After ‘a world of change,’ artist Martine Syms brings forth ‘Neural Swamp’
Based in LA, Syms is the second winner of the prestigious Future Fields Commission in "time-based media," awarded by the PMA and an Italian foundation.
When L.A.-based video, mixed-media artist Martine Syms was named winner of the second Future Fields Commission, a prestigious program run by the Philadelphia Museum of Art and the Fondazione Sandretto Re Rebaudengo, Turin, Italy, it was a whole other world.
The biennial commission is aimed at encouraging works on the outer fringes of “time-based media” — video, filmmaking, performance, or some combination.
It was awarded to Syms in 2018 with a debut for her project scheduled for Philadelphia and Turin in 2020.
Needless to say, that did not happen. The pandemic intervened and a lot of the world stopped, particularly for the art world. For Syms, whose work involves a fusion of digital media, live performance, video, writing, and animation, the stoppage was not entirely a disaster.
She deftly took advantage of the changes imposed by isolation, even as openings and rehearsal and all manner of tech work were delayed and delayed and delayed again.
The real world may have stopped but the virtual world of “time-based media” accelerated and proliferated.
Neural Swamp is what emerged, a time-based, AI-driven video installation on view at the art museum until Oct. 30.
The art museum and the Turin-based foundation jointly funded Neural Swamp and together established the Future Fields Commission in Time-Based Media in 2016, awarding each recipient $125,000 for the production of newly commissioned work. Its first recipient was Rachel Rose, whose inaugural project, Rachel Rose: Wil-o-Wisp, was presented at the two institutions in 2018-19, accompanied by an illustrated catalog. Martine Syms is the second recipient.
The Syms commission is web-based artwork. Each institution has a manual that outlines the works’ components, which are produced and installed according to the artist’s instructions, along with a kit for all technical elements needed to present the artwork — monitors, speakers, the MacStudio computer, and so forth.
“The works in the show have really changed for me over the time that I’ve made them,” Syms, 34, said in a recent telephone interview from her L.A. studio. “There are parts of the process that I thought, ‘This isn’t what I wanted to do. This wasn’t my plan.’ I know we’ve all had that feeling during various parts of the last few years.”
“I lived with it a lot in fragments and in process,” she said. “So really this is the first time I’ve seen it as it was in my head.”
The large gallery in the contemporary wing of the PMA, where Neural Swamp’s three large video screens have been installed, is bathed in an overpowering green light, as though the entire gallery has become a green screen, with those moving through it now characters in a virtual landscape.
The monitors project video avatars — Athena, a Black professional golfer, Dee, a character she meets, and an invisible narrator named Jenny — in dialogue across the gallery, their voices and language driven by neural networks. Actors were used to establish the visual contours of motion, AI drives the language based on what raw voices and scripts have been loaded into it.
In other words, based on the literal voices entered into the computer, the AI software “creates” the specifics of dialogue and language, both in sound and sense. The work unfolds like a digital ready-made, revealing itself as the AI learns different ways of articulation.
The dialogue is funny and pointed, in the way an absurdist script can be. On a recent visit, Athena and Dee were seeking to make sense of their conversation:
Athena: I didn’t understand what I was going to write for. I don’t think it works with what I’m saying right now. I’m not doing that.
Dee: Very funny.
Athena: Normally, it’s weird. I never thought of myself.
The narrator: Jimmy appears from behind a chair.
Athena: Can you hear me?
The language, sequence, and sound is completely generated by AI, an unlikely evolving skit or playlet.
“We were set to shoot March 2020,” Syms said. “So that did not happen. Even when we shot each of the actors, it was very much a closed set later that year. We didn’t have them in the same space together. Initially, that felt like such a constraint. Like, what am I going to do with this? This isn’t my original idea. And then, like I said, I started to think, ‘How can I use this kind of technology almost as a tool for my writing?’”
Syms took footage of real actors who “cycle through” Neural Swamp with “generated voices, one that’s based on my voice and two that are based on celebrity voices,” she said.
She then took a scene from a draft script and ran it through her AI engine “to see what else would come out,” she said.
“They’re just asking each other on a first date,” said Syms. “And that goes through responses that are more interested to less interested, to more curt, to more lengthy. I don’t know, but I’ve been in acting workshops where the words you’re saying are kind of innocuous.”
The gallery where Neural Swamp is on view contains some companion video pieces, including one called Meditation, which features Kita, a character based on the virtual-reality host of BET’s Cita’s World, the first show created around a virtual reality Black avatar. It ran on TV for a few years at the turn of the millennium and captured Syms’ imagination with its music videos and computer-generated visuals.
Kita also makes a guest appearance in Neural Swamp. Syms found the actress who worked on the original show, kittie KaBoom (a.k.a. Kali Troy), and worked with her to create Meditation and other videos that Syms characterizes as “essays” about the current moment.
“She’s amazing, kitty KaBoom,” said Syms.
Meanwhile, at the Neural Swamp installation, a PMA security guard, encountered during a recent visit, stood transfixed before the Meditation video.
She watched Kita floating in an environment reminiscent of wallpaper for a screen world.
Amanda Sroka, associate curator of contemporary art at the museum, and the curator for Neural Swamp, also watched the video.
The guard turned to Sroka. “Is she real?” she said, indicating Kita. Sroka said that the character is based on a TV character, which prompted the guard to do a quick search for kitty KaBoom on her smartphone. Finding her seemed reassuring.
“She is real,” the guard said, turning back to the screen.