The Statistical Self
“Every man has his hobby-horse,” goes the proverb, and mine is the difference between the subjective and the objective. Though that sounds exotic and philosophical, it’s not. The subjective is the perspective of limited, first-person experience. The objective is the perspective of unlimited, god-like experience. As Stalin supposedly said, “One death is a tragedy, a million deaths is a statistic.”
Read Joyce or Proust and you are immersed in the subjective; read a geology textbook and you are immersed in the objective. (Whatever you want to say about the findings of physics depending on human observation, human beings matter little to, say, shale.) That subjective-objective distinction pops up in all sorts of places and many disagreement turn on it.
The philosopher Thomas Nagel built a whole book around this idea, as I recall, called The View from Nowhere. This was that last philosophy book I enjoyed reading; it was published in 1989. I found his device of looking at a group of ancient problems through that lens to be brilliant. Reviewers in the New York Review of Books, the Times Literary Supplement, and The New Republic loved the book, while many professional philosophers disdained Nagel’s views and approach. David Lewis, a onetime colleague of Nagel’s at Princeton, introduced him once by saying that Nagel was always profound. And always profoundly wrong.
I think the subjective is in decline, especially with the rise of what might be called the “statistical self.” The history of psychology, as I read it, swings back and forth between focus on the subjective and the objective. Fraud deified the subjective, though with a twist: we didn’t have access to our own subjectivity. We had to search around for it, with a psychoanalytic flashlight, until we found our true self. But it was there. Frankl and Maslow stressed the traditional self: how it feels to have and reach goals. Skinner eliminated the self by fiat. Buddhist psychology claims the self is an illusion.
I agree that the self is an illusion. But so is color, pattern in music, continuous lived time, and many other phenomena that most of us would miss were they removed from our lives. Perception is the only true magic: it is filled with illusion. Just because something is illusory, though, doesn’t mean it is harmful, nor does it mean the illusion can’t be useful. Illusions, like many ideas, are tools. And certain illusions, like the self, I suspect, will never be extirpated—though that is an empirical question.
And yet, I think the self as traditionally conceived is in decline. In its place is a self constructed from more and more traditionally objective data than ever before. Do I really mean “than ever before?” Yes. I don’t think that’s hyperbole, because today really is different from many other ages. We have, today, more data than ever before. And much of that data we use to construct the self.
As Timothy Wilson pointed out in his brilliant book Strangers to Ourselves, often the best way to know what we are like is to ask someone else. This idea is reflected in the old joke about two behaviorist psychologists. Passing his colleague in the hall one day, the first one says to the second, “You’re doing well today. How am I?”
Take a trivial example: iTunes. Many of us rate our music. I do. I take three stars out of five to be the mean rating, then try to determine how much above or below the mean any particular song should be. This is a pretty haphazard process. I listen to a song, rate it or modify the rating, and don’t give much thought to the larger pattern that these ratings compose. But then, from time to time, I group together all my favorite songs—“favorite” meaning “five star.” And I see that there is a pattern to what I like. Subjectively, I never would have seen that pattern. Armed with the data, I update my self conception with this new information. And each time I do something like this, my self gets more objective. (This seems to be analogous to the interplay between “top down” and “bottom up” processing in the brain.)
I think Seth Roberts’ self experimentation work is the perfect example of this. These days, whether with Excel or R, almost any reasonably intelligent person can perform experiments on themselves. It’s not hard to learn the basic statistical methods. Computers relieve the need for all the drudgery that psychology grad students used to have to go through. Better and cheaper instruments make data capture easier than ever. (Think of the heart rate monitor as a trivial example). As this self-experimentation gets even more widespread, I think our sense of self will change.
There will always be something it feels like to be a human being. We will always have a self, I think. It’s useful for a social animal to be able to model itself. And I have little doubt that cats, dogs, hamsters, and perhaps spiders have simple models of themselves. For instance, generally a spider knows not to chew off its own leg when it gets hungry. How the spider knows this is, I’m sure, far different from the way human beings know it. Our bodies are built differently, with a more complicated cognitive apparatus. Still, in each case the self is being modeled.
But, just like psychological conceptions of the self in professional psychology oscillated between focusing on the subjective (Freud) and the objective (Skinner), our own theories about ourselves move between focusing on how it feels to be us (think of a human baby an an extreme example of pure human subjectivity) and focusing on how it “is” to be us—from the outside, using reports of others, statistics, and other measurements. I suspect that, as more and more data becomes available, the self of the future will be the statistical self.
[composed and posted with ecto]