What Is Philosophy's Most Preposterous Notion? - A Suggestion.

1

1

What Is Philosophy’s Most Preposterous Notion? – A Suggestion.

Abstract.

There is much that is preposterous in philosophy. I suggest that the willingness of

certain philosophers to accuse their colleagues of talking nonsense is an extreme case

of this. I call the theory and practice of it ‘nonsensicalism’.

I interpret accusations of talking nonsense as claims that the other means nothing by

what he says in spite of sincerely believing that he does mean something. He is thus

subject to an ‘illusion of meaning’ (IOM). I spell out what this involves.

There are four main problems confronting the nonsensicalist:

a) Are IOMs possible? Here it is necessary to focus on the utterer of the supposed

nonsense. Too many writers on the topic of philosophical nonsense, even those who

consider it from a relatively critical standpoint, are content to discuss the sounds or

marks on paper and neglect the utterer. But we are concerned with a supposed error –

someone is supposed to have been misled – not just with whether certain sounds or

marks constitute a possible sentence of the language. One way of correcting the

tendency to forget the utterer is to take seriously Wittgenstein’s view that an

acceptable sentence of the language can be nonsense if it is uttered in inappropriate

circumstances. The question whether IOMs as understood here are possible seems

ideally suited for investigation by the methods of analytical philosophy. It is therefore

especially surprising that it should have been neglected.

b) If IOMs are possible, how might they be diagnosed? The answer is not obvious.

Normally, if one criticises what someone says, one does so on the basis of what one

2

2

thinks he means. One cannot do that if one is trying to show that he means nothing.

Those who have advocated an ‘austere’ conception of nonsense have shown greater

awareness of this difficulty than have other nonsensicalists (though still not enough).

c) The nonsensicalist must give us good reason for believing that he is talking about

the ordinary concept of meaning and not some narrower notion of his own invention.

d) It is not obvious that the concept of meaning is precise enough to do the work

nonsensicalists require of it.

The rest of the paper is devoted mainly to a consideration of problem (b).

One suggestion for dealing with it is that we should abandon the aggressive

employment of the notion of nonsense in favour of a more cooperative, therapeutic

approach. Could someone who is perplexed by philosophical difficulties come, with

the aid of someone else, to see that this is because he has been taken in by nonsense?

Could he perhaps see this unaided? Unfortunately, at the time of writing, the idea of

therapeutic nonsensicalism is woefully underdeveloped.

Another suggestion for dealing with this problem of diagnosis involves trying to come

to terms with a troublesome point in Wittgensteinian exegesis. Wittgenstein has a

tendency to call grammatical rules ‘nonsense’ when they are presented as if they were

universal generalisations about (possible) reality. On the one hand, this seems to offer

a way of getting a grip on nonsense, of diagnosing and specifying it. On the other, it

seems to go against the rigorously ‘austere’ view he takes elsewhere. Could it really

be true that someone who thus misconstrues a grammatical rule means nothing

whatsoever?

3

3

I conclude that, even if there is something in the notion of philosophical nonsense, the

way it has been employed so far has been pretty preposterous.

I want this paper to be in part a review of the critical literature on the idea of

philosophical nonsense – those writings which, whether sympathetic or not, do not

just treat it as unproblematic. The ‘austere’ conception is discussed in the main text,

but I have also thought it helpful to include a brief appendix on those few writers I

know of who in one way or another take issue with nonsensicalism.

________

Cicero, Chris Coope and W.E.Johnson on Preposterousness.

Cicero memorably claimed that ‘nothing is so absurd that some philosopher has not

said it’ (De Divinatione, II, 58). Similar remarks have been made by philosophers

and non-philosophers ever since and it is difficult to disagree. But what is the most

absurd thing philosophers have ever said? Let me suggest a possible candidate: that

much – or most – philosophy is sheer nonsense. My suggestion might seem

paradoxical. Have I not just admitted that philosophy is full of absurdities? Yes, but

when philosophers accuse each other of talking nonsense they are using the word

‘nonsense’ literally. They really do mean nonsense (non-sense). They are accusing

each other of meaning nothing by their utterances. I suggest that this may well be as

preposterous as it sounds.

4

4

But of course it won’t sound preposterous to everyone. In a recent paper Chris Coope

(Coope, 2009) mentions Cicero’s remark and comments that the word ‘preposterous’

might almost have been invented to describe philosophical theses. (Coope, 2009:

196) He seems to think that philosophical preposterousness is on the increase, at least

in some branches of the subject. (Coope, 2009: 198-99 n.44) And one reason for this,

he suggests, is the loss

of the insight that philosophers run the risk not only of saying things that are

untrue or unwarranted, but also of saying things that only appear to make

sense … So often what is offered as the results of philosophical enquiry

should be regarded instead as the raw material for it. It is thought perhaps that

as we are no longer positivists we can afford to relax in this regard, both in

regard to propositions and questions.1 It is as if we were all now prepared to

go along with W. E. Johnson’s remark (expressing his exasperation with

Wittgenstein) ‘If I say a sentence has meaning for me no one has the right to

say that it is senseless’. (Coope, 2009: 199)

On Coope’s view the idea that some philosophers talk nonsense, far from being

preposterous, may be our best defence against preposterousness in philosophy.

Where do your sympathies lie – with Coope or with Johnson?

The Supposed Error In Question.

Let me now state as clearly as I can what I take philosophical accusations of talking

nonsense to consist in. I may seem somewhat dogmatic but the more disputable

points in my interpretation will be discussed later. When one philosopher accuses

another of talking nonsense he is accusing her of meaning nothing whatsoever by an

5

5

utterance even though she genuinely believes she does mean something. There are

several points to note:

a) In everyday speech the word ‘nonsense’ is most often used to dismiss an assertion

as manifestly false, highly improbable or totally unsupported by evidence. Clearly it

is not the meaningfulness of what is said that is being impugned. In so far as what is

said is absurd, it is absurd because of what it means. Analytical philosophers have for

a century or so avoided this usage (usually at least: there are cases I am unsure

about2).

b) It is assumed that the accused believes she means something. She is not being

accused of deliberately talking nonsense. Accusations of charlatanry are not unknown

in philosophy and perhaps some analytical philosophers have suspected certain

Continental philosophers of merely pretending to talk meaningfully, but I shall not be

concerned with this.

c) When one philosopher accuses another of talking nonsense he does not just mean

that he does not understand her. (Is there not a sense in which he is claiming to

understand her – though not what she means – better than she understands herself?)

d) Nor does the accuser just mean that what she says is not an acceptable sentence of

the language. Of course his suspicions might have been aroused by something

deviant about the wording of the utterance3 but the question will always be whether

the accused meant anything by it. Could she produce an acceptable reformulation if

asked?

e) If producing an unacceptable sentence of the language is not sufficient for

producing philosophers’ nonsense, it is not thought (at least by Wittgensteinians) to

be necessary either. Producing what is a possible sentence of, say, English does not

guarantee that one means anything by it.

6

6

f) It is of course possible to repeat something one has heard, thinking it has a meaning

when it has none (for example, if one wants to know what it means). This is not what

the accuser has in mind. He is accusing the other of wrongly believing that there is

something she means by what she says, not just that it has a meaning.

g) Consider these formulations: ‘She is wrong to think there is something she means’;

‘She is wrong to think she means anything’; ‘She is wrong to think she means

something’. I suggest we avoid the last. It could mean that she is wrong about

whether she means anything, but it could also mean that she is wrong about what she

means. The latter, if it is possible, would be a different error.

This at any rate is what I take accusations of producing philosophical nonsense to

involve. It might be held that the same error occurs outside philosophy. Perhaps we

are at risk whenever we engage with abstractions (though this might be held to be a

manifestation of the philosopher in us all). Or perhaps it occurs in certain mental

illnesses or drug-induced states. I am not aware that a great deal has been written on

this point.4 Here I shall be mainly concerned with specifically philosophical

accusations of talking nonsense.

Charles Pigden has shown that they occur in Hobbes, the classical empiricists, Kant

(on some interpretations), the pragmatists, the logical positivists, Wittgenstein (both

early and late) and many (though not all) analytical philosophers. (Pigden: 2010)

They are common today (pace Coope), if not quite so common as in the middle

decades of the last century. But are they ever justified? Judging by the way many

philosophers write, one would think the answer was pretty obvious and in the

affirmative.5 I would not myself go to the opposite extreme and claim that they are

7

7

obviously never justified, but I would claim that they face immense difficulties which

ought sooner or later to occur to anyone who considers them critically.

In what follows I shall refer to the view that there is such a thing as philosophical

nonsense, which can be detected if one is sufficiently astute, as ‘nonsensicalism’. I

shall refer to the supposed error of thinking one means something when one means

nothing as an ‘illusion of meaning’ (IOM). ‘Illusion of meaning’ is, I think, slightly

preferable to ‘illusion of meaningfulness’, since the latter seems more likely to be

taken to include the error mentioned in (f) above of wrongly thinking something has

meaning without necessarily seeming to see a meaning in it oneself. To experience an

illusion of meaningfulness need not be to experience an illusion of meaning.6

The Difficulties Facing Nonsensicalism.

a) Are IOMs possible? This is the first question that ought to strike anyone who

makes a serious attempt to examine the notion of philosophical nonsense. We are

asked to accept that it is possible for someone to believe sincerely that he means

something by what he says when he means nothing. Suppose that it is not. Then

much philosophical polemic from Hobbes onwards has been simply misguided.

Philosophers have been accusing their opponents (and occasionally their own

previous selves) of an error that no one could conceivably make.7 Yet, to put it

mildly, very little effort has been made by those who make such accusations to

demonstrate that we are dealing with a genuine possibility. It is rather as if

philosophers frequently accused each other of self-deception, yet made no effort to

confront the notorious paradoxes that the notion of self-deception generates.

8

8

We have what seems to me to be a very interesting problem, one that lies at the

intersection of the philosophy of language and the philosophy of mind. It is of the

first importance for the assessment of the legitimacy of a certain style of philosophy

that is still widespread; yet it is hardly ever tackled head-on. It is particularly

surprising that analytical philosophers should have neglected it, since it is a question

about logical possibility and one would have thought that debate about what is and is

not logically possible was one of the most characteristic features of analytical

philosophy. Trying to describe a clear and incontrovertible case of someone’s

thinking there is something he means when there is not, or trying to demolish the

attempts of others to achieve this8 – would not this be an almost paradigmatic case of

analytical philosophy in action? (Ask those who dislike analytical philosophy.9)

It might be objected that my complaint is unfair, since some philosophers have made

suggestions about the sorts of things that might mislead someone into thinking he

meant something when he meant nothing. They have indeed, but it is not clear that

their suggestions directly address the question whether IOMs are possible. They often

seem to be assuming that we have before us a case of such an error. Consider a few

of the better-known suggestions. Someone mistakenly thinks he means something by

an utterance because he is misled by emotions or imagery accompanying the

utterance; by superficial grammatical resemblances between the combination of

words he utters and genuinely meaningful sentences (grammatical illusions); or by the

assumption that he can transfer a word from one context to another and it will carry its

meaning with it ‘like an atmosphere’ (PI, I, 117).10 If we already know that someone

is mistaken in thinking he means anything, then these suggestions might explain or

help to explain how he was misled. But reflect: emotions and imagery often

9

9

accompany meaningful utterances; genuinely meaningful sentences, however novel,

will surely bear grammatical resemblances to other uncontroversially meaningful

ones; and we often successfully transfer a word to a new context. (The last two points

are just variations on a familiar Chomskian theme.) Clearly we need some

independent reason for thinking that someone does not mean anything before we set

about trying to explain away his conviction that he does mean something.

I strongly suspect however that if someone were to describe a clear and convincing

case of an IOM, the description would include an explanation of how the victim of the

illusion was misled. The case would be that of someone’s uttering certain words, say

‘It’s 5 o’clock on the sun’. It would have to be

(i) clear that he believes that he himself means something by them; and

(ii) equally clear that he does not mean anything by them.

Both conditions would have to be satisfied and seen to be satisfied by all who

consider the matter. There must be nothing doubtful or borderline about the case. It

seems to me that an explanation would have to be given of why he says it and that this

could not be done without giving some indication of how he comes to think he means

something by it. (It may occur to the reader that, since bizarre sentences like this are

regularly discussed as philosophical examples, as is the case here, this is enough to

explain how someone could utter them. I return to this matter in the appendix – under

Falsidal Theories.)

I am not entirely sure about pathological cases. Could someone, because of mental

illness or under the influence of a drug (or even perhaps hypnosis) come out with ‘It’s

5 o’clock on the sun’, thinking there was something he meant by it, yet being wrong

10

10

about this? One would have a sort of explanation of why he said it (the illness or the

drug) but it would not go far towards explaining how his belief that he means

something by it arose. There is so little knowledge about these matters that probably

no one is in a position to be very definite. In any case the suggestion that such

pathological cases might be possible hardly amounts to meeting the demand for a

clear and incontrovertible case of an IOM. Nor is it clear how much such cases would

have in common with those of alleged philosophical nonsense.11

b) The Problem Of Diagnosis. Suppose IOMs are possible. How might one detect

them? Suppose someone makes sounds or marks on paper. How can one tell whether

she means anything? The first thing to note is that in daily life we would only rarely

consider the possibility that she does not mean anything. Occasionally one might

wonder whether someone had said something or was merely clearing her throat, or

whether she was writing or merely doodling. But normally we are very ready to

assume that people mean something by the sounds or marks they make, even when we

have not the slightest idea what it is they mean. If I hear people apparently chatting in

a language I do not understand, and perhaps do not even recognise, the possibility that

they do not mean anything will probably not even cross my mind.

We are not in fact well-equipped to detect an absence of meaning.12 No doubt this is

partly just the usual difficulties with a negative existential. But there is more to the

matter than that. Suppose someone says something I do not understand. Or suppose

she says something that is not an acceptable sentence of the language. Or suppose I

do not understand her because what she says is not an acceptable sentence of the

language. None of these possibilities entails that she does not mean anything. If she

11

11

is deliberately talking nonsense, then perhaps she will eventually come clean, but we

are concerned with the case where she believes she does mean something. So how

could I show that in fact she means nothing? How do I go to work on someone’s

words to show that there is no meaning behind them? To echo W. E. Johnson, what

could give me the right to call what she says ‘senseless’? This I call ‘the Problem of

Diagnosis’.

The logical positivists maintained that you could show that a claim was meaningless

by showing that it was neither tautologous nor verifiable, but they never succeeded in

explaining how you could test it for tautologousness or verifiability without first

assigning it a meaning.13 Indeed they seemed eventually to admit the difficulty by

conceding that the test was only for factual meaning. Presumably an utterance could

be meaningful without being factually meaningful.14 They faced further difficulties of

course, but the point to emphasise here is that they were no longer claiming to detect a

complete absence of meaning.

Might some other approach to the diagnosis of nonsense fare better? There seems to

be a feeling that a more piecemeal approach such as that of the later Wittgenstein

might succeed where positivism failed. Instead of applying some general criterion of

meaningfulness, could one examine individual utterances on their own merits and

perhaps show that the particular utterance of a particular philosopher was not backed

by meaning? It seems to me that the problem remains. One could see it is a dilemma.

It would be self-defeating to attribute a meaning to the utterance – one might thereby

uncover all sorts of difficulties with it but hardly the difficulty that it had no meaning.

On the other hand, if one refrains from assigning it a meaning or assuming it has a

12

12

meaning, one might be able to say truly that one does not understand it or that it is not

an acceptable sentence of the language or both, but in none of these cases will it

follow that the utterer means nothing by it.15 So how is one to proceed?

c) Coercive Theories Of Meaning. Charles Pigden has drawn attention to an

argument used by Richard Price against David Hume. (Pigden. 2010) Hume, on the

basis of his own theory of meaning, argued that many of the most basic concepts of

everyday life and of science are in reality mere pseudo-concepts. In particular, our

concepts of the self and of necessary connexion were dismissed. There are no

impressions from which they could have been derived and so the words that are

supposed to refer to them are in fact meaningless. Price replied that we do in fact

have, understand and use these concepts and therefore it is Hume’s theory of meaning

that is suspect. Moreover, many of the concepts Hume rejected (or would be

committed to rejecting if he applied his theory to all relevant case) are essential to

Newtonian physics, the scientific success story of the age. ‘It is much more likely

that Hume is mistaken than that Newton’s theory is nonsense.’ (Pigden, 2010: 177)

But, says Pigden, Price’s argument would still have much force even without the

‘august backing’ of Newton. (Pigden, 2010: 178) Why cannot one reply to Hume or

anyone else who accuses one of talking nonsense, ‘I do mean something and therefore

it is your theory of meaning, which implies that I do not, that is mistaken’? I suspect

that many readers will be taken aback by this argument. Their initial astonishment

might perhaps be expressed as follows:

We are concerned with illusions of meaning. We are assuming that the

accused believes he means something, that he has the impression of meaning

13

13

something. But it does not follow that he does mean something. Explanations

of how someone could be misled into thinking he meant something when he

meant nothing have been given. Even Hume offers an explanation of how we

come to think we can meaningfully talk of such things as ‘the self’ and

‘necessary connexion’.16 And Twentieth Century philosophers, particularly the

later Wittgenstein, have made a number of suggestions.

But we have already seen that you need good reason to think that someone does not

mean anything before you start trying to explain away his impression that he does

mean something. And this, I think, is the real force of the Price-Pigden argument. It

transfers the burden of proof to the accuser. No longer need the accused feel that the

onus lies on him to ‘give sense’17 to his words. He can demand to know what entitles

the accuser to think he can overrule someone else’s sincere claim to mean something.

Pigden suggests that the aim of a theory of meaning should be to account for what we

pre-theoretically judge to be meaningful and what we do not. (Pigden, 2010:178-79)

Pre-theoretical judgments of meaningfulness constitute our initial data. If our theory

leads to wholesale dismissals of utterances as meaningless even though they seem

meaningful, it has failed to achieve its aim. It is surely irrational to reject the data

because they fail to fit the theory. Pigden does not entirely rule out the possibility that

a very successful theory of meaning – ‘one that helps to solve pressing problems in

psychology or the social sciences for instance’ – might lead us to make readjustments

to our conception of meaningfulness. (Pigden, 2010: 179) But he does not believe

that any theory of meaning, from Hobbes to Dummett, comes near to being so

successful. Rather, they have all been ‘coercive theories of meaning’, theories

specifically designed to stigmatise as meaningless the utterances of one’s

14

14

philosophical opponents (Pigden, 2010: 156) – or, one might add, questions that

philosophers have found themselves unable to answer.

The reader must judge for himself the extent to which Pigden’s historical claim is

justified. But surely nonsensicalists ought to regard the charge that they are

employing tendentious theories of meaning as a serious one. Take Wittgenstein.

Whether or not he is best regarded as an ordinary language philosopher (On this see

Uschanov, 2001; Read, 2010), he would surely not relish being told that he was not

using the word ‘meaning’ in the way it is actually used ‘in the language which is its

original home’. (PI, 116)

Now suppose some theorist of meaning were to attempt to meet Pigden’s challenge by

arguing that he was not seeking to impose his own revisionary concept of meaning on

others but using the ordinary concept of meaning. If he wanted to make accusations

of talking nonsense, he would have to claim that our ordinary concept of meaning,

when properly understood and applied, shows that philosophers are sometimes wrong

to think they mean anything by their words. He might, for example, try to meet my

challenge (p. 9) to describe a clear-cut, incontrovertible case of an IOM – and then,

armed with the knowledge that our ordinary concept of meaning does allow for IOMs,

he could go on to look for them in the philosophising of others. One question he

would have to ask sooner or later is:

d) Is Our Ordinary Concept Of Meaning Precise Enough To Do The Work Required

Of It? Suppose it turns out to be a fuzzy concept and that the fuzziness occurs in

philosophically crucial areas. I do not see how this could be ruled out a priori. In

15

15

fact I can think of one reason to be suspicious of the assumption that there is a sharp

boundary between the meaningful and the meaningless. Would it not make the

problem of the origin of language even more difficult than it already is?18

Now the idea that meaning(fulness) is a vague or fuzzy concept is likely to be more

threatening to nonsensicalists than to others. For one might find that philosophers are

doubtful about, or disagree about, whether certain utterances are meaningful without

there being any principled way of deciding the issue.19 One could stipulate of course –

and this is just the sort of situation in which people do stipulate20 – but why should a

philosopher accept a stipulation that rules out his own utterances as meaningless?

And would even the stipulators have any confidence that, by stipulating that a

troublesome thesis or question was meaningless, they had really got rid of it?

Wittgenstein may have held that utterances such as ‘Is Wednesday a fat or a lean

day?’ and ‘Arrange the vowels in order of their darkness’ were borderline meaningful.

As far as I know, he never explicitly says so, but some of his remarks strongly suggest

it. (BrB, p. 148, Z: 185) No doubt nonsensicalists can take this in their stride. But

did Wittgenstein ever consider the possibility that the concept of meaning has much

vaguer boundaries than his concept of ‘secondary sense’ allows for? I do not know.

The remark in On Certainty that ‘[d]oubt gradually loses its sense’ needs to be

weighed in this connexion. (OC, 56, emphasis added.)

The Obviousness of the Problems.

Let me sum up the problems with the notion of philosophical nonsense. First, we

need to be sure that IOMs are possible, that it really could happen that someone

16

16

thought he meant something when he meant nothing. We need to be able to describe

a clear and indisputable case where we could say: If this were to happen, we would

have an instance of an IOM. Such a description would probably include an

explanation of how the person was misled. Certainly, it would be preferable if it did.

Second, we need to know how such an error, assuming it to be possible, could be

diagnosed. How does one operate on someone’s utterance to show that behind it there

lies no meaning whatsoever?21

I suggest these two problems are the ones to be tackled directly. The other two are

perhaps more in the nature of dangers to be constantly borne in mind. Are we talking

about the concept of meaning as it is generally understood or are we subtly (or not so

subtly) substituting something narrower that suits our own polemical purposes? And

how clear and definite is the concept of meaning? Does it really have the sharp

boundaries it will need to have if it is to serve as a precision instrument for probing

and eliminating the seemingly interminable disputes and unanswerable questions that

have plagued (or should it be: constituted?) philosophy?

In my view none of these problems is particularly subtle or difficult to grasp. Anyone

who reflects critically on the notion of philosophical nonsense and does not just

accept it on the authority of his chosen masters will soon find he has to confront them.

Indeed it is difficult to imagine what a critical examination of the idea that

philosophers talk nonsense that failed to engage with them would be like. The

problems are also statable without much reliance on technical terminology. The

reader will have to take my word for it but I made no special effort to avoid technical

terminology, as I might have done had I been writing specifically for the layperson.

17

17

And yet they are problems with which philosophers are only beginning to grapple.

Why this should be so will, I think, be something of a puzzle for future historians of

philosophy.

How Should We Approach The Question Of The Possibility Of IOMs?

Philosophers of course frequently argue about whether particular sentences (or

‘sentences’22) are meaningful. Sometimes these are theses seriously propounded or

questions seriously asked by philosophers. But sometimes they are sentences

invented by philosophers that are supposed to be superficially grammatical (and

perhaps analogous to ones seriously uttered by philosophers) but obvious nonsense.

And sometimes they are offered as sentences that may look like nonsense but which

do in fact make sense, whatever other philosophers might say. And sometimes they

are just supposedly simple cases whose meaningfulness or otherwise it ought to be

relatively easy to investigate. Here are some examples of the second, third and fourth

types23, some of which will be discussed later – ‘The good is more identical than the

beautiful’; ‘Virtue is square’; ‘Virtue is not square’; ‘Virtue is a fire-shovel’;

‘Quadratic equations go to race-meetings’; ‘A thing is very similar to itself’; ‘[That is]

an honest geranium’; ‘[That is] a dead rainbow’; ‘It’s 5 o’clock on the sun’; ‘It isn’t 5

o’clock on the sun’; ‘’What time is it on the sun?’; ‘Every rod has a length’; ‘I

perceive I am conscious’; ‘Suppose everything doubled in length overnight’; ‘There is

a white rabbit between two chairs whenever no observations or verifications are being

made’; ‘Quadruplicity drinks procrastination’; ‘Chairman Mao is rare’; ‘This stone is

thinking about Vienna’; ‘Julius Caesar is a prime number’; ‘The class of men is a

man’; ‘The signal flashed black’; ‘Friday is in bed’.

18

18

I believe that arguing about whether such sentences are meaningful is not such a good

way of approaching the question of whether there is such a thing as philosophical

nonsense as might be supposed. Considering such sentences in isolation encourages

us to neglect the utterer – in particular, to forget to ask why anyone might utter them.

However I do not say that nothing can be learnt from such an approach, so let us

begin by doing the very thing I am unhappy about.

Views according to which such sentences as those above are nonsensical are

sometimes contrasted with what are called ‘falsidal’ views.24 This name – which was

apparently originally ‘falsidical’ (Magidor, 2013: 95.) – hardly seems a good one.

First, if one opts for saying that a sentence is false rather than nonsense, one will

presumably have to say that its negation is true. Second, one will have to deal with

questions (which are not normally considered to be either true or false). Third, even

with what seem to be assertoric, fact-stating sentences, one might wish to cater for the

possibility of the failure of bivalence. I shall however retain the term ‘falsidal’ here.

The views in question are not as well-known as they ought to be25, and discussing

them under a new name would hardly tend to enhance their prominence. (A fair few

philosophers have taken to using the jocular ‘sensical’ as the opposite of

‘nonsensical’. Perhaps they would favour renaming ‘falsidal’ views ‘sensical’ or

‘sensicalist’.)

Enough has been written on sentences like those listed to suggest (to me at any rate)

that there may often be no definite answer to the question of their meaningfulness. I

have already mentioned the possibility that the concepts of meaning and

19

19

meaningfulness might not be precise enough to bear the weight that philosophers have

wanted to put on them.

The arguments produced on both sides strike me as inconclusive. On the one hand,

techniques have been developed for showing how a sentence like ‘Friday is in bed’

can be seen as false and therefore meaningful.26 But it is not clear to me that this

shows that, in spite of the fact that to many (most?) people it initially seemed to

convey nothing, it was really meaningful all along. On the other hand, arguments that

supposedly compel us to see such sentences as meaningless often seem utterly

question-begging. Hacker and Baker tell us that in order to give sense to ‘honest

geranium’ one must be able to say what it would be like to encounter an honest

geranium. (Hacker and Baker, 1984: 338-39) This sort of demand would condemn

all necessary falsehoods as nonsense by fiat. Even contradictions would have to be

regarded as nonsense27 as well as arithmetical falsehoods.28 Surely this is too short a

way with the problem. Cora Diamond notes that Frege would have considered ‘Julius

Caesar is a prime number’ to be false rather than nonsense. (Diamond, 1991: 96-97)

Can his view simply be swept aside on the grounds that one cannot describe what it

would be like to discover that Caesar was a prime number? (What in any case is the

status of the ‘cannot’ here?)

At this point I propose to cut the Gordian Knot by maintaining that we have not quite

got hold of the right problem anyway. As already suggested, is it not an error, or at

least a potential source of error, to focus so exclusively on the question whether

certain combinations of sounds or marks do or do not constitute meaningful sentences

20

20

of, say, English? Such a question makes no mention of any utterer. And talk of

philosophical nonsense surely concerns an utterer.

Consider again that old warhorse ‘It’s 5 o’clock on the sun’. My impression is that

there is considerable agreement that there is something odd about this sentence,

something wrong with it even, and in large measure about what this oddity or

wrongness consists in. A convention has not been established for telling the time on

the sun (Or should it be ‘at particular places on the sun’? And what counts as on the

sun?). Our convention for telling the time of day is based on the position of the sun in

the sky and so does not work with – or even suggest any natural extension to29 –

telling time on the sun. Suppose one philosopher were to claim that, until such time

as a convention is established for telling the time on the sun, the English language

assigns no meaning to ‘It’s 5 o’clock on the sun’. If another philosopher were to

dispute this claim, we would have a debate that as yet made no mention of any utterer

of the problematic sentence. (I have not forgotten that the disputants themselves

would utter it.) The one who thinks it meaningful could say it was false. After all,

‘It’s 5 o’clock on the sun’ could be reformulated as ‘The time on the sun is 5 o’clock’

and this seems to contain a definite description, so why not apply Russell’s Theory of

Descriptions? A case could surely also be made for saying that it was meaningful but

without a truth-value. It is not obvious that the sentence, however odd or defective it

might be, is meaningless.

But even if agreement could one day be reached about this (something I am inclined

to doubt), how much light would it shed on the question of philosophical nonsense?

What we would still need to know is whether someone could come out with it

21

21

thinking she meant something by it, even though she in fact meant nothing by it.

Could we describe a case satisfying the conditions specified on p. 9? –

(a) It was perfectly clear that she herself thought she meant something by the

sentence. (She did not just think that it had a meaning; she was not just entertaining

the possibility that it might be given a meaning; she was not just playing with the

idea, perhaps in a work of fiction, that it had been given a meaning.)

(b) It was equally clear that she meant nothing whatsoever by it. (She was not just

saying something false, for example, or something meaningful but without a truth-

value, or something very vague or something that was very badly expressed.)

It does not seem to me at all obvious that such a case could be convincingly

described, even if it were generally agreed that ‘It’s 5 o’clock on the sun’ does not at

present have any meaning.

Of course, this particular sentence may not be a good one to work with. I bring it up

because many philosophers seem to think it is closely analogous to some of the

supposedly nonsensical utterances of other philosophers. The way it fits into

Wittgenstein’s discussion at PI, 350-51 is rather complicated, but the way it seems to

strike many philosophers might be expressed thus: ‘If someone said that, he would

obviously be talking nonsense and, although it is not a philosophical thesis, it

parallels the sort of nonsense philosophers do talk and helps to show how it is

possible to talk nonsense unwittingly’.30 I have in fact had it brought up in discussion

as a counter to my own scepticism about nonsensicalism – that and ‘What time is it on

the sun?’. But this does not mean it will turn out to have been a good choice. Perhaps

believers in the notion of philosophical nonsense should look elsewhere for a locution

that might help them describe a clear case of an IOM.

22

22

The Context Of Utterance.

Another reason why one should not expect an examination of sentences without

reference to utterers to effect any spectacular clarification of the question of

philosophical nonsense emerges when we consider an important aspect of the later

Wittgenstein’s view of nonsense. He contends that an acceptable sentence of the

language can be nonsense if it is not uttered in the right sort of circumstances.

Suppose I utter what is a genuine sentence of English. This does not guarantee that I

mean anything by it. After all, my utterance might be some kind of automatism, due

perhaps to an electrode inserted in my brain. But Wittgenstein would go much further

than this. An ordinary sentence of the language, when used in a philosophical way,

can be nonsense – though I suppose that in such cases one should say that it is not

really being used at all.31

One could make a long list of sentences for which one can think of a (non-

philosophical) use – sometimes easily, sometimes only with difficulty – but which, on

Wittgenstein’s view, become nonsensical when philosophers utter them in the way

they do – e. g. ‘This is here’ (PI, 117); ‘Is this foot my foot?’; ‘Is this body my body?’;

‘Is this sensation my sensation?’ (PI, 411); ‘This is how things stand’ (PI, 114, 134); ‘I

know I am in pain’ (PI, 246); ‘This tree doesn’t exist when nobody sees it’; ‘Only my

pain is real pain’ (BB, p. 57); ‘I know that that’s a tree’ (OC, 349); ‘I know I have

always been near the surface of the earth’ (OC, 264). Unlike the sentences discussed

in the last section, for which one can make a case for saying they are excluded from

the language and to which many people react with incomprehension, these are ones

that one has to accept as possible sentences of the language as it now is. So, if for

23

23

example we are to inquire whether ‘Only my pain is real pain’, as said by a

philosopher, is in fact philosophers’ nonsense, we will have to proceed directly to the

question whether someone could say it, believing that there was something he meant

by it and being mistaken in this. The question whether it is a possible sentence of the

language has already been answered and should not distract us.32

Before proceeding further I would like to make two complaints about Wittgenstein’s

treatment of this issue. First, he often writes as if he thinks that to dismiss something

as nonsense is to imply that it is excluded from the language. The much quoted

‘sense that is senseless’ passage (PI 500) concludes: ‘a combination of words is being

excluded from the language, withdrawn from circulation’. In a parallel passage in

Philosophical Grammar (p. 130) he speaks of its being excluded ‘like some arbitrary

noise’ and in another well-known passage he compares a piece of philosophical

nonsense to ‘Ab sur ah’. (AWL, p.63) I at least would have found it much easier to

reach an understanding of Wittgenstein’s views on nonsense if he had juxtaposed such

remarks with reminders that on his view acceptable sentences of the language can also

be nonsense if they are uttered in the wrong sort of circumstances.33

My other complaint concerns On Certainty. This work contains numerous examples

of sentences which Wittgenstein thinks are perfectly acceptable in the right sort of

context but which become nonsense in wrong ones. Unfortunately he seems to me to

take a step backwards in one respect. Such works as the Blue Book and the

Investigations make some attempt to explain how the supposed nonsense-talker has

been misled, how he comes to utter his philosopher’s nonsense. But in On Certainty

one again and again meets examples of his imagining someone’s saying something in

24

24

circumstances that make it very odd indeed – sometimes positively insane, e. g. 350,

464 – but with no explanation of why the person says it. It is surely not a sufficient

answer to say (what is no doubt true) that Wittgenstein intends the examples to sound

as bizarre as they do.

But the point I want to emphasise – a point which can hardly be emphasised too much

– is that if we are to take the idea of philosophical nonsense seriously, we need a

convincing example of an IOM. Whether it is something a philosopher has actually

said or merely something analogous to what philosophers say, whether it concerns a

sentence that has a non-philosophical use or one that has none, we need a clear and

convincing case. Even those who are quite certain that other philosophers talk

nonsense and perhaps accept that they themselves have talked nonsense in the past

should appreciate this need. They presumably want to convince doubters like myself.

Part of the problem is that among many philosophers the idea of philosophical

nonsense is too familiar. If it were more clearly recognised what a remarkable error

an IOM would have to be, the task of showing such an error to be possible might

seem more urgent.34 What little has been done I do not myself find remotely

convincing. But perhaps this is because the problem has not been squarely confronted.

As already suggested, the question whether IOMs are possible seem eminently

suitable for investigation by the methods of analytical philosophy. Who can say with

complete confidence what the outcome of such an inquiry would be?

25

25

How Should We Approach The Problem Of Diagnosis?

I have drawn attention to the difficulty of diagnosing IOMs, even on the assumption

that they are possible. One way of deepening one’s appreciation of the difficulty is to

consider the work of a number of Wittgensteinians who have advocated what is most

often called an ‘austere’ conception of nonsense. The basic idea here is that if

something really is nonsense, it does not consist of meaningful parts. If ‘Julius Caesar

is a prime number’ is nonsense, then none of the words composing it has its usual – or

indeed any – meaning. It is not that the meanings of the words combine or interact to

produce in this case nonsense, as they combine to produce sense in ‘Seven is a prime

number’. On this view, the constituents of ‘Julius Caesar is a prime number’ are only

meaningful in the sense that they could be used in genuinely meaningful locutions.

They are not meaningful in the sense of having any meaning in this particular

locution.

I shall defend the austere view only to the extent of explaining why I find it

plausible.35 One can perhaps best appreciate the strength of the austere view by trying

to formulate a rival non-austere view.36 It would be that in a piece of nonsense the

meanings of the constituent parts contribute to the meaninglessness of the totality.

One would therefore be able to parse, partition or segment the meaningless whole into

parts each of which played its own role in generating the absence of meaning of that

whole. Merely trying to state the view with care gives an indication of how difficult it

is going to be to defend.

This is not to say that it is not a tempting view. Indeed Diamond calls it the ‘natural’

view. We are, she says, inclined to think that, assuming ‘Julius Caesar is a prime

26

26

number’ and ‘Scott kept a runcible at Abbotsford’ are both nonsense, they are

nonsense in different ways. The latter is nonsense because English assigns no

meaning to ‘runcible’; the former because of the meanings English assigns to its

constituent words. (Diamond, 1991: 95-97) But on the austere view they are

nonsense in the same way. In neither case do the constituent parts have any meaning.

The fact that ‘prime number’ has a meaning in ‘Seven is a prime number’, for

example, is irrelevant. Diamond sums the matter up by saying that on the natural

view there is both ‘negative nonsense’, due to a failure to give meanings to certain

sounds or marks, and ‘positive nonsense’, due to the clashing of meanings, whereas

on the austere view there is only ‘negative nonsense’:

There is no ‘positive nonsense’, no such thing as nonsense that is nonsense on

account of what it would have to mean, given the meanings already fixed for

the terms it contains. (Diamond, 1991: 107)

The austere theorists derive their conception of nonsense from Wittgenstein (and also

Frege). Well-known passages in Wittgenstein forcefully express the view that if a

combination of words is senseless, ‘it is not as it were its sense that is senseless’. (PI,

500) It is non-sense and has no more sense than ‘Ab sur ah’ (AWL, p.63) and thus no

grammatical structure or logic. I shall not take issue with their interpretation of these

passages37 (though I shall shortly be drawing attention to a tendency in Wittgenstein

that is not easily reconciled with the austere view). What I do wish to suggest is that,

in spite of the austere theorists’ commitment to Wittgenstein’s philosophy, their

conception of nonsense helps to bring out the seriousness of the Problem of

Diagnosis.

27

27

Suppose I want to prove that what someone has said is self-contradictory. It is

difficult to see how I could do this other than by considering what she means by the

constituent parts of her utterance. In a simple case I might be able to point out that

what she says here contradicts what she says there – two meaningful sentences that

cannot both be true. But I might have to show that a single sentence is self-

contradictory, by inspection or by deriving a more blatant contradiction from it. In

neither case will I be able to avoid considering what she means. If, for example, I

assume she is using a term univocally and I turn out to be wrong, this will in most

cases spoil my argument – though I might have grounds for complaint about how she

expressed herself. In proving that someone is guilty of self-contradiction, meaning is

of the essence.38

This is not going to work if I want to prove that something is nonsense. For nonsense

cannot figure in inferences. I could not deduce a piece of nonsense from an assertion

of Plato’s and conclude from that that Plato was talking nonsense. And on the austere

view I could not prove that an assertion was nonsense by showing that its parts

combined to produce an absence of meaning. It is beginning to look as though

nonsensicalism, at least if we take an austere view of nonsense, deprives the

philosopher of what one might have thought was his only legitimate way of

proceeding – by reasoning, argument, inference.39

The problem is: How could one reach the conclusion that an utterance is nonsense by

means that are in some way rational, more rational at least than mere arbitrary

dismissal? What part could reason(ing) play in the rejection of it as nonsense?

Edward Witherspoon, after producing a remarkably astute critique of the way some

28

28

nonsensicalists (including some of his fellow Wittgensteinians) go about diagnosing

philosophical nonsense40, raises the question of what an austere approach ought to be

like. (Witherspoon, 2000) I have read the relevant passage (Witherspoon, 2000: 345,

final paragraph) times without number and still cannot see that his answer amounts to

more than this: You ask someone you suspect of talking nonsense to explain what she

means; perhaps she succeeds and perhaps she doesn’t; and if she doesn’t, perhaps she

will come to acknowledge that she has been talking nonsense.

Faced with the question whether one can legitimately accuse others of talking

nonsense, Lars Hertzberg has written:

[I]t would be a misunderstanding of the philosopher’s task to suppose that we

should be trying to establish ‘what can be said’ and ‘what cannot be said’.

This would presuppose the existence of general rules laying down the sense of

utterances in various circumstances, but of course, there are no such rules.

What we may end up saying, at most, is things like, ‘’You can’t say this and

mean that’, or, ‘’If you say this here, it will come out as something quite

different from what you mean to be saying’ or maybe even just, ‘I wouldn’t

say that if I wanted to make that kind of point in this situation’, such remarks

getting their point from the particular philosophical difficulty we are trying to

straighten out. On the other hand, the temptation to express such a point by

saying, ‘That wouldn’t make sense’ is a mark of philosophical impatience

(Hertzberg, 2001: 14-15 – the reference is to the corrected version on the

Internet).

29

29

This sounds like a thoroughgoing abandonment of nonsensicalism. In fact it sounds

more definitive and decisive than the scepticism about nonsensicalism that I am

urging here. Now there are followers and interpreters of Wittgenstein who play down

this aspect of his work.41 But Hertzberg’s article is about Wittgenstein and about

nonsense. It seems that he is quietly but deliberately abandoning a major feature of

Wittgenstein’s philosophy, both early and late. There is a footnote to the paragraph

quoted above which reads: ‘We may note in this connection that many of the

invocations of nonsense in the Philosophical Investigations are tentative or

conditional’. Yes, they are, and I shall later be mentioning examples of a certain

tentativeness; but ‘many’ is not ‘all’42. We must ask: Should philosophers abandon

nonsensicalism in all its forms, Wittgensteinian or otherwise?

One response might be to try to develop a non-austere conception of nonsense, but I

am afraid nothing plausible occurs to me and I leave the suggestion to others. There

are however two ideas current in the literature that might seem to help with the

defence of nonsensicalism. Both derive from the later Wittgenstein and both are

undeveloped and therefore difficult to assess.

The Therapeutic Approach.

Wittgenstein sometimes speaks of his method as therapeutic or like the treatment of

an illness. Sometimes he likens it to psychoanalysis. Many of the relevant passages

are well-known and frequently quoted. Yet I do not think that Wittgenstein or his

followers – or anyone else – has provided us with a clear picture of what therapeutic

philosophy would consist in.43

30

30

First, let me mention one possible misunderstanding. A characteristic feature of

Twentieth Century nonsensicalism was the dismissal of questions as well as theses as

nonsense (In this perhaps it contrasted with earlier forms.) Clearly the aim was to

get us to stop racking our brains over pseudo-problems and, equally clearly, such an

aim could be called ‘therapeutic’. Even logical positivism could claim to be

therapeutic in this sense.44 But the kind of therapeutic philosophy I want to discuss

would be cooperative. It would not be a matter of one philosopher accusing another

of giving non-answers to non-questions but of one philosopher helping another out of

his difficulties by showing him that there was something radically wrong with his

questions and with the answers he was considering.

Suppose I am perplexed by some philosophical problem. Suppose also that I am open

to the possibility that my perplexity arises from my having taken nonsense for sense.

Could a philosophical therapist help me to see that this was indeed the case? We

would have to work together. It would not be a matter of his refuting me, though of

course if he were successful, I would eventually abandon questions and answers to

which I had previously been attached.

The first thing to note is how utterly remote such a therapeutic session would be from

the actual practice of most nonsensicalists. Kenny tells us that ‘Wittgenstein does not

think the sceptic can be answered, only silenced’. (Kenny, 1973: 218) This sounds

far less like a cooperative endeavour than did the traditional enterprise of taking the

sceptic’s doubts seriously and trying to persuade him that we do know the things he

doubts. If the sceptic is to be given nonsensicalist therapy, this must be something

very different from a peremptory silencing, even though it is supposed to lead to his

31

31

no longer voicing sceptical doubts. Or take Peter Hacker. He has written

voluminously, eruditely and illuminatingly on virtually every aspect of Wittgenstein’s

work. I should imagine that he quotes almost all the remarks that are suggestive of a

therapeutic approach somewhere; some he quotes more than once. Yet in his

philosophising he is as aggressive, as polemical, as any positivist. He thinks it

possible to show that those who voice sceptical doubts, for example, are talking

nonsense, as are those who claim that a machine could think.45 He does not wait for

sceptics and AI theorists to come to him asking to have their confusions sorted out.

He does say that for philosophers to be cured of their inclination to talk nonsense they

must want to be cured (Hacker, 1986: 247), but apparently this does not mean that

one cannot recognise, simply from what they say, that they are in need of a cure.

I want to suggest that, if there is any way of dealing with the Problem of Diagnosis, it

will be by a cooperative, therapeutic approach and not by straight attempts to refute

those one sees as opponents.

The passages in Wittgenstein that I find most suggestive are those where he insists

that in discussing philosophical problems one must secure the other’s agreement

about certain things – for example, about how he is using certain words or about

whether a certain form of words expresses his feelings or about whether he is

influenced by a certain analogy.46 Obviously, this sort of thing occurs quite frequently

in ordinary philosophical argument. I might begin to wonder whether we are talking

at cross purposes and raise the question whether you are using a term in the same way

as I am. Or I might suggest to you that you are thinking along certain lines

(particularly if I am claiming to diagnose a fallacy). But Wittgenstein seems to be

32

32

envisaging something more thoroughgoing. At every step the other’s agreement must

be secured.47 Presumably, if the therapy is successful, the other will come to see that

at some point in his argument he has ceased to use words in a meaningful way, or,

better, that he has ceased to use them at all.

Well, could this be done? Is there a procedure consisting of small steps each of which

is taken with the agreement of the ‘patient’ that would or could culminate in such a

realisation on his part? It seems to me that, if there is, then self-diagnosis or self-

therapy ought to be possible. It ought to be possible for someone, even without the

help of anyone else, to come to realise that, though he previously thought he meant

something by an utterance, he was wrong. It also seems to me that the Problem of

Diagnosis has still not been circumvented. I can certainly detect faults in my own

thinking, fallacious reasoning or inconsistency for example. But this requires

attention to the meaning of what I say (perhaps to myself). Just what operations I

could perform on my own utterances or silent soliloquies that would expose them as

devoid of meaning is not clear. (For more on this see my ‘Nonsense and

Argumentation’, especially pp. 8-9, on academia,edu) They seem to me to mean

something. What grounds (non-arbitrary grounds, if that is not a pleonasm) could I

have for discounting this appearance? I do not say dogmatically that there could be

none but I believe the onus is on the nonsensicalist to make suggestions.

Philosophers can change their minds, of course. And presumably it sometimes

happens that a philosopher who once thought certain questions or theses meaningful

no longer does so. Philosophers who dismiss (sentences purporting to express)

sceptical doubts as meaningless seem fairly numerous. I should imagine that most of

33

33

them were ‘taken in’ by such doubts when they first met them. (The question ‘How

can I tell I am not dreaming?’, for example, often occurs to people who have never

studied philosophy.) How did they come to the conclusion that they were victims of

IOMs? Some autobiographical details would be welcome.

Distorted Reflexions of Grammar.

Consider the following passage from Moore’s report:

[Wittgenstein] said that … both the Realist and the Idealist were ‘talking

nonsense’ in the particular sense in which ‘nonsense is produced by trying to

express by the use of language what ought to be embodied in the grammar’;

and he illustrated this by saying that ‘I can’t feel his toothache’ means ‘”I feel

his toothache” has no sense’ and therefore does not ‘express a fact’ as ‘I can’t

play chess’ may do. (M in Klagge and Nordmann, 1993: 103)

Many commentators have noticed a tendency in Wittgenstein to talk in this way and

some, such as Peter Hacker (whose terminology has influenced me here) have made it

central to their interpretations of Wittgenstein. It seems relevant to the Problem of

Diagnosis since it suggests a way in which one could specify the nonsense one was

accusing the other of talking. One could tell from a consideration of what she says

the specific mistake she is making. One could point to the rule one thinks the

metaphysician misconstrues as a highly general truth about reality and try to show

that what she actually says is best seen as a nonsensical response to her dim and

distorted awareness of the rule. But the fact that one could do this might make one

wonder whether what she says is well described as totally without meaning.

34

34

Why does Wittgenstein speak of nonsense here? Why is one talking nonsense if one

tries ‘to express by the use of language what ought to be embodied in the grammar’?

Such an error would no doubt concern meaning but it is far from obvious that it would

deprive one’s utterance of meaning – even if one accepts that ‘I can’t feel his

toothache’ means that ‘I feel his toothache’ has no sense. If the passage were an

isolated one, one might be inclined to discount it as not representing Wittgenstein’s

considered view or as a merely rhetorical use of the word ‘nonsense’ or as a piece of

misreporting by Moore. But in fact one could cite numerous passages in which

Wittgenstein seems to take this line. In Do Philosophers Talk Nonsense? I try to

explain why he adopts this manner of speaking but without any claim to complete

success. (Dearden, 2005: 53-56) Here I shall take a different line. I shall suggest

(a) that Wittgenstein is wrong to talk in this way;

(b) that in many of the relevant passages his employment of the term ‘nonsense’ is

tentative and undogmatic, as though he himself is unsure whether it is the best

way of putting the matter;

(c) that he might be talking of an error that really could occur and that if he is,

this could well be important;

(d) that describing precisely what the error would have to be is not easy – let alone

deciding whether it is possible.

a) Suppose it is possible for someone to present us with what is really a rule of

grammar, thinking she is stating a highly general truth about reality (or possible

reality). It would be an error but I cannot see any good reason for thinking she does

not mean anything. To use the kind of metaphor favoured by Peter Hacker, she is

seeing a rule of grammar in a distorting mirror or through a distorting lens and

35

35

reporting on what she sees. Perhaps one could say that she is wrong about what she

means – her error somehow concerns what she means (or perhaps understands) by the

rule of grammar. To me it seems to be pervaded by, shot through with, meaning. No

doubt it is very unlike an ordinary factual error, but not because she is talking without

meaning anything.48 Given that in many passages Wittgenstein is very strict and

‘austere’ in his use of the word ‘nonsense’, in view of his rejection of ‘senseless

senses’, this way of talking can only be confusing. Some writers suggest that

Wittgenstein is using the word ‘nonsense’ in a technical sense when he applies it to

distorted reflexions of grammar.49 But, apart from the question ‘Why doesn’t he say

explicitly that that is what he is doing?’, the potential for confusion remains.

b) Consider what Wittgenstein says in the Blue Book (p. 30) on the subject of

adopting ‘a wrong method of brushing aside [a] question’:

It is similar when we ask, ‘Has this room a length?’ and someone answers, ‘Of

course it has’. He might have answered, ‘Don’t ask nonsense’. On the other

hand ‘The room has a length’ can be used as a grammatical statement. It then

says that a sentence of the form ‘The room is … feet long’ makes sense.

Compare PI 252:

‘This body has extension.’ To this we might reply: ‘Nonsense!’ – but we are

inclined to reply ‘Of course!’ Why is this?50

Is there any significance in the fact that he says that one might reply, ‘Nonsense!’

rather than something stronger, such as that one ought to reply thus? Obviously, it is

dangerous to place great weight on nuances like this, but Wittgenstein chose his

words with care. It is possible, I suggest, that he would have been content to

emphasise that someone who says, ‘This body has extension’ has not succeeded in

36

36

saying anything about the world and to forgo the claim that he has not succeeded in

saying anything whatsoever.

c) Since I do not think that these distorted reflexions of grammar would be well

described as ‘nonsense’, I could stop there. But the issue is an interesting one, so I

shall say a little more. Has Wittgenstein uncovered a possible error? Could someone

misapprehend a rule of grammar as a fact about (possible) reality? I do not know. I

do not think the question has been properly investigated. But I do think it could be

important. Wittgenstein applies, or has been interpreted as applying, the idea to many

philosophical problems. Take that of colour-exclusion: Why cannot something be

both green and red all over at the same time? A proof – or even a plausible argument

– that this seeming impossibility is not a matter of the essential nature of colours but a

distorted reflexion of the grammar of colour-words would surely be of great

philosophical interest. So we need to ask: Could someone think he was talking about

the essential nature of colours when he was really talking about the grammar of

colour-words?

d) The way I have just formulated the problem makes it sound as though he would

have to be wrong about what he meant, in the sense of thinking he meant one thing

when really he meant another. And it is not obvious that this is how the error should

be characterised or that, when so characterised, it is possible. There was a view,

fashionable in the Fifties (and no doubt based at least in part on BB, pp.57, 59 and PI,

400-403), that metaphysical claims were disguised linguistic recommendations and

this really did seem to imply that the metaphysician was wrong about what he meant.

He was not supposed to be aware that he was making a disguised linguistic

37

37

recommendation. He was not being accused of dishonesty. But it is, to say the least,

unclear how someone could make such an error.51

But how should the supposed error with which we are concerned be characterised?

With IOMs the problem was fairly easy to formulate. Can someone think she means

something by an utterance when she means nothing? True, there were several points

that it was easy to forget – such as that we are chiefly concerned, not with whether the

utterance is of an acceptable sentence of the language, but with whether the utterer

means anything by it. But that did not make the central problem difficult to formulate

clearly and briefly. As regards distorted reflexions of grammar, however, I am not

sure how best to formulate the problem.

As I have said, it seems that someone making the error in question would have to

mean something. It is also perhaps true to say that she would have to be in some way

wrong about what she meant, if only because the phrase is so vague. Should we

formulate the problem as: ‘Could someone see a rule of grammar as a highly general

(probably necessary) truth about the world?’? The fact that this formulation uses a

metaphor might worry us. I am not sure that it would help to attempt to conceal this

by the use of more formal, Latinate words – ‘apprehend’ for ‘see’, for example. How

about: ‘Could someone misunderstand a rule of grammar as a highly general

(probably necessary) truth about the world?’? This is in fact the best I can do at

present and I commend the problem thus formulated to the reader.

38

38

Conclusion: On Being A Little Less Preposterous As A Philosopher.

It will be evident that my position is much closer to W. E. Johnson’s than to Chris

Coope’s. Johnson boldly asserted that if he claimed that a sentence had meaning for

him, no one had a right to contradict him. I would simply ask what right anyone has

to contradict me if I say I mean or understand something by a sentence. I do not think

anyone has yet come near to establishing such a right – not even the later Wittgenstein

or his followers at their subtlest. And it should be pretty obvious by now that it is

going to be very difficult to do so, though I concede that the question of the logical

possibility of IOMs has not been properly investigated and that something might yet

be made of the therapeutic approach.

It is not often mentioned that there are no uncontroversial examples of philosophical

nonsense. Perhaps it is thought too obvious to require mention. After all, the claim

that some philosophical thesis or question is nonsense implies that someone has been

taken in by an IOM, so at least one person must at some time have failed to recognise

something as philosophical nonsense. But one thing that emerges from the foregoing

discussion is that there are no uncontroversial examples of what one might call

‘analogue philosophical nonsense’ – sentences (or, better: utterances) that parallel the

sorts of thing philosophers come out with and are obvious nonsense. And this ought

to make one wonder. Philosophers may debate interminably about whether this or

that famous argument is valid, but at least there are in elementary logic

uncontroversial examples of valid and invalid arguments.

One might with mild cynicism reply that philosophy is the subject where nothing is

ever finally and definitively established. But there is an irony here. Nonsensicalism,

39

39

at least in its Twentieth Century forms, was motivated to a large extent by

exasperation with philosophy’s failure to arrive at agreed solutions to its problems.

(Often an invidious comparison with science was made or implied.) Surely, it was

thought, there must be something wrong with the questions philosophers were

asking.52 But nonsensicalism has merely given philosophers new things to disagree

about.

So how preposterous is nonsensicalism? Well, I have admitted that it is just possible

there is something in it. But the way philosophers have blithely forged ahead without

regard to the very obvious difficulties with the notion of philosophical nonsense is

truly preposterous. And the claim to know better than someone else whether she

means anything – at least when this is not part of some kind of therapeutic

interchange – is preposterously arrogant. I do not think one is likely to find anything

in the metaphysicians of the past that is significantly worse. And are there no cases

where nonsensicalist methods have been used to defend views that are in themselves

preposterous? What of Norman Malcolm’s views on dreaming? (Malcolm: 1959;:

2005, 18-33.)

Appendix: Past Critics Of Nonsensicalism.

Over the years, I have been on the look-out for philosophers who have exhibited some

scepticism about the notion of philosophical nonsense. I have not found many but I

think my finds are worth bringing together. To me they usually seem to err in the

direction of giving too much attention to the meaning or lack of it of particular

sentences and too little to the state of mind of those uttering them. But that does not

40

40

mean they cannot be studied with profit. With the possible exception of Popper, I

have not included anyone just for being critical of logical positivism.53

a) Richard Price. As already mentioned, Charles Pigden has emphasised the

importance of Price’s criticism of Hume.

b) Karl Popper. Popper is well-known for his disagreements with the positivists and

for his view that philosophers have become too concerned with questions of meaning

and meaningfulness. I do not know, however, that he ever mounts a general attack on

nonsensicalism. In Unended Quest he mentions the criticism of the positivists that

they had to understand, accord meaning to, a claim in order to apply the verification

principle to it. (Popper, 1976: 80) This, I have suggested, can be generalised into a

problem for all nonsensicalists, but I do not know that Popper ever generalises it in

that way. He writes as if the criticism was made early on, when the logical positivists

were still in business, but, frustratingly, he does not say who first made it. Was it

perhaps Popper himself? There will no doubt be experts on Popper who will be able

to give a more rounded picture of his attitude to the question of philosophical

nonsense than I can. A similar comment applies to

c) W. V. O. Quine. In Word and Object he writes:

[T]here has been a concern among philosophers to declare meaningless, rather

than trivially false, such predications as ‘This stone is thinking about Vienna’

(Carnap) and ‘Quadruplicity drinks procrastination’ (Russell). Here we

witness sometimes just a spontaneous revulsion against silly sentences and

sometimes a more remote project of cutting meaningful language down to

41

41

something like empirical size. But since the philosophers who would build

such categorial fences are not generally resolved to banish from language all

falsehoods of mathematics and like absurdities, I fail to see much benefit in

the partial exclusions they do undertake; for the forms concerned would

remain quite under control if admitted rather, like contradictions, as false (and

false by meaning, if one likes). (Quine, 1960: 229)

This seems a pretty clear repudiation of the entire nonsensicalist enterprise. On the

other hand, Pigden accuses Quine of resorting to nonsensicalist tactics in ‘Two

Dogmas of Empiricism’. (Pigden, 1987: 182-85) I leave it to experts on Quine to sort

out the details and development of his views on meaning and its absence.

d) More on Falsidal Theories.

A number of contributors to the Australasian Journal of Philosophy put forward and

debate ‘falsidal’ alternatives to nonsensicalism.54 I have already mentioned certain

reservations I have about the former and the name they have been given. My main

complaint is that more needs to be said about those who might utter the locutions

whose meaningfulness is being debated. But the nonsensicalists too have been at fault

here. Only the later Wittgenstein and some of his followers have made much attempt

to put the utterers of supposed philosophical nonsense at the centre of the discussion.

Michael Bradley’s view will serve as an example of the ‘falsidal’ approach. He

writes:

The method which I shall now describe is a very obvious, even simple-minded

one … [It] is based, naturally enough, on a general consideration about truth.

That consideration is that if an entity e has a property incompatible with its

42

42

having property P, or lacks a property required for its having P, then it is false

that e has P, and true that e does not have P. (Bradley, 1978:204-5)

Using this principle, he argues that ‘Friday is in bed’, ‘[The number] 7 likes dancing’

and ‘The Battle of Hastings likes tomato soup’ are false rather than nonsense and that

their negations are true rather than nonsense. Most of the paper is concerned with

arguing that he is not covertly appealing to (and cannot be forced into appealing to)

the notion of nonsense, and that his position does not commit him to counter-intuitive

conclusions such as that ‘The Battle of Hastings is indifferent to tomato soup’ is true

and indeed necessarily true. Much depends, as might be expected, on attention to

negation and what exactly is being negated.

The dispute between philosophers like Bradley and their opponents55 illustrates just

how far some philosophers are willing to go in discussing the meaningfulness or

otherwise of combinations of words without asking why anyone might utter them. It

might be replied:

Philosophers are interested in the distinction between sense and nonsense

mainly because some of them hold that it, rather than that between truth and

falsehood, is the key distinction we need when assessing philosophical claims.

But rather than starting with claims such as ‘God exists’, ‘Time is unreal’ and

‘I cannot tell whether I am now dreaming’, which are clearly philosophical

and about which philosophers tend already to have strong views, it is better to

consider trivial, perhaps even ridiculous, examples and try to establish general

principles about when one should ascribe meaning and when not. Then, it is

to be hoped, one will be in a better position to assess philosophical claims.

43

43

But my claim is not about the discussion of examples that are in some way

unphilosophical, trivial or ridiculous56 but about the neglect of the relationship

between the suspect sentences and those who might come to utter them. If we are to

get the problem of philosophical nonsense properly in focus, we need to ask whether

IOMs are possible and if so, how. These are questions about the utterers of alleged

philosophical nonsense. Nevertheless, the mere fact that Bradley can make a

plausible case for his view ought to worry the nonsensicalist. Perhaps traditional

philosophers were right to emphasise the true/false rather than the sense/nonsense

polarity.

Let me now deal with something I have been postponing, perhaps to the annoyance of

some readers. I have complained of the tendency of falsidal theorists and their

opponents to ignore the question why anyone might utter the problematic sentences.

Surely, it will be said, the very existence of the debate provides an answer to this.

Philosophers discuss, and hence utter, sentences like ‘Friday is in bed’ in the hope of

shedding light on the traditional questions of philosophy and the answers suggested to

them. Not only that, but if someone were to prove57 that ‘Friday is in bed’ is

nonsensical rather than false, then he would have met my demand for a demonstration

that IOMs are possible. After all, Michael Bradley and others seem to see a meaning

in such sentences. These two points I must concede. But what of the traditional

problems of philosophy? Do IOMs occur here? And even the demonstration that

IOMs are possible would now explain how they are possible. I still maintain that

there is a limit to what can be achieved by discussing sentences without reference to

an utterer. It must inevitably be ancillary to any discussion of the main issues of

philosophy.

44

44

e) Hugo Meynell. In ‘On Understanding the Unintelligible’, Meynell criticises the

way philosophers use the word ‘unintelligible’ to reject the views of their opponents.

(Meynell, 1973-74) It seems to me that his arguments apply with little or no

alteration to the parallel use of ‘nonsensical’, ‘senseless’ and ‘meaningless’. His main

claim is that, when a philosophical thesis is called ‘unintelligible’, the charge usually

proves on inspection to be one of inconsistency. This would make his view of such

charges a falsidal one, assuming that contradictions are simply false, which he clearly

does assume. (Meynell, 1973-74: 109) But he is discussing genuine philosophical

theses, not invented examples like ‘Friday is in bed’. Unfortunately, he discusses a

limited range of examples (all of which are theological), so it is not clear that he has

any general technique for showing that philosophical accusations of unintelligibility

are better reinterpreted as accusations of inconsistency. As regards the examples he

does discuss, he is very plausible in arguing that the accuser attributes a meaning to

the thesis he calls ‘unintelligible’ and on the basis of this attributed meaning finds an

inconsistency in it. But, leaving aside questions, virtually all philosophical theses

have been called unintelligible (nonsense, meaningless, senseless) by somebody. Can

all such charges be reinterpreted as ones of inconsistency?58

At one point Meynell seems to accept the positivistic claim that ‘judgments for or

against which evidence will not ever and could not conceivably ever be forthcoming’

are meaningless. (Meynell, 1973-74: 111) This is surprising. He argues that one has

to accord meaning to a claim in order to show that it is inconsistent. Why does one

not have to accord meaning to a claim in order to show that there could never be

45

45

evidence for or against it? Whatever his answer to this would be, the general thrust of

his paper is clearly anti-nonsensicalist.

f) A. C. Ewing. In his paper “Meaninglessness’ Ewing uses an argument that sounds

very like the Price-Pigden argument. Can we justify

giving a definition of meaning that would make meaningless many statements

that prior to the definition were held by everybody to have meaning? If a

definition does this it is, prima facie at any rate, not an account of the way the

word defined is usually used by people … [It] can only be justified if an

independent argument is given to show that in the cases where meaning is

attributed to statements which cannot be verified the term ‘meaning’ does not

mean anything. [Surely he means: the term ‘meaning’ is wrongly attributed.]

(Ewing, 1968: 23)

But this argument is only used in passing and against verificationism. Ewing may not

have appreciated its general availability as a weapon against ‘coercive theories of

meaning’.

He goes on to give a falsidal account of ‘Quadratic equations go to race-meetings’.

(Ewing, 1968: 28-31) It is similar to Bradley’s treatment of ‘Friday is in bed’. But

his opposition to nonsensicalism is not total:

There are, I think, cases where philosophers have uttered meaningless

expressions thinking they had a meaning … [N]o doubt some philosophers

have thought probability was a quality, so that you could say A was probable

significantly without asserting or understanding any data to which the

46

46

probability of A was relative, while probability is a relative term. (Ewing,

1968: 27-28)

Nevertheless, as with Meynell, it can be said that his general position is anti-

nonsensicalist.

g) A. N. Prior. I have left to the end what I consider the gem of the collection: Arthur

Prior’s article ‘Entities’ (Prior, 1954, 1976). Prior is best known as a pioneer of tense-

logic. I am not sure how much influence the article in question ever had. Its title

would hardly have helped advertise it. And the fact that it appeared the year after the

Philosophical Investigations would have pretty well ensured that it was submerged

beneath the tide of post-positivist, late-Wittgensteinian nonsensicalism. It is notable

for several things.

i) Rather than giving us yet another critique of logical positivism, Prior directs his fire

against Russell’s Theory of Types.

ii) He suggests a way of dealing with Russell’s Paradox that does not make use of the

notion of nonsense. It is subtle and is complicated by the fact that he translates the

paradox, which Russell formulates in terms of classes, into one about properties. I

shall not attempt to expound it here except to say that he regards ‘Virtue is square’ as

false and ‘Virtue is not square’ as true, and accepts that the same applies to statements

involving ‘self-inherence’59 (Prior, 1976: 26-27) Thus ‘Squareness is square’ is false

and ‘Squareness is not square’ is true.

iii) He finds what I call ‘nonsensicalism’ (or something very like it) in Aquinas and

the Thomist tradition (Prior, 1976: 29) and also in the British Hegelians. (Prior, 1976:

26) If he is right about this, then not only does the polemical use of the notion of

nonsense go back even further than Pigden claims, but philosophers who would be

47

47

regarded by many nonsensicalists (and not just positivists) as paradigmatic talkers of

nonsense turn out to be among those who accuse others of talking nonsense.

iv) He stresses the role of Russell’s Theory of Types in stimulating philosophers to

look for instances of nonsense elsewhere in philosophy and regards this as an

unfortunate influence.60 (Prior, 1976: 26) If Russell had found some other way of

dealing with his paradox, the history of philosophy during the past century might have

been very different.61

I hope I have said enough about the above writers to indicate why their work on the

notion of philosophical nonsense should repay careful study.

Footnotes.

1. Perhaps some philosophers are as simple-minded as this. But is there not also a

marked tendency for those philosophers who still make accusations of talking

nonsense to treat the positivists as scapegoats? They say in effect, ‘The positivists

made the mistake of trying to set up criteria of meaningfulness, aiming indeed at a

single criterion of meaningfulness. But we have moved on since then’.

2. Paul Boghossian, in discussing the Sokal hoax, uses the word ‘nonsense’ to

describe some of the deliberate mistakes Sokal inserted into his article for Social Text.

(Boghossian, 2000: 173, 184) But the burden of his argument is that the editors cannot

have understood what Sokal was saying; otherwise they would have spotted the

mistakes. This seems to be a case of a philosopher using the word ‘nonsense’ in the

colloquial sense. Sokal mentions that he did include a few sentences that were not

intended to mean anything, but only a few. (Sokal and Bricmont, 1998: 248-49)

48

48

3. How often this is the case I am not sure. Certainly I do not think the dictum of

Heidegger’s that has provoked such English renderings as ‘The Nothing nothings’,

‘The Nothing noughts’ and ‘The Nothing noths’ is typical. (See Rundle, 1990: 214)

4. I discuss drug-induced and schizophrenic ‘word-salads’ in, Dearden, 2005: 71-77.

See also my ‘Louis Sass and Rupert Read on Schizophrenia’ on academia.edu.

5. Pigden notes that those who do not make philosophical accusations of talking

nonsense usually just ignore those who do. (Pigden, 2010: 180 n. 18) There has yet

to occur a straight confrontation between the two schools of thought. How there can

persist this remarkable and rarely noticed split running right through the middle of

philosophy, indeed of analytical philosophy, is a mystery to me.

6. An interesting illustration of this can be got by modifying an example of

Wittgenstein’s. Writing of the experience of ‘knowing how to go on’ he imagines

someone on whom

light was always seeming to dawn …he exclaims, ‘Now I have it!’ and then

can never justify himself in practice – It might seem to him as if in the

twinkling of an eye he forgot again the meaning of the picture that occurred to

him. (PI, I, 323.)

Could not this happen to someone searching for the meaning of some dark saying?

7. It is of course perfectly possible that the reasons given for declaring a thesis to be

nonsense might involve drawing attention to genuine defects in it. Unverifiability, for

example, might well be a defect in certain circumstances. And think of

Wittgenstein’s insistence that (what he regards as) useless locutions like ‘A thing is

identical with itself’ are meaningless – no one would deny that uselessness is a defect.

49

49

8. One could look at the question the other way round. Someone asserts that one

cannot be mistaken in thinking there is something one means and someone else tries

to construct a counter-example to this.

9. I am not among these. Yet I feel that analytical philosophers have been

exceedingly remiss in their neglect of this question. The explanation cannot be that

they are afraid that the investigation would undermine analytical philosophy, since not

all analytical philosophers make nonsensicalist accusations. Of the founding fathers of

analytical philosophy – Frege, Russell and Moore – none made much use of them in

their polemics. (On Frege and Russell see Pigden, 2010: 163-65)

10. I discuss about ten such suggestions in Dearden, 2005:.81-96, as well as one of a

more general nature that is arguably implicit in the later Wittgenstein (97- 105). None

of the former are adequate as they stand. The latter is somewhat more promising.

11. I suggest (Dearden, 2005: 75-77) that, if there is such a thing as philosophical

nonsense, it will have to be a far better imitation of sense than are many psychotic and

drug-induced utterances.

12. Stephen Mulhall apparently thinks otherwise. We have an ‘everyday capacity to

distinguish sense from nonsense’ which we can deploy in a philosophical context. It

is ‘an ability that can equally well be laid claim to by any competent speaker, and

hence by any philosophical interlocutor’. (Mulhall, 2007: 10) His talk of

distinguishing sense from nonsense (Mulhall, 2007: 6-10) seems to me to be a little

glib and to mask a crucial distinction: the failure to find sense is not necessarily the

detection of the absence of sense.

13. I do not know who first pointed out this difficulty. See the appendix under: Karl

Popper.

50

50

14. Ayer speaks of the factually meaningful and the factually meaningless. (Ayer,

1946: Chapter 1) ‘Cognitively meaningful’, ‘empirically meaningful’ and

‘empirically significant’ were also used.

15. Any sequence of sounds or marks could be used meaningfully. The

Wittgensteinian Anthony Kenny says of ‘The class of men is a man’ that ‘there is

nothing in that set of sounds which disqualifies it from being given a meaning’.

(Kenny, 1973: 43) I cannot see how anyone could seriously disagree. Yet pondering

this point alone should be sufficient to bring out the problem with diagnosing the

talking of nonsense.

16. Hume’s position is by no means as clear-cut as is often assumed. Much of what

he says about the self and necessary connexion reads better as an account of how we

acquire certain false or unjustified beliefs than of how we come to think we mean

something by certain words when we mean nothing by them. Is this just another

example of his tendency to forget what his official position is or does it mean that his

nonsensicalism is not as thoroughgoing as might be supposed? I hope to discuss the

matter elsewhere.

17. I have been unable to decide whether this talk of ‘giving sense’ is a specifically

philosophical idiom or ordinary English. The earliest philosophical occurrence of it

that I have been able to find is in the Blue Book (p. 7) where Wittgenstein says that

the phrase ‘a locality where thought takes place’ ‘has sense if we give it sense’.

18. David Lumsden makes a similar point. (Lumsden, 2005) It might be objected

that the ‘proto-meanings’ one might postulate when trying to explain the origin of

language would lie at the opposite end of a scale of sophistication from whatever

states of mind lie behind talk of, say, classes of classes. No doubt; but all I am doing

51

51

is warning against simply assuming that there is always a sharp distinction between

meaningfulness and meaninglessness.

19. David Wiggins, speaking of the difficulties the positivists had formulating the

verification principle, remarks that they were ‘difficulties we still have in seeking

some principled philosophical exclusion of that which really is nonsense’. (Wiggins,

2006: 178 n. 11, emphasis added) Might not part of the problem be a certain

indeterminacy in the notion of the meaningful?

20. Perhaps when Pigden says that a very successful theory of meaning might lead us

to make revisions to what we consider meaningful (Pigden, 2010: 179) he is

envisaging a situation in which some ways of making the concept of meaning more

precise than it is now would seem more natural or convenient or scientific than others.

He does mention the possibility of borderline cases of meaningfulness. (Pigden,

2010: 179)

21. Wittgenstein himself uses the word ‘operations’ in this connexion. (AWL, 64)

22. If one were really strict, one would perhaps enclose every occurrence of

‘sentence’, ‘thesis’, ‘question’ and many other words in scare-quotes whenever

meaningfulness was in question. Alternatively one could precede them with words

like ‘supposed’ and ‘putative’. I do not think much would be gained by this, since no

one is likely to forget what the issue is.

23. David Stove in ‘What is wrong with our thoughts? – A neo-positivist credo’, the

final chapter of The Plato Cult and Other Philosophical Follies (Stove, 1991), gives a

long list of mainly philosophical claims most of which he thinks are in some sense

nonsense (Stove, 1991, 187). But in what sense? He thinks it a mistake to dismiss

such claims as simply meaningless. (Stove, 1991: 189, 194-95) And he is clearly not

using ‘nonsense’ in the colloquial sense to refer to the patently false. (Stove, 1991:

52

52

194) And he does not think many philosophical claims can be shown to be self-

contradictory. (Stove, 1991: 66, 194) In fact he does not claim to know what is

wrong with most of them. (Stove, 1991: 194-202) I can only assume that he has a

hunch that if we did know what was wrong with them, we would feel that ‘nonsense’

was an appropriate and revealing word to use of them.

24. Of course there will probably be philosophers who think (of the sentences that

have the form of assertions) that some are false and some are nonsense, though I

suspect there will a tendency for philosophers to make a clean sweep of all of them –

one way or the other.

25. (a) Googling ‘falsidal’ will show what I mean. I have come across one interesting

discussion. (Daneau, 2006) On the question of what one can say about God the

author maintains that certain Wittgensteinian doctrines must be rejected if theologians

are to maintain the via negativa:

The most important is his view that ‘category errors’ are nonsense. I hold to

the ‘falsidal’ view - that category errors are just false, not nonsensical. I can’t

see how you could make negative predications of God if category errors were

nonsense, because every negative predication would bring with it a positive

predication. So saying, ‘God is not coloured’ would either be false (if God

really was coloured), or nonsense, on Wittgenstein’s view. The same would

be true of things we want to say like ‘God is not temporal’.

(b) Ofra Magidor’s recent book Category Mistakes has made me aware that there is a

larger literature in philosophy and linguistics than I had realised on whether absurd

sentences like ‘Friday is in bed’ are best regarded as false, truth-valueless or

meaningless. But I would still say that the debate is not particularly well-known and

would continue to emphasise that, if it is to shed much light on the notion of

53

53

philosophical nonsense, far more needs to be said about the utterers of the

problematic utterances.

26. See appendix under: More on Falsidal Theories.

27. Perhaps this will not seem so very terrible to some readers. But when we come to

consider the austere conception of nonsense we will see that regarding contradictions

as simply meaningless is not an attractive option.

28. Baker and Hacker (1986: 265-66, 279, 289) do seem to countenance this.

29. It would be very easy to set up a convention for telling the time of day (at

particular places) on Mars that exactly paralleled ours on earth, especially since the

Martian day is similar in length to ours. Even so, however natural the convention

might seem, it would have to be established. It is, so far as I know, not already in

place.

30. For a recent example see Heather Gert, 2010: 142, 148 n.4.

31. Neither Wittgenstein nor his followers seem very careful on this point. Rupert

Read however expresses it thus: ‘Metaphysical use is, roughly, only a variety of use in

the same kind of way as a decoy duck is a variety of duck’. (Read, 2010: 65)

32. As Wittgenstein says (BB, 57), it might mean: The other people are only

pretending. It will be noticed that I have not committed myself very far as to whether

Wittgenstein is right to recognise this kind of philosophical nonsense. I think he is

right to this extent: if IOMs are possible, then it will be possible for someone to be

mistaken in thinking there is something he means by ‘p’ even when ‘p’ is an

acceptable sentence of the language.

33. I am indebted to Steve Kupfer for emphasising to me that PI 500 is misleading as

it stands.

54

54

34. I could cite dozens of examples of the complacent assumption that IOMs are

possible. The passage from Coope quoted earlier is one. Here is another:

‘Wittgenstein always tries to demonstrate how easy it is for us to employ language in

a way that deprives it of its sense without realising that that is what we are doing.’

(Pears, 1988: 461) It is clear from the book as a whole that Pears agrees that making

this error is not only possible but easy.

35. Diamond formulates the view very clearly in Diamond, 1991, but without

attempting to prove it. A good statement of the case for it is in Conant, 2000.

36. Called the ‘Carnapian’ view by Witherspoon (2000) and the ‘substantial’ view by

Conant (2000) and others.

37. The austere view will have to be extended to cover the case of talking nonsense.

The austere theorist will have to say, I take it, that if someone does not mean anything

by an utterance, he does not mean anything by any part of it, even if the parts appear

to be familiar words of the language. (Dearden, 2005: 51-53 and ‘What Talking

Nonsense Might Be’ on academia.edu.)

38. There is supposed to be a view that contradictions are simply nonsense, i.e.

meaningless. Hilary Putnam mentions it but does not attribute it to anyone. (Putnam,

2000: 227.) He dismisses debate about the matter as ‘futile’. As far as I can see,

anyone who did regard contradictions as nonsense would have to recognise ‘positive

nonsense’, nonsense that is nonsense on account of the meanings of its constituent

parts, and thus reject the austere conception.

39. Hacker (1986: 208) seems to accept this with a curious nonchalance. Proof,

apparently, is obsolete in philosophy. One does not refute the sceptic by proving that

we know what the sceptic doubts but ‘by showing that sceptical doubts make no

sense’. If showing that something is nonsense is not proving that it is, then what is it?

55

55

And what kind of cogency or analogue of cogency does it have? (Dearden, 2005:

118.)

40. He argues that they in effect attribute ‘quasi-meanings’ to utterances in order to

prove that they lack real, honest-to-goodness, meanings. But it is doubtful whether

anything could play this role of being a candidate meaning, an entity that one might

investigate to see whether it is a real meaning. In a similar vein, David Pears suggests

that for Wittgenstein, though there are candidates for reality, namely possibilities,

there are no ‘candidates for possibility’. (Pears, 1987: 172, 189; 1988: 215-16, 356,

412, 448-9)

41. When I come across a new book on Wittgenstein I look in the index for entries on

‘nonsense’, ‘senseless’, ‘meaningless’ and so forth. Often I do not find any and this is

not always because the index has not been conscientiously put together. While this

neglect of Wittgenstein on nonsense might not quite be a case of Hamlet-without-the-

Prince (Berkeley-without-his-Immaterialism), it is surely not far off. Such selectivity

cries out for some justification.

42. My own feeling is that it is in the Blue Book that Wittgenstein is at his most

tentative and undogmatic in diagnosing nonsense.

43. A good introduction to this aspect of Wittgenstein’s philosophy is provided by

Gordon Baker. 2004: Chapters 8-10.

44. See A. J. Ayer, 1946, especially Chapter 8 – ‘Solutions of Outstanding

Philosophical Disputes’ (actually, attempted dissolutions). Some of Wittgenstein’s

remarks about his method’s resemblance to the treatment of an illness are only

therapeutic in this sense, e. g. PI, I, 254-55, where he speaks of ‘what a mathematician

is inclined to say about the objectivity and reality of mathematical facts’ as

56

56

‘something for philosophical treatment’. There seems to be no suggestion that

mathematicians are aware of their need for treatment.

45. That scepticism is nonsense is something he often claims (see, for example, n.36).

For his views on the status of ‘Artificial Intelligence’, see Hacker, 1990: 160-69.

46. They are also the ones most reminiscent of psychoanalysis, though I shall not

develop that point here. I have found the excerpt from the Big Typescript (sections

86-93) in Klagge and Nordmann (1993: 160-199) particularly useful in giving some

idea of Wittgenstein’s conception of therapy. If my account of it seems somewhat

impressionistic, I can only say that it is the best I can do at present. The work of other

commentators on this topic strikes me as being equally so, whether or not they would

be prepared to admit it.

47. A particularly strong statement of this is in WVC, 183:

Controversy always arises through leaving out or failing to state clearly certain

steps, so that the impression is given that a claim has been made that could be

disputed.

Two comments seem in order. First, the strategy of breaking down one’s argument

into steps that are too short to admit of doubt is more than a little reminiscent of

Descartes. Second, let me point out again how different this procedure would be from

the polemical approach of most of philosophy’s nonsense-hunters.

48. Hacker remarks (Hacker, 1996: 241) that ‘the motivations behind many

nonsensical philosophical assertions may include a grain of truth or an insight that is

distorted’. The distorted insight could presumably be into a rule of grammar. Surely

this concession of Hacker’s fits badly with the claim that the one who has the insight

means nothing by what she says.

57

57

49. For example John V. Canfield says of the passage already quoted from Moore’s

report:

Wittgenstein gives a clear account of how he uses the term ‘nonsense’

hereabout. Nonsense occurs if someone tries to assert as a ground level claim

in a language game something that is part of the grammar of that game. In

general, Wittgenstein has a tendency, which survives to the Philosophical

Investigations, to restrict ‘say’, ‘sense’ and related words to what can be said

as a contingent claim in some or another language game.

Danièle Moyal-Sharrock, in ‘The Good Sense of Nonsense’, also takes this line (as

the title to some extent indicates), though she is not primarily concerned with

Wittgenstein’s later philosophy. According to her, Wittgenstein’s use of the word

‘nonsense’ is not always pejorative but sometimes discriminatory: nonsense can

demarcate or elucidate the bounds of sense. (Moyal-Sharrock, 2007)

50. Cf. PG, 129, where he is less tentative.

51. Not surprisingly, Morris Lazerowitz, who makes a serious attempt to develop the

disguised linguistic recommendation view, appeals to psychoanalytical considerations

and makes much of the metaphysician’s unconscious motivations. (1955, 1968)

52. The plausibility of this argument was never very great. Has it not been decisively

rebutted by Colin McGinn’s suggestion that philosophical problems are just those

problems that the human intellect is not adapted – or only very badly adapted – to

handling? (McGinn, 1993) Note that his suggestion does not have to be right. The

mere fact that it could be right is enough. McGinn’s thinking here, it should be noted,

is in part inspired by Chomsky (McGinn, 1993: 155-56 n. 8).

53. In any case the positivists were, as Scott Soames has brought out, among their

own most astute critics. (Soames, 2003.)

58

58

54. Examples include Arthur Prior (1954; 1976) of whom more later, Robin Haack

(1971) and Michael Bradley (1978).

55. These latter include Ross Brady and Richard Routley (1973) and Leonard

Goddard (1970). Further references are given in Bradley (1978).

56. Nor am I joining with those who say that philosophers have become too

concerned with language.

57. Perhaps one should not expect conclusive demonstrations in philosophy. Well

then, imagine someone producing a very powerful argument that ‘Friday is in bed’ is

nonsense and also a very convincing account of where a falsidalist like Bradley goes

wrong. (I have no idea how this could be achieved.)

58. Meynell does not, it should be noted, discuss Wittgenstein’s nonsensicalism.

59. Although he thinks that statements about self-inherence are perfectly meaningful,

he denies that there is a property of self-inherence (or of non-self-inherence) and

indeed exploits this denial to avoid Russell’s Paradox. (Prior, 1976: 27, 29-30) But I

doubt whether my mentioning this will convey very much to anyone who has not read

the paper.

60. As already pointed out, Russell was not himself much inclined to accuse his

philosophical opponents of talking nonsense. The careful wording of Prior’s

comments on Russell suggests to me that he was aware of this.

61. It has been claimed that the sort of move made by Russell in his Theory of Types

will not always serve its purpose since similar paradoxes can be constructed even if

one is prepared to dismiss the seemingly-meaningful-but-paradoxical as in fact

meaningless. (Ushenko, 1955) If this is so, then we have an additional reason for

viewing the influence of that theory with some suspicion.

59

59

References.

Ayer, A. J., 1946, Language, Truth and Logic, 2nd edn. London: Gollancz.

Baker, G. P., 2004, Wittgenstein’s Method – Neglected Aspects, ed. Katherine J.

Morris. Oxford: Blackwell.

Boghossian, P., ‘What the Sokal Hoax Ought to Teach Us’, in The Sokal Hoax – The

Sham That Shook the Academy, ed. the editors of Lingua Franca, 2000,

University of Nebraska Press.

Bradley, M., ‘On the Alleged Need for Nonsense’, Australasian Journal of

Philosophy, 56, 1978.

Brady, R. and Routley, R, ‘Don’t Care Was Made to Care’, Australasian

Journal of Philosophy, 51, 1973.

Canfield, J. V., 1981, Wittgenstein: Language and the World, University of

Massachusetts Press.

Conant J.., ‘Elucidation and Nonsense in Frege and Early Wittgenstein’ in Crary, A.

and Read, R. (eds.), 2000, The New Wittgenstein. London: Routledge.

Coope, C., ‘The Doctor of Philosophy Will See You Now’ in Anthony O’Hear (ed.).

2009, Conceptions of Philosophy, Cambridge: C. U. P.

‘Daneau’, ‘The Problem With Analogy’:

www.theologyweb.com/campus/archive/index.php/t-85793.html.

Dearden, I., 2005, , Do Philosophers Talk Nonsense? – An Inquiry into the Possibility

of Illusions of Meaning. New Romney: Teller Press.

Diamond, C., ‘What Nonsense Might Be’ in Diamond, C., 1991, The Realistic Spirit

- - Wittgenstein, Philosophy and the Mind. Cambridge, Mass: MIT Press.

Ewing, A. C., ‘Meaninglessness’ in Ewing, A.C., 1968, Non-Linguistic Philosophy.

London: George, Allen and Unwin.

60

60

Gert, H., ‘Knowing that the Standard Metre is One Metre Long’ in Jolley, K. D.

(ed.), 2010, Wittgenstein – Key Concepts. Durham: Acumen.

Goddard, L., ‘Nonsignificance’, Australasian Journal of Philosophy, 48, 1970.

Haack, R., ‘No Need for Nonsense’, Australasian Journal of Philosophy, 49, 1971.

Hacker, P.M.S., 1986, Insight and Illusion, rev. ed. Oxford: Clarendon Press.

Hacker, P.M.S., 1980, 1986, 1990, 1996, Analytical Commentary on the

‘Philosophical Investigations’, (4 vols. The first two with G.P. Baker)

Oxford: Blackwell.

Hacker, P. M. S. and G. P. Baker, 1984, Language, Sense and Nonsense.

Oxford: O.U.P.

Hertzberg, L., ‘The Sense is Where You Find It’ in T. McCarthy and S Stidd (eds.),

2001, Wittgenstein in America, Oxford: Clarendon Press. Corrected version,

to which references are given here, at:

web.abo.fi/fak/hf/filosofi/…/The_Sense_Is_Where_You_Find_It. pdf.

Hume, D., 1978, A Treatise of Human Nature, Selby-Bigge/Nidditch (eds.). Oxford:

O .U. P.

Hume, D., 1975, Enquiries Concerning Human Understanding and Concerning the

Principles of Morals, Selby-Bigge/Nidditch (eds.). Oxford: O. U. P.

Kenny, A., 1973, Wittgenstein, Harmondsworth: Allen Lane The Penguin Press.

Lazerowitz, M., 1955, The Structure of Metaphysics. London: Routledge and Kegan

Paul.

Lazerowitz, M., 1968, Philosophy and Illusion. London: Routledge and Kegan Paul.

Lumsden, D., ‘How Can a Symbol System Come Into Being?’ in Dialogue, XLIV,

2005.

Magidor, O., 2013, Category Mistakes, Oxford: O.U.P.

61

61

Malcolm, N., 1959, Dreaming. London: Routledge and Kegan Paul.

McGinn, C., 1993, Problems in Philosophy: The Limits of Inquiry. Oxford:

Blackwell.

Meynell, H., ‘On Understanding the Unintelligible’, in Analysis, 34, 1973-74.

Moyal-Sharrock, D., ‘The Good Sense of Nonsense: A Reading of Wittgenstein’s

Tractatus as Nonself-repudiating’, in Philosophy, 82, 2007.

Mulhall, S., 2007, Wittgenstein’s Private Language: Grammar, Nonsense and

Imagination in ‘Philosophical Investigations’, 243-315. Oxford: Clarendon

Press.

Pears, D. F., 1987 and 1988, The False Prison: A Study in the Development of

Wittgenstein’s Philosophy (2 vols.). Oxford: Clarendon Press.

Pigden, C., ‘Two Dogmatists’ in Inquiry, 30, 1987.

Pigden, C., ‘Coercive Theories of Meaning, or Why Language Shouldn’t Matter

(So Much) to Philosophy’ in Logique et Analyse, 210, 2010.

Popper, K. R., 1976, Unended Quest: An Intellectual Autobiography. Glasgow:

Fontana/Collins.

Prior, A. N., ‘Entities’ in Prior, 1976, Papers on Logic and Ethics, Geach and Kenny

(eds.). London: Duckworth. (Originally in the Australasian Journal of

Philosophy, 32, 1954.)

Putnam, H., ‘Rethinking Mathematical Necessity’ in Crary, A. and Read, R. (eds.),

2000, The New Wittgenstein. London: Routledge.

Quine, W. V. O., 1960, Word and Object. Cambridge, Mass: MIT Press.

Read, R., ‘Ordinary/Everyday Language’ in Jolley, K. D. (ed.), 2010, Wittgenstein:

Key Concepts. Durham: Acumen.

Rundle, B., 1990, Wittgenstein and Contemporary Philosophy of Language. Oxford:

62

62

Blackwell.

Sass, L. A., 1994, The Paradoxes of Delusion: Wittgenstein, Schreber and the

Schizophrenic Mind. New York: Cornell U.P.

Soames, S., 2003, Analytic Philosophy in the Twentieth Century, vol. I, The Dawn of

Analysis. Princeton: Princeton University Press.

Sokal, A. and Bricmont, J., 1998, Intellectual Impostures. London: Profile Books.

Stove, D., 1991, The Plato Cult and Other Philosophical Follies. Oxford: Blackwell.

Uschanov, T. P., ‘The Strange Death of Ordinary Language Philosophy’:

www.helsinki.fi/~tuschano/writingds/strange/

Ushenko, A. P., ‘A Note on the Liar Paradox’, Mind, 64, 1955.

Wiggins, D., 2006, Ethics: Twelve Lectures on the Philosophy of Morality. London:

Penguin.

Witherspoon, E., ‘Conceptions of Nonsense in Carnap and Wittgenstein’ in Crary, A.

and Read, R. (eds.), 2000, The New Wittgenstein, London: Routledge.

Wittgenstein, L: I use the following fairly standard abbreviations for his main works –

BB – Blue Book; BrB – Brown Book; OC – On Certainty; PG – Philosophical

Grammar; PI – Philosophical Investigations; TLP – Tractatus Logico-

Philosophicus (tr. Pears and McGuinness); Z – Zettel. In addition:

AWL – Wittgenstein’s Lectures, Cambridge 1932-35, from the notes of Alice

Ambrose and Margaret MacDonald, ed. Alice Ambrose, 1979. Oxford:

Blackwell.

WVC – Ludwig Wittgenstein and the Vienna Circle [1929-32], shorthand notes

recorded by F. Waismann, ed. B. F. McGuinness, 1979. Oxford: Blackwell.

M – Wittgenstein’s Lectures in 1930-33, notes by G. E. Moore, in Philosophical

Occasions, Klagge, J. and Nordmann, A. (eds.), 1993. Indianapolis: Hackett.

63

63

BT – Excerpt from the Big Typescript in Philosophical Occasions, op. cit.

64

64