What caused the election catastrophe? Economy? Racism? Personality? Sexism?Russians? etc. The arguments here continue in full force, cycling around the same old issues. I think this is one of the rare cases where a bit of explicit philosophy can help us focus on the important questions, reduce bullshit, sharpen analytical methods, and maybe even soften acrimony. What follows may seem dry at points, but it’s an attempt to channel the very strong political emotions many of us feel toward productive ends. Although my conclusions sound trite, they are directly opposite to the spirit of numerous recent much-recommended stories here.
What do we mean by the statement “A caused B”? There’s one meaning that is both testable and relevant to the question of what we should do.
If, event A had been replaced by some alternative A0, would B have still happened?
The answer obviously depends on both A and A0, and may also be probabilistic. There usually are multiple causes for any event, especially a very close election, so attempts to find “the” cause of anything, although common around here, are usually completely pointless. We’re trying just to find what the likely outcomes would be of various actions.
What does this mean for the causes of the election results? Most importantly, it allows us to set aside many possible “causes” as irrelevant and not worth discussing. It lets us focus on what we should do, not on whom we should blame.
1. It’s meaningless to discuss the effects of A unless there’s an alternative A0, that can plausibly be imagined in a world otherwise very similar to ours. If there’s no such plausible alternate world, there’s no real meaning to asking what would have happened without A. E.g. if one were to say “the cause of DJT’s election is America’s racist history”, that’s an empty statement, because any political worlds in which that weren’t true would be so different from ours that we don’t really know how to discuss what would happen in that collection of imaginary worlds. A surprising amount of the discussion here still focuses fruitlessly on such fixed features of our political landscape (e.g. most minorities vote D, most southern whites vote R) rather than on the meaningful questions of what features might change. We might as well discuss the effects of the Earth going around the Sun.
2. Many causes have plausible alternatives but are entirely outside our control. The behavior of the Russian government is a prime example. One could imagine a world like ours but without their one-sided snooping revelations, and try to make arguments about how its election results would be different. That’s meaningful, but not really relevant to our question of what we should have done or, more importantly, what we should do next. (Unless John Podesta is reading this, in which case the relevant counterfactual alternative would be a world in which he chooses a real password.)
3. Finally, there are causes which we (potential readers of this blog, people we talk to,...) could change. That’s where we need to focus the discussion.
Obvious examples of things we could have changed include making (false) statements that HRC would be as bad as Trump and making (false) statements that HRC didn’t need various categories of inferior voters so that it was good to insult them. Both of these claims were common around here. The latter one still is. It’s pretty obvious that as a group we didn’t do our best to cause a Trump defeat, since making these statements must have hurt the vote margin.
There are also causes which we can’t control but can influence: choice of candidates, positions on issues, etc. These form the core of most rational political arguments. The most central one that is certain to come up again concerns how much to focus on the role of the very rich in draining the resources needed for basic social needs.
What are some of the arguments we should throw out? A big one concerns whether this was an election about (fill in the blanks: racism, the economy,..) All those things go into any election. The argument is meaningless. A meaningful question would pose some alternatives: e.g. the 2016 campaign vs. one with more of a class-struggle theme. Then arguments can focus on what the alternative consequences would have been: smaller net donations, more enthusiastic volunteers and voters, and so forth.
A focus on well-defined causal questions doesn’t just change which things we talk about. It also leads to much more serious analytical methods (under the headings “potential outcomes approach” or “counterfactual causation”), a sharp contrast to the crude correlation-slinging that often passes for analysis here.
There is plenty of uncertainty, and plenty of room for bitter dispute, in figuring out what effects our actions have even if we confine the discussion to meaningful choices. Let’s at least quit wasting time on the meaningless questions.