I have a paper for my Aristotle class coming due, so I’m trying to figure out something interesting and original to do, rather than just the usual boring analysis of Aristotle’s views on substance from the Categories. (I could also write on his various arguments for the Prime Mover, his theories of the four causes and of chance/spontanaeity, or a topic of my own choosing. But his account of substance in the Categories is rather interesting to me, so substance it is!)
It would be all too easy to dismiss Aristotle’s views of substance in the Categories as immature and unrefined. His distinction between primary and secondary substances seems absurd on its face. And yet, as I become more deeply immersed in Aristotle’s thinking, I find myself more entranced by its complexity and insight, even as I clearly see the failings.
In the discussion of substance in Chapter 5 of the Categories, Aristotle’s primary goal seems to be to establish, contra Plato, the ontological primacy of individual particulars. Individual trees, houses, books, and horses are the most real kinds of things, not the universal categories TREE, HOUSE, BOOK, and HORSE. Qualities, quantities, relations, change, and so on are also dependent upon these individual particulars for their existence. Thus Aristotle argues that particulars are metaphysically primary.
Not being any sort of Platonist myself, I have few disagreements with this broad outline of Aristotle’s argument. But questions and problems do arise with Aristotle’s distinction between primary and secondary substances, particularly why Aristotle speaks of both particulars and universals as substances. But before we can examine this issue, we must understand Aristotle’s basic arguments about substance in the Categories.
Aristotle begins his discussion of substances by asserting substances are “most strictly, primarily, and most of all” individual particulars, such as this flower or that book (2a13-15). These individual particulars are, unlike other things, “neither said of a subject nor in a subject” (2a14). But what does that mean? Aristotle’s explanation of “said of” and “in” a subject can be found earlier, in Chapter Two of the Categories.
To be said of a subject or not said of a subject is simply the distinction between abstract, universal terms and concrete, particular terms. The individual boy Eric is not said of any subject, but the universal term “boy” is said of Eric. My particular copy of The Fountainhead is not said subject, but the universal term “book” is said of my particular copy of The Fountainhead. So universal and abstract terms of said of subjects, while particular and concrete terms are not.
The distinction between in a subject and not in a subject is, unfortunately, not quite so clear. Aristotle describes that which is in a subject as “in something, not as a part, and cannot exist separately from what it is in” (1a25). That which is not in a subject seems to be the subject itself. In other words, Aristotle seems to be distinguishing here between attributes (that which is in a subject) and entities (that which is not in a subject). So the gray of my cat’s fur, for example, cannot exist separately from my cat, so the gray is in a subject. But the cat itself is not in any subject, as the subject is the cat (3a13-15). Similarly, knowledge of history cannot exist separately from someone’s mind, so it is also in a subject. But that someone’s mind is not in any subject. So to be in a subject is to be an attribute of an entity, while to not be in any subject is to be the entity itself.
Aristotle’s distinctions can thus be understood as a simple four-square grid:
||Said of a subject
universal and abstract
|Not said of a subject
concrete and particular
|In a subject
e.g. white, long
e.g. this white, that length
|Not in a subject
e.g. cat, book
e.g. this cat, that book
(This chart was inspired by a similar one found in some notes on Aristotle by Robin Smith.)
(Perhaps the most confusing aspect of Aristotle’s account so far is his failure to distinguish between the things-in-the-world and our language-to-describe-those-things, as Bryan Register pointed on in paper on Aristotle’s views of substance. Thus it is not always clear whether Aristotle’s distinctions regarding “said of a subject” and “in a subject” refer to ontology or language.)
Returning now to our original topic of substances, we are in a better position to see what Aristotle means by his distinction between primary and secondary substances. Primary substances, being neither said of nor in a subject, are individual, concrete particulars, such as this dog and that table (2a13-15). Such individual particulars are primary substances because they are “most strictly, primarily, and most of all” substances (2a13-15). But Aristotle goes on to extend his concept of substance to the species and genera of those primary substances, such as dog/animal and table/furniture (2a15-17).
However, not all secondary substances are equal. Later in Chapter 5, Aristotle ranks secondary substances, arguing that “the species is more a substance than the genus” (2b7) on the grounds that the species is a “more informative and apt” description of some particular than the genus. For example, it is more useful to describe a particular elm tree as “an elm” rather than merely as “a tree.” Additionally, Aristotle offers us an argument from analogy to justify his division of secondary substances as follows. Primary substances are “substances most of all” because they “are subjects for all the other things” (2b16-7). Similarly, “the species is the subject for the genus,” but the genus is not the subject for the species (2b20-1). For example, Eric is man, man is animal, but animal is not man. Thus species are more primary, more substance, than are genera. However, within these ranks of primary substance, species as secondary substance, and genus as secondary substance, Aristotle denies any further ranking. To use Aristotle’s examples, an individual man is no more or less a substance than the individual horse. And that individual man is no more or less a man than an individual horse is a horse.
The most basic question at this point in Aristotle’s account is why he chose to apply the term “substance” to species and genera at all. Was he conforming himself to ordinary Greek usage in some way? Or is his terminology a purely rhetorical device for comparison and contrast with Plato’s views of the Forms as most real? Although reasonably certain answers to these questions may be impossible, Aristotle’s overall project in the Categories, as well as some of his particular comments about substance in Chapter 5 may give us some hints as to why he spoke of primary substances as particulars and secondary substances as universals.
Much of Aristotle’s comments in the Categories makes little sense if we suppose him to be speaking of ontological categories. After all, he begins the Categories with a discussion of the linguistic terms homonyms, synonyms, and paronyms (1a1-15). The next chapter speaks “of things that are said,” distinguishing between concepts and propositions (1a16). The chapter on relatives opens with the linguistic statement “we call relatives all such things…” not an ontological claim that “relatives are all such things…” (6a37). Similarly, in Chapter 12 Aristotle opens with “one thing is called prior to another in four ways,” not “one thing is prior to another in four ways” (14a26). In the chapter on quality, Aristotle first offers his basic definition, then remarks “but quality is one of the things spoken of in a number of different ways” (8b25-6). The subject of the chapter is then those different usages of the term. And finally, Aristotle’s final, short chapter on “having” deals not with the ontological meaning of the term, but rather the “number of ways” in which it is spoken (15b17). So although the Categories sometimes deals with ontology, as in his discussion of the contrariety of relatives, Aristotle’s basic focus in Categories is surely on the ways in which such ontological terms are used in language (6b15-26).
By keeping this linguistic perspective in mind, we can make more sense of Aristotle’s discussion of primary and secondary substances. In short, the seemingly strange distinction seems to be grounded in concern about “informative and apt” description rather than ontology (2b10). The evidence for this approach comes in Aristotle’s account of why only species and genera are secondary substances. Aristotle writes:
It is reasonable that, after the primary substances, their species and genera should be the only other things called secondary substances. For only they, of things predicated, reveal the primary substance. For if one is to say of the individual man what he is, it will be in place to give the species or the genus (though more informative to give man than animal); but to give any of the other things would be out of place–for example, to say white or runs or anything like that. So it is reasonable that these should be the only other things called substances.
In other words, species and genera are designated as secondary substances simply because they capture the basic nature of individual particular, of the primary substance. Imagine, for example, that you are walking with a young child. She points to a tree and asks “What is it?” To say “It’s green and brown” or “It’s alive” or “It has leaves” or “It what paper is made of” may be truthful, but not nearly as helpful to the child as saying “It’s an elm tree.” The species and genera are so useful to us largely because they often imply or hint at those other possible descriptions, such as having leaves, being alive, and so on. In short, the species and genera, the secondary substances, are most “informative and apt” because they usually encapsulate a great deal of knowledge.
Additionally, in attempting to designate particulars, there is a certain progression of usefulness from primary substance to species to genus, in that the species is more informative than the genus, just as an ostensive “this” or “that” is more informative than the species. For example, if I am gardening and tell my husband to bring me “the plant” (the genus), he might not know which plant I mean. If I tell him to bring me “the geranium” (the species) there still may be confusion as to which geranium. Additionally, he might not be familiar enough with plant classifications to distinguish geraniums from other types of plants. But if I say “that geranium right there” or “that plant right there” or even “that thing right there,” clearly designating the particular primary substance of one geranium, then I have provided the most useful indication of my wishes to him.
So from this perspective of what is “informative and apt,” a continuum from individual particulars as primary substances to species and genera as secondary substances at least seems less strange (2b10). Nevertheless, his use of the rather ontological language of primary and secondary substance to describe how we may best use language is decidedly uninformative and unapt.
Of course, this linguistic interpretation of secondary substances is not to deny that Aristotle is making significant ontological arguments in this chapter on substance. Contrary to the Platonic view of the Forms as most real, Aristotle clearly argues for the ontological primacy of particular entities on the grounds that “all the other things are either said of primary substances as subjects or in them as subjects” (2a35-6). With respect to entities, he notes that “animal” is predicated of the abstract term “man” only because animal is predicated of individual men. Similarly for attributes, where color is in body in general only because it is in individual bodies. Thus Aristotle asserts the ontological primacy of particulars in concluding that “if the primary substances did not exist it would be impossible for any of the other things to exist” (2b5-6).
With respect to this argument against Plato, Aristotle’s talk of primary and secondary substance may also serve a rhetorical function. The use of the term “substance” for both universals and particulars allows him to more directly compare them, showing how some universals are “nearer” to particulars than others (2b8). Additionally, the talk of secondary substances may be a superficial concession to Platonism, allowing Aristotle to say that even if we grant universals the status of substance, they are still dependent upon individual particulars, upon primary substance, for their existence — and thus less real.
In any case, Aristotle’s designation of universals as secondary substances, whatever its purpose, does not seem to be ontological in any significant sense.
So it seems that I have written much of my paper already. Hooray for blogging!