I’ve been getting a lot of my subject matter from Ryan lately, it seems…
Well, in any case, he put a link on Twitter to Sam Harris’ TED talk about science and morality, and how science could feed into morality. It’s well worth looking at and thinking about a little.
Morality has to do with distinguishing “right” from “wrong,” and Harris has a very good point that scientific methodology could be applied to help make that distinction. However, while I listened to his talk, a very important point came to mind. Let me set this up with the statement that many concepts or measures in this universe don’t come out to binary extremes. (Quantum states of spin-1/2 particles, for instance, are an exception.) In most cases, it’s not a question of just being on one side or the other; it’s a question of how far towards one side or the other your measurement comes out. I think the same is true of morality: how right is one thing compared to another? How wrong are the alternatives?
In answering such questions with scientific processes – not an idea I disagree with, in principle – we would likely end up at some kind of optimization problem. Given all the scientific data about the possible reactions and effects of a particular decision, how can we make the most “right” decision? That’s a pretty straightforward problem to approach scientifically. However, we must be careful about how we define “most!”
As an example, if you drive you have probably had the experience of getting stuck at a stoplight somewhere, getting frustrated, and saying to your passenger or yourself, “Wow, these lights are stupid. I’d love to meet the guy who designed them, they could be a lot better than they are.”
The operative word there is “better,” and the question is, how do you tell which stoplight timings are better than others? Probably, the guy who designed them actually chose the best timings. But what he considered “the best” is maybe not what you consider “the best.” Maybe he maximized the traffic flow on the main street instead of the cross street. Maybe he minimized the average number of red lights cars encounter along a certain route. Maybe he found the timing that gave the least amount of wait time at certain intersections, while also giving the highest possible rate of cars through the intersection, during rush hour on average Thursday mornings. Which one of these definitions of “best” is best? And why is it so? There is an assumption underlying the process here, and it can have a dramatic effect on the results.
I think we have to keep that point in mind while considering Harris’ points. We have a lot of data on actions and consequences. We can use scientific processes such as optimization to try and synthesize that data into a decision about what is right and what is wrong. But we have to bear in mind the assumptions that underlie that process, be up front about them, and be willing to entertain other possibilities.