I. Z'S ONTOLOGY. MAKING A CASE FOR FORM

In Met. Z Aristotle never denies that ordinary particular objects are substances. On the contrary, as early as in Z 2, he clearly says that particular sensible substances are the things everyone is prepared to call substances and never seems to reconsider his view on this particular point.Footnote ⁶ However, one of Z's most characteristic claims is that the form of sensible substances, which are in the new setting analysed as composites of matter and form, is substance as well. Sometimes, form is even called “primary substance”.Footnote ⁷ Thus, the main problem in interpreting Z's ontology is to understand how the substantiality of form relates to that of the sensible object of which it is the form.

The claim that form is substance is not as simple as it might seem at a first glance. It seems to contain two distinct claims, which I shall label (C1) and (C2):

1. (C1): Form is the substance of particular sensible objects (i.e. composites of matter and form)

2. (C2): Form is a substance in its own right

That form is the substance of particular sensible objects is a claim Aristotle states often times in the course of Z's argument.Footnote ⁸ The meaning of this claim is that form is what explains why a sensible object is a substance, and so can be called the substance of the particular object whose substantiality it explains. If we ask why a particular sensible substance is a substance, the answer is that it possesses a form of a certain kind conferring upon the sensible object itself the essential characters it possesses. Among the different reasons why we should take the form of a sensible object to be its substance the most important is that form is the essence of a sensible object.Footnote ⁹ In other words, what makes of a particular substance the kind of thing it is, is its form and not its matter. It is form that confers upon a particular sensible substance the essential characters by which we classify it as a substance of a certain kind. Although this is highly controversial, one logical consequence of Aristotle's claim should also be that when we define a certain kind of sensible substance we should mention only its formal characteristics leaving out any reference to the material ones.Footnote ¹⁰ For a definition is supposed to perfectly reflect the content of an essence. Thus, form is clearly the substance of sensible objects by being their essence, i.e. by being the factor or principle which explains why they are the kind of objects they are.

(C1), therefore, is an authentic piece of Aristotelian doctrine no interpreter can seriously have doubts about. One of the main interpretative problems, by contrast, with regard to Z's ontology is whether endorsing (C1) allows us to endorse (C2) as well. (C2) is the claim that form is not only the substance of a sensible object, but should also be considered a substance on a par with the sensible object of which it is the form. Thus, to pass from (C1) to (C2) means to say that the fact that form is the substance of a sensible object allows one to conclude that form is a substance in exactly the same sense as we say that the sensible object possessing form is a substance.

It is not surprising, therefore, that it is with regard to the move from (C1) to (C2) that the different interpretations of Z's ontology are at variance. Most interpreters, for instance, grant the move from (C1) to (C2) and consequently believe that form is not only the substance and essence of sensible objects but is also a substance in exactly the same sense of “substance” as sensible objects are called substances. What is more, they also insist that, being the substance and essence of sensible objects, form must be more substance than the sensible objects of which it is the substance and essence. But how can they justify the conclusion that form is substance in the same sense as sensible objects and even more of a substance than them? The best strategy is to put emphasis on an explanatory criterion of substantiality. In other words, the idea is that, since form explains why a sensible object is the kind of substance it is, form itself must be more of a substance than the sensible objects whose substantiality it explains. Furthermore, if form must really explain the substantiality of a sensible object, it must be substance in the same sense of “substance” as the sensible object itself. Thus, on this interpretative line, the results of Z's enquiry mark a major departure from the ontology in the Categories. For form is not only a substance in the same sense as particular sensible objects, but is also more substance than them in that it explains their being substances. Thus, in some sense form simply replaces particular sensible objects in the role of primary substance the Categories assigned to them. Let me call this first interpretative strategy “incompatibilism” in that it maintains that Z's ontology supersedes the one Aristotle proposes in the Categories and so is incompatible with it.Footnote ¹¹

Other interpreters resist the view that form is a substance in the same sense as we say that a particular sensible object is a substance and so do not license the move from (C1) to (C2). Accordingly, they maintain that we should sharply distinguish between the notion of “substance” in the absolute sense and that of “substance of”. In other words, on their view form is substance only in so far as it is the substance of particular sensible objects, i.e. the principle explaining their fundamental characters, but is not a substance in the same sense as the particulars of which it is the form, i.e. is not an object alongside particular sensible things. Supporters of this line of interpretation do not deny that the explanatory criterion of substantiality plays an important role within Z's enquiry. But they insist that the explanatory criterion only governs the notion of “substance of” and not also that of “substance” in the absolute sense. The substance of particular sensible objects is in fact the principle explaining why such objects are the kind of things they are, but this does not mean that such a principle is a substance on a par with the objects it contributes to explain. The notion of “substance” in the absolute sense is not governed by the explanatory criterion, but rather by the criterion of complete and independent existence. Only those things which are independent and full-fledged members of a certain kind count as substances in the absolute sense. Thus, particular sensible objects are substances in the absolute sense, whilst their forms are not substances in this sense, in that they are not complete and independent objects, but only principles of independent objects. Clearly, this second line of interpretation presupposes that Z's ontology fundamentally agrees with the Categories with regard to the notion of substance in the absolute sense. Particular sensible objects remain, even within Z's framework, primary substances when “substance” is taken in the absolute sense of the term. What Z does is to introduce a further level of discourse, namely the explanatory or causal level. On this new level, form turns out to be what explains the substantiality of particular sensible objects, and so is entitled to be called the substance of those objects. However, the explanatory level does not enter into conflict with the level of independent and complete existence. For it is at the level of independent and complete existence that we are supposed to evaluate the different candidates' claims to the title of primary substance in the absolute sense. Therefore, since it defends a conciliatory interpretation of Aristotle's ontology with regards to the relation between the Categories and Met. Z, this second line of interpretation can be rightly labelled “compatibilism”.Footnote ¹²

Aristotle's test itself is not of much help in adjudicating between the two competing interpretations of Z's ontology, in that it seems to present different and sometimes conflicting criteria of substantiality. What is more, Aristotle's discussion of the different criteria of substantiality does not allow us to decide whether he distinguishes in any significant way between the notion of “substance” in the absolute sense and that of “substance of”. In the Categories substantiality is evaluated on the basis of the so-called subject criterion. Particular sensible objects are primary substances because they are ultimate subjects, i.e. because everything else is predicated of them and there is nothing of which they themselves are predicated.Footnote ¹³ “Predication” should be taken here as an umbrella term, including both the relation of inherence and that of essential predication. Thus, in the Categories accidents (both particular and universal) “are in”, i.e. inhere in, primary substances, whilst natural kinds (species and genera) “are said of” primary substances, i.e. express what primary substances essentially are.Footnote ¹⁴ Both the “being in” and the “said of” relation count as cases of predication, which should be clearly taken in this instance as an eminently ontological notion, even though it may have and as a matter of fact has some linguistic counterpart.

Now it is clear that in the central books of the Metaphysics the subject criterion is no longer regarded as the main criterion of substantiality. The reason is explained by Aristotle in Met. Z 3. The subject criterion of substantiality points to the wrong candidate for the title of primary substance, i.e. matter.Footnote ¹⁵ Aristotle does not tell us whether the conclusion that matter is primary substance is the result of a correct or of an incorrect application of the criterion and hence we do not know for certain whether he really believes in the argument leading up to the conclusion – the famous “stripping-away argument”.Footnote ¹⁶ It is not my intention here to go into all the details of the stripping-away argument, but the main idea seems to be the following.Footnote ¹⁷ Matter might be thought to be the ultimate subject of predication because it is what is left when all the attributes of a sensible substance are stripped away. Accidents in fact are predicated of particular sensible objects, which are composites of matter and form. And form, which is responsible for the particular object's essential features, is in turn predicated of matter.Footnote ¹⁸ Thus, if we remove from a sensible substance all the items which count as predicates – accidents and form – the only item we are left with is matter, which is not predicated of anything further. In other words, if we endorse the subject criterion of substantiality – or at least if we push to the extreme one of its possible applications – we end up regarding matter as the ultimate subject of predication and hence as primary substance. However, although Aristotle assigns some degree of substantiality to matter,Footnote ¹⁹ he does not think that matter should be taken to be substance to a higher degree than form or the composite of matter and form. The outcome of the discussion, therefore, is that the subject criterion of substantiality should be abandoned in favour of – or at least integrated by – some other criteria. But which ones? At the end of Met. Z 3 Aristotle says that matter does not qualify as a primary substance because it does not satisfy two important requirements for something to be a substance in the strict sense of the term, namely being a τóδε τι and being separable (χωριστόν).Footnote ²⁰ Thus, one natural suggestion is that being a τóδε τι and being separable are Aristotle's criteria of substantiality in Met. Z. Let me explore, briefly, this suggestion.

Aristotle says in Met. Z 3 that both form and the composite are τóδε τι and separable. However, when summing up the results of Z's enquiry in Met. H 1, he suggests that form and the composite are τóδε τι and separable in two different senses.Footnote ²¹ Take for instance the notion of being separable.Footnote ²² The composite of matter and form – Aristotle says – is separable in an absolute sense (ἁπλῶς).Footnote ²³ This means, presumably, that the composite of matter and form enjoys an autonomous existence and hence is separate, i.e. ontologically independent, from all the other autonomous and independent objects. Form, by contrast, is not separable in an absolute sense, but only “in account” (τῷ λóγῷ).Footnote ²⁴ What Aristotle means by this is that form can be taken apart in thought from all the other constituents of a sensible object – and in particular from matter – and so be defined without making reference to any of them. As can be seen, the sense of being separable according to which form is separable is quite different from that in which the composite is so. It can be argued – although this point is admittedly more controversial – that form and the composite are also τóδε τι in two different senses.Footnote ²⁵ The composite is a τóδε τι by being an individual or particular object. After all, this is the sense that the expression τóδε τι already carries in the Categories, where particular objects are described as τóδε τι and contrasted with universals, which are described as πoιóν τι.Footnote ²⁶ Now, although some interpreters contend that form as well is a τóδε τι by being a particular object, I side with those who think that form is not a particular object, but rather a principle which can be replicated in and shared by different objects.Footnote ²⁷ Thus, if form is a τóδε τι, it must be so not in the sense of being a particular but in some other sense. One reasonable suggestion is that form is a τóδε τι by being a fully determinate thing, i.e. something which cannot be further determined from a formal point of view. And being fully determinate is compatible with not being a particular, but rather some kind of sharable and replicable entity.

The fact that form and the composite are τóδε τι and separable in two different senses may invite the conclusion that Aristotle sees in fact a difference between the notion of “substance” in the absolute sense and that of “substance of”. For instance: the way in which the composite of matter and form is separable and τóδε τι seems to agree with its being a substance in the absolute sense. For the composite is separable by being an autonomous and independent entity and is τóδε τι by being a particular being of a certain kind – which both things clearly suggest that the composite satisfies the independent existence criterion and so is a substance in the absolute sense. Form, by contrast, is separable only in account, and is a τóδε τι not by being a particular thing of a certain kind but rather a structural principle which is replicable in and sharable by many objects. This might suggest that form is not a substance in the same sense as the composite, but is rather the substance of the composite, namely the principle that explains why the composite is the kind of substance (in the absolute sense) it is.

However, the conclusion that Aristotle neatly distinguishes between the notion of “substance” and that of “substance of” seems to be in conflict with the importance he accords to the explanatory criterion throughout Met. Z. In Met. Z 17, for instance, Aristotle explores the idea that substance is a principle and a cause of some sort.Footnote ²⁸ The qualification “of some sort” (τις) is important, for Aristotle's main point is that substance is a principle and a cause when by “a principle and a cause” we mean “a formal principle and cause”. In some cases, also the end and the agent can count as the substantial principles of a sensible object, but only because in such cases they happen to coincide with the formal cause.Footnote ²⁹ In other words, it is the form of a sensible object that is substance in so far as it is the principle and cause of the object's being what it is.Footnote ³⁰ Being a cause and a principle means to perform an explanatory function. More specifically, Aristotle's main idea in Z 17 is that form is substance because it is what explains why a certain piece of matter is an object of a certain kind.Footnote ³¹ It is because a certain pack of flesh and bones possesses a certain kind of form, i.e. the human form, that it can be called a human being. In other words, form is an explanatory principle by being the essence of a sensible object, i.e. what explains why the object in question possesses the characteristics which distinguish it from objects of different kinds. Now, the important thing about Z 17's argument is that Aristotle by no means confines its conclusions to the notion of “substance of”. On the contrary, he presents Met. Z 17's investigation into the notion of substance as cause as a further (and perhaps the final) step of his enquiry into the general notion of substance.Footnote ³² Thus, one can reasonably argue that the investigation into the notion of substance as cause concerns the notion of substance in the absolute sense or, more precisely, that Aristotle does not intend to distinguish between the notion of “substance” in the absolute sense and that of “substance of”. If this is the case, the fact that form turns out to be substance according to Z 17's argument should be read as the thesis that form is substance in the same, absolute sense as the composite of which it is the form. What is more, since form explains why a certain kind of composite is the kind of substance it is, form should be considered to be more substance than the composite.

In conclusion, Met. Z's argument does not provide decisive evidence in favour of either of the two general interpretations of Aristotle's ontology I have just presented. Some things Aristotle says may suggest that he distinguishes between “substance” in the absolute sense and “substance of” and evaluates substantiality in the absolute sense on the basis of the independent existence criterion. However, the bulk of Z's argument speaks in favour of the view that he does not distinguish between the two notions of substance and takes the explanatory criterion as the most appropriate criterion to establish what is substance in the strict sense of the term. Given the ambiguities of Aristotle's text, it is not surprising that medieval as well as modern interpreters have come up with very different readings of Aristotle's ontology and have put emphasis on one or another of the different criteria of substantiality he introduces in the course of Met. Z's argument.

II. AVERROES AND THE EXPLANATORY CRITERION OF SUBSTANTIALITY

1. According to Averroes, the aim of Aristotle's enquiry in Met. Z is to investigate into the causes and principles of sensible substances, i.e. of the particular objects of our ordinary perceptual experience – the objects which the Categories describes as primary substances and which are presented in the Metaphysics as composites of matter and form.Footnote ³³ Thus, for Averroes Met. Z starts from where the Categories left off and proposes a new sort of investigation, which mainly focuses on the notion of cause or explanation. Met. Z's causal investigation concerns the substance of particular sensible objects, i.e. the ontological constituent which explains why particular sensible objects possess the substantial character they possess. For the constituent of a sensible object that explains why it is a substance counts as a cause and a principle of the object in question. More particularly, according to Averroes, Met. Z tries to determine which ontological constituent of a sensible object is its substance by being its essence or quiddity, i.e. by being the item we point to in response to the question about what the object is.Footnote ³⁴ As we shall see in some more detail later on, according to Averroes's reconstruction the item that plays the role of the substance of sensible objects in the sense of their essence or quiddity is form. Thus, Met. Z's causal investigation ends up singling out form as the substance and the essence of sensible objects, i.e. the principle explaining why sensible objects are what they are.

Thus presented, Averroes's understanding of Met. Z's strategy is neutral with respect to the two rivaling interpretations of Aristotle's ontology which I have presented in Section 1. What we have yet to establish in particular is whether the notion of “substance of” introduces, according to Averroes, a sense of “substance” different from the absolute sense of the term, i.e. the sense carried by expressions such as “x is a substance”. For, as we have seen, compatibilists hold that being the substance of something and being a substance represent two different ways of being substance, which need not necessarily enter into conflict with one another. For them, both form and the composite can be called primary substance in that they are so called according to two different senses of “substance”. Therefore, Met. Z's claim that form is primary substance does not clash with the view in the Categories that it is particular sensible objects that are the basic or fundamental entities in the world. For incompatibilists, by contrast, there is only one sense of “substance”, i.e. the absolute sense of the term. Thus, the fact that form is the substance of particular sensible objects implies that it is also a substance in the same sense as the sensible object of which it is the form. What is more, being what explains why particular sensible objects are what they are, form must be substance (in the absolute sense) to a higher degree than the composite.

Now, it seems clear to me that Averroes endorses an incompatibilist reading of Aristotle's strategy in Met. Z and consequently regards the explanatory criterion as the fundamental criterion of substantiality.Footnote ³⁵ Admittedly, the Arabic commentator explicitly starts from Aristotle's results in the Categories. Particular sensible objects, which are analysed in the new framework as composites of matter and form, remain the most clear and paradigmatic cases of substance. After all, as Averroes himself puts it, they are the particular ostensible substances, i.e. the substances which are immediately given to us in our ordinary perceptual experience.Footnote ³⁶ Z's task is precisely to single out the substance of such objects, i.e. the principle explaining why particular objects are substances. However, Averroes is an incompatibilist because he does not consider the enquiry into the substance of particular sensible objects to be without consequences for the general question of substance, i.e. for what we should take to be substance in the absolute sense of the term. On the contrary, on his view the very fact that form is the principle explaining why a particular sensible object is a substance is the reason why form is more substance than the sensible object whose substantiality it explains. Thus, even though Averroes tries to preserve as much of the Categories ontology as possible, he is clearly of the opinion that, at the end of the day, form replaces ordinary particular objects in the role the Categories assigns to them.

In general, Averroes seems to subscribe to a general explanatory principle according to which if x is the cause/explanation of y's being F, then x is more F than y.Footnote ³⁷ Averroes's endorsement of this principle clearly emerges from his treatment of the relation between substance and accidents, which Aristotle deals with in Met. Z 1. One of Aristotle's main points in the chapter is that substances, by which he clearly means in this context particular sensible objects, are prior in being to accidents and hence in some sense more real than them. Averroes's understanding of Aristotle's argument hinges upon the notion of causality and explanation. On his reading, substances are prior in being to and more real than accidents in that they are the causes of accidents' existence.Footnote ³⁸ Presumably, substances are the causes of accidents' existence because substances do not depend on accidents in the way accidents depend on them. As is known, it is not easy to provide justification for the asymmetrical relation of dependence between substances and their accidental properties which both Aristotle and Averroes intend to defend.Footnote ³⁹ For it seems that just as accidents cannot exist without being in some subject or other, so substances cannot exist without having some accidental properties or other. However, it is reasonable to think that Averroes believes, like many other medieval philosophers, that accidents are individual and unsharable properties or, to put it otherwise, that each substance possesses accidental properties which it does not share with any other thing.Footnote ⁴⁰ Thus, at a certain point in time no two substances possess the same accidents. Furthermore, accidents do not migrate from one substance to another, i.e. it is not the case that two substances have the same accidents at two different times, either. Hence an accident is the accident it is because it inheres in one particular substance and not in another. This view on accidents allows one to justify the asymmetrical relation of dependence which is at the centre of both Aristotle's and Averroes's ontologies. For according to the theory of individual accidents, an accident depends for its own existence on one, individual substance, whilst an individual substance does not depend for its existence on its having certain accidents in particular, even though of course it must have some accidents or other. Substances change their accidents over time, whilst accidents do not migrate from one substance to another. Thus, it is not unreasonable to maintain, as Averroes does, that substances are the causes of accidents' existence. Be that as it may, the important point is that Averroes employs when discussing the relation between substance and accidents the general explanatory principle I have mentioned earlier on. Substances are prior in being to and more real than accidents because they explain their being.

It is not difficult to see that, when applied to the case of the relation between sensible substances and their ontological constituents, the explanatory principle immediately yields an explanatory criterion of substantiality. The idea in this instance is that the substance of particular sensible substances is more substance than them precisely because it explains why they are substances.Footnote ⁴¹ The substance of particular sensible substances is form, which is the substance of sensible objects by being their essence or quiddity. Form, therefore, is more substance than the particular objects of which it explains the substantiality. The crucial point in the argument is the implicit claim that talking about the substance of something does not introduce a sense of substance different from that we use when we say that a thing is a substance. This claim in conjunction with the explanatory criterion of substantiality yields the conclusion that form is more substance than the composite and so is primary substance. In this perspective, it is no chance that Averroes regards Met. Z 17's exploration of the notion of substance as cause as the culminating point of Z's argument. For it is precisely in the commentary on Z 17 that Averroes's causal interpretation of Aristotle's doctrine of substance as well as his endorsement of an explanatory criterion of substantiality become most evident. Although some of the details of Averroes's commentary are somehow difficult, he clearly emphasises the role of form as the substance of sensible objects and hence as primary substance. He observes for instance that form is that in virtue of which matter exists.Footnote ⁴² I take it that Averroes's point is not, strictly speaking, that form is responsible for the very existence of matter. What he means must rather be that matter, which is in itself only a potential substance, becomes a substance of a certain kind only when it is informed by a form, which confers upon it certain essential characters. A pack of flesh and bones, for instance, is a human being only when and only because it is joined to the form of human beings, which bestows upon it the essential characters of human beings. Averroes also adds that the explanatory role form plays with respect to matter enables it to be described as the substance of the sensible object resulting from the composition of matter with form.Footnote ⁴³ Thus, form is the cause in virtue of which a sensible object is a substance. And – as Averroes makes clear when presenting Z 17's causal investigation – the cause of something's substantiality is itself a substance, where “substance” should be taken in the absolute sense of the term.Footnote ⁴⁴ In conclusion, it seems clear to me that Averroes takes the results of Z 17's causal investigation to concern substance in its absolute sense.Footnote ⁴⁵ This further confirms his endorsement of the explanatory criterion of substantiality and of an incompatibilist strategy.

It is interesting to note that Averroes gives an explanatory reading also of those criteria of substantiality which he does not consider to be the most important for Z's general argument. His treatment of the notion of subject, which Aristotle examines in Met. Z 3, is a case in point. As is well known, at the beginning of Met. Z 3 Aristotle presents four candidates for the title of substance: the essence, the universal, the genus and finally the subject, which is precisely the theme of Met. Z 3.Footnote ⁴⁶ At least the first three items on Aristotle's list (the essence, the universal and the genus) are interpreted by Averroes as candidates for being the substance of particular sensible objects, i.e. the principle explaining why sensible objects are substances.Footnote ⁴⁷ Thus, the suggestion seems to be that, for instance, the essence of a particular sensible object can be taken to be the principle explaining why it is a substance and hence count as the substance of the object in question. This view is defended by Averroes in his commentary on Met. Z 4. Likewise, the universal and the genera that are predicated of particular sensible objects might be taken to be the substances of such objects, i.e. the principles explaining why sensible objects are substances.Footnote ⁴⁸ Does Averroes give the notion of subject the same causal reading as he proposes for the other three candidates for the title of substance? With regards to this point, it is important to realise that Averroes assigns, in his commentary on Met. Z 3, two different roles to the notion of subject. On the one hand, when commenting on the first lines of the chapter, Averroes refers the expression “the subject” not to a criterion of substantiality but rather to one particular kind of entity, i.e. the particular sensible objects which in the Categories are described as primary substances.Footnote ⁴⁹ This interpretation, which is then taken over by Aquinas, is suggested by the very definition of subject Aristotle puts forward in Z 3.Footnote ⁵⁰ For the subject is defined as “that of which the other things are predicated while it itself is predicated of nothing further”, and this definition is strongly reminiscent of Aristotle'scharacterisation of primary substances in the Categories, as Averroes does not fail to point out.Footnote ⁵¹ Thus, Averroes's first suggestion is that Aristotle's discussion of the subject in Z 3 is first of all a metaphysical analysis of particular sensible objects, i.e. the subjects par excellence, and their substantiality. However, in the rest of his commentary on Z 3, Averroes clearly interprets the notion of subject as a criterion of substantiality and in particular as a criterion to single out the substance of particular sensible objects. The general intuition here is that the ontological constituent that plays the role of ultimate subject for a sensible object's properties counts as the substance of the sensible object itself, i.e. the principle explaining why the sensible object in question is a substance.Footnote ⁵² In other words, since in Met. Z sensible objects are not regarded as unanalysable wholes but are rather analysed into ontological constituents, the suggestion might be advanced that it is not the sensible object as a whole that plays the role of subject for its own properties, as the Categories maintains, but rather one of its ontological constituents. As a matter of fact, Averroes rejects the suggestion that the subject criterion can univocally identify the substance of sensible objects.Footnote ⁵³ His reason for doing so is that the subject criterion seems to single out the wrong ontological constituent, i.e. matter, as the substance of sensible objects.Footnote ⁵⁴ The connection between matter and the subject criterion of substantiality is explored by Aristotle in the course of the stripping-away argument. On Averroes's interpretation, the argument is guided by the principle that the ultimate subject of a sensible object's properties, i.e. the substance of a sensible object, is the ontological constituent we are left with when we strip away from a sensible object all its non-substantial properties. Thus, the suggestion Aristotle explores in the stripping-away argument is that matter is in fact what we are left with once all the non-substantial properties of an object are stripped away and so counts as the ultimate subject and the substance of particular sensible objects. Strictly speaking, according to Averroes the application of the subject criterion contained in the stripping-away argument is an incorrect one. For matter is not what we are left with once all the non-substantial properties of an object are stripped off. To get at matter we need to strip off not only the non-substantial properties of an object but also its substantial form. Thus, the argument's fault consists precisely in overlooking the role of substantial form. The analysis of the notion of essence, by contrast, will show that it is form and not matter that is the substance of particular sensible objects.Footnote ⁵⁵ This indicates – so Averroes thinks – that the stripping-away argument is not genuinely Aristotelian, but is rather the attempt by some ancient philosophers (presumably the Presocratics) to vindicate the role of matter as the substance of sensible objects. However, the fact that the criterion can be so easily misapplied is enough to show, according to Averroes, that it is intrinsically ambiguous and so cannot be the fundamental criterion when it comes to singling out the substance of sensible objects.Footnote ⁵⁶

Apart from the details of Averroes's reconstruction of the stripping-away argument, the point which I wish to emphasis is that the explanatory criterion of substantiality plays a major role also within the discussion of the notion of subject. For the implicit thought throughout Averroes's commentary on Z 3 is that, if the conclusion of the stripping-away argument were true, i.e. if matter were the constituent we are left with once all the properties of an object are removed, then matter would be not only the substance of the sensible object in question, but also primary substance tout court.Footnote ⁵⁷ To put it more precisely, matter would be primary substance precisely because it would be the substance of a sensible object. After all, this suggestion is not unreasonable. If all the predications concerning a sensible object are true because there is a certain piece of matter that underlies the properties corresponding to the different predicates we attribute to the sensible object, then in some sense matter is the real substance within the sensible object, its substantial core.

2. My analysis of Averroes's interpretation of Aristotle's ontology in Metaphysics Z would be incomplete without hinting at some of the ontological consequences of the Arabic Commentator's view that form is the substance of sensible objects by being their essence or quiddity. Averroes's discussion of essence is particularly vast and sophisticated. It touches upon some general issues, including for instance the relation between metaphysics and logic. Here, however, I am only interested in showing that Averroes's view on the essence of sensible substances licenses the move from (C1), i.e. the claim that form is the substance of sensible objects, to (C2), i.e. the claim that form is a substance in the same sense as sensible objects. As we shall see in the next section, this aspect distinguishes Averroes's doctrine of substance from Aquinas's. For the way the Dominican Master takes Aristotle's view on essence does not allow, by contrast, for the move from (C1) to (C2).

In Met. Z Aristotle discusses the issue of essence in two times, i.e. in Chs. 4–6 and in Chs. 10–12, in the course of his treatment of definition. The first discussion is a very general and abstract one, which does not touch upon the ontological constituents of sensible substances. In Met. Z 4, for instance, Aristotle raises the question as to what things have an essence in the strict sense of the term and responds to it on the basis of a Categories-style and so very sketchy ontological framework. His view is that substances have an essence in the strict sense of the term, whilst accidents and accidental compounds (i.e. things such as white man and the like) either do not have an essence at all or, if they have an essence, they have one only in a secondary sense of the term. It is not until Z 10–11 that Aristotle raises the question of what the essence of sensible substances is made of, i.e. the question of which ontological constituents of a sensible substance are parts of its essence and which are not. Thus, it is especially in Averroes's commentary on Z 10–11 that we find indications as to his view on the essence of sensible substances. In Met. Z 10–11 the discussion of essence is carried out by Aristotle through the closely related notion of definition, the assumption being that at least in the case of substances a definition perfectly reflects the content of an essence. Therefore, once we know which constituents of a sensible substance figure in its definition, we also know which constituents are parts of its essence and which are not. This perfectly square with Averroes's general approach to the question of essence. For since his commentary on Z 4 his view seems to be that the study of essence is simply the study of the predicates figuring in a definition, and in particular the study of the relations they bear to one another and to the object defined.Footnote ⁵⁸

From within the hylomorphic model which dominates Met. Z, deciding which constituents of a sensible substance are parts of its essence and definition means raising the question whether formal or material parts, or both, enter into the essence and definition. More particularly, since everybody agrees that formal parts are essential parts of sensible substances, the problem boils down to establishing whether material parts, in addition to formal ones, enter into the essence and definition of sensible substances. As is known, the view that matter is part of the essence and definition of sensible substances is not without some support in Aristotle's text.Footnote ⁵⁹ Moreover, it is also suggested by some theoretical considerations about the nature of sensible objects. After all, sensible objects are necessarily material. Unlike other kinds of object, such as for instance mathematical and geometrical objects, sensible objects can only be made of one kind of material. A circle, for instance, can be realised in different kinds of material, whilst human beings only exist, of necessity, in flesh and bones.Footnote ⁶⁰ Thus, it seems reasonable to think that matter should be part of the essence and hence of the definition of sensible substances. As a matter of fact, there are texts in Averroes's commentary on Z 10–11 where the Arabic commentator seems to espouse the view that matter is part of the essence and definition of sensible objects. He contrasts in fact the objects in the definition of which matter does not appear with those whose definition contains, by contrast, matter.Footnote ⁶¹ And the most natural way of interpreting Averroes's contrast is to think that the things in the definition of which matter does not appear are the forms of sensible substances, whose definition makes no reference to the matter such forms exist in, whilst the objects the definition of which contains matter are the composites of matter and form, i.e. sensible substances. However, my opinion is that, despite appearances to the contrary, in Averroes's doctrine the essence and definition of sensible substances contains their form and formal parts alone. In particular, I think that Averroes admits of two different kinds of formula concerning sensible substances. One kind of formula is a rough description of what a certain sensible substance is, i.e. a general characterisation of the way a given sensible substance concretely exists. This first kind of formula includes a reference to the type of matter a given sensible substance is made of. For sensible objects concretely exist as material objects. The other kind of formula, by contrast, is the definition of the essence of a sensible substance, i.e. its true definition. And this second kind of formula makes reference only to the form of the sensible substance in question.Footnote ⁶² For form is the essence or quiddity of a sensible substance, i.e. what a sensible substance essentially is.Footnote ⁶³ This view too finds support in Aristotle's text and seems to represent in particular Aristotle's main line of argument in Z 10 and 11.

If my reconstruction is correct, then, Averroes believes that form is the essence of sensible substances. This claim gives some content to the view that form is the substance of sensible objects by being their essence or quiddity. In other words, if we ask what a sensible object essentially is, we should answer that it is essentially a certain kind of form.Footnote ⁶⁴ The matter a sensible substance is made of, although being necessary to the concrete existence of the form of the substance in question, is not part of the essence and so should not be mentioned in the definition that spells out its content.Footnote ⁶⁵

It is not difficult to see which consequences this view has also for the issue of (C2), the claim that form is not only the substance of sensible objects but also a substance in its own right. For Averroes's claim seems to be not only that form is the principle accounting for a sensible substance's essential characteristics, but also the real substance we are confronted with when we analyse the structure of a sensible object. When asked what a sensible substance essentially is, we should answer that it is a certain kind of form. Thus, by being the essence or quiddity of a sensible object, form turns out to be a substance in its own right and a substance to a higher degree than the sensible object of which it is the form. As is clear, even in the case of his doctrine of essence and definition, Averroes appeals to an explanatory criterion of substantiality. For the reason why form is a substance and an essence in its own right is precisely that it accounts for the sensible object's essential properties. This aspect can also be elucidated from a slightly different perspective. So far, I have presented the question of essence and definition as if the thing we wish to define were a sensible object and form were the item we should point to in response to the question as to what the sensible object in question is. This is certainly also the perspective from which Averroes himself starts, in observance of the Aristotelian principle that we always start an enquiry from the things we are more familiar with to arrive at those that are less familiar to us.Footnote ⁶⁶ And certainly sensible objects are more familiar to us than their constitutive principles. However, at the end of the enquiry, the picture gets somehow reversed. Once the investigation into the essence and definition of a sensible object is concluded, we realise that the real object of definition is not the sensible object we started with but rather its form. In other words, form is not only another object of definition alongside the sensible substance of which it is the form, but is also in some sense the only object of definition, the real thing that is there for us to define.Footnote ⁶⁷ Thus, even if it is generally correct to say that form is the item we should point to when trying to define a sensible substance, it would be even more correct to say that a sensible substance has a definition only in so far as its form has one. For in some sense form is the only object of definition, in that it is the substantial core of a sensible object.

III. AQUINAS AND THE INDEPENDENT EXISTENCE CRITERION

1. Like Averroes, Aquinas too thinks that the aim of Z's enquiry is to investigate into the causes and principles of sensible objects, i.e. of the things which the Categories describes as primary substances.Footnote ⁶⁸ Z's causal investigation basically consists in examining the internal structure of sensible objects and determining which of their ontological constituents is the one accounting for their fundamental characters and in particular for the fact that they are substances. Thus, for Aquinas too, Met. Z is fundamentally about the substance of sensible objects. Unlike Averroes, however, Aquinas is not of the opinion that the investigation into the substance of sensible objects also bears on the issue of substance in the absolute sense of the term. On the contrary, he denies that the ontological constituent which accounts for a sensible object's substantiality is a substance on a par with the sensible object of which it explains the substantiality. Thus, if form turns out to be the ontological constituent of a sensible object which accounts for its substantiality, it will be only the substance of the sensible object in question and not also a substance alongside the sensible object itself. In other words, Aquinas does not license the move from (C1), the claim that form is the substance of a sensible object, to (C2), the view that it is also a substance in the same sense as the object of which it explains the substantiality.

According to Aquinas's understanding of substance, therefore, the absolute use of “substance” and the notion of “substance of” introduce two irreducibly different senses of “substance”, which play at different levels and so never enter into conflict with one another. The notion of “substance” in the absolute sense applies to objects, and in particular to those objects which can be regarded as basic or fundamental. The notion of “substance of”, by constrast, applies to the principles of fundamental objects. The implicit assumption behind the distinction between two senses of “substance” is that the principles of objects are not themselves objects, but rather ontological constituents which do not occupy the same ontological level as the objects of which they are the constituents. Thus, not only does Aquinas believe that the notion of “substance” in the absolute sense and that of “substance of” represent two radically different senses of being substance, but he also maintains that the notion of “substance of” is derivative with respect to that of “substance” in the absolute sense and so introduces a secondary sense of being substance. In other words, when we use the term “substance” in its proper and primary sense, we always want to refer to the fundamental objects in the world and not to their ontological constituents. As can be easily seen, Aquinas's distinction between two different and irreducible senses of substance serves a compatibilist strategy. For the Dominican Master the question of substance in its primary sense is solved once and for all in the Categories, at least in so far as the sensible world is concerned. It is sensible objects, the objects of our everyday experience, that are primary substances. Met. Z does not re-open the question of substance in the absolute sense of the term, but rather introduces a different level of analysis by proposing a causal investigation into the nature of primary substances, i.e. an investigation into the substance of the objects the Categories regards as primary. Since the notion of “substance of” introduces a different sense of “substance” with respect to the notion of “substance” in the absolute sense, the results of Met. Z's causal investigation do not clash with those of the Categories.

If my reconstruction is correct, the important question becomes: Why does Aquinas take this view? Why does he not allow for the move from (C1) to (C2)? The answer to the question I shall try to defend in the following is that Aquinas does not endorse an explanatory criterion of substantiality, but rather a different one, which I shall be calling “independent existence criterion”. For him, only those things which enjoy a complete and independent existence count as substances in the absolute and primary sense of the term. Before going into the details of my argument, I would like to respond to one preliminary objection which might be raised against my reconstruction. Clearly, Aquinas believes that separate substances, i.e. immaterial substances, are substances in the absolute sense of the term. Moreover, he also believes that they are more substantial than sensible objects and play a causal role with respect to them.Footnote ⁶⁹ Therefore, it might be thought, Aquinas endorses after all an explanatory criterion of substantiality. For the reason why separate substances are more substances than sensible objects seems to be that they are the causes of sensible objects. I have two replies to this objection. The first is that Aquinas sharply distinguishes between the internal and the external causes of a sensible object.Footnote ⁷⁰ The internal causes, which are sometimes called “principles” in the strict sense of the term, are the ontological constituents of a sensible object, i.e. matter and form, whilst the external causes are the agent that brings an object into being and the end towards which an object tends and orients its characteristic behaviour. Now, separate substances are clearly external causes of a sensible object. Thus, even if the explanatory criterion of substantiality should apply to the case of external causes, this does not mean that it equally applies to the case of internal ones. In other words, it may well be the case that the external causes of a sensible object are more substantial than it, whilst its internal causes are not, in that the explanatory criterion does not apply in their case. As a second reply, I wish to point out that separate substances perfectly satisfy the independent existence criterion. Separate substances are complete and independent objects. Thus, the suggestion could be advanced that it is because they satisfy the independent existence criterion that separate substances are substances. Of course, they fact that separate substances are the (external) causes of sensible objects may play a role in explaining why they are more substances than sensible objects, but may not be the reason why they are substances in the first place. Thus, there is no indication that Aquinas applies the explanatory criterion of substantiality across the board as Averroes does. I suggest, therefore, that we leave aside the case of separate substances and focus on understanding why, according to Aquinas, the ontological constituents of sensible objects – and form in particular – are not substances on a par with the objects of which they are the constituents.

Probably, the place in his Commentary on the Metaphysics where Aquinas makes more explicit his endorsement of the independent existence criterion is Aristotle's discussion of subject in Z 3. In the Categories, ordinary objects are primary substances because they are ultimate subjects of predication. Since Aquinas thinks that sensible objects remain in the Metaphysics primary substances despite their being analysable into matter and form, it is clear that he cannot reject altogether the subject criterion as a criterion of substantiality. However, the Dominican Master acknowledges that the hylomorphic analysis complicates the picture considerably.Footnote ⁷¹ For one thing, the hylomorphic analysis can easily lead to misapply the criterion, as Aristotle shows in the stripping-away argument.Footnote ⁷² As Averroes already suggests, people can be mislead into thinking that it is matter that underlies all the properties a sensible object possesses and so it is matter that is the ultimate subject of predication. This mistake is more easily made if someone overlooks – as Presocratics for instance did in Averroes's and Aquinas's reconstruction – the role played by substantial form and believes that a substance is composed of a material substratum plus a bunch of accidental properties. On this view, matter turns out to be what is left when we strip off all the properties of a sensible object. Of course, for Aquinas the view in question is just mistaken and so the stripping-away argument is not genuinely Aristotelian.Footnote ⁷³ However, both the view in itself and the argument supporting it show at least that the subject criterion is not entirely unambiguous in that it can be easily misapplied and be taken to point to the wrong candidate for the title of substance. For another thing, Aquinas seems to suggest that there is a sense in which matter is after all an ultimate subject. For even if it is true that it is the composite of matter and form, i.e. the sensible object, that is the subject of accidental properties, it is also true that matter is the subject of substantial form.Footnote ⁷⁴ Thus, even if matter is not subject for everything else that exists – for accidents are predicated of the composite and not of matter –, it is in some sense an ultimate subject in that there is no thing in which matter exists as in a subject. All this further suggests that the subject criterion by itself is not capable of singling out primary substances and so needs to be complemented and integrated by some other criteria of substantiality. Aquinas follows very closely Aristotle in thinking that the things that count as primary substances, besides being subjects of predication, must also be τóδε τι (hoc aliquid) and separable.Footnote ⁷⁵ The two additional criteria rule out matter as a valid candidate for the title of primary substance and leave on the table form and the composite.Footnote ⁷⁶ Aquinas's specific interpretation of the two criteria further narrows down the list and promotes the composite to the title of primary substance.Footnote ⁷⁷ So, which is Aquinas's interpretation of the notions of being τóδε τι and being separable?

On my view, Aquinas's understanding of the notions of τóδε τι and being separable shows his endorsement of the independent existence criterion of substantiality. Let me start with separability. For Aquinas to be separable means to be able to exist per se and not in virtue of something else.Footnote ⁷⁸ Separability is supposed to govern both the relation between substance and accidents and that between a particular sensible substance and its ontological constituents. As is natural, I am particularly interested in the way separability applies to the case of a sensible object and its ontological constituents, but it may be useful to say also something about the case of substance and accidents. It seems clear to me that Aquinas is here endorsing the theory of “particular accidents” I have alluded to earlier on. On this view, an accident is the accident it is because it inheres in one particular substance and not in another. Thus, a particular sensible substance is separable from its accidents, i.e. exists independently of its accidents, not because it can exist without accidents – which it cannot – but because there is no particular accident a particular substance must necessarily possess. A substance in fact changes its accidents – even all its accidents – over time and so there is no accident in particular it must possess to be the substance it is. It could have any other accident and still remain the substance it is. Each particular accident, by constrast, is not separable from the substance it inheres in because in order to exist it must inhere in the particular substance it exists in and not in another. The notion of separability can be easily applied to the case of the relation between a sensible object and its ontological constituents.Footnote ⁷⁹ In this case, it is evident that to be separable means to enjoy an autonomous and independent existence. Thus, matter is not separable in this sense because matter cannot exist without a form giving it a certain structure and organisation.Footnote ⁸⁰ Neither is form separable in the sense specified because a form – at least in the case of the forms of sensible objects – only exists in a material substratum.Footnote ⁸¹ Therefore, only the sensible object as a whole, the composite of matter and form, is capable of independent and autonomous existence.

Even more important is Aquinas's understanding of the notion of τóδε τι (hoc aliquid). In some of his philosophical works the Dominican Master puts foward what looks like a definition of being a τóδε τι.Footnote ⁸² According to the definition, something qualifies as a τóδε τι if it satisfies both of two conditions: (i) it is capable of existing per se; (ii) it represents a full-fledged member of a certain kind or species. Condition (i) is nothing but separability. And we have already shown that neither form nor matter are separable in the way the composite of matter and form is so. As to condition (ii), Aquinas's view is that it is not satisfied by matter and form, either. For neither matter nor form alone represent by itself a full-fledged and complete member of a certain kind or species. It is only the composite of them that is a full-fledged member of a certain kind or species. Matter and form are only constituents of a sensible object, and so are only ontological parts of a complete and autonomous member of a certain kind. Therefore, matter and form, in so far as they do not satisfy the two conditions for something to be a τóδε τι, are not τóδε τι or are so only in a diminished sense, i.e. in that they are ontological constituents of what is a τóδε τι.Footnote ⁸³

The fact that neither matter nor form are separable and τóδε τι does not mean for Aquinas that they are on a par. On the contrary, the Dominican Master insists that form is that in virtue of which a sensible object is separable and τóδε τι.Footnote ⁸⁴ In other words, form is the main factor we should take into account when it comes to explaining the substantial character of a sensible object. However, its being that in virtue of which a sensible object is separable and τóδε τι is enough only for form to qualify as the substance of the sensible object in question, but not also for its qualifying as a substance in its own right. For to be a substance in the absolute sense of the term is necessarily linked for Aquinas with enjoying some kind of autonomous and independent existence – which form does not do. In conclusion, Aquinas's understanding of the notions of being separable and being a τóδε τι clearly shows that he regards the independent existence criterion as the main criterion of substantiality.

2. As in the case of Averroes, some more light can be shed on Aquinas's understanding of substance by looking at the way he conceives of the closely related notion of essence. We have seen in Section 1.2 that for Averroes form is the substance of sensible objects by being their essence or quiddity. The meaning of this view is that a certain kind of sensible object, when considered from the point of view of its essence, is nothing other than a certain kind of form. Form is what a sensible object essentially is. Consequently, the definition of a certain kind of sensible object – or at least its true definition, the definition of its essence – must make reference only to the form characteristic of the kind of object in question.

In his commentary on Met. Z 10 Aquinas explicitly criticises Averroes's conception of essence by arguing that it is both a philosophically untenable position and a bad interpretation of Aristotle's Metaphysics.Footnote ⁸⁵ According to the Dominican Master, the essence of a sensible object does not only include form, but also the kind of matter a sensible object is necessarily made of.Footnote ⁸⁶ By using the expression “kind of matter” I wish to capture Aquinas's famous distinction between common and individual matter. Aquinas thinks that all the members of a certain species have the same (in some sense of “same”) essence. Thus, his view is clearly not that the material features which are proper to one individual object of a certain species, i.e. the individual material features that mark off one individual of the species in question from another, fall within the essence. Aquinas's claim is rather that the essence of a sensible object contains the type of matter all the individuals of a certain species are made of, i.e. what is usually labelled “common matter” as opposed to “individual matter”. In any event, the important point for our purposes is that Aquinas rejects Averroes's claim that the essence of a sensible substance is confined to form alone. Consequently, also the definition of a sensible object makes reference to both its form and its matter.

Aquinas's view on essence affects his general conception of substance. So far, I have said that Aquinas endorses (C1), the claim that form is the substance of sensible objects. If this is true, however, it is also true that Aquinas implicitly distinguishes between “the substance of” and “the essence of” a sensible object. For, on his view, the essence does not only contain form, but also common matter. Admittedly, there is a perfectly understandable sense of the expression “the substance of” according to which it is equivalent to “the essence of”. This is for instance the sense of the expression Aristotle employs at the beginning of the Categories in order to define the notions of homonymy and synonymy.Footnote ⁸⁷ And it is clear that Aquinas too employs the expression “the substance of” in this sense. More technically, however, he sharply distinguishes between the substance of a sensible object and its essence. In other words, the principle which plays the role of the substance of a sensible object is different from that which plays the role of essence, even if the two principles may be closely connected. This distinction between “substance of” and “essence of” is reflected in Aquinas's well-known distinction between forma partis and forma totius. Footnote ⁸⁸ The forma partis is the standard Aristotelian form. Its task is that of giving actuality to matter and turning it into an actual and complete sensible object. By playing this function form qualifies as the substance of a sensible object. The forma totius, by contrast, is the essence of a sensible object. Its task is explaining why a sensible object belongs to the species or kind it belongs to. Forma partis and forma totius – and hence the substance of and the essence of a sensible object – are distinct not only with regard to their functions, but also to their content. For the forma totius, i.e. the essence, contains in addition to form also matter. In other words, the substance of a sensible object and its essence are two mind-independently distinct principles. Thus, even if Aquinas shares Averroes's view that form is the substance of a sensible object, he rejects his further claim that it plays such a role by being its essence or quiddity.Footnote ⁸⁹

In the light of the foregoing considerations, the suggestion might be advanced that the reason why Aquinas does not allow for the move from (C1) to (C2) is that he includes matter in the essence of sensible objects. In other words, it might be thought that for Aquinas form is not a substance in the same way as the composite because a sensible object is, essentially, not only a certain kind of form, but rather a certain kind of form in a certain kind of matter. Thus, the fact that a sensible object is essentially a composite of matter and form and not just a certain kind of form would explain why form cannot be considered to be an object on a par with the sensible thing of which it is the form. There is certainly something to this suggestion. For it is evident that Aquinas sees a certain correspondence between the view that the composite is more substance than form – because the composite is substance in the absolute sense of the term, whilst form is only the substance of the composite – and the claim that the essence of a sensible object contains matter in addition to form. Just as form by itself is not a complete and independent object, so it by itself is not capable of accounting for all the essential features of a sensible object. As Aquinas sees things, composites seem to come first, because they exhibit completeness, which form lacks. However, it is important to understand that, strictly speaking, the reason why Aquinas does not regard form as an object on a par with the composite of which it is the form, has little to do with his view on essence, but is rather connected with his endorsement of an independent existence criterion of substantiality. It is because form does not exist autonomously and independently that it cannot qualify as an object and a substance in its own right. This point is easily realised if we move from the notion of form to that of essence. Aquinas believes that the essence of a sensible object contains both matter and form. However, he does not think that essence, either, counts as an object and a substance in its own right. The reason is that essence, exactly like form, is a principle, an ontological constituent of a sensible object and so, unlike the object of which it is one of the constituents, does not enjoy independent existence. I have said in fact earlier on that Aquinas distinguishes between common matter, i.e. the type of matter all the individuals of a certain kind possess, and individual matter, i.e. the material characteristics which mark off one individual of a certain species from another. On Aquinas's view, all the individuals of a certain species possess the same essence, that is do not differ essentially. Thus, common matter falls within the essence, in that all the individuals of a certain species are made of the same kind of matter, whilst individual matter falls outside it, in that one individual possesses material characteristics which another individual does not possess. Aquinas's distinction between common and individual matter is not unproblematic, but one consequence, which is particularly important for our purposes, is clear enough. If the essence of a sensible object contains form plus common matter, but not individual matter, it is clear that the essence is not a complete and independent object but rather a part or constituent of an independent object. For a sensible object includes besides its essence also individual characteristics, which fall outside the essence and hence add on to it to constitute a complete individual. In other words, an individual is not an essence, but only has an essence, i.e. needs something else besides its essence to be a complete individual.

Therefore, the fact that Aquinas does not regard either form or essence as substances in the same sense as the sensible object of which they are the principles and constituents clearly shows that for him substantiality in the strict sense of the term must be evaluated on the basis of an independent existence criterion.