22 May 2010

The Ideal of Sufficient Reason

Leibniz is one of my favorite philosophers, tempering excessive philosophical ambition with admirable moral sensitivity. His sensitive side stems, in part, from his commitment to what is known as the praejudicium Leibnitii: "If I were wrong, I would like to err in favor of other persons, and not be mistaken to their disadvantage." His ambition stems, in part, from his mad and unrelenting commitment to the principle of sufficient reason (PSR): "no fact can be real or existing and no statement true without a sufficient reason for its being so and not otherwise." This is a mischievous principle. Parmenides used it to argue that nothing changes; Aquinas, to argue that God exists; Spinoza, to argue that nothing is contingent. No tenet of common sense seems able to escape PSR's withering gaze.

One might have doubts about whether PSR is true, either because it seems to conflict so drastically with common sense or because there seem to be some facts that are just brute (think quantum indeterminacies). What interests me, however, is not so much whether PSR is true as the ideal of which PSR seems to be a manifestation. I've been thinking about the ideal of sufficient reason, pursuit of which demands that reason until there can be no more reason. This is a tempting ideal. Its pursuit fuels the regress argument from epistemology, requiring any justified belief to have its justification grounded in some reason, which in turn is grounded in some further reason, and so on. If there are no endless regresses of justification-giving reasons, pursuing the ideal vindicates epistemological foundationalism (or skepticism!).

Metaphysics, too, is no stranger to the ideal. Reality that accords with the ideal of sufficient reason requires any existent to have its reality grounded in some reason, which in turn is grounded in some further reason, and so on. If there are no endless regresses of reality-giving reasons, pursuing the ideal vindicates either metaphysical foundationalism (where the basic entities have no deeper reason) or nihilism (where there is no foundation and therefore no reality).

Pursuing the ideal of sufficient reason produces structures (epistemological, metaphysical, or otherwise) that are asymmetric, because what is a reason for something cannot have that something as its reason (whence the ban upon "circular" reasoning).

Given the ideal's prevalence and attraction, I've been wondering whether there are any alternative ideals one might pursue in the philosophical quest for understanding the world's structures (be they metaphysical, epistemological, or something else). Philosophers from the Chinese tradition seem to pursue a different ideal, the ideal of the round. Pursuit of this ideal demands inclusion until there can be no more inclusion. An ideally inclusive structure differs from an ideally rational structure, because the structure of inclusion is not asymmetric. For example, if a baseball team is ideally round, the team includes each of its members and each member includes the team; and if a system of justified beliefs is ideally round, each belief includes every other as part of its justification.

I'm currently working on tracing the consequences commitment to the ideal of the round has for metaphysics. My hunch is that a reality that accords with the ideal of the round has a holistic, rather than a linear, structure, offering a satisfactory alternative to metaphysical foundationalism. I'll report back with progress on how this hunch is playing out, as well as more information about the ideal of the round.

For now, though, I'm wondering: Are there examples, from politics, ethics, aesthetics, or wherever, in which the ideal of sufficient reason manifests itself? Well-known arguments with structures analogous to the epistemological regress argument?


  1. Andrew Cling22 May, 2010 18:22

    The ideal of the round seems to have an affinity with ideals of coherence in epistemology and elsewhere. A system is coherent, roughly speaking, if it is consistent and properly integrated. One idea is that a system is coherent (=round?) if, and only if, every element in the system is implied by some other element in the system AND the system is consistent.

    Goodman invented the idea that Rawls later called "reflective equilibrium," the state in which the general moral principles to which one subscribes imply the particular moral judgments that one finds plausible, and vice versa. That's a political application of an epistemic value, I think.

    I cannot think, offhand, of a political ideal per se that is an analogue of "the round." (Though, on a certain reading, Rush Limbaugh approximates the ideal.)

  2. andrew cling23 May, 2010 07:55

    John Post offered a general argument against all principles of sufficient reason. As I recall, it went like this. PSR, as Jones notes, requires an endless regress of reasons. No such regress can be circular, that is, contain a reason both sooner and later. So any regress of reasons must be infinite. But--and this is the key move--we can construct an infinite sequence of reasons for any proposition such that any member would explain its predecessor. So if PSR were true, we would have an explanation for anything. Therefore PSR is not true.

    At least that's how I think it goes. The actual details are in the section called something like "Infinite Parades of Explanation" in Post's fascinating book *Faces of Existence.* (I was fortunate to be Post's research assistant while he was writing the book. I learned lots of philosophy while learning a lot about how to write philosophy.)

    I just remembered that Post has that section posted here:


  3. andrew cling23 May, 2010 08:02

    Upon further reflection, I think Rawls' use of RE is an ideal of the sort you want. Once our system of moral principles and judgments is fully integrated by the elements' being mutually supporting there is no further question of justification. I think that this kind of coherence is not supposed to be a sign of something beyond the system--truth, for example--but is itself a kind of ideal goal for the principles of a moral theory.

  4. Thanks, Andy. I'll have to look at Rawls' RE as a potential Western instance of pursuing the ideal of the round, and also see what kinds of objections people level against RE.

    Post's arguments are relevant and helpful. John Worrall (1996) makes a sympathetic point: "the worst of all possible worlds is one in which, by insisting that some feature of the universe cannot just be accepted as 'brute fact,' we cover up our inability to achieve any deeper, testable description in some sort of pseudo-explanation--appealing without any independent warrant to alleged a priori considerations or to designers, creators and the rest."

    I think pursuing the ideal of sufficient reason is consistent with accepting Post's arguments against versions of PSR, because one can always weaken PSR to something like "for every X, if X can have a reason, then X does have a reason; and if X cannot have a reason, X is brute."

    This weak PSR seems to appear in non-theistic arguments for foundationalism in metaphysics. Here's Jonathan Schaffer (2010): "There must be a ground [foundation] of being. If one thing exists only in virtue of another, then there must be something from which the reality of the derivative entities ultimately derives."

    I'm focusing not so much on the soundness of these arguments as on the mindset that gets them going to begin with -- a mindset for which understanding requires reasons and reasons are asymmetric, so that understanding requires hierarchical structures (ending either not at all, in "necessary" entities, or in "brute" entities.) I'm after examples of this mindset that are from neither metaphysics nor epistemology. Then, from there, the project is to develop a competitor to reason-based structures in metaphysics, based upon an alternative ideal.

  5. 3 preliminary things, before I answer the question you ask about examples of PSR in other subfields of philosophy. Andy beat me to the punch -- I was also going to discuss Rawls. My take on Rawls is a bit different than Andy's.

    1. I know the passage you're thinking of in Parmenides, but I'm not really sure it's PSR that drives his argument for monism; I think it's rather a particular conception of predication. To be something requires that it be all and only that thing, which simply denies the possibility of it ever having been anything else. (This is the "predicational monism" reading that belongs to Curd and that I think is right.)

    2. I sometimes think that PSR can be quite closely connected to common sense: I occasionally explain it to my students with reference to our general desire to find causes, reasons and explanations for events and behaviors.

    3. I think you're right that PSR often drives both epistemological and metaphysical quests for foundations in philosophy; as you put it in a later post -- that there is some kind of hierarchical and asymmetrically principles or prejudices at work in this way of thinking. In this respect, though, PSR has always seemed to be a victim of itself. I mean, does PSR have a sufficient reason? But I like the ideal of the round because it seems it could simply include itself in its own "justification." In fact, as I read your remarks about it, it would have to include itself.

    As to Rawls. I find him to be an interesting lesson here. In the early work, he endorses reflective equilibrium, but this is in direct conflict with his famous claim to be thinking of justice sub specie aeternitatis. It can't be the case that we are justifying the principles of justice by appeal to a balance of our historically grounded considerations and that we are also discovering what justice really is for all time and independent of our historical considerations.

    In the later work, Rawls drops the ideal of justice sub specie aeternitatis. Now, he speaks of trying to justify the principles of justice politically. This means that the justification appeals to no single moral doctrine or ideal, but rather it appeals to what all parties in the original position would find to be in their best interest regardless of the various party's different conceptions of the good. Rawls acknowledges that this makes the entire enterprise non-foundational, and that it only provides a justification of liberalism for persons who located at our place and time in human history.

    But he cheats. In order to make sure that parties in the original position actually work out, he must make assumptions about the reasonability and rationality of persons that cannot be justified by reflective equilibrium, because they are necessary to put persons in a state of reflective equilibrium.

    In this respect, Rawls is not that different from the entire long history of political thought since Plato. This thought has almost always tired to anchor considerations about the best form of political life by appeal to ultimate standards. But here I'm getting into a topic that I should perhaps use for my own post….

  6. One other thing: if it's the "mindset" you're after, you might want to look at philosophers who attack that very "mindset" -- Wittgenstein comes to mind, but I think Nietzsche's actually better here. Let me know if you're interested in tackling that rather big mountain; I can give you some guidance.