Aristotle on Knowledge, Truth, and Error (6 Arguments)

This article focuses on Aristotle’s epistemology—his theory of knowledge—and the manner by which he attempts to distinguish truth, knowledge, and error. This article begins with a discussion of error and accidental knowledge. It then moves on to discuss two forms of error. The relation between truth and causation is then discussed. Aristotle’s idea of self-evident truth or truth by definition is then set out. The article concludes with a discussion of truth as universal by nature and the bearing that this definition has on Aristotle’s conception of science.

Much of Aristotle’s philosophy is, at least implicitly, the philosophy of science, which can be taken to mean the systematization of ways of knowing a particular subject matter. Although what constituted a science for Aristotle was quite different from what constitutes a science for us today, much of our modern-day conception of science owes something to Aristotle.

Aristotle’s conception of truth and error has to be understood in terms of its contribution to his philosophy of science. Aristotle’s epistemic ideal is one of certainty and reliability. For instance, Aristotle offers one general definition of knowledge as follows:

“We think we know a thing (in the unqualified sense, and not in the sophistical sense or accidentally) when we think we know both the cause because of which the thing is (and know that it is its cause) and also that it is not possible for it to be otherwise.”

2. Two Forms of Truth and Error

Broadly speaking, there are two ways in which we can conceive of error in the context of knowledge or understanding. Firstly, there is error as a lack of certainty—the absence of a firm basis on which to assert a certain thing. Secondly, there is error as a kind of perceptual inattentiveness or misinterpretation.

Though Aristotle is an empiricist, his conception of error proceeds from the first conception. Aristotle seems to be attempting to show the possibility of excluding all possible errors from human knowledge. This would seem to be too restrictive a criterion for error in the sense that it would involve counting various things as errors when we happen not to know all that we could about them. There are things we can reasonably claim to know that we could not reasonably claim to know absolutely.

There is a distinction to be drawn here between a conception of error that functions as a test for a certain claim and a conception of error that constitutes an orientation or a method. As a way of providing a principle for the execution of science, Aristotle’s conception of error appears to work much better than as a test for individual propositions, insofar as the failure to fully realize the scientific ideal does not ipso facto diminish it as such. Indeed, on one conception, an ‘ideal’ is the highest possible standard (perhaps even an unrealizable standard) that scientific activity can orientate itself around. This is one way of thinking about the concepts of truth and error in Aristotle.

To return to the definition Aristotle gives of knowledge above, a question can be raised: why is it that knowing something means knowing its causes? Surely there is a difference between knowledge of something and knowledge of what caused it. This may consist, perhaps, of the difference between knowledge of a thing and knowledge of some previous thing (or subsequent thing).

We might even want to distinguish between knowledge of something and the more demanding conception of knowledge as consisting of knowledge of all the facts about a certain thing, which would presumably include knowledge of a thing’s causes.

Nonetheless, this conception of knowledge does raise a series of problems. For one thing, it appears to suggest the possibility of an infinite regress which undermines the possibility of knowing anything. That is, if knowledge of the causes of a thing constitutes a precondition of knowing that thing, then there seems to be no good reason not to think that knowledge of the causes of those causes are themselves conditions of knowing said thing (and so on and so forth).

Aristotle is aware of this as a potential issue, which is why he holds that there are certain things that are self-caused, or in his terminology, ‘self-explanatory.’ Often such self-explanation is understood in relation to the nature of a certain thing. Why is X how it is? Because it is in the nature of X to be this way.

4. Truth by Definition

For something to be self-explanatory is for it has to be a certain way by definition. But whatever it is for something to be ‘true by definition’—whatever the criterion for ‘truth’ is in this sense—is not the same as simply stating what it is for something to be true. Rather, it involves a statement of what is essential to that thing.

A large part of science, both in Aristotle’s conception and in that of many modern-day philosophers of science, involves explaining various properties with reference to certain essential facts about things. Nowadays, those facts might be couched in terms of ‘structure’ or ‘laws’ rather than essences.

The question, then, is whether knowledge of something is always causal. The problem we are confronted with is that there is a great deal of variation with which the term knowledge is typically treated. To know something can, of course, mean to know of something, to know something to be true, to know something with total certainty, or to know something to some degree of certainty. To say that we know is to say something very different in different discursive contexts and with respect to very different subjects.

The emphasis here is on the chain of reasoning that gets us to the essence of a certain thing. This allows us to get to the second part of the definition Aristotle offered above—the idea that to know something is to arrive at knowledge that could not be otherwise. Only universal propositions can be truly known.

Yet this seems to run into what by now looks like the characteristic Aristotelian problem: demanding too much. Do we really have no knowledge of contingent things? For one thing, it seems like a lot of science is concerned with contingent knowledge—with particular arrangements of the world, from planets to animal species. Are astronomy and biology any less scientific—any less concerned with systematic knowledge—because they concern contingent things?

Aristotle tries to develop quite a broad conception of universal knowledge. For instance, he tries to argue in favor of including such things as astronomy in this category. Nonetheless, he seems to recognize how unintuitive an idea this is, conceding that “to say that all knowledge is universal . . . is in a way true and in a way not true . . . It is clear that knowledge is in a way universal and in a way not.”

It also seems as though this conception of knowledge is in tension with Aristotle’s conception of metaphysics, which stresses the contingency and perishability of the most basic substances.

6. Truth, Error, and Scientific Understanding

It seems like Aristotle’s conception of what there is finds itself in tension with his conception of what we can know. One possible counter Aristotle could offer is to observe that universal or timeless truths need not be about imperishable or unchanging objects. For instance, it is at least arguable that we can make claims about human nature, even though individual human beings are mortal. Of course, in this case, one still has to posit a certain kind of universal attribute held by every version of a thing, but the basic point is that there can be universal truths about perishable things.

We’ve already observed that the claim to know is extremely contextual. The value of conceiving of knowledge in terms of universal truths must therefore be evaluated in the context of Aristotle’s overall project, which is scientific and (therefore) systematic. The idea here is presumably that we should focus on knowing something in such a way that this knowledge is unchanging (or changes as little as possible). In this way, knowledge provides a standard by which we can judge whether a particular scientific theory functions as well as it can.

This is an idea that will recur over and over again in the philosophy of science: that theories of science function best when they come closest to grasping things themselves and provide principles that are as consistent across as wide a range of experimental contexts as possible.