The Foundations of Logic
Logic as a Whole · Toward Beginning of Logic
From his file: FOUNDATIONS OF LOGIC.doc.
THE FOUNDATIONS OF LOGIC
In order to discover the foundations of logic, it will be useful to consider certain sophistical objections which conclude that defining, reasoning, and error are impossible. Since these objections take away the whole of logic, they must destroy the very foundations or first beginnings of logic. Hence, a solution of them should enable us to hit upon the beginnings or foundations of logic.
Here is the first objection: Defining a thing requires the cutting off, or separating, of that thing from all other things. Hence, in order to define, one must know the differences between the thing to be defined and all other things. But one cannot know the difference between two things unless he knows both of those things. Hence, if one is going to know the differences between the thing to be defined and all other things, one is going to have to know all things in the world. But this is impossible since the latter are infinite to us. Hence, one cannot define.
The second objection is thus: One cannot say that something is anything other than what it is. One cannot say, then, that a cat is anything other than a cat; a dog, than a dog. etc. The only kind of statements, therefore, which we can form truthfully, are tautologies: such as a cat is a cat, a dog is a dog etc. Since all these statements are obvious and one does not reason to or try to prove the obvious, there is clearly no need for reasoning, or consequently, an art to direct us in reasoning. This same argument also makes defining impossible since every definition must be said of the thing defined and every definition has many parts and these parts will be different from each other, and the thing defined, and will be said of the latter.
The third objection is as follows: We cannot make an error. If a man should make an error, he would say either (1) that something is what it is not or (2) that something is not what it is. Both of these would be a statement containing a subject and a predicate. Let us take (1) first. Before one forms a statement, one must grasp both the subject and the predicate which in this case are different. But if one grasps only what two things have in common without perceiving their difference, he does not grasp two things but rather one - what they have in common. For example, if I grasp only what dog and cat have in common (four-footed animal), I do not grasp dog and cat, but rather four-footed animal. Hence, one can form a statement using that common thing, but one cannot form a statement whose subject and predicate are different. If one grasps, not only what the subject and predicate have in common, but also their difference, one could not make the mistake of thinking they were the same. One's thinking they are the same presupposes one's grasping their difference (otherwise one would not grasp both of them) and it is impossible to think these two things at the same time: that they are different and the same. Hence, it is impossible to make an error by saying (1) that something is what it is not. Error in the other case (2) is even more clearly impossible, because no one would say that something is not what it is. No one would say that the cat is not a cat etc. Hence, error is altogether impossible, and there is no need of a logic to help us avoid it.
In order to answer the second and the third sophistical objections, we must discover whether it is true that a thing can only be said of itself. If it is not true, there must be something which is apt to be said of something other than itself. Since reasoning, error and defining take place, what is it that is apt to be said of something other than itself? That which is found in many things, the universal, is apt to be said of many things and not just of one thing itself. The universal is something one which is found in many things and which, consequently, can be said of many things.
We must be careful not to confuse the universal with the class which is like an integral or composed whole. (An integral or composed whole is put together from its parts, but not said of them. A chair, for example, is composed of seat and back and legs, but not said of them.) But the universal (or universal whole) is said of each member that it contains; as man is said of Socrates by himself and Plato and so on. But a class, like the American nation, is not said of the individuals that compose it. This citizen is not an American nation. There is a tendency for the human mind, in talking about the univeral whole, to fall back upon the class or integral whole, because the latter can be sensed and imagined, while the former cannot; and what can be sensed or imagined is before for us in our knowing. The young Socrates is represented in the Parmenides as falling back upon the integral whole when he is trying to understand the universal whole.
It is fitting that logic begin with a consideration of the universal or predicable (what is able to be said of many). If definitions or arguments are made in the reason, then they are made out of what is in the reason. But things are in the reason when they are understood, and a thing is universal when understood. When we sense or imagine a triangle, the triangle is an individual or singular triangle, But when we understand what a triangle is, we are understanding what is universal. What a triangle is is common to all triangles. Hence, all the instruments of logic, being made out of things as they are in reason or the understanding, are based ultimately in universality. Thus, as St. Albert the Great teaches us, the first thing to be considered in logic is the universal. or the predicable (what is able to be said of many).
Already we can solve the first objection for the man who knows the universal knows infinite things in some way; as a man who knows what a number is, knows something common to an infinity of things. Hence, if one knows that a thing is not a number, he knows the difference between the thing he is defining and all numbers, even though these are infinite. He doesn't need to know all numbers in particular or to go through all of them before he can separate what he is trying to define from all numbers.
But the second objection still remains, for reasoning and defining still could not take place unless many things could be said of one thing which was not identical with any of the former many things. (This would give us a second coincidence of the one and the many in logic.) In answering the second objection, we shall also be able to give a fuller solution of the third objection. Let us see, then, if universal or predicable can be divided into various kinds.
That which is predicable, or able to said of many things, either belongs to their nature (what they are) or it does not belong to their nature. If it states their whole nature (such that they differ only numerically), then such a predicable is said to be the species of those many things. If it is said of many which differ (not only numerically but) in species, it is called a genus, provided that it states their nature in general (as animal states in general the nature of the cat and the dog). If it states that whereby one species or many species differ from each other as regards their nature, it either follows upon their nature (as the ability to talk follows upon human nature) and in this case it is called a property; or else it can be present or absent while the nature is the same (as green or red can be said of a triangles), and in this case it is called an accident
All of these (although they may be based on parts of the thing) signify in the manner of a whole, like two-legged is said of the whole man although it is based on a part of him. His two legs could not as such be said of him.
When our reason predicates one thing of another, it says this is this, like man is an animal. The word is indicates that our reason takes the subject and the predicate to be the same in some way. But as there are five predicables, the meaning of same is different in each case. If many predicates are said of one subject (as plane figure, three-sided, have its interior angles equal to two right angles, and green are said of triangle) they cannot be the same as the subject in the same way. If they were the same as the subject in the same way, they would be wholly the same as each other. Then they would not be many.
The strictest sense of being the same is when the subject and the predicate are identical, as when one says that Socrates is Socrates. One can also say that Socrates is a man, is rational, is an animal, is able to laugh, and is sitting. But each of these five belongs to or can belong to something else that is not Socrates; so that none of them is as much the same as Socrates, as Socrates is the same as Socrates. And of those five, sitting is the least the same as Socrates for he can lose this without ceasing to be Socrates. Again what Socrates is is more the same than what follows upon what he is, and the species and the difference are more the same than the genus since they are closer to being convertible with Socrates. And the species more than the difference, since it states the whole nature of Socrates.
The second wand third objections against logic (no need for reasoning or defining, and the impossibility of error) assume that the word is, or the word same have only one meaning, identity. If that were so, people would never reason, or make errors or, for that matter, define since a definition must have many parts each of which is truly said of the thing defined. But since they have more than one meaning, it is possible to conclude something different of the subject than what was said of it in the premisses; as in a demonstration, the most perfect kind of argument, where one can conclude a property of the subject in a conclusion and predicate the genus and the difference of the subject in the premisses. error is possible since a man can know things are not identical (and hence are two), and yet make a mistake about their being or not being the same in one of the five other senses (the five predicables). Definition is possible since one can make explicit what is the nature of a thing since both the genus and difference pertain to the nature.
The five predicables are thus the foundation of logic. A complete consideration of the predicables is made in Porphyry's Isagoge which should be read thoroughly with the commentaries of St. Albert the Great and Cajetan.