Principle of Sufficient Reason: Trichotomy?

[Astro Cat]

Does every state of affairs have an explanation or a cause? Stated simply, the Principle of Sufficient Reason (henceforth PSR) is that for every fact F, there is a sufficient reason for why F is the case. The notion is deeply connected to causal relations and determinism: we're often employing the PSR when we answer questions of the form "why is x the way it is" with "because of y."

For every explanation we give to explain some fact, though, it quickly becomes clear that there's going to be a regress: if x is explained by y, then y is the way that it is? On the PSR, there would be some sufficient explanation z -- and so on. Where does the ultimate sufficient explanation lie, then, if we follow this regress "all the way back?"

For instance, I could ask "why is grass green," and come up with a sufficient explanation involving the most efficient wavelengths to absorb and reflect from a star with the sun's properties. I could ask "why does the sun have those properties" and come up with some sufficient explanation for how a Sun-like star develops. I could ask "why would it develop that way" and come up with some sufficient explanation involving the physics of star formation, and so on (I am skipping a lot of things we could ask in between).

So really, in order for the PSR to work, we don't just need one sufficient explanation, but a whole interlinked chain of sufficient explanations in order to call some apparently contingent fact explained. But therein seems to lie the problem: it seems like in order to answer the question of whether some contingent fact is ultimately explained, we encounter a trichotomy -- where none of the options of that trichotomy either ultimately explain the apparent contingency, or where contingency doesn't exist at all!

If we follow a chain of sufficient explanations backwards, the trichotomy (as far as I can see it) is this:

----------

1) There is an infinite chain of explanations. Under this horn of the trichotomy, we would say that the apparently contingent fact we're interested in has a sufficient explanation, which has a sufficient explanation, and so on down an infinite regress where each fact is sufficiently explained by the fact "preceding" it. This might on its surface seem satisfactory even with misgivings about infinities because, indeed, when we zoom in to individual facts, each one is sufficiently explained by the one preceding it.

However, the problem here is that this entire chain of facts sufficiently explained by their preceding fact is itself a sort of fact, and we are left with the question: "why this chain of contingencies, and not some other one?" There's nowhere "further back" to go in order to sufficiently explain this meta-fact! So is the PSR ultimately satisfied? I would say not!

2) There is some modal ontologically necessary fact that serves as the foundation of a finite chain of sufficiently explaining facts. On this horn of the trichotomy, we get around the meta-problem from the first horn by supposing there might be some modally necessary grounding fact. This might seem sensible since we don't have to give a sufficient explanation for a necessary thing by definition: after all, that thing is the way it is because it's necessary to be so.

The problem with this idea is that modally necessary causes must have modally necessary effects: how could it possibly be otherwise? If A necessarily --> B, then B itself is necessary. If B --> C, and B is necessary, then B necessarily --> C, which then makes C necessary. We end up with a chain of modally necessary things and never arrive at a contingent fact at all! This would be a sort of Spinozan fatalism in which there are no contingencies, which renders the PSR moot as there are no contingencies to sufficiently explain.

3) There is some finite, first step in the chain of explanations which has no explanation. Perhaps there is true randomness, and if we look backwards through a chain of explanations we'll eventually hit some explanation that just is, yet isn't itself modally necessary. It's obvious from the start that this horn abandons any pretense of the PSR from the start, so not much more needs to be said about it.

----------

None of these three options seem to retain the PSR. What are we to think about this? If the universe has contingent facts, then the PSR tells us those facts have an explanation. Yet those explanations require explanations on the PSR. If we have an infinite chain, then we have to wonder "why that infinite chain and not some different one" in a sort of meta-fact that requires explanation. If we have a chain we try to anchor in finiteness by giving some foundational cause or explanation that's modally necessary, then we don't have a contingent fact to explain at all because that modally necessary foundation just causes a chain of necessary effects of necessary causes. Lastly, if we suppose that at some point there's an explanation that doesn't itself need to be explained (yet isn't modally necessary), then we're simply proposing true randomness: causeless causes, and we have already abandoned the PSR.

What to do about this?

[GE Morton]

None of these three options seem to retain the PSR. What are we to think about this? If the universe has contingent facts, then the PSR tells us those facts have an explanation. Yet those explanations require explanations on the PSR. If we have an infinite chain, then we have to wonder "why that infinite chain and not some different one" in a sort of meta-fact that requires explanation. If we have a chain we try to anchor in finiteness by giving some foundational cause or explanation that's modally necessary, then we don't have a contingent fact to explain at all because that modally necessary foundation just causes a chain of necessary effects of necessary causes. Lastly, if we suppose that at some point there's an explanation that doesn't itself need to be explained (yet isn't modally necessary), then we're simply proposing true randomness: causeless causes, and we have already abandoned the PSR.

Your last statement there--- "we have already abandoned the PSR" --- is not true. The PSR begins with an indubitable but inexplicable fact, namely, that something exists. We can't answer the question, "Why does anything exist?," but given that it does, we can ask, "Why does X exist"," for any value of X. The PSR chain may be long for any X, but it will terminate at the "brick wall" of "Something exists."

Notice, BTW, that any effort to provide a cause for "something exists" will be circular, since any proffered cause will itself be embraced by "something."

[Astro Cat]

None of these three options seem to retain the PSR. What are we to think about this? If the universe has contingent facts, then the PSR tells us those facts have an explanation. Yet those explanations require explanations on the PSR. If we have an infinite chain, then we have to wonder "why that infinite chain and not some different one" in a sort of meta-fact that requires explanation. If we have a chain we try to anchor in finiteness by giving some foundational cause or explanation that's modally necessary, then we don't have a contingent fact to explain at all because that modally necessary foundation just causes a chain of necessary effects of necessary causes. Lastly, if we suppose that at some point there's an explanation that doesn't itself need to be explained (yet isn't modally necessary), then we're simply proposing true randomness: causeless causes, and we have already abandoned the PSR.

Your last statement there--- "we have already abandoned the PSR" --- is not true. The PSR begins with an indubitable but inexplicable fact, namely, that something exists. We can't answer the question, "Why does anything exist?," but given that it does, we can ask, "Why does X exist"," for any value of X. The PSR chain may be long for any X, but it will terminate at the "brick wall" of "Something exists."

Notice, BTW, that any effort to provide a cause for "something exists" will be circular, since any proffered cause will itself be embraced by "something."

Could we not apply the PSR to "something exists" in the form of "why that thing and not another thing," if the something is apparently contingent?

[Sy Borg]

There is an explanation for everything, but whether humans can work them out is another matter.

The horns need not be seen as antithetical, rather as tendencies.

So, yeah, there probably is an infinite chain of factors behind any dynamic, but not an infinite number of significant factors for whatever intents and purposes.

And yes, there may be an ultimate grounding fact, but it is not an anthropomorphic magical spirit entity posted in the Iron Age. Ultimately, if there is some kind of first cause, no one knows.

If there is no first cause as such, but our reality emerged from true randomness through raw probabilities, that would be hard to prove too.

[Astro Cat]

There is an explanation for everything, but whether humans can work them out is another matter.

The horns need not be seen as antithetical, rather as tendencies.

So, yeah, there probably is an infinite chain of factors behind any dynamic, but not an infinite number of significant factors for whatever intents and purposes.

And yes, there may be an ultimate grounding fact, but it is not an anthropomorphic magical spirit entity posted in the Iron Age. Ultimately, if there is some kind of first cause, no one knows.

If there is no first cause as such, but our reality emerged from true randomness through raw probabilities, that would be hard to prove too.

I get that it's difficult for humans to work out ultimate explanations; but the purpose of making the trichotomy explicit was to show that even if we don't know what the ultimate explanations are, the PSR doesn't seem to survive intact.

GE Morton might have had a good point that on the PSR we're making at least one brute assumption that something exists and we just aren't expecting ourselves to explain that whether it's modally necessary or not; but I'm not sure whether that solves the conundrum (as with my last response to him). It might, though, I haven't decided. It was a good point to make.

[Pattern-chaser]

Does every state of affairs have an explanation or a cause? Stated simply, the Principle of Sufficient Reason (henceforth PSR) is that for every fact F, there is a sufficient reason for why F is the case.

Am I right in saying that this topic is about, at least partly, "why?" questions? It seems so, to me. And if it is, then it might be worth considering "why?" questions in themselves. I gave this a lot of thought recently, and came up with this.

A "why?" question cannot be answered unless its context/environment is fully understood. If it is fully understood, the answer to the question emerges almost of itself; it has become obvious. But, if we don't have a complete understanding of the context of our "why?" question, then that question becomes impossible to answer.

[GE Morton]

Could we not apply the PSR to "something exists" in the form of "why that thing and not another thing," if the something is apparently contingent?

We could, but that would be the same as asking why anything exists. "Why" questions only make sense when there are conceivable alternatives, and we, being imaginative creatures, can conceive all kinds of alternatives (physicists have speculated on what a universe in which some of the fundamental constants had different values would "be like"). But there can be no sufficient reason for any of those alternatives. You have to begin from the one you experience, and pursue the PSR until you arrive at one of its features.

[Astro Cat]

Could we not apply the PSR to "something exists" in the form of "why that thing and not another thing," if the something is apparently contingent?

We could, but that would be the same as asking why anything exists. "Why" questions only make sense when there are conceivable alternatives, and we, being imaginative creatures, can conceive all kinds of alternatives (physicists have speculated on what a universe in which some of the fundamental constants had different values would "be like"). But there can be no sufficient reason for any of those alternatives. You have to begin from the one you experience, and pursue the PSR until you arrive at one of its features.

I have been fairly busy, just wanted to thank you for your responses and I’ll have more to say when I’m back. I like your thinking on this though.

[EricPH]

There is an explanation for everything, but whether humans can work them out is another matter.

We can't work these questions out in our lifetime.

So, yeah, there probably is an infinite chain of factors behind any dynamic,

When you extrapolate back, the word 'infinite' makes little sense. Sufficient reason leads to two possibilities, something/s had no beginning, or something/s did not come from anything. Sufficient reason leads us to think; both these options require some kind of magic, so we dismiss them. If we are searching for the truth, sufficient reason seems lacking.

If there is no first cause as such, but our reality emerged from true randomness through raw probabilities, that would be hard to prove too.

It would be impossible to prove in our life time.

[Sy Borg]

So, yeah, there probably is an infinite chain of factors behind any dynamic,

When you extrapolate back, the word 'infinite' makes little sense.

Yes, that was loose language. Thanks.

It should have been "there probably are virtually countless factors behind any dynamic".

[Pattern-chaser]

Sufficient reason leads to two possibilities, something/s had no beginning, or something/s did not come from anything. Sufficient reason leads us to think; both these options require some kind of magic, so we dismiss them.

I disagree. Both of these options require some kind of behaviour that is currently unknown and not-understood, so we set them aside, pending the arrival of new information. We don't dismiss them unless we have clear evidence that they are incorrect. To do otherwise would be illogical.

[GE Morton]

When you extrapolate back, the word 'infinite' makes little sense. Sufficient reason leads to two possibilities, something/s had no beginning, or something/s did not come from anything. Sufficient reason leads us to think; both these options require some kind of magic, so we dismiss them. If we are searching for the truth, sufficient reason seems lacking.

Well, that is not so, for the first alternative. No magic is involved in assuming that "something has always existed." As Robert Nozick put it (which argument he credits to his 9 year-old daughter), "Something cannot come from nothing. Therefore something has always existed."

[EricPH]

"Something cannot come from nothing. Therefore something has always existed."

Science doesn't have a clue what that first something was. Supposing hydrogen always existed, then there would only be hydrogen in existence today. It would not have anything else to react with. It seems that 'many somethings' needed to have always existed.

Sufficient reason tells us that everyone who has died on this planet sa far. Has died in ignorance of how the universe came to be. Everyone in our generation will die in ignorance too.

If we are truthful, we can only believe what we choose to believe. Sufficient reason tells me that God created the universe and life. Sufficient reason can also conclude there is no God. So what does that say about sufficient reason?

[GE Morton]

Sufficient reason tells us that everyone who has died on this planet sa far. Has died in ignorance of how the universe came to be. Everyone in our generation will die in ignorance too.

If something has always existed, there was no "coming to be."

If we are truthful, we can only believe what we choose to believe. Sufficient reason tells me that God created the universe and life. Sufficient reason can also conclude there is no God. So what does that say about sufficient reason?

Sufficient reason only takes us to the "brick wall" of that which has always existed. There are no reasons for its existence (because it has always existed).

[Sy Borg]

Sufficient reason tells us that everyone who has died on this planet sa far. Has died in ignorance of how the universe came to be. Everyone in our generation will die in ignorance too.

If something has always existed, there was no "coming to be."

What about state changes, like the Big Bang?