BP Ben wrote:appind wrote:thanks for your post. i get that affirming the consequent doesn't affirm the antecdent, it's a well known fallacy. and conditionally speaking, it's plain to see how necessary condition obtaining can't deductively affirm the sufficient condition obtaining. my question was more related to intuitive aspect in that how does it square with the fact that correlation, which is only a necessary condition of a causative relationship, strengthens. correlation obtaining doesn't affirm causation either.
intuitively, if one is given the conditional "if it's night, then it's cold" and "it is cold" (necessary cond NC obtains) then it just feels that at the very least NC obtaining removes from consideration the scenario that "it's not cold" which would have led to deduction that "it's not night." so NC obtaining makes one feel that a conclusion of "it's night" would get strengthened a tiny bit by precluding the scenarios of NC not obtaining. why does this mirage occur?
Ah, cool OK. I see the problem now -- sorry I didn't address this in my last post: The confusion lies in the fact that causal relationships and conditional relationships are totally different things, and you have to treat them differently.
Check the bolded and enlarged. It's true, in some sense, that correlation is a "necessary condition" for causation. If you have a causal relationship, there must be a correlation. (Which is, itself, a conditional statement.) And, as a matter of fact, we know that a specific correlation can totally strengthen a specific causal statement. So, voilà, through this brave feat of semantics, we've bridged the divide between conditional and causal relationships! Right?
Let's use your example, with a slight twist. Imagine that you've decided to spend a year studying penguin mating rituals in Antarctica. You there? Good. Now, say the conditional statement: "If it's night, then it's cold."
That's a totally valid and true conditional statement, but it has no causal implications at all. In Antarctica, it's literally always cold. If you're blindfolded and you pull your hand out of your glove to see if you get frostbite, that information will tell you exactly nothing about the time of day. Nor does the correlation between night and cold, in this specific case, help at all to strengthen the causal claim about night and cold. Because there is no causal relationship. Cold correlates with everything.
Now imagine you're in San Francisco. Here, it's possible to say that night causes cold. It's in the 70s in the mid afternoon, then the sun sets, and the temperature drops 20 degrees. You put on some socks and a sweatshirt, and shiver your way to the bar. But even in San Francisco, we have to separate our conditional statements from our causal statements:
You can say, "If it's night, then it's cold." But it's sometimes cold during the day too. So, in conditional logic land, the fact that it's cold--absent any other information--doesn't tell us that it's night time. But then you can make the (totally separate) causal statement at the same time: "Night causes cold." In order to strengthen this causal statement, you would need both variables to be present. If it's night AND it's cold, then that can be said to strengthen the claim, "Night causes cold."
*Note that in this final example, it's not correct to say that a necessary condition is "strengthening" a sufficient condition. That conflates the two totally separate statements (conditional and causal) that we're making, and it ignores the fact that you need BOTH variables from the conditional statement to be present before you can use the correlation to strengthen the causal statement.
I hope that helps to clear things up! Please feel free to drop by again if you come across an LSAT question that deals with this issue, and we can go over it together.
If you drill hard enough, you will get 170+.
That is both a conditional and a causal statement.
i should have probably been clearer in my post and thanks for your response. i mean, i get that my example is not a causal relationship and conditionality and causality are two entirely different things, so in my mind i am not actually conflating one with the other. they're different. my question is not whether logic books say correlation strengthens causation, but it is why logic books say so. in other words, what's is it about the relationship between two things, causation between A/B and correlation between A/B, that makes it so special that we take correlation to act as strengthener for causation, even though there is a well-known logical fallacy that correlation obtaining doesn't mean causation. i understand this relationship between causation and correlation is sort of unique in this respect, but what makes it unique is i guess what i am getting at. sorry if the issue is still not all that clear, but please let me know if so, and i'll elaborate.