[lit-ideas] Ostension and Criterion in Wittgenstein and Grice

  • From: "" <dmarc-noreply@xxxxxxxxxxxxx> (Redacted sender "Jlsperanza@xxxxxxx" for DMARC)
  • To: lit-ideas@xxxxxxxxxxxxx
  • Date: Sat, 10 May 2014 07:06:42 -0400 (EDT)

We are considering various claims, notably related to Witters, but since I  
tend to interpret what Witters is after alla Grice, I added him 'into the  
The claim concerns aspects of language. While Omar K. thinks  
self-referentiality is as issue, and while it may, in most cases, I think  
concern is a broader one, with meta-language as such. L1 used to speak  about 
need not be self-referential. Yet Witters does not seem to allow for  
'metalinguistic' explanation of linguistic phenomena -- We are mainly dealing  
with the interpretation of Witters's doctrine as propounded by McEvoy.
One of the 'paradigms' was 'naming' -- but this issue should apply to any  
aspect of communication, meaning, or language -- however vague these terms 
The show/say distinction appears largely in TLP, and while the distinctions 
 have been identified, it should be pointed out that these extremes in the  
distinction SHARE a few features too. McEvoy is concerned with the object 
of  what is allegedly being said and the object of what is shown as being 
'sense'.  So, for one, the alleged object of these two activities (to say and 
to show) is  a common one. It is true that Witters is alleging that the sense 
cannot be said.  This impossibility is described as being 'a priori' or 
'analytic': a conceptual  impossibility, as it were, rather, than, as the 
author of 'Definitions' in the  Stanford Encyclopedia notes, a mere 
So, we have two 'propositional attitudes', involved in 'saying' and  
'showing' that share an alleged object (the sense). With these common aspects 
mind, we now turn to the idea of 'criterion'.
Criteria are NOT used by Witters with regard to the say/show distinction,  
but my point was that they can. Criteria _show_ and criteria _say_, we say, 
the  alleged _sense_. McEvoy should have no problem with criteria SHOWING. 
How can I  demonstrate or provide evidence that criteria _say_? 
McEvoy is right that Witters was perhaps casual in his use of 'criterion',  
and it's only interpreters (notably Albritton, but also Wellman) who have 
tried  to make sense of this term. For example, it has been noted that 
Witters  sometimes uses 'outward criterion', as being the correct phrase, and 
Wellman  wonders if this IMPLICATES that Witters allows for inward criteria 
(Albritton's  example: I may have an inward criterion for my having a 
i.e. for my  personal use of "Ouch!", say). Wellman concedes that the use 
of 'outward' in  'outward criteria' may be rhetorical, rather than 
But the point emphasised by Witters is that criteria are public, and  
shared, by alla Grice, U and A, utterer and addressee. 
Criteria are notably DEFEASIBLE, Witters seems to be saying. Baker has  
expanded on how 'defeasibility' applies to meaning notably in the theories of 
H.  L. A. Hart, and indeed Grice has also stressed this 'ceteris paribus' 
aspect of  'communication' and 'meaning'. 
The reductio for Witters would be to admit that criteria (which allegedly  
SHOW sense) themselves can only be _shown_ rather than formulated 
explicitly.  Implicit criteria for the 'appropriate' use of this or that. 
It was this vagueness that Grice set to criticise some 10 years after the  
"Investigations" were published, in the William James lectures. He notes 
that  what we may feel is some deep scepticism (if that's the word) behind 
Witters's  campaign (but formulated by 'meaning = use') is best challenged by 
detailed  analysis. Indeed, Grice claims that the fashion by 1967, given the 
fuzziness  with which notions like 'meaning' and 'use' are employed, should 
be that meaning  ≠ use.
I propose 'ostension' as a term for Witters's 'show'. McEvoy is right that  
in the translation provided in Philosophical Investigations for the 
Augustine  passage 'show' (indeed 'shewn') is used, but 'ostension' may do for 
cases.  Ostension need NOT be related to 'definition' (as in 'ostensive 
definition')  although meaning is usually involved. And ostension should 
perhaps be  distinguished from DEIXIS and self-referentiality. Thus, 'this' as 
'this king  of France is bald' is DEICTIC. And 'This!' with finger pointing 
in the direction  of a rat may be an ostensive definition of 'rat'. As Gupta 
notes, philosophers  are usually NOT interested in, say, an ostension (or 
ostensive definition) of  things like 'know'. "You what to know what 'know' 
means? It means THIS' -- how  can you _show_ the meaning of 'know'? 
Piero Straffa was an aristocrat (almost) and Palma seems to minimise that  
factor, although it is true that the gesture he used did show SOMETHING to  
Witters (he is credited in the Preface to the PI that Witters wrote and was 
only  posthumously published). It is not that kind of gestures that _show_ 
stuff that  Witters is into, though. He is into, it seems, the activity of 
the professional  philosopher (as Witters was -- Straffa was an economist) 
including 'showing'. It  would be the role of the philosopher or philosophical 
linguist to _show_ things  rather than _say_ things -- perhaps even to show 
senses, too.
There is a broader interpretation, where Witters may be claiming that  
'showing' is all-pervasive among communicators themselves. It is communicators  
who rely on showing rather than saying. Witters's problem seems to be that 
every  time communicators aim or purport or set to _SAY_, say, the 'sense' of 
this or  that, they find themselves in what Plato (or Kant) would call an 
'aporia' and  they are left with nothing but showing, which, Witters seems to 
think, ends up  being PRETTY effective, if not VERY effective, or _AS_ 
effective as it  can.

Or stuff.

In a message dated 5/9/2014 3:54:30 P.M. Eastern Daylight Time,  
donalmcevoyuk@xxxxxxxxxxx writes:
The explanation may be that it is not a  'technicism' as far as 
Wittgenstein is concerned, given the purposes for which  he deploys it, but it 
one in the hands of some commentators. As I  understand it, Wittgenstein is 
not using the notion of 'criterion' in the strict  sense in which we might 
say the positivist 'verificationist theory of meaning'  is offered as a 
strict criterion of "sense" - that is, a definitive (rather than  partial and 
defeasible) yardstick. In Investigations Wittgenstein's sense of  'criterion' 
is looser: more elastic and flexible, as it is deployed to reflect  the 
sometime elasticity and flexibility of many different uses of  language.

Well, but one may grant that there is a way out in the former  claim then:


"This post  may help explain why I do not  find what W writes about 
(mentioned in a  previous post by  JLS) to be at all incompatible with the 
interpretation I am   suggesting."

That post did perhaps attempt to SHOW that, though.  


It may be counter-argued that we COULD (and perhaps should)  make use of 
the idea of 'criteria' _sic_ in the plural. Or even in the  singular.

Witters may not have had something technical or  'jargo-technical' or 
'krypto-technical' (as I think Grice's word is) when he  said 'criterion', 

I would grant that there's no use of having one  word only in the plural. 
If Witters does speak of 'criteria' sometimes _sic_ in  the plural, we should 
not be prohibited from using 'criterion' in the singular.  I propose the 
capital "C" to represent "Criterion".

The 'grammar' of  'criterion' is a difficult one. I propose to relate it to 
'utterance' tokens,  symbolised by 'u'. Then we have an Utterer and an 

By uttering  u, U makes himself understood if he uses x according to 
Criterion C, which, to  echo Omar K., is one shared or recognised by the 


The idea would be that a 'criterion' is good, and good  enough, and, a 
criterion is something that may be included in what Aristotle  would call 'ta 
legomena', or 'ton legomenon': what IS SAID.

So, if there  is a criterion, which can be made explicit, on occasion, and 
upon request, in  something that is being said (a dictum), then, we would 
have a general 'theory'  according to Witters that would shed light on things 
like 'naming',  etc.

There would be a criterion C for name N -- and it is concordance  with this 
criterion that marks the appropriate use of "N".  


For the record, here is 

Gupta, Anil,  "Definitions", The Stanford Encyclopedia of Philosophy 
(Spring 2014 Edition),  Edward N. Zalta (ed.), URL =  

on  'ostensive' DEFINITION.

This, Gupta says, "occurs with ostensive  definitions."

"We can teach a boy a term, say ‘meter’, by giving the  ostensive 
definition “This stick is one meter long,” while showing the boy a  meter 

"There is a mystery here, too."

"Before the  definition is given, the boy does not understand the sentence 

‘This  stick is one meter long’."

"Yet, by using the sentence in a particular  way, the boy is brought to 
understand one of the constituents of the sentence  and the sentence itself."

"How can a sentence that lacks meaning for the  boy impart to him an 
understanding of a term and, at the same time, an  understanding of itself?"

It is THEN that the reference to Witters is  given:

"Ostensive definitions look simple but, as Ludwig Wittgenstein  observed, 
they are effective only because a complex linguistic and conceptual  capacity 
is operative in the background. It is not easy to provide an account of  
this capacity."

Perhaps one problem is with 'definition', and another  with 'philosopher'.

Gupta goes on:

"Definitions sought by  philosophers [are not of the ostensive kind]."

He gives an example that  we have discussed profusely in Lit-Ideas:

on knowing what is  false.

Gupta writes:

"When an epistemologist seeks a definition of  “knowledge,” she is not 
seeking a good way of teaching young children the word  ‘know’."

He concludes the section:

"The philosophical quest for  definition can sometimes fruitfully be 
characterized as a search for an  explanation of meaning. But the sense of ‘
explanation of meaning’ here is very  different from the sense in which a 
dictionary or an ostensive definition  explains the meaning of a word."

The entry provides two Gupta  references:

Gupta, A., 1988/89, “Remarks on Definitions and the Concept  of Truth,” 
Proceedings of the Aristotelian Society, 89: 227–246.
–––, 2006,  “Finite Circular Definitions,” in Self-Reference, edited by 
T. Bolander, V. F.  Hendricks, and S. A. Andersen, Stanford: CSLI 
Publications, pp. 79–93.
Gupta,  A. and Belnap, N., 1993, The Revision Theory of Truth, Cambridge 
MA: MIT  Press.


To change your Lit-Ideas settings (subscribe/unsub, vacation on/off,
digest on/off), visit www.andreas.com/faq-lit-ideas.html

Other related posts:

  • » [lit-ideas] Ostension and Criterion in Wittgenstein and Grice