The principle of sufficient reason (PSR) in a nutshell says that things happen for a reason. Occam’s razor suggests to not postulate things that bring in additional assumptions without doing any explanatory work. But isn’t the postulate that things must happen for a reason an assumption itself?

In every case of an event that we found to have an explanation for its existence, we found it by finding evidence for it, not by merely assuming that a further explanation exists.

Contrary to the widespread intuition that by default everything has an explanation, it seems instead that one should assume there is no explanation for X since that is by default more parsimonious. After all, what is simpler than X occurring for no reason or perhaps even necessarily? If one postulates Y to explain X, that Y will now be left unexplained. But now you have X and Y rather than just X which is arguably necessarily more complex (atleast from an ontological perspective).

Of course, once you find evidence for the fact that X has an explanation, the postulate that X must have an explanation isn’t an unneeded assumption anymore, and hence now there is no violation of Occam’s razor.

So, does PSR without any antecedent evidence for a further explanation or reason behind something, violate Occam’s razor?

  • 1
    PSR is an assumption in the same way that Occam's Razor is an assumption. In fact, in another, greater sense ALL postulates are assumptions because we cannot prove them. If we could prove them, then they wouldn't be postulates, they would be conclusions. Commented Mar 23 at 17:56
  • Occams Razor is useful when one has a bloated set of options and needs a way to start somewhere. It is not useful when one doesn't have any principles, and is first establishing their fundamental Axioms. Occams Razor isn't a universal law of thought, it is a "best practices" heuristic for specific circumstances. There are other relevant criteria for assuming things than the number of assumptions one wishes to make. Commented Mar 24 at 21:23
  • More formally, Occams Razor is a "should not" principle, and PSF is a Postulate about causality. They can't contradict each other. Even if we had a "negation" of Occams Razor, to favor complexity, it wouldn't violate Occams Razor- as Occams Razor is a "should", rather than a "must". Commented Mar 24 at 21:38

8 Answers 8


No. To consider them at odds would be the product of committing a category mistake. The Principle for Sufficient Reason is an observation about how causality and explanation are universally applicable to events, where as Occam's Razor is a philosophical razor about how parsimony should govern arguments and therefore theories. The first is an Is, and the second is an Ought.

But while they are generally orthogonal, they do intersect at a point. That is, parsimony in arguments helps us to deal with the glut of potential explanation and description that results from the consequences of accepting the PSR. If everything must have a reason or cause, then one must decide what those reasons and causes are. This inevitably leads to arguments, and when made, those arguments and theories should be governed by parsimony. An example would help.

In philosophy of mind, one is tasked for explaining why minds exist. The PSR says that the existence of mind must have a reason or cause, and therefore an explanation must be devised. But Occam's razor suggests, whatever the explanation, simpler is better. An eliminative materialist may argue that folk psychological theories are loaded with all sorts of claims about what the mind is and does, and simply say that such complexity doesn't do a good job of reducing to biology. Less is better. In the extreme, a logical behaviorist will argue that mental states are best not considered at all, and that one should consider understanding the mind as an illusion to account for behaviors. Most physicalists say that the mind has nothing to do with the soul, because given theories of mind that align them with the soul are unfounded and unparsimonsious given the soul is non-material and superfluous.

All three philosophical positions attack other philosophical positions that are more complex and attempt to eliminate entities that populate other theories. An eliminative materialist wants to do away with many complex mental states. A logical behaviorist often seeks to dismiss them entirely, both positions which are usually physicalist completely dismiss any talk of the soul as non-existent. Thus, Occam's razor is applied while attempting to satisfy the PSR.

Thus, parsimony and causality are two different ideas that, with a good thinker, align quite well. They are not oppositional because one is a claim about the necessity of explanation, where as the other is a heuristic to govern the construction of explanation.

  • 1
    I think you've chosen a rather awkward example. PSR is about things and their causes or reasons. Now you may have thoughts that come from your mind, but you don't see your deepest mind, you see thoughts (in your conscious mind). Insofar as mind is being-like it is not a thing. "'Being' cannot indeed be conceived as an entity." (Being & Time ¶1.2). PSR says "no thing is without reason" so it's application to mind is questionable. Commented Mar 24 at 9:40
  • 1
    Here’s the kicker though: what if rather than trying to decide between causes/explanations, we have to decide between an explanation existing for X vs. an explanation not existing. Let’s use the universe as an example. In this case, how do we decide between the two? It seems that Occam’s razor would allow us to prefer no further explanation for the cause of the universe until there is evidence. Assuming that there is an explanation moves back the mystery by one step. Here, the very notion of deciding whether PSR is true may be influenced by the Razor Commented Mar 24 at 13:14
  • @ChrisDegnen My understanding of PSR, I believe, is consistent with the characterization in the SEP. Would you cite a passage which supports your contention, for me?
    – J D
    Commented Mar 24 at 16:10
  • 1
    @Baby_philosopher That's intriguing, and I hadn't considered it. I'm not able to produce a rejoinder. Let me think about that. Thank you for you provoking thought.
    – J D
    Commented Mar 24 at 16:13
  • @JD One part is that mind is being-like, which I may have to answer later if that is requested. The second part is from Heidegger's The Principle of Reason, page 51 : re. PSR and being, "one might be inclined to understand this in the sense of "being has a reason," that is, "being is grounded." The popularly understood and presumably valid principium rationis never speaks of this. According to the principal of reason, only beings are ever grounded." Commented Mar 24 at 16:48

You've not been paying attention to the answers to your earlier question about Occam's razor, which should have made it clear to you that an argument with fewer assumptions need not be better if the assumptions are unrealistic. In your current question, you suggest we might assume things happen for no reason- why do you consider that to be a realistic assumption when virtually every experience you have about the world suggests it is untrue?

  • 2
    Quantum mechanics suggests that certain things happen without cause. And since it’s the most fundamental form of reality, uncaused things happen gazillions of times per day! Commented Mar 23 at 15:24
  • 1
    @MarcoOcram, using that logic, something that happens once every few lifetimes is safely deemed to be impossible as every experience most people have suggests it doesn't happen. Commented Mar 23 at 21:20
  • 1
    @Flash_Steel absolute nonsense- your suggestion and mine are not equivalent. Commented Mar 23 at 22:07
  • 1
    Your argument boils down to "virtually every experience in your life suggests A so assuming not A is unreasonable." Commented Mar 24 at 6:34
  • 1
    @Flash_Steel and that argument holds in the case in which not A directly contradicts virtually every experience. If you say that events do not have causes, I can point to plenty that do, so your statement is incorrect. If you say that some events do not have causes, that is a different matter. Even then the issue of plausibility arises. I have never seen a giant pink rabbit, so I would be skeptical of any theory that required one to exist in order to explain everyday phenomena. Commented Mar 24 at 6:40

OP: Contrary to the widespread intuition that by default everything has an explanation, it seems instead that one should assume there is no explanation for X since that is by default more parsimonious.

Identifying the reason for a thing is part of understanding what the thing is. If you don't want to find its reason it will still have one, even if it is just that the thing appeared in your eyeline. This quote from Heidegger's The Principle of Reason (pages 118-9) might clarify things for you.

In showing the extent to which the principle of reason founds all principles—that means, first of all, founding every principle as a principle—Leibniz raised the nihil sine ratione, nothing without reason, to a supreme fundamental principle. This character of the principle of reason becomes clear in the complete Latin title Leibniz gave the Principle. Leibniz characterizes the principle of reason as the principium reddendae rationis sufficientis. We will translate this title while discussing its individual determinations. The principium rationis is the "principium reddendae rationis." Rationem redder means: to render the reason. ...

  1. How come a reason is always a rendered reason? ...

A reason is a rendered reason, quod omnis veritatis reddi ratio potest, "because a truth is only the truth if a reason can be rendered for it."49 For Leibniz, truth is always—and this remains decisive—propositio vera, a true proposition, that is, a correct judgment. Judgment is connexio praedicati cum subjecto, the connection of what is stated with that about which a statement is made. That which, as the unifying unity of subject and predicate, supports their being connected is the basis, the ground of judgment—it gives the justification for the connection. Reason renders an account of the truth of judgment . "Account" in Latin is called ratio. The ground of the truth of judgment is represented as ratio.

Accordingly, in a letter to Arnauld, Leibniz writes: "Hanovre le 14 juillet 1686: il faut tousjours qu'il y ait quelque fondement de la connexion des termes d'une proposition, qui se doit trouver dans leur notions. C'est là mon grande principe, dont je croy que tous les philosophes doivent demeurer d'accord, et dont un des corollaires est cet axiome vulgaire que rien n'arrive sans raison, qu'ont peut tousjours rendre pourquoy la chose est plustost allé ainsi qu'autrement. . . ." In translation: "it is always necessary that there be a foundation for the connecting of the parts of a judgment, in whose concepts these connections must be found. Precisely this is my grand Principle about which, I believe, all philosophers must concur—and this common axiom remains one of its corollaries—that nothing happens without a reason that one can always render as to why the matter has run its course this way rather than that."50

The grand Principle is the principium reddendae rationis, the fundamental princi­ple of rendering reasons.

  1. Leibniz, Philosophische Schriften, ed. Gerhardt, 7:309 [Philosophical Writings, ed. Parkinson, p. 75].

  2. Briefwechsel Leibniz, Arnauld, und dem Langrafen Ernst v. Hessen-Rheinfels, ed. C. L. Grotefend, (Hannover: Verlag Hahnschen Hof-Buchhandlung, 1846), p. 49. Cf. Leibniz, Philosophische Schriften, ed. Gerhardt 2:56, 62. [Discourse on Method, Correspondence with Arnauld, and Monadology, trans. George R. Montgomery (LaSalle, Ill.: Open Coun, 1973), p. 132].

  • 1
    ""because a truth is only the truth if a reason can be rendered for it."" Surely not. Reality presumably is what it is even if we are unable to find a good reason for the fact that it is. Commented Mar 23 at 10:39
  • 2
    @Speakpigeon If there is a spoon it is for the reason that at some time it has made an appearance. In Kantian parlance, if a thing has never been seen and no-one knows anything about it then it does not attain Kant's standard for existence, i.e. "Logically, [existence] is merely the copula of a judgement." A598/B626. It's reasonable to say undiscovered things exist though, but that's all they exist as, and their reason is that undiscovered things have previously been discovered. Commented Mar 23 at 10:56
  • 1
    "a spoon" Reasoning by examples is bad logic. - 2. "Kant's standard for existence," Kant is dead so why should we care about his standard for existence? - 3. ""Logically, [existence] is merely the copula of a judgement."" Equivocation between what we mean by existence and the possible logical treatment of the notion of existence. Commented Mar 23 at 11:10
  • 1
    3. "It's reasonable to say undiscovered things exist . . . their reason is that undiscovered things have previously been discovered." No. We have zero reason to give for the existence of undiscovered things. A reason for our belief that there are undiscovered things is not a reason for the existence of any of these things. Commented Mar 23 at 11:12
  • 2
    I doubt physics can solve the paradoxes Kant addresses as the Standard Model breaks down at the Big Bang and also any black holes. It is one of the things that makes both so fascinating to study. Currently anything at the time of the Big Bang or before, or beyond the event horizon of any black hole is metaphysics. As these subjects are metaphysics, I think the only way forward is for philosophers, mathematicians and physicists to work together. Commented Mar 24 at 14:18

But isn’t the postulate that things must happen for a reason an assumption itself

Hume would agree, in A TREATISE OF HUMAN NATURE

Though distant objects may sometimes seem productive of each other, they are commonly found upon examination to be linked by a chain of causes, which are contiguous among themselves, and to the distant objects; and when in any particular instance we cannot discover this connexion, we still presume it to exist.

So, does PSR without any antecedent evidence for a further explanation or reason behind something, violate OR?

Occam's Razor says "Entia non sunt multiplicanda praeter necessitatem," Wikipedia translates this as "Plurality must never be posited without necessity". Wikipedia - Occam's razor (I'll change the link to actual source material if anyone leaves one in the comments.)

If PSR is taken to mean every effect/entity has a cause then this assumption comes with an infinite chain of causes and effects spanning way back before the Big Bang which is a huge metaphysical assumption.

Therefore, if you have something which is as good at predicting outcomes or modelling reality as PSR but with less baggage then choosing PSR still would violate OR as it is not necessary to make the huge metaphysical assumption. Without a rival principle or axiom then OR does not apply.

If PSR is taken to mean one can rationalise why something is the way it is this seems more of a tautology than a useful principle and does not require the reason correlating with reality.


Occam's razor is consistent with the PSR, because in general, a world in which there are reasons for things is a simpler world.

If it is necessary to describe a world containing A and B, a description that knows no connection between them will have to describe each individually, whereas a description that knows B is a consequence of A can describe both just by saying "A and its consequences". Knowledge of how things are connected (causally or otherwise) allows redundancies to be removed from their descriptions, which makes them more to Occam's taste.


The principle of sufficient reason (PSR) in a nutshell says that things happen for a reason. Occam’s razor suggests to not postulate things that bring in additional assumptions without doing any explanatory work.

The principle of sufficient reason claims that all events have an explanation.

Occams Razor is not used to create explanations. It's used to evaluate competing explanations of the same phenomenon that have equally passed all scientific or philosophical scrutiny: All things being equal, choose the simpler explanation. Occam's Razor is applied after exhausting all other comparisons.

So PSR and Occam's Razor are partners: PSR and Occams Razor claim that all events have an explanation and that this explanation can be as simple as possible but not simpler.


Apparently to invoke the principle of sufficient reason is only useful when it is used as a heuristic: Now search for the reasons or explain why there is no reason!

Otherwise the principle is superfluous because it has no explanatory power. Then it should be skipped according to Occam's razor.

  • 2
    "Otherwise the principle is superfluous because it has no explanatory power. Then it should be skipped according to Occam's razor." But Occam's razor also says that once we have a sufficient reason, we can dispense with Occam's razor . . . Commented Mar 23 at 10:34
  • @Speakpigeon Could you please give a reference for your last sentence about Occam's razor, thanks.
    – Jo Wehler
    Commented Mar 23 at 14:33
  • First, give me a reference for your claim that according to the principle of Occam's razor a principle with no explanatory power should be skipped. Commented Mar 23 at 17:26
  • @Speakpigeon The common formulation of Occam's razor reads: "Non sunt multiplicanda entia sine necessitate (Entities are not to be multiplied without necessity)!"
    – Jo Wehler
    Commented Mar 23 at 18:22
  • "Entities are not to be multiplied without necessity" Good. So? Commented Mar 24 at 10:29

The principle of sufficient reason and Occam's razor are two different ways of expressing the same idea, namely that since human beings are only logical, there is no compelling argument outside logical arguments.

Here is how Spinoza formulate the principle in 1663, before Leibniz:

Nothing exists of which it cannot be asked, what is the cause (or reason), why it exists.

Spinoza was articulating Descartes’ views. This is markedly different from Leibniz interpretation. According to Spinoza, Descartes thought that nothing exists of which we cannot ask why it exists. Sure, we can always ask, but this of course doesn't mean that we can always give a reasonable answer to our question, or indeed whether there is always a reason (or a cause) to everything. And there cannot possibly be.

  • 1
    "human beings are only logical." In what possible world?
    – J D
    Commented Mar 23 at 10:43
  • @JD I ought to feel jealous . . . How many possible worlds do you know?! - 2. If you think we are not logical, why are you even asking anything at all? Commented Mar 23 at 10:56
  • Re: "this of course doesn't mean that we can always give a reasonable answer" - but an answer can be given even if it's trivial, like "the thing appeared in my eyeline". It's there because I'm seeing it. Maybe that's all that can be known, but it is a cause. Commented Mar 23 at 11:04
  • @ChrisDegnen "It's there because I'm seeing it" You must be kidding. No, not even that. Kids, no, toddlers, learn to think of existence as distinct from mere appearance. You are confusing the reason that we can decide that something exists (because we see it), with the reason why the thing exists to begin with whether we can see it or not. Commented Mar 23 at 11:17
  • Well, Vulcan is one possible world, where the beings are devoted to reason and logic. Another possible world is an Earth governed primarily by the principles of the Enlightenment. But neither are good descriptions of what goes on our world currently. Both of the possible worlds I mentioned are governed by beings that are much more (but still not only) logical. And I ask, because your claim doesn't seem only logical. It seems motivated by emotional aspirations rather than parsimonious description of the state of affairs.
    – J D
    Commented Mar 23 at 11:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .