SB logo
en's Homepage

Expound and Evaluate Plato's Doctrine of Recollection

In simple terms, Plato's doctrine of recollection states that rather than 'learning' in the common sense, what is actually occurring when people think about something, and eventually find an answer or solution to a problem, is that they are 'recollecting' things that they already knew. The reason that Plato proposed such a theory was because of a very real problem, namely that it seems to be an impossible task to be able to come to know something, without in fact already knowing it. As Meno points out - if we don't know what something is if we try to find it how will we know when we have? When, for example, we say that we don't currently know what 358721 multiplied by 54 is, how is it that by some means we can come to discover that it is in fact 19370934? Surely if we don't already know that 358721 X 54 = 19370934 then when someone presents this to us as a fact we should not have a faculty that enables us to recognise that it is.

One of the main problems with the debate in The Meno is that Plato denies that he is discussing mere beliefs (true or otherwise) and claims to be talking about knowledge, yet it is not until his much later work, Theaetetus, that he properly attempts to define what the difference between these things is (and, even then, he fails to draw a conclusion as to what exactly this difference is). So Plato sets out his discussion without a clear formulation of what exactly he is looking for in 'knowledge' that cannot satisfactorily be achieved except by a process of recollection. The problem, however, of how we can say that we now know something (i.e. be able to identify something as correct) when we did not already know it is reminiscent of Plato's third definition of knowledge in Theaetetus. Having refined the definition from 'Knowledge is true belief' to 'Knowledge is justified true belief' Plato suggests that whilst a belief must be correct for it to be said to be knowledge, it also needs some extra thing (which he calls a logos) which gives us a reason for the belief we have. The problem he found with this definition is that it effectively defines knowledge as true belief, plus the knowledge of some evidence to confirm it. This sort of recursive definition is not going to be very useful to precisely define what knowledge is, and yet for our current purposes with The Meno it might be very useful indeed.

Whilst we may not know that 358721 X 54 = 19370934 we do know that 1 X 4 = 4 and that 2 X 4 = 8, and that 7 X 4 is 28, and so on, and that if there are multiple digits we carry the leftmost ones, and that once we have multiplied all the numbers by 4 we can multiply them again by 5, shifting each one up one column and adding together the products derived in the two stages. Once we have performed this process we should be satisfied that the number produced at the end is in fact the answer to 358721 X 54. Whilst we have failed, perhaps, to justify our original claims, and the axioms by which we perform the multiplication process, these axioms are already knowledge, and we do not need to justify our use of them. So whilst it might be true that we inately know (whether from a previous incarnation or not) our times-tables up to ten, Plato cannot claim that those things we base on such recollections are recollections themselves. Moravcsik correctly complains that Plato should not be using this sort of theory to explain the process of forming concepts from prior-known propositions, since this sort of deductive reasoning from accepted axioms really should not require any prior knowledge to be recollected. The axioms could happily fit as Plato's logoi, to be used as the evidence with which to prove more complicated things. Whilst we cannot continue our abstraction indefinitely, and can never justify the most basic building blocks of our beliefs, there is clearly a very large wealth of knowledge that could be established using such building blocks, and in view of such a concession clearly Plato's suggestion cannot be as far reaching as he might like - it can only be used to debate very basic facts, and where they came from.

Coming, however, to the question of whether even such as concession as this - that basic propositions would have to be recollected for a previous incarnation - we will have to consider how fundamental to the debate the stranger parts of Plato's suggestion is, that it would be easy for people arguing against to use as red herrings to attack the argument. The question of reincarnation, for example, might enter into the debate if one were to argue that since Plato's theory relies on such a concept it is clearly unprovable at best, ridiculous at worst. Yet this would be an unfair criticism of the doctrine of recollection. As I have said, the doctrine is indeed formulated due to a very real problem, and to dismiss it for its accompanying baggage would seem rather hasty and foolish. There is no need to suggest that reincarnation must be taken literally, if at all, and it would probably not seem at all strange if one were to argue that it might be that some genetic make-up were so arranged as to create a brain with various simple mathematical or linguistic functions.

Far from being a completely out-dated theory, even much more recently Jung seriously suggested that there might be a form of 'genetic memory' which would allow a generation to benefit from the discoveries of a previous one. Whilst few would accept such theories, there is no simple way to suggest they are completely mad, and the possibility of accepting some form of evolution of thinking seems quite natural.

There is, however, no need to accept any suggestion that reincarnation is 'to blame' for any knowledge we may have - the doctrine would work just as well (though not to prove the more metaphysical aspects of Plato's thinking) were we to say that the mind simply does have embedded ideas which with careful thought we are able to recollect. Even Plato seems a little confused if he is suggesting that simply be virtue of reincarnation we could explain knowledge, stating:

'Thus the soul, since it is immortal and has been born many times, and has seen all things both here and in Hades, has learned everything that is' (My italics).

It isn't clear whether this is simply a slip on Plato's part, or perhaps even a common mistake in the translation, but it seems to me clearly ridiculous to be basing a doctrine that says we never learn anything on the idea that we learned it all in previous lives, and thus already have the knowledge. Overall, therefore, it would seem unwise for someone trying to put forward Plato's theory to suggest that reincarnation was a fundamental part of the doctrine and could not be separated from it. Clearly this would mean the death of the theory if it were true, but I don't think many people would claim it was. In fact, even Socrates covers himself, moving the doctrine yet another step further away from Plato's own opinion, by attributing it to other people. It might, therefore, be claimed that whilst he would have probably discarded the stranger parts of the theory, what he was attracted to was the core idea. There isn't much evidence for such a saving grace in the text, but it might be just this sort of embarrassment (that it doesn't especially make sense to introduce religious concepts of reincarnation) that led Socrates to distance himself from at least parts of the doctrine.

Another unnecessary distraction would be that of Plato's suggestion that learning cannot occur since it would have to be a process of giving, where a teacher gives knowledge to a pupil, but that it cannot be giving since when someone gives something to someone else they don't have it any more. This seems, perhaps, to be not only taking the analogy far too far, but also to make ridiculous conclusions in light of it. If giving is a process which requires the first person to lose the thing that they give then clearly knowledge is not given, but that certainly doesn't rule out the possibility of there being knowledge. It is not merely coincidence that people talk about 'giving of their wisdom' or the such-like, rather than saying 'giving their wisdom' since 'give' is clearly not being used in a normal sense. This seems an unimportant point, but one where attention should be taken not to dwell on it rather than more fundamental parts of the dialogue.

Socrates insists that he is not teaching the slave-boy during his demonstration, and yet he seems to expect the audience to be just as accepting of his word when he 'alerts' Meno to the fact that he is not instructing and merely questioning the slave-boy, as the slave-boy is of everything Socrates says. It is not clear what the distinction is between instructing and questioning (and it certainly doesn't seem satisfactory to put it down simply to mood of verbs). There seems little proof that this is not instruction, though the main thing often picked up by scholars is that the fact that he does not always lead the slave-boy in the right direction must show that no teaching is going on. This seems a poor excuse at best. It is a frequently used tactic to lead people down the wrong path when teaching them, if only so they hit upon a reductio ad absurdum, whereupon they realise that since the conclusions they have been led to are clearly wrong they must have made a mistake along the way.

Clearly if Socrates had left the slave-boy in mid lesson, when the slave-boy believed that doubling the sides of a triangle would double the area, he would have succeeded in badly misleading him - but surely far from suggesting that Socrates isn't teaching the slave-boy it should just show that he is teaching him very badly. I would suspect that Plato would deny that getting someone to believe something that wasn't true could be classed as 'teaching', as he would probably claim that no knowledge has been transfered. This, too, seems something of a cop-out; since it would allow someone to claim to be a good teacher on the basis that everything they taught was correct. This would, of course, always be true if the claim that I put in Plato's mouth (fairly or not) were true, since only correct things would count as being taught. Plato's system, however, seems to be just as selective as this. We are told that, using recollection, the slave-boy 'intuitively' grasps true belief' - but he also 'intuitively grasps false beliefs', so how can we claim that it is an internal faculty that is telling him what is true or not. As Dominic Scott puts it:

'If we can derive from within ourselves false as well as true judgements, we shall need to decide which are which. But how are we to make this decision? Is there to be another process of recollection to help us find out? If so, we have an infinite regress on our hands'.

Again we find ourselves stuck at a dead-end when trying to claim that all knowledge must come from recollection, and again we surely must come to the conclusion that even if recollection sometimes comes into the equation, it by no means can be the be-all and end-all of acquiring knowledge.

Many philosophers attempt to pick on Socrates' use of a diagram to show that clearly the slave-boy's acquisition of knowledge is an empirical process. The response of those trying to defend the demonstration is to claim that the diagram is unimportant, and in fact doesn't help him when it comes to the vital question of what effect doubling the sides will have on the area. These people point to the fact that the realisation that to halve a square you can cut it in two diagonally, producing two identical triangles, comes not from inspection of the diagram, but from understanding that with a square all four sides are equal, as are all four angles . This serves, however, only to set the question one stage further back; it may well be that the slave-boy knows that they must be exactly congruent triangles from his knowledge about squares, but this serves merely to draw into question where he came to know what he does about squares, and how the axioms that allow him to see the congruency of the triangles arises. This surely does not prove that he is recollecting that the triangles are the same, any more than he is recollecting the important features of a square.

Overall, therefore, Plato has certainly failed to prove that all knowledge must be 'acquired' through recollection. At best it might be said that he has picked up on a very important question, in asking how it is that we can develop knowledge where previously there appeared to be none, but his solution to the problem is not terribly satisfactory. If we are to accept that recollection is the source of knowledge, we have still not satisfactorily explained where the knowledge has actually appeared from - and if we are simply accepting that it is contained in our minds a priori then I cannot see how it is useful to introduce the word 'recollection' into the discussion.