Virtual Colloquium: Chad McIntosh, “How to be a Rational Foundationalist”
February 24, 2017 — 6:00

Author: Kenny Pearce  Category: Existence of God  Tags: , , , , , ,   Comments: 5

This week’s Virtual Colloquium paper is “How to be a Rational Foundationalist” by Chad McIntosh. McIntosh is a PhD student at Cornell writing a dissertation entitled Rational Foundationalism. His work has appeared in Religious Studies and Res Cogitans. His blog, Appeared-to-Blogly, hosts a detailed outline and bibliography on natural theology.

How to be a Rational Foundationalist

Chad McIntosh

Many thanks to Kenny Pearce for inviting me to be a part of Prosblogion’s Virtual Colloquium. I have been a reader of Prosblogion for many years, so it is an honor to contribute. The paper attached below is a massively condensed version (you can think of it as one of those APA 3k mutilations) of the main idea of my dissertation, Rational Foundationalism. Despite what the title suggests, it is an exercise in metaphysics, not epistemology. Feedback, via comments below or email, are most welcome!

Consider a thingy-version of the PSR, where every thing that exists has an explanation of its existence, either in itself or in some other thing. It is widely held that if the PSR is true, then there must be an “ultimate ground” of the cosmos, such as God. It can’t be “turtles all the way down,” as they say; the buck stops with a being whose raison d’etre is in itself. Such arguments have received a lot of attention and are well-known, especially to readers of this blog.

But there hasn’t been as much attention given to what it means for something to have its explanation “in itself,” and traditional proposals strike me as either implausible or incoherent. Saying something’s explanation is “in itself” if it exists necessarily is implausible, because clearly necessary things, like numbers, if they exist at all, still require an explanation. And saying God’s explanation is in himself because his existence is identical to his essence (Aquinas) seems like bootstrapping, and saying God’s explanation is in himself because his existence and essence depend on each other (Leibniz) seems viciously circular, like a chicken-and-egg scenario. I propose a very different way for how something can have its explanation “in itself,” which is roughly as follows.

First, I assume that there are things whose existence isn’t explained by their causes, but by their grounds. For example, another way of saying that the sufficient reason for the existence of the number 2 is God is to say the number 2 is fully grounded by God. So the question I’m asking is how the “ultimate ground” itself gets fully grounded.

An immediate barrier to considering this question along these lines is that, at least according to contemporary terminology, an “ultimate ground” is something fundamental, which, by definition, is ungrounded. Well, so much the worse for contemporary terminology. The heart of the concept of fundamentality is independence, which can be given a richer meaning than just “not dependent;” i.e., ungrounded. In their affirmations of the divine attribute of aseity, for example, theists often describe God as being not dependent on anything distinct from or external to himself, or on anything ad extra. Such descriptions leave open the possibility that God might yet depend on something ad intra.

But what could that be? A very simple answer, much to the consternation of divine simpletons: parts! So let’s say that something is fundamental iff it is fully grounded in its parts but nothing but its parts. Of course, this just pushes the question back a step to how the parts get fully grounded. But our rejection of bootstrapping scenarios means that the parts can’t fully ground themselves, and our rejection of turtles-all-the-way-down scenarios eliminates turtle gunk (parts fully grounded in parts all the way down), and our rejection of chicken-and-egg scenarios eliminates chicken parts (one part fully grounding another part and vice versa). As I argue in the paper, the only way for the parts to get fully grounded is if there is a minimum of three parts, each of which partially grounds the others. So, any one part gets fully grounded by the other two. This is possible because, as the examples I discuss show, partial ground, unlike full ground, can be symmetric without being viciously circular so long as the grounding structure is minimally tripartite. The Devil’s in the details which, in this case, are in the paper.

A fundamental being, then, must have at least three parts. Why stop at three? I don’t go into this in the attached paper, but here’s a closing thought. We’re all familiar with Ockham’s Razor: don’t multiply entities beyond necessity. Jonathan Schafer has recently proposed his own version—Schafer’s Laser, we might call it—according to which we shouldn’t multiply fundamental entities beyond necessity. Fine, there exists just one fundamental being. Schafer thinks it’s the cosmos, which grounds its many, many parts. I have a parsimony principle of my own to recommend, Chad’s Eraser: don’t multiply parts of a fundamental entity beyond necessity. And on my view, all we need is one fundamental being with just three parts.

This a surprising picture of what a fundamental being must be like. Perhaps it is less surprising to Christian theists. But with the PSR, the proposed understanding of fundamentality, and the coherence of the examples of symmetric partial grounding, we have an argument for the audacious and bizarre conclusion that there exists a fundamental being that is, in essence, triune. If you prefer a cute name for this attribute, it would be not aseity, but triseity.

The full paper is here. Comments welcome below!

  • Aristotle

    This is an excellent blurb–interesting and entertaining. I look forward to reading the entire paper. Question though: What work is the distinction between ungrounded and independence really doing? Is it allowing you to say that God is independent and yet is still grounded? (albeit in this ad intra way?)

    February 24, 2017 — 20:28
    • Thanks for the comment, Aristotle. That’s right: theists have given cosmological arguments for the existence of a “fundamental,” “independent,” “self-existent,” “a se” being. I think there’s something to these descriptions, and, furthermore, think the contemporary notion of ground can be of help in seeing what. But there is pressure from the PSR to resist understanding fundamentality as “ungrounded.” But our new understanding shouldn’t do violence to the intuitive core of the concept of fundamentality, which I think is independence. So I’m proposing that there is indeed an understanding of independence–and so fundamentality–that is consistent with being grounded.

      February 25, 2017 — 10:23
  • Chad!

    It’s been a long time, my friend. Thanks for sharing this paper with us.

    I don’t know much at all about the grounding literature. So maybe you can explain something to me in a bit more detail. You say:

    “In this way it avoids the priority problem: when a and b ground c, a alone, as a mere partial ground of c, is not ontologically prior to c. What’s prior to c in this case is not a, but a with b.”

    This seems very odd to me. It seems to me that the following is true: If a and b fully ground c together, then *both* a and b are prior to c”. Call this the Conjunction Principle (CP). In fact, what you say earlier when you are discussing “full grounding” circularity seems to support CP. You say:

    “If we wanted to know the full story on why y exists and are told it’s x, we would justifiably assume that x is ontologically prior to y: to find the reason for y, we have to go back to x… it seems that priority tracks full ground: if x fully grounds y, then x is ontologically prior to y.”

    Right. So if the full story for why c exists, its full ground, is “a and b”, then we can justifiably assume that “a and b” are ontologically prior to c. Hence CP.

    Where is my reasoning going wrong?

    February 26, 2017 — 16:24
  • Alan White

    I’m going to ask what may be a clueless question, but occurred to me while reading your very intriguing post.

    The knowledge of the existence of relations is notoriously not reducible to the question of the knowledge of the existence of relata. For example, fraternal triplets might go to their graves thinking they sustain that biological relationship, when actually they were three separate adoptees who were never told the truth. So the triplets as relata exist, but the fraternal relation doesn’t.

    So it seems to me that providing the reason that some real relationship exists with any given relata is a separate issue, even if a priori (I mean that e.g. the relation between two different numbers as greater-than may be dependent on the matter of defining different numbers as relata of the relation, but the relation itself is still some function beyond those definitions). And that leads me to another question: does your position overall require an Anselmic stance about a unique necessary being?

    February 26, 2017 — 19:04
  • Tim Perrine

    Hi Chad,

    Thanks for the post. Hope all is well! I’ve got a couple of questions, though maybe your response will just be read a longer paper of mine!

    First, a (probably) minor point. I’m not sure I understand why metaphysical foundationalism and metaphysical rationalism are inconsistent. I take the former to be that there is a being or existent that is not dependent on any other being or existent. The latter is that the existence of every being has an explanation. So formed, these are not inconsistent. I guess you might have in mind a stronger a stronger form of metaphysical foundationalism according to which the fact that a being exists is not dependent on any other fact? Anyway, this is probably just a terminological issue.

    Second, I’m worried about the application to theistic issues. (Though the word ‘trinity’ doesn’t appear in the paper, somehow I have a feeling that issue is simmering below the surface.) Taking classical mereology as stating the truths about anytime anyone talks about ‘parts’ strikes me as clearly problematic. Plausibly, my pinky toe is part of me, and I am part of my department. But it is not at all plausible that my pinky toe is part of my department. (The department web page is not incomplete for not having an entry and photo of it.) But in classical mereology, parthood is transitive. A superior understanding, I think, is taking CM as stating formal principles regarding an term ‘part’ that can interpreted in various ways and then the interesting questions become not whether CM is true, but whether certain usages of ‘part’ can be modeled using CM.

    I raise this because it is not at all clear to me that when we speak of God having ‘parts’ we can understand that language as being a model for CM. But your argument in section 5 requires a principle of classical mereology. But if we cannot model God-part-talk in terms of CM, then we need some independent reason for thinking that principle would hold if we wanted to take God as our fundamental entity, which I take it many theists will want. Can you say more about how you think part-talk applies to God?

    Third, I’m not sure I understand how the proposal in the final section should work. Suppose we have an object O with three parts, a,b,c. You seem to want to say that the following is possible:

    O is fully grounded in a,b,c
    a is fully grounded in b,c
    b is fully grounded in c, a
    c is fully grounded in b,a

    But here is my worry about these claims. I take it that almost everyone in the grounding literature accepts the following principle: if G (or Gs) fully grounds H, then G (or each of the Gs) is more fundamental than H. I think most people think the principle is self-evident. Further, anyone who thinks there is such a thing as a grounding explanation needs an account of what distinguishes grounding explanations from other explanations (like causal explanations, conceptual explanations, etc.). This principle helps explain the difference: in non-fundamental explanations, this principle need not hold. But this principle leads to a contradiction with your claims. For your claims imply that (e.g.) a is less fundamental than b, i.e. (by definition) b is more fundamental than a. But your claims would also imply that: b is less fundamental than a, i.e. (by definition) b is not more fundamental than a. So I’m worried that I don’t fully see exactly how your positive proposal is to work or if you really want to say those four things.

    Anyway, thanks for an interesting paper!


    February 27, 2017 — 10:33
  • Leave a Reply to CAMcIntosh Cancel reply

    Your email address will not be published. Required fields are marked *