The impossibility of Naturalist brute facts

1.) Scientific explanations require laws and initial conditions. More generally, some some intelligible process arising from a given fact.

2.) Take the given fact. Facts are given differently from axioms. Both the fact and the axiom enter explanation as a proposition, but  the axiomatic proposition is one where we see the unity of subject and predicate while we do not see this for the fact.

3.) Either an explanation of the proposition is logically possible or not. If not, we must either have an axiom or what nowadays gets called the “brute fact”. Note that this is not a “brute fact” as first coined by Anscombe to describe more or less basic levels of explanation, but as it nowadays gets used to forestall theistic arguments and/or to establish some Naturalisms.

4.) Brute facts only exist if there is some proposition that (a) cannot logically be the conclusion of some line of reasoning and (b) is not axiomatic. In other words, there must be some proposition whose unity cannot be explained either by its own terms or any other. The “cannot” there needs to be very strong: there would be a logical impossibility in the unity of the proposition arising from its own terms or by those terms in unity with others, since allowing for the first logical possibility would make the proposition an axiom (and so not a brute fact) and the latter would be to deny it was a brute fact altogether.

5.) But if it is logically impossible to unify terms either by themselves or another term, we have no more reason to affirm predicate than to deny it. I stress the logical impossibility because we see all sorts of facts around us that are neither axiomatic nor for which we see any obvious connection. It’s presently a fact that  “My couch is 15 feet from the overgrown raspberry garden”. There is nothing axiomatic in this, and I’m a bit puzzled at the thought of what I would use to conclude that (a ruler maybe?) Now if I wanted this to be a brute fact I’d have to keep insisting it was not an axiom (which would be easy) while making the additional claim that there is a logical impossibility in its being a conclusion of some previous propositions. But that’s (a) nuts and (b) an argument that it could not be a fact at all.

Brute facts and Euclideanism

(N.B. What follows is a sort of cosmological argument that identifies an incoherence in Naturalism as a rejection of the supernatural.)

As an explanation, a brute fact has two components (a) it is what it is and (b) it should be accepted. It is a justified tautology, the justification for which is that physical laws only explain the actual world in conjunction with initial conditions, and these conditions are ultimately historical accidents.

The appeal to brute fact thus rests on a hylomorphic understanding of nature where laws : initial conditions :: form : matter. Since the actual world is a composite of intelligible forms and unintelligible matter it contains features about which one cannot say anything more than that “they are what they are”, where the point of the tautology is that the intelligibility of our explanation, and of nature itself, doesn’t reach to one of its essential parts, sc. matter/ initial conditions.

All sides therefore allow some unintelligibility in nature, but we disagree whether all explanations of nature are naturalist. In one sense, of course, it is self-evident that they must be, in exactly the same way that all explanations of Euclidean shapes must be Euclidean, i.e. so as long as we are building up Euclidean explanations we obviously have to stay within Euclid’s foundational axioms, definitions and postulates.

But it’s just this analogy between Euclideanism and Naturalism that rules out Naturalism in the sense of the rejection of supernaturalism. Remember that Euclideanism was superseded precisely at the moment we recognized that the fifth postulate was only a postulate and not axiomatic nor provable, and the presence of matter or initial conditions proves Naturalistic explanations are likewise only postulated and not axiomatic or provable.

In other words, the fifth postulate is a brute fact of Euclidianism. It is neither axiomatic nor provable, but simply (a) is what it is and (b) needs to be accepted (if one is going to give Euclidean explanations). Nevertheless, in geometry we draw a correct conclusion from this that we fail to draw so far as we are Naturalists that reject the supernatural. The demonstration of the brute facticity of nature is simultaneously the demonstration of a domain of natures of which the physical world is only a specialized instance that, like Euclideanism, we find easier to understand.

Notice that the brute facticity of both the fifth postulate and matter/ initial conditions prove the necessity of something beyond Euclid and nature for the same reason, namely that they prove something is possible which can be known to be necessary as soon as it is known to be possible. If we  distinguish possible from actual existence in either of these then for them to actually exist would be in some way or another a brute fact, in which case the supposed “thing other than a brute fact” that was just shown to be possible would not be other than a brute fact.

Whether brute facts are possible

Hume gives what is still the best account of what it would be to reduce phenomena to brute facts:

[The] ultimate springs and principles are totally shut up from human curiosity and enquiry. Elasticity, gravity, cohesion of parts, communication of motion by impulse; these are probably the ultimate causes and principles which we shall ever discover in nature; and we may esteem ourselves sufficiently happy, if, by accurate enquiry and reasoning, we can trace up the particular phenomena to, or near to, these general principles.

But success at reducing material activity to, say, gravitation is evidence that the statement “matter is gravitational” involves the same sort of predication as “squares are quadrilateral”, and the more success we have at the reduction the more confident we become. Successful reduction therefore drives out the idea that the explanans is simply a contingent, given brute fact and instead is taken as evidence that it involves per se predication.

Hume wants to divide reductive success from discovery of something ultimate, but the division seems unreasonable: Reductive success consists in finding the ultimate. The sense that we really can go no further in analysis is the grasping of an ultimate. We don’t take our failure to find a further reductive explanans as evidence that there is one out there we can’t find.

Sure, we can be mistaken about what is ultimate. Per se predicates are very difficult to find and very prone to refutation by experience. But we already knew that from Aristotle.

The basic issues are (a) whether sense experience ever gives insight or whether it is simply homogeneous repetition, and (b) whether the defeasibility of insight requires that insight never actually occurs. Both claims strike me as untrue to the experience of repeated events, and the second seems to conflate a fallible activity with one that cannot happen.

Most Empiricist or Kantian epistemologies suffer from just this sort of overlooking the reality of sensation giving rise to insight. Even Aristotle seems to do this – though he was probably the most eel-balanced empiricist who ever lived, when asked to explain insight he mumbled an enigmatic metaphor about soldiers fleeing from a battle. The difficulties in accounting for insight are very real – it’s not clear that one can define insight in a positive way, and there is certainly no formal-logical account of the process. The temptation to brush it aside altogether is unavoidable. That said, even the sharpest critique of the reality of insight – say, the grue-bleen problem- is still proposed as an insight into cognition.

The theory of recollection is still probably the most rigorous theory of insight on offer.

If God’s a brute fact, why not the universe?

Objection: If you’re going to take God as a brute fact, why not just take the universe as one?

I hate the term “brute facts”, but why bother fighting over names? Let the term mean “a proposition for which there is no explanation, even in principle.”

NB it’s crucial to note that facts are propositional, by which I mean nothing more profound than that single terms like “Unicorn” or “first” are not facts. Neither is “God” or the “universe”. The point of the objection has to be about predicating existence, as in “If you are going to allow for no explanation for why there is a God, why not allow for there being no explanation for why the universe exists?”

The first response is that this is not how cosmological arguments work. We don’t start off with a single “Brute fact” post-it note that we can stick to any existence claim we want. We take explanations where we find them, and if we have one for the universe (i.e. the totality of space-time or motion or whatever) then we take it.

But given what brute facts are, some examples of them are  “Socrates is Socrates” or “man is human” or “A brute fact is a proposition for which…” at least when taken unqualifiedly and not in some exotic sense like “Why was that individual (Socrates) named what he was?” or “what is the principal of identity when applied to the teacher of Plato?” Notice that these exotic senses end up proving the rule, since the proposition only needs an explanation when we find a way in which the predicate is not immediate to the subject.

If this is right, then we have a very good reason to take “God exists” as a brute fact in a way the universe cannot be, since a brute fact is something about which we could know that it exists simply by knowing what one was. But what we mean by “the universe” (the solar system, galaxies, etc.) is not something that tells us that there is such a thing. We know that there is a universe by looking at it, not by defining it. We know what it is only after we see that there is something there.

God is not like this at all. We know only that there is such a thing, but we can’t even take the first step to knowing what God is since we cannot place him essentially within a genus of things that are the same as he. We know God only by his effects, in such a way as to know that there is nothing in his essence making him homogenous with the effects themselves.

Brute facts and knowing that

What we call a brute fact is what Aristotle would call knowledge that something is true without knowing why it is true.  Ironically, while brute facts occur most often in attempts to halt cosmological arguments the conclusion of the cosmological argument is itself a case of only knowing that something is true.

Aha! So as long as we posit a BF in either case, why not take the universe as one and treat God as a superfluous addition? But this is to be tricked by the mere repeating of a word: if the universe is a BF it is because there is only one of them and so we cannot run multiple experiments on it, or because it is (probably) fallacious to extrapolate from truths about parts of the universe to truths about the whole (cf. Lee Smolin’s Time Reborn). But God is a BF because to know what God is belongs only to the blessed.

More simply, the B-facticity of the universe is given to sensation while the B-facticity of God is not. One wonders how much of the talk about “brute facts” involves a failure to recognize the false assumption that the first sort of facticity is the only kind.


Brute facts as pointers to cosmological arguments

Science doesn’t reduce everything to brute facts: diabetes does not reduce to an autoimmune attack on T-cells from the ATP/P2X7R pathway as a brute fact but because that’s what diabetes is, likewise with the reduction of malaria to a parasite. There are other times when things are reduced to brute facts, like when the motion of one part of a system is explained by the motion of another part, or the time of one clock is coordinated with the time of another. There are good reasons for this sort of reduction – namely the unification of many phenomena to some primary instance – but they are to be divided from the sort of explanation that actually tells us what something is. But if explanations should reduce to seeing what something is, and brute facts are sometimes appropriate ultimate explanations of physical things, then Brute facts are appropriate explanations only when the what-something-is is non-physical. We take some mover as a brute fact because the First Mover transcends the physical; we take some cause in a physical system as a brute fact because the First Cause likewise is supernatural; we take the conserved (and therefore necessary) existence of matter, momentum, energy, etc. as brute facts because what is necessary in itself is not and cannot be a physical being. Such facts, therefore, are the opposite of an alternative to a cosmological argument. They indicate the physical being proximate to divine causality, in exactly the same way as Aristotle’s primum mobile is the physical being proximate to the First Unmoved Mover.

Brute facts vs. explanations

Barry Miller critiques the claim that per se causal series are necessarily finite, saying that every explanation is relative to something taken as a brute fact, and that there is no reason for such a series to terminate, even if it is a per se causal one. He’s right in the sense that we might, for all we know, keep finding per se movers forever, or perhaps even hit on a proof that we must keep finding such things forever. Perhaps the body moves so far as it has a force, but this force moves so far as it is energy, and the energy moves so far as it is mass-energy, and so on forever, with higher and higher levels of physical unification and transcendence. Leibniz argues either for this or a variant of it. Miller still argues that some causal series terminate, and do so in a way that makes a cosmological argument, but he denies that a per se causal series is necessarily one of them.

But Miller is muddling unexplained facts with inexplicable ones, since any explanation might be said to trace back to the first, but only the second are brute facts. If all “brute” fact means is one which we are not giving an explanation for, then we can make any fact such by simply remaining silent or dropping dead after we utter it.

But my main objection is that explanations don’t trace back to brute facts at all, since to say that they do commits the fallacy of the accident. No one seeks to rest an explanation on the unexplained or inexpicable, even if the thing he is targeting will necessarily have such a property. Saying that the one explaining is looking for the unexplained is like saying that a man who wants a wife is looking for a non-abstraction, or someone looking to make chocolate cake is looking for non-mustard. No one looks for the brutish or non-explained when he makes explanations, but for what is axios or worthy of serving as a starting point. You might push an explanation only so far, but you don’t choose to stop when you hit brutishness but when you hit on something axiomatic in a given domain, or at least for something that doesn’t assume the existence of what you’re trying to explain.

It’s this latter sense of explanation that is at work in the causal series that STA uses in his cosmological arguments. St. Thomas – like everyone giving explanations – isn’t targeting a brute fact as an explanans but something that can account for some fact of experience without assuming it.

But even if all this talk of brute facts wasn’t the fallacy of the accident, there is the broader problem whether any such things exist at all. in practice, they seem to work as justifying a stoppage of thought – hey, if you get to yell “brute fact!” at some point, why not yell it at nature and not God? But this is a ridiculous account of explanation, and one that ends up at odds with itself, since no one sees the point of explaining as to find something rationally impenetrable to rest logical consequences on. If it comes to that, there’s no reason to attempt explanations in the first place. After all, if what your’e targeting is what lacks a rational account, you’ve got that before you even bother giving an account at all. If the point of explaining was to get to brute facts, what you start with is as good as anything else.

Appeal to brute facts in Medieval Cosmological arguments

The critique of cosmological arguments by appeal to brute facts (i.e. we must have something unexplained, so why not the universe?) fails to see that, for these sorts of arguments an explanation does not terminate by finding of fact of any kind, or even with a proposition, but with a subject about which the feature one was trying to explain can be said per se and first of all, or a predicate which is said of a known subject in this way. The search for who dropped the atomic bomb does not terminate in a fact, but with Truman, about whom that predicate is said first of all; the search for what heat is terminated not in a fact but in fire (for Medievals) or molecular motion (for us) which is said of the subject per se and first. Even if we took “Truman dropped the bomb” as a fact, it functions as an explanation so far as we discover an identity between the thing we were looking for (bomb dropper) and the thing we find (Truman).  It’s not just that the predicate and subject are convertible, but convertible in a certain way that we understand as being simply what the thing is, and the cause of all those things that have that feature in a secondary way (the way that, say, Tibbits dropped the bomb or fire is hot).

Brute Facts

For Anscombe, a brute fact was opposed to a moral fact. Moral facts required certain dispositions of the perceiving subject (like maturity or good education) in order to be perceived whereas brute facts did not. Discerning that the sky is blue certainly seems different from discerning that a loving father’s discipline is good, though both are certainly facts, and one must mark off this distinction by fixing some sort of adjective. I’m not crazy about the adjective “brute”, but it will have to do.

Anscombe’s division has its limits. Aristotle (and all who followed him) makes a great deal out of the fact that sick persons cannot discern even what are called “brute facts” by the above definition. Men in fevers think spicy foods are bland, sweet foods are bitter, pungent things are odorless, etc. Indeed, Aristotle continually appeals to this as an analogy to explain the discernment of moral facts, that is, he argues that moral facts require a right disposition just as brute facts do. But again, this only shows that Anscombe’s division has its limits, which she no doubt admits herself.

Somewhere along the way, however, brute facts became unmoored from their relation to moral facts. As it stands now, a “brute fact” is supposed to be something that has utterly no explanation. It is not clear that there are such things, or even if they are possible. The first difficulty is that this account distorts what we first suppose a brute fact would have to be. If brute facts are “unexplainable facts” then statements like “the sky is blue” or “two plus two is four” cease to be brute facts, since one can give an explanation for why both are so. Another difficulty is that it is doubtful we have any criterion to discern brute fact, since no one has a very good sense of how to divide what can be explained from what can’t be. It might mean something to speak of facts that have no  explanation for the moment or on this or that hypothesis or school, but to claim that something has no possible explanation simply speaking is an odd claim – and at any rate if we could ever discern such a thing we would call it a mystery and not a brute fact. But doesn’t it make more sense to say that a mystery is the opposite of a brute fact?

Another problem with the account is that “explanation” means far more than one thing. It is not the same thing to prove (or explain) that something is so and to prove why it is so, as any scientist would tell you. That said, both require different sorts of explanations and proofs – sometimes even very elaborate ones.  If we take brute facts simply (that is, without some limiting qualification), it follows we can’t even establish that they are so. But if this is the case, then how in the world are they facts?

But perhaps a brute fact is supposed to be that which is not explained by something else, but itself explains other things. In this case, a brute fact is identical with a first principle or universal law or an axiom. But, here again, this is just not what we call “a fact”. When we speak about the “facts of a biologist” or the “facts of physicist”, or “the facts of an experiment” we aren’t talking about their fundamental axioms and laws. The facts of the science are things known even prior to the science itself while the fundamental axioms are not. Who thinks that the goal of science is to explain facts by facts (brute or not). If this were true, how is there any room for theory? Aren’t facts exactly the things we are supposed to explain?

And how would we distinguish such facts from the tautological or sheerly happenstance, since none of these things has an explanation either? There is no explanation for either a.) why did the earthquake happen while I stepped in the bathtub? or b.) Why are butterflies butterflies? But one can explain the lack of explanation- on the one hand there is mere chance conjunction of things, on the other hand there is only logical reflexive duplication. Neither have the same explanation for why they have no explanation. Both are at least mildly ridiculous. So is this what a brute fact comes to – the ridiculous?

Why call merely inexplicable facts “brute”?

Calling a fact brute is a curious mix of the obvious and utterly obscure. In the philosophical discussions of the last few years, it means just “an inexplicable fact”, and so it does no more than negate the possibility of an explanation, or at least to negate any explanation by extrinsic causes (like agent causes). The word “brute” even frequently becomes the focus of attention, and so there are actually articles that parse degrees of brutishness or describe facts as relatively brute. But why focus on such a seemingly arbitrary word when “inexplicable” would do just as well?

Brute seems more like an idol or a totem. The brute is what is irrational and overpowering – the bête noire that lurks in the background. Brute is meant to give a fundamental character to things as just there but not as manifest or self evident. The mind sees nothing but is simply bound or overcome by a force that makes no sense and is incapable of doing so.

Brute facts thus seem more like a mythology, or at least a total view of explanation. The idea is that explanation itself at some point breaks down, becomes ridiculous, or no longer functions while at the same time there is some given that must be held to. There is, however, another crucial element – for the failure of explanation cannot be a mystery, that is, a sublime or superintelligible reality. “Bruteness” indicates a harshness or violence against the intellect, a frustration in the reaching for a goal while the solemn and lofty character of mystery is a direct negation of this sort of frustration.

And so we hang onto brute because it alone conveys the dark mythology of fundamental frustration.

« Older entries