Find Enlightenment

Aristotle on 'Substance'
by Bryan Register

Date: 1 Dec 98
Forum: University of Texas at Austin
Copyright: Bryan Register

Note: The author may or may not still agree with the views expressed in this paper.

I shall argue that Aristotle employs a different sense of the word 'substance' in his two works, the Categories and the Metaphysics. This confused usage stems from the different concerns of the two texts.

The Categories proposes a ten-fold distinction in basic kinds of things. The distinction is reached by attention to language. Aristotle writes "Of things said without any combination, each signifies either substance or quality [or one of eight other things]" (1b25-26) Aristotle is focusing on a linguistic distinction between kinds of words. He presents no evidence for his ten-fold distinction other than the grammatical distinction between the ten kinds of words. Apparently, he believes that each kind of word corresponds to a kind of object. The number of kinds of objects equals the number of kinds of words. Each word is a member of a certain kind of word, and each object is a member of a certain kind of object. The words of kind one will all make reference to objects in kind one, words of kind two will all make reference to objects in kind two, and so forth. The kind which we will focus on is substance.

In Categories 5, Aristotle defines substance as "that which is neither said of a subject nor in a subject" (2a14-15) This definition refers back to Categories 2, in which Aristotle had distinguished between four different kinds of words: those which are said of but not in, those which are in but not said of, those which are both said of and in, and those which are neither said of nor in, a subject. Since the concern is with kinds of words, we can assume that by 'subject' Aristotle is referring to the grammatical subject or to its referent.

An example of what is said of but not in a subject is 'man': " is said of a subject, the individual man, but is not in any subject." (1a21) It seems that words which are said of but not in the subject are common nouns, or kind terms. 'Man' is a common noun, so it does not refer to a property of the subject but rather the class of which the subject is a member. That which is said of but not in a subject is the predicate term in a subject-predicate sentence which employs the 'is' of identity, rather than of predication. Saying that "Bob is a man" is a different sort of thing from saying "Bob is black". We are tipped off to this by the necessity to include the indefinite article before the predicate in the former sentence. The indefinite article signifies the predication of an identity (class-membership), rather than of a property, of the subject. As may be, Aristotle's point is that "that which is called a substance most strictly, primarily, and most of all" (2a13-14) is not the referent of a common noun. Substance is not a category into which things fit (except of course that 'substance' is a category term, but its referents, individual substances, are not categories).

An example of what is in but not said of a subject is "the individual white". (1a27) Aristotle says that a particular instance of whiteness "is in a subject, the body, but is not said of any subject." (1a27-28) Aristotle does not seem to be speaking of shades of color, but rather of individual instances of the color. For instance, the color of pages two and three of The Complete Works of Aristotle are the same, but each page has a different individual instance of that same color. Thus we might say that page two is white1, while page three is white2, even though white1 and white2 are instances of the same color. However, we lack terms for individual instances of whiteness in the natural language. Thus, though it is the individual whiteness, not an abstract universal whiteness, which is in each white subject, it is the general term, 'white', which we predicate of subject-terms whose referents are white. Aristotle is thus claiming that substance is not a property of things.

Aristotle does provide a positive characterization of things which are neither in nor said of a subject: "Things that are individual and numerically one..." (1b6) He gives as examples "the individual man or the individual horse" (1b5) Aristotle's idea seems to be that substance is that which the referent of a proper name or definitely referring expression. The word which refers to substance always sits in the subject place in a sentence. In fact, this seems to be the mark of substance: the word referring to it is always a subject, never a predicate.

Aristotle also posits a class of 'secondary substances': "The species in which the things primarily called substances are..." (2a15-16) Secondary substances are the classes of which primary substances are members. The mark of secondary substance would seem to be its capacity to serve as predicate in a subject-predicate sentence which employs the 'is' of identity and in which the subject term refers to a primary substance, and its capacity to sit in the subject place only in a subject-predicate sentence which employs the 'is' of identity and which takes as its predicate the genera of the secondary substance (that is, a tertiary substance). The sign of substance, both primary and secondary, is the capacities of the words which refer to them. Thus the distinctions between primary and secondary substance and between both kinds of substances and other kinds of things are based on a classification derived from language, rather than an examination of phenomena.

Aristotle's comments about substance in the Categories deal primarily with the language we use to speak about substances. He writes that "...both its name and its definition are necessarily predicated of the subject." (2a19-20) That common nouns and definitely referring expressions can always take the category and the definition of the category-term as predicates is not a truth about the world but a truth about language. It is based, not on an examination of substance, but on an examination of the words which refer to substances.

Aristotle claims that "All the other things are either said of the primary substances as subjects or in them as subjects." (2a35-36) Those things which are said of primary substances are the general terms for properties of various kinds. Those things which are in the primary substances are their particular properties. It is unfortunate that Aristotle does not distinguish between substances and the words which refer to substances. Through failing to distinguish between things and words for things, he tends to equate words and their referents in a confusing manner, as in the above sentence which implicitly treats properties and words which refer to properties as both things. Thus usage suggests an ontological status for language which language does not possess. It may be this error that leads Aristotle to distinguish his categories along linguistic rather than phenomenal grounds.

There are two claims in the Categories which suggest that Aristotle's concern is not solely grammatical. The first is that there are ten, rather than four, categories. In Categories 2, Aristotle distinguishes between four kinds of words: words for categories of things, words for categories of properties, words for particular instantiations of properties, and words for particular things. But in Categories 4, he announces his ten-fold distinction of kinds of things. The distinction between, say, being-in-a-position (e.g. is-sitting) and being-affected (e.g. being-cut) is not a grammatical one.

The second claim is that "It seems most distinctive of substance that what is numerically one and the same [i.e., substance] is able to receive contraries." (4a10-11) Kinds of things and kinds of properties are not in the world, they are mental constructs or timeless entities of some kind or other. Thus they do not change. Particular instantiations of properties cannot change while remaining what they are; if we repaint a green surface red, the green has ceased to be whether we regard it as having changed to the red or as having ceased to be in deference, we may say, to the red. Only substance can change while remaining the same; if we repaint a white room pink it is the same room as before, but has accepted a property contrary to one of its earlier properties. We may say that it is 'substantially' the same room. Aristotle did not learn this from attending to language, but the point does have a linguistic side. At different times, the word referring to a substance can take both some A and some ~A as predicates.

Now let us turn to Aristotle's view of substance in Metaphysics 8. Here, Aristotle claims, based on the dialectical discussion of Metaphysics 7, that

...the generally recognizable substances... are the sensible substances, and sensible substances all have matter..., and in another sense the formula or form..., and thirdly the complex of matter and form, which alone is generated and destroyed, and is, without qualification, capable of separate existence... (1042a24-31)

Substance, then, means three things: matter, form, and the combination thereof. Aristotle discusses each of these in the first two chapters of book 8.

He says that

...matter also is substance; for in all the opposite changes that occur there is something which underlies the changes... and similarly in respect of substance there is something that is not being generated and again being destroyed, and now underlies the process as a 'this' and again underlies it as the privation of positive character.(1042a33-b3)

We should focus on the passage 'in respect of substance' there is generation and destruction. Aristotle seems to mean that 'ordinary' substances, if I may so call them, which on the one hand are the referents of definitely referring expressions and on the other hand are combinations of form and matter, come into and go out of being. However, whenever this occurs, the referent or combination is replaced by another referent or combination. For example, if one were to take a skein of yarn and unwrap it only to sew it into a garment, one has destroyed the skein but produced a garment. The yarn itself remains in existence throughout the process of destruction and generation, and underlies both the skein and the garment. The yarn is the matter of both skein and garment, and it takes on alternating forms. Aristotle wishes to call the matter substance because he said of substance in the Categories that it admitted change over time. Matter does, we see, admit change over time, because it becomes the matter of different things over time.

Likewise, allegedly, for form. Aristotle provides an example of form: "a covering for bodies and chattels"(1043a16) as opposed to matter: "stones, bricks, and timbers" (1043a15). These are the form and matter, respectively, of a house. Aristotle provides the additional example of stillness. Here he says of 'still weather' that it is "Absence of motion in a large extent of air; air is the matter, and absence of motion is the actuality and substance [in the sense of form]."(1043a23-25) Moreover, 'a calm' is "Smoothness of sea; the material substratum is the sea, and the actuality or form is smoothness."(1043a24-25) We can take it that, just as in Book 7 'snub' is a synonym for 'concave' which applies only to noses, 'smooth' is a synonym for 'having an absence of motion' which is appropriate to liquids. Thus Aristotle is saying that 'absence of motion', the form, is present in both air and sea. Likewise, just as stones, bricks, and timbers could be made into something other than a covering for bodies and chattels - matter can take on different forms - a covering for bodies and chattels can be made of something other than stones bricks, and timbers. One kind of form can inform different kinds of matter, just as instances a kind of matter can be informed by different forms. Aristotle concludes that "It is obvious then, from what has been said, what sensible substance is and how it exists - one kind of it as matter, another as form or actuality..."(1043a25-27)

Since Aristotle provides no additional argument for form being substance, we may take it that the comments explained above constitute his argument. The view seems to be as follows. Matter is substance because it can take on different forms. Thus, since form can inform different instances of matter, it must be substance, too. However, this argument is either faulty or makes a serious ontological commitment. Recall that matter is substance not only because it can take on different forms, but because it can take on different forms through time . If Aristotle wishes to make a parallel claim for form, he must defend the thesis that form can pass from one instance of matter to another through time .

How might such a thesis be understood? Let us review several examples of form. Aristotle has said that the form of a house is shelter, while the form of still weather and a calm sea is absence of motion. In De Anima, he explains that soul is the form of the organism while body is the matter. Form, through these examples, appears to be structure or function. Shelter is the function of a house, and this function implies certain kinds of structure for a house. Living is the function of living beings, and thus the form or animating principle (soul) of living things is their function and causes their structure. Absence of motion is the physical organization of the matter of air or sea in virtue of which they are still. Thus it seems that form is the arrangement of the matter.

When we consider how Aristotle deploys his notion of form in, for example, the De Anima, we can respect the notion as a substantial advance for scientific explanation. In Plato's early dialogues, form becomes reified into a separate existent on its own to which things have a special relationship. This theory is quite implausible. On the other hand, Democritus, as Aristotle observes, tends toward reductionism in his explanations of things (1042b11-15), which is, again, implausible. Aristotle follows Plato in recognizing the explanatory value of form, and follows Democritus in recognizing that only physical objects exist. He advances beyond them both, however, by showing how form can exist and thus possess its explanatory value in a world made up only of physical objects.

However, this does not explain how form can change matter over time. Aristotle talks in several places about the transference of form. The sculptor has an idea of the sculpture which possesses its form and thus he can transfer that form into the marble. The father possesses the form of his species in his semen and thus can transfer that form into the menstrual blood. The senses take on the form of external objects without their matter. These cases are all cases of the form being transferred. But these cases do not involve some single instance of a form ceasing to inform one instance of matter and beginning to inform another. Rather, they are all cases of the form moving without the matter.

Two possibilities occur at this point. One possibility is that there are single instances of forms, and these single instances can exist without matter; this is why they can be transferred without matter from one place to another. There are two reasons to reject this view.

First, Aristotle's examples of the transference of form without matter are usually scientific. They have to do with the physiology of perception and reproduction. Aristotle's accounts of these processes are wrong, and so cannot help him establish this point. Second, we have discovered that form is the arrangement of matter. But there does not seem to be any way that an arrangement of things could exist without the things. We cannot have the skein without the yarn.

The other possibility is that Aristotle does not believe that there are single instances of forms. Individual things are structured the way they are by deriving that structure from a form which structures many other things. Since a form can thus inform different ordinary substances over time, it can underlie change and be a substance. This notion is dangerously close to Plato's theory and should be abandoned for similar reasons.

We can therefore reject the view that form is substance, because contrary to Aristotle, it cannot underlie change. Form, rather, is the arrangement of physical matter and does not exist other than as that arrangement. Aristotle's claims about it should be regarded as the somewhat flawed beginnings of scientific accounts of things.

The last kind of substance, the combination of form and matter which I have called 'ordinary' substance, can underlie change of accidents. There was never any question but that ordinary physical objects would count as substances.

There remains to ask whether matter counts as substance by the other criterion mentioned in the Categories ; can the word which refers to it be used only as a subject term? No. We can say either that "Bronze is a metal", in which case 'bronze' is a subject-term and a term referring to matter, or we can say that "The sphere is bronze", in which case bronze is a predicate-term and a term referring to matter.

The doctrines of the Categories and Metaphysics do not fully cohere. The linguistically defined substance of the Categories is only one of the three kinds of substance discussed in the Metaphysics, the kind which is the combination of matter and form. There is an explanation for this.

When Aristotle is discussing substance in the Categories, his concern is linguistic, not scientific. Thus he isolates a kind of term which plays a certain role in logical discourse and specifies the role it plays. However, in the Metaphysics, Aristotle's concern seems to be scientific. He wishes to accord a certain key role to matter and form, which are merely components of ordinary substance. Thus he mistakenly calls them substances.

Aristotle uses a linguistic criterion to identify substances, and then illegitimately extends the use of the word 'substance' to mean certain things which do not match this criterion. The account which Aristotle gives of ordinary substances may be very helpful to the sciences, but it is not a linguistic account. He should therefore have respected the difference between his two concerns and employed separate vocabulary for each.

Find Enlightenment at