A.J. Ayer

In this paper I offer an examination and defence of A.J. Ayer’s claim that a private language (or what he refers to as a ‘primitive language game’) is possible. In the first part of the paper, I argue that Ayer’s concept of a percipient can be reduced to an irreducible act of recognition of percepts. In the second half of the paper, I turn to Ayer’s Symposium piece. Ayer claims that Wittgenstein’s argument against a private language has to lay claim to certain fundamental philosophical assumptions which cannot withstand thorough scrutiny. The first assumption claims that any descriptive statement, in order to be understood, must be related to some publicly observable object either directly or indirectly. The second assumption states that in order to attach meaning to a sign it is necessary that others should also be capable of understanding it. The argument for a basic kind of recognition, or what Ayer calls primitive recognition, is mobilised against the second assumption. In order to attack the first assumption, Ayer utilises a cluster of arguments designed to highlight the radical contingency of the properties of ‘public’ and ‘private’ as they are attached to objects. Ayer also argues that it is possible to construe, without logical contradiction, a possible scenario whereby a person could understand a descriptive statement for someone else’s sensations regardless of public criteria holding sway.

There are passages in Wittgenstein where he compares his method to psychotherapy and one or two where he seems to suggest that the ‘patient’ has the last word on his ‘illness’ and ‘cure’.  This paper tries to take these seriously, especially in connexion with the idea that victims of philosophical confusion talk nonsense.

Philosophical accusations of talking nonsense made in a polemical spirit are bound to fail, since the only way one can effectively criticise another’s utterance is by attaching some meaning to it.  But if the other acknowledges the possibility that his puzzlement arises from his having been taken in by nonsense, then perhaps a co-operative investigation of this suggestion might be fruitful.  Indeed self-diagnosis ought in principle to be possible.  Ought it not to have happened at least once in the history of philosophy if philosophical nonsense has ever been exposed? 

But there are difficulties.  For example:
a) We must ask whether or not philosophical nonsense, when exposed as such, is supposed to lose all appearance of meaningfulness.  Either answer is problematic.
b) The acknowledged similarity of the methodology under investigation to that of psychoanalysis exposes it to many of the familiar objections to the methodology of psychoanalysis, especially those that draw attention to the misleading influence of suggestion by an authority-figure.
c) The proposal that relentless cross-questioning of a suspected nonsense-talker will eventually lead her to ‘dry up’ (if she really is talking nonsense) seems to involve a version of the so-called Socratic Fallacy.

I originally entitled this paper ‘Why are there no uncontroversial examples of philosophical nonsense?’, but since this seemed apt to provoke rather superficial responses, I decided to re-title it.  In it I ponder the fact that, not only are there no uncontroversial examples of philosophical nonsense, there are no uncontroversial examples of what one might call ‘analogue philosophical nonsense’.  These would be non-philosophical cases of someone’s believing he means something when in reality he means nothing.  (Someone’s saying, ‘It’s 5 o’clock on the sun’ is sometimes offered as an example.)  I ask what conditions such cases would need to satisfy.  What would count as a compellingly clear case of an illusion of meaning?  Why has no one described one?

.Taking as my starting-point Cora Diamond’s paper ‘What nonsense might be’, I extend her ‘austere’ conception of nonsense to encompass the talking of nonsense.  We need to focus on the utterer as well as the utterance.  This brings out how difficult it is going to be to DIAGNOSE the talking of (philosophical) nonsense, without falling into the ‘sense that is senseless’ trap (PI, I, 500).  It is not clear to what extent Diamond sees the problem here.  I suggest that if there is a way round the difficulty, it will involve adopting a therapeutic approach.

Introduction


In his early essay On Particulars and Universals,1 as well as in his Russell and Moore: The Analytical Heritage, Russell, and The Central Questions of Philosophy,2 A.J. Ayer tackles the problem of the status and meaning of universals. Ayer’s main concern throughout is to come to a decision on exactly how their use as general terms can be understood. In OPU, the younger Ayer enters the problem by asking how universal colour properties are able to be considered via resemblance, and in this sense he inquires whether sense data (which he calls “primary particulars” in this work) are amenable to universalization as opposed to more complex entities (such as the individual Socrates), which he calls “secondary particulars”.3Although Ayer has not reached his mature position on resemblance in the 1933 work, the seeds are sown for what will be his final position taken on universals in the 1970s. In the 1933 paper, Ayer maintains that it is possible to form basic universals out of sense-data known as “primary particulars”, which are inextricably bound up with, but cannot be translated down to, the fundamental relation of resemblance. Interestingly, what we shall see is that Ayer moves beyond this Humean position in order to modify his stance slightly on the key relation of ‘resemblance’. Ayer is not eliminating resemblance. On the contrary, this is simply not possible; Ayer instead moves further away from a nominalist position by pushing the resemblance relation downstairs and replacing it with a basic identity-forming relation that is prior to resemblance in order of providing for generality of expression.

Edward Witherspoon distinguishes Wittgenstein’s conception of nonsense from Rudolf Carnap’s.  The latter does not fully take into account the fact that, if something really is nonsense, it has no more meaning than ‘Ab sur ah’ and therefore cannot have any logical properties.  He finds the Carnapian conception even in professed Wittgensteinians such as Peter Hacker and Gordon Baker who think of themselves as rejecting it.  My question is where this leaves the notion of PHILOSOPHICAL nonsense.  Could anything that was utterly devoid of content play the crucial deceptive role that nonsensicalists, including Wittgensteinians, assign it?  Could it ever be detected if it did?

Since the demise of logical positivism accusations of talking nonsense in philosophy have mainly been made by Wittgensteinians.  But some of these latter are as free with words like 'nonsense', 'senseless' and 'meaningless' as any logical positivist.  One such is Rupert Read, who is prepared to write off the whole of metaphysics as nonsense.  Condemnations of this sort have in recent years largely been ignored by other philosophers, which is surprising in view of their apparent importance.  It is equally surprising that the evident problems with accusations of talking nonsense have been largely ignored by those who are still prepared to make them.

Cook’s paper ‘Wittgenstein on privacy’, though published over fifty years ago, is still one of the most impressive attempts to take seriously what philosophical nonsense would have to be.  Cook argues that someone who says, ‘It is not even logically possible to feel another’s pain’ is experiencing ‘a queer sort of illusion, produced by seeing one pattern of grammar on analogy with another and quite different pattern of grammar’.  I argue that he has identified a genuine philosophical illusion, indeed a grammatical illusion, that is quite close to what nonsensicalists are postulating – but I am not sure that it is IDENTICAL with it.

Stuart Hampshire in his review of Gilbert Ryle’s 'Concept of Mind' draws a distinction between two ways of using psychological terms.  The simple ‘He doesn’t believe in life after death (or the coming revolution)’ has a different use and different implications from ‘He doesn’t REALLY believe in life after death (or the coming revolution)’.  I ask whether there is any similar distinction with regard to meaning:  ‘He didn’t mean anything by what he said’ as against ‘He didn’t REALLY mean anything by what he said’.

Experimental philosophy’s much-discussed ‘restrictionist’ program seeks to delineate the extent to which philosophers may legitimately rely on intuitions about possible cases. The present paper shows that this program can be (i) put to the service of diagnostic problem-resolution (in the wake of J.L. Austin) and (ii) pursued by constructing and experimentally testing psycholinguistic explanations of intuitions which expose their lack of evidentiary value: The paper develops a psycholinguistic explanation of paradoxical intuitions that are prompted by verbal case-descriptions, and presents two experiments that support the explanation. This debunking explanation helps resolve philosophical paradoxes about perception (known as ‘arguments from hallucination’).

How do philosophical accusations of talking nonsense relate to the layperson’s notions of meaning and meaningfulness?  If one were to explain carefully what philosophical nonsense was supposed to be, would one be greeted with incredulity?  How do people react to being accused of talking nonsense themselves?  Is anything analogous to an illusion of meaning recognised in everyday life?  In dream-reports perhaps?  Or in the utterances of schizophrenics or those under the influence of drugs?  Or in certain jokes and hoaxes?  Are contradictions felt to be meaningless?  I argue that, though it is clear that experimental philosophy could shed some light on these questions, their subject matter creates special difficulties.  First, it is more difficult than usual to formulate questions and produce ‘vignettes’ that do not subtly encourage certain responses at the expense of others.  Second, the fact that philosophical nonsense is a metaphilosophical concept ensures that its investigation is going to be more indirect than would be that of knowledge or intention, for example.

Mackie’s claim that in general courage benefits its possessor seems inconsistent with his ‘error theory’ of value.  But how plausible is it in itself?  I suggest that his arguments for the claim fail in the same way as the arguments of those who more explicitly embrace ‘virtue ethics’ fail.  We do not know any way of objectively weighing risks against other considerations, of deciding whether something is ‘worth the risk’.  Since I have no PROOF that there could not be a way, I prefer to speak of ‘ignorance’ rather than ‘error’ here.

Davidson and Kolnai in different ways emphasise that practical syllogistic reasoning can only tell one that one has a reason to do or not to do something.  It cannot adjudicate between conflicting reasons.  It can tell one the means to certain ends but not which ends to pursue.  This paper considers possible responses to this predicament.  What if anything can be done to put deliberation - the 'weighing' of competing reasons - on  a rational basis?

I argue that Mackie's thesis about courage and self-interest is neither consistent with his 'error theory' of value nor convincing in itself.  The question of the objectivity of value needs to be distinguished from that of whether one value can objectively outweigh or outrank another.  This is so even when the only values under consideration are self-interested ones.