The Spirit Concept Argument

Moreland's argument goes like this: even supposing there are no thinking immaterial substances (spirits, let us say), we still grasp what they might be. We have the concept of a spirit. Further, if have this concept, then it is not so much as possible that we are thinking material substances. For, given that we do indeed have the concept of a spirit, it is possible that we are material substances only if we have the resources to distinguish properly the concept of a spirit from that of a thinking material substance; and there are none. So we are not thinking material substances. We are, instead, spirits. That's the quick version. Here's a more careful presentation:

1. 1. If someone understands what it is for something to be an entity (or purported entity), then she has an adequate concept of that thing.

2. 2. If someone has an adequate concept of a thing, then she has a distinct positive concept of that thing.

3. 3. Therefore, if someone understands what it is for something to be an entity (or purported entity) – a divine spirit, for example – then she has a distinct positive concept of that thing (from 1 and 2).

4. 4. We understand what it is for God to be a divine spirit.

5. 5. Therefore, we have a distinct positive concept of God's being a divine spirit (from 3 and 4).

6. 6. If thinking matter is metaphysically possible, it is not the case that we have a distinct positive concept of God's being a divine spirit.

7. 7. Therefore, it is not the case that thinking matter is metaphysically possible (from 5 and 6).

8. 8. We are either thinking material beings or thinking spirits.

9. 9. Therefore, we are thinking spirits (from 7 and 8).Footnote ²

The argument crucially deploys the concept of a spirit; so let us call this it Spirit Concept Argument. Before looking at some details, I make five observations.

First, the Spirit Concept Argument is formally valid. If its premises are true, so too is its conclusion.

Second, though the argument as stated makes use of ‘God’, it is not in any relevant sense a ‘religious’ argument. It does not, for example, require trust in any sacred text or tradition; nor does it require the actual or even possible truth of theism. What it requires, instead, is that we have the conceptual resources to understand what it might be for there to be immaterial thinking things. Theists, of course, have special reason to affirm this, but many others will join them as well.

Third, the Spirit Concept Argument deploys concept-talk. If you are, like me, a little distrusting of concept talk (I'm not sure I know what concepts are supposed to be), you may be leery. But in this case, a permissive attitude is in order, for the argument can be translated into other vocabulary. Instead of talking about ‘having a concept’, for example, we might instead talk about ‘competently using a word’, ‘understanding the satisfaction conditions of an open sentence’, ‘knowing what property a word expresses’, and the like. So even though the argument at hand is unashamedly conceptual, it needn't assume that guise.

Fourth, the Spirit Concept Argument can, at first glance, seem to be a bit of a magic trick. It appears to pull a thick thesis about what we are from a remarkably thin conceptual hat. I think it is more than just a trick, though, in at least this sense: reflection on the argument is metaphysically instructive. Just as tricky ontological arguments invite research on the nature of modality, so too the Spirit Concept Argument invites excursion into the jungle of ontological categories and their relation to our basic concepts. Even should the argument be unsound, then, it might still prove worthy of our time and cognitive resources.

Fifth, the Spirit Concept Argument is significant in its own right and on a number of dimensions. It thus deserves careful attention. Here's why. The argument is wide in its intended audience. It includes as a target anyone who thinks they have a concept of a spirit. Those who think there is a god and those who think there is no god (all of whom would have a concept of a spirit, I suppose) have reason to pay attention. The argument is ambitious in its conclusion, claiming nothing less than that we are wholly immaterial thinking substances, giving us further reason to pay attention. It is surprisingly modest in its premises, beginning from seemingly innocuous ideas about concepts and their place in our intellectual economy. Each premise is, I think, individually and initially plausible and appealing. And yet together the premises imply a substantive metaphysical result – a sure sign that a closer look is in order. The Spirit Concept Argument is, finally, timely, for it promises to undermine various theistic forms of materialism according to which, though there are immaterial spirits (gods, angels, and the like), we are wholly material beings. If the Spirit Concept Argument succeeds, these increasingly popular forms of materialism are mistaken and no theist should endorse materialism about human persons.Footnote ³

Despite the above remarks, some may still doubt the utility of examining the Spirit Concept Argument, favouring instead a cursory dismissal and an even more cursory assessment of the mental capacities of dualists.Footnote ⁴ This reaction (however tempting it may be) is not the way of true philosophy. Scorn is no replacement for reasoned argument. The thing to do here is not to close the ears or to make fun of the substance dualists when they argue for their position. It is, instead, to show where their arguments go wrong. This is especially the case when the dualist has (like Moreland) been so kind as to offer a formally valid argument with numbered premises and conclusions. The least we materialists can do is pick a premise or two and say why they are false; and that is my task in this note.

Thus the Spirit Concept Argument in broad outline, a few introductory remarks, and a rant for good measure. It's time for the details. In the sequel, I'm going to focus on the crucial second and sixth premises.

Against the Positive Requirement

The second premise imposes a strict requirement; a concept is adequate only if it is positive, says the premise. Let's call this the Positive Requirement. Moreland's case for the Positive Requirement opens:

Moreland does not offer an account of what it is for a concept to be positive (there will, no doubt, be difficult cases involving disjunctive and conjunctive concepts). Nor does he elaborate on what adequacy might here amount to (‘adequate for what purpose?’ one wonders). Some might raise a fuss here and turn these nagging worries into objections. I'm willing to set these details aside, though, since I think the premise is false in broad outline (and not just in detail).

Here's why. We have an adequate concept of the abstract. We possess the concept of abstractness. We use it in theorizing and appear to do so with full competence. We can reliably classify items as either abstract or concrete (concrete and abstract divide reality exclusively and exhaustively, I shall suppose). People and planets are concrete, while numbers and propositions are abstract. And yet there is very little we can say to explain what abstractness comes to.Footnote ⁶ We can say only what it is not. So, for example, if concreteness is a matter of having causal powers, abstractness is a matter of lacking them. Similarly, if concreteness is a matter of being in space and time, abstractness is a matter of not being within space and time. When it comes to the abstract and the concrete, other roads are closed, and so we travel the via negativa. There is, then, a convincing counterexample to the Positive Requirement, for we have an adequate concept of the abstract, but it is not a positive concept.

Moreland correctly anticipates that there will be trouble for the Positive Requirement in this neighbourhood; he suggests ‘real/unreal, true/false, good/evil, male/female’ as potential problem cases. In reply, he distinguishes two varieties of concept pairs. In the first variety (priority pairs, let us say), one concept is more ‘conceptually basic and the other is a privation of some sort or another (real/unreal, true/false, good/evil)’.Footnote ⁷ In the second variety (egalitarian pairs, let us say), neither concept is more basic, nor is either a privation of the other. One interesting (and perhaps distinguishing) feature of egalitarian pairs is this: each member of the pair has its own positive concept.

Moreland then argues that spirit/matter is not a priority pair. It is not as though the concept of a spirit is merely the concept of non-matter, for ‘the number two satisfies this privation condition [non-matter] but it is not a spirit’.Footnote ⁸ So spirit/matter must, instead, be an egalitarian pair, and, accordingly, ‘there needs to be a distinct positive concept of matter and spirit for there to be an adequate concept of each’.Footnote ⁹

In reply, I note that the conclusion Moreland appears to draw here (that spirit/matter is an egalitarian pair) does not follow from the premise (that the concept of a spirit is not merely the concept of non-matter). Here is what does follow: spirit and matter are not complements; they do not exhaustively and exclusively chop up reality. For it is possible for something to be non-matter without being a spirit (this is what number two example illustrates; other examples might include other abstracta – possible worlds, sets, propositions, and the like). Moreland's reasoning here, then, does not support the thesis that spirit/matter is an egalitarian pair; it thus does not support the thesis that ‘there needs to be a distinct positive concept of matter and spirit for there to be an adequate concept of each’.

Categorial ontology and the Positive Requirement

I have offered a counterexample to the Positive Requirement and in doing so have offered a rebutting defeater to one premise of the Spirit Concept Argument. I have also criticized Moreland's reasoning for and defence of that premise and so have offered an undercutting defeater to that premise. Those are my tree-remarks. Now for some forest-remarks. These assume the shape of a theory of categories that vindicates my basic response to the Positive Requirement and a perfectly general diagnosis of its falsity.

To give a theory of categories is do ontology in the grandest and most ambitious sense. It is to say, in the broadest possible terms, what the joints of reality come to.Footnote ¹⁰ Specifying what something is, furthermore, involves locating it within whatever categories there are. Consider, then, a theory of categories in the form of this chart:

It is not always easy to read truth-evaluable theses off of a chart. Luckily, the present case presents few difficulties. According to the theory charted above, for example, every item is a thing (there is no non-thing category, and no category higher than ‘thing’). Similarly, every thing is either abstract or concrete, and every concretum is either material or immaterial. Both material and immaterial things, furthermore, cleave into thinking and unthinking categories. We could also offer examples or alleged examples of items within each of the four base-level categories (conscious organisms as thinking material things; rocks as unthinking material things; gods and demons as thinking immaterial things; bare particulars as unthinking immaterial things).

The ontology encoded in the above chart – the Chart Ontology, if you like – is coherent. We have a grip on the concepts involved, and we can say something about what the Chart Ontology does and does not imply. The Chart Ontology is amenable to theism in at least this sense, for example: it allows that there could be immaterial thinking things (spirits). There is, in the Chart Ontology, room for such items. The Chart Ontology does not, furthermore, settle any questions about our nature. It does not say where we belong on the chart, and in particular does not say whether we are thinking immaterial things or thinking material things. But if you wanted to say what we were, using the vocabulary of the Chart Ontology, it's plain how you'd do that: point to a node, and say ‘that's our place in this world’.

The Chart Ontology is coherent. It may not be true. But it is coherent. And to talk about the Chart Ontology is obviously to do ontology (a point that will prove important below). To argue about whether the Chart Ontology is true is to argue about what, in the broadest possible terms, the joints of reality come to.

Reflection on the Chart Ontology can illuminate the Positive Requirement's failure. Here's how. Every juncture within the Chart Ontology involves a concept and its complement. The highest juncture involves the concept concrete and its complement, abstract. The second-highest junctures involve the concept material and its complement, immaterial. And the third-highest junctures involve the concept of thinking and not-thinking. These are all perfectly serviceable concepts, so far as I can tell; they are adequate. The Chart Ontology is, after all, coherent. But half of them appear to be negative. So there cannot be any general requirement that adequate concepts be positive. Consider this reply:

The reply is tempting. But not for students of the Chart Ontology – or anyone who recognizes what the Chart ontologists are up so. For students of the Chart Ontology are engaged in the serious business of carving up reality. They are saying what things there are. They are saying what things are. And yet they freely deploy negative concepts like not-material in doing all that. My claim, again, is not that the Chart Ontology is true. Rather, I claim only that it is coherent, and that its proponents are plainly engaged in categorial ontology.Footnote ¹¹

I conclude, then, that there is no requirement – whether in sound thinking or in the abstruse art of ontological categorizing – to deploy only positive concepts. The Positive Requirement fails, and so too does the Spirit Concept Argument.

Against Key Link

We now have the resources to appreciate another problem with the Spirit Concept Argument. According to the sixth premise, there is a key link between what concepts we have and what is possible. It says, recall:

Let's call this Key Link. There are plenty of complaints one might direct against Key Link. There may be, for example, no general bridge between our conceptual resources and modality (perhaps reality is not so respectful of our concepts as we might wish). But I do not propose to advance that kind of objection; I will not attempt to unlink mind and modality. Instead, I'll show how the reply I've developed to the Positive Requirement applies here with equal force. So even if the Positive Requirement could be shored up or restricted to avoid my criticism, the Spirit Concept Argument would nonetheless be vulnerable to attack.

Before criticizing Key Link, let me offer a brief word on its behalf in the form of a brief speech. I'll put things in my own words, but they will express, I hope, Moreland's own case for Key Link:

There is something appealing about the basic idea here, I think. If some kinds are indeed real (not made-up or gerrymandered) and distinct, then they must have distinct natures. The speech above proposes that the distinctness at hand must involve non-overlap. This non-overlap condition is the core of Moreland's case for Key Link. Here's how he puts things:

But if what I have argued above is true, this just isn't so. We can use mental concepts to characterize the nature of a spirit, even if those properties could be had by both material and immaterial beings. Indeed, we can do so in our most serious and metaphysical moods, even when doing old-fashioned categorial ontology. To demonstrate my claim here, I need only point again to the Chart Ontology. Can the Chart ontologist say what it is to be a material thinking thing? With ease. Recall that, according to the Chart ontologist (and categorial ontologists in general), saying what something is involves pointing to a node within the chart. To say what a spirit is, then, the Chart ontologist need only point to the proper node in the chart (in this case, the thinking immaterial thing node). To be a spirit is to fall within that node in the grand hierarchy of being; it is to be a thinking material concretum. That mental properties characterize spirits as well as material thinking things is no more a problem than that immaterial characterizes both some abstracta (possible worlds, say) and some concreta (gods and demons, say).

Coda

In sum, I have shown that two premises of the Spirit Concept Argument are false. The argument does not support substance dualism. The offending premises, furthermore, do not merely falter in the presence of some philosopher's trick; they are false for quite abstract and general reasons.

Substance dualism is on the move, but I remain unmoved by this argument on its behalf.Footnote ¹³