Symmetry as a methodological move, Part II

We've been discussing the oft-misunderstood principle of symmetry in actor-network theory. Last time I emphasized that symmetry is a methodological stance meant to get at certain things and not others–not wholesale anthropomorphization. Let's pursue that line further by performing a thought experiment. I emphasize that this is just a thought experiment for reasons that will become clear below.

Aristotle believed and taught that heavy objects fall faster than light ones, in direct proportion to their weight. Galileo disagreed, and described a classic experiment in which he stood at the top of the Leaning Tower of Pisa with two balls, dropped them, and they hit the ground at about the same time. (It's unclear whether Galileo actually performed the experiment.)

Suppose you want to perform a similar experiment. Since I'm at the University of Texas campus, let's imagine that you ascend UT's Tower to the observation deck. You bring a one-pound weight and a ten-pound weight.

Photo © 1980 Larry D. Moore (CC-BY-SA), available  at http://en.wikipedia.org/wiki/File:Uttower1.jpg

(NOTE: You can only do this as a thought experiment. The observation deck is inaccessible except during guided tours due to a tragic incident. Also, dropping weights from the observation deck is a misdemeanor. Obviously, I do not endorse actually performing this experiment.)

Leaning over the railing, you drop the two weights at the same time. Not surprisingly, you discover that Galileo was right and Aristotle was wrong. The ten-pound weight does not fall ten times faster than the one-pound weight; they hit the ground at roughly the same time.

Interesting, you think. And then you decide to try a second variation of the experiment. You ascend the tower again, this time with a one-pound weight, a ten-pound weight, and a 150-pound hipster.

(NOTE: Again, you can only do this as a thought experiment for obvious legal, moral, and ethical reasons.)

When you reach the observation deck, you repeat the experiment, this time dropping all three objects at the same time. It turns out that no matter how ironically the hipster falls, in terms of gravity, he behaves exactly as the other weights do. For the purposes of the experiment, the hipster is exactly the same as the other two weights; you treat them symmetrically, and so does gravity.

But when you descend the elevator to the bottom of the tower, you discover that what is treated as symmetrical in one network of meaning is not treated as symmetrical in other networks. Dropping the weights is a misdemeanor. Dropping the hipster is a felony. The law of gravity treats them as symmetrical, but state and federal law do not.

When Latour describes humans and nonhumans as symmetrical, he means that differences among actants (both human and nonhuman) are generated within a given actor-network rather than preexisting them; we can't presuppose those differences. Consider Galileo's experiment as well as the previous illustration of elevator capacity: in these narrowly defined situations, humans and nonhumans act in exactly the same way. (If the tower experiment were a physics problem, you wouldn't even specify the materials, you'd just plug in the weight values.) But of course these networks of meaning are always entangled with other networks, as they should be. There are many good reasons why you would never actually perform this experiment, and they go beyond trouble with the law to basic questions of empathy and morality, questions that apply to all people,  few if any artifacts, and to some extent to nonhuman animals. 

But Latour talks about situations in which people are actually interacting: scientists, technologists, the public. Aren't these social rather than technical? And in social situations, don't we have to treat humans and nonhumans differently? 

The answer is: it depends on your methodological aims. Even in terms that we consider very human–cognition and persuasion–it sometimes makes sense to take a symmetrical viewpoint in one's methods. We'll see an example in the next post in this series.