Wednesday, February 18, 2009

Truthmaker Maximalism Without Remorses?

Let truthmaker maximalism be the view that every true truthbearer has a truthmaker and let me put aside questions about the nature of both truthmakers and truthbearers and assume, for the purposes of this post, that truthbearers are propositions and that truthmakers are ordinarily facts. Now, according to naive truthmaker maximalism (NTM), the proposition that p is true if and only if it is a fact that p. For all its naivete, NTM seems to be a nice, simple view of truthmaking. However, most truthmaker maximalists seem to be unwilling to embrace it, mostly because they seem to feel uneasy about admitting in their ontology certain kinds of facts as the truthmakers for certain uncontroversially true propositions. Two standard examples of facts truthmaker maximalists seem to be queasy about are negative and general facts and quite a bit of ink has been spilled in an effort to explain how (some) negative and general propositions can be true in the absence of negative and general facts.

The problem is that all these attempts sacrifice much of the simple charm of NTM to a worry I can't really understand because I still don't get what's so wrong with, say, negative and general facts. Yes, unlike Quine, I don't have a love for desert landscapes but not many metaphysicians seem to love them these days and, of course, I would have problems with someone thinking that negative or general facts are fundamental facts, but I don't see any problem with the view that such facts supervene on more fundamental, more respectable facts. In fact, as far as I can see, if you think that something makes true the propositon , that something better be the fact that Socrates is not a fool. So, if you and I are both remorseless truthmaker maximalists and we both agree on which propositions are true and which are false, we should also agree on what facts are there even though we might disagree as to which of those facts are fundamental and which are not. In general, if you accept that a propositions <p> is capable of being true or false, you should also accept that there is a fact of the matter as to whether it is true or false and that this is the existence or non-existence of the fact that <p> not whether or not its fundamentality.

Maybe I'm wrong in assuming that NMT has become a minority view among truthmaker maximalist (please let me know if you think my perception of state of the play is somehow distorted), but, if I'm not, can someone please explain me what's so wrong about superveneint negative and general facts that makes it preferrable to abandon a view of truthmaking as nice and simple as NMT rather than admitting them in one's ontology?


  1. If "the fact that p" is understood in an ontologically innocent way—the way in which we can talk about them even though they are no addition of being—I see nothing wrong with NMT.

    But talk of supervenience seems unhelpful. After all, the xs supervene on the ys if the xs just are the ys. They also supervene if there is a common cause. And they supervene if...

    Saying that negative and general facts supervene, it seems to me, is just a way of naming the problem, not solving it.

    For myself I see no reason to think negative and general facts aren't fundamental.

  2. I don't think anything non-fundamental can do truthmaking work, but that's because of my broaded meta-ontological views. Still, I'd like to know *what* the fundamental facts are upon which, say, the totality fact supervenes on your view. As I see it, the problem is exactly that we can't point to any such facts.

    To my mind, there are two big problems with totality facts. The first is the weird necessary connections: how can it be that this thing simply *can't* exist with a unicorn? The second is that the totality properties look peculiar: they're extrinsic, and the only point in admitting them seems to be to solve truthmaking problems.

  3. Jon,

    The fact that p was not meant to be ontologically innocent. You can think of it as an Armstrong-style state of affairs if you wish. (I just find the fact terminology more in line with ordinary language and more flexible and avoids the ambiguity with Plantinga-style SoAs).

    I was using supervenience somewhat loosely (as it seems to be often used) to express asymmetric supervenience (The A-facts supervene on the B-facts but the B-facts do not supervene on the A-facts) and I agree that talk of supervenience is still very abstract but I don't think it's just a way of naming the problem. In any case, as I mention below, one could think of the supervenience relation as being token-identity in some cases. The fact that Socrates is not a fool may be token identical with the fact that Socrates is wise.

    In any case, if you don't have any problem with negative and general facts being fundamental, then I guess you don't have a problem with them existing, so probably you don't any good reason to think they don't either.


    I dislike totality facts as much as you do (in fact I don't even understand how there could be any such facts), but I don't see any good reasons to think that remorseless truthmaker maximalists (RMTs) need such facts. As far as I can see, RMTs could maintain for example that the fact that all apples are red is simply token-identical with the fact that this apple is red, that apple is red and so on for all apples (or, if you prefer, with the fact that is the sum of those particular facts) (even if, of course, it is not type-identical with the sum of those facts facts because if there was one more red apple there would be a different sum fact but all apples would still be red). In other words, once you have all the particular positive facts in place, I don't see the need to postulate the further "fact" that there are no other apples--of course there are no other apples, if there were other apples, wouldn't there be further facts about the colour of each of them?

  4. Some believe that a truthmaker must necessitate the truth it makes true. If so, negative facts are not token-identical with positive facts. Then the problem is what you said it is: how to understand such sui generis facts.

    True, one can reject the idea that truthmakers must necessitate their truths. But I suspect it is adherence to this idea that leads many to reject positions like your NTM.

  5. I just came to say what Gonzalo said! The reason I wouldn't accept your position is that I think truthmaker necessitarianism is on a stronger footing than maximalism (despite the best argument for it begging the question!). I don't want the truthmaker for 'all the apples are red' to be able to coexist with a green apple. Any objection to such a necessary connection is, I think, a reason to deny that there is such a truthmaker rather than a reason to abandon maximalism.

  6. I was just wondering how are you guys understanding 'fundamental' in this context. In particular, would admittance of "non-minimal truthmakers" be compatible with the contention that only the fundamental can truthmake?

    Nice blog btw!!

  7. Dan - for me at least, what's fundamental is what really exists: what can be said to exist using the most natural quantifier. I can admit that there are non-minimal truthmakers and that only the fundamental can truthmake.

  8. Before the discussion gets too side-tracked by my contentious suggestion at 4:01 pm, let me note that that suggestion is only ONE of the things remorseless truthmaker maximalists could say and I find it encouraging for NTM that nobody so far seem to disagree with the more genral claim that general and negative facts supervene on particular positive facts (while the latter do not supervene on any of the former). The only objection I have heard so far is Jon's objection that this is somehow trivial. which I don't think it's the case for the following reasons.
    First, someone who believes that the A facts supervene on the B facts is someone who believes that A facts exist, while I take it that a number of truthmaker maximalists don't believe that negative and general facts exist.
    Second, if the A facts supervene on the B facts (but nor viceversa), if then the existence of certain B facts necessitates (is a sufficient condition for) the existence of certain A facts (those A facts cannot fail to obtain if those B facts obtain). If such a necessitation relation was somehow trivial, I do not see why it would be not trivial when it obtains between the existence of truthmakers and the truth of the corresponding truthbearers.


    I'm not sure I follow your argument, but first, accoridng to the general view I was suggesting, the truthmaker for Socrates is not a fool is the (negative) fact that Socrates is not a fool. Second, you seem to think that that negative fact cannot be token identical with the postive fact that Socrates is wise because the latter does not necessitate the truth of the proposition that Socrates is not a fool but as far as I can see it does. How can it be possible that Socrates is wise and nevertheless it is false that Socrates is not a fool or to pick a less contentious example how is it possible for this apple to be red all over and for it not to be true that it is not blue? Am I missing your point?

    Ross at 12:36pm (yes I'm addressing one of Ross's temporal parts),

    Obviously, the truthmaker for the proposition that all apples are red cannot coexist with the existence of a green apple because the former is the fact that all apples are red and that fact cannot exist if there is a green apple. You seem to be thinking that, if that general fact is identical with a certain conjunctive fact and the latter can coexist with a green apple, the former can as well but (if the naive truthmaker maximalist go down the token identiity route) this is not the case for those tokens of the conjunctive fact that are identical to tokens of the fact that all apples are red!

    Dan and Ross at 4:44pm,

    I don't think that what is less fundamental exists any less really than what is more (or even most) fundamental--I believe that tigers and chairs exist as much as the fundamental particles that make them up. As I see it, the question of fundamentality is a question of what depends on what not of what is more real than what. (So probably this is also why I disagree with Ross at 1:12pm)

  9. Gabriele,

    I didn't mean to suggest triviality. Rather, I meant the following: *Either* you mean it in an ontologically innocent way, a way consistent with there not actually being any fact that p, *or* you do but you haven't given a theory. You didn't mean it in the innocent way, so let's take the other way. You're original proposal was just supervenience. But that's just a correlation; it doesn't tell us anything else about the nature of the relation between the two.

    For example, it might be that the negative facts are fully distinct and ontologically real, and yet of brute metaphysical necessity are correlated in this surprising way with other, fundamental, ontologically real facts. That's consistent with supervenience.

    So is this: The negative facts *just are* the more fundamental facts.

    So is this: The negative facts are instantaneously caused by the more fundamental facts (which causation is necessary).

    And so on.

    Then, you offered one version (4:01), according to which they are token identical but type distinct. I confess to not really finding the token-identity/type-distinct (multiple realization?) move too plausible (in this or any other context). That's not an argument, of course.

  10. So now I'm confused: I took Gonzalo and myself to be *exactly* objecting to your claim that the general and negative facts supervene on the particular positive facts. We offered your maximalist an 'out', though: denying necessitarianism (this is the position I took in an early paper of mine). according to this position, what you claim as an obvious truth (in response to my earlier temporal part) is denied: the truthmaker for 'all apples are red' could indeed coexist with a green apple. *In that world* it is not a truthmaker for 'all apples are red' but, since we're denying necessitarianism, that doesn't mean it's not such a truthmaker in the actual world.

    You want to have your cake and eat it: to hold that the truthmaker for 'all apples are red' both necessitates that all apples are red and is token identical to some conjunctive positive fact. I simply can't understand how any position like that could be true, given that all the positive facts could be as they actually are and an extra green apple exist.

  11. Jon,

    I don't think that someone can be both a superveneintists and an eliminativists about a certain sort of facts. In any case, the view that I was suggesting in the post was meant to be a view according to which the facts in question exist and supervene on some more fundamental facts. In my comment at 4:01 I said that I agreed that this general view can be filled in different ways (and I suggested one way the supervenience could e cashed out could be done and I sympathize with your worries about it but since there are people that are happy to hold an analogous view in philosophy of mind, I don`t see why one couldn`t do so in this context.)


    (Let me set aside the token identity issue for the moment, which is proving to be a red herring) I don`t see what your objetionto the view that negative and general facts supervene on the particular positive facts was. As far as I understand it, all that the latter view implies is that two worlds cannot differ with respect to the general or negative fact without also differing with respect to some particular positive fact and, as far as I can see, nothing in what you said seems to suggest this not to be the case.
    In particular, suppose that at the actual wold all apples are red, the fact that all the positive particular facts at the actual world *could* coexist with the fact that there is a green apple in no way seems to show that the fact that all apples are red does not supervene on the positive particular facts at the actual world, for the world at which those particular facts coexist with the fact that an apple is green is not a world at which all apples are red but it is also a world that differ wrt some positive particular fact (i.e there being a green apple).
    Compare this with the view that the nomic facts supervene on the Humean facts and suppose that it is a law that all apples are red. The fact that all the actual Humean facts could coexist with there being a green apple does not seem to show that the fact that it is a law that all apples are red does not supervene on the Humean facts (if it did nobody in their right mind would maintain that the nomic facts supervene on the Humean facts).

    Maybe this can be made clearer by the followin story. When God created the world, He, She, or It needed only to decide what the fundamental facts were. Once He, She, or It did so, He, She, or It did not need to decide anything else because all other facts were necessitated by those fundamental facts (God gets them for free). Once God creates a world in which each apple is red, He, She, or It also thereby creates a world a which all apples are red. Could have God created a world in which all the fundamental facts that actually obtain obtain but there is also a green apple. Of course He, She, or It could have done so but that wouldnt be a world at which the fact that all apples are red obtains. Again, this is pretty much a story, but the idea is--give me all the particular positive facts and you already have all negative and general facts because the totality of the particular positive facts that exist necessitates the existence of negative an general facts, which in turn necessitate the fact that the true negative and general propositions are true. (Note that positive particular facts necessitate general and negative facts just like all facts necessitate what we could call truth facts)

  12. The story of the proper part of the discussion I'm going to comment on so far: at 9:17, Gabrielle writes:

    Second, if the A facts supervene on the B facts (but nor viceversa), if then the existence of certain B facts necessitates (is a sufficient condition for) the existence of certain A facts (those A facts cannot fail to obtain if those B facts obtain).

    Ross at 3:23 claims confusion, since he took himself and Gonzalo to be objecting to the supervenience claim, and then Gabrielle at 11:08 says "Hey, what in the truthmaker necessitarianism thing threatens supervenience?"

    So, here's what I think is going on: Ross's confusion stemmed from the quoted bit above, and Gabrielle's confusion about Ross's confusion comes from the fact that Ross's confusion is not in fact due to anything about supervenience, but rather to do with something about entailment.

    More precisely. Supervenience: it's not possible for worlds to differ in B-facts without differing in A-facts. Entailment: For every possible A fact, there is some set of B facts such that the obtaining of those B-facts entails the obtaining of that A-fact. Problem: Supervenience will entail Entailment only if the B-facts are closed under (perhaps infinitary) conjunction and negation.

    This is pretty important here, because positive particular facts are not closed under negation, so Supervenience doesn't entail Entailment. Ross and Gonzalo's point: positive, particular facts don't entail negative or general facts. (Even if the facts that a1 is a red apple, that a2 is a red apple... are all the apple-involving facts ad that all apples are red doesn't mean these facts entail that all apples are red: these could all obtain along with there being one extra, green apple.) Gabrielle's point: general and negative facts do *supervene* on positive particular ones: you can't differ a world's general and negative ones without differing their positive particular ones.

    Punchline: I'm guessing Ross was confused by the italicized quote. Am I right about this?

  13. Right - sorry, I muddied the waters: it's not the *superveience* claim I'm objecting to, but the thought that any truthmaker or collection of truthmakers for positive facts can provide a necessitating truthmaker for the general facts. It's *that* that I just don't see how it could be the case. Of course, maybe the supervenience claim is all we need: but then that would be to *abandon* maximalism.

  14. Ross,

    I think there is still a misunderstanding here (and I'm afraid it's all due to my red-herring comment at 4:01). The NTM thinks that the truthmakers for negative truths are negative facts and that negative facts exist but they are not fundamental facts--they are supervenient facts. So, NTM does not claim that the subvenient particular positive facts are the truthmakers for any negative truths--negative facts are the truthmakers for negative truths but the existence of negative facts supervenes on that of particular positive facts.
    So, I don't see why NTM requires the rejection of maximalism. Every truth does have a truthmaker according to NTM, it's only that some truthmakers are not "stand-alone" fundamental facts but supervene on more fundamental facts.

  15. Given that both substance dualism and reductive materialism are consistent with some supervenience theses, it seems odd to suppose that a solution to the mind-body problem is a supervenience thesis. Likewise, given that radically different truthmaker theories are consistent with some supervenience theses, it seems odd to suppose that a solution to the mind-body problem is a supervenience thesis.

    That's why, it seems to me, it's not surprising that most of the attention has been given to your 4:01 proposal. For it is the one that attempts to say something about *why* the supervenience thesis is true.

    I don't know how helpful it would be to run through the arguments against non-reductive materialism here, but that's what I imagine one would do in response to your 4:01 proposal, other than worry about Necessitation, as others have.

  16. My first ever blog comment, maybe my last. A couple of weeny points. First, does 'truthmaker necessitarianism' require that truthmakers necessitate truths? If so, I don't like it. Truthmaking is, if it is anything, a relation, indeed an *internal* relation: if you have the relata you have the relation. But you do need the relata. It would be wrong to think of this in terms of the relation itself or one of the relata being necessitated by the other. Second, I don't see a problem in supposing that totalities 'supervene' in the innocuous sense that if you have all the parts of the totality, you have the totality. But wait! You could add another! In that case you'd have the original totality plus an additional part: a new totality. But consider: I have exactly five buttons on my coat (one has gone missing). To express this totality I need a 'that's all' clause. But it doesn't follow that we need an additional that's all *fact*. God can create the totality by creating the parts, then stopping. No need to add a totality fact (and certainly no need to add negative facts). Not so with language. Totality clauses are ineliminable. So two descriptions, 'there are five buttons on my coat' and 'there are exactly five buttons on my coat' can have the same truthmaker.

  17. Hi John,

    Glad to see you contributing, and I hope it's not your last!

    A question for you: Suppose instead of the way Necessitation is typically put, you put it this way: Necessarily, if the truthbearer, P, and the truthmaker, T, exist, then the T is a truthmaker for P.

    That avoids your worry about propositions being necessary existents. (It's not consistent with your last point, of course.)

  18. Gabriele: okay, I think I'm finally understanding the view at least. But just to help me get to grips with it, can you say what the advantage is meant to be if you have negative facts but say that they're not fundamental and that which one obtains supervenes on what positive facts you have. Why is that better than just having fundamental negative facts?

  19. Jon,

    I see your point but as I said I do not find it too worrying insofar as the supervenience claim is not trivial (in the sense that it is not something that all participants to the debate accept). I am not suggesting supervenience to be a solution to the problem but to identify a family of solutions to the problem, which may be quite different from one another.


    I guess simply that one wants to have as few brute facts as possible and that fundamental facts are brute facts, so the fewer the better.

    What really worries me about all this is that I cannot think of any plausible account of such general facts along the lines of particular Armstrong-style states of affairs (I agree that the token identity one is not particularly plausible but I cannot think of anything better). So, a remorseless truthmaker maximalist might just have to stick to an ontology of unanalyzable facts (i.e. facts that do not have say, particulars or universal as their components), which I find to be rather at odds with my other metaphysical views.