With her new album, Deathwatch, the modus operandi has become that much clearer: she’s not so much exposing the structure of R&B tracks as she is cracking them open, stripping them—ribbon-like—into pared-down glitches, revealing the magic of the machine, and then infusing it with something heart-pumpingly human. It’s pretty highbrow stuff, surgical in its precision and cyber-ganically grooving. We spoke to her over e-mail about Deathwatch and her ethos.

Let’s start from the beginning. Tell me about what you were working on before your solo albums, before Pit er Pat. You have a clear interest not only in music, but in sound and its construction. When did that interest originate, and what sort of art were you making before?
I started playing piano when I was three. I connected with music really strongly; it was the most important thing for me. I think I understood the communication much more so than any other form. I practiced all the time, as you must do to get the physical parts down. But apart from the physical, I think that communicating with music just made sense to me. It was something I could easily feel and know.

Sometime in high school, I got interested in video art. Looking back, I can see similarities in how I edit sounds and how I was editing video: trying different things together, without knowing ahead of time what the end should be, and paying attention to rhythm, the length of line, melody. It’s similar to intuition, in that I know why one thing should follow another, or be with another, not in words, but in the meaning that is communicated with sound through pitch, duration, timbre, volume.

I think I’ve become a lot more focused on constructing sounds because, after Pit Er Pat, I didn’t want to play keyboard parts that came from the root of piano pieces or songs. I had a tendency to write parts and use sounds that had a very similar tone, and I wanted some very different tones. The old ones were not adequate to describe what I wanted to communicate, [which is] not dissimilar to why I stopped using my voice in the same way that I used to. I can only change my voice so much, you know. I can’t sound like Johnny Cash, and the tone of my voice was not at all what I wanted to use. By editing out the beginnings and endings of words, and layering vocals to make chords, I could make new tones. Making vocalizations that were not words was another way to change the tone.

Also, learning how to manipulate and make new sounds is really fun for me. I always try to keep things fresh for myself and not repeat something that I already know how to do. I think the arrangements end up more surprising because of this, as well. If I’m familiar with the sound, then I will have a tendency to use it some way I’ve used it before. But if it’s new, then it is unknown to me how it can be used. It leaves the possibilities open, which, I think, is how it’s best for me to hear.

Deathwatch and DIN have the quality of pared-down skeletons, the limbs of the tree, the most minimal parts of a particular sound exposed. Can you talk to me a little bit about this?
I think that you can listen more to what each sound is doing when there are fewer layers. They work together to give a complete sound, but you can also single them out and hear each voice. I don’t feel it’s necessary to complicate the arrangements. I think for some people, this makes it sound less polished or finished, but that’s just compared to things they’ve become accustomed to hearing. I think because each moment is an event unfolding, rather than serving a familiar structure—like a drum fill leading you into a chorus, or the rhythm of a lyric anticipating the following line—it is helpful to have simplicity. Then you can more closely follow what is happening, where it is going, what the structure is doing.

Although the kind of idea above seems mechanical—pleasantly disjointed—there’s something very human to all of your music. I was able to dance around to “Sex Hats.” This definitely speaks to how visceral it is to move and dance; doesn’t take much for it to happen. But it’s also a testament to how much of a groove Deathwatch has, as pared down as it can be. I’d love for you to comment on the human element in your work, the stuff that makes it movable.
I love hearing that you were dancing! Yeah, I have heard people say that they find my rhythms very disjointed or unusual. Maybe it just takes letting go of where you think the accents should be, or accepting a new way of moving your body. It was really important to me when I started working on DIN—that I keep the human performative element very strong. I had never made things entirely on the computer before and I was wary of allowing the machine to strip it of its juice, its life. That’s partly why I didn’t use a grid or sequencers. I wanted to place each drum hit as a drummer would, allowing for variation. Also, that way, I wouldn’t just decide certain parameters and then program it and let it play out. I had to listen to how it was developing and make decisions along the way [about] how it should change moment to moment. I wanted it to feel exciting, not flat.

I think the way that our bodies respond to music is really important. I wasn’t making background music. It was something that wanted to really involve you. Using bass and drums to connect with the body is effective. I was really focused on the bass tones in Deathwatch. Usually bass is used underneath other sounds—we feel it and our bodies are given cues from it, but we’re not consciously focusing on it. I wanted to bring it forward, to focus on that connection. What does it tell us? I think it’s something very interior. Whereas a melody resonates closer to our heads and our chest, and therefore our centers of language and our hearts, the bass sits deeper. It communicates less verbally. We feel it in our bodies and want to move to it. It has us think less in terms of our ego, more towards general existence. I don’t know if this is true, but I think it’s something I wanted to do with this project, to see what the possibilities were in communication. More than what I get from a typical song structure, what are the possibilities of what different sounds in different places can communicate to us?

There have been a couple years between DIN and Deathwatch. How was the recording process different for each? What’s grown or shifted for you?
I didn’t want to fall into a formula. With Deathwatch, I spent more time making sounds first, and then having them tell me what should happen. I tried to stay super open, to not have an idea from the beginning, so that I could listen for possibilities—listen for things that surprised me, or for what I wanted to hear follow.

Deathwatch feels simultaneously more rhythmic and more hard than DIN — something about the lack of vocals. What’s the difference for you? Is it just a progression or growth?
I’m not sure what you mean by “hard.” It’s darker, I think, in tone. I think that’s partially due to what I was experiencing at the time of the making of each record, what I wanted to look into and allow to come forward. I think while making DIN, I wanted to keep a lightness or playfulness because I was experiencing a lot of darkness, so I needed to not focus too much on it and try to get myself into a better place. And with Deathwatch, I was a lot stronger when I made it, so I could bring in darker tones without getting sentimental. I could accept dark elements without them being negative or indulgent. They just exist as what they are.

I do think there is growth between them. I learned a lot while making DIN, and when I finished it, I felt like I was just beginning. I was happy with it, but I knew that I could take it much further. I feel that way again. After having finished Deathwatch, there is still a long way to go. But each step is an important part of the process. It’s good to document and finish something in order to move forward.

I feel a little ignorant asking this, but I must—partly for clarification, partly because it can be useful to have an artist explain, in her own words, what’s been written. I understand that when you recorded DIN (and maybe Deathwatch), it was edited visually—no grid or sequence, but “its distance in time from the others is measured by visual space.” Can you explain this?
I think I touched on this before, but I’m glad you asked. I think I explained why it was a way for me to retain human feeling in performance while using a machine. As far as editing visually, I only meant that because I didn’t have a grid, I had to visually measure the distance on the timeline between, say, kick drums, and place them where it looked like they should be. When the track is playing, you can watch the timeline and see where things should fall. But then, of course, I would listen to it, and if it sounded off, then I would move it until it sounded right to me, maintaining a handmade quality.

This was incredibly time-consuming and other engineers are baffled when I explain this process. They think it’s a huge waste of time, but they usually concede that it also seems like it’s working for me. In the beginning this was also a way for me to make beats based on what I was hearing in my head. Because I had never written drum parts before, I didn’t know where to place things in a sequencer so that they would match what I was hearing in my mind. So, it was easier for me to lay the hits out as I heard them. It was a way for me to allow more interesting things to happen.

You’ve mentioned that this process leaves an opportunity for chance. Why is this linear process, with its opportunity for chance or even mistakes, important?
Because surprises are exciting! Standard song structures have certain cues that tell us what is coming. They can communicate very specifically to us. But I wanted to try to have the sounds say something else. I don’t want to be too safe; I want to accept unpredictability and be able to roll with it. Finding something unknown, something I haven’t heard before or didn’t know was coming, can possibly communicate something that I wouldn’t have found if I had planned it intellectually.