John Duns Scotus: Hylomorphism and Pluralism about Substantial Forms

        Famously, Duns Scotus, like most medieval scholastic theologians following after Thomas Aquinas, argue that for an individual animal substance to subsist, such a substance requires two substantial forms -- a bodily form (for the body) and an animating form (its soul). The living animal that results from this combination will thereby have two substantial (i.e., essential) forms. This proposal, however, is open to the following objection, developed some years earlier by Aquinas. If the point of positing a substantial form in an individual substance is to explain the *unity* of that substance, i.e., what gathers together and shapes its matter into numerically one individual of a specific kind, it will now be (at best) difficult to give an adequate account for why one should regard the combination of the individual animals matter + its first form + its second form as a *single* substance. So, in the absence of such an explanatory principle, why some combinations of form/matter entities form substances and others do not will remain (at best) mysterious.
       In an attempt to answer this objection, Scotus suggests that a material substance is a unity if and only if the it has at least one property that its various parts or constituents lack. In consequence, Scotus concludes, following Aristotle, that acts of sensation require the right sort of body (i.e, matter + bodily form of the right type) and soul, the union of which allows the processing of sense data. Scotus also argues that the soul alone is insufficient for being alive; that is, he thinks animal life requires both a body and a soul, such that, when the two are combined, the composite that they form is itself alive.
        But Scotus' principle, I submit, is open to objection. For on the present criterion, it seems a heap of rocks will count as a substantial unity. (Recall, once more, that for Scotus, a material substance counts as a unity, as one *thing*, if and only if it has at least one property that none of its parts alone could have.) For surely, one might argue, the heap has the property 'being a heap' -- a property that none of the rocks taken individually are said to have. My instinct here is that Scotus' Aristotelian hylomorphism would not permit him to countenance such a consequence.
        But in case anyone has doubts, one could simply pile on a piece of wood to the heap, and, according to Scotus' criterion, this heap + the piece of wood would seem to count as a substantial unity (whether it would still count as a 'heap' or not might be open to question -- but that it counted as an assortment of entities that, on Scotus' definition, collectively make up a unified thing, seems beyond question). But this seems clearly wrong. For even if one insisted on calling the heap a 'thing', it would be far-fetched to call it a *unified* thing. It is certainly not unified in the sense in which a living organism, say, is unified.
        Nor would it do to argue that a heap is not a single thing (even though taking such a position is something that Aristotelians in general would also seem to affirm) but merely an aggregate. For the very thing one wants to know is what makes it true that a given aggregate or assortment of entities make up a substantial unity and not a mere heap. That is, why, on Scotus' definition, should one count [one form + another form + a certain hunk of prime matter] as making up a unity, a unified thing, while not counting a rock heap as such a unity? 
       But perhaps the defender of Scotus' definition will deny that there is such a property as 'being a heap.' Perhaps instead such a critic will argue that a heap comes into being from the merely extrinsic properties that all the rocks in the heap collectively share. Here, of course, if one insists that the properties in question are extrinsic, one can consistently deny that the rocks in the heap constitute a single, unified entity. But by 'extrinsic,' one had better not mean merely 'relational.'
       For clearly, it seems, the components or parts of *any* substantial unity that includes any two constitutive entities will be such that its components, when joined, relate to each other in such a way that is necessary for their composing the substantial unity in question. For example, the relation that an animal's body bears to its soul is one of such a constitutive unity -- and an animal exists only if these two constituents are related to each other in just the right way. And clearly this such relationship does not and should not count as extrinsic; for, if the two entities are related only by extrinsic properties, it is very difficult to see how they could possibly constitute or make up a single substantial unity.
        So, here, Scotus will want to insist that the relations that obtain between the rocks in the heap are merely extrinsic, while those between, say, an animal's body and its soul is not. But, one can I think safely conclude, two or more entities make up a substantial unity only if they relate to a third entity -- i.e., the *thing* or individual substance which they collectively make up or constitute. (For, if one of these properties is extrinsic, it is again difficult to see how it could be included among the features or components of a single substantial unity.) And moreover: all these relational properties such entities share are intrinsic only if there is some property which belongs to the whole -- and for which these components are parts -- and which does not belong to any of the components taken individually.
        But which relations count as intrinsic and which count as extrinsic depends on one's already having sharply drawn boundaries or criteria for which groupings of entities count as unified substances and which do not. In particular, such groupings will be determined by either (a) intrinsic-only, (b) extrinsic-only, or perhaps (c) partly intrinsic and partly extrinsic sets of relational properties. (That is, for (c), there will be some groupings that include an arbitrary number n of entities such that some of these entities are related to each other intrinsically and others are related to each other extrinsically.) Here, only the groupings that fall under (a) will count as substantial unities or unified substances. But here is the difficulty: Scotus' definition gives no criteria by which to determine which groupings fall under (a) as opposed to either (b) or (c).
        And clearly, I think, it will not do to protest that only *things* can be substances. For if we have no criteria for what counts as a substantial unity, we have no criteria for what counts as a substance. And if something is a thing only if it is a substance, a claim which Scotus certainly accepts, it follows that without criteria for what counts as a substantial unity, we have no criteria for what counts as a thing. Hence, one has good grounds for rejecting that definition and continuing to press the charge that positing more than one substantial form in an individual substance makes it difficult to explain that substance's intrinsic unity.