08 July 2015

Theories (in theory)

It’s been a surprisingly long time (not really)! Much has happened (this is true)!

Last week I got the surreal experience of attending HCIC, a conference mostly for up-and-coming and established researchers in the field of HCI (often but not exclusively in North America, mostly because it’s a very small conference often held on the west coast). And then there was me.

I can’t say everything about what happened mostly because there was too much good stuff to recap in a single place. Too many valuable talks, conversations, and experiences; they all sort of blended together. As a counter-example, at CHI there were thousands of people, dozens of talks going on all at the same time (this is what’s called a “multi-track program”, I think), and tons of “noise”, by which I mean not only loudness but also low-quality moments (milling around outside a room with too many people with whom you didn’t share quite enough to initiate a conversation at random).

One conversation that stood out, however, was about theory. In particular, we got to discussing the value of having “native theories” in HCI, ones that we own rather than import from cognitive science, anthropology, sociology, and other social sciences. The crux of this conversation came about, however, when the discussion became more broadly about theories as we import them.

In one camp, it seemed that we were talking about theories as answers to questions we have about the world. In this camp we point to the theory of evolution, the theory of relativity, the theory of Mendelian genetics, the Bohr atomic model. Some theories have been disproved or revised (Mendelian genetics disproved, Bohr atomic model significantly revised). Importantly, these theories have a sort of truth value to them, and more importantly, that is a rather direct proxy for their value. The people in this camp often proudly champion the positivist flag.

In another camp, however, it seems like we’re talking about theories as ways to ask questions we have about the world. In this latter camp, we’re talking about Actor-Network Theory, Symbolic Interactionism, Social Constructionism, and others. These theories don’t have truth values per se; they’re not disproved, but they are arguably evaluated for their usefulness as a lens with which to interrogate the world. Functionalism might give you a way of looking at things that exposes relationships and motivations that you wouldn’t see if you approached it with Interactionism in mind. Or it might. This is why we learn these theories, to think about what they offer as interrogative tools.

I’ve gotten to talk with Terry Winograd about this with my advisor (Michael Bernstein) and Terry pointed out that these have been categorized (among others, by him) as “structural-correlational” and “ontological orientation”. What I think he means by this is that there are theories that describe structures or at least offer correlational insights; they are defined by their relationships to the world and, implicitly, suppose that there is an objective environment/world/universe to be measured.

If you’re thinking “duh, idiot”, then you might be a positivist. And that’s great! I think positivism is a reasonable position to take, in moderation. Here’s my critique: we know from experience as adults that things tend not to be one-sided. Our understanding of the world is framed by our experiences, upbringings, and world views. Critics of positivism take this further and argue that all the “knowledge” we’ve built up is predicated on the bases of our beliefs and the paradigms to which we subscribe. Knowledge is constructed, not some objective thing we go out there and find.

If you’ve just read that and thought “of course, that’s stupidly obvious”, then jeez can you give me a break? First the positivists and now you. But more importantly, you may agree with me that positivism might need to cede a little room to other ways of thinking. The problem is, they both seem to share this word “theory”, albeit with different meanings, and subscribers to one view or another end up stumbling over each other as they try to communicate (an admirable, but treacherous, endeavor).

We might need to better formalize what we mean when we talk about theories, or we’ll continue to have trouble trying to foster collaboration between both groups, let alone facilitate dialog.

If you have something to say about this post, email me or tweet at me.