Correlation is not causation! As I discussed in part 1 of this three-part series, imputing causation from correlation, usually erroneously, is one of the “Big Three” LSAT flaws. (The other two are Necessary/Sufficient and You haven’t proved you’re right so you’re wrong.)
In part two of this video series (the video is here; I can’t embed it because it’s too large) I discuss an important issue in certain correlation/causation problems, namely whether it’s possible that the speaker (in the problem) got cause backwards. In other words, if the problem reasons from the fact that A and B go together (i.e., are correlated) that A causes B, an issue may be that perhaps B causes A, not the other way around.
Often this reversal of causation is logically impossible, usually because of timing. For example, B can’t cause A if A happened first. (We assume, reasonably enough, that the arrow if time goes in only one direction. Even if your view of metaphysics allows for exceptions I promise you they’re not relevant on the LSAT.) And often while it’s logically possible, the reversal doesn’t make factual sense. For example, rain probably causes people to carry umbrellas but it just isn’t reasonable to consider take account of the logical possibility that umbrella usage causes it to rain.
The factual situations in which we do see reversal of causation show up almost always involve serious science, usually medicine or biology. If you see blood proteins or cell types or some other medical terminology that you don’t recognize (or think most others won’t) in a logical reasoning problem, you can confidently expect that the issue will be correlation/causation generally and more specifically, whether cause can run the opposite direction from what the conclusion states or assumes.
Cause running backwards can show up in any conclusion-based problem but the four most common are weaken, strengthen, necessary assumption, and flaw questions. The structure of the answer choices for these question types is pretty predictable:
- To weaken the argument, choose the answer that says cause does, or at least might, go backwards;
- To strengthen it, choose the one that says in some way that cause doesn’t, or probably doesn’t, go backwards (which can be a lot harder to spot than the weaken version);
- The (necessary) assumption is that cause doesn’t go backwards; and
- The flaw is that it might.
Watch the video and let me know what you think.