Principle of Sufficient Reason (PSR)

The Principle of Sufficient Reason is a powerful and controversial philosophical principle stipulating that everything must have a reason or cause. This simple demand for thoroughgoing intelligibility yields some of the boldest and most challenging theses in the history of metaphysics and epistemology. In this entry we begin with explaining the Principle, and then turn to the history of the debates around it.

Principle of Sufficient Reason

I think it would be a shame if a discussion over PSR was embedded deep in some other thread somewhere. So here’s hoping Erik and KN will take up any discussion around it here in this thread. (Yes, you’re not just talking to each other.)

But I’ll start by taking the first shot as is my right having created the OP. 🙂 I think there’s a more fundamental disagreement between Erik and KN than the PSR, and that would be over the very possibility of metaphysics itself. Or I could be all wet!

Please discuss. Thank you.

138 thoughts on “Principle of Sufficient Reason (PSR)

  1. Erik: Math will not solve it, because the debate is really over the interpretation of math.

    Fair enough, but I’d add that the structure of reality implied by the math of QM will impose a corresponding structure on any interpretation. One needs to understand the formalism to understand these constraints.

    I’d also venture that a detailed discussion of the metaphysics meeting these constraints will require mathematics for a comprehensible exposition.

  2. I could be misreading the literature, but in the case of QM, it is best to stick to math and not interpret. From what I’ve read in the arena of interpretations, it looks like apologetics.

    There are some things a layman can follow regarding what can and cannot happen, but I’m not aware of any consensus opinion about what is really happening under the covers.

  3. BruceS: Saying that only a single electron was in the apparatus at a time is a standard way of describing the experiment in my experience.

    Saying that only a single electron was in the apparatus at a time is the atomistic way of describing it, taking atoms to be tiny little balls revolving around each other like a mini solar system. Such assumptions were challenged as soon as QM.began to took form. The best men to formulate QM were those who held no atomistic assumptions or were able to drop them, if they held them.

    BruceS:
    BTW, there is a fascinating book on Pauli and Jung, which I suspect you are already familiar with, but I link it just in case.

    And another one, just in case.

    BruceS:
    I don’t really understand those analogies. To me, the Doppler effect is a change in frequency due to different relative velocity, and we hear thunder much later than the corresponding lightening flash because light travels much faster than sound.I had never related these two phenomena to time dilation.

    Sound travels slower due to the heavier density of its medium than light. The difference of frequency resp. relative velocity is how Doppler effect works. And if physical phenomena are to behave analogically, then it follows straightforwardly that the chronometer, if you put speed or gravity on it, should work differently than it did when it was at rest.

  4. BruceS: I’d also venture that a detailed discussion of the metaphysics meeting these constraints will require mathematics for a comprehensible exposition.

    It’s best to avoid the metaphysics of QM (in my opinion). Just do the math and use the data.

  5. Neil Rickert: It’s best to avoid the metaphysics of QM (in my opinion).Just do the math and use the data.

    Right, but we know you’re attitude to philosophy!

  6. Erik: The best men to formulate QM were those who held no atomistic assumptions or were able to drop them, if they held them.

    Yes, they formulated ideas like wave/particle duality to explain the experiments. But this was not intuitive to the formulators. Hence the Bohr quote and many similar from other experts as linked by my post with Bohr’s quote.

    Sound travels slower due to the heavier density of its medium than light.

    Not sure what you mean by “its” medium. The speed of both vary by medium. And they travel at different speeds in the same medium. Light speed does depend on density of the medium it is traveling in (but not in a simple way).

    The difference of frequency resp. relative velocity is how Doppler effect works. And if physical phenomena are to behave analogically, then it follows straightforwardly that the chronometer, if you put speed or gravity on it, should work differently than it did when it was at rest.

    Dopper effect: 1842. SR 1905. What took them so long!!?.

    ETA: By the way, saying “x is at rest” sounds like an absolute description of rest/motion. But it is relative motion that matters. For two non-accelerating frames in relative motion, the effect is symmetric: If A is moving relative to B, A will say that B’s clocks are slower and B will say that A’s are. And both will be right from their point of view.

  7. BruceS: Dopper effect: 1842. SR 1905. What took them so long!!?.

    That mistake won’t be repeated. Erik is busy right now calculating how fast angels can dance on the heads of pins placed in various media.

  8. BruceS: Yes, they formulated ideas like wave/particle duality to explain the experiments.But this was not intuitive to the formulators.Hence the Bohr quote and many similar from other experts as linked by my post with Bohr’s quote.

    Planck and Pauli were not so puzzled. You would find quotes of different nature from them.

    BruceS:
    Not sure what you mean by “its” medium.

    You mean air? If yes, then we mean different things. I will do my best to adjust.

    BruceS: Dopper effect: 1842. SR 1905. What took them so long!!?.

    Why does any discovery take the time it takes? It all looks simple and straightforward only in hindsight, as some say.

    By the way, discoveries themselves are not so important. It’s the importance attributed to them which is important. Gun powder in China was not as explosive discovery as it was in Europe. Same with printing press.

    BruceS:
    ETA:By the way, saying “x is at rest” sounds like an absolute description of rest/motion.But it is relative motion that matters. For two non-accelerating frames in relative motion, the effect is symmetric:If A is moving relative to B, A will say that B’s clocks are slower and B will say that A’s are.And both will be right from their point of view.

    My bad of trying to cover velocity time dilation and gravitational time dilation in a single sentence. But you are better familiar with the physics anyway.

  9. Erik: Planck and Pauli were not so puzzled. You would find quotes of different nature from them.

    I don’t know what Pauli said about wave/particle duality, but he a did say this

    Physics is very muddled again at the moment; it is much too hard for me anyway, and I wish I were a movie comedian or something like that and had never heard anything about physics.

    (I think it was about understanding the details of the spectral lines in certain situations, but that’s from memory for the context of the quote.)

    As for Planck, I believe that he never accepted QM as it evolved after his invention of the quantization of energy.. Wiki says he never accepted Bohr’s interpretation.

    You mean air? If yes, then we mean different things.

    Yes, I was using the word as I understand it to be used in the science of sound and light, or at least its non-QM version for light. I’d be interested in at least reading what it means to you. You can have that last word if you want it; I’ll read it but not reply.

  10. BruceS: I don’t know what Pauli said about wave/particle duality, but he a did say this (I think it was about understanding the details of the spectral lines in certain situations, but that’s from memory for the context of the quote.)

    Remember the book that you recommended to me, and another book that I recommended in return? Read them. They contain what Pauli actually said throughout decades about physics, about everything connected to it in many contexts, and more.

    As for Planck, since you are rather familiar with the topic, you should be well aware of his panpsychic sympathies. Other QM physicists have been more qualified on this, but Planck was not.

    BruceS:
    Yes, I was using the word as I understand it to be used in the science of sound and light, or at least its non-QM version for light.I’d be interested in at least reading what it means to you.You can have that last word if you want it; I’ll read it but not reply.

    We obviously didn’t go to the same school. Not even to different schools in the same country. And we don’t share the same first language. And I am not aware if “medium” is an established concept in QM, apparently not. So, the way I synthesise the concept may be very well unfamiliar to you for all these reasons, but it’s just that – unfamiliar. I constantly have to adjust my own wording to make it more agreeable to people with atomistic and empiricist presuppositions and this is annoying by itself.

    By the way, the Bell Theorem discussion you linked to is great stuff. I found this page particularly interesting, because it states rather clearly that the clash between Einstein and QM was not over math, but over presuppositions, i.e. over how to interpret. For example,

    According to EPR [Einstein et al.], an element of reality exists independent of the act of observation. I.E. all elements of reality have definite values at all times, EVEN IF WE DON’T KNOW THEIR VALUES. In fact, EPR says that any other position would be unreasonable.

    This is not a conclusion from math. This is not something required by any law of logic. This is just a loudly held assumption (“intuition”, if you like). Bell’s procedure, according to the webpage, was to identify and list such related assumptions and devise a way to test them. Which, in my opinion, is how everybody worthy to be called a scientist must proceed. To test any hypothesis is ultimately a form of testing one’s own presuppositions.

  11. Erik:

    “According to EPR [Einstein et al.], an element of reality exists independent of the act of observation. ”

    This is not a conclusion from math. This is not something required by any law of logic. This is just a loudly held assumption (“intuition”, if you like). Bell’s procedure, according to the webpage, was to identify and list such related assumptions and devise a way to test them. Which, in my opinion, is how everybody worthy to be called a scientist must proceed. To test any hypothesis is ultimately a form of testing one’s own presuppositions.

    I promised not to reply, but I thought it worth adding that one of the fascinating “what-ifs” is “what if Einstein was aware of the results of the test of Bell’s inequality?” Most interpret these as showing Einsteins intuitions were false. How would Einstein have responded?

    I agree interpretations matter, but the structures of the world in the interpretation have to be consistent with the structures embedded in the math. Einstein understood the math of his time, but Bell’s inequalities, their subtler successors, and the experimental tests added constraints to possible metaphysical structures. Namely, contrary to Einstein’s intuition, there can be no local hidden variables (there are minority views that choose other constraints, like superdeterminism.)

  12. BruceS:
    I agree interpretations matter, but the structures of the world in the interpretation have to be consistent with the structures embedded in the math.

    I happen to agree completely with this statement, but note that this (=the statement itself and agreement with it) requires a specific interpretation of math. Namely, it requires the assumption that math is identical to the metaphysical structure of reality – something that I indeed happen to hold. However, there’s dispute about it. Some physicists regard math as simply a “useful tool”, nothing more. Many physicists acknowledge the value of mathematical proof when it happens to confirm what they like, but they dismiss it when it appears to confirm what they don’t like, e.g. reality of infinity. Heck, some physicists even dismiss the value of experiments when experiments confirm what they don’t like, e.g. Einstein never retracted his complaint that QM was essentially incomplete, or John Clauser, himself part of the team that experimentally verified quantum entanglement, refuses to “buy” (some theses of) quantum mechanics, as he puts it.

    These are humanly understandable attitudes and, as such, unstoppable. No evidence, no proof, no rational argument can change it.

  13. Erik: . Namely, it requires the assumption that math is identical to the metaphysical structure of reality

    I think what you are saying is what Ladyman calls Ontic Structural Realism, and it is one of the standard views among scientific realists in analytical philosophy (and is favorably regarded by KN).

    But getting into the details of rational versus empirical ways of fleshing that comparison out would be getting into the type of discussion you and KN are having. Which would be a knife-to-a-gunfight situation for me if I tried to get involved.

    On Einstein: even though it seems his intuitions in EPR turned out to be wrong, they have been amazingly fruitful for research. Bohr’s reply to that paper is generally considered to be vague and unhelpful, and has been mainly forgotten.

  14. BruceS: Which would be a knife-to-a-gunfight situation for me if I tried to get involved.

    More like bringing an issue of Nature to a jesuitical dinner at the local rectory.

  15. BruceS: I think what you are saying is what Ladyman calls Ontic Structural Realism, and it is one of the standard views among scientific realists in analytical philosophy (and is favorably regarded by KN).

    FWIW, I think that discussion of structural realism is missing the key early papers on the subject. In particular there’s no mention of Wittgenstein, Newman’s classic criticism of Russell’s statement of the position (http://www.jstor.org/stable/2249202?seq=1#page_scan_tab_contents See also Champagne, http://www.ceeol.com/aspx/issuedetails.aspx?issueid=3dedbeb1-264a-4a7a-a7be-89c5cf90bba1&articleId=e926ce8a-6da5-4002-a3fe-406284db3933of ), or Putnam’s discussion of the Skolem (cats/cherries) version of the same (or similar) difficulty in Reason, Truth and History. It also misses interesting discussions of the position by Schlick. FWIW, not too long ago, I exchanged a couple of emails with Stathos Psillos, an expert on this matter, regarding Schlick’s views on the subject. That correspondence can be found here: https://www.scribd.com/doc/203211505/Psillos-Horn-Correspondence-on-Schlick

  16. BruceS: I think what you are saying is what Ladyman calls Ontic Structural Realism, and it is one of the standard views among scientific realists in analytical philosophy (and is favorably regarded by KN).

    Let me get a bit philosophical about this.

    We have two ways of doing mathematics. We can start with something simple (say, the empty set), and then build up. This gives a discrete combinatorial way of looking at, say, arithmetic.

    Alternatively, we can use geometric methods. We start with the world (the universe of mathematics, if you like) as a whole, and then find ways of dividing it up into parts. This is what gives us the real numbers — start with the idea of an infinite line, and divide it up.

    So we have the logical/combinatorial method of building up. And we have the geometric method of starting with the whole and dividing down.

    The two do not meet. The continuum hypothesis sits in the gap, and has been proved undecidable. We do our best to fit the two together. But that gives us the axiom of choice (disputed by mathematical intuitionists), non-measurable sets and the Banach-Tarski paradox.

    Now look at physics. We have two ways of modeling reality. We have the geometric method of starting with the whole and dividing down. This gives us classical physics. Or we have the discrete combinatorial method of building up. And, it looks to me as if the two do not meet, with quantum physics attempting to bridge the gap. So peculiarities such as Schrödinger’s cat seem analogous to the issues with the continuum hypothesis, the axiom of choice and the Banach-Tarski paradox. (Aside: that’s the gap where the ID proponents should be trying to put their god of the gaps, because we are unlikely to ever fill that gap).

    So if you want to the idea that the math is identical to the metaphysics of reality, are you talking about the discrete math, or the continuous (geometry based) math? Or are you really wanting to put issues such as the Banach-Tarski paradox into metaphysical reality.

    And maybe you can see why I am skeptical as to whether there is such a thing as “metaphysical reality”.

  17. BruceS: I think what you are saying is what Ladyman calls Ontic Structural Realism, and it is one of the standard views among scientific realists in analytical philosophy (and is favorably regarded by KN).

    Not at all, I would say. The section discusses whether ‘continuity in scientific change is of “form or structure”‘, but scientific change would be quite irrelevant to (my) metaphysics. Scientific change is a social phenomenon, not metaphysical. Scientific change has no bearing to the structure of reality, only some relevance to common awareness about the structure of reality, which is a whole different thing.

    I diametrically differ from KN on a number of issues. I am foundationalist/essentialist, he is not. I am rationalist (defined as ‘the view that “regards reason as the chief source and test of knowledge”‘), he is not. I affirm the universals and rational principles (such as the PSR) right up to ontology and I evaluate everything in terms of them (the so-called top-down approach of metaphysics), he does not. Etc.

    Neil Rickert: And I see no basis for that assumption.

    It works. That’s the basis.

  18. Erik: Not at all, I would say. The section discusses whether ‘continuity in scientific change is of “form or structure”‘, but scientific change would be quite irrelevant to (my) metaphysics.

    I understand that you came to your conclusion that metaphysical reality is mathematical structure from a complete different route: that’s what I what trying to get at by referring to rationalist versus empirical.

    I also understand that you may end up with a different mathematical structure than one that is based in mathematical physics. (I should add that I think Ladyman would only include mathematical structure of a “mature-enough” science in the structures he thought were candidates for reality).

    I also understand your point of view about science. And I understand that there are fundamental differences in your view and KN’s on many core philosophical issues.

    But still, I think there is a commonality in the two views in terms of what they say about the structure of reality. Not, to repeat, in what particularly that structure is or how to construct and argue for one particular structure out of all the logical possibilities afforded by math.

    I apologize if suggesting maybe you and KN sort of agreed on something touched a raw nerve!

  19. Neil Rickert: Let me get a bit philosophical about this.

    Except for saying that I believe Ladyman relies on a version of the “no-miracles” argument to conclude that mature science tell us something about reality, I don’t know enough about OSR to reply to this.

  20. Erik: For Aristotle, it makes sense to say “final cause” i.e. that towards which the thing moves is somehow a cause for the thing. In Aristotle’s system there are causes throughout, no complementarity at all. I wouldn’t be surprised if for some A-T philosopher retrocausality (an effect that temporally precedes its cause) sounds plausible or explanatory. Not to me.

    Nor to an A-T. That’s not what is mean by the final cause. It doesn’t involve retro-causality. That’s a myth. It comes about when people conflate the efficient cause with the final cause.

    If we were asked to explain the appearance of your post from which I quoted, one answer might be that you wanted to respond to walto. As the final cause (that for the sake of which), that is logically prior to your banging it out on the keyboard, but the typing on the keyboard (the efficient cause) came before it’s appearance here on TSZ.

  21. BruceS: I think what you are saying is what Ladyman calls Ontic Structural Realism, and it is one of the standard views among scientific realists in analytical philosophy (and is favorably regarded by KN).

    I don’t mean to intervene in this on-going conversation, but I just wanted to note that recently I’ve had some grave doubts about OSR. By all means, please defend it if you wish, but leave me out of it!

  22. Mung: As the final cause (that for the sake of which), that is logically prior to your banging it out on the keyboard, but the typing on the keyboard (the efficient cause) came before it’s appearance here on TSZ.

    Three quick notes:

    (1) it’s not clear to me that intention or goal always precedes action. In some (many?) cases the intention is realized qua intention through action — what Searle (I think) calls “intention-in-action”. One can voluntarily do something (one is not coerced, the action is not a reflex, etc.) and yet also, at the same time, have a realization of “I didn’t know what I trying to do until I tried doing it!”

    (2) it’s not clear to me that the intention-action relationship is a good model of causation generally. This is a standard objection to final causation, but it’s worth raising again. And usually the intention-action relationship is construed by contemporary philosophers as a kind of efficient causation — one has a mental representation of the end-state to be achieved, and the mental state then efficiently causes the physical actions that realize the end-state as represented.

    (3) based on (1) and (2), the intention-action relation is neither necessary nor sufficient for final causation — it is not necessary, because one can understand intentionality in terms of efficient causation; it is not sufficient, because the intention-action relation seems so unique that it is not a promising start for understanding why, for example, salt dissolves in water.

  23. BruceS: Have you thought about putting this stuff on academia.edu?Scridb wants me to pay to download it appears.

    Or do you need an actual educational job to use academia?

    Your paper looks a bit too technical for me right now.

    I have a bunch of stuff on academia.edu but also some links to my scribd page there. I didn’t know they charge you to download stuff at scribd. I have never paid for anything there and have downloaded hundreds of books and papers. It may be that the trick is to upload stuff yourself. If other people look at what you’ve put there, maybe they give you a free pass.

    Anyhow, I can put that correspondence on academia.edu for you, but it’s just a couple of emails.

  24. I wanted to add that you should let me know if I should bother with that, but I got a “You do not have permission to edit this post” message twice, and now the edit link has disappeared.

    (It’s so dastardly that I’d accuse me of wrecking the site so I could lie about something if this weren’t happening to me.)

  25. walto: I didn’t know they charge you to download stuff at scribd.

    I can read it online, but I cannot download it. But maybe all it requires is creating a login at scribd, in order to download.

  26. Kantian Naturalist:
    (2) it’s not clear to me that the intention-action relationship is a good model of causation generally.

    I don’t disagree with you here. It was a quickly obtainable example to show the difference between final causation and efficient causation and to show how the charge that final causation somehow entails that the cause precedes the effect is a misunderstanding of final causation. If I succeeded in that I am pleased.

    If people still think that teleology requires that an effect comes before what it causes I’d like to know why.

    You accept teleology, and you obviously don’t think that to accept final causation requires one to believe that one accept the existence of “an effect that temporally precedes its cause.”

    So I disagree with Erik. And you disagree with Erik. Either that or you and I are speaking completely different languages 😉

    Some day we’ll probably want to explore what you mean by teleology and what I mean by final cause. Final. End. Telos. That for the sake of which.

    I’ve steadfastly declined to say that finality requires any sort of inherent intentionality. As I presently understand things, Aristotle had finality without intentionality and Aquinas agreed that intentionality was not inherent in some things but that they still had to derive their “end directed” natures from some being with intentionality. The Fifth Way.

    Thank you for your comments.

  27. walto, you should just change your online name to dastard. 😉

    ETA: Dastard Lee?

  28. Mung: If people still think that teleology requires that an effect comes before what it causes I’d like to know why.

    It should be obvious. That’s what the expression “final cause” seems to imply, given how we otherwise think of “cause”. That’s why I prefer to use “purpose”. But that has problems, too, for it is used in different ways.

  29. Mung:That’s not what is mean by the final cause. It doesn’t involve retro-causality. That’s a myth. It comes about when people conflate the efficient cause with the final cause.

    I’m not confusing anything. I disagree with the concepts of formal and final causality precisely because I know what they mean. To be clear, I disagree with the naming. That which is being denoted by the words, I take them to be real, but it’s a correspondence or correlation rather, not causation.

    I type because I feel motivated to do so. This is a cause which seems to precede the effect as normal, but my complete motivation is to produce a certain kind of result, which looks like a cause that lies posterior to the effect, and this is where it gets dubious. Moreover, my motivation does not produce the desired result necessarily, so it goes contrary to how cause and effect are linked by the usual definition. Therefore I prefer to call it correlation between ideal and practical reality, a kind falling short from ideal, the way Plato put it.

    ETA: Teleology is real, but it’s not a type of causality. It seems to make much more sense to say that causality is a type of teleology.

  30. Mung: If people still think that teleology requires that an effect comes before what it causes I’d like to know why.

    Neil Rickert: It should be obvious. That’s what the expression “final cause” seems to imply, given how we otherwise think of “cause”.

    You think of causes in terms of efficient causes. I get that. In A-T there are four causes. Material, Formal, Efficient and Final. Why would anyone conflate any or all of these unless it’s simply due to ignorance?

  31. Erik: I’m not confusing anything. I disagree with the concepts of formal and final causality precisely because I know what they mean. To be clear, I disagree with the naming. That which is being denoted by the words, I take them to be real, but it’s a correspondence or correlation rather, not causation.

    If you know what final causality means you would be surprised if for some A-T philosopher retrocausality (an effect that temporally precedes its cause) sounds plausible or explanatory.

  32. Mung: In A-T there are four causes.

    Yes, but that terminology is outdated. People today are more likely to think in terms of emotions, psychological drives, goal directly behavior, etc. The expression “final cause” should be dropped, except for its historic interest.

  33. The terminology isn’t outdated Neil. To say that it’s outdated is to say that it’s been replaced, and that is simply not the case.

Leave a Reply