First Philosophy and the Kinds of Substance

Joseph G Defilippo. Journal of the History of Philosophy. Volume 36, Issue 1. January 1998.

On a certain interpretation Aristotle’s Metaphysics contains two incompatible conceptions of metaphysics or, as he calls it, first philosophy. At two points in the treatise he identifies first philosophy with theology (E.1,1026a19; K.7, 1064b3). Along with this identification comes a certain view about the nature and number of theoretical sciences. We are told in Epsilon.1 that there are three: natural philosophy, mathematics, and theology. Natural philosophy deals with nonseparate, mutable substance, whereas the objects of mathematics are nonseparate but immutable. It is left to theology to study substance that is both separate and immutable (and therefore eternal). Hence it is prior to the other two theoretical sciences and more worthy of honor. But for Aristotle first philosophy is not merely a compartmentalized science concerned with a single genus or kind of substance; he means it to be a universal science of “being qua being.” Indeed, it is the status of first philosophy as the primary theoretical science that is supposed to provide for its universal scope. As he says in Epsilon. 1, it is universal “in this way, because it is first” (1026a30-31). Large and wellknown difficulties loom in the way of this tantalizing idea.

First, it is not clear how theology’s position of primacy is the cause of its universal scope; if anything, divine substance seems to be a special item within a more general ontology. Nor, if we look elsewhere in the Metaphysics, is it obvious that Aristotle does consider theology to be a general science of being. When he argues for the legitimacy of a science of being in book Gamma he makes no mention of theology. Nor does he do so in books Zeta-Theta, his most detailed treatment of sensible substance; from these central books theology seems a peripheral part of a general understanding of being. Moreover, Metaphysics A, which contains Aristotle’s single extant treatment of divine substance, reinforces a view of theology as one special area of metaphysics. His rarified conception of the highest God-incessantly thinking himself, and causing the motion of the first heaven through no volition of his own-certainly discourages the idea that theology proper should be identified with the universal inquiry into being.

These sorts of objections have been dealt with in various ways in the past. W. Jaeger famously proposed an elaborate developmental account, according to which Aristotle’s original metaphysics, represented by book Lambda.6-10, was entirely theological; the immanentist view of the middle books of the Metaphysics then developed as Aristotle weened himself from his early, Platonic, nurture. Accordingly, the discrepancies between the program of E.i and the actual procedure of Zeta-Theta are to be explained by the survival of Aristotle’s earlier views alongside his later ones. To resolve the apparent conflict between the two conceptions of first philosophy, one need only relegate the appropriate texts to the “original” Metaphysics and recognize a development to the maturely Aristotelian position. Jaeger went so far as to maintain that “book A represents a stage that was still purely Platonic and did not recognize the doctrine of sensible substance as an integral part of first philosophy” (221). Of course, such an approach implies a devaluation of theology in Aristotle’s maturest stage of thought. Nowadays, few scholars would accept Jaeger’s view of Aristotle’s development. Nevertheless, the theology of A continues to suffer relative neglect because of its perceived lack of philosophical connection to the arguments of the middle books of the Metaphysics. In this respect many contemporary scholars have a view of the relation of A to the rest of the Metaphysics which parallels Jaeger’s.

G. E. L. Owen answered Jaeger’s developmental account with a groundbreaking discussion of how Aristotle employed a particular kind of predication, which he termed “focal meaning,” to argue for a science of being; Owen’s argument entailed, in part, that Aristotle’s earlier views were, if anything, more hostile to Plato’s metaphysics than his later ones. At about the same time G. Patzig employed the notion of focal meaning—though not the term—more directly to the problem of the relationship between theology and what could rightly be called a general metaphysics. He concluded essentially that theology is general in that the existence of every substance is dependent on and caused by the highest kind of substance, God, or the unmoved movers. Thus Patzig offered the following interpretation of E.1, 1026a23-32: “Aristotle is envisaging here a philosophical discipline that is both a first and a general philosophy, and a substance that is so superior to all other substances that it can at the same time be called in a certain sense general” (38). Regardless of whether one is willing to accept Patzig’s characterization (as indeed I am), even to understand such a view as he attributes to Aristotle we must have a clearcut explanation of what the superiority in question amounts to. Yet Patzig never really delivers such an explanation. At one point he suggests that the relationship between God and sensible substance is a special kind of part– whole relationship, “by which the part supplies in a way the content and principle of the whole” (38). Later, he states correctly that the unmoved mover must be in a way the cause of the existence of all other beings (42). But in the passage of his article which is supposed to explain the causal and explanatory role of divine substance, he falls back again on the part-whole relationship (45). Such an explanation is, I believe, too vague to provide good support for Aristotle’s view. More importantly, though, it is unmotivated by the Aristotelian texts, which do not explain priority among the kinds of substance in terms of a part-whole relationship.

In an attempt to correct the shortcomings of Patzig’s interpretation, M. Frede has articulated more clearly than any previous commentator the use of focal meaning to explain the universal scope of theology. According to his interpretation, divine substance differs from other kinds of substance qua substance. It has a way of being peculiar to itself, and this way of being is the focal way of being: “Now, this way of being, peculiar to divine substances, I want to suggest, is the focal way or sense of being in terms of which all other ways of being have to be explained” (87). The way of being in question is that which belongs to pure forms, in terms of which even sensible substances must be understood, since even they in a sense are most properly forms.

Frede’s interpretation provides a sensible and persuasive account of how a double application of focal meaning (first to being in relation to substance, then to substance in relation to God) could render theology universal. This is a great accomplishment, for the point-though commonly asserted-had not been adequately explained by previous commentators. There is, however, a serious objection to be raised against an interpretation of this sort. For, though Aristotle clearly uses focal meaning to explain the dependence of being on substance, he just as clearly does not use it to explain the dependence of substance in general on God. In this paper I will first give reasons why I think it would have been a bad idea for Aristotle to use focal meaning in the way Frede et al. suggest. I will then try to explain the relevant texts in terms of a different kind of priority relation, which I believe does explain the universality of divine substance in a way that is motivated by Aristotelian texts.

1. Metaphysics Gamma.2 and Focal Meaning

Since we are concerned with understanding in what way theology may be understood to be a science of being qua being, it is first necessary to be clear about how Aristotle conceives such a science of being. His argument for its existence occurs in Metaphysics Gamma.2, which begins with the following statement (1003a33-34):

Being is said in many ways, but with reference to one thing [pros hen] and not single nature and not homonymously.

“With reference to one thing and a single nature” is an obscure phrase. One can get a purchase on its meaning, though, from the gloss that follows it: “and not homonymously.” We know from Categories 1 that when a word is predicated homonymously of different things, this means that though the word is the same in each case, the logos corresponding to it is different. The example Aristotle uses to illustrate homonymy in the Categories is the Greek word zoon, which may mean both “animal” and “painting.” Thus, when zoon is predicated of man it actually has a different signification from when it is predicated of, for example, the representation of a man. Aristotle says that in a case of this sort there is a unique account (idios logos) of what it is for each thing to be a zoon (1a5-6). His point is that there are two different essences each of which may be referred to by the same word, and which have only the word in common. If the conception of homonymy with which he is operating in Metaphysics Gamma is the same as the one defined in the Categories, we can understand the gloss “and not homonymously” as meaning that the essences signifled by the different applications of the word “being” do not have “unique accounts,” i.e., accounts which belong only to the particular kind of being. This alone, however, does not explain everything one would want to know, for to say that the respective accounts of two different essences are not strictly unique to those essences leaves a variety of possibilities for the way in which they are not unique. Fortunately, Aristotle indicates clearly the way he has in mind.

He does this by means of a comparison between “healthy” and “medical” on the one hand and “being” on the other (1003a34-b5). First it is explained that everything healthy “is said with reference to health” (pros hugieian). This explanation is then illustrated by the following examples:

one thing (is called healthy) by preserving health;

another thing by producing health;

another thing by indicating health;

another thing by being receptive [dektikon] of health. Similar examples are then given to illustrate the pros hen predication of .medical” in reference to the art of medicine:

a man (is called medical [iatrikos, i.e., a doctor]) by having the art of medicine;

another thing by being well-suited to the art of medicine;

another thing by being a function of the art of medicine.

It is important to understand correctly the significance of these examples, in order not to be misled about the proper application of pros hen predication to the case of being.

Things are truthfully called “healthy” or “medical” by standing in one of a number of relations to health or to the art of medicine. Thus the truthcriterion of the application of these terms is the success or failure of the thing of which they are predicated to satisfy one of these relations. The examples are easily understood in this way. “Healthy” is predicated of an apple because it satisfies the relation of preserving health, of penicillin because it satisfies the relation of producing health, and of rosy cheeks because they satisfy the relation of indicating health.

It is instructive to consider again Categories 1. There things are defined as synonymous that have both the same name and the same logos corresponding to that name. Synonymy is thus sharply contrasted with homonymy. (I leave out consideration of the third relation defined there, paronymy, which is irrelevant to present purposes.) Given the definition of homonymy from the Categories and the careful qualification in the first sentence of Metaphysics Gamma.2 of the multivocity of “being” as not homonymy, it is reasonable to assume that in the absence of some qualification to the contrary Aristotle normally supposes that something’s being said in many ways at least suggests that the ways in question are homonymous. Recall too that in Categories 1 the example used to illustrate synonymy is the predication of “animal” (zoon) of both man and cow. Since man and cow are both species of the same genus, it is predicated synonymously of them both, i.e., the logos of “animal” does not vary when it appears in the respective definitions of the two species.

It is easy to understand pros hen predication in its application to “healthy.” It is first of all a kind of predication that falls under neither synonymy nor homonymy. That is to say, it neither involves genus-species, species-individual, or genus-individual predication, nor does it involve predicating of different things a term that has entirely different significations in each case. “Healthy” is a pros hen predicate because all the things of which it may be truly predicated stand in some appropriate relationship to health. “Healthy” is therefore to be understood as having many different significations, some of which will correspond to the following:

healthy1 = preserving health (e.g., apple)

healthy2 = producing health (e.g., penicillin)

healthy3 = indicating health (e.g., rosy cheeks)

healthy4 = receptive of health (e.g., a body)

It should also be clear, however, that “health,” which appears in all four definitions, is used synonymously. The signification of “health” does not vary, for the nature to which the various healthy things stand in an appropriate relation is the same in each case.

Still, health is not the genus of healthy things, at least not in the way in which animal is the genus of man and cow. Animal is a universal genus, which by the addition of appropriate differentiae gets specified into its various species, all of which are animals. But health is not the universal genus of healthy things, for, as we can see from the examples, not all healthy things are instances of health. The health of a human being, e.g., is a specific nature, which cannot be further specified by the addition of differentiae. Its universal applicability results from the fact that every healthy thing is necessarily related to it in some way.

It should be plain how the comparison of being to health is meant to work. “Being” is predicated of things that stand in an appropriate relation to some one nature, whose logos appears in the same formulation in all the significations that correspond to “being.” This interpretation is easily verified by examining the detailed application of pros hen predication to being, which begins at 1003b5. 1003b5-6 differs slightly from the opening line of Gamma.2: ” ‘being’ is said in many ways, but always with reference to a single principle [arche].” Why arche, instead of the neutral “one thing” (hen), or “nature” (phusis) as before? The change of wording may simply be due to variatio, but perhaps it reflects a significant point. From as early as Metaphysics A. 1 (982a2) and throughout AGamma.1 as a whole Aristotle has repeatedly asserted that the knowledge he is seeking is the knowledge of certain principles and causes-the “highest” principles and causes. One should therefore take notice when he says that being is said in many ways but always in reference to a single arche. This signals that he is bringing the application of focal meaning directly into line with the inquiry into the principles and causes that are the proper subject of his sought-after science. The use of arche also suggests a causal relationship between the various beings and the single nature to which they are related. It is not merely that there is a common element in the different definitions of “being,” but also that this common element signifies a nature that is ontologically primary.

Aristotle identifies this principle at 1003b6 as substance (ousia). Thus, things may be called beings for the following reasons (among perhaps others):

by being substances;

by being affections of substances;

by being a path to substance.

According to the preceding analysis of the “health” example, one should understand “being” as a predicate that may be attached to things. Therefore, one effect of applying focal meaning to being as a concept is to outline the criteria of such a predication being made truthfully. Something is truthfully called a being in virtue of standing in an appropriate relation to substance. While there will be many different definitions of “being” as it is used in its various applications, ousia will appear in all these different definitions. More importantly, though, it is supposed to appear synonymously in the different definitions. The common presence of substance is what allows being to be “said in many ways, but not homonymously.” This state of affairs reflects the double priority-ontological and logical-of substance; not only do things depend on substance for their existence, but qua beings, they can only be explained by appealing to substance.

Aristotle begins to elaborate the pros hen equivocity of “being” at 1003b11. He starts again with the illustrative example of “healthy” things, and makes a comparison to similar cases. Here he has a new point to make (kathaper oun). Just as there is a single science of all healthy things, so too there is a single science “in the case of the other things [epi ton allon]. For not only is it the task of a single science to investigate [theoresai] things that are so-called according to a single <species> [kath hen], but also things that are so-called with reference to a single nature [pros mian phusin]” (1003b12-l4).

That there is one science of healthy things is, one assumes, an empirical fact, which it would be counterintuitive to deny. But this assertion is intended to offer factual support for the more general and less intuitive contention that pros hen, as well as kath hen, predicates can comprise the subject matter of a single science. Aristotle explains this by saying that even pros hen predicates are said “in a way” kath hen. This statement might lead one to wonder whether he has just destroyed the point of introducing focal meaning in the first place, for kath hen predicates are just those terms that are predicated synonymously. Obviously, one key to understanding pros hen predication is to understand in what way it is actually kath hen.

The structure of the argument under consideration (1003b11-16) is as follows:

  1. It is the job of a single science to investigate kath hen predicates.
  2. But it is also the job of a single science to investigate pros hen predicates. For (a) even these are a kind of kath hen predicate; and (b) we see that there is in fact a single science dealing with healthy things, which are pros hen predicates.
  3. Therefore, since being is a pros hen predicate, there is a single science that investigates beings insofar as they are beings.

Since kath hen predicates are just those things that can be divided directly according to genus and species, when Aristotle says that there is a science of them we are getting straightforwardly the view from the Organon that there is a single science for a single genus. This view of science flows from Aristotle’s demands, first, that a science be demonstrative, and secondly, that demonstration cannot transfer to another genus (An. Po. I.7, esp. 75b8-12). It is this conception of science that Owen argued normally would rule out a science of being qua being, since “being” is predicated across all genera without itself introducing a genus. Even if one disagrees with Owen that the Analytics rule out a Gamma-style science of being qua being (cf. Code, “Owen on the Development”), it is clear that Aristotle is primarily concerned here to shape a conception of metaphysics that will fit the Analytics’s conception of science.

Clearly, then, pros hen predication is the conceptual tool that Aristotle uses to obviate implicit objections to a science of being. The assertion that pros hen predicates are nevertheless organized under a single science is justified by the further assertion that they can somehow be understood as kath hen predicates (as in va). Therefore, since kath hen refers to the straightforward predication of a species or genus term, it is clear that Aristotle means to say that “being” actually is “in a way” predicated synonymously of the different kinds of being. And if we now refer back to the characterization of focal meaning which Aristotle offered at the beginning of the chapter, we can see that it was designed to support this conclusion. For, as we have seen from 1003a33-b10, pros hen predication requires the synonymous inclusion of the focus in the different definitions of the term related to it. Thus, the explanation at 1003b14-15 (kai gar tauta tropon tina legetai kath hen) is to be understood precisely in terms of the synonymous appearance, e.g., of “substance” in the definitions of different beings. The qualifying phrase “in a way” (tropon tina) describes two situations. (i) Although the definitions corresponding to the various uses of “being” are different, an element appears in each of them whose own definition is the same in each case. The various beings are said kath hen only “in a way” because for this to be strictly true would require not just a common element, but the whole definition to be the same. (2) Even in virtue of the common element, the non-substantial beings are only called beings in a qualified way, for they are not said to be substance-the primary instance of being-at all. “Substance” cannot be predicated of them, although it necessarily turns up in the account of every being when it is considered as a being.

2. The Parts of Philosophy

Thus far Aristotle has provided a clear, though still indeterminate, outline of how a science of being is to be justified and understood. Since all being is dependent both logically and naturally on substance, i.e., since substance is the arche of being, a science of being will just be a science of substance, and this science will be the true occupation of the philosopher (1003b16- 19; cf 1004b17-26): “And generally a science is strictly of the first instance, i.e., that on which the other things depend, and through which they are understood. If, moreover, this first instance is substance, then the philosopher must grasp the principles and causes of substances.” Depending on how successfully Aristotle can defend a conception of substance that fulfills the needed role, this should be the beginning of an ambitious and inspiring project. But things are not as simple as they seem. Though Aristotle has now established the possibility of a science that studies all of being, qua being, he then tells us that there are in fact parts of philosophy, and that these parts correspond to kinds of substance. This additional point has the potential to complicate immensely the pros hen unity of being.

This qualification is introduced at 1004a2-9: “There are as many parts of philosophy as there are kinds of substance [ousiai], so that there must be a first and a second of them. For being divides directly into genera. Therefore, the sciences will correspond to these. Indeed, the philosopher is so called just as the mathematician is. For philosophy has parts, and there is a first and a second science, and the rest of the disciplines too come in order.” The first obvious thing to note is that the plural ousiai in the first sentence. must be understood as “kinds of substance,” for it would be absurd to claim that there are as many parts of philosophy as there are individual substances. Now, up to this point Aristotle has concerned himself only with the dependence of being on substance and has said nothing to indicate systematic or generic divisions of substance itself. In fact, the very nature of pros hen predication might seem to suggest that there should not be kinds of substance, for if there were, being might not be said “with reference to one thing and a single nature” (pros hen kai mian tina phusin), but rather with reference to (pros) a number of natures, namely, as many as there are kinds of substance. Moreover, this passage tells us that there are divisions within philosophy coordinate with the different kinds of substance, which suggests that there is notjust one discipline dealing with being and that different kinds of being may be said pros different kinds of substance.

For example, one can easily imagine that many beings that are dependent on sensible substance do not have a direct relationship to divine substance. Such beings would be so called in reference to sensible substance but not in reference to divine substance. In this case there would be not one science of being, but rather, perhaps, three, corresponding to the three realms of ousia which Aristotle recognizes in Metaphysics Lambda. 1, and which he presumably has in mind here. Or should we now apply pros hen predication to substance? In this way a single science of being could be preserved by referring all beings to one of a number of kinds of substance and then referring these kinds to another focus, presumably divine substance. It is tempting to understand Aristotle in this way, but I believe this line of interpretation is ultimately unsatisfactory. It is unsatisfactory first because Aristotle does not say that focal meaning applies to substance in this way, even though to do so would have presented no difficulty whatsoever, and indeed would have been a natural point for him to make if that is what he had in mind. And secondly, it is difficult to see how other substances would have the kind of logical dependency on God that pros hen predicates have on their focus. For if God were the focal meaning of substance, he would have to appear in the definitions of the other kinds of substance; and this is something that Aristotle shows no tendency to advocate, nor is it easy to imagine such a view not being artificial and contrived.

Aristotle needs the pros hen predication of being in order to allow a universal science of being, which will fulfill the requirements set by the common conception of the nature of wisdom (A.2, 982a8ff.), yet not fall prey to certain of his own objections to a universal science. This requires in turn that a knowledge of substance is prior to and required for a knowledge of all other kinds of being. But since there are kinds of substance too, it seems that either even the appearance of “substance” in the different definitions of “being” will be in some way multivocal, or that “substance” in this context must be able to stand for some generic notion which can be understood as subsuming all of its species. But if “substance” is understood as the genus of the different kinds of substances, then it will itself not be substance, for Aristotle explicitly says that no universal is a substance (Zeta. 13), and yet all genera are universals (Metaphysics gg8b17-18, 1014a17, b9, 1069a27).

The dilemma may be brought into relief by taking as an example the definition of one particular being, say, the quality redness. Strictly speaking, Aristotle does not admit definitions of items in categories other than substance. However, since one may ask ti estin of anything at all, there is a less strict sense in which we may speak of the definition even of things like redness (Metaphysics Zeta.4, 1030a11-27, b3-6). The account of redness as a being will be of the form “such a quality in such a substance.” The signification of “substance” here will depend on the generality of the definition. If it is an account of some particular redness, the substance will also be a particular. And since it will be a particular member of some species, we can substitute for its name its genus and differentiae to get a logos of redness as it applies to members of that species. But this account will not necessarily constitute a definition of the essence (logos tes ousias) of redness, for nothing prevents redness from belonging to other species of the same genus or to other genera. We will have to move up the classificatory scale of substances until we reach a genus sufficiently large to exhaust all possible cases of redness. In this case, the genus in question will be sensible substance, or perhaps some subsection of it. The definition of redness will then be of the form “such a quality in sensible substance of a certain kind.” Now, Aristotle also recognizes the existence of non-sensible substances, but redness cannot be defined in relation to these, simply because it is a sensible quality. Accordingly, we cannot widen the extension of the term “substance” as it appears in the last definition of redness, for this would actually make the definition false by allowing “red” to be predicated of subjects of which it cannot, ex definitione, be predicated.

Thus, there are items in Aristotle’s ontology that will need to be understood in reference to different kinds of substance from those in terms of which other items in his ontology may be understood. And since it is Aristotle himself who introduces the divisions in philosophy that arise from the divisions within the realm of substance, it is difficult to believe that he himself did not see this. In Gamma he does not indicate how he thought of substance generally in this connection. The three obvious interpretive possibilities-that substance is a genus, a pros hen equivocal, or a strictly homonymous notion-are either unsupported by the evidence, philosophically unsatisfactory, or both.

I want to suggest that the answer to this problem lies in Aristotle’s indication that the parts of philosophy and the kinds of substance coordinate with them have an order. As he ambiguously puts it, the kinds of substance as well as the branches of philosophy come ordered into first, second and, perhaps, successive members (1004a2-g; cf. Alexander [25o.3o-32]). Certainly, it is this assertion that grounds Aristotle’s characterization of metaphysics as first philosophy; and indeed this is the first occurrence of the term in this sense (cp. 993a15) in the Metaphysics. The question that must now be raised is, what justifies the identification of this “first” philosophy with the study of being qua being?

Here we may consider Aristotle’s assertion, in this same chapter, that “knowledge is of the first, of that on which the rest depend, and through which they are understood” (1003b16-17). The application of this principle to the doctrine of the ordered kinds of substance leads to the conclusion not only that being in general is understood in terms of its dependence on substance, but also that this knowledge is strictly a matter of knowing some first instance of substance. It should be clear that here already we have a doctrine wherein a universal knowledge of being is held to be contained not merely in the knowledge of one area of being, namely substance in general, but even within one specific kind of substance. Given this interpretation of Gamma. 1-2, then, the traditional problem about the consistency of special and general metaphysics should not arise if Aristotle’s views are at all well-founded. We can see this also from the fact that theology has not yet been mentioned. Aristotle is committed to a special science that is also general just on the basis of the arguments he uses to argue for the possibility of the general science.

3. Ordered Series

The question to be considered now is, how can the kinds of substance be related in such a way that knowledge of the first kind also involves knowledge of the successive kinds? To begin an answer, we must consider the kinds of relations that hold among members of an ordered series.

There are at least three different texts (EE 1.8, EN 1.6, Metaphysics B.3) from which it is clear that an ordered series is to be contrasted with a genus that is divisible into species by means of differentiae. This contrast is derived at least in part from Academic discussions regarding what sorts of Forms there may be. Aristotle tells us, for instance, that the Platonists did not allow Forms of “things in which there is a prior and posterior,” i.e., of ordered series. This restriction lies behind the Academic denial of a Form of “number” in addition to the Forms of the individual numbers (EN 1096a17- 19; cf. EE 1218a2-9, Metaphysics 999a7- 12).

In the passages from the two Ethics Aristotle’s aim is to hoist the Platonists with their own petard. It seems that their objection to a Form of number was based on a principle of ontological economy. To recognize a series as ordered according to a standard of priority is automatically to posit the beginning of the series as the first member. If one should then posit a Form which is above the series as a whole, and a fortiori above the first member, this would destroy the series as one had first understood it. The hypothesized first member would now be displaced by the Form; in this case one would have no grounds on which to exclude a Form of the new series that has a Form for its first member. This principle from the Academy thus prohibits the unnecessary and arbitrary production of separately existing entities. Aristotle uses it himself in arguing against the Form of the good, for in the EN the presence of the good in the category of substance implies its priority over the good in other categories.

Thus, if substance forms an ordered series, this set of arguments would require Aristotle not to posit a separate instance of substance above the series. But this is something we would not expect him to do in any case. What is needed is an argument that would prohibit a proper definition of substance which applies synonymously to all members of the series, while still recognizing a kind of commonality among the members. Without such an argument, Aristotle is not justified in claiming that there are first and second philosophies which treat first and second substance respectively. He will rather have to say that there is just one philosophy which treats all of substance, but only in a generalized and abstract sense, and that autonomous, specialized branches of this general philosophy actually treat the different kinds of substance. In this case philosophy would be more like mathematics, a result which he tries assiduously to avoid.

In Metaphysics Gamma.2 Aristotle does not elaborate the point that substance comes in a series. But it turns out that other things too come in series and, concerning these, he has more to say about the relations between the first member and its successors. The most extended and useful passage in this regard occurs in the second book of the De Anima.

At the opening of De Anima II Aristotle announces his intention to start over again in the wake of book I and attempt the most general possible account (koinotatos logos) of the soul (412a1-6). This he does over the course of the next two and a half chapters. Then, in the middle of chapter three (414b2off.), he shifts ground and launches an argument to show there cannot be a general definition of soul. The denial of a general definition is a methodological correction to the opening chapter of the book, and it is based on the fact that the kinds of soul form an ordered series (414b20-32):

It is clear, then, that it is in the same way as with figure that there could be a single definition of soul. For neither is there a figure beyond the triangle and the figures that follow it, nor is there a soul beyond the kinds that have been mentioned. But even if in the case of the figures there were a general definition that corresponded to all of them, it would be specific to no single figure. So too in the case of the kinds of soul that have been mentioned. Therefore, it is ridiculous in these instances, as in others, to seek a general definition which is not specific to any being without also seeking a definition according to the proper and individual form. The case of soul is parallel to the case of figure. For what is prior always subsists potentially in what follows it. This works in the case of both figures and ensouled things; for example, triangle is in rectangle, whereas the nutritive capacity is in the perceptive.

Aristotle’s claim that it is ridiculous to seek a general definition of the soul is not at all in conflict with his stated intention at the beginning of II. 1 There he said that he would try to give the “most general” account of the soul. This need not imply that there is a proper general definition of the soul, for the most general account one can produce may fail the requirements of a definition, one of which is certainly that it apply in the same way to all the items it is supposed to define. Aristotle is telling us that the general characterization of the soul as “the first actuality of a body potentially having life” does not qualify as a true definition.

As elsewhere (cf. Metaphysics 999a9), rectilinear figure is used to illustrate the kind of thing that comes in a series and which therefore does not have a general definition. His first point is that there is no figure beyond the individual kinds of figure. This is the familiar principle from the Academy, mentioned briefly above. There is some useful expansion here, though, for Aristotle tells us that even if there were a general definition, it would belong specifically to no figure. The point seems to be the following. We first of all have the accepted principle that when one has an ordered series there is nothing else beyond the series, i.e., the series is supposed to be fully explicable just, and only, by appeal to its individual members. This should not, however, deter us from attempting to formulate a general characterization of the series. What we must realize, however, is that this general characterization corresponds to nothing in reality in such a way as to define it, i.e., to signify its essence. Since the common characterization (koinos logos) defines no specific series member and there is nothing beyond the series, Aristotle’s point is that it defines nothing at all.

This can be made clearer by considering a genuine Aristotelian genus. Let the general definition of animal be “ensouled with perception.” For any animal, insofar as it is an animal, the statement that it has a perceptive soul will denote the same state of affairs, namely, the presence of a certain set of capacities. It is also true that this definition of animal will be specific to none of the individual species within the genus of animal, for, since as the definition of the genus it belongs to all of them, it will not actually define any of them. This is fine, though, because it will be specific to the genus itself, if we understand the genus as the indwelling nature (enhuparchousa phusis) which is common to all the species (cf. Philoponus, ad loc. [256.28ff]).

It is a different matter, however, with figure. Take the series of rectilinear figures, beginning with triangle. There is no single thing they all have in common in the way in which all animals are alive and have perceptive faculties that perform precisely the same functions. A triangle is a three-sided figure, a rectangle a four-sided one and so on. We can legitimately characterize all figures as bounded planes, but it will still be true that for a triangle to be a bounded plane is a different thing from a rectangle’s being one. For a triangle to be a figure involves having three sides, while for a rectangle it involves having four, so that even in virtue of being the same kind of thing, namely a figure, the triangle differs from the rectangle. For a man to be an animal, however, involves precisely the same thing as for a cow to be one, namely, having perception. In other words, “animal” is predicated synonymously of both man and cow, whereas “figure” is homonymously predicated of both triangle and square.

There is another respect in which a series differs from a genus, which plays a prominent role in the ancient commentators. This has to do with the way in which the series as a whole is ontologically dependent on the first member. In the De Anima this dependence is expressed by Aristotle’s insistence that in a series the prior member is contained potentially in what follows it. On the one hand, this principle makes the series of rectilinear figures dependent on the existence of triangle, for since all of them can be dissected into triangles, if triangle did not exist neither would any of the figures, and hence figure in general. On the other hand, since triangle exists only potentially in the other figures, it cannot serve to ground a general definition of figure, for though one might be tempted to define rectilinear figure as “composed of triangles,” being a triangle is a different thing for an actual triangle than it is for the potential triangles contained in the other rectilinear figures. This sort of consideration prevents Aristotle from attempting a general account of the soul based on the primacy of the nutritive soul, for he sees the nutritive capacity of the perceptive and intellectual souls as categorically different from the nutritive soul simpliciter.

An important difference, then, between genera and ordered series like soul and figure is that a genus is, while an ordered series is not, predicated synonymously of its species. This is taken by Aristotle and his commentators to mean that in a genus there is a single nature to be found in all the instances that come under the genus in question. A genus is therefore logically prior to its species in the sense that it is knowable without reference to any of them, even though it does not exist separately from them. In one sense, however, it seems quite wrong to say that one can know a genus without knowing its species, for we would not take anyone very seriously who claimed to have knowledge of the genus animal and yet had nothing to say when questioned about different animal species. It must, then, be in some special sense that the above is true.

First, it does seem reasonable that the notion of animal as “ensouled substance with perception” would not require reference to any particular species in order to be understood. This does not mean that the genus “animal” would still exist-as a Platonic Form would-even if none of its instances existed, but rather that one cannot specify an instance or instances that would be necessary for its existence. In fact, the definition might still hold perfectly well of completely different animal species from the ones currently existent on earth. Logically, then, the formulation of the essence of animal is not dependent on any one of its species, and is therefore dependent on none of them. It is in this sense that knowledge of what it is to be an animal does not presuppose knowledge of any of the species of animal qua those species, but only insofar as they are animals. This is precisely not the case with soul and figure, and specifically because these are a kind of multivocal (pollachos legomenon). Since the terms “soul” and “figure” do not signify a single essence across all their instances, there is not just one thing that one needs to know in order to have knowledge of soul or figure in general. Nevertheless, souls and figures do have something in common with other souls and figures, namely, the relationship in which posterior members of the respective series stand toward the prior ones. This relationship is the potential presence of the prior members in those posterior to them.

4. The Kinds of Substance

When Aristotle says that the kinds of substance come in an order of priority, we should expect this order to display the logical and ontological features which other ordered series do. And if this is so, then “substance” will itself be a kind of multivocal (pollachos legomenon), but not a pros hen predicate. The reason for this is simple. We have seen that pros hen predicates are characterized by the synonymous appearance of a focus in the different definitions that correspond to a given term. This kind of equivocation, however, does not work in the case of ordered series, for though posterior members do have in common some relation to the first member of the series, the common relation in question is that of containing that first member potentially; the first member does not appear synonymously in the definitions of the posterior ones. Thus, for example, though an animal soul is potentially a vegetable soul in virtue of having the nutritive capacity, what it is to be the nutritive capacity of an animal is actually different from what it is to be the nutritive capacity of a vegetable. Even in virtue of what they have in common, the prior and posterior members of a series will differ in their formulable essences.

It is not difficult to see how these features of an ordered series might apply to substance. In Metaphysics A.1 Aristotle distinguishes three kinds of substance, one immovable and two sensible. Of the latter two, one is eternal and the other corruptible (1069a30-36). At this point there is some doubt expressed as to what will actually qualify as immovable substance, but the book’s later chapters specify the unmoved movers.

The first distinction among the kinds of substance is perceptibility. Since substances are perceptible in virtue of their matter we can alternatively describe this distinction as materiality. The second distinction is between the two kinds of matter, one eternal and the other corruptible. Thus, by combining the two pairs of opposites, immaterial-material and incorruptible-corruptible, we get the threefold classification of substance which is made up of the (immaterial and incorruptible) unmoved movers, the (material and incorruptible) heavenly bodies and the (material and corruptible) substances of the sublunary world.

Now, Aristotle’s view of the natural world requires unceasing and therefore circular motion on the part of the circle of fixed stars. It is well known that in keeping with this requirement he attributes to them a special kind of matter that undergoes no motion except change of place, or local motion (e.g., Metaphysics 1042b5-6, 1044b6-8, 1069b24-26). He thinks in addition that local motion is one of three types, of which the other two are quantitative and qualitative changes, and that it is the “first” of these types of motion. This is obviously true on a cosmic scale, for the circular motion of the heavens is causally prior to all other motions in the cosmos. But the priority of local motion is even stronger than this, for Aristotle believes that some local motion must be involved in every individual motion, even qualitative and quantitative changes.

In terms of their characteristic motions the three kinds of substance display the sort of progression one should expect from an ordered series as one moves from the first to the third. The first substance, as a pure actuality, does not undergo motion at all. The second undergoes only local motion, which is directly dependent on the existence of pure actuality (Metaphysics 1071b1220). The third undergoes in addition the two posterior kinds of motion, which depend for their existence on the first kind of motion, and therefore also on actuality. Significantly, this is true not only on a cosmic scale, but also in the case of each individual motion.

This ordering is one of increasing susceptibility to change as one moves away from first substance. Now, Aristotle defines motion as the actuality of the potential as such. He is also careful to distinguish this kind of actuality from actuality proper, which in Metaphysics Theta.6 is sharply distinguished from motion. It is also a prominent feature of Aristotle’s metaphysics that matter is in some sense potentiality; the matter of something is potentially what its form is actually (e.g., Metaphysics 1043a14–21). Since, moreover, motion is the actuality of the potential as such and matter is potentiality, it is easy to see why material substance must undergo motion while the unmoved mover cannot. And given the dichotomy between motion and actuality proper, it is also easy to see why the more material something is, the less fully actual it can be. The order leading from the unmoved mover to the substances of the sublunary world is defined by decreasing actuality, increasing susceptibility to kinds of change, and increasing materiality. Now, though, it might seem difficult to justify substance as an ordered series. According to the criteria of actuality, movability, and materiality it is not apparent that the posterior members of the series will include the former in a systematic way. Indeed, the unmoved mover might seem too radically different an entity from the two kinds of sensible substance for this construal to work.

I have already noted that according to Aristotle the matter of something is potentially what its form is actually. It would be a mistake to talk about degrees of actuality without keeping in mind that what is actual is always a form of a certain kind. Therefore, although there is a sharp break between the unmoved movers and the two sensible substances according to the criteria of motion or matter, there is in fact a continuous order when one considers substance from the viewpoint of form. Divine substance is pure form, the heavens are form compounded with local matter, which is sufficient only for local motion but necessary for the other two, whereas the substances of the sublunary world are form compounded with corruptible matter and hence are subject to every kind of change.

If we treat the order of the kinds of substance as an order of the modes in which form may have existence, substance clearly forms a series similar to those of soul and figure. The parallelism can be represented schematically as follows:

At this point, one must be careful not to misapply the results of the comparison between the two series. For example, the order of souls has as its first member the least worthy of the three, whereas just the opposite is true of substance. On the other hand, the comparison yields some intriguing insights. First, it provides a legitimately Aristotelian understanding of the kind of priority involved among the kinds of substance, one which is supported moreover by the words of Metaphysics Gamma.2. Secondly, if we understand substance as an ordered series, it is possible to see how “substance” can be multivocal without sliding into the difficulties that ensue if it is understood as a pros hen equivocal with God as the focus. Thirdly, although Aristotle does not in practice separate the two movable kinds of substance and assign them to different disciplines, at least once he does talk of three studies (pragmateiai): “one of the unmoved, one of the moved but imperishable, and one of the perishable” (Physics II.7, 198a28-31). Finally, this interpretation allows for an Aristotelian understanding of the contention in Epsilon.1 that first philosophy is universal because it is first (1026a30-1). As in all ordered series, the first member is (potentially) present in the successive members. And since being is universally dependent on substance, it will therefore be universally dependent on the first kind of substance: form without matter. In addition, if we treat form as what is common to all of being, then there is a focus which is in a sense synonymous; all beings are beings in virtue of their relation to form. As we shall see, there are other things going on in Epsilon.1 as well, but this interpretation of the priority of first substance fits with what is said there at least as well as any other I know of.

I believe these considerations strongly favor the interpretation of substance as a series. They are not conclusive, however, for we must first see how it matches Aristotle’s explicit remarks about the scope and subject matter of first philosophy as theology. These remarks come in Metaphysics Epsilon.1, and I shall turn to that chapter next. Before moving on, though, I would like to point out that the idea of substance as a series suggests an interpretation of first philosophy even before we reach Aristotle’s explicit statements on the topic. If divine substance is “first” in virtue of being form without matter, and it is form that is common to the three kinds of substance, and therefore to all of being, then we should expect first philosophy to be primarily a science of form.

5. Theology in Metaphysics Epsilon.1

Metaphysics Epsilon. 1 repeats Gamma. 1’s assertion that there is a science that, unlike all other sciences, studies being without restricting itself to some special realm of existence (1025b7-9; cf. 1003a24-26). But whereas Gamma. i is followed in Gamma.2 by a justification for the existence of such a science, the same assertion in Epsilon.1 is followed by an attempt to demarcate its boundaries.

The first difference adduced by Aristotle between the science of being and the special sciences is that the latter give no account of the “what it is” (ti estin) of their primary subject matter, but rather proceed from it. They either establish their subject matter inductively by means of perception or assume it hypothetically, respective examples of which are medicine and mathematics (1025b4-5). Doctors do not conduct philosophical discussions of what a man is; they do reason about therapies based on their experience as to which bodily states are healthy and which treatments encourage the development and maintenance of these states. Correspondingly, the mathematician, qua mathematician, does not inquire into the nature of number or line; he posits definitions and then reasons out the attributes of the posited entities (1025b12). These special sciences differ from one another in the exactness of their demonstrations, but they remain alike in this respect, that they do not try to attain demonstrative knowledge of what their primary subject matter is (1025b1416): “Therefore it is clear that there is no demonstration of the essence, nor of the `what it is’, but there is another kind of indication” (1025b13-16). And since Aristotle thinks that the existence of a class of entities can be explained only by a science that also establishes their essence, it turns out that the special sciences do not explain the existence of their objects either.

This opening section of Epsilon. 1 shows that the science of being qua being will ask the ti estin question of its primary subject matter; that this subject matter will be not merely some kind of being, but being quite generally; and that in the course of demonstrating the essence of its subject matter, it will also provide an explanation of its existence. Minimally, then, the science that is being sought is a theoretical science.

Physics is a theoretical science, but it is concerned with one particular realm of being, “that which is able to be moved, and with substance that is for the most part formal, and not only separate” (1025b27-29). It becomes clear from the lines that follow that by “separate” Aristotle means “separate from matter.” For physics is concerned with seeking essences and giving definitions, but the things whose definitions it tries to give, since they are all movable, are compounds of form and matter. Aristotle uses the familiar example of “the snub” to illustrate his point: some objects of definition contain matter, like the snub, which has a certain form-hollowness-compounded with a certain kind of matter-the flesh of a nose. Others do not contain matter, for example, the form of the snub. Physics is therefore a theoretical science which is concerned with the essences of material, movable, objects (1025b2l126).

The next theoretical science mentioned is mathematics. Aristotle’s remarks here are sparse and indefinite. At this point he is only willing to say that it is unclear whether the objects of mathematics are both separate and immovable, though it is clear that at least some of the mathematical sciences study their objects qua separate and immovable (1026a8-10).

Aristotle begins to delineate the conditions for the third and final theoretical science at 1026a10. Its object must be eternal, immovable, and separate; it is therefore prior to both physics and mathematics. First a word about separation. In order for the differentiae among these three sciences to work, one must understand “separate” as meaning “separate from matter” throughout the chapter. This is clearly the case at 1025b28, where “and not only separate” (ou choristen monon) is used as a gloss on “for the most part” (hos epi to polu). The point there is that physics does deal with formal substance, but only for the most part, and not exclusively with substance that is separate from matter. Likewise, Aristotle’s uncertainty as to whether the objects of mathematics are separate arises because it is unclear whether lines and numbers, for example, can exist apart from matter, though it is clear that the mathematician considers them in abstraction from matter. Thus, the relevant sense of separation in the discussion of mathematics is again separation from matter. And this overall delineation is borne out when we come to 1026a13- 16: “For physics is concerned with things that are not separate and not immovable, whereas some of the mathematical sciences are concerned with things that are immovable though not perhaps separate, but existing in matter. The first science is concerned with things that are separate and immovable.” First philosophy deals with substance that is separate from matter, whereas physics deals with substance that is enmattered and therefore movable. This division of theoretical sciences fits perfectly with the conception of substance as an ordered series.

The presence of mathematics here requires some explanation. First, we must note that Aristotle conceives of physics as second philosophy, and later on says that it would be first philosophy if there were no immovable substance (1026a27-29). This suggests that he considers the important relationship to be between metaphysics and physics, and specifically because these two sciences deal with kinds of substance, whereas mathematics does not. The proposal of mathematics as a candidate for first philosophy is due to two factors. (1) The first part of the chapter has established that the science of being qua being must at least be a theoretical science, for since it is not concerned with only one area of being it will differ from the special sciences in that it provides an account of the essence of its subject matter. And since mathematics is regarded as a theoretical science it merits at least brief consideration as a candidate for the first philosophy. (2) There are historical reasons for bringing up mathematics in this connection. Here Aristotle is unwilling to assert a definite view about the independent existence of mathematicals. This is not because he does not have one, but rather because certain strands of Platonism in the old Academy make mathematical objects the primary existents to which all other beings are somehow reducible. The fact that there are respectable philosophers who hold views of this sort means that there is something to them, and therefore that they must be dealt with before a definite pronouncement can be made (for which see M.1-3). Yet the very brief mention of mathematics as compared to physics is an indication even now that its credentials for being the first philosophy really are poor.

The chapter opened with the pronouncement that the principles and causes of being qua being are being sought. This distinguished the inquiry at hand from those special sciences that are concerned with one particular area of being, and which therefore do not provide accounts of the essences of their subject matters. This last point seems to mean roughly the following: the science that is concerned with being as such fulfills its function by asking what things are, by inquiring into their being, whereas the special sciences are concerned only with the properties of beings of a given type, and therefore are not required to demonstrate that they are of a given type, but may proceed from more or less commonly accepted notions. These notions will be provided by sense perception, experience and, in the case of mathematics, hypothesis. One difference between a science of being qua being and most special sciences (though not, perhaps, mathematics) will be that it is a theoretical science. Hence, from 1025b18–lo26ai6 Aristotle has looked among the acknowledged theoretical sciences for the one that is most appropriate to his inquiry.

But even here he has not succeeded in isolating a study of being qua being, but rather two or three theoretical sciences arranged according to a certain priority, with the first science being concerned with immaterial being that has its existence as immaterial. Thus, the section that follows (1026a16-23) devotes itself to delineating some of the features of first philosophy in virtue of which it is first: its causes are eternal, for they are the causes of the heavens (a17-18); and the nature which it investigates is divine, and hence it is theology, and therefore the worthiest discipline of all (a1g-23).

There is something one might consider unsatisfying about Aristotle’s procedure so far. After sharply distinguishing between a general science of being and the various special sciences, he has followed a line of thought that isolates a special kind of being as the subject of first philosophy. Jaeger dealt with this problem by cutting up the chapter to correspond to different stages of Aristotle’s development. This procedure, however, is artificial and pays insufficient attention to the fact that Aristotle himself notices the difficulty and attempts to solve it. If, then, a contradiction or inconsistency remains, he will have no defense, for he is aware of the difficulties involved and seems to have chosen his procedure deliberately. Having first established that he is looking for a certain kind of theoretical philosophy, he considers the theoretical sciences from the point of view of priority. In Gamma.2 he made a similar move when arguing for a division of philosophy according to kinds of substance. There too first philosophy was universal in virtue of being first, and I have outlined a way in which this division could be intelligibly understood in Aristotelian terms.

Aristotle recognizes that the understanding of first philosophy as a science of being in general may naturally give rise to confusion (1026a23-27): “One might be in aporia as to whether the first philosophy is universal or whether it is concerned with a particular genus and some single nature. For it is not the same way in the mathematical sciences, but on the one hand there are geometry and astronomy, which are concerned with some single nature, and then there is universal mathematics, which is common to all” (1026a23-27). This aporia arises because the preceding characterization of first philosophy seems from one perspective to be self-contradictory. The comparison of geometry and astronomy to universal mathematics makes this clear. In the case of these mathematical sciences there is on the one hand a special science or set of special sciences which are differentiated by their subject matter, and then a universal science beside them, which is not related to any specific subject matter within mathematics, but rather provides proofs and principles which are common to all the mathematical sciences. If first philosophy is like universal mathematics, there is a problem with specifying it as the study of being qua being. If we emphasize its universality, it gets reduced to little more than a methodology, which provides proofs that can be used by the more specialized branches of philosophy. In other words, it becomes dialectic; and Aristotle’s objections to dialectic as a universal science of being are well known.so On the other hand, if we emphasize the superiority of its subject matter, it gets narrowly defined by that subject matter and ceases to be the universal science of being that was sought in the first place.

Aristotle says explicitly, however, that first philosophy is not like mathematics in this respect. He may be wrong about this, but there can be no doubt that he believes the aporia is dissolved when one makes the necessary distinction between a science like first philosophy and sciences like mathematics.

It is now left to discern the nature of this distinction from the last section of Metaphysics E.i (1026a27-32): “If then there were no other substance beyond the natural compounds, then physics would be the first science. But if there is an immovable substance, this is prior and is first philosophy, and is universal in this way, because it is first. And it would be the job of this science to investigate being qua being, both what it is and its attributes qua being.” It is not obvious why physics would be first philosophy if there were no unmoved substance. Commentators have had difficulties with this statement, for it seems that even if there were no independently existing immaterial substance, it would still be legitimate to have a first philosophy that gave definitions of form without matter, and such a science would presumably not be physics. Indeed, in the Metaphysics Aristotle tends to identify primary substance as form,3 and this identification does not depend on the independent existence of some entities that exist as forms without matter; it depends rather on recognizing the true subject of physical change. There is no easy way to resolve this difficulty. Progress can be made, though, by considering that Aristotle’s point is actually that an inquiry such as the one we find in the middle books of the Metaphysics would be a part of physics if there were no immovable substance.

In this case, if we were to ask the question, “what are the substances?”, we would get a list of material objects, all of which are therefore subject to movement. Qua movable they would be objects for physics to study. Now, the natural philosopher would have to give accounts of these things that take account of both form and matter. At some point, therefore, he will have to be concerned with form in itself. This concern with form, though, would merely be a part of physics, for it would be pursued as part of an effort to explain movable things. Thus, the inquiry of Metaphysics E. would still take place if there were no separate substance; it would, however, be part of a different, more narrowly conceived, enterprise. It may seem unsatisfying that the conception of first philosophy should be dependent on the actual existence of entities that are far removed from everyday experience. Yet the interpretation of substance as an ordered series helps to explain why this is so. I suggested earlier that the ordering of substances according to the ways in which form may have existence offered an intelligible explanation of the line in our current passage, wherein Aristotle says that first philosophy is universal because it is first. This explanation involves recognizing that the first kind of substance, God, is purely form, and considered as such is also universal, for it is form that is common to all substances, even the two material kinds. Thus, the inquiry into the form of sensible substances is part of an inquiry into form in general. This inquiry is also a study of being qua being because even in the case of sensible substance what we get in answer to the ti estin question is a form, an essence without matter, so that the study of form and the study of being coalesce. One can see now, though, that if there were no separate substance, but only the heavens and the sublunary world, the “first” substance would not be form, but form plus local matter; and since local motion is the primary change, the most universal study of substance would be concerned primarily with form and its involvement with the kind of matter that provides the minimal conditions of change. At this point one should recall the passage in Physics II.7 (198a28-31; cf. above, p. 21), in which Aristotle explicitly mentions a separate pragmateia of “moved but imperishable things.”

In conclusion, Aristotle’s conception of a “first” philosophy that is also universal is grounded in the application of two kinds of priority. The first is pros hen predication, which provides the logical explanation for the dependence of nonsubstantial entities on substance. The second is the kind of priority exhibited by ordered series, which provides a way of understanding the logical and ontological relations between two radically different kinds of substance-material and immaterial-in such a way that it can be part of the same enterprise to know them both.

Other problems remain. In order for this account to be convincing, it must be shown that Aristotle is primarily concerned with form even in his treatment of sensible substance. And in order for Aristotle’s overall view to be judged a good one, it must be shown that the knowledge of God is relevantly informative about the world; otherwise, any talk of first philosophy as the study of form will remain empty, for it will be difficult to see how divine forms have much to do with the forms that are embedded in matter all around us. The problem of course is that Aristotle’s discussion of God is notoriously indeterminate as to the precise content of his knowledge, and correspondingly indeterminate as to the precise content of his nature.