rule-folllowing
I came across the following argument today in a draft of a paper I read on how Plato might have responded to Kripke's version of Wittgenstein on rule-following (and this paper isn't, I believe, the first time I've come across this argument).
Proposition1: Any action a can be construed to be in accordance with rule r.
Conclusion: Words don't mean anything at all.
Perhaps this would make more sense if we added the following propositions:
P2: The meaning of words = their use in our language or form of life.
P3: The use of words is rule-governed.
(P3.5: If any action a can be construed to be in accordance with rule r, then any use of a word is as good as any other -- we can always construe that usage as having been used in accordance with the rule which allegedly governed it.)
First, the problem arises only with regard to the interpretation of rules. There must exist, Wittgenstein tells us, a way of following rules which doesn't involve interpreting them. I confess that I read Kripke's book years ago, and I don't recall whether or not Kripke addresses this -- but it seems the argument above overlooks it.
Second, I'm curious about P2, the proposition that the use of words is rule-governed. It strikes me that there are two kinds of rules.
1: Rules you formulate ahead of time, then ask someone to follow. "Start with a number, add 2 to it, then keep on adding 2 to whatever number you end up with," for instance. These are rules in a "strong" sense.
2: Rules that you formulate after-the-fact, which characterize or describe or quantify a certain already-existing phenomenal patterns. "Any object will fall to the ground when there's nothing below it, holding it up," for example.
I'd put the rules of language in the second category. Our use of words tends to cohere into certain patterns, but those patterns are always evolving. Any one of us can disrupt those patterns at any time, and that disruption -- if it's catchy enough -- could eventually become the "correct" pattern.
I could easibly be wrong about this, but I think that Kripke-Wittgenstein's "paradox" exists only with regard to the first kind of rule. Wittgenstien raises the issue when trying to determine when it is possible, to say with certainty, that someone is following a rule. Only with regard to the first kind of rule does this concern arise -- for only then does the question of "following the rule" even have relevance.
4 Comments:
In his remarks on the Golden Bough, Wittgenstein notes that he could invent a ritual but doubted that it would catch on. I think we should think a bit about whether it is easier to be truly disruptive than to be truly catchy. It is hard to disrupt the rule for counting from one to ten. Any effort is only likely to do so if we come up with a useful (i.e., amenable to usage) alternative.
My portable version of "Wittgenstein on Rule Following" reduces his argument to a regress that starts with the question "What is the rule for applying the rule?", i.e., for deciding when the rule is relevantly "in force".
Consider stopping at a red light when crossing the street. Suppose you do stop and suppose the light is red. Is your stopping accounted for by saying "When the light is red you don't cross the street." The first time you break this rule (at 4 in the morning on an empty street, for example) you show that there are contextual factors that determine either (a) when the rule should be applied or (b) what the rule means. But (b) actually amounts to importing the exceptions from the context into the content of the rule. P2 (meaning = use) is such an interesting Wittgensteinian discovery (people will argue about whether it is his) because it allows us to reformulate (a) and (b) to mean the same thing:
(a) when should the rule be applied = how the rule is used
(b) what the rule means = how the rule is used
What Wittgenstein is saying is that referring to the rule doesn't solve the problem.
We may as well just talk about how crossing signals are used directly. The "rule" becomes a sublime intermediary that distracts us from examining (ordinary) usage.
Thanks very much for your comments, Thomas.
I, too, doubt that any intentional disruption we'd come up with would catch - but my point is that sometimes new ways of using words (or new words) do catch on. Patterns shift over time, languages "evolve".
In part because of this, I'm inclined to make a distinction between "rules" like the rules of grammar which merely describe phenomenal patterns and rules which are prescriptive laws or commands. I guess you could talk about "descriptive rules" vs. "imperative rules". The distinction may not always be clear -- some rules may be a bit of both, depending on context But the idea that I'm toying with is that enough of a distinction can be made enough of the time to justify the claim that Wittgenstein's discussion on rule-following addresses more of the latter kind of rule than the first -- and thus his discussion can't be generalized to a discussion about language in general (and remains, therefore, a discussion about a certain kind of language game we're calling "rule-following".)
As for what Wittgenstein has to say about following rules, I'm absolutely with you -- thank you for your discussion.
I suppose my post was written out of frustration at having come across the claim (one too many times) that "Wittgenstein's discussion on rule-following attempts to show that words, language, etc., are absolutely meaningless." I tend to think of this as a strawman version of Wittgenstein, an oversimplification set up with the intention of being an easy target.
This post has been removed by a blog administrator.
(The irony of having typos in a comment like this is not funny enough...)
I agree with you about the Wittgensteinian straw man.
Is using a language like following a set of rules?
I think Wittgenstein's remarks here lead us to seeing how using a language is like following a set of rules, how using a word is like applying a rule. (A word's meaning is the rule for applying it.)
But Wittgenstein is careful to point out that this does not mean that an imperative can be articulated for each word which captures it's meaning.
There are some games whose imperatives are clear (on the background of a set of less clear imperatives about what a game is).
Consider the work of editor. On the basis of a descriptive account of the rules of usage he prescribes for a particular text, correcting its "errors". Now, the author may refuse to accept the changes the editor makes on the basis of some superior sense of what is possible with the particular word or phrase. The editor is still right to point out that it is "wrong" (or at least unusual).
I'm just thinking out loud here...
Post a Comment
<< Home