Friday, January 20, 2012

PSR Revisited

I have returned from a brief hiatus from blogging. Work, school, the holidays, and life in general have been calling. I am glad to be back!
The other day at work a colleague expressed some frustration with an issue she had been dealing with. She received an e-mail asking her if there was a “particular reason” she had done something. Exasperated, she said something like “Of course there is. I mean, there’s a particular reason for everything.” I found this extremely interesting because she is a self-described atheist. Not only that, she has a very strong educational background in philosophy.
This got me thinking that it sounded awfully like the Principle of Sufficient Reason (PSR). Interestingly, in everyday conversations in which we do not know there to be theistic implications we tend to grant, almost as a prima facie truth, the PSR. This is what has led Alex Pruss to label the PSR as “self-evident” to anyone who understands it.[1] Further, he contends a major reason for denying it is in the case of “fear that acceptance of the PSR will force one to accept various theological conclusions.”[2] Surely, one must have a better reason to reject the PSR than his or her desire to avoid God.
This also allowed me to consider the question: “what is the reason that everything has a reason?” In her statement, my co-worker intended something like “for every state of affairs X there exists some reason or state of affairs explaining X.” She thought it to be plainly false that there are events or things that are simply inexplicable. So what could be the reason or state of affairs which explains the state of affairs of everything having a state of affairs which explains it?
It seems to me that one cannot always simply appeal to some other, further, explanation—at least not without being willing to accept an infinite regress. But since we know an infinite regress does not—indeed, cannot—explain the entire state of affairs such as we wish to do, it cannot be acceptable. Nor can the series of explanations run in a circle, with each part explaining some other part so that the whole state of affairs is explained. The reason is twofold.
First, suppose we wanted to know why a series of dominoes had fallen. It wouldn’t do any good to explain each domino’s falling in terms of some other domino’s striking it, for it simply doesn’t apply in the case of the initial domino. Suppose we could get simultaneous motion on the dominoes, however. We still do not understand why the entire state of affairs exists at all. That is just to say we do not have an explanation in the relevant sense. Second, the reason or state of affairs which explains the state of affairs of everything having a state of affairs which explains it is just not this type of thing that can be explained by its individual parts. It seems to be a self-evident truth of metaphysics, not a happenstance principle forced onto the universe by perception.
It seems to me the most plausible solution to the PSR’s truth is ultimately God. Why? Because the truth of the PSR is most plausibly an expression of logic and truth. Logic and truth are a part of reality. So we can use an argument to our advantage suggesting God is the sufficient reason for just anything and everything that does exist, or comports with reality/truth, including the PSR and its truth.
1.      The PSR is true.
2.      If the PSR is true and reality exists, then reality has its explanation either in its own necessity or another cause.
3.      There is nothing real external to reality.
4.      Reality exists.
5.      Therefore, reality has its explanation in its own necessity.
6.      At least some part of what is real could have failed to exist.
7.      If some part of what is real could have failed to exist, that part is contingent.
8.      Reality would be different were any parts of it different (explanation of possible worlds).
9.      There could have been different realities (or possible worlds [analytically true]).
10.  Therefore, the reality which now exists is not necessary (analytically true).
11.  Therefore, reality does not find its explanation in its own necessity.
But now consider:
12.  If God is a necessary being, then he is part of reality.
13.  God is a necessary being.
14.  Therefore, God is part of reality.
15.  Therefore, reality finds its PSR in God (analytically true).
(2) is definitional, (3-4) are demanded of any serious thinker, and (5) is a conclusion. (7-8) are also definitional, and (9) is true if (6) is true, and the same goes with (10) as a consequence of (9). (11) results from (3) and (10), which of course makes (5) and (11) contradictory. (6) may be opposed by a “hardliner” who just wants to claim a modal collapse and that everything is necessary, but this is not a road well-travelled. (12-13) are definitional as a hypothesis, and (14) is the result if we grant these. But those engender (15), if the other foundational premises are true.
Therein lies the rub. The atheist/objector may cite (2) and insist, via modus tollens, that the PSR (1) is false. However, any attempt at reasoning that the PSR is false will rely on premises that are less obviously true than the PSR. Additionally, we have seen that even an atheist, when they are left with their intuitions and experiences, will believe strongly that the PSR is true. But if the PSR is true, the major explanation of the universe and reality is likely to be God.

[1] Alexander R Pruss, The Principle of Sufficient Reason: A Reassessment. (New York: Cambridge University, 2010), 14.

[2] Ibid.

All posts, and the blog Possible Worlds, are the sole intellectual property of Randy Everist. One may reprint part or all of this post so long as: a) full attribution is given (Randy Everist, Possible Worlds), b) all use is non-commercial, and c) one is in compliance with the Creative Commons license at the bottom on the main page of this blog.


  1. The first argument, as it stands, is logically invalid up to the conclusion in (5). The deduction goes as follows: From 1 and 4, we get the antecedent of 2, so we modus ponens and get the disjunction "Reality has an explanation in its own necessity" or "Reality has an explanation in another cause." Premise 3 does not negate this second disjunct. In order to do so, this disjunct would either need to say, "Reality has an external explanation," or premise 3 would need to say a bit more like, "There is nothing external to reality, and the parts of reality cannot explain the existence of reality." That is, "another cause" just doesn't cut it.

    Under either of these revisions we run into issues. If we go with the prior, then because God explains reality, God is a part of reality, and parts are internal to their wholes, it looks like reality doesn't have an external explanation and must be necessary if the PSR is true. Under the latter revision, God can't explain reality at all. Thus, it looks like reality comes out as a necessary being here too. Note that it is no help to only talk about *contingent* parts because, in the manner in which we're speaking, there are no contingent parts of reality because "reality" is a rigid designator and we're applying Leibniz's Law very strongly (as is seen in premise 8).

    I think the best way out here (if we're looking to preserve (12)) is to deny 4 and say that reality isn't a literal being which exists and has parts. But in that case, we'd need to revise the argument to be much more clear about what being(s) are in need of explanation that we're paraphrasing as "reality".

  2. Hi Jake. :) I don't see it as logically invalid--after all, the antecedent of (2) is affirmed by (1) and (4), and (3) does in fact rule out another cause--for it is not really the case that some contingent part of reality (which is logically subsumed under reality) explains reality--that would just be to say reality explains itself and is not found in its own necessity, which is viciously circular. I just took that as a definitional truth of "another"--that is, other than reality (and its own necessity or lack thereof). Perhaps that is my fault for not explaining such.

    I take God to be necessary and the foundation of truth/reality, so that consequence doesn't bother me. Perhaps the one problem for the theist may be that one may find this particular reality necessary. I'm not sure that's the case, however, in the case of a God with free will. That is, the necessity of there being a reality at all is found in God, and which reality "plays out" is itself flowing from the necessary truth that God exists, and is the foundation of reality.

    1. If God exists and has free will in the libertarian sense (which you seem to be invoking), then God's decision is uncaused and PSR is false.

    2. Hi Nightvid. I don't see why, as libertarian free will is not construed by me (or almost any theist) as uncaused, but as rather self-caused (not where the choices cause themselves, but where the person or "self" causes the choices as the faculty of the will just is the ability to choose).

  3. Regardless of what else we might say about the argument, the argument must either be invalid or a reductio, because if we accept all of the premises and assume that (5) and (11) are validly deduced, then we should accept the conjunction of them. That is clearly absurd, so something is going wrong somewhere. Because we both accept (11) on independent grounds anyway, we can point to 1-5.

    Now, it looks like there are a few ways that we might be equivocating here. First, I'm inclined to deny that reality exists in the same way as tables, chairs, and people exist, and this can cause problems for the scope of the PSR (but getting into that will be a digression) Secondly, I don't generally read "another cause" (ie, non-identical) the same as "external cause" (non-identical, non-internal). If we do want to read these the same, then it's either the first way or the second. If it's just the first way, then the third premise is false because there are many things that are not identical to reality. If it's the second, then, insofar as God is internal to reality, premise 3 denies that God can be an explanation of reality. Thirdly, one might want to draw a distinction between being ontologically external (like the way tables and chairs are from each other) and being explanatorily external (like the way tires, chassis, motors, etc are to cars and roads). This distinction could probably be motivated by some counterfactual analysis. But if we were to do that, then I'm not sure how we could motivate a premise like "Reality requires something explanatorily external to it" which doesn't just come down to clashing intuition, as the distinction looks like it would require the counterfactual analysis, and a counterfactual analysis would probably require some substantive premises regarding dependence relations for contingent entities which themselves depend upon the intuitions under question.

  4. Thanks for your continued discussion Jake. Yes, I do intend it as a reductio. I also think I see what you're saying here.

    I take reality just to be "that which is true or that which exists," (this avoids demanding that abstractions exist in order to correspond to reality, or that other possibilities which are not actual are not real--which is not the sense in which I mean real in the other premises), which just makes (4) definitional in a generic sense. I realize (8) is the intuition that reality could have been different, so that this particular reality is not necessary. But if it is not necessary, on the first analysis (1-11) we seem to have an impossible situation. This is why I insert (12-15).

    Since God is a necessary being, his "part" of reality is necessarily true. So what's true is that for a generic reality, the explanation why anything is real at all is either of necessity or in an another, external cause. In the case of specific realities, they may vary, so that we may ask why we have this specific reality and not another--but that's beside the point. The point in (8) is to show that reality can't find its ultimate explanation in itself. On an atheistic analysis, we should find that reality is necessary and hence, this specific reality is necessary as well (as there's just nothing to distinguish one reality from another--why should this exist from a necessary truth that some reality exist contingently and not necessarily?). But I now see that premise should have been incorporated in order to avoid confusion.

    I do recognize this won't be compelling, if only because of the clashing intuitions you mention. It would probably have been better to have the argument from contingency in its contemporary form, as it seems simpler and better constructed than what I have here.

  5. Randy,

    Generally, I hold no brief for metaphysical speculations as I find them to be more reports of one's own psychology than any deep insight into the so called 'nature of things'.

    That said, that *something* exists necessarily hardly seems to be a logical truth. It would seem entirely possible that there should be nothing rather than something- the domain of quantification (that which our particular and universal quantifiers range over) is empty.

    Essentially, why there exists something rather than nothing is an open area of inquiry in physics, not philosophy, and there have been some interesting conjectures from that quarter.

  6. Hi Aaron. I don't see any reason to think physics can find anything outside itself for why there is something rather than nothing, so it seems it's ill-equipped to ensure a metaphysical job is done well. ;) And further note it seems there really isn't much of an escape from the PSR even with physics--for there is an assumed explanation for why it is there is something and not nothing. Otherwise, physics is just going to assert the universe's existence as a brute fact, for no reason at all!

  7. Randy,

    Thank you for replying to a comment on a superseded post. There is much I could say on this topic, but I will try to stay concise.

    First, I have no use for metaphysics as I view it as pure obscurantism, and thus metaphysical speculations really amount to naught for me. A quote from C.S. Peirce aptly captures my view:

    "Metaphysics is a subject much more curious than useful, the knowledge of which, like that of a sunken reef, serves chiefly to enable us to keep clear of it."

    Second, I would like to press you on what, exactly, an explanation is. You say the business of science is formulating 'explanations' of physical phenomena. While I may agree prima facie, we may reduce so called 'explanations' to predictive hypotheses: The business of science is a matter of framing hypotheses which imply past observations and which imply future observations under specifiable conditions, which in turn would then serve to confirm or disconfirm said hypotheses. 'God', I would argue, admits of no logical deductions of observable criteria, and is thus cognitively insignificant. In other words, 'God' explains nothing (or, one could argue, 'God' explains everything and thus explains nothing).

    Third, that matter exists as a brute fact is an open question to be answered by physics. If physics makes room for brute facts and this offends our metaphysical sensibilities, that is our intuitions, so much the worse for our intuitions: Science is on much firmer epistemic grounds than philosopher is or could ever hope to be. (As an aside, I am not sure what 'intuitions' are if not one's personal prejudices.)

    Lastly, Alex Pruss's business about denying the PSR stems from a “fear that acceptance of the PSR will force one to accept various theological conclusions” is silly. First, Peter van Inwagen, a prominent theist, rejects the PSR. Second, quite a few atheists accept the PSR (e.g. Arthur Schopenhauer and Quentin Smith).

  8. Hi Aaron, thanks for the response! As far as Pruss is concerned, it is evident he doesn't think this is a necessary condition for rejection of the PSR, but a strong motivator. Nor could we conclude that some atheist's acceptance of the PSR functions as a counterexample, for his claim is not that all people who hold a fear of God would do this.

    As far as metaphysics is concerned, it's only concerned with logic and "the way things work." I don't see any argument contained therein, implicit or otherwise.

    Moving to the business of explanations, I would say an explanation is just a reason, thing, or state of affairs in virtue of which some other thing, event, or state of affairs has obtained as true and not some other thing. I don't see that as being away from science's mission, but a part of it.

    Next, I don't see how physics can have the tools to answer the question! It must rely on philosophy to know whether or not there even are brute facts, much less whether the universe just is a brute fact. Only on an assumption of naturalism would we be forced to work only with physics, which of course would be question-begging here. Also, it's noteworthy science both cannot operate apart from philosophy (even while philosophy can operate in certain areas apart from science) and cannot operate apart from intuition. For the former, just any conclusion reached will depend upon reasoning. For the latter, what justifies any inductive reasoning whatsoever?

    Take the apply falling from the tree to the ground, or a man who releases a ball from shoulder length. If he does this on Earth today, is he justified in thinking it will drop? If not for intuition, it's difficult to see how. For if he says "it has dropped every other time I have done it," he is just assuming a principle that cannot be justified apart from its own truth. He has no reason to think it will not drop. If he says, "well we've seen multiple experiments confirming Earth's gravity and gravity and space--physics confirms it everywhere," but problems abound, of the same variety. Perhaps some mathematical reasoning may come into play here, but that misses the point. We would be forced to conclude that unless the man had knowledge of these mathematical truths, he is not in fact justified in assuming the apple will fall. This is absurd. Our intuition--the shared intuition that drives science to this day--is that if X happens under specified conditions over and over and over, controlling for other factors, we are justified in assuming it's going to happen again. Science cannot account for itself. It desperately needs philosophy, and we do well not to abandon it.

  9. Randy,

    Again, thank you for your response.

    Re: 'As far as metaphysics is concerned, it's only concerned with logic and "the way things work."'

    Logic is the development of systematic techniques for the assessment of arguments for deductive validity and inductive support. The area of inquiry into 'the way things work' is science, the development of systematic empirical techniques & methods for the investigation into the physical world, i.e. the domain of physical 'things'. Neither science nor logic require one to make recourse to 'metaphysics'.

    Re: 'I would say an explanation is just a reason'

    A 'reason' is a psychological term which involves intentionality, etc. An explanation / hypothesis is a linguistic entity which describes & predicts some state of affairs under specifiable conditions; think of explanations as linguistic instruments through which we account for existing data and predict future patterns of sensory stimuli. Though in common parlance many certainly conflate the two, they are distinct. So, e.g., the 'God' hypothesis (so called) permits of no logical deductions of observable criteria, and thus it is not explanatory- indeed, it is not even cognitively significant.

    Re: 'Only on an assumption of naturalism would we be forced to work only with physics, which of course would be question-begging here.'

    Not at all. The issue is one about methodology, not about whether a non-physical personal deity (whatever that means) exists. Even if one were to exist (whatever that type of 'existence' would amount to), it is not at all clear that (1) it did create us (we could still be the result of purely physical processes) and (2) that methodological naturalism is not the appropriate methodological approach.

    Traditionally conceived, philosophy was concerned to provide a firm foundation upon which to build science. However, the history of philosophy is largely a history of its cannibalization by the special sciences, which shows in dramatic relief the problem-solving poverty of traditional philosophical analysis (cf. Leibniz, Descartes, Malebranche, Kant, etc.) and the problem-solving success of scientific methodology. As I said previously, science is on much firmer epistemic grounds than traditional philosophical analysis can ever hope to be.

    Having said this, I should offer the following caveat. Philosophy, as conceived by naturalists, is consonant with science- indeed a part of science- differing only in abstraction: scientists tell us what exists & how these things interact, whilst philosophers analyze the connective tissue of science via logical analyses of concepts such as 'causation', etc. So, conceived in this sense, I can agree in part with you in that science without philosophy is blind, and philosophy without science is empty.

  10. Continued...

    Re: Science's dependency on intuition.

    'Intuition' is often ambiguously used to connote different things, i.e. subconscious reasoning processes, so-called mystical experiences, or some queer cognitive faculty that modern anatomical science has yet to identify. I suspect you are using the term in the latter sense, in which case the lion's share of modern cognitive science research shows that 'intuition' amounts to little more than our preconceived personal and cultural prejudices and is thus not the type of thing which justifies beliefs. In other words, 'intuitions' are evidence of nothing except for the contents of our psychology.

    Now, if science is in an important way premised upon 'intuition' (in the sense in which you are using the term), science is founded upon base irrationalism, much like pseudoscience, faith healing, and every other nonsense under the sun are, and thus science can make no claims to epistemic authority. However, there is something importantly different about science and pseudoscience & mysticism- look at the successes of the former and the failures of the latter.

    It is not that one 'intuits' (whatever that means) the epistemic justifiability of a evidence-gathering method, but rather we look at its reliability and truth-tracking ability in an instrumental sense (we would explicate 'reliability' via something like epistemic iteration [see my post over at FSPB for a presentation of epistemic iteration])- if a method, e.g., induction, continues to produce successful results, we continue to employ it and we partly assess the rationality of beliefs, hypotheses, claims, etc., by virtue of it.

    So, we could run an argument for the rationality of inductive methods over alternatives, say, extispicy (reading animal entrails), in the following way:

    First, let us use a standard disquotational schema for truth:

    DS: 'p' is true if and only if p

    Second, let us consider a standard principle of epistemic justification:

    EJ: S is justified in believing p at t if and only if S’s evidence supports p at t and S believes p at t on the basis of the evidence.

    I take EJ to be true analytically, but by ‘evidence supports p’ I take it that, on the evidence, p is more likely to be true than not-p, where not-p is the set of all alternatives to p. It seems clear to me that it is plausible to say that the evidence makes p more likely to be true than not only if it is plausible to say that the evidence tracks the truth of p, or reliably discriminates p from its competitors.

    Essentially, your options for response are limited. DS is uncontroversial enough and you are, at the terminus of your analysis, committed to EJ, so via some variant of the problem of induction you need to reject that epistemic iteration delivers an appropriate notion of reliability.

    We can pragmatically justify inductive methods in the following way (this is not to imply, however, that this is the only, or even best, way):

    Pace Hume we agree that we cannot know a priori if nature is appropriately uniform so as to permit inferential methods. If nature is not, no rule of inference will work, inductive or otherwise. If nature is, some rule of inference will work. If some rule(s) of inference will work, clairvoyance, extispicy, or any other claptrappery under the sun may or may not work. If some rule(s) of inference will work, induction must work, since if any method works, standard inductive methods or not, the success of the method can be exploited inductively. So, e.g., if clairvoyance works, that is, leads to more accurate forecasts than not, then we can exploit clairvoyance inductively. The method via which we would discover the operable rule(s) of inference would be epistemic iteration. In nuce, we have nothing to lose if we reason inductively, but we have a world to gain.

    Thus, reason obliges that we reason inductively.

  11. Hi Aaron, I think we are getting far afield so this may be my last set of comments (it all depends lol). First, I don't see science as accounting for all of "how things work" unless we presuppose naturalism, which of course I am not wont to do. Even naturalism in our methodology is presupposing what it tries to prove--namely, because it comes dialectically as a move dismissing metaphysics.

    Even pragmatism must be itself a philosophy, and if philosophy is simply reasoning or "thinking hard about something" as Plantinga would simply put it, the logical order literally cannot run from philosophy to support science, but rather the other way around.

    The comments about intuition are true only in the case intuition is not itself justified, which I think the case cannot be made that it wouldn't be (by definition, in fact). For instance, when you say, "if a method, e.g., induction, continues to produce successful results, we continue to employ it and we partly assess the rationality of beliefs, hypotheses, claims, etc., by virtue of it," why assume that this is good? Pragmatism, I suppose. But why suppose what is pragmatic is epistemically justified? Yet we know the princple you list is justified, but it cannot escape the circular logic. It just assumes what it tries to prove, which is the problem with EJ. Unless, of course, "evidence" is defined so as to include background knowledge, which of course includes intuition, which solves the problem. ;)

  12. Randy,

    I agree that the conversation has gone far afield (for that I apologize). Needless to say, we could go on with it, though I will restrain with the last words: Regarding the argument justifying the rationality of inductive methods, you have not understood me at all (though, I was less than clear and left out much detail).

  13. Randy,

    The first paragraph of the comment you posted on Jan 26 displays a fundamental misrepresntation of atheism. Atheism is the state of not believing any god(s) exist(s), not a fear of a god. The latter, I suppose, would be "theophobia", not atheism.

  14. Hi Nightvid. I don't understand what you're saying here, inasmuch as I never claim (nor do any of my claims entail) that atheism's definition is a state of fear of God. It wouldn't even follow from my comments that all atheists reject the PSR due to a fear of God. It only follows that many reject it because of theistic implications as a motivator.


Please remember to see the comment guidelines if you are unfamiliar with them. God bless and thanks for dropping by!