There are many ways to formulate a substance theory; Aristotle, Descartes, and Locke are all three famous for the substance theories they held. The notion indeed has a very long and distinguished history.
And we hope that some articulate, well-informed philosopher will right now take a minute or two to explain that history and generally correct this article! Just click "edit this page"!
There are a number of arguments for substance theory that have reoccurred in one form or another in the long history of the substance theory.
The argument from grammar
Let us begin with something that can be called the argument from grammar, which goes something like this. When we say, for example, "Snow is white," there is a subject, snow, and we are saying of it that it is white. It makes no grammatical sense to speak of "whiteness" disembodied, without snow (or some subject) that is white. The only way to make a meaningful claim about anything is to speak of a subject, and we predicate various properties of this subject. This subject of predication is called a substance. So in order to make meaningful claims about physical objects (or about minds, if we are advocating a substance theory of the mind), we have to refer to substances. Since we make meaningful claims about objects all the time, substances exist.
Modernly, many philosophers have held that the argument from grammar is no good. The bundle theorist definitely rejects it. The argument says: just because there is a grammatical subject in a sentence, we are referring to what might be called a metaphysical subject, a substance. How, one might ask, does that follow? One might well think that it does not follow. In any event, the argument from grammar does get the following right: in order to make meaningful statements about bodies, we have to talk about subjects. But, the critic wonders, why must the subject of a meaningful statement refer to a bundle of properties? For example, consider the sentence, "Snow is white." Is it not possible for us to understand that sentence as referring simply to a bundle of properties like containing ice crystals, being cold, being a few feet deep, etc.? In that case, we would say that this bundle of properties includes the property of being white, and that is the way to cash out the meaning of the sentence "Snow is white." The bundle theorist, at least, maintains that this seems to be a reasonable way of understanding this claim that snow is white. If so, then we need not talk about any mysterious substances. That, then, appears to be a way to show that we can make meaningful statements about bodies without referring to substances.
The argument from conception
Here is another commonly-discussed argument for the substance theory. Whenever we conceive of one of an object's properties, like the redness of an apple, we must conceive of the object that has that property. One cannot conceive of redness, or any other property, all by itself. The point is that whenever one conceives of a property, one must conceive of it as a property of something. What it is a property of, the substance theorist maintains, is just a substance. There is no conceivable thing such as redness all by itself, or being four inches wide all by itself--that is nonsense. It is always a substance that is red, or that is four inches wide. Similarly with all other properties. Therefore, substances exist.
Here is a criticism of the argument from conception. The important premise in the argument is that we cannot think of properties all by themselves. Therefore--this is the conclusion--they must be properties of substances. Does that conclusion follow from the premise? A critic answers in the negative. Why could we not say that objects need only be associated with a bundle of other properties, which bundle we call an object? An individual property, we might concede, cannot exist by itself. But that, the critic maintains, does not mean that substances have to exist: maybe these bodies that exist are just bundles of properties, and an individual property cannot exist separately from such a bundle.
The inherence relation
One difficult and commonly-raised problem for the substance theory is the problem of specifying what the so-called inherence relation is between a substance and its properties. For example, what is the relation between the apple, considered as a substance, and its redness? The substance theorist might say a property inheres in a substance. That is the word often used: "inheres." A property's inherence in a substance is a bit, but only a bit, like being part of the substance. But it is definitely different from just being a part. When we say, for example, that the apple is red, we are saying that redness inheres in the apple. But then what is inherence? Can any good sense be made of it? It seems that all one can say is that it is what we mean when we say the apple is red--which is to suggest a circular explanation.
The substance theorist may just say that the name of the relation is "inherence," and that inherence is a primitive concept. It cannot be explained any further, but it also does not need to be explained any further. We know what it means to say that the apple has the property of redness, or the property of being juicy. It does not matter that we cannot explain what this talk of a substance "having" properties, or a property "inhering in" a substance, amounts to in any other terms. We have to start somewhere--we cannot define everything, or if we try we will run out of words.
The contents of the foregoing section is also found at inherence relation. Please keep these article consistent.
Bare particulars
Another problem is perhaps more serious. Bare particular is what a substance is called when considered independently of its properties. It seems that substance theories are committed to the existence of bare particulars. But, the critic maintains, the very notion of a thing with no properties is absurd. We just cannot conceive of a thing without any properties. John Locke is famous for describing a substance as "a something, I know not what." It seems that as soon as we get the fuzziest notion of a thing in mind, we are thinking of some property or other. The problem is not just that it is physically impossible that we might stumble across a bare particular, or a propertyless thing on our strolls about town. The point is that the very notion of a propertyless thing is strange: we just have no such notion, and perhaps cannot have such a notion.
That at least is what the bundle theory's advocate might say. Indeed, we might say that this argument against the substance theory is one main argument for the bundle theory; so see also bundle theory, where this article is developed further.
The contents of the foregoing section is also found at bare particular. Please keep these articles consistent.