A compositional fine-tuning argument

Assume naturalism about the human mind. Our best naturalistic account of the human mind is functionalism. But functionalism faces multiple too-many-minds problems. The most famous of these are the Chinese Room and its variants like Schwitzgebel’s consciousness of the United States argument. But a more troubling bevy of problems comes from abundant ontologies. Thus, as Dean Zimmerman noted (building on Unger), where I am there are many clouds of atoms that differ from me in an insignificant way—say, an atom in some insignificant skin cell. On functionalism, each of these clouds should have the same conscious states as I do. Or, as Johnston argued, I have many personites—temporal parts of my life that are intrinsically just like the life of a person could be. On functionalism, they will have the same conscious states as me. The clouds of atoms and personites are not just a consequence of functionalism but also of other naturalistic accounts of mind.

But why are the too-many-minds problems problems, beyond the fact that they are counterintuitive? After all, we have good reason to think that the mind is mysterious enough that the true theory will have some counterintuitive consequences.

I think the best answer is ethics. If a country has a person-level mind, then it would be a murder-suicide for the citizens to vote to dissolve the country. But it is not wrong for the citizens to vote to dissolve a country for, say, economic reasons. If the Zimmerman argument is right, then where there is a person feeling pain, there are many other beings with human-level consciousness feeling the same pain. But the number of being that coincide with a specific person rapidly increases with the size of the person—the more cells they have, the more clouds of atoms there are that differ with respect to a few insignificant atoms. Consequently, if we have a choice between relieving an equal pain in two smaller persons or one much larger person, we should always relieve the pain in the larger one, because the number of conscious atom clouds coinciding with the larger person is likely much larger than the total number of atom clouds coinciding with the smaller ones. In other words, crucial intuitions about equal treatment of people are undercut. Something similar is true on the Johnston arguments if the number of personites is finite, and if it’s infinite we have other ethical problems. On the other hand, there is no immediate serious ethical problem in saying the Chinese Room is conscious.

Given functionalism, I think there is only one way to block the ethically problematic too-many-minds cases: deny that the alleged entities exist. There are no countries. There is only one human-shaped cloud of atoms where I am. There are no personites. But we better not go all the way to blocking all complex objects—we will get other ethical problems if we conclude with the early Unger that humans don’t exist. In other words:

1. If functionalism and ethical realism are true, restricted composition is true.

Restricted composition says that some but not all (proper) pluralities of atoms compose a whole. Note that (1) also applies to some other naturalistic theories than functionalism.

But it’s not enough that restricted composition be true. What we need is a carefully fine-tuned restricted composition. If we restrict composition too much, there will be no humans—and that’s ethically unacceptable. If we don’t restrict composition enough, there will be too many minds of an ethically problematic sort. In other words, restricted composition must be fine-tuned to fit with human ethics.

That’s difficult to do. For instance, van Inwagen’s life-account—that a plurality composes a whole if and only if it has a life together—has the problem that clouds of atoms that differ from me insignificantly have a life together just as I do.

Given naturalism, I think any restricted composition account that fits with ethics will involve seemingly arbitrary choices. Thus, one might start with van Inwagen’s account, but have an incredibly fine-grained account of what counts as “a life together” such that only one of the clouds of atoms nearly coinciding with me has a life together—namely, the cloud constituting me. But such a fine-grained account will have a ton of free parameters, and will be an implausible candidate for a metaphysically necessary account of restricted composition. Thus, the account will not only be fine-tuned but will likely be contingent.

How do we explain the fine-tuning of restricted composition for ethics? It’s hard to see how to do it other than by supposing that fundamental reality is value-driven. There are two main value-driven theories of fundamental reality: theism and axiarchism, where the latter is something like the view that reality must be for the best. Thus we have an argument for theism or axiarchism. And axiarchism, as Rescher noted, plausibly implies theism, since it’s for the best that there be a perfect being. So, either way, we get theism.

We can also run this argument in a Bayesian way. Assume naturalism about the earth ecosystem as a background belief, and assume as part of the background that the physical simples are arranged as they are. On atheism, it is extremely unlikely that composition is fine-tuned for ethics. On theism, it is at least moderately likely. So, we have significant evidence for theism.

Objection: God can’t control which cloud of atoms composes a whole, because whatever is the answer, the answer is metaphysically necessary.

Response: First, as noted above, it is likely that any ethically fine-tuned restricted composition theory has a bunch of parameters that appear contingent, and hence is likely contigent. Second, God is creator and has power over being itself. It seems quite plausible that where there is a bunch of particles God can lend his power to create an entity composed of the particles. Third, if God exists, likely modality itself is grounded in God—all reality necessarily reflects the goodness of God. But if so, then divine goodness may help to explain surprisingly good features of necessary truths, such as a fine-tuned but necessary theory of composition. Fourth, we don’t need to be certain of any of the above. All we need is that one of these stories is an order of magnitude more likely on theism than the fine-tuning of restricted composition is given naturalism (where the probabilities are all epistemic).

If my argument succeeds, it yields a dilemma:

2. Either naturalism about humans is false or God exists.

One may ask whether some variant of the above fine-tuning argument applies if naturalism about humans is true. I expect it does, but the exact shape of the bump under the rug will be different for different non-naturalistic stories. For instance, on Cartesian theories, there will be the question of why there is exactly one soul per human body. On strong emergence, we can ask why consciousness arises in exactly one of the human-shaped clouds of atoms where I am.

The explanation of our reliability is not physical

1. All facts completely reducible to physics are first-order facts.

2. All facts completely explained by first-order facts are themselves completely reducible to first-order facts.

3. Facts about our epistemic reliability are facts about truth.

4. Facts about truth are not completely reducible to first-order facts.

5. Therefore, no complete explanation of our epistemic reliability is completely reducible to physics.

This is a variant on Plantinga’s evolutionary argument against naturalism.

Premise (4) follows from Tarski’s Indefinability of Truth Theorem.

The one premise in the argument that I am not confident of (2). But it sounds right.

First-order naturalism

In a lovely paper, Leon Porter shows that semantic naturalism is false. One way to put the argument is as follows:

1. If semantic naturalism is true, truth is a natural property.

2. All natural properties are first order.

3. Truth is not a first order property.

4. So, truth is not a natural property.

5. So, semantic naturalism is not true.

One can show (3) by using the liar paradox or just take it as the outcome of Tarski’s Indefinability of Truth Theorem.

Of course, naturalism entails semantic naturalism, so the argument refutes naturalism.

But it occurred to me today, in conversation with Bryan Reece, that perhaps one could have a weaker version of naturalism, which one might call first-order naturalism that holds that all first order truths are natural truths.

First-order naturalism escapes Porter’s argument. It’s a pretty limited naturalism, but it has some force. It implies, for instance, that Zeus does not exist. For if Zeus exists, then that Zeus exists is a first-order truth that is not natural.

First-order naturalism is an interestingly modest naturalist thesis. It is interesting to think about its limits. One that comes to mind is that it does not appear to include naturalism about minds, since it does not appear possible to characterize minds in first-order language (minds represent the world, etc., and talk of representation is at least prima facie not first-order).

Duplicating electronic consciousnesses

Assume naturalism and suppose that digital electronic systems can be significantly conscious. Suppose Alice is a deterministic significantly conscious digital electronic system. Imagine we duplicated Alice to make another such system, Bob, and fed them both the same inputs. Then there are two conscious beings with qualitatively the same stream of consciousness.

But now let’s add a twist. Suppose that we create a monitoring system that continually checks all of Alice and Bob’s components, and as soon as any corresponding components disagree—are in a different state—then the system pulls the plug on both, thereby resetting all components to state zero. In fact, however, everything works well, and the inputs are always the same, so there is never any deviation between Alice and Bob, and the monitoring system never does anything.

What happens to the consciousnesses? Intuitively, neither Alice nor Bob should be affected by a monitoring system that never actually does anything. But it is not clear that this is the conclusion that specific naturalist theories will yield.

First, consider functionalism. Once the monitoring system is in place, both Alice and Bob change with respect to their dispositional features. All the subsystems of Alice are now incapable of producing any result other than one synchronized to Bob’s subsystems, and vice versa. I think a strong case can be made that on functionalism, Alice and Bob’s subsystems lose their defining functions when the monitoring system is in place, and hence lose consciousness. Therefore, on functionalism, consciousness has an implausible extrinsicness to it. The duplication-plus-monitoring case is some evidence against functionalism.

Second, consider Integrated Information Theory. It is easy to see that the whole system, consisting of Alice, Bob and the monitoring system, has a very low Φ value. Its components can be thought of as just those of Alice and Bob, but with a transition function that sets everything to zero if there is a deviation. We can now split the system into two subsystems: Alice and Bob. Each subsystem’s behavior can be fully predicted from that subsystem’s state plus one additional bit of information that represents whether the other system agrees with it. Because of this, the Φ value of the system is at most 2 bits, and hence the system as a whole has very, very little consciousness.

Moreover, Alice remains significantly conscious: we can think of Alice as having just as much integrated information after the monitoring system is attached as before, but now having one new bit of environmental dependency, so the Φ measure does not change significantly from the monitoring being added. Moreover, because the joint system is not significantly conscious, Integrated Information Theory’s proviso that a system loses consciousness when it comes to be in a part-to-whole relationship with a more conscious system is irrelevant.

Likewise, Bob remains conscious. So far everything seems perfectly intuitive. Adding a monitoring system doesn’t create a new significantly conscious system, and doesn’t destroy the two existing conscious systems. However, here is the kicker. Let X be any subsystem of Alice’s components. Let S_X be the system consisting of the components in X together with all of Bob’s components that don’t correspond to the components in X. In other words, S_X is a mix of Alice’s and Bob’s components. It is easy to see the information theoretic behavior of S_X is exactly the same as the information theoretic behavior of Alice (or of Bob for that matter). Thus, the Φ value of S_X will be the same for all X.

Hence, on Integrated Information Theory, each of the S_X systems will be equally conscious. The number of these systems equals to 2ⁿ where n is the number of components in Alice. Of course, one of these 2ⁿ systems is Alice herself (that’s S_A where A is the set of Alice’s components) and another one is Bob himself (that’s S_∅). Conclusion: By adding a monitoring system to our Alice and Bob pair, we have created a vast number of new equally conscious systems: 2ⁿ − 2 of them!

The ethical consequences are very weird. Suppose that Alice has some large number of components, say 10¹¹ (that’s how many neurons we have). We duplicate Alice to create Bob. We’ve doubled the number of beings with whatever interests Alice had. And then we add a dumb monitoring that pulls the plug given a deviation between them. Suddenly we have created 2¹⁰¹¹ − 2 systems with the same level of consciousness. Suddenly, the moral consideration owed to to the Alice/Bob line of consciousness vastly outnumbers everything.

So both functionalism and Integrated Information Theory have trouble with our duplication story.

Aristotelian functionalism

Rob Koons and I have argued that the best functionalist theory of mind is one where the proper function of a system is defined in an Aristotelian way, in terms of innate teleology.

When I was teaching on this today, it occured to me (it should have been obvious earlier) that this Aristotelian functionalism has the intuitive consequence that only organisms are minded. For although innate teleology may be had by substances other than organisms, inorganic Aristotelian material substances do not have the kind of teleology that would make for mindedness on any plausible functionalism. Here I am relying on Aristotle’s view that (maybe with some weird exceptions, like van Inwagen’s snake-hammock) artifacts—presumably including computers—are not substances.

If this is right, then the main counterexamples to functionalism disappear:

• the Chinese Room (and relatives)

• rocks (on computational reinterpretation arguments)

• organizational consciousness.

Next recall Leibniz’s mill argument: if a machine can be conscious, a mill full of giant gears could be conscious, and yet as we walked through such mill, it would be clear that there is no consciousness anywhere. But now suppose we were told that the mill has an innate function (not derived from the purposes of the architect) which governed the computational behavior of the gears. We would then realize that the mill is more than just what we can see, and that would undercut the force of the Leibnizian intuition. In other words, it is not so hard to believe that a mill with innate purpose is conscious.

Further, note that perhaps the best physicalist account of qualia is that qualia are grounded in the otherwise unknowable categorical features of the matter making up our brains. This, however, has a somewhat anti-realist consequence: the way our experiences feel has nothing to do with the way the objects we are experiencing are. But an Aristotelian functionalist can tell this story. If I have a state whose function is to represent red light, then I have an innate teleology that makes reference to red light. This innate teleology could itself encode the categorical features of red light, and since this innate teleology, via functionalism, grounds our perception of red light, our perception of red light is “colored” by the categorical features not just by our brains, but by the categorical features of red light (even if we are hallucinating the red light). This makes for a more realist theory of qualia, on which there is a non-coincidental connection between the external objects and how they seem to us.

Observe, also, how the Aristotelian story has advantages of panpsychism without the disadvantages. The advantage of panpsychism is that the mysterious gap between us and electrons is bridged. The disadvantages are two-fold: (a) it is highly counterintuitive that electrons are conscious (the gap is bridged too well) and (b) we don’t have a plausible story about how the consciousness of the parts gives rise to a consciousness of the whole. But on Aristotelian functionalism, it is teleology that we have in common with electrons, so we do not need to say that electrons are conscious—but because mind reduces to teleological function, though not of the kind electrons have, we still have bridging. And we can tell exactly the kind of story that non-Aristotelians do about how the function of the parts gives rise to the consciousness of the whole.

There is, however, a serious downside to this Aristotelian functionalism. It cannot work for the simple God of classical theism. But perhaps we can put a lot of stress on the idea that “mind” is only said analogously between creatures and God. I don’t know if that will work.

Where does the evolutionary argument for naturalism work?

I’ve never been moved by Plantinga’s evolutionary argument against naturalism in general, but I’ve also always found it plausible that naturalism and evolution undercuts cognitive reliability in certain areas, such as metaphysics. It seems, on the other hand, really plausible that we would get cognitive reliability for empirical things (largely because of the fact that naturalism makes causal theories of content very likely).

One might go on to conjecture that Plantinga’s argument works everywhere outside of empirical areas. I thought so until I realized that there is a significant area of normativity where the argument doesn’t work: prudential value judgments. This is because that life and reproduction is good for living things, and many other goods, noncoincidentally, contribute to life and reproduction. But at the same time, we are evolutionarily selected for successful promotion of life and reproduction. A being that believed that life is bad would be unlikely to promote its own survival, and hence unlikely to pass on its genes.

Plantinga’s standard answer to similar objections in the empirical arena is that behavior does not come just from beliefs, but from a combination of belief and desire. But this response is rather implausible in the prudential rationality case. It is extremely plausible that there is a conceptual link between a mental state representing something as good for one and having a desire for that thing. If there were no correlation between a mental state S respecting a thing and having a desire for that thing, that mental state just is not a belief that the thing is good for one.

A schematic argument against naturalistic reductions

Here is an argument schema.

1. If X is reducible to the natural, then likely the vast improvement of natural science over the last three hundred years would have led to a much better knowledge of X.

2. If X is not reducible to the natural, then it is not likely that the vast improvement of natural science over the last three hundred years would have led to a much better knowledge of X.

3. The vast improvement of natural science over the last three hundred years has not led to a much better knowledge of X.

4. So, probably, X is not reducible to the natural.

Some options for X:

• ethics

• aesthetics

• value in general.

I think the best response would be to dispute (1), by saying that (1) is only plausible if we know how to do the reduction. The mere existence of a reduction, when we do not know how to run it, is not enough.

Maybe. But I still think we get some evidence against reductionistic theories in ethics, aesthetics and value in general from fact that great progress in science hasn’t led to great progress in these areas.

Gratuitous and objective evil

Suppose, highly controversially, that no defensible atheist account of objective value is possible. Now consider a paradigmatic apparently gratuitous horrendous evil E—say, one of the really awful things done to children described by Ivan in the Brothers Karamazov. The following two claims are both intuitive:

1. E is gratuitous

2. E is objectively evil.

But if there is no defensible account of objective evil on atheism, then (1) and (2) are in serious tension. For if there cannot be objective evil on atheism, then (2) cannot be true on atheism. Thus, (2) implies theism. But on the other hand, (1) implies atheism, since E gratuitous just in case if God existed, then E would be an evil that God has conclusive moral reason to prevent.

On our initial assumption about atheism, then, we need to choose between (1) and (2). And here there is no difficulty. That the things described by Ivan are objectively evil is way more clear than that God would have conclusive moral reason to prevent them, even if the latter claim is very likely in isolation.

Is a defensible atheist account of objective value possible? I used to think there was no special difficulty, but I’ve since come to be convinced that probably the only tenable account of objective value is an Aristotelian one based on form, and that human form requires something like a divine source. That said, even if objective value is something the atheist can defend, nonetheless knowledge of objective value is very difficult for the atheist. For objective value has to be (I know this is controversial) non-natural, and on atheism it is very difficult to explain how we could acquire the power to get in touch with non-natural aspects of reality.

But if knowledge of objective value is very difficult for the atheist, then we have tension between:

1. E is gratuitous

2. I know that E is objectively evil.

And (3) is still, I think, significantly more plausible than (1).

Naturalism and perdurance

According to one version of naturalism, the only objects that have causal influence are objects posited by a completed science.

According to perdurantism, changing persisting objects have temporal parts which have the changing properties more fundamentally.

The changing properties of objects are causally efficacious. Thus, if perdurantism is true, the temporal parts have causal influence.

But the temporal parts of changing persisting objects are not among the objects posited by our current science. For instance, our current physics holds electrons and quarks to be fundamental, i.e., not made up of other objects studied by physics. The temporal parts of electrons and quarks are thus not studied by physics. Yet they have causal influence. If the completed physics is relevantly similar to our current physics in this regard, it won’t include temporal parts, either. And hence perdurantism posits causally efficacious entities that are unlikely to be posited by a complete science.

Thus perdurantism does not appear to fit with naturalism understood as above.

Chinese Room thought experiments

Thought experiments like Searle’s Chinese Room are supposed to show that understanding and consciousness are not reducible to computation. For if they are, then a bored monolingual English-speaking clerk who moves around pieces of paper with Chinese letters letters—or photographic memories of them in his head—according to a fixed set of rules counts as understanding Chinese and having the consciousness that goes with that.

I used to find this an extremely convincing argument. But I am finding it less so over time. Anybody who thinks that computers could have understanding and consciousness will think that a computer can run two different simultaneous processes of understanding and consciousness sandboxed apart from one another. Neither process will have the understanding and consciousness of what is going on in the other process. And that’s very much what the functionalist should say about the Chinese Room. We have two processes running in the clerk’s head. One process is English-based and the other is a Chinese-based process running in an emulation layer. There is limited communication between the two, and hence understanding and consciousness do not leak between them.

If we accept the possibility of strong Artificial Intelligence, we have two choices of what to say about sandboxed intelligent processes running on the same hardware. We can say that there is one person with two centers of consciousness/understanding or that there are two persons each with one center. On the one person with two mental centers view, we can say that the clerk does understand Chinese and does have the corresponding consciousness, but that understanding is sandboxed away from the English-based processing, and in particular the clerk will not talk about it (much as in the computer case, we could imagine the two processes communicating with a user through different on-screen windows). On the two person view, we would say that the clerk does not understand Chinese, but that a new person comes into existence who does understand Chinese.

I am not saying that the proponent of strong AI is home free. I think both the one-person-two-centers and two-person views have problems. But these are problems that arise purely in the computer case, without any Chinese room kind of stuff going on.

The one-person-two-centers view of multiple intelligent processes running on one piece of hardware gives rise to insoluble questions of the unity of a piece of hardware. (If each process runs on a different processor core, do we count as having one piece of hardware or not? If not, what if they are constantly switching between cores? If yes, what if the separate the cores to separate pieces of silicon that are glued along an edge?) The two-persons view, on the other hand, is incompatible with animalism in our own case. Moreover, it ends up identifying persons with software processes, which leads to the unfortunate conclusion that when the processes are put to sleep, the persons temporarily cease to exist—and hence that we do not exist when sufficiently deeply asleep.

These are real problems, but no additional difficulty comes from the Chinese room case that I can see.

Requests and naturalism

If someone asks me to ϕ, typically that informs me that they want me to ϕ. But the normative effect of the request cannot be reduced to the normative effect of learning about the requester’s desires.

First, when you request that I ϕ, you also consent to my ϕ, and hence the request has the normative effects of consent. But one can want something done without consenting to it. For instance, if I have a lot of things on my plate, I might desire that a student give me their major paper late so that I don’t have to start grading yet, but that desire is very different in normative consequences from my agreeing to the lateness of the paper, much less my requesting that it be late.

Second, considerate people often have desires that they do not wish to impose on others. A request creates a special kind of moral reason, and hence imposes in a way that merely learning of a desire does not.

Moreover, we cannot understand requests apart from these moral normative effects. A request seems to be in part or whole defined as the kind of speech act that typically has such normative effects: the creating of a permission and of a reason. Moreover, that reason is a sui generis one: it is a reason-of-request, rather than a reason-of-desire, a reason-of-need, etc.

There is something rather impressive in this creation of reasons. A complete stranger has the power to come up to me and make me have a new moral reason just by asking a question, since a question is in part a request for an answer (and in part the creation of a context for the speech acts that would be constitute the answer). Typically, this reason is not conclusive, but it is still a real moral reason that imposes on me.

Consider the first time anybody ever requested anything. In requesting, they exercised their power to create a moral reason for their interlocutor. This was a power they already had, and the meaningfulness of the speech act of requesting must have already been in place. How? How could that speech act have already been defined, already understandable? The speech act was largely defined by the kinds of reasons it gives rise to. But the kinds of reasons it gave rise to were ones that had never previously existed! For before the first request there were no reasons-of-request. So the speech act had a meaningfulness without anybody ever having encountered the kinds of reasons that came from it.

This is deeply mysterious. It suggests an innate power of the human nature, a power to request and thereby create reasons. This power seems hard to reconcile with naturalism, though I do not have any knock-down argument here.

Monotheism and anthropomorphism

Xenophanes famously lambasted Greek religion for its anthropomorphism:

if cattle or lions had hands, so as to paint with their hands and produce works of art as men do, they would paint their gods and give them bodies in form like their own-horses like horses, cattle like cattle.

Two and a half millenia later, accusations of anthropomorphism continue to be made against monotheistic religions, typically by naturalists.

I was thinking about this, and had an odd thought. According to monotheism, the root of all explanation is the activity of God. According to standard naturalism, the root of all explanation is the activity of the fundamental physical entities, either particles or fields. But humans are more like fundamental physical entities than like the God of the monotheistic religions. The difference between us and the fundamental physical entities is merely finite. The difference between us and God is infinite. Thus, in an important sense, it is standard naturalism that is more anthropomorphic in its fundamental explanatory agents than monotheism.

If we do not feel this—if we feel ourselves more God-like than electron-like—then we are infinitely elevating ourselves or infinitely demoting God or both.

That said, the three Western monotheistic religions do think that the physical universe is made for us. Thus, while the religions are not anthropomorphic, they do have an anthropocentric view of our physical universe. Interestingly, though, to some (albeit lesser) extent so does the most plausible current naturalist view, namely a multiverse theory together with the weak anthropic principle.

Naturalism and lovability

1. If naturalism is true, Stalin is not lovable.

2. Everyone is lovable.

3. So, naturalism is not true.

Here, by “lovable”, I don’t mean that it is possible to love the person, but that it is not inappropriate to do so.

Premise 2 follows the intuition that it is permissible for every parent to love their children. It also follows from the more controversial claim that everyone should love everyone.

The intuition behind premise 1 is something like this: Stalin’s actions were so horrible that the only plausible hypotheses on which he is lovable are that there is some deeply mysterious and highly valuable metaphysical fact about his being, such as that he is in the image and likeness of God, or that his Atman is Brahman, a fact incompatible with naturalism. For if all we have are the ordinary naturalistic goods in Stalin, these goods are easily outweighed by the horrors of his wickedness.

A superpower

Imagine Alice claimed she could just see, with reliability, which unprovable large cardinal axioms are true. We would be initially sceptical of her claims, but we could imagine ways in which we could come to be convinced of her having such an ability. For instance, we might later be able to prove a lot of logical connections between these axioms (say that axiom A₁₂ implies axiom A₁₄) and then find that Alice’s oracular pronouncements matched these logical connections (she wouldn’t, for instance, affirm A₁₂ while denying A₁₄) to a degree that would be very hard to explain as just luck.

Suppose, then, that we have come to be convinced that Alice has the intuitive ability to just see which large cardinal axioms are true. This would be some sort of uncanny superpower. The existence of such a superpower would sit poorly with naturalism. An intuition like Ramanujan’s about the sums of series could be explained by naturalism—we could simply suppose that his brain unconsciously sketched proofs of various claims. But an intuition about large cardinal axioms wouldn’t be like that, since these axioms are not provable.

Now as far as we know, there is no one exactly like Alice who just has reliable intuitions about large cardinal axioms. But our confidence in the less abstruse axioms of Zermelo-Fraenkel set theory—intuitive axioms like the axiom of replacement—commits us to thinking that either we in general, or those most expert in the matter, are rather like Alice with respect to these less abstruse axioms. The less abstruse axioms are just as unprovable as the more abstruse ones that Alice could see. Therefore, it seems, if Alice’s reliable intuition provided an argument against naturalism, our own (or our experts’) intuition about the more ordinary axioms, an intuition which we take to be reliable, gives us an argument against naturalism. Seeing the axiom of replacement to be true is just as much a superpower as would be Alice’s seeing that, say, measurable cardinals exist (or that they do not exist).

An argument against naturalism from the concept of the numinous

1. If naturalism about our minds is true, then the correct account of intentionality is causal.

2. On a causal account of intentionality, our possession of an irreducible concept is caused by something which falls under that concept.

3. The concept of the numinous is irreducible.

4. Therefore, if naturalism about our minds is true, our possession of the concept of the numinous is caused by something numinous. (1–3)

5. If there is anything numinous, then naturalism in general is false.

6. If naturalism about our minds is not true, then naturalism in general is false.

7. So, naturalism in general is false. (4–6)

What do I mean by “the numinous”? Since I claim it to be irreducible, I had better not try to define it. But I can point to it by means of our experiences of the holy, the uncanny, etc.: see Rudolf Otto’s book on the holy.

I think the best objection to the argument is to say that numinous can be reduced to the negation of the natural. But that objection seems to me to be mistaken. Imagine some simple particle-like thing that doesn’t interact with anything else in a way that is governed by the laws of nature. That thing wouldn’t be numinous. Likewise, not all magic is numinous: quite a bit of the magic in the Harry Potter stories is not numinous at all (there is nothing numinous about the chocolate frogs).

From naturalism to divine command theory by way of vagueness

Plausibly:

1. There is a collection K of atoms forming an early stage of the developmental history of a particular human, where it is vague whether it is morally permissible to disperse the atoms in K.

Pro-choice readers might take K to be the collection of atoms in a late fetus or an early newborn. Pro-life readers might take K to be the collection of atoms at some early point during the fusion of gametes. (Note that (1) does not say that the atoms in K form a human.)

It is widely held that:

2. Vague matters always depend on semantic plasticity.

But:

3. Whether it is permissible to disperse the atoms in K does not depend on any semantic questions.

Claims (1)–(3) look incompatible. So we should abandon at least one of them. Personally, as a pro-life Aristotelian dualist, I am happy to abandon (1)—whether it is permissible to disperse the Ks depends on whether there is a human form uniting the Ks, and that’s not a vague matter. But apart from a Markosian-style brute fact view of composition—a view that seems implausible—it seems hard for a naturalist to deny (1).

Claim (2) is pretty plausible.

This means that if we accept (1), we should deny (3). But how to deny (3)? How could the question of whether it is permissible to disperse the atoms at some developmental stage depend on semantic questions? Surely it is a totally abhorrent idea that whether it is acceptable to disperse the atoms in K is determined by our linguistic performances?

Indeed! But it need not depend on the semantics of our linguistic performances. There is exactly one major moral theory that make that allows for a reasonable denial of (3): divine command theory. Commands are linguistic performances. Whether some act violates a command is in part a linguistic matter. Given divine command theory, semantic plasticity in the terms of the commands can ground vagueness in the obligations constituted by the commands. If your commander says: “Shave until you’re bald”, your obligation seems to be vague precisely because of the vagueness of language.

And that the semantics of God’s linguistic performances matters to the right to life of K is far from as objectionable as some claim on which our linguistic performances have the determining role. (Though our linguistic performances have some role to play, since they may help define the words that God is using if God speaks our language.) Thus, we can imagine that God says: “Thou shalt not murder.” But perhaps “murder” is vague. And the vagueness in this word then translates to vagueness as to whether the dispersal of the atoms in K is permissible.

If the above is right, then we have an argument for a very surprising thesis:

4. If physicalism (about us) is true, then divine command theory is the correct moral theory, and hence God exists.

This is yet another in a series of observations that I’ve been making over the years, that theistic naturalism has resources that its atheistic cousin lacks.

All that said, I think physicalism (even about us) is false, and divine command theory is not the correct moral theory. Though I do think God exists.

Quasi-divinization and love

When we deeply love someone, we are apt to raise them to a quasi-divine status in our hearts.

If naturalism is right, this is misguided, for the evolved clouds of particles that are the people we love do not in fact have any quasi-divine status.

If theism is right, then this quasi-divinization could well be appropriate: for persons participate in God in such a way that they are in God’s image and likeness. But although not necessarily misguided, the quasi-divinization is dangerous, lest it cross the line into idolatry. (See C. S. Lewis’s Four Loves.)

I think that the theistic outlook on the quasi-divinization in love better fits with the plausible observation that this kind of deep love is sometimes both laudable and yet still morally dangerous, while on the naturalistic outlook, it is merely misguided.

What if my kidney got smart?

Suppose the cells in my left kidney mutated, and the kidney grew neurons and started engaging in the same sorts of computations that my brain consciously does. (The idea is not as outlandish as it may seem. We will no doubt one day be able to make replacement kidneys in the lab. And if so, why not replacement kidneys with neurons?)

Question: Would I come to have a new mode of kidney-based consciousness on top of my brain-based consciousness?

I don’t know the answer to this as a genuine hypothetical question. But I have a strong intuition that there is no metaphysical guarantee that I would have a new mode of kidney-based consciousness. The mere fact that my kidney functions computationally like a brain doesn’t guarantee that I think with it.

It’s an interesting question which views about persons and mind can agree that there is no guarantee of my consciousness through kidneys.

Brainists, who think that we are brains, will happily agree. I think with my brain because I am my brain. The kidney would, perhaps, think, but it wouldn’t be me thinking, because the kidney isn’t even a part of me.

Dualists of all sorts can agree: for there is no guarantee that kidney-based computation gives rise to consciousness, since the connection between neural function and mental function on dualism can be contingent.

Some non-dualist animalists, however, will have a problem. For a non-dualist animalist will identify us with the animal, and then many of them will presumably want to say that the animal thinks provided that it has an organ that engages in certain kinds of neural behavior. But now it seems like I would have to be thinking through the kidney if it were to engage in this neural behavior.

But it’s not quite so simple. For it could be that the neural behavior that defines thought has a normative component. Thus, to think may require the neurons to appropriately engage in certain behaviors. But neurons in the kidney would not have proper function.

Thus, perhaps, the no-guarantee constraint only rules out one of the views I’ve considered: non-normative non-dualist animalism.

Deflating particular normative states of affairs?

Some actions lead to a future normative state of affairs. A promise or valid command leads to a future state of affairs of my or someone else’s being obliged to fulfill it. A wrongful action leads to a state of obligation to repent. And so on.

Here are two views of these states of affairs:

• Deflationary: There is nothing more to these states of affairs than general conditional moral normative facts, such as the facts that you should keep your promises, together with the fact of the triggering action, such as that you’ve promised to ϕ.

• Non-Deflationary and Causal: These states of affairs are metaphysically irreducible aspects of reality that are caused into existence by their triggering actions.

In fact of the deflationary view is that it’s deflationary, and hence supported by Ockham’s Razor.

But I think there are some reasons to accept the non-deflationary view. First, suppose you now come to the time where the normative state of affairs obtains: you must now fulfill your promise or you must now repent. Then the deflationary view implies an odd sort of “tyranny of the past”. What obliges you is not anything about the present, but something about the past. Your present obligations, on the deflationary view, do not supervene on the present state of the universe. This might especially bother presentists, but I think it’s also a bit worrying to eternalists like me.

Second, the “general conditional moral normative facts” the deflationary approach deals with will have to have extremely complex antecedents. For instance, for a command, there will be a fact of the form:

• If you were validly commanded to ϕ, and you have not yet fulfilled the command, and the command wasn’t changed by a higher authority, and circumstances have not relevantly changed, and …, then you should ϕ.

My worry about this is that there might be an infinite number of possible ways for a command obligation to disappear that would have to be put in the “…”. But perhaps not. Perhaps all I’ve said above is enough.

However, there is some reason not to be persuaded by this consideration. It is reasonable to think that human beings have normative powers: our actions can create reasons and obligations for ourselves and others. But one way for the obligation from a promise or command to disappear is for the non-normative circumstances to change. For instance, if I promised to do a minor errand, and a giant herd of yaks blocked my way, so that I could only do the errand via an unreasonably large detour, I might be off the hook normatively. But it seems implausible that a herd of yaks has the causal power to annihilate normative facts. So, it seems, even the non-deflationist may want the normative states of affairs to be conditional: “I should do the errand unless it becomes unreasonable.”

Third, the phenomenology of being released from an obligation—say, by being forgiven or a promisee’s releasing you—is an experience as of a load being removed. That “load” felt like a real thing which was annihilated.

Fourth, being forgiven changes your obligations by removing your guilt, at least assuming repentance. But it seems that God could forgive you a sin without announcing the forgiveness in any way, or in any other non-normative way changing the world. In such a case, God’s forgiveness would have a contingent normative effect. But God is simple, and hence all contingent facts about God are grounded in necessary truths about God and contingent facts about creation. But then if there is no non-normative change in creation due to the forgiveness, there must be a normative change in creation due to it.

On the non-deflationary causal view, divine forgiveness consists in God’s destroying the normative state of affairs of your being guilty. On the deflationary view, it’s got to be grounded in something like a divine “I forgive you” speech act, whether specific to your case (e.g., God telling you in your heart that you’ve been forgiven), or general (e.g., God’s announcing that anything the Apostles forgive is forgiven by God in John 20:23). But in the case of our forgiving someone the speech acts are announcements of a contingent state of affairs of forgiveness that goes beyond the announcements. That state of affairs, in the case of a simple God, cannot be internal to God. And it seems like it’s normative.

I find the last two considerations fairly powerful, but not conclusive. Of course, I accept divine simplicity, but the claim that God can forgive without any announcement isn’t completely obvious. Divine forgiveness could be like a Presidential pardon, which must be promulgated.

For us non-naturalists, it would be cool if we could argue for the non-deflationary view. For on this view, naturalism is false: we have causal powers that go beyond those described by the sciences, namely the causal power to produce normative states of affairs.

Update: Here's an argument in favor of deflation. While a particular obligation feels like a something ("a load"), what we cite as reasons for action is often not a resultant normative state but the original triggering action: "You promised!" or "I need to make it up to her given what I did." But on the causal non-deflationary view, the original triggering action is not even a part of the reason for action: it is, rather, a cause of the normative state, and the normative state itself is the reason. Of course, this isn't conclusive, because it could be that we mention the triggering action as evidence for the resultant reason. We likewise say: "I need go home because I left the kettle on." That I left the kettle on is no reason to go back home. That the kettle is still on is a reason to go home, and that I left the kettle on is evidence that that the kettle is still on.

Artifacts and non-naturalism

One of the reasons to be suspicious of artifacts is that it seems magical to think we have the power to create a new object just by thinking about things a certain way while manipulating stuff. If Bob gets some clay and exercise his fingers by randomly kneading it, he doesn’t make a sculpture or any other new object out of it. But if his identical twin Carl intends to shape the clay into a sculpture, and in doing so moves his fingers in exactly the same way that Bob did, and produces exactly the same shape, then—assuming artifacts exist—he creates a new object, a sculpture. It seems magical that our thoughts should affect what object exists in the world, even when the thoughts make no difference to our manipulation of the world.

When I discussed arguments with this in my Mid-Sized Objects graduate seminar, I found, however, that there was a lot of friendliness towards the view that, yes, we are capable of this magic, though some demurred at the word “magic”. And in particular, a student pointed out that we are in the image of a God who can create.

This has made me think that a non-naturalist can think that our thoughts have effects that are not screened by the movements of our bodies. Thus, it could well be that Carl’s thoughts causes the world to be different. For instance, on a hylomorphic view, Carl could have the power to create a scu;tural form for a piece of clay by his thoughts. Or on a variant of Markosian’s brute composition view, Carl could have the power simply to cause a new object composed of the clay.

In fact, this suggests an interesting new argument against physicalism, where physicalism is understood as the claim that all causal powers reduce to those of physics. Intuitively, the correct ontology includes more things than van Inwagen’s ontology of particles and organisms and but not all the things from the mereological universalist’s bloated ontology. In particular, intuitively, the correct ontology does include Carl’s new sculpture, but Bob hasn’t produced anything new, and hence the correct ontology seems to require a non-natural “magical” power over composition facts to be found in Carl’s (and presumably, albeit in this context unexercised, Bob’s) mind. And if our ontology is to include, as common-sense would suggest, galaxies, planets, mountains and rocks, we need powers in things to produce such objects—i.e., to ensure that their particulate parts do compose something—and these powers are not to be found in physics.

Markosian’s apparently preferred version of the brute composition view can almost accommodate this. On that version, the composition facts supervene on the arrangement of particles: there are infinitely many necessary truths that specify which arrangements of particles compose. But these necessary truths would include lots of arbitrary parameters (e.g., encoding the difference between some stones that are just lying there and a hillock). We don’t want necessary truths with arbitrary parameters. It is much better if any such arbitrary parameters are relocated to the laws of nature or, better, the causal powers of things.