For instance,a guy is watchig footbal match in a vidoe club.He admirly watches how one of his favorite players turns around the ball.He yells”My favorite plays well than any other players. One guy beside him compares his favorite to one spectacular and world class player”Have you seen Lionel Messi playing? the guy beside asks. He answers back”I hate Lionel Messi without reason” Those in the club watching ask for his reason but he vehemently states that he hates Messi with no reason.

If we go with the more general, we’re back to the point that it’s trivial that everything has a sufficient reason, because everything becomes a sufficient reason for itself. (If y is the case, y is the case). Simply stipulating that x not be y doesn’t seem to quite work, either (see the response to Jason below).

Jason,

I actually think that there’s good reason for a proponent of PSR to find an interpretation of the principle that allows contingent things to have sufficient reasons; a great many PSR proponents, in any case, do hold that there are contingently existing things, and are committed to saying that there are sufficient reasons for them. The real question is, as you say, whether this can be done without making PSR trivial.

I like the general approach in your suggestion about trivial and substantive sufficient reasons. But I don’t think your suggestion actually gets us out of the problem. For suppose instead of making the sufficient reason for y the process up to and including y, I made it the process up to and arbitrarily close to the actual production of y (but not including it). Then we have an r that is not identical to y, nor contains y as a part, but also seems trivial.

1.
I wonder if instead of:

a ’sufficient cause of x’ is anything that has an outcome of x such that if it occurs the outcome occurs.

one can go with more general:

a ’sufficient reason for x’ is any y, such that if y is the case, then x is the case.

2.
Maybe we can distinguish pseudo-random for really-random processes, based on something like this:
A process is really random if, all things being same in a set of possible worlds, it still has different outcome in those different possible worlds. Flipping the coin will fail to satisfy this condition, as all things being same, it will always (presumably) produce the same outcome.

You say:

Rather, what I’m saying is that the sufficient reason(which we can call “d”) does not itself *have* to have a sufficient reason (although it might). So I think e can have a sufficient reason (namely, d) for its occurence, while d itself has no sufficient reason for it’s occurence.

I agree with that. The argument doesn’t though work on the issue if d has sufficient reason itself, but if there is a sufficient reason (and based on some further thought I have, probably would require ‘necessary and sufficient’) for “d is a sufficient reason for e”. Call it meta-reason if you want.

So when you say:

There is a suffient reason for d’s being sufficient to produce e *if d occurs*…

, we agree on that too. Now, of course it is open question what this meta-reason might be. I left the issue open in the post, and worked on two assumptions: 1)that there is this kind of meta-reason which determines why e1 happened which sufficient reason and e2 without them. and 2)that there is no such meta-reason.

So, given the assumptions that either there is no such meta-reason, or there is no such meta-reason (but not both), and that both 1) and 2) come as impossible, argument concludes that there can’t be that e1 happens with sufficient reasons, and e2 without sufficient reasons.

Actually, I did have that in mind. But I think such an understanding of “sufficient reasaon” would trivialize PSR, because on such a reading even contingently existent beings have sufficient reasons for existing (and these reasons would themselves have sufficient reasons, etc.). So perhaps we can distinguish between trivially sufficient reasons and substantively sufficient reasons along these lines: a sufficient reason r for some entity x counts as a trivially sufficient reason for x if either r or some part of r is identical to x, and r counts as a substantively sufficient reason for x otherwise. This is supposed to capture the idea that a non-trivial r ought to make a *non-redundant* contribution to x’s existence or occurence.

What do you think?

I think this raises, actually a question about what we are talking about in raising the question of sufficient conditions. According to a very common way of characterizing necessary and sufficient conditions (in terms of implication) everything does necessarily have a sufficient condition: itself (since p always implies p, p is always a sufficient condition of p). What we often mean when we talk about sufficient conditions in this sense is a sufficient cause. But this is also tricky to characterize in the context: On one way of looking at ’sufficient cause of x’ is to regard it as anything that has an outcome of x such that if it occurs the outcome occurs. But it would seem an outcome of a random process can indeed have a sufficient cause in that sense, namely, the random process as it unfolds in this particular case. Thus flipping a coin would not be the sufficient cause of its landing heads, but the whole trajectory of the coin from flip to final outcome would be. But we might want an account of ’sufficient reason’ that doesn’t allow this.

First, I don’t think the opponent of PSR is committed to (1), only that there doesn’t *have to* be a sufficient reason for its having a sufficient reason rathen than not.

Second, I don’t think (2) is really a contradiction, because I don’t think it follows that if an event e has a sufficient reason then there must be a a sufficient reason for e’s having a sufficient reason. So if “e’s having a sufficient reason” lacks a sufficient reason, I don’t think it follows that e lacks one.

You say: “that the sufficient reason was sufficient reason for the event is negated by saying that there is no sufficient reason for it to be sufficient reason”

But I’m not saying that the sufficient reason for e is not sufficient to produce e. Given this particular sufficient reason, e must indeed happen. Rather, what I’m saying is that the sufficient reason(which we can call “d”) does not itself *have* to have a sufficient reason (although it might). So I think e can have a sufficient reason (namely, d) for its occurence, while d itself has no sufficient reason for it’s occurence. There is a suffient reason for d’s being sufficient to produce e *if d occurs*, but there need not be any sufficient reason for d’s occurence. Compare this with propositional logic: It can be logically necessary that p implies q even if p itself is not logically necessary. If p is given, q is *conditionally* necessary, but it is not absolutely necessary, because p itself is not absolutely necessary.

OK, so, if opponent of the PSR accepts that:

1) There can’t be a sufficient reason why some specific thing happens with a sufficient reason and other without a sufficient reason.

We go to the (2)

2) From (1) we get that there was no sufficient reason why the thing that happened with sufficient reason, happened with sufficient reason rather than with no sufficient reason.

And to me (2) sounds as contradiction, because it says that both there was a sufficient reason for the event, and also that there was no sufficient reason for the event (that the sufficient reason was sufficient reason for the event is negated by saying that there is no sufficient reason for it to be sufficient reason).

Can’t we rephrase (1) in terms of sufficient conditions though?

There can’t be sufficient conditions for some specific thing having sufficient conditions and other specific thing not having sufficient conditions, as there can’t be sufficient conditions for the specific thing happening without sufficient conditions.

(BTW, I was having defense of principle of sufficient reason on mind when writing the post, so, ‘reason’ stands for ’sufficient reason’ in the post. Thanks for pointing to the ambiguity.)

I agree with Richard; I think (1) depends on an equivocation on “reason”. Specifically, I thing there is an equivocation between something’s being a reason for x in the sense that it is a *sufficient* condition for x’s existence or occurence, and something’s being a reason for x in the sense that it is a *necessary* condition for x’s existence or occurence. Nothing can exist or occur without having all the necessary conditions for its existence or occurence met, and in that sense everything does indeed have a reason. But I see no reason why everything has a sufficient condition for its existence or occurence. Now, when you say in (1) that

“There can’t be a reason why some specific thing happens with a reason and other without a reason, as there can’t be a reason why something happens without reason.”

I think you rely on the equivocation mentioned above. For there can indeed be a reason why something happens without a reason, in the sense that an event e could lack a sufficient condition for happening, and thus be said to happen for no reason, while the fact that e has no sufficient condition itself is a necessary condition for e’s happening without a sufficient condition.

What do we mean by saying that “it is an outcome of *a random process*” if not that there is no reason why the “outcome” is as it is?
So, I don’t think that we can specify “it is an outcome of a random process” as a reason for something being without a reason. It, seems to me, that “outcome of a random process” is just a way of saying that there is no reason why the “outcome” is as it is.

Interesting post!

One question that I have, is what you mean in premise 1 when you say that there can’t be a reason why something happens without reason. So, take some random process, say like the ones we are told occur in quantuum theory, then there is no reason for one coutcome over the other, in the sense that there is causal reason, or no determining reason, but in a more general sense it is easy to see that the reason for the outcome is that it is the outcome of a random process…so it looks to me like the argument that you present equivocates on ‘reason’