm also inspired by other female composers who work in developing alternate musical languages, such as Pauline Oliveros, Yoko Ono, even Marina Abramovich's early performance work was described through her as rhythm. The ordering of disparate elements into communication is a political act—in the original "Cyborg Manifesto," Donna Haraway emphasizes a constant struggle for language, to see language as technology, to build new languages. This is how my musical work is most feminist. I have always found myself trying to argue a space for artistic creation that is feminist through its materiality, and furthermore through the emphasis on the body as a source of knowledge.


Wait, what do you mean by "materiality"? I understand the emphasis on the body as a source of knowledge being feminist, but I thought materiality would refer to the physical stuff you're using to make the work—is that wrong? Do you consider the body and its responses to emotions your artistic materials?
I guess the materiality of my work—and I'm drawing these dangerous analogies here, but I can remember a historical moment when men wondered if women have souls. The objectification of women in modern society is well-documented and known. Women today are used as political tools among those in power; their personhood becomes drawn up in their rights to their bodies, the right to deny sex as well as their ability to bear and raise children. There are layers of interrelation that all humans have, where we are objectified, reduced to our utility, treated as objects rather than subjects, reacting to and acting through a subject or group of people more powerful than we are. I think in this way I want to make work that tries to bridge the gap, to create music that has the subject trying to learn the language of the object, to assume that in fact the thing we assumed to be an object in fact has a voice, and it is important to listen. To have a voice is to have agency and power.

When I speak of a feminist "materiality," I can personally relate to the feeling that I am regarded as an object that sometimes speaks, because I live this as a woman. I understand that there are many other people that positions like this can speak to, so these elements remain abstract, but discovering or seeking out voices in electronic objects or processes is something that gives me joy. I also acknowledge that this position is part fantastical, part cynical as fuck. Maybe fantastical cynicism is something we should add to creepiness as a useful tool.

Do you put stock in the idea that feelings are somehow inherently female and, on the flip side, logic/reason inherently male? Would you say logic governs the technological aspects of your work? There seems to be a tension between programming—inherently ordered and logical—and the spontaneity you use.
I'm not sure that I see the programming aspect of my work as distinctly logical and ordered as opposed to the ways our bodies process emotions. When you are alone in the woods and hear a twig snap, your body automatically widens your eyes, increases cortisol production; your breathing becomes deeper, and depending on how your fight or flight reaction is, your blood will be pumping heavily, readying you to flee; or freezing up, preparing you to play dead. This too is ordered and logical, for your body's survival.

I guess what I'm wondering about the emotion/logic dichotomy has more to do with all of us being "special snowflakes" in our emotional responses. Like, I struggle to reconcile that individuality of response with the logic of survival instincts. Yes, there are certain emotions and emotional reactions we've evolved to have, but on topics not so obviously evolutionary as, say, fear or sexual arousal, emotions/responses are not so logical, right? Someone might react to smoking weed with paranoia, someone else with sleepy contentment. Or you might flip out when your roommate leaves his milk on the counter, but your other roommates are like, "Dude, who cares?"
Well, the special snowflakes of emotion thing is being investigated currently; emotions beyond survival instincts present challenges to research that I'm sure scientists probably won't figure out for a while now. This is because emotions were first conceived of in this way by James-Lange theory in the 1880s—Darwin, yo!—when they were first imagining that perhaps it isn't because we feel scared that our eyes widen and our breath becomes sharper, but rather, the other way around! This was drawn up at a time when evolution was a hot topic. Through the centuries the conversations continued, but adequate tools for measuring physiological markers of emotion weren't really in existence until the 1970s. So we're not even 50 years into examining the ideas of 200 years ago. Things as subtle as regret, jealousy, and mild irritation (though I think I can tell when the actors are annoyed with me) are difficult, because at a certain point we humans are probably not the sum of all our parts. There might be something more.

"Swarming Emotional Pianos" was produced by the Canadian chamber music collective Innovations en Concert. For more information visit their website.

Follow Lauren Olyer on Twitter.

Broadly is a women's interest channel coming soon from VICE. Follow us on Twitter, Facebook, and Instagram.