Tuesday, March 17, 2009

On Suits on Games

There has been some discussion recently about the Suits analysis of games. In The Grasshopper: Games of Life and Utopia, Bernard Suits famously gave an account of what it is to be a game, in which he holds that games have the following three elements:

(1) They are aimed at goals that can be described independently of the games themselves. For instance, in golf you aim at getting your ball in the hole; but, of course, you don't need to recognize the rules of the game to recognize that a ball goes in a hole in as few strokes as possible.

(2) They have rules that place impediments in the way of doing things in the most efficient ways. For instance, soccer players cannot pick up the ball with their hands and run down the field.

(3) In playing the game you voluntarily accept these rules because they make the game possible.

As Suits put it, a game is a voluntary attempt to overcome unnecessary obstacles. What is noteworthy is that these criteria, if not read in a very particular way, make following traffic laws a game. If I am trying to drive to a conference, I have a goal that can be described independently of driving and the rules governing it. The rules that govern driving, namely, traffic laws, impede doing things in the most efficient ways; for instance, even if there would be, in my circumstances, no harm in violating the rules, I have to stay on the road, and stay in my lane, and if a sign says, "No Right Turn," I can't turn right even if the conference is right there, to the right. And I voluntarily accept traffic laws because they make my driving possible. We just have no good name for the game, and a lot of people appear not to have any fun playing it. Indeed, most rule-following activities where the rules are not necessary but are reasonable turn out to be games on this account. When evangelical Christians have a passover seder -- which happens on occasion at many evangelical churches in the United States -- they are involved in a voluntary attempt to overcome unnecessary obstacles. Rabbinical law puts up a lot of obstacles; they are being accepted by the evangelicals because they are the rules for a passover meal. And the particular things you do in a passover can be described in a way independent of the passover itself. So is it a game? Perhaps, but we seem to be stretching the term beyond recognition, with no definite benefit in doing so.

There is a further issue that the account seems better suited for describing games that are sports than all games. If you play games with children to any great extent, it's hard to deny that there are games that have no unnecessary obstacles; to take an extroardinarily simple card game, in 52 Pick-Up you throw a pack of cards up into the air, and then the game begins: everyone picks them up as quickly as they can. And that's the whole game (there is no way to win, you just play), a game for little children, but a game nonetheless. Unnecessary obstacles make games challenging; they don't make them games.

But there is something useful about Suits's analysis. The basic elements are themselves attempt to specify a more general analysis in which games have three components: 'prelusory goals', 'lusory means', and a 'lusory attitude'. And the idea, a right one, is that playing a game is to use means to an end, in accordance with rules, with a particular sort of attitude. The problem with (3) as it stands is that it pretty clearly falls short of capturing the sort of attitude we have with regard to a game we are playing (an attitude notably lacking on the highway, for instance). One problem with Suits's argument is that he is unable to capture the attitude element in a noncircular way: the lusory attitude can't be just any sort of attitude a game-player might have (so that it doesn't matter whether, for instance, someone is playing a game for fun or for money), but, as the Grasshopper describes it, the attitude without which it is not possible to play the game. The problem with (2) is that it pretty clearly goes beyond what is required for something to be a lusory means; in fact, the lusory means merely have to be appropriate to the prelusory goal, and the rules that determine what counts as appropriate may be of just about any sort, as long as they are consistent with the lusory attitude and the prelusory goal.

And so, in the end, what we find out from Suits is that a game is an attempt to reach a goal appropriate to a game by means appropriate to that goal, according to rules that are accepted because they are the way you play the game. That does indeed clarify what a game is. It even provides necessary and sufficient conditions, since every game will meet these conditions, and meeting these conditions is playing a game.* And because of its circularity that makes it resistant to counterexamples. I don't think there is anything wrong with this. But if we use it as an explanation of what makes a game a game, it is, we should be quite clear, a Father Noel account of games.**

----

* The finding of necessary and sufficient conditions is often seen as an important goal in analytic philosophy, or, at least, it has been seen as such. But it's misleading to leave it at that. No matter what you're talking about, you always already have in hand a necessary and sufficient condition for the thing you're talking about: itself, since everything is both a necessary condition for itself and a sufficient condition for itself. But this sort of necessary or sufficient condition is not counted, of course. And it's possible to have necessary and sufficient conditions that are different from the thing being explained. For instance, in a list of biological species it may be that being a renate is necessary and sufficient for being a chordate. But this sort of necessary or sufficient condition doesn't illuminate anything, at least on its own, and therefore this doesn't seem to be what is primarily in view. What people are really looking for is a definition (in a broad sense of the term), and the 'necessary and sufficient conditions' part is just two of several requirements being used to sort out good attempts at definition from bad attempts at definition -- in particular, the adequately proportioned to each other, or equivalent. And when analytic philosophers look for either necessary or sufficient conditions, they don't look for just any necessary or sufficient conditions; they look for those that will contribute to a definition that will clarify and illuminate. In other words, they are looking for elements of definitions that fit certain values; explicative definitions, they are called. And this goal, being value-dependent, allows all sorts of judgment calls about what counts as the right sort of necessary or sufficient condition.

** Father Noel was one of the objects of Pascal's criticism; according to Pascal, he defined light as the luminary motion of a luminous body, which meant one could not understand the definition unless you already knew what was supposed to be defined. Such 'definitions' are not, contrary to what Pascal wanted to imply, useless; for instance, this one makes a claim relating light to motion and to body, which could be the beginnings of an advance in one's understanding of light. But regardless of whether it is true, and regardless of whether it can be the beginning of a research project, it will not help you to understand what light is. It is not what you are really looking for; at best it is a first baby-step toward it.