Adam Parrish sits in the back of a Columbia University cafe, engrossed in his MacBook Air, blending in with any given student taking summer classes. But at 32 years old, Parrish is teaching such a class. He’s a programmer by trade and a summer interloper at Columbia, teaching journalism grad students how to code for their journalism careers on the Web (to build infographs, data models, etc.).
But Parrish’s labor of love is the Twitter account @everyword, which has auto-tweeted words individually once every half hour for the last seven years. Today, it ends, already having gone viral and accumulating nearly 100,000 followers and 110,000 tweets.
Fusion sat down with Parrish at the sunset of his Twitter phenomenon to get his reflections on Twitter’s reaction to @everyword, the tech industry, Parrish’s love of fellow Twitter phenomenon @horse_ebooks, and word association with the Twitter bot’s most-popular words.
Fusion: You’re a programmer by trade. How long did it take you to program @everyword?
Adam Parrish: Honestly, not very long at all. It’s a very, very simple script. It took me about an afternoon. I posted an open-source version of it on [software-development project site] GitHub.
What comes to mind when you see the word "sorry?"
I think of the singer Feist, weirdly, because it makes me think of the first line from her album The Reminder, "I'm sorry / two words / I always think / after you've gone." I have a lot of instant word-to-song associations.
Words like these make @everyword seem conversational, or meant to start a conversation. Was that part of the intent with @everyword? Did you want people to talk to it? When that started happening, what'd you think of it?
I'm not surprised that people attribute a personality to @everyword. I find that most people who respond to @everyword aren't talking to it, so much as talking through it—responding to the text of the tweets, or elaborating on them, using them as the basis for telling jokes. "Sorry," as a word, is attached to a conventional "script"– it elicits a particular set of responses, like "It's okay" or "I forgive you"—even though the tweet came out completely automatically, with no intention behind it. But as a follower, it's funny to cast yourself as the other person in the conversation and give the "expected" response.
Which word was the first that got a Twitter response so large and/or fast that it took you aback?
I don't remember, specifically. I feel bad, actually, for not paying more attention the reactions of @everyword's audience. I do remember being rapt as I watched the responses to "sex" come in, constantly refreshing to see more as they came in. That was fun, and really the first time I thought to myself, "Oh my God, I've created a monster."
It’s fascinating how many responses—replies, favorites, retweets—there have been to certain words, particularly ones that users view as eccentric, loaded, even vulgar. "Sex" stands as the most retweeted and favorited tweet of @everyword. What are your thoughts on that?
I just think people are having fun. Words like that have a cross section of a lot of people who recognize and have associations with that word and know that it’s going to be funny to their friends. ["Sex"] is just one of those words where that is the case. It’s a fun word. It’s a fun word to talk about.
It’s funny how a word by itself like that, a series of them, has become kind of a Rorschach inkblot test for Twitter. Is that something that kinda happened, you think, or is it something you intended? Or are you just happy it turned out that way?
I’m definitely thrilled at how the response turned out. It’s amazing so many people have connected with it. One of the things I was going for [in the creation of @everyword] is to show that words, even when you take them out of their usual context, still have all of this weird baggage associated with them. The word is like a weird cultural object that exists.
I studied linguistics as an undergrad. One of my jobs with the Linguistics Department was to transcribe conversations between kids. It was really hard. It was really hard to take a conversation and transcribe it word for word. Because all of these units that you think exist, like sentences and words themselves, it’s hard to come up with, definitively, an idea of where they’re separating ideas.
Probably a lot of sentence fragments and disjointed thoughts, there.
Well, the larger point I’m trying to get to [with @everyword] is that we all talk like that. And the boundary where one word ends and another begins—when you’re looking at the actual phenomenon of speech—it’s very hard to separate those out. What is a word, what isn’t a word and how you draw the boundary between those is something that is very much socially constructed, not constructed from the fact of what language is.
How @everyword is interesting to me from that perspective is: if you bring up this cultural object of the word and just strip it of everything that normally surround it, people still have all of these ideas as sort of its own thing. It’s like a self-driven vehicle that’s pulling all of this baggage behind it.
Tell me a little bit about what and who you are. Where are you from?
I’m from Utah, originally. I do come from a Mormon background but I do not practice.
You mentioned your job as an undergraduate in your Linguistics Department. Where did you go to school?
University of California, Berkeley. I studied linguistics but I got an undergraduate degree in computer science before that.
What was your initial aim when you started that tract of education? And is @everyword the culmination of those two degrees coming together, interdisciplinarily? Is this what you most want to be doing with your career?
That’s a great question. When I started studying computers, originally, it was because I latched on to the family computer when we got it when I was a little kid. I love computer programming. It’s very fun to think about tasks in that way and break down the world in ways programming encourages you to do. But I think I had an idea that I didn’t want to do that for a living. And I always loved language a lot, because I read [Lord of the Rings author] J.R.R. Tolkien as a kid.
So in digging deeply into computer-programming languages and human languages, do you find a similarity between them?
Yes, I did, but not in the way people usually draw that similarity. They think those languages are self-contained mathematical systems, that you can draw a diagram of a sentence and translate that diagram into a computer program and distillate its meaning or something like that. That always seemed really weird to me. It doesn’t really seem like a good idea. I try to focus on the human side of things.
It’s a lot like video game design, another interest of mine. You’re creating a system of rules for people to engage in, in order to have some kind of experience. I’ve made text adventure games.
Anything out there that people could find?
Yeah, this game called Earl Grey that I made with Rob Dubbin a couple of years ago. He’s a writer for the Colbert Report. He did all the writing, I did all the programming. It’s a really funny game.
Have you been able to monetize @everyword? Is that something you’re thinking about? Has it opened up other opportunities?
Nobody has approached me or anything…. It’s definitely been good for me in terms of visibility, after the Gawker piece a couple of years ago, it was something I could talk to people about but only secondarily.
On your blog that you said @everyword wasn't a "GRE review," it's an "art project." Do you think @everyword has become more of an opportunity for study, or contemplation, for Twitter users? Is it like that for you?
I can easily imagine that being the case. I really don’t know why people [bother to] follow @everyword. (laughs) It’s amusing, kind of soothing and adds a pacing to my Twitter feed… I’ve read some people reacting to it in an oracular fashion, foretelling events that happen in their lives or in the news. Or that there’s a synchronicity or serendipity to it.
What comes to mind when you see the word "swag?"
I remember going home for Christmas when I was 18, my first year of college, hanging out with a friend, walking around my [Salt Lake City suburb] hometown. We were walking past a display of Christmas "swag"—[what the store called] wreaths and decorative stuff– and just thought it was the weirdest word. We’d never heard that word before referring to that thing. It was such a nonsense word, the most ridiculous thing ever. This was like 14 years ago. And now it’s a hip-hop term.
I think this is one of those instances where it’s funny to imagine @everyword as a person who has gotten to this point [in the word list] and they’re like, "Ugh."
So is @everyword is over? This is it?
I’d really like to run @everyword again. I’m not exactly sure what form that would take. I don’t want to do the same word list again. I may take a little break and then bring it back, eventually.
Is there a demand for that?
That's the number-one request that I get. There were no followers of the account until it got to I or J [words]. I didn’t even follow it for the first year it existed.
What’s the last word going to be?
The last word is a surprise, I’m not going to give that away. There’s a little twist ending that should be amusing to people. Either amusing or annoying. For people who don’t understand why these words are at the end of the list, they will be annoyed. For people who do understand why these words are at the end of the list, they will be happy.
This whole process has taken seven years. You started it when you were in grad school. What were you like when you started @everyword and what are you like now? Have you changed?
Yes and no. Not fundamentally. I’ve gained a lot more confidence in my interests. It’s weird to look back. I was interested in this stuff but now I know it. I have some expertise in this area. Stuff I used to dream about doing, I’ve done or I know how to do it.
I know now that I don’t actually want to be doing [programming] for a living. Working in startups in New York is kind of a drag. It’s kind of boring. The tech sector is a mess for a whole bunch of reasons, especially with issues like sexism in the industry, which is really awful and terrifying… The whole way that the scene is funded… It’s not a great place to work. I’m hesitant to say that because I still need work from that area to pay the bills.
The @everyword bot is especially interesting in terms of last year’s @horse_ebooks phenomenon. To a lesser extent, the cult following that @everyword has developed is similar to that of @horse_ebooks. How did you feel when it was revealed to be an art project and not truly a random, automatic bot, like @everyword actually is?
I was so heartbroken by @horse_ebooks and I can’t really talk about it without getting super emotional, breaking down and blubbering. I have so many thoughts about it. It was wonderful for what it was… the idea that @everyword is in that category, that’s good to hear.
@Everyword isn’t as enigmatic or complex as @horse_ebooks was. It’s a lot more straight-ahead.
You have to bring more of yourself to @everyword to get something out of it. It’s human in that way.