It’s All Causation
March 26, 2011 11 Comments
Until we come up with a coherent idea of what we mean by causation, I say teachers should accept this on tests (via Saturday Morning Breakfast Cereal, one of my favorite webcomics):
It reminded me of the Simpsons episode “Much Apu About Nothing” in which Apu is finally taking the citizenship test. I couldn’t find a YouTube clip, but at least I found the quote I was looking for:
Proctor: All right, here’s your last question. What was the cause of the Civil War?
Apu: Actually, there were numerous causes. Aside from the obvious schism between the abolitionists and the anti-abolitionists, there were economic factors, both domestic and inter–
Proctor: Wait, wait… just say slavery.
Apu: Slavery it is, sir.
Or perhaps we could learn to phrase our questions better.
The typical answer is something like “if you were to replay history, but remove the causes of event A, things would have turned out much the same way except without event A.” There are several obvious problems with this sort of explanation. It reminds me a little bit of the frequentist explanation of probability (“the probability of event A is the fraction of times event A would happen if you do something a lot of times”). As we Bayesians know, the true meaning of probability is a state of belief in a proposition, taking some subset of the relevant information as given (and assigning a maximum entropy prior to the rest). I think what historians or others mean when they refer to the causes of a historical event is some concise subset of the information we call history which, in combination with background information about people and societies, is sufficient for you to believe that the event would follow with high probability. (Since we don’t have enough processing power to evolve the dynamics of the universe from 14bya-1914CE, knowing fundamental physical law is not sufficient.)
“As we Bayesians know, the true meaning of probability is a state of belief in a proposition, taking some subset of the relevant information as given (and assigning a maximum entropy prior to the rest).”
Well put!
It was a big shock to me when I came to that conclusion – that ‘probability’ was an expression of our own confidence based on the evidence we take as given. For a while I was concerned that “as our information gets better fewer things are possible” (when I was first exploring the topic of free will.) I got over it when I reframed the issue in my mind to be “as our information gets better we have the power to make undesirable things less probable.”
Objectively, a “possibility” is just a proposition we don’t have enough information to reject. So the fact that there are fewer possibilities as our information gets better should not be surprising. Leaving out QM for the moment, if the universe is deterministic, there is only one (perfectly determined) future, and with perfect information, we could reject all conflicting propositions. And all QM really says is that some of the necessary information is fundamentally inaccessible (because before a collapse event, all the information that exists in the universes that are about to diverge is exactly the same). Similarly, when we talk about free will, what we’re really saying is that information about our own future thoughts is fundamentally inaccessible to us (if we knew, we’d already be thinking it!). So no matter how much information we have about the external world, all the possibilities that could actually result from the different actions you might take are still valid, since you’ll never have perfect information about what you will do until you’ve done it.
Obviously relevant is the wonderful Carl Sagan quote – “To bake an apple pie from scratch, you must first create the universe.”
One of the more defensible attempts to at least precisely cache out the folk conception of a cause is Mackie’s ‘INUS’ condition – A caused B if A is an insufficient but necessary component of an unnecessary but sufficient condition for B’s occurrence. Here’s a nice summary of the idea I found online –
“[T]o say that short circuits cause house fires is to say that the short circuit is an inus condition for house fires. It is an insufficient part because it cannot cause the fire on its own (other conditions such as oxygen, inflammable material, etc. should be present). It is, nonetheless, a nonredundant part because, without it, the rest of the conditions are not sufficient for the fire. It is just a part, and not the whole, of a sufficient condition (which includes oxygen, the presence of inflammable material, etc.), but this whole sufficient condition is not necessary, since some other cluster of conditions, for example, an arsonist with gasoline, can produce the fire.”
Read more: Causality – Inus Conditions – Sufficient, Fire, Effect, Mackie, Nonredundant, and Insufficient http://science.jrank.org/pages/8545/Causality-Inus-Conditions.html#ixzz1HjhABLCh
I love that Carl Sagan quote! It always makes me think of my dad, who would make that sort of comment whenever Julia or I would complain that the food was taking a long time in a restaurant. “Well, first they have to plant the wheat…”
I’ll have to think about the INUS description for how we use the word ’cause’ as it relates to the overdetermined assassin scenario.
This is kind of a nitpicky point, but if the house fire actually was an arson, under this condition it would still be valid to say that a short circuit caused the house fire. You could patch it by saying that A has to be true to be a cause of B–but perhaps there was a short circuit, although nothing caught fire from the spark, and then the arsonist dropped his match on the gasoline. This is basically parallel to the arguments (http://en.wikipedia.org/wiki/Gettier_problems) against the definition of knowledge as justified true belief.
(Basically, Aristotelian logic is inadequate, or at least extremely inconvenient, for reasoning under incomplete information.)
For every Cause there is an equal ation?…..Cosmos Law..stirred not shaken.
spread the humor.
Wow – just shared that Sagan quote with my wife, who says she remembered him saying that in an undergrad class he taught. The Mackie condition made my brain hurt. One problem with cause and effect is the assumption that all events are based on things that took place in the past. Maybe empirically, …..
A reference to Judea Pearl’s book on causality is appropriate:
http://bayes.cs.ucla.edu/BOOK-2K/
It is a Bayesian account of causality.
Hilarious! 😀
A history teacher once told me that when she was in school, her exam was an essay on “How WW2 could’ve been prevented.” She wrote a big essay discussing politics and economics, while her classmate turned in her essay after a couple of minutes. Turns out she just wrote the 10 Commandments. That was supposed to be profound. I would’ve flunked her so fast, her head would spin.