Thanks for the comment!!

I don’t think that there would be a difference in terms of sense impressions…That is what I think Hume’s mistake is…why, do you think this matters somehow to whether or not we see the causation?

What would be the difference, in terms of sense impressions, if A was in fact only correlated with B?

Thanks for the comment!

Depending on the details of the situation I guess I would say that in the case you describe it <em?looks like A causes B and C. The experiment would be done to find out whether or not it actually does cause both or only one. So, for instance, if we found out that B was just a one time thing we would likely conclude that though it looked like A caused B in the first experiment it in fact did not it only causes C. This sort of thingseems to be what scientist do all day long in the lab (of course they do other things, my point is only that this is one of them). I think we use this ‘looks like’ talk all the time when we talk about causal relations. It looks like the bullet kills the guy but it is CGI etc…What I do not say you observe is the necessity of the relation. That is a seperate matter (according to me)…So, I don’t have to see that the table is 5ft to the left in order to see it to the left and I o not have to see that the relation is necessary in order to see A cause C. The fact that there are cases where we can be fooled (i.e. where somthing looks like it causes something which it doesn’t in fact cause) seems to me to just be evidence that we are observing something. Compare: It is because I know what a red wall looks like that I can be fooled into thinking that a certain wall is red (as when say, it is really a white wall with a red light shone on it). Does that help?

to reiterate: how can you be thought to “observe” causality in any way other than that which hume describes if you need repeated trials to infer the causal link?

I think that you are on to something in the comairison that you are making. I guess I would opt for option 1*, I think something like 2* is what Hume and Quine might think, but it is not something that I like. I am very anti-time-slice myself…

I was thinking about the object permanence as a theory (re our discussion on Brood Comb), and how it connects to this issue.

With the object permanence one can give (among others) those two possibilities:
1.The objects are seen (or perceived in general) qua permanent objects.
2.What one sees is time-slices (whatever this means, I haven’t seen those yet ), and the objects are theoretical constructs based on built (naive) physical theories

It seems that one can do analogous division of options for case of causality (for making the case even more analogous one can point to Baillargeon researches that indicate that at 2.5 months, infants expect for a thing to be displaced when hit by another, which is also the time where they seems to be aware of object permanence). The options would be like this:
1*.One object hitting another is seen qua hitting, or qua event in which one object affects another or qua causal relation.
2*.What is seen is merely objects and their movement, or time-slices, and that one object hits another is a theory that is developed.

I wonder where would you stand if things are put like that… Would your saying that causation is observable is along 1*? Or would you go with something like 2* and try to say that causation is observable because the theory is somehow added to the perception?

I have a couple of things to say.

1. I agree that we do not se the necessary-ness of the relation, and it is possible that the relation may not be necessary (that is, it may all boil down to constant conjunction). Either way this is a seperate question from the question of what gets observed.

2. As for the other point about how we can tell the difference between causation and corelation, or even coincidence…I don’t know…but what ever we say notice that this is a general problem for perception, not just the perception of the causal relation. I mean, how do you tell the difference between a red wall and a white wall with a red light shone on it? Is there some perceptible thing that will cue you to the difference? I don’t think so. But this is no reason to despair because our sensory systems are generally reliable and so we have a defeasable justification to believe that things are the way that they seem to us.

3. One reason why I might think that turning the light switch matters, and not the bird chirping, is that I can set up experiments where I control the circumstances and thereby narrow down what it is that is doing the causing. At any rate, it is fairly easy to rule out the bird…This is actually the way that science usually works, we observe how things work and then devise ways to interfere with the thing of interest to see what happens.

what do you “observe” that allows you to dinstinguish between causality and mere coincidence?

in fact, the only way we could make such a distinction between (1) and (2) is by saying that the lights seem ALWAYS to go on after i flip the switch, whereas birds do not always chirp when i do this. hence hume’s contention that we call “causality” is just the observation of constant conjunction. the conjunction between the flip of a switch and a light’s illumination may be more consistent than that between the flip of a switch and a bird’s chirping, but there is not a difference in kind between our perception of either phenomenon.

(this of course leads into the problem of induction; i.e., why suppose that two events that have been conjoined in the past should be conjoined in the future?)

That’s a good point about Hume’s theory of perception. He certainly talks as though we see the billiard ball, I mean that is what the word ‘impression’ conjures up. We have this picture that the object itself makes the impression, but it is not clear that that is what Hume means at all. Since Descartes most philosophers had endorsed the view that our ideas were really judgments based on sensory experience (the empiricists, of course, thought that ideas resembled objects, whereas rationalists like Descartes, and later Kant, denied that). But there is a difference between the billiard ball and causation for Hume in that the former but not the later can be traced back to an impression. The relation between the impression and the object is a different question all together (one that I, personally, think that a Kripke kind of story can be given).

But, I just realized you were probably replying to the last comment and not the post…ah I see…then yes I see the analogy that you are making. In the case of seeing objects, we say that what counts as seeing them is making a correct judgment based on our impressions, so we can say the same thing in the causal case; what counts as seeing the necessary connection is a. seeing the connection, and b. judging (correctly) that it is necessary. That would mean that if one accepts a Humean theory of perception we should be more inclined to think that we could come to see THAT the connection was necessary. Nice point. Is this what you meant?