An Argument from Our Agency

Longtime commenter Ilíon Troas and I have been corresponding privily about topics tangential to my recent post on error and free agency. In a recent message to me, he shared the following startling argument, and courteously agreed to my suggestion that we should publish it here as a guest post. A more expansive version may be found at his blog, Iliocentrism; here, I reproduce only the core of his argument. It is this argument that prompted the train of thought in me that resulted in my even more recent post on causation.

We theists recognize two general categories of causation: mechanistic (i.e., “cause-and-effect”) and agency (“ground-and-consequent”). Most people, including most God-deniers, will initially agree that these two categories are real, and distinct, and unbridgeable … until they see where the argument is going.

From recognition of the unbridgeable distinction between mechanism and agency, I argue that agency cannot “arise” from mechanism – this is what the God-deniers who haven’t denied agency from the start will then deny and this denial can then be shown absurd and thus false – and thus that agency is, and must be, fundamental to [the] nature of reality.

But, as there is no such thing as agency unless there is an actually existing agent, it follows that *an actually existing agent* is fundamental to the nature of reality.

That is, *we* cannot be agents unless God (who is an agent) exists; or put another way: the fact that we *are* agents proves the reality of God and simultaneously proves the falseness of atheism, in all its forms.

On the other hand, atheism in all its forms denies, and must deny, true agency. For, as per the little argument above, to acknowledge the reality of agency is to acknowledge the reality of God.

Some *atheists* will try to posit random causation, or ‘randomness’ as a causation – and these people will frequently try to subsume agency under ‘randomness.’ But, this is absurd, and thus seen to be false. For, to speak of ‘randomness’ is to speak of a lack of correlation between two or more things. That is, to speak of a “random cause” is to literally speak of a “cause” which is not correlated with its alleged effect – literally, it is to speak of an effect which is not caused by a “cause,” and of a “cause” which does not cause an effect.

28 thoughts on “An Argument from Our Agency

  1. The Aristotelian-Thomist proof for God seems to ultimately rely on an argument from contingency. We live in a world of contingent beings – beings that are actually here, but in an actual sense might not be here. There must be an explanation for why these beings are here, when they might actually not be here. But this question cannot be answered by reference to another contingent being, or even by an infinite regress of contingent beings. Therefore, there must be a necessary being that can explain the existence of the contingent beings.

    I think this proof works on its face. However, the concept of “contingent being” is subject to challenge. Take an obvious example of a contingent being – the moon.

    -The moon is here, but I can logically conceive of it not being here. Therefore the moon is a contingent being.
    MEPHISTO: Ah, but that is an illusion of your mind. The moon is in fact there necessarily, as a result of deterministic processes that have been unfolding since the beginning of the Universe. It never could have not been there at this point.

    -Well, I know as a matter of science that the moon was once not there. It is here now. It will one day not be here. Therefore, the moon is contingent.
    MEPHISTO: Ah, but there is no “moon”, only the continuous flux of particles revealed by physics. Your mind notes an interesting pattern in those particles and calls it “moon”, but there is in fact no point at which those particles assume an attribute of “moonness” in reality outside of your own mind.

    Your post points to a quick and easy way to rebut these challenges. A neat feature of the Aristotelian-Thomist proof is that we do not need to prove *everything* in the Universe is contingent. We only need to prove that *one thing* in the Universe is contingent in an actual sense, and this will demand an explanation involving a necessary being.

    What is the one thing we know to be contingent? Our self, with agency and a free will known through personal knowledge. I can do X or not do X, therefore contingency is meaningful. I can create X object or not create X object, therefore these man-made artifacts are contingent in a meaningful sense. If free will is not an illusion, a necessary being is required.

    [I don’t think this is the only way to rebut Mephisto. Natural objects may in some sense be there deterministically, but the ultimate laws of the universe that put them where they are, are themselves contingent. They could have been set at different values. We know this as a matter of math. What accounts for how the values are actually set (to say nothing of the fact that they are fine tuned)?

    Furthermore, “moonness” may not have mind-independent reality, but there is *something* there that did not exist once and might not exist in the future, if only the subatomic particles, and these call for an explanation as to why they are, came into being. Something cannot ultimately come from nothing.]

    • Yes. What’s more, Mephisto’s replacement of the moon by the continuous flux of particles revealed by physics is no more than kicking the can down the road. That road terminates upon a simple question: granting arguendo that our concepts have no correlates in reality – i.e., granting nominalism – what is the correlate in reality of the concept of nominalism?

  2. Hm, well if I understand you (I am still in metaphysics kindergarten): I suppose Mephisto would say that concepts are just byproducts of the brain, which is just another part of material reality. Evolution by natural selection caused our brains to respond to certain to certain things and patterns because doing so conferred evolutionary fitness. That’s all these supposedly rarified “concepts” are.
    Three rebuttals come to mind:
    1) “Natural selection did it” can be used to waive away anything that troubles the materialist, But how credible is a Darwinist explanation for something? Finches beaks – sure. Eyeballs – perhaps. DNA – strains credulity. Consciousness built around abstract concepts – absurd.
    2) If natural selection did it, why do I experience concepts *qualitatively*? I experience *redness*; I am not just a zombie or meat computer that mindlessly avoid red things because they have a pattern of being poisonous or hot and thus inhibit my reproductive fitness. I eat a *banana* for breakfast; I don’t just mindlessly consume yellow crescent shaped things when I encounter them because they increase my chances of replicating my genes.
    3) Some concepts, like nominalism, are so abstract as to be completely detached from the material world and the phenomena therein. Where do they come from?
    Not sure if that’s what you’re getting at.

    • Those are all good points, but they are not what I was getting at. What I was getting at was that if nominalism is correct that there are no real correlates of our concepts – so that, e.g., there is nothing “out there” in reality that corresponds to our notions of redness or banananess, such concepts being rather nothing more than, and *only,* heuristics we invent and employ for the sake of convenience in parsing our sensory inputs – then, not only are there no such things really as red things or bananas, but *there is no such thing as nominalism.* It’s not a real thing. It’s a fake thing.

      Thoroughgoing and consistent nominalists will cheerfully agree to this conclusion. Not so many of them, however, will feel quite comfortable with the ultimate reductio of nominalism: namely, that there are no such things really as nominalists, including themselves.

      If concepts don’t truly pertain to anything real, but rather only speciously pertain, they can’t be truly true; they can be only speciously true.

      • I suppose the only conclusion one can draw from the reduction of nominalism is that there are no explanations, but forgive me because that is also an explanation.

      • Exactly. You’ve nailed it, and what is more, by exemplification thereof, you have identified a sure test of theoretical truth: if a notion redounds to its own impossibility, we can be sure it is false.

      • “there are no such things really as nominalists, including themselves.”

        That does not follow

        “Ockham was the father of Nominalism” refers to a concrete person, unit and individual, brought home to us by sense- experiences and informations we have of him, although the predicate applies a concept or notion to him; the notion, like all abstractions or generalisations existing in the mind and nowhere else.

        Similarly, “Philip was the father of Alexander” refers to two concrete individuals.

      • On nominalism, father, reference, concrete, person, unit, individual, home, us, sense, experience, information, we, him, predicate, application, concept, notion, abstraction, generalisation, existence, mind and nowhere are all concepts with only specious apposition to reality.

        A somewhat closer examination of their denotations reveals that Ockham, Philip and Alexander suffer from the same defect.

        Argument cannot be conducted except by concepts, so that if it purports to establish true statements it implicitly presupposes realism. Indeed even intentionally false statements presuppose realism.

      • There are immediate objects of sense-experience.

        Thus, Bronwen, lying under my desk is real – I can see and feel her. “Welsh Springer Spaniel” is a label, a notion, not a thing. “Dog” is a yet higher abstraction.

        Of course, such labels or classes are essential for “convenience of discourse,” as Berkeley notes and can vary considerably between cultures – Wagon-loads of books and learned papers have been written on Homer’s scheme of colours and no botanist now finds Linnaeus’s categories useful.

      • It should hardly surprise us to discover that this model or that, this categoreal scheme or that, are somehow or other inadequate to the domains of their appertinence. That’s how models are, by definition. But that the topo map cannot capture every last detail of the terrain does not entail that the topo map is false, or inapposite, to that terrain. Nor a fortiori does it indicate that maps are otiose per se.

        However good a topo map might be, a better is possible. So models improve over time, insofar as they are tested.

        There are indeed immediate objects of sensation, to be sure. That this is so is one of the verymost basic conceptual abstractions from raw experience. But, the notion that Bronwen now has anything to do with Bronwen then – that, i.e., there is really such a thing as Bronwen, then and now – is an abstraction. That does not mean it is a false abstraction, or that the abstraction is wrong headed. On the contrary.

        There are no unicorns or griffins on Earth, apparently. Not anymore anyway, so long as we can now tell. So ancient taxonomies have been corrected. But that a given category in a conceptual scheme turns out to be vacuous or wrongly specified does not entail that concepts per se have only specious apposition to reality.

        Indeed , it is hard to see how we might be able to tell that a model was somehow errant were it not for the fact that many – perhaps most – surviving models are predominantly reliable.

        Nominalism leaps from “whoa, this concept didn’t pan out” to “no concepts can pan out.” It is an unwarranted saltation. If it were true, then the concept that no concepts can pan out could not possibly pan out.

      • “But, the notion that Bronwen now has anything to do with Bronwen then is an abstraction.”

        And a hotly disputed one. Is what exists at a given instant the whole of the object, a time-slice of it, or a time-point instantiation? Or think of Bronwen’s world-line on a Minkowski diagram.

        According to Bergson, what we do is “to take the ceaseless, living flow of which the universe is composed and to make cuts across it, inserting artificial stops or gaps in what is really a continuous and indivisible process. The effect of these stops or gaps is to produce the impression of a world of apparently solid objects. These have no existence as separate objects in reality; they are, as it were, the design or pattern which our intellects have impressed on reality to serve our purposes.”

        Think of a still from a motion picture.

        Joad found this reminiscent of Dedekind’s creation of a new irrational number at every gap in the continuous number line at which there is no existing real number.

      • Sure. I am myself a Bergsonian of the Whiteheadian persuasion, so I am amenable to such arguments. But – and – Minkowski diagrams, world lines, flow, living, cuts, indivisibility, process, stops, gap, solid, object, existence, reality, design, pattern, intellect, impression, purpose, instant, whole, slice, point, time, instance, and so on are all likewise abstractions. Understand, I don’t mean to question whether any of those abstract concepts have true apposition to reality. I think they do. It’s just that there is no way to proceed intellectually other than by the presupposition that concepts can have true – or, therefore, false – apposition to reality. Nominalism proposes that our concepts can have no true – or, therefore, false – apposition to reality. It proposes to forestall thought, and indeed animal action.

        As to that trenchant quote of Bergson, we should recall always that the integral qua basic character of reality – the One – nowise vitiates the differential qua basic character of reality – the Many. To think that there is only the Many is atheism; to think that there is only the One is acosmism.

      • “Nominalism proposes that our concepts can have no true – or, therefore, false – apposition to reality”

        Wittgenstein points out the distinction between analysing a phenomenon (e.g. thought) and a concept (e.g. that of thinking), and therefore the use of a word. “Nominalists,” he argues, “make the mistake of interpreting all words as names, and so of not really describing their use, but only, so to speak, giving a paper draft on such a description.”
        (PI 384)

        That seems right.

      • I think I grasp your point. If concepts are just useful heuristic delusions generated by our brains with no mind-independent reality, then that includes not just nominalism but the nominalist himself. His very self is a delusion thrown off by his brain, which should disconcert him because it contradicts his rich and very real qualitative experience of “self”.

        It’s analogous to the logical positivist saying: “Why don’t I believe in metaphysics? Because metaphysics is about asking “why” questions, and there are no “why” questions, only “how” questions. That’s why.” You just did the very thing you deny has meaning – ask and answer a why question.

        I think we have called nominalism into serious question, but I don’t think we’ve quite defeated nominalism, if the nominalist is willing to fall back on “the self and its manifold concepts, including logic and math, are just heuristic delusions of the brain that evolved to increase reproductive fitness; nothing contained within the brain is “out there” in reality in a substantive sense”.

        I think to completely defeat nominalism you need to demonstrate that math and logic are “meta-true” in that they have an existence or shape reality in an affirmative way and thus are not mere passive processing tools of the mind.

        Perhaps this has been done. Quantum mechanics apparently shows that electrons are not little balls whizzing around – they are probability clouds, they are math. Therefore the fundamental structure of reality is math. Maybe Godel has shown math has its own truth beyond the mind’s utility. I don’t know; theoretical physics and math are far above my pay grade. But if this could be demonstrated:

        1) Math and Logic have mind-extrinsic reality
        2) Therefore, nominalism is affirmatively demonstrated to be false (not just highly subject to question as we have established)
        3) Therefore, classical metaphysics has been conclusively demonstrated to be true and the ultimate nature of reality; it is not just a logically coherent closed system which have good reasons for believing in extrinsic to the system itself.

      • All pertinent points, but it’s simpler than that. Nominalism is straightforwardly autophagous.

        1. Concepts have only specious apposition to reality.
        2. This syllogism consists entirely of concepts.
        3. This syllogism in its entirety has only specious apposition to reality.


      • “His very self is a delusion thrown off by his brain”

        Which by-the-by, it really is. The “self” is simply a misconstrual of the reflexive pronoun, under the influence, one suspects, of the Cartesian Ego.

        This led to a tour de force on the part of Locke, who raises the question of re-identification: might not the thinking substance which thought the thought “I did it” — the genuine thought of agent-memory — nevertheless be a different thinking substance from the one that could have had the thought: “I am doing it” when the act was done? Thus, one can detach the identity of the “self” from the identity even of the thinking being which does the actual thinking of the I-thoughts.

        After all, as Hume pointed out, we are aware of our perceptions, but not of an “I” or “self” doing the perceiving; all we perceive is the thinking that thinks this thought. But, if “I” is just the name of the whole collection of perceptions, what, then, makes “I” a unity? Asking what “I” refers to is like asking what “It” refers to in “It is raining.”

        “Self-consciousness” is assumed to be consciousness of a “self.” The self is then reified into something that some things (people, for example) are or have: a sort of mysterious, immaterial entity. How this self is connected with this particular, living human body is by no means obvious; that is because it is a piece of nonsense.

        In reality, “self-consciousness” simply means “consciousness that such-and-such holds of oneself.” It goes without saying that here “oneself” is simply the indirect reflexive, the reflexive of indirect speech. Understanding indirect speech we know what the related direct speech is. That is all. There is nothing at all mysterious about “He washed himself,” “She cut herself.”

        “So-and-so holds of me” is not at all problematic; “me” refers to this living human body that is currently operating this keyboard and thinking these I-thoughts (and who, in distant childhood, learned to have them by being asked what I had done, was doing &c).

        If, instead of his famous, “I think, therefore I am,” Descartes had tried, “I am jumping up and down,” the fallacy of the “self” would have been obvious enough,

      • 1. Concepts have only specious apposition to reality.
        2. This syllogism consists entirely of concepts.
        3. This syllogism in its entirety has only specious apposition to reality.

        I spent my commute turning this one over in my head, it’s a tough one! How would Mephisto try to rebut?

        1. Nominalism is the belief that concepts only exist in the mind
        2. Nominalism is a concept
        3. Nominalism only exists in the mind, but that’s OK because we aren’t deploying the assumption that it has real world significance to reach grounded metaphysical conclusions (e.g. non-subjective truth vs. falsity), only realism is trying to do that.

        @Michael Paterson-Seymour I don’t think you’ve done anything to address the ways I have called nominalism into serious question. My qualitative experience of self is not a grammar error. The only people who are just undergoing the act of perceiving without self reflecting on the mundane or profound implications of those perceptions for the self are those operating under a profound intellectual disability. Why would I get sad if I’m just the act of perceiving, and not the perceiver? Curiously, the intellectually disabled don’t seem to get sad precisely because their self-perception is attenuated.

      • Curtis, Mephisto’s retort as you have stated it works only if we excise the mind from reality; if, that is to say, we suppose that there is no such thing as the mind; for, if there are such things as minds, so that minds are parts of reality, then the conclusion of my syllogism tells against minds as well as it does against the rest of reality: the notion that concepts have only specious apposition to reality has itself only specious apposition to the mindful bits of reality.

        NB also that both the premises of Mephisto’s retort are general statements about reality – about the relations throughout existence of minds, concepts, and reals – that both purport to be true and are constituted of concepts. So, Mephisto’s syllogism is after all trying to state metaphysical truths. And nominalism does the same thing. It says, “it is universally true that there are no universals.”

      • “[I]f, that is to say, we suppose that there is no such thing as the mind

        “Mind” is an abstract, general idea, not a “thing.”

        Actual people do certain things, knowing, learning, discovering, imagining, pretending, hoping, wanting, feeling depressed, feeling a pain, resolving, doing voluntarily, doing deliberately, perceiving, remembering and so on.

        Because these activities have some similarities (“similarity” and “difference” are themselves abstractions), we bundle these very different activities together and label them “mental.” Then we postulate that they are all the product of a single faculty (which we cannot begin to imagine) and, if we are Dualists, we round it off by treating “Mind” (with a capital M) as a substance or entity, in Aristotle’s sense of substance (what exists without either being predicated of or existing in anything else)

        Minds don’t think; people think. Intelligence and intention are not “internal mental processes.” When we see someone making a cup of coffee, we see, not the results of intelligence and intention; we see intelligence and intention in action

      • We agree. To reify mind as such is to fall prey to the Fallacy of Misplaced Concreteness. Minds are not themselves substances, but rather aspects of substances. The same is true of intelligence, acts, and intentions. We call all such properties of substances “things” not so much to indicate that they are substances in themselves, as to indicate that they are real aspects of substances. Thus as carefully and properly interpreted, the statement “there is such a thing as minds” should be taken to indicate only that minds are real aspects of some real substances.

      • Hm, yes, that seems to work. I’ll have to digest it a while. Thanks for drilling down on this with me, even though we got off the original topic.

  3. Some philosophers have simply denied “mechanical” causality altogether (one thinks of the Occasionalism of Malebranche or the Subjective Idealism of Berkeley). In effect, they claim that our sense-data is produced immediately by God. So, too, does Kant, for whom causality is an a priori principle or form by which we organise and sort sense-experience.

    St John Henry Newman flirted with the idea: “What are the phenomena of the external world, but a divine mode of conveying to the mind the realities of existence, individuality, and the influence of being on being, the best possible, though beguiling the imagination of most men with a harmless but unfounded belief in matter as distinct from the impressions on their senses? This at least is the opinion of some philosophers…” – Arians of the Fourth Century

    Phenomenalism comes pretty close to it, in confining our knowledge to our perceptions.

  4. The notion of “random causes” is not as absurd as it appears at first sight.

    Thus, Bertrand Russell observes, “The principle ‘same cause, same effect’ … is … utterly otiose. As soon as the antecedents have been given sufficiently fully to enable the consequent to be calculated with some exactitude, the antecedents have become so complicated that it is very unlikely they will ever recur.”

    In other words, by the time you pack enough material into C to guarantee that E will follow, C is so complex that we cannot reasonably expect it to occur more than once. But any C that occurs only once will inevitably be succeeded by any E that just happens to occur on that occasion.

    This is not to say we cannot find regularities or patterns of functional dependence, but these are not general, exceptionless laws and we should not mistake them for such.

    In his sadly-neglected Grammar of Assent, St John Henry Newman gives a homely illustration of the same concept:

    “I call the characteristics of an individual accidents, in spite of the universal reign of law, because they are severally the coincident of many laws, and there are no laws as yet discovered of such coincidence. A man who is run over in the street and killed, in one sense suffers according to rule or law; he was crossing, he was short-sighted or preoccupied in mind, or he was looking another way; he was deaf, lame, or flurried; and the cab came up at a great pace. If all this was so, it was by a necessity that he was run over; it would have been a miracle if he had escaped. So far is clear; but what is not clear is how all these various conditions met together in the particular case, how it was that a man, short-sighted, hard of hearing, deficient in presence of mind, happened to get in the way of a cab hurrying along to catch a train. This concrete fact does not come under any law of sudden deaths, but, like the earth’s yearly path which I spoke of above, is the accident of the individual. It does not meet the case to refer to the law of averages, for such laws deal with percentages, not with individuals, and it is about individuals that I am speaking…”

    • In other words, by the time you pack enough material into C to guarantee that E will follow, C is so complex that we cannot reasonably expect it to occur more than once.

      Yes. On Gödelian Incompleteness, the formal specification of any concrete is not finitely completable; and if events are to be different from one another, so that there is a Many, their specification strings must somehow differ, so that each thing is unique. But from these premises, it simply does not follow that:

      … any C that occurs only once will inevitably be succeeded by any E that just happens to occur on that occasion.

      I.e., from the fact that each event together with its causal inputs is unique, it does not follow that each event succeeds its causal inputs as a matter of pure happenstance, without any order or reason.

      If events do indeed follow their causal inputs as a matter of pure happenstance, then there is no order of nature, period full stop; and so, we cannot hope to make even a beginning of understanding. That impossibility of understanding rules out the understanding that we cannot hope to understand. If we can’t succeed at all in understanding, then neither can we fail thereat. There’s no way to obtain a defective instance of x if there are no such things as x in the first place.

      NB: the suggestion that each event succeeds its causal inputs as a matter of pure happenstance is itself a perfectly general notion – an hypothesis of a metaphysical law – of the very sort that it avers is unwarrantable.

  5. Would someone care to expand – in poor man’s language – on ‘ground and consequent’? I’ve read Ilion reference to Lewis’s kettle. Is it only descriptive? Could it lead to abstract reasoning that produces, if not physics then something like it?

    • Ground and consequent denote formal terms of logical arguments: grounds are premises, while consequents are conclusions. So, they are strictly formal, whether as abstractions from concrete events (which is how we usually employ them), or as purely conceptual (as in mathematics, logic, and metaphysics). The findings of physics are ideally expressed in formal equations or functions – the acme of success in natural history consists in finding and specifying the formal relations that characterize a broad category of empirical data – so mathematical physics is ground → consequent reasoning. In concrete terms, you can grasp the distinction between that sort of purely formal consequentiality and actual, causal consequentiality, by considering that the equations on the page do not by themselves act or move anything. Meanwhile the physicist who reads them does definitely act and move things.


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.