Can We, or Should We, Take the Sceptical Challenge Seriously?

To continue in this vein, Kripke allows that '125' is justified as an answer to '68 + 57', based on the mathematics of the case. As he puts it on p. 12: "The problem is not "How do I know that 68 plus 57 is 125?", which should be answered by giving an arithmetical computation . . . ." On p. 13, Kripke insists again that the sceptical problem is not a sceptical problem about arithmetic ("at least not initially", says Kripke, which is itself a rather odd hedge since it's not clear that he ever does raise a sceptical problem about arithmetic), as follows:

"How do I know that 68 + 57 is 125?" (Why not answer this question with a mathematical proof?) At least at this stage, scepticism about arithmetic should not be taken to be in question: we may assume, if we wish, that 68 + 57 is 125.

Now, if the sceptical problem does not involve scepticism about arithmetic, then what is it sceptical about? According to Kripke, the sceptic is sceptical about the agreement between my past usage and my present usage. As Kripke says, the sceptic "merely questions whether my present usage agrees with my past usage, whether I am presently conforming to my previous linguistic intentions". (K, p. 12). The sceptic is ultimately sceptical about our ability to determine whether or not one "present answer" (e.g., '125') rather than another (e.g., '5') can be said to agree with my past usage, can be said to conform to my previous linguistic intentions. Although the sceptic's scepticism is expressed in several different ways, it essentially comes to this: there is no telling what "present answer" agrees with my past usage because all the facts which we have to go on here leave us unable to determine the exact nature of our past usage, or the exact nature of our previous linguistic intentions. As such, there is no telling what conforms and what doesn't.

I think that one mistake of commentators has been a failure to ask whether anything really hangs on our not being able to determine whether some particular present answer ('125') agrees with past usage. This failure is no doubt due to the fact that commentators have been more concerned with challenging the sceptic's case against our being able to determine which of several answers conforms with past usage. If someone claims that there are no facts which show whether I meant plus or quus in the past and consequently, no facts which show what I should presently say in order to agree with my past usage (or, as Kripke likes to put it, no facts that compel me to answer one way rather than another), our first and most natural response is to prove such a person wrong. I am suggesting that we ask why there need be facts of the sort denied by the sceptic. I don't see that there need be. If I am correct then there will be no reason to take the sceptic seriously, for even if we grant him the nonexistence of "meaning facts", he cannot get to his pet conclusion, viz., that the concepts of meaning and intending one function rather than another make no sense.

In order to bring sanity back to the discussion, we need to ask the sceptic: Why must I determine my past meaning in order to be "compelled" to answer one way rather than another, or in order to be justified to answer one way rather than another? By my lights, the proper answer to this question is: I don't. I contend the sceptic is guilty of simply assuming that past meaning must be determined in order for us to be justified to answer one way rather than another. I also contend that there is a good reason the sceptic has simply assumed this, viz., no sound argument can be given for it since it's false. The truth is that the sceptical challenge or problem is little more than a colossal red herring. The justification of one answer rather than another never is, and most importantly, need not be, something that involves an appeal to past usage or past meaning or past instructions.

Surely we don't want to say that any time someone presently conforms to past usage that s/he thereby has used language correctly. For example, if I had used 'hoi polloi' to designate the rich and powerful, or had intended to so use it, it would be a bad thing if my present use agreed with my past use or my past intentions, since 'hoi polloi' refers to the masses, not the rich and powerful. Upon discovering my error, I would want to make sure that my present use of 'hoi polloi' NOT AGREE with my past usage. As such, the correctness, legitimacy, or justification of a "present" response can be seen not to be tied to having my present response agree with my past usage.

By the same token, if my present use of a sign is taken to be at odds with my past usage, it doesn't mean that I am thereby using a term incorrectly or saying something unjustifiable. Again, the 'hoi polloi' case reveals this. If I presently use 'hoi polloi' to refer to the masses, where before I had used it to refer to the rich and famous, then although my present use fails to agree with my past usage, I am now using 'hoi polloi' correctly, justifiably. It is clear then that there is no connection at all between correct (or incorrect) usage of signs and agreement (or disagreement) between past and present usage. As such, our alleged inability to determine whether it is '125' or '5' which agrees with out past usage does nothing to threaten the legitimacy of either answer.

Another way to see that the sceptic is wrong to link our alleged inability to determine whether '125' or '5' agrees with our past usage with our inability to justify answering '125' rather than '5', is as follows. I can justify the answer '125' by claiming that I am using '+' to denote plus and whether this use of '+' agrees with my past use is, it seems, by the way. For meaning such and such presently is not a matter of agreeing with past usage or meaning. Nor is it the case that the justified answer is the answer that agrees with past usage. We justify one answer over another by providing an explanation. Typically, the explanation appeals to a function. If the function in question compels its users to the answer given in a particular case, then the answer is justified, regardless of past usage.

To be sure, the sceptic can insist on identifying, (i) being justified to say '125' rather than '5', and (ii) showing that '125', rather than '5' agrees with past usage. But for the reasons given above, I reject the identification of (i) and (ii). And if we reject this identification, we can rightfully reject the idea that the sceptic has undermined our ability to justify saying '125' rather than '5'.

This leaves an interesting question, viz., what now of the sceptic's case against our ability to determine whether we meant plus or quus in the past? Does the fact that past usage is irrelevant to questions about justifying present answers block the sceptic's case for our inability to tell which of plus or quus we meant in the past? I am not sure. Much depends here on how we understand the sceptic's assumptions about the relationship between meaning and past usage. For example, if the sceptic supposes that saying that someone means plus rather than quus requires, BY APPEAL TO PAST USAGE ALONE, something which justifies his/her saying '125' rather than '5', then my argument so far does nothing to challenge the sceptic's claim that we have no basis for saying that someone meant plus rather than quus in the past. On the other hand, if the sceptic supposes that saying that someone means plus rather than quus requires only that s/he be able to give a "functional explanation" for saying '125' rather than '5', then clearly my argument does block the sceptic's case for concluding that there is no telling whether someone meant plus rather than quus in the past.

The bottom line here is that the sceptic misleads us by making us think that something important hangs on determining whether we meant plus or quus in the past. In particular, the sceptic holds that we must be able to determine which of plus or quus we meant in the past in order to be able to determine the "present answer" which agrees with our past usage. And we must be able to determine this, according to the sceptic, in order to properly justify our "present answer". For without it, according to the sceptic, any answer we give is no better than, indeed, it is nothing more than, "an unjustified stab in the dark" (K, p. 17), no better than a "mere jack-in-the-box unjustified and arbitrary response" (K, p. 23). But none of this follows, since the sceptic is wrong to tie the justification of our present answer to the determination of agreement between it and our past usage. Once we appreciate this, we will no longer attempt to "refute" the sceptic by showing that there are facts that show which of plus or quus we meant in the past. That is, we will no longer take the sceptical problem seriously.

Top of Page | WRPL Ultimate HomePage