For more than 20 years now, I’ve been writing occasionally on the theme of motives, always making the same points:

1. Because, as Rebecca West famously said, “There’s no such thing as an unmixed motive,” it’s very hard to tell what anyone’s truly dominant motives are;
2. What people actually do is more important than what you think their motives are;
3. There’s no reason to think you can understand the motives of others if you don’t have a firm grasp on your own.

I was thinking about that third point especially last night while listening to the most recent episode of FiveThirtyEight Politics podcast. The hosts prided themselves on looking into the motives of the people who make polls, but it never occurred to them that their own project might be motivated also.

Some writers in the so-called “rationalist community” will preface their posts or essays with some statement of “confidence interval” or “epistemic status” – Scott Alexander does this, for example. I think everyone who writes about the motives of others should append to their discourse a “personal motive estimation” – an account of what they believe their own motives to be. In the spirit of full disclosure.