no previous
no next

Later Socratic Dialogues 1948


Lecture 1 (8th July 1948)

Dialogues to be considered: Parmenides, Theaetetus, Sophist (1949), Philebus.

Reading: Burnet, Cornford, Taylor (Chapter 1 Parmenides, Zeno and Socrates; Proceedings of the Aristotelian Society 1916. See also Philosophical Studies.)

You have the general question over the later dialogues and their relation to the earlier dialogues and treating this especially in terms of the Burnet-Taylor view we can say that for Burnet the dialogues up to the Republic is the philosophy of Socrates and the Republic is quite considerably Socratic though having at least suggestions of Platonic views or problems. (N.B., opposing views of Fields and Hardie.)

Burnet emphasises the break, Taylor minimises it.

Taylor tends to discount any appearance of refutation that later dialogues exhibit. This is all the more a possible line because in certain matters especially on the ethical and political side Plato in the later dialogues does not seem to go far beyond the position of Socrates. And in the first place that helps anyone who so desires to discount the difference on the logical side and in the second place it has a tendency itself to reduce that logical difference - that is to say we find in later dialogues especially in the Sophist the materials for a logic fundamentally opposed to the Socratic but Plato himself is prevented by his ethical and political tendencies from working out this position to its logical conclusion.

Now this sort of question is important for the interpretation of the Parmenides. We have Taylor regarding it as a jeu d'esprit - taking the view that the theory of forms is sound - that any supposed refutation of it must be invalid and that the presumption at least is in favour of Plato's being aware of the fallacies of the argument that he puts in the mouth of Parmenides. But on the empirical view that the theory of Forms is false - that it can be validly refuted and if that is combined with the view that some of the arguments here attributed to Parmenides

  ― 2 ―
are valid - the presumption is in favour of Plato's being aware of their validity and thus of his rejecting the theory of Forms as Socrates held it.

Parmenides a reply to the Phaedo.

I would say that the dialogue is certainly the result of Plato's increased interest in Eleaticism and it amounts to this that, as against the view that there are two available theories of being, the Eleatic view and the Socratic (Pythagorean), these are not the only possibilities - they can be both rejected and for essentially the same reasons that neither solve the problem of the one and the many (unity and multiplicity) or of the relation between being and becoming. In fact the Eleatic dialectic as a form of criticism can be turned against Eleaticism. In the same way the Theaetetus is concerned with knowledge and may be taken as showing that the two previous theories of knowledge - the Sophistic and the Socratic - may be rejected, again for the same reason that neither can account for error or knowledge as a process and that, even if Plato is not quite clear on the matter. It is because each of them assumes fixities or certainties of thought which cannot be reconciled with uncertainty and inquiry and the unhistorical character of the dialogue (Socrates is again the leading speaker) is shown by the fact that we are presented with a secret doctrine of Protagoras - something said to have never been published but revealed to his intimates - an indication that Plato is trying to state the Sophistic view even more strongly than Protagoras did in order more conclusively to show its untenability. Secondly, by the fiction by which Socrates presents a doctrine revealed to him in a dream - the point being that this is a sign that it is a restatement of Socrates' position which of course he couldn't be represented as explicitly attacking. We can say that the appearance of Parmenides as the leading speaker in the Parmenides indicates that for Plato the Eleatic theory of being is superior to the Socratic theory but the second point shows that it too can't be maintained - and the reappearance of Socrates as the leading figure in the Theaetetus would similarly indicate that the Socratic theory of knowledge was better than the Sophistic although it is also open to fundamental criticism.

Sophist - leading figure the Eleatic stranger (unhistorical character for 1st time - Plato?)

Philebus - ethical theorising (hedonist view, intellectualist view). Socrates the leading speaker.

This indicates that the intellectualist view better than the hedonist but the final decision is a compromise embracing elements from both doctrines and the dialogue introduces much logical material which goes beyond anything in the early dialogues and couldn't be Socratic.


  ― 3 ―

Lecture 2 (15th July 1948)

The argument starts off from an account of Zeno's criticism of the doctrine that things are “a many”. Zeno contended this leads to contradictory results and incidentally remarks that he was trying to give the partisans of the many as good as they give - the point being that they had tried to show that Eleaticism led to absurdities - he in his turn was showing that Pythagoreanism led to just as great absurdities. It is a general feature of Rationalist criticism to dub views self-contradictory or self-refuting as against which the true view would be held to be self-establishing, ie., you get a division between the certain and the absurd taking the place of the empirical division between true and false and you'll notice that in the somewhat obscure and partially confused account that Socrates gives of method in the Phaedo the same points are made; i.e., the question is raised whether hypotheses do or do not lead to contradictory consequences (contradict one another); i.e., rationalistic refutation as against the empirical refutation which consists of showing that an hypothesis has consequences that contradict fact or what is assessed as fact by the persons arguing independently of the hypothesis and not as a consequence of it - which of course is the actual way in which Socrates proceeds in the dialogues and which is the account of the Socratic method presented by Burnet in Thales to Plato. It does not really establish Eleaticism if Zeno shows absurdities in the doctrine of multiplicity - just to set one group of arguments against another. The criticism is not complete until the unsoundness of one set of arguments has been shown and of course if we reject the notion of absurdity it would mean showing one of the conflicting hypotheses false because it has false consequences and that is connected again with the view that in fact both Pythagoreanism (Socraticism) and Eleaticism are false and untenable because they are both doctrines of ultimates and the real outcome is that there are no ultimates, either one or many - that there are only empirical or historical things, i.e., no irreducible or absolute realities - no “things-in-themselves” or no limits and hence it is suggested that that is really what Gorgias shows in supporting his thesis “There are none”, namely, no ultimate reality as contrasted with empirical reality though again that doesn't commit us to the sophistic conclusion that empirical reality is mind-dependent - that is only to treat mind as an absolute in the objectionable way.

One way in which the point could be put is if you formulate a notion by combining incompatible elements. For example where XeY:


then XYaZ

XeZ (argument depends on XiY)


then XYeZ

Unit: (indivisible magnitude) an incompatible … [?]

  ― 4 ―
It is perfectly easy to show many units coalesce into one but can criticise the One in the same way.

Now the particular arguments of Zeno on the many that is mentioned here is that “if things are many they must be both like and unlike”. What does this mean? If we are going to treat that as involving a contradiction then we have to be clear what is meant by “like” and “unlike”. In ordinary discussion we find no difficulty in saying that two things are both like and unlike - namely in the sense they have some characters in common and another character(s) in which they differ.

(1) AaX


(2) AaY


The point rather is that like and unlike are incomplete expressions. We should say A and B are alike in being X but are unlike in the fact that only one of them is Y. The general statement A is like B ennumeration of the term (A or B) without any predicate yet having been applied to it. However, Zeno's argument has more force than this - it is directed against the Pythagorean units each of which was just one or has unity as its whole character so that we couldn't distinguish one unit from another - would have to say that they were alike in all respects. On the other hand, if there are two of them we must be able to show some difference between them - they must be unlike if we are to say this is unit X and this is unit Y. Trouble is caused then because it isn't a question of respects in which things can resemble and differ from another but of total natures, of what a thing just is, and of course we get the same difficulty with any rationalistic theory. In Thales, e.g., if two things both just water - water the true substance or constitution then you can't find any difference between one and the other - in other words what we may call “substatialism” is a form of rationalism and is open to the general objections to rationalism. Now, it is partly to escape such difficulties that the later Pythagoreans and Socrates following them made the division between empirical and rational - that later Pythagoreans replaced the doctrine that things are numbers with the doctrine that things are like (approach) numbers - that what is just one or just six stands above experience (history) and things are to be described according as they approach one or other of these standards and Socrates adds to those mathematical standards (including the just equal) such standards as the just beautiful - aesthetic and moral standards - and thus we get the doctrine of forms - a doctrine which separates the empirical or historical from the unhistorial or absolute but in the same way as there is a difficulty (impossibility) of distinguisihing and relating what a thing absoutely and essentially is and what it relatively or accidentally is so there is the difficulty (impossibility) of distinguisihg or relating the two different realms - the realm of the historical

  ― 5 ―
and the realm of the unhistorical (becoming and being) - it is that difficulty that Socrates is vainly trying overcome in the Phaedo with the doctrine of participation and its that attempt that is now being shown up in Parmenides. What is being shown is that you don't reduce your difficulties by formulating those two realms. Empirically speaking there is no difficulty in saying A and B both like and unlike when we mean in different respects. It would be impossible to say things like and unlike in the same respect but this point is obscured by Socrates when he argues that things could not partake of one another because if we are dealing with the same respect there would be the same opposition whether we are concerned with particular examples, A and B, or with general possibilities, a question of comparability or otherwise of general characteristics. No difference in either case and if it is a question of different respects, i.e., likeness in respect of X no opposition to likeness in respect of Y and in fact the two cases are not distinct at all but both can be understood only empirically.

N.B. Socrates difficulties in Phaedo with relations in general (smaller and greater).

  ― 6 ―

Lecture 3 (22nd July 1948)

And this is linked up with the Socratic treatment of relations or with the fact that in spite of the Euthyphro he constantly uses relational predicates as if they were on the same footing as qualitative predicates and especially such predicates as great and small i nthe argument on generation from opposites and again in the final argument, an example which crops up again in the Parmenides and whereas there might be some force in speaking of beauty itself, or goodness itself, it is a very different matter to speak of greatness itself or smallness itself - these being unintelligible unless you get the other term of the relation. But of course if Socrates were to complete these expressions, if he were to make it a question not of being great as such but of being great in comparison with Simmias he couldn't very easily say that that was something that had its reality apart from particulars - or if we take his two important examples Beauty and Equality we could see some force in speaking of being beautiful or even the beautiful in itself which we couldn't see in speaking of the Equal itself - that which is just equal - because equality implies two things and even if we were to admit that never in sense-experience do we encounter things that are absolutely equal still the notion of absolute equality would be a notion of two things absolutely equal and couldn't be said to be conceivable apart from particulars and thus Socrates' use of relational examples actually weakens his case - gives something that obviously can't be thought about apart from particulars - though it can be said that even in the case of qualities his position will not hold - that it is impossible to think of absolute beauty except by thinking of something that is beautiful or of absolute humanity except by thinking of someone who is a man. The doctrine of the Universal apart from the particular is rendered plausible by the fact that there is no single particular which we have to think of in order to think of the universal or the kind and that leads us to the mistaken conclusion that we can think of them apart from all particulars.

Fallacy of composition:



does not give:


N.B. valid:





The confusion regarding relations and qualities is facilitated by the whole theory of Forms which can be said to give a relational account of all qualities - to treat being beautiful as partaking of the form beauty or coming under the form beauty and this is connected with the line of argument against Socrates that in the very attempt to speak of forms as entities apart from particulars he is making them particulars -

  ― 7 ―
he is making the particular and the form the two terms of a relation - two terms both required for the relation and he is making the relation indistinguishable from those relations which he must admit to exist between two particulars, e.g., Socrates is smaller than Simmias - something whose truth would seem to be wholly in the realm of particulars or becoming (sense-perception) - certainty not accounted for by saying that there is such a thing as smallness or more accurately comparative smallness.

Another way of expressing this criticism is to say that if there are to be two terms, Particular and Form, then each of these things must have characters of its own independent of the relation - otherwise nothing to enter into the relation and therefore Socrates is wrong in saying tha the particular has its whole character, is entirely determined by this relation. But if we admit that the particular has a character of its own apart from the form then we are rejecting the theory of forms as an explanation of predication or of the characters of things and when at the same time we treat the form as having characters of its own we are treating it as a particular and the whole theory of forms breaks down.

The position is that even in speaking of Forms we are speaking of them as if they were particulars - objects of experience having various characters like objects of sight and sound and in the same way the relations which Socrates puts up as holding between particulars and forms are themselves relations found in sense-experience. When in particular we speak of various beings having a share in something we are speaking of a matter of observation - we can see the partaker and the partaken of and we can see the partaking going on and what Parmenides is largely concerned to do in the opening part of the dialogue is to take the literal meaning of the Socratic formulations of the relations between particulars and forms and show that this cannot be what Socrates requires forhis theory because his theory requires a relation between two things on different levels (orders of beings) and no relation we are acquainted with can fulfil this requirement because all relations we are acquainted with are relations among particulars and themselves matters of observation just as particulars e.g., “copying” on which Taylor regards the criticism of Parmenides as unsound because it turns similarly on the question of likeness between copy and original and leaves the priority of the original - but to that we can say (i) that the copy or original has likeness with whatever consequences that involves; (ii) copying is a process we know empirically - we understand what we mean by one perceptible picture

  ― 8 ―
is a copy of another perceptible picture but doesn't help us to understand what is meant by saying that a perceptible particular is a copy of an imperceptible universal. Socrates has in fact to say that partaking, copying and so forth don't quite express his meaning - none of these empirical relations is the relation between the empirical and the rational entity which means then that Socrates is just not able to state his theory.

  ― 9 ―

Lecture 4 (27th July 1948)

The particular ways in which Parmenides brings out the weaknesses of the Socratic position are not all important in themselves but are at least important in showing that Socrates' position can't be coherently stated. If you take the original position on the question whether all things have forms or not you can see in the Socratic uncertainty the weakness in the doctrine of Forms, i.e., if simply a question of accounting for predication clearly the theory would apply to everything and those who say Socrates could easily have said this (thus avoiding the inconsistency) don't explain why he should have fallen into the inconsistency and it should be noticed that the things whose participation or non-participation in forms is in question are already “kinds”, i.e., the question is whether Man or Fire or Water is a form - again whether Air, Mud or Dirt have forms where it must be clearly understood that the so-called particulars in these cases have characters of their own so that the theory of Forms is not being offered as the solution of the problem of predication but as involving something additional - as involving some kind of value or worth so that worthless things like mud are not accorded a form though they have a character. These things however worthless they are taken to be must be just something or just nothing and so it is not simply in the interests of precision that the Forms are introduced or it is not precision that sensible things lack. Unless we were attaching some ethical or pseudo-ethical character and not instead a logical character to the Forms we would see that our logical problem was solved by not going beyond particulars - it is only because they are taken as in some sense higher that forms are spoken of as separate from particulars. That is, if not separate - if concerned just with the character things have then no sense in speaking of A Form - on the other hand if we do speak of A Form we are speaking of it as a particular something so even then we don't get beyond particulars.

N.B. In Phaedo oscillation between accidental and essential participation.

Now similar points come out in connection with the question of “sharing”, especially whether each particular has the whole of the form in it or has only a part of the form in it - when in ordinary speech we refer to participation we treat the thing partaken of as being broken into parts such that each partaker has a different part and that leaves the problem how having different parts of the same thing can involve - be what is meant by “having the same character” or “being alike”.

The position is similar to the situation in Stout's theory of universals,

  ― 10 ―
namely that qualities of particular things are particular so that if two particular things - e.g., a curtain and a carpet - are red then each has its own redness and the question is then how it comes about that we are able to say two things have a common character at all. Stout says the two characters have the same kind, belong to the distributive unity of red and that is why we seem to be able to attach the same predicate to the things though actually each has its own different character. (Cf., F. P. Ramsey, “Foundations of Mathematics” .)

The main lines of criticism are these: we seem only to have pushed the difficulty back a step - just as difficult for two qualities to be of the same kind as for two things to be of the same quality. Stout doesn't explain how we attach to the two things exactly the same predicate - how we can say each of them has a quality of the kind red without having to break up having a quality of the kind red into a number of different instances. A second difficulty of the distributive unity is this - the suggestion that we know the kind red as red1, red2, red3… and so on - the objection to all Nominalist or neo-nominalist doctrines that unless we have recognised the quality independently of the distributive unity we should never have been able to form this distributive unity - should never have known what terms to take together in this unity - applies exactly to the Berkeleyan theory of Universality. Or, to put it otherwise, that unless we had known that each of two things has a certain quality we should never have known that they resembled each other, e.g., similarity not prior to quality. In fact it might be said that we come to distinguish qualities sharply and especially to find names for them more in terms of differences than in terms of resemblances - by seeing a thing stands out from ints surroundings - by having a quality which they have not - than by seeing that in the ordinary way of putting [?] things a certain quality reappears or is repeated in a number of instances. I would say in fact that when Stout contends that the qualities of particular things are particular he doesn't show us what he means by calling one set of these things entities things and the other set qualities, and thus doesn't show how particular things may not be of the same kind without having to have an intermediate class of qualities distributed among them in the same way as he says that qualities can be of the same kind without themselves having qualities throught the medium of which they are brought under the kind; i.e., if he

  ― 11 ―
can avoid the regress at the stage of the qualities could equally have stopped it at the things without introducing what he calls particular qualities at all - on the other hand, if he takes this first step he will have to go on taking similar steps and there will always be something between whatever point he has arrived at and the generic unity or kind and that sort of difficulty comes out in the further part of Parmenides' criticism concerning likeness.

We can say that there is a sense in which the redness of the carpet is different from the redness of the curtain - the sense in which the redness of the carpet means the curtain's being red which is just a way of referring to the proposition “The curtain is red” and the state of affairs the curtain's being red is a different state of affairs from the carpet's being red. But to assert that is not in the least to assert that the curtain has red1, the carpet red2 - the two situations are different though they are both occurrences of redness - in that sense we could talk of the quality as a thing, i.e., could talk of “the carpet's being red” as a thing - could make that the subject of a proposition just as the carpet is the subject of a proposition, but this still doesn't mean that there are a number of different rednesses which we can call particular. It really means that there is no logical distinction between things and propositions, i.e., in what we call a thing there is implied the same distinction of place and character that is explicit in the proposition or, as I have put it, when we are talking of things we are talking of sorts of things.

  ― 12 ―

Lecture 5 (5th August 1948)

The difficulty about saying that the carpet and curtain have the same quality red is that this seems to imply on the other side of the Parmenidean dilemma a certain entity, the quality red which is completely in several places at the same time. That of course is the result of treating a quality as a thing and it is a position that is unavoidable if we speak about a form which is clearly meant to be a distinct entity - which in fact on the Socratic view is the only real entity. That of course leads Socrates into similar difficulties at the next stage - the Forms being instances or particulars of Reality or “real being” and the difficulty being to show that anything but the supreme Form “real being” exists at all - the conclusion to which the Eleatics would want to force him. So the Eleatic criticism applies alike to the many particulars which are said to come under a given form and the many forms which on the same showing would come under “formality” or “real being”. Now the question is if we didn't treat the quality as a thing which could either be broken up into separate pieces for the separate things which were said to have the quality with the consequence that they had nothing in common or could be said to be wholly in each particular with the consequence that the particulars would have to be identical or else it would be again a different things that would be embodied in each of them. The question is what we can treat it as being and the answer or part of the answer is that we can no more treat the thing as an independent and self-explanatory category, or a something that can stand by itself, than we can so treat the quality - that thing and quality amount to subject and predicate and there are different functions of the same entity the term or sort of thing or otherwise that instead of having on the one hand things and on the other hand qualities (two classes we could never get together again once we had separated them) we have complex situations in which place and character respectively are alwys themselves complex and embody a similar distinction. Thus (cf., “The Meaning of Good” ) the term red thing and the term red are the same term - the use of the expression thing meaning merely we are thinking of it as a subject, that is as locating something while when we use red simply this usually means that we are thinking of it as a predicate or describing something - but admitting that difference of emphasis the term in either case means “processes of a certain kind going on” and without including kind or generality in any term we couldn't have significance - couldn't say anything at all. Now that means that in recognising one situation we are recognising kind or being of a certain kind and we don't have to look for repetition or resemblance in order to recognise the kind or more correctly we have repetition already in the form of continuance - we

  ― 13 ―
have phases of a process (though the division may be rather arbitrary and in no case implies units) but still a process that could be divided into P1, P2, P3 etc., such that R in P1 passes into R in P2 and that into R in P3 and so without any question of a different red we do have this repetition - we do have a class of phases of the process - phases if we like which can be said to resemble one another but to suggest we started with comparison [?] would be to suggest that we have a unit phase - a phase of no duration - out of which the whole process can be built up, i.e., would be to fall into the Pythagorean error.

This is connected with the sort of question raised by Xenophanes concerning identification - how can we ever know that we meet the same person again - or see the same sum on two different occasions - how could we ever know that it is the same sun on two different occasions - how could we even know that it is the same sun before and after going behind a cloud and of course it must be admitted that we do make false identifications, e.g., conjuring depends on the fact that people tend to identify whatever went behind an obstacle with whatever comes out on the other side. But there is no reason for thinking that we never make true identifications or that we could have any coherent experience at all without recognising this sort of identity - without recognising the continued existence of what we have not continuously observed - but, of course, what enables us to do this is the recognition of the continuous existence of something we do continuously observe and this recognition of continuous existence is part of the recognition of fact or objective activity, in any way whatever. Now within this continuity we have resemblance but resemblance is still a notion subsequent to continued existence - you could say, I think, that for Alexander continuity and existence are the same thing and at the same time you might say that cyclic continuity or the continuance of a certain activity through phases of differing activities enables us to form the conception of interrupted continuity and thus to think that a thing subsequently observed not merely resembles a thing previously observed but is the same thing. But it is the same thing not in having an essential thinghood or inner substance which it preserves throughout a period (its “history”) but as being the same sort of thing - as maintaining the same character in continuous process so that in one sense Person X today is not the same as Person X yesterday namely that the X processes that we now observe actually different from the X processes we then observed but in another sense it is the same person namely there is an X process of which the X processes yesterday and the X processes today are parts or phases. Now Heraclitus would argue that this maintenance of character is always cyclical, that it is through ups and downs like that of sleeping and waking, that any persistent thing persists but also that in connection with each phase of the

  ― 14 ―
cycle there are cycles, that a person keeps awake or keeps normally [?] active by continuous interchange with his surroundings and it is in terms of such exchanges - in terms of law or formula of such cycles that we can speak of existence and character.

Question of Existence - bound up with question of kind (character). Presentation of some special character (interests). Stout - principal error situation of categorical names [?]. (Kind and existence.)

  ― 15 ―
If two orders of being (“reality”) then at least real difference. Question of third order of being; if so, no connection; but, if over-all reality admitted, different orders unnecessary.

Burnet takes Third Man argument to be an obscure way of expressing this criticism.

Lecture 6 (9th September 1948)

If we take the form as a standard and the particular as something that approaches this standard, that comes up to the standard to a greater or less extent then we are compelled to form the notion of something that fully comes up to the standard - of a perfect particular as contrasted with the imperfect particular on the one hand and the perfect form on the other hand - of an intelligible particular coming between the sensible particular and the intelligible form.

(Question of 2nd segment of the divided line; but should be a difficulty for the Socratic doctrine.)

And if we don't take this as breaking down the original distinction then we are involved in further difficulties - the difficulties in finding the relation between forms and intelligible particulars on the one hand and intelligible particulars and sensible particulars on the other hand without having to insert in each case some intermediate entity and so on indefinitely. Another way of putting it is that what intervenes of forms the connecting link between form and particular is the proposition - that between the particular Socrates and the universal man you have the notion of being a man which is applied to Socrates and others - you have the propositional function “X is a man” (or the class of propositions — is a man) and this would be the Third Man that these Eleatic critics are, however obscurely, referring to. Or putting it otherwise unless the proposition “Socrates is a man” shows that Socrates and man are terms of exactly the same order it would be unintelligible to say Socrates is a man - the notion of something having the human character would be one that we had to form unless we were to treat forms and particulars as occurring in different universes but that would not be a notion that we could understand. Essentially, then, the Third Man is a reproach or objection to the Socratic theory but it is also at times treated by Socrates as something he could embody in his theory - in the Republic the ideal soul and the ideal State are what we might call perfect or intelligible particulars and it is by this and not by the pure form that we are to judge and find the shortcomings of historical particulars

  ― 16 ―
—a natural enough development since we can't find the pure form and make it a standard or measuring rod without making it a particular; or again if we take the doctrine of the Phaedo we can say that the soul is the Third Man or intelligible particular - that it is the intermediate link or go between since it is supposed to be able to pass back and forward between the realm of being and the realm of becoming - since it is in the same sense a historical thing and in some sense an unhistorical thing. But if this passage shows anything it really shows that the distinction between the realm of being and becoming should not be made and if there were really separate realms of reality then nothing could pass between one and the other.

Question of forms fictional question, but brushed aside in Parmenides.

  ― 17 ―

Lecture 7 (16th September 1948)

The actual examples of regress that Parmenides uses are different from that. You could take the argument involving sharing to involve a regress, or Stout's theory of universals to involve a regress along the lines that just as the particular qualities of particular things are inserted between the things themselves and the kind, so you would have to insert qualities of qualities between the particular qualities of the kind and so on ad infinitum without getting any nearer being able to show how what was different could be the same, e.g., if the original conception of difference weren't such as to be able to include similarity - if we needed to go outside the thing at all in order to get the kind. And that illustrates the point that a regress is just a way of expressing an original contradiction or a duality of conception of something - the things shall we say in this instance being conceived of as both of the kind and not of the kind from which in theories of this sort the particular is distinguishable. But it is only in being of a kind that it can be called a particular (participator) at all - not just as a particular without qualification, it is a particular X. There is the same problem in the Phaedo, in the Socratic treatment of a thing as being entitled to be called X because it comes under or partakes of the form X - the point being that we can't distinguish the particular from the form nor a particular which partook of X from a particular which didn't unless there was something which partook of X in the particular itself by which we distinguish it and if there were it would be by that character of its own that we should describe it and not by its relation to something else outside it. Therefore the treatment of the form as broken up in various particulars would give us the same problem of bringing the various parts of the form under the form as we had in bringing the various particulars under the form, and in that way we would be led on to an infinite regress apart from special difficulties raised by Parmenides of how the presence of predicability of a part of the form would enable us in any way to predicate the whole form of the thing.

The next kind of regress that Parmenides brings up is that of the similarity of the form to the members of the class whose similarity it is intended to explain which gives us a second class of similars - namely the previous class plus the forms whose similarity will require a further form to explain it and so on ad infinitum. Now this is where Taylor raises the difficulty regarding the confusing of the asymmetrical relation of copying with the symmetrical relation of likeness - a difficulty which however is met by the fact that where the asymmetrical relation exists the symmetrical relation

  ― 18 ―
also exists and any problem there might be of explaining this second relation (resemblance) would have to be faced. Secondly, Taylor objects that the form is not predicated of itself or not in the same way as it is predicated of particulars and so the question is not one of resemblance (having the same predicate in the same manner). But here we must notice that the Parmenides is looking back at the Phaedo where it is quite clearly implied that the form is predicable of itself as of particulars. Socrates says if anything besides beauty itself is beautiful it is beauty itself that makes it beautiful - the suggestion being that the beautiful object derives its beauty not just from beauty itself but from the beauty of beauty itself - where in fact the phrase that is rendered “beauty itself” or “absolute beauty” could also be rendered the “just or perfectly beautiful” (whole nature to be beautiful); the point being that there is no clear distinction between the form and the perfect particular and indeed that there can't be so long as we talk about the form as something specific, something particular. So that Parmenides is bringing out difficulties that Socrates has already involved himself in and that he must involve himself in as soon as he talks about this or that form instead of just postulating a proposition. And the regress here could be said to arise from the contradiction or duality of conception of the form itself, namely the attempted conception of it as something “above” particulars and yet the necessity of regarding it as a particular involved in saying “it” at all.

I have suggested then that a regress is just a way of bringing out an initial contradiction by leaning [?] now on one side and now on another - a contradiction if you like between the declaration that something is unintelligible and the unavoidable answer that it is intelligible. (Cf., “Universals and Occurrences” ).

With reference to the arguments of Mr. Merrylees who takes it that the assertion of “X is Y” must be understood as “X is expressive of (the Y)” - the universal or form Y, I argue that if, in order to make “X is Y” intelligible we have to say X is expressive of the expressive of the Y… ad infinitum. So that we never arrive at the exact interpretation of the assertion. If, on the other hand, we are supposed to stop at the 2nd stage - if “X is expressive of Y” is intelligible with one interpretation then we could have stopped at “X is Y” - taken it as intelligible without further interpretation. Once again an initial contradiction - it is contended that the propositional form needs interpretation but in the very contending it is implied that the propositional form doesn't and can be understood as it stands.

  ― 19 ―
N.B. difficulty about relations raised by Parmenides. For example, smaller than Simmias: particulars but whole esp. [?] of same generality as qualitative predicate. Doctrine of Forms rejects situational logic: anything a situation, involves relations. Emphasises the separate term.

Lecture 8 (23rd September 1948)

I have contended that the arguments put forward by Parmenides even if they are not all effective, nevertheless are quite sufficient to dispose of the doctrine of forms and, in general, to dispose of dualism and this would be to set up the alternatives of a complete empiricism - a doctrine of the sensible as the only reality or as we may call it, a complete idealism denying the sensible altogether. It is nothing againt this presentation that it leads us in the empirical direction since as we have seen it is still only in empirical terms that we could understand what Parmenides says not merely in opposition to Pythagoreanism but in support of his own position which means that the only anti-empirical alternative very soon breaks down and we are driven back on the empirical.

The longer and the main position of the Parmenides does show the breakdown of Eleaticism - the impossibility of stating it consistently - the position being stated with an extraordinary elaboration which might lead us to think that Plato when he wrote the dialogue wasn't very sure of his own position, but which still shows that Eleaticism fails on the same sort of grounds as those on which it had destroyed Socraticism (Pythagoreanism) - the position being capable of fairly simple statement. We may take it that Plato was concerned at this period with questions of method and that he had not arrived at a very satisfactory position. That is, he is not satisfied with the sort of position that he attributes to Socrates in the Phaedo, yet hasn't got a clear alternative procedure or hasn't realised the implications of his alternative procedure especially its empiricist implications. First of all just as in the Phaedo we have the talk about the consequences of a hypothesis without its being realised that these are not consequences of the hypothesis alone but follow from the hypotheses together with other propositions which are assumed to be matters of positive information.

And when we take consequences in this way it has to be noted that we may take as the consequences of a hypothesis what are not its consequences at all because the other proposition which we were certain was true is actually not true. This is specially important in regard to Falsification of a hypothesis because when we say

  ― 20 ―
the hypothesis AaB has been falsified because along with the true proposition XaA implies the false conclusion XaB we are taking as matters of definite knowledge XaA and XoB (? AoB) which prove the hypothesis false. But if one of our premises itself were false then AoB is not proved. That is, AaB is not destroyed or falsified. The point that Parmenides specially insists on is that of seeing not merely what follows from a hypothesis but what follows from denying it (what is true if the hypothesis is false). Now here again the case of falsification would be the most important but of course the falsification of the contradictory of the hypothesis is actually the proof of that hypothesis which then properly speaking is not a hypothesis but a conclusion - is no longer being tested by its consequences but is being by premises independently known. Hence in effect when Parmenides says we should consider the consequences both of a proposition's being true and its being false he is saying that we should both test a proposition of which we are doubtful by its consequences and try to prove it from propositions of which we are not doubtful. Suppose again our hypothesis is false then we are staring from the hyposthesis AoB and along with the true premise AaY drawing the conclusion YoB so if we get a falsification here (i.e., YaB) which gives us two data AaY and YaB ? AaB is proved and this is the hypothesis we started from so that we need no longer treat it as a hypothesis. It is misleading then to talk about the method of testing both a proposition and its contradictory - what we should speak about are two distinct questions, Testing and Proving.

Turning very briefly to the hypotheses which are professedly shown to have contradicted [?] any conclusions we find them in their positive form “if it is true” and “if one is” and whatever may be meant by that distinction the vital point is that in Eleaticism we have the same sort of an initial contradiction as was seen to lead to an infinite regress in Pythagoreanism - namely that the Eleatic one must be and can't be distinguished from its own being - that Parmenides wants to say that this entity is and yet also wants to say this entity is the very same thing as being in which case it would be meaningless to say “it is”. That sort of contradiction runs right through the pre-Socratics, all of whom want to identify being material or being of a certain substance or, as we may put it, want to take being as a term not as the copula. This implies an initial contradiction which leads to further contradictions when we try to work with it logically - to treat it as if it were a definite or specific thing. In falsifying the conditions of discourse they are led into contradictions by the fact that they can't help using this discourse.

Cf. Naturalistic fallacy - same thing both identified and distinguished.

Theaetetus: Theory of Knowledge

  ― 21 ―

Lecture 9 (30th September 1948)

See Burnet, Taylor and Cornford. You couldn't have a worked out theory of knowledge in the Eleatic theory of reality because knowledge implies distinction between knower and known which Parmenides couldn't admit (though he couldn't consistently deny it). He does make the superficial distinction between the way of truth and the way of opinion, so there can be false opinion but only inconsistently can he admit this even with the purpose of refuting it. On strict Eleaticism there couldn't even be such a theory. Still you do get in this distinction of Parmenides the suggestion of a theory of knowledge and you do find the question of knowledge taken up by the 5th century philosophers such as Empedocles and Anaxagoras, though of course they take up the question in a physiological way and not so as to bring out and, if possible, solve the logical question. (Democritus also has theory of knowledge - very crude and unphilosophical.) What we can say about the Sophists (as Burnet says) is that they were influenced by the great variety of philosophical theories to adopt the view that there could be no settled truth in these matters - no independent reality that we can know but simply appearances, simply what we accept - but I think it can be said that they were influenced most of all by Eleaticism as providing a kind of criticism that could be turned against any theory. In that way Gorgias can be llinked with Parmenides and of course the [?] Euthydemus gives a further example of universal refutation that comes from Eleaticism and also the positive doctrine of Eleaticism that there is no multiplicity, no motion or change, is something that no one could possibly believe. If that is the outcome of rigorous argument the only thing to do is to abandon rigorous argument and go back to persuasion or what various people find plausible. Now of course this is the sort of position that has been met by modern realism and is already largely met by Socrates - by the contention, namely, that even to say something seems so to us is to admit the conception of being so or of truth and the particular truth that it does seem so to us and that even to say we are persuaded by an argument which is no rigorous (i.e., is plausible) is to imply the possibility of rigorous argument, it to imply the reality of implication so that the relativist or humanist position can't be maintained (see “Marxist Philosophy” ). In fact we find Socrates arguing against the doctrine of convention or nomos

  ― 22 ―
that there is a nature or physis in human affairs as much if not more than other affairs, i.e., something that is just so - though Socrates for the most part confuses the issue by taking his “nature” as something behind or beyond experience, his being as something as beyond becoming instead of something in becoming. The Sophists then are right in holding that if we are to speak of anything at all it must be something that appears to us but wrong in taking appearance or relativity to us as the character of what does appear - taking it to exist in so far as it is ours. Now coming to the dialogue in which Socratism and Sophistry are confronted on the theory of knowledge we find the position taken up that knowledge is perception or sensations - that it is aesthesis - which really means something immediate or a faculty that presents us with immediate objects, a sort of usage borne out in the Kantian expression transcendental aesthetic for those conditions of our knowledge which he thinks to be immediate or intuitional and also in the use of the term “aesthetics” for what was thought to be immediate feeling as contrasted with discursive thought. The general distinction of Logic, Ethics and Aesthetics corresponding in theories of this kind to the distinction of the faculties of reason, will and feeling - first definitely separated in the Kantian theory (Cognition, Conation, Feeling). Now Socrates identifies the doctrine (hypothesis) that knowledge is sensation with the doctrine of Protagoras that “man is the measure of all things”, that “I perceive this” is the same as “this appears to me” and that again as “this is so to me” or “this is to me” and similarly what you perceive is to you and the meaning of being and knowledge alike is being to a person or to a mind. Now it is to be noticed in connection with this identification (of being perceived with relative existence) that it depends on the insistence of Socrates in treating knowledge as meaning certainty - that he assumes that to ask what is knowledge is to ask what is there of which there can be no doubt and one answer (cf., Theaetetus) is the immediate data of experience - what is given or impressed on our minds and the realist (empiricist) criticism of this is that there is nothing immediate - nothing immediate in this sense “impressed on our minds”. The least we can know - the least we can say we perceive or sense is a proposition or complex situation and that is bound up with the formal possibility of contradiction, with the denial that anything is certain in itself. But Socrates never even considers the possibility of holding that there is no

  ― 23 ―
intelligible type of cognition and yet admitting that in sense perception we can sometimes get truth. We can know what is the case and I would argue that the break-down in the dialogue of the two theories of knowledge proposed is due just to this fact that we cannot in any such theory establish a basis of certainty - that we can't in any such way get the sort of thing that Socrates describes as knowledge - the upshot being that there is no such thing - this knowledge can be usefully understood not as being above all possibility of error - as having arrived at the safe and certain but just as not as a matter of fact being mistaken in a given instance. Now Socrates goes on to develop the position of Protagoras, maintaining that while Protagoras openly contended that “what I perceive is to me” he held a secret doctrine which he revealed only to his closest followers that nothing is, that everything becomes or moves (is in motion) and it is in terms of the theory of universal motion that an account of sensation and of the relativity of knowledge is to be given.

  ― 24 ―

Lecture 10 (7th October 1948)

If we take the meaning of sensation in a modern terminology knowledge is sensation would mean the reception of a datum, something simple and given in regard to which it might be said we can't be wrong though equally it could be said we can't be right because the datum is not supposed to be complex or propositional from which it could be argued that it isn't known at all and that it gives no assistance whatever in the treatment of matters in which we can be wrong which any theory must admit in some sense to exist. In fact that is one line of the Socratic criticism of the Protagorean view attributed to Theaetetus, namely that even a doctrine of data must admit matters in which we can be wrong but cannot account for them. Now this may have been what Theaetetus meant - he may have meant “seeing is believing” in the special sense of being quite certain but if we rendered aesthesis as perception, if we took the view that knowledge is perception there would be no reason at all for identifying it with the view that knowledge is relative to the individuals because there is no reason whatever for saying that perception is relative to the individual, that perception is of something peculiarly mine and couldn't possibly be yours. And while on any view we might have to admit a knowledge of something not immediately confronting us, e.g., as we have a knowledge of some things that happened yesterday, that doesn't exclude the view that perception or knowledge of what is immediately in front of us occupies a specially important position in the theory of knowledge - that is something which any knowledge either is or is in some sense based on. [Note inserted by DMA: I take this to be base empiricism - connect with JA's confusion about Universal connection.]

I was referring then to what Socrates presents as the secret doctrine of Protagoras, namely, as against the superficial view that what appears is, that nothing is but everything proceeds or goes on and Socrates presents as this secret doctrine an account of the processes which take place when we perceive - processes which however he, in this theory, identifies with what is perceived so that not merely is it said that when we perceive a white colour there are movements from us to something outside of us and movements from that thing to us (our eyes specifically) but it is assumed that these processes, these cyclic exchanges (Heraclitus) are the white colour or that this cycle is some sense both whiteness in the thing and seeing in us and this is supposed to explain why different people have different perceptions and also why the same person has different perceptions under different conditions, e.g., dreaming, insanity, diseased, i.e., this relativist theory of knowledge would speak of someone becoming sensible which means his becoming sensible of something and something becoming sensible which means its becoming sensible to someone - sensation then or perception always involves exchanges and difference of sensation is explained by the exchanges being different. It is contended, however, that

  ― 25 ―
while on this theory Protagoras would not say that one view was truer than another but only that it is relative to different conditions he would say that some views are better than others and that there is a distinction between the wise and the unwise in that the wise is he to whom what is better appears and who can make it appear also to others. Now even if Protagoras didn't hold a theory of motion such as is attributed to him here he did hold a theory of “usefulness” and took teaching to be the inculcation of better or more useful beliefs, beliefs which are relative to a better state and we can accept the arguments of Socrates against such a theory of relative truth even if we consider that the theory of motion - of knowledge as involving cyclical exchanges is not invalidated at all and similarly with the theory that knowledge is perception or at least that perception is knowledge, even if we think that there is no ground for combining any of these views with relativism.

The position is that in putting forward his doctrine of useful beliefs Protagoras is committing himself to a different sort of distinction between truth and falsity - to the doctrine that certain views are better or are more useful - not meaning that it is more useful to take them as more useful and so on indefinitely - a regress which would prevent Protagoras from even stating his position. In other words here again we have an initial contradiction, a contradiction between maintaining truth is usefulness and the meaning of truth involved in that very case.

Now in order to say that the true is the useful Protagoras has had to admit the existence of absolute facts, to admit the existence of the true as actual or occurring and the same is the case with the doctrine of motion or at least with the attempt to import relativism into it. If, for the sake of argument, we admit that knowledge involves interaction between mind and something else then in order to say this we have to assume a knowledge both of this mind and of this other thing and no simply a knowledge of what passes between the two. To say that A knows X only when there are motions from A to X and motions from X to A, i.e., only when there is a cycle passing through A and X is not the least reason for saying that it is these passages or intermediate motions which are known and not X itself or in other words knowledge of X depends on interaction with X doesn't imply knowledge of X is knowledge of that interaction of what in this theory is called the swifter motions as contrasted with the slower motions and in fact it regularly happens that we know let us say a book without having any knowledge of the processes, e.g., physical processes going on between us and the book and if we did know these intervening processes it would be on this theory because there were further processes or cycles linking us and them, but that would be no reason for saying that we knew those further intervening conditions.

(Cf., Kemp Smith, Prolegomena. - must make realist assumption, e.g., on “subjectivist” view must assume [?] of sense organs (not “mediated”).)

  ― 26 ―
To sum up the realist will not deny that what particular thing a person knows depends both on his own state at the time and on the state of the thing with which he is confronted but the fact that my knowing X depends on, or has conditions, doesn't make it relative or dependent knowledge - does nothing to show X isn't an independent thing.

Realists will still admit the possibility of error, will admit that the change and complexity of things along with change and complexity in ourselves or the variety and development of our interests - these sometimes lead us to believe what is not the case but even so, even admitting the occurrence of error, this is to have some recognition of what error is, and what brings it about and that means a direct knowledge of some facts even if the person who knows them continues to be in error in other things. But we certainly could give no account of error on the theory of relative knowledge - on the theory that our objects are just something that characterises ourselves at a given time or thqat characterised our interactions with other things - on that showing there wouldn't be error.

  ― 27 ―

Lecture 10 (7th October 1948)

(Any account of error assumes certain account of knowledge that is not erroneous, so no ground for “relativity of knowledge”.)

The reason why the question of error with the associated questions regarding appearance is evaded in the discussion lies in the assumption already mentioned with which the whole dialogue begins, namely that what we are looking for as knowledge is some type of cognition (the exercise of some faculty) which always gives us truth, which is certain and incapable of error, as against the recognition of the fact which I take to be part of empiricism that it is by the exercise of the very same faculty that we on the one hand know rightly and on the other hand make mistakes. Indeed whatever may have been Plato's precise view of the matter I take the actual outcome of the dialogue to be that if what we are seeking as knowledge is certainty then we cannot work out any theory of knowledge just as it is the outcomes of the Parmenides that if we set up anything ultimate or self-sustaining in the way of being we cannot work out any theory of being.

Realising apparently that there is a difference between the theory of Heraclitus and Protagoras, Socrates further examines the claim of the theory of universal motion to give an account of knowledge - the argument as presented by Burnet as follows.

“When we say everything moves what do we mean by moves? There are two forms of motion, first motion from place to place and second motion from state to state, in other words motion is either locomotion or alteration and if motion is universal it must include both. Since then everything not only moves its place but also alters its state we cannot ascribe any quality to what moves for what we call qualities are nothing but perpetual processes going on between what acts and what is acted upon and accordingly in the vewy moment of being named, the quality is gone.” (Thales to Plato, p. 245.)

(N.B. Sophistical character of the argument: if motion is universal it must include both, but view [?] in Heracliteanism.

Leaving aside the distinction in Heraclitean theory between persistence as involving equality of exchanges and change as involving inequality of exchanges we can advance at the outset a general criticism of the Socratic view here - namely that if a thing had to stay in order to be named - if whatever could be said to be known must be something abiding then the very expressions “motion”, “becoming”, “change”, “process” would be meaningless so that Socrates would not merely have to deny that everything is in motion but would have to deny that the words “change of state” could convey

  ― 28 ―
anything and this criticism emphasises again the importance of the Parmenides in opposition to the doctrine of the Phaedo - the doctrine of the secondary or derivative or incomplete being to what becomes or of process - the point being that unless we accord complete being to what becomes we can give no account of its distinction from or relation to what is but doesn't become, to the true being of the forms. And I would argue this is to deny being which doesn't become, to assert with Heraclitus that true being belongs to becoming or to process.

The next part of the argument against the view that “knowledge is sensation” is the argument from “common sensibles” - that is from things cognised by various senses or at least common to things cognised by various senses. In terms of senses we are said to see colours and hear sounds but if we are aware of anything common to a colour and a sound, this, it is argued, cannot be given by either seeing or hearing but must be due to some other faculty, something we might call “thought” so that there is knowledge other than sensible knowledge and therefore knowledge cannot be identified with sensation. Such characters are same, being, other, like and unlike, also unity and number, odd and even also aesthetic and ethical predicates good and bad, fair and foul. For example, we say sound and colour both are or colour is other than sound or is the same as itself. As Burnet puts it “not one of these common properties has any specific instrument by which it is apprehended as was the case with such properties as sweetness, hardness etc. - it seems rather that in these cases the soul is its own instrument and acts by itself”. Here we have a distinction between sensible qualities ( [?]) known by means of bodily affections and the apprehension of common qualities by means of comparisons and reflections, processes which take place within the soul itself. (As Burnet remarks here we have the beginning of a theory of categories in place of a theory of form.) These categories are recognised by the soul without the help of sense and are used by it to organise the manifold of sense and this kind of cognition leads us to reject the hypothesis that knowledge is sensation and to consider that knowledge resides in the activity of the soul when it is concerned with what is, with being and truth which cannot be apprehended by the affections of the body - that activity is called judgment and the second part of the dialogue is concerned with whether knowledge can be identified [?] with judgment. (i.e., modification of the term used for “opinion” in the Republic.)

It should be noted (cf., Phaedo) that the particulars are not supposed to be known by the body but to be known by the soul through the body or using the body but the forms are known by the soul in itself or using

  ― 29 ―
its own resources and here also the soul would seem to be what knows anything whatever, but if that is so what it knows by means of would seem to be unimportant, would not seem to imply different kinds of knowing since in each case the soul is concerned with something external to itself. It doesn't matter whether the soul is knowing Forms, properties, categories, sounds or colours and it doesn't matter what instruments it is using, we don't have any ground for making a distinction among cognitions as such.

Now if we take what is called an act of comparison or reflection, take the judgment “Sound is not colour”, then whatever is acquainted with that fact knows difference or otherness, knows sound and knows colour. If it were one act of mind that knew sound and another that knew colour and a third that knew otherness or difference there would be nothing to know the fact that sound is not colour.

But in point of fact we don't merely know differences between sound and colour and so on, we know differences within sound and within colour - we know in what are called seeing and hearing not simple sensations or data but complex situations and it is only because we do that any further coordination - any knowledge of additional connections and distinctions - is possible. The very admission that some objects have likeness and unlikeness to one another is an admission that they are complex - that they have not a single quality X but many qualities - in fact what we are said to see by means of our eyes is not colour by itself but coloured things, things having colour or being of this or that coloured kind, i.e., situations or propositions and that is why we can connect what we see with what we touch, why we can say the thing I touch is the thing I see and this is opposed to the view that we recognise one and only one type of quality by means of a given sense, the point being that when we use our eyes we can make certain discriminations better than we can make others (particularly discriminations of colour) but we still see a thing as having a variety of characteristics not just as being coloured and this means that sense knowledge itself is what Plato is here calling judgment - that is knowledge of propositions, so that the occurrence of judgment does nothing to show that there is any type of knowledge other than sense knowledge.

  ― 30 ―
Question why ethical and aesthetic characters included in “common sensibles”. Suggests they are comparisons; Socrates ethical judgments essentially comparative. N.B. Phaedo [?] any quality - question of more or less?

Lecture 11 (21st October 1948)

According to Socrates we have certain simple sensations given to the soul which then makes comparisons and discriminations among them but on the view that we know nothing less than a situation these primary data would be denied and anything that could be called sensory would be judging or awareness [?] of a situation, awareness of something that already embodies the categories - at any rate embodies the form of the proposition as well as particular terms. On the theory of the pure datum on the one hand while we couldn't have a false sensation we equally couldn't have a true sensation, i.e., we wouldn't have knowledge at all, not having a complex object, and, as already said, we couldn't connect a … [?] knowledge of with an actual knowledge and mistakes about complexes. The position is the same whether the simple entities from which we are supposed to start are recognised as sensations or intelligible objects and this point is brought out in the discussion of knowledge as thought when Socrates refers to a theory which has been revealed to him in a dream, a theory of intelligible elements of known complexes - this dream theory being parallel to the secret of Protagoras with the suggestion in both cases that Plato is going beyond the doctrine that the [?] actually held - trying to see whether they can't be made consistent with a slight modification and that being found impossible in both cases the upshot of the dialogue is that we can't work out any theory of certain or infallible knowledge - that we have always to admit the formal possibility of error - the possibility of contradicting any proposition so that there is no generic distinction between the true and the false - no character whereby we recognise the true wherever we see it, but we have in each case to find out the concrete truths ourselves and at the risk of error. The position regarding simple entities (sensible or intelligible) is similar to that of Locke's simple ideas - namely that no proposition can be asserted about them except possibly that any one of them is not any other. The discussion in the later part of the Theaetetus shows the impossibility of asserting any affirmative relation between any simple and another and even to assert a negative relation is to assert a situation in which they both occur - in fact is to imply the complexity. If we look back to the Phaedo we see

  ― 31 ―
that Socrates can have affirmative predication only by admitting that forms are complex as when he says that three is odd, both 3 and odd being general terms which have various instances and as regards negative predication if we stick to simple forms and say that any one is not any other we have to drop the Socratic notion of specially opposite forms, a notion which depends on a propositoinal logic with the possibility of obversion [?]. Now, if admitting that there can be no relation between simples we recognise that all knowledge is of situations the distinction between sensibles and intelligibles loses all point.

As regards the contention that the dream theory in the theory of Socrates himself modified so as to remove certain inconsistencies Burnet and Taylor though admitting the parallel with the secret doctrine of Protagoras do not regard this as a secret doctrine of Socrates, i.e., do not regard the doctrine of intelligible elements as a logical development of the theory of forms having to be presented in a roundabout way precisely because Socrates is conducting the argument. They suggest it is the theory of some Pythagorean most probably Ecphratus of Syracuse. Now obviously the theory has a Pythagorean character - you might even say that the Socratic theory is being reformulated in Pythagorean terms just as the theory of Protagoras was reformulated in Heraclitean terms but on their own showing the Socratic doctrine was always of Pythagorean character - in fact Burnet maintains that Socrates was looked up to as a leader by the Pythagorean schools in Greece after the departure of Philolaus. If then the parallel between the secret doctrine of Protagoras and the dream of Socrates is to have any point the latter must be regarded as an attempt to restate Socraticism in a more consistent form and while the problem of the relation of elements to one another and the complexes they are said to compose is certainly one that affects Pythagoreanism it also applies to Socrates in the Phaedo who, according to Burnet, is endeavouring to overcome the separation between form and particulars by treating the particular as a meeting place of forms - as being constituted by the coming together of various forms just as in the Theaetetus a syllable is said to be constituted by the coming together of various letters. Now the difficulty of the doctrine of the Phaedo, e.g., that on this view there is still no such thing as a particular, has been criticised with reference to the Parmenides but the main point is that if individual forms are what is, then no relation whatever, e.g., coming together, can be asserted between one form and another, or that if we start from unitary beings, rational entities, things which are in themselves then we can give no account of situation. In the last part of the Theaetetus it is shown that if we take the so-called letter as ultimate or being in themselves we can give no account of so-called syllables (complexes of elements) and we have to reject the notion of

  ― 32 ―
that which is in itself and take a propositional view of things as is done to some extent later in the Sophist. The position is that intellectual knowledge, knowledge by way of thought, the knowledge that soul is supposed to have in itself must be like sense knowledge, knowledge of situations and must be equally liable to error. The suggestion is made that error arises not in pure sensation or in pure thought, but in a maladjustment of the two, so that a present situation is referred to the wrong thought, e.g., when being acquainted with both Theaetetus and Theodorus and seeing the two of them in the distance we think it is the other. But apart from other difficulties it is pointed out in the dialogue that people make mistakes when they are entirely concerned with what are called thought objects, e.g., mistakes in pure calculation (5+7=11) working with not things but numbers themselves. From the point of view of the logic of situations the position would be that we are always concerned with objective situations and 5+7=12 is a situation of the same order and one we could just as consistently … [?] about as the situation Theaetetus and Theodorus are both present or Theaetetus is in the distance.

This line of argument is closely connected with Hume's theory of belief, namely, that belief is constituted by a lively idea associated with or present impression, impression being immediate perception and ideas reproductions of such impressions (real [?] in front of us, not in front of me). The point is that we see how a belief that is entirely composed of what Hume called ideas, e.g., “Caesar crossed the Rubicon” where it is not a question of association between something given and something reproduced or recalled.

We could say in fact that there are in the Theaetetus at least indications for the refutation of all the theories of knowledge of the modern period (Descartes to Hegel), for the refutation, e.g., of a view like Locke's that there are simple ideas or impressions - atoms of cognition so to speak - out of which complexes may be constructed. The point is that if impressions mean anything an impression if a particular thing and incidentally a complex thing and as against any correspondence theory to know such an impression is not to know something else of which it is the impression. The criticism is not fully developed in Theaetetus but at least the insoluble difficulties of a correspondence theory are shown and we are driven towards the solution that all cognition if of complex situations and that a distinction between inner and outer if it has any truth at all is merely a distinction between the location of situations and not between kinds of situation or kinds of knowledge we could have of them.

Lecture 12 (28th October 1948)

The hypothesis that knowledge is thought doesn't meet the difficulties because in the case of what are considered to be specially thought objects there is no possibility of error any more than with pure data so that we still have not a theory covering the facts of cognition. But when the suggestion is made that error may occur in a false relation between a thought-object and sense-object this means that the hypothesis that knowledge is thought is already abandoned and no account whatever has been given of the kind of cognition that is not certain. And when further it is admitted that we can be in error even in the case of what are called objects of thought (e.g., 5+7=11) we can no longer speak of error as concerned with the relation between a thought and a sense object. We have also seen that even the latter type of error - reference of a sensation to a wrong memory image - is not really explained by the correspondence theory apart from the fact that memory images are different from the “common predicates” which were first formulated as objects of thought and further objections to a correspondence theory appear, e.g., the “wrong reference” is due to the fact that image was not sharply impressed to begin with or has been worn away in the course of time; then the present representation would be a quite correct one - we should be correct in saying that the present sensation resembles the image as it is now and error would arise only in terms of the things perceived at one time and another - only if we said this thing which I now see is the same as the thing which I saw then, i.e., judgment would be concerned with situations and the mistaking of situations wouldn't be accounted for by any doctrine of images - of things intervening between us and external realities. If, in fact, there are internal objects then knowledge of a relation between an external object and an internal object is on exactly the same footing as knowledge of a relation between two external objects or two internal objects - that is, it is knowledge of a situation.

Now before coming on the theory of elements of thought Socrates suggests that error in pure thinking (5+7=11) may be explained by distinguishing between having knowledge and possessing knowledge, e.g., we may possess a coat without having it on and here we have the dovecote simile - the suggestion of various pieces of knowledge flying about in our minds and of our trying to catch them and catching the wrong one; but then when we do catch one we should know what it is and if we were said to have pieces [?] of error in the mind as well we should still know one of them when we got it and the fact that it was different from another one wouldn't matter. What all this leads to is the rejection of the view that knowledge is in separate pieces - of the very sort of view that Socrates proceeds to bring up, as contrasted with the view that any knowledge

  ― 34 ―
is of situations and that any constituent of situations is itself a situation - is not an ultimate element. Theaetetus, then, admitting that we can't identify knowledge with true judgment suggests Knowledge is True Judgment with a reason (doxos), account, or explanation of it and this leads on to the dream theory of Socrates that there are elements which are the basis of all knowledge - which are the objects of simple apprehension and have to be put together to form a judgment so that there elements are the reasons or ground of the judgment and putting them together is discourse or judgment itself - just as we are able to discourse in the sense of speaking by putting letters together to form syllables. Now this hypothesis of letters and syllables is cogently criticised in the dialogue - the criticism as summed up by Burnet being that either the syllables are only the sum of the letters so they won't be anymore unknowable than the letters are or else it itself is an indivisible unity in which case it would be the object of simple apprehension and we would be no further forward.

Now one way in which this latest hypothesis is formulated is that to know anything is to know it with its differentia which is just one way of expressing the position involved in the Socratic distinction between Knowledge and Opinion - that we have to show not only that something is so but why it is so and it is thus the Socratic position that is here being criticised by Socrates. It would be straining verisimilitude to make Socrates say it was his position but it is precisely against a doctrine of the kind he held that the criticism applies namely that knowledge of the ground or reason requires the same sort of explanation as the theory it was supposed to explain. To say that we have discovered grounds or a cause of something is not to say that it is any more certain than the objects of opinion - if to be an object of knowledge is to be deduced from premises that means that these premises will need to have grounds assigned to them ad infinitum - i.e., it would appear that they are objects of opinion and thus knowledge is grounded in opinion - a thing is shown to be certain by its dependence on the uncertain. Now one way to take to avoid this difficulty is to argue that there are self-grounded propositions on which all demonstration depends which will mean that it is these self-evident entities that are the objects of knowledge but this brings us back to the same sort of difficulty as in the doctrine of elements: firstly, how we can describe the self-evident at all and secondly how we can derive anything from it, how if fact we can derive complexity from simplicity. It would in fact be impossible to derive anything but the self-evident from the self-evident - a poitn which involves

  ― 35 ―
a rejection of Taylor's distinction between historical and demonstrative science - a rejection in fact of demonstative science and with it of self-evident propositions themselves. The whole doctrine of Rational Science involves the attempt to find some non-propositional knowledge on which propositional knowledge can be based - to find as in the theory of Leibniz some essence or notion from which various characters can be unfolded but the very recognition of their being wrapped up in this essence is a recognition of a situation, not of simplicity and so here, in reference to a thing's differentia (essence) if this is treated as something simple it doesn't help us to see how the subject has the predicates empirically assigned to it and if it is treated as something complex then we are starting from something just as empirical any anything it explains - namely that a number of characters do hang together. Starting then with the non-propositional we could not proceed to any proposition - starting from the propositional we have something on the same logical footing as any proposition we could arrive at - open to the same doubt and criticism so that there is no question of distinguishing Rational or Essential or Primitive truths from derivative. We noted the absurdity of holding that an opinion becomes knowledge by being inferred by opinions, in other words taking conclusions as of higher order than premises but it is equally absurd to take premises above conclusions - to think that a conclusion is more subject to doubt than what it has been inferred from for if the premises are objective facts and the inference formally correct then we are just as sure of the conclusions as of the premises.

Finally in regard to any position like that of Socrates (or Taylor) in reference to any recognition of demonstrative science the attempt finally is to derive propositions from the non-propositional - from pure forms or concepts and this can't be done - the Form of the Good can't establish the goodness of anything in particular - the theory of Forms cannot show even that there are particulars - let alone argue that a particular is of this or that kind.

no previous
no next