The futility of insisting on exactness
For @tim wood
This video appeared on reddit recently. It offers some thoughts about whether we can ever go beyond being pragmatic about defining words. Tell me what you think:
This video appeared on reddit recently. It offers some thoughts about whether we can ever go beyond being pragmatic about defining words. Tell me what you think:
Comments (36)
*In the nebulous social goal situation we call 'philosophy', semantic drift equates to what Wittgenstein called 'language on holiday'.
We imagine that the people around us are following the same rules we are, yet there could always be a situation just beyond the horizon that reveals that this isnt true.
[quote=Wikipedia]Similar skeptical reasoning can be applied to any word of any human language. The power of Kripke's example is that in mathematics the rules for the use of expressions appear to be defined clearly for an infinite number of cases. Kripke doesn't question the validity in mathematics of the '+' function, but rather the meta-linguistic usage of 'plus': what fact can we point to that shows that 'plus' refers to the mathematical function '+'.[/quote]
Why doesn't Kripke "question" the validity in mathematics of th the '+' function?
Is the mathematical function '+' so well-defined as to leave no room for doubt, in which case Kripke would be admitting a solution to Wittgenstein's paradox viz. that being able to state the rule is a good indication that a person is following that particular rule.
One argument against this "solution" is that each word sub-unit in a rule is itself subject to the same problem of conforming to multiple interpretations of a rule.
My response is that definitions, traced back to their origins, all start off with the obvious. Nobody will ever make an error in what "water" means or what "pain" means. We begin from the obvious, unquestionable meanings of such words and build our word-empire from there. Let's take the example of a chair, a favorite it seems.
A chair is, "obviously", more complex than its constituents. In the meaning of "chair" the only rule-following error that can be made is to mistake "chair" to mean a component or property of a chair. I mean we may mistake "chair" to mean wood or metal or other simpler constituent property of "chair". This type of error is impossible or highly improbable because the meanings of these simpler components of a chair is already understood - bridges already crossed.
:joke:
Even if this is true, the Empire is apt to be filled with fudge factor.
That doesnt help us to know by observation of people whether they understand words exactly the same way we do.
That's a good question. What fact would you point to to assure yourself that a meeting of minds has taken place?
Quoting tim wood
Correct focus? What's that and how does it work?
Science seems pretty useful. I guess you meant something else though?
This problem occurs in natural language, because natural language tries to establish correspondence with the real, physical world.
This problem does not occur in formal languages, because a formal-language sentence does not "mean" anything. It just acquires (or does not) its truth status from other formal-language sentences that already have such status. Example: S1 is true, S2 is true, S3 is false. What is the truth status of S1 and not S2 or S3?
It is a tremendous advantage that S1, S2, and S3 do not mean anything.
Natural language obviously has its applications but is often useless in situations where you should rather use a meaningless formal language. In that case, the solution is simply to strip away all meaning.
I’ve recently used the game tic-tac-toe as an example. The rules are clear enough, and limited enough, for us to learn without any disagreement about how to play and how to win. For young children there are possible misinterpretations and mistakes because they don’t have the capacity to see all the possibilities in their minds eye. The same goes for us and a ‘chair’. We lack the capacity to see all possibilities of chair because the ‘limit’ chair operates in is ... well, ‘limitless’ as far as we can see.
As we’re able to extend our thought further into time by the facility of ‘language’ we’re able to encapsulate set parameters for our understanding and expand our abstract capacities.
The fallacy is the overextension of rules beyond set parameters. The thing is we’re always, at least partially, inclined to do this. We also do reap the reward of the occasional happy accident of such over-extensions. A so-called ‘cognitive bias’ can prove fruitful in realms no person without such bias would even consider looking. The human ‘flaw’ is necessarily a ‘boon’ in some occasions.
Ok.
Quoting I like sushi
Yep.
?
Some people seem to be able to read people better than others, is it simply that they are very attentive to the clues that the other person gives off, or that in some way they are able to directly experience what the other experiences?
In uttering words we attempt to convey what we experience, to give rise to that experience in others. With some people sometimes it seems that a few words do the job, or even no word and just looking at each other is enough, while with some other people it seems that even after repeated clarification the essence of the experience doesn't get through to the other side.
We can try to be as clear as we can, but then we can't be clearer than that. Sometimes no amount of words or explanations can convey what a picture can convey or what hugging someone can convey. Sometimes people are in a state where they aren't ready to listen or to understand something because they feel in a specific way or because what they are told is in conflict with some deeply held belief.
Words are one tool to communicate, not the only one, so in the situations where that tool doesn't seem to work it seems indeed futile to insist on using it with exactness and hoping that if only we use it precisely enough it will start working.
One example I like to use is that of the dictionary. Even if we all used the exact same dictionary, with all the exact same definitions for each word, we would still have zero guarantee that we would understand one another, because there is a missing link between the words and what experiences the words refer to. A dictionary relates words with one another, not with actual experiences, feelings, perceptions.
Without that link it doesn't help to strive for more exactness by using the same definition of a word, and then the same definitions of the words that make up the definition of the word and so on and so forth, because eventually the original word is defined in terms of itself, circularly. What breaks the circularity is the link between words and experiences, which more precise definitions do not provide.
But we don't have an objective link, each of us has their own personal link, and words alone do not help to clarify these links, rather we need to interact with people in other ways in order to uncover what they experience and to communicate what we experience more precisely. Maybe exactness could be reached, but not through words.
I like philosophy and math because of present/claimed rigor. Exactness is great because it allows us to isolate a concept or theory in a sterile room of sorts for better analysis.
However...
I think the great Wittgenstein was aware that language functions a bit differently - it's not just exactitude and precision that it aims for. We have science, philosophy and math for such activities.
Language requires a certain amount of flexibility to minimize an exponential increase in the number of words we need to learn. Everyone knows that words usually have multiple meanings and that's for a very good reason - to prevent memory overload and wasting time over unnecessary hair splitting. I think it's this flexibility that leads to the Wittgenstein paradox.
I think you're right. Natural language is in use where experiences vary (as Leo pointed out). It would take time and special personalities to even uncover this.
Boring :(
An error in my post which I hope to "correct"...
It isn't about ambiguity as I thought. Actually it's about multiple disparate rules that concur ONLY in a particular set of instances. The effect being an inability to determine if a given number of people are actually following the same rule or not.
However it doesn't seem so bad because the relationship as in the example of "plus" and "quus" is that of containment. The "quus" rule lies within the "plus" rule. The concurrence between "quus" and "plus" is true in the specified range of numbers.
What I mean is there is no right vs wrong/true vs false in the matter of rules as far as Wittgenstein's paradox is concerned. It's more about degrees of correctness. In our example the "quus" rule isn't wrong/false which would be a real problem. Rather the "plus" rule is more correct than the "quus" rule.
In short there is no error and the difference in the rules is just a question of degrees of correctness.
There is no black and white. Simply shades of grey.
I'm very gratified when I find I'm not the only person on the planet who thinks x. But yea, retract claws, abort missile launch. What's for lunch?
Quoting TheMadFool
Which says that communication can't be rooted in rule-following, though we may be able to note what looks like rule-following.
Have you ever noticed that post-structuralists seem to isolate themselves in a domain of words and phrases? When you ask them what's beyond that realm, they say there is no escaping it. Which makes one wonder: is there really a bubble of words floating around in the universe somewhere?
Rather, human communication is a form of our overall interaction with the world. In the same way we look around asking questions and getting answers just with our senses, we look at each other and encounter something we take to be subjects just like ourselves. I agree that the way we think about ourselves is influenced by language, but I'm saying language itself is a form of sensing that applies only to psyches.
I think that the history of divinity might reveal that we started off thinking of the world as a giant psyche. Praying to the sun was the way we interacted with it. Somewhere in there is the notion that science as we know it is the offspring of religion in a deeper way than some people realize. What if speech is actually a primary way to sensing and interacting for humans?
I can conclude that communication never really happens, or that something accounts for the existence of communion that isnt obvious.
Except that one such intermediate shade is sufficient, usually, to facilitate perfectly reliable separation of black and white.
Not that such a separation is possible between the categories designated by "offside" and "not offside", which are assumed to exhaust the domain of football-play-states, and must therefore eventually overlap, however keen the effort and eyesight of referees.
Consideration of which is, I would think, a better use of the example than using it for discussion of Plus and Quus. The definition aspect of offside is an unproblematic equation of "offside" vs. "not offside" with "nearer to the opposing team's goal-line than are the ball and at least all but one of that team's players" vs. "not nearer... etc." Which, as you point out, isn't about induction or generalisation, and isn't what creates disputes.
And not that Plus and Quus aren't fascinating as anything, e.g. Blue and Grue. But they don't depend on necessary vagueness ('open texture') of terms. I.e. on the impossibility of keeping one term separate from another (such as its complement) without a margin for error, an impossibility that does explain the sporting disputes.
What I mean to say is that the problem may not be with language but with the users of language. Might we be mistaking ignorance/incomplete knowledge for a linguistic paradox?
In math there's a similar problem with induction. Suppose we're given only a three numbers in a sequence like so:
A = {1, 4, 8,...}
We're asked to find the next number in the sequence.
There are two possible rules given what we do know:
1) 1 + 3 = 4 and then 4 + 4 = 8
2) 2^0 = 1, 2^2 = 4, 2^3 = 8
As you can see the problem isn't with mathematics
As with the off-site rule we need some more more data points to complete our understanding of the rule being applied.
... or as many as you like. I think you're back on Plus/Quus.
Just saying, offside rule disputes may be a good example of a different (interesting) problem, but not this one.
I think we agree. It's not a problem with mathematics or language. It's that on reflection, we note that there's no way to verify that communication has the clarity and precision we assume it has.
More data wouldn't resolve that. It's similar to the problem of induction.
Quoting frank
I guess the example given is insufficient to capture the essence of Wittgenstein's paradox because both the off-side example and the plus-quus example are about acquiring more information.
Could you guys give me a better example? Thanks.
How would you know you have all the required info? What fact would you point to to show that you have it?
Presumably plus-quus isn't in doubt as a case of Kripgenstein, at least? But compare blue-grue if you want another (or very close).
I was questioning offside as a suitable example of (yes) generalising from limited (but presumed non-vague) information, and recommending it instead as a case of vagueness, specifically the ineliminability of vagueness in measurement with no margin of error. (I.e. a case of the futility of insisting on exactness, where the latter is assumed absolute rather than relative to a margin of error.)
The required info would be instances where the disagreement arises, thus revealing that different rules are/were in play.
Quoting bongo fury
I think vagueness requires a continuum to exist in. The classic heap paradox illustrates that quite well I believe. However Wittgenstein's paradox seem to be about clear and distinct rules. No continuum.
Well it illustrates the opposite, because it proceeds step by discrete step. But a continuum can make the vagueness ineliminable, yes.
Quoting TheMadFool
"No vagueness", and yes, my point exactly.
If the meaning of words were how they are used, and the the way we are taught the meaning of words is by using them to point to things in the world, not to point to things in our head, or our experiences, then when we use words we would be using them as we've seen others use them and how they taught us to use them. There would be no pointing to our experiences in using words, only pointing to the world outside of our experiences.
When someone says, "I have a stomach ache and feel like I'm going to vomit." How does one to learn how to use those words if we can't see what it is that they are pointing to when using them? I can't see their feelings. I can see their behavior, and it is that I would associate the meaning of the words, not something I can't observe, like their feelings. So when I use that same phrase, would I be pointing to my behavior, or my feelings? How would I know that a person is pointing to their feelings if I can't see their feelings?
If we say that those words point to their actions when using them, or the meaning of the words are how they are using them along with their actions, then I could say I have a stomach ache and hold my stomach as if I appear to be protecting my upset stomach, then I could do that without having the feeling of having a stomach ache, and I'd be wrong in how I'm using it, or I'd be lying.
It seems to me that we all have similar feelings or experiences and it is those similar experiences and feelings that we understand others as having an are pointing to when they speak. Meaning isn't use, or else there would be a lot of misuse going on because we can't always see what a person is pointing to when using them, as in the upset stomach example. Meaning is the relationship between some idea in your head and the words that others know that point to that shared feeling or experience.
If no disagreement arises, the parties would never know a gap in understanding exists.
I guess Liebniz's law of identity would apply here viz. that indistinguishability implies identity. In fact Wittgenstein's paradox admits that it's the impossibility of telling two rules apart from a fixed number of instances as the source of the paradox.
In my humble opinion the paradox requires a high degree of similarity between the rules involved but is absent when the rules under scrutiny are violently opposed in meaning i.e. there are absolutely no instances of concurrence between the rules because they contradict each other. I mean a total absence of agreements between rules would imply a clear division in the meanings of given rules. Therefore, the rules aren't so fundamentally different i.e. there are no obvious fatal contradictions in our thinking to make the paradox a real unmanageable problem for language.
Ambiguity and vagueness seem to part of the problem too but for now I'm still in the dark as to how exactly they weigh in. What do you guys think?
Oh gosh, is that my fault, for suggesting that offside is about vagueness, not induction? And you thought I meant induction is about vagueness, not the opposite? Or you thought it would be just a shame to examine induction without ambiguity and vagueness in the mix (even if that was good enough for Hume, Goodman and Kripke, and probably Wittgenstein)? And hey what about Leibniz's law too?? You want it all in the pot!!
You mad fool!
You can't convict Fred of murder just because the witness can't tell the difference between him and the perpetrator. Eye witnesses are known to give crap evidence.
:joke: :grin:
Well the claim seems to be that it's impossible to know whether people engaged in discourse are talking about the same thing. I agree that many, possibly infinite, rules or definitions may have enough similarities for one to pass of easily as any other in the list. However, there's got to be an underlying similarity between them that makes us not see the difference. For instance in the plus-quus example, subtraction can never be one of the other rules mistaken as plus or quus. There is clearly a limit to the confusion. That's to say, even admitting of vagueness or ambiguity or the problem of induction or whatever else, we can have [I]meaningful discourse[/i].
Also although some subjects involving inherently vague or ambiguous terms may be rendered difficult, if not impossible, by Wittgenstein's paradox, there are limits to the paradox which, in my opinion, provides enough room for a reasonable conversation, as this one we're having.
Quoting frank
My contention is that the necessity for some alignment between different rules/definitions so that they may both apply simultaneously as in the paradox, reveals that the problem isn't catastrophic either to language or logic.
It just means that if you want to locate the basis of communication in rule following, one of your challenges will be to explain telepathy.