Does Plato have a satisfactory account of the difference between knowledge and belief?
To a certain extent, that depends on where one looks. Plato is known for two different approaches to a theory of knowledge. In the Meno and Phaedo, Plato presents his theory of knowledge as recollection. But he does this as a more or less incidental part of another purpose. In the Theaetetus, on the other hand, Plato offers three more astute and targeted analyses of what knowledge is. Even though he has Socrates reject all of them, it is Plato’s third analysis in the Theaetetus that is generally recognized as being “Plato’s definition of knowledge.”
The earlier Platonic view of knowledge (the recollection theory) is based largely on the then extant popular tendency to view knowledge as familiarity. In a key scene in the Meno, Socrates draws upon one of Meno’s slaves in order to demonstrate the Theory of Recollection. By showing that the slave can display “knowledge” of things he could never have learned, Plato attempts to demonstrate that all nature is interconnected. If one learns one point, it is possible to “recollect” all of the rest (Meno 81d). Hence, within the Recollection concept of knowledge:
S knows that P if and only if
(i) P is true, and
(ii) S can recollect that P.
Yet, if the scene is read with a focus on the different conversational roles played by Socrates and the slave, it becomes clear that Socrates is leading the slave in an exercise in deductive reasoning, rather than recollection. The slave’s participation is clearly minimal. He contributes no data to the reasoning. There is nothing here for the slave to “recollect”, as the “answer” he is supposedly “recollecting” is inherent in the data already presented to him by Socrates. The “knowledge and correct account” that the slave supposedly displays is clearly nothing more than elementary deductive logic, as carefully guided by Socrates. Since all of mathematics (particularly the arithmetic and geometry of Plato’s time) is a deductively reasoned edifice drawn from a small set of premises, mathematical examples (the geometrical properties of a square in the Meno, and the arithmetic properties of equal in the Phaedo) are the only possible ones Plato could have employed to support his theory of knowledge as recollection. The more empirically supported alternative that it is an ability to reason, rather than preexisting knowledge (a “recollection”) of the answer, that is innate within each of us is not addressed — either in the Meno, or in the Phaedo.
To draw upon the often problematic “analytic / synthetic” dichotomy — Plato’s recollection theory is conceivable only for analytic (a priori) answers — answers that are inherent (logically deducible from) the data already available. That two sticks (with lengths already in hand) are equal is determined from the meaning of the word “equal”. The recollection theory is totally incompatible with synthetic (a posteriori) answers — answers that are dependent on further investigation of the world around us.
The more significant failing of the recollection theory of knowledge described in the Meno and Phaedo, at least for the context of this essay, is that Plato makes no distinction between recollection as belief and recollection as knowledge. He is assuming that knowledge is the recognition of an answer that one recollects. But this is indistinguishable from a true belief.
By comparison to the passing treatments of knowledge in the Meno and the Phaedo, in the Theaetetus it is Plato’s entire purpose to examine the competing theories of knowledge. Oddly enough, however, nowhere in the Theaetetus does he examine his own earlier supposition that knowledge is recollection. The closest he comes is in his rejection of the theory of knowledge as perception in the first analysis of the Theaetetus.
Socrates and Theaetetus here explore that theory — a theory he attributes to Protagoras (“Man is the measure of all things”). Socrates rejects this hypothesis because of the obvious extreme relativism it would impart to knowledge and judgement. But in passing, without specifically addressing the point, he has dismissed his earlier theory that knowledge is recollection. (How else could the soul gain familiarity with the forms in the other realm?)
In the second analysis, Socrates and Theaetetus explore the hypothesis that knowledge is true belief (in some translations, “true judgement”). Socrates rejects this hypothesis also because we can form true beliefs/judgements without having any rational reasons to form them, any relevant basis upon which to distinguish one alternative from any other. Plato’s refutation of knowledge as true belief/judgement is not very convincing, however, because it presupposes the hypothesis treated in the third analysis — that knowledge is to be distinguished from belief on the basis that knowledge is true belief accompanied by some “account”for forming the belief. On the basis of this third analysis, therefore, even though he has Socrates reject it, we can suggest that Plato believed that knowledge is true belief accompanied by some form of rational “account”of why we believe the belief to be true. This he specifically contrasts with true belief unaccompanied by a rational account — something he clearly maintains is distinct from knowledge. The only problem he identifies with this analysis, and the reason he has Socrates reject the definition, revolves around the word “account”. In Plato’s judgement all interpretations of “account”are deemed inadequate because they are to some degree circular.
Plato is nevertheless credited with the earliest documentation of the now generally accepted standard “Tripartite” definition of “knowledge”:
S knows that P if and only if
(i) P is true; and
(ii) S believes that P; and
(iii) S is justified in believing that P.
To modern philosophers, as with Plato, the key difficulty with this definition remains the proper understanding of “account”. Or in modern parlance, the proper understanding of just what constitutes the necessary “justification” that qualifies a true belief for the honorific of “knowledge”. That branch of philosophy known as epistemology is primarily concerned with an exploration of just exactly what it means for S to be justified in believing that P. After over two millennia, an entire branch of philosophy continues to pursue the problem identified by Plato in the Theaetetus.
But assuming that we adopt as Plato’s account of “knowledge” this “tripartite” definition provided in the third analysis of the Theaetetus, we can see that Plato has a clear difference in mind between “belief” and “knowledge”. Knowledge is a special kind of belief, a belief that satisfies some additional conditions. In order to be knowledge, a belief must not only be true, it must be accompanied by some form of “account”(“justification”). So it is fair to say that Plato has offered at least a clear account of the difference between knowledge and belief.
One can certainly argue that Plato’s account of the difference between knowledge and belief is not perfect. Philosophers are today still exploring the details that stumped Plato, and have not yet reached any generally accepted consensus as to what constitutes a proper understanding of justification. But one would certainly have to argue that Plato’s account of the difference between knowledge and belief is sufficiently satisfactory to act as the foundation for the entire field of epistemology. That Plato’s account is satisfactory can also be argued by pointing out that his tripartite understanding of knowledge is still the most common understanding today. In the field of epistemology, even critics of modern concepts of justification adopt as their starting point Plato’s tripartite definition of knowledge.