April 12, 2012

Laozi's and Analytic Types' Consanguinous Philosophy of Language

It's here.

I disagreed, however, with this assertion:
"The question we should ask is not, ‘What does the text literally and mathematically say’, as if interpreting the Daodejing is like adding numbers, but rather, ‘What does the text mean?’"
If we take that kind of rigor to the text, we don't lose much of the text to inconsistency, and we can very neatly outline what the referent of '道' must be when one reads it with such rigor.  In so doing, we better understand "what the text means."

But there's an element to "namelessness" that this article does not discuss, and that is in the ambiguity of "having a name."  My wager is that the Laozi is not raising a highly skeptical philosophy of language, since the text provides sufficient descriptions to isolate a clear referent for '道.'  I've generally assumed, as I thought other readers did, that "無名" in chapters 1, 32, 37, and 41 similarly refer being "nameless" in the sense of being "faceless," the way we English speakers might talk of "not making a name for ourselves."

My impression with the Laozi, more generally, is that it establishes a definition for the set U, and then attempts to extract further moral lessons from that amoral entity.  Laozi's Daoism, in this light, is just cargo cult set theory.

6 comments:

  1. But how can mathematical rigor be applied to text that lacks common objective meaning; isn't the Daodejing ambiguous natural language? This reminds me of Carnap's attempt to replace natural language with formal; I think you make a good point, just trying to get it.

    ReplyDelete
  2. wait, is your point that whether or not you cook up meanings for the text, you can define it? That's cool, if I get it!

    ReplyDelete
  3. Actually, mathematical rigor (i.e. formal consistency checking) can be applied to any collection of propositions, even entirely fictional ones.

    While words and sentences have infinitely many ambiguities, we have a number of methods by which we can narrow the possible meanings and acceptable readings to a finite few.

    For example, the sentence, "Two men have four eyes," doesn't tell us which picture to draw. How many eyes does each man have? Do you mean it literally, or idiomatically (that they wear glasses, that they were "real, manly men" and not just adult male humans)? Does any man have the eyes in his head, or are they in his pocket? Whose eyes do these men have? The explicit addition of details actually makes the problem worse, not better, because every bit of information brings its own ambiguities.

    (This, by the way, is exactly where logical atomism will fail miserably, and thus cannot be the correct representation of language or thought.)

    One easy filter for this and other issues is a consistency filter. It's probably illogical to call headless bodies men, so we probably wouldn't imagine that one of the men was headless.

    Another easy filter is an empirical proximation filter, wherein we would deliberately try to decode the sentence to match the world as we observed it. Thus, we would assume, generally, that each man had one eye in each of the two eye sockets in his normal human head.

    If we take this approach to the Daodejing, we can provide 道 with at least one consistent, existent referent, one for which everything that is said about 道 in the Daodejing is still true, assuming that his domain of discourse is "the real world" (the universe that we inhabit, etc.). I've never encountered a translation or commentary of the text which dared to provide a plainer referent to 道 than the one that present-day set theory and my reading of the text provides -- that 道 is just U -- so I'm currently committed to that view.

    ReplyDelete
  4. I still don't understand why applying some such filter to, say, "1:00 p.m. is in the afternoon" (P) is a formal method. We can conventionally narrow the acceptable readings of P to a finite few, but we are not thus narrowing the POSSIBLE meanings of it, right? If in some pragmatic sense we are, that sense seems inadequate and categorically different from the sense in which it's possible (and necessary) that the Godel sentence is true and unprovable.

    Take the question whether, and to what extent, a given meaning of 1 is consistent with 2:

    1: Two men have four eyes.
    2: Men in the observable world generally have one eye in each of two sockets.

    I should say that a 'literal and mathematical' analysis can be used to answer this apparently nonempirical question. Now, I know nothing about the Daodejing that I didn't glean from skimming Wikipedia, but I thought "how should we interpret 道?", or "how should we interpret 'two men have four eyes'?" were not similarly amenable.

    If they were, then whether either men was headless would be a matter of formal consistency (whatever that might mean). (If I'm not mistaken, this was the kind of claim Frege was trying to make, even if it committed him to a stupid ontology.) This is consistent with your (plausible) claim that since it's probably illogical to call headless bodies men, we probably wouldn't imagine one of the men headless.

    Good point about logical atomism.

    ReplyDelete
    Replies
    1. You bring up some interesting points, and I think there's a tidy way to resolve some of the issues.

      I don't have a fully laid out theory of meaning, but I side more deeply with pragmatists and empiricists generally than I do with others. In that sense, the pragmatic shift in meaning is the only kind which really can concern me, because I have to rely on observable behaviors and human use of language, not anything resembling "meaning" in some non-empirical sense.

      The issue at hand, with say, "Two men have four eyes," is that we can, in principle, illustrate entirely distinct states of affairs from limited conveyed information. Any potential picture that we can draw without logical inconsistency is, quite literally, a possible meaning, because it is something to which we could actually refer to something (since we can't refer to the logically absurd).

      That leads me to conclude that any referent, phrase, or sentence is meaningful if, and only if there is at least one coherent empirical situation that we can illustrate in our heads. Call it a consistency criterion of meaning.

      That means everything from Santa Claus to the n-trillionth digit of the Euler constant, which we could clarify or discover, but rarely feel compunction to clarify or discover, has a consistent referent, and so has a meaning.

      Back to this sentence: "Two men have four eyes." Out of this sentence come infinitely many variations, some so minute that we could scarcely describe them accurately (for instance, if you opened your right eye and closed your left eye, then alternately opened your left eye and closed your right eye -- different pictures). But we don't need the exact picture from any one speaker, but just a mutually recognized picture that we can use, one that's "good enough for government art," so to speak.

      That ought to help answer how we can measure the consistency of two sentences, say, "Men in the observable world generally have one eye in each of two sockets," with, "Two men have four eyes." Where we assume to look for a picture for which both statements are true, we find an intersect that offers fewer pictures than the sentences do individually, but still offers at least one.

      In this light, the formal and natural languages aren't so remote from each other as one might assume. The only difference is that logic allows a general method for dismissing certain proposed illustrations of a given utterance, by proposing a sort of illustration game in which certain pictures are classified by variances in their illustration (e.g. putting an x in the shaded region of a Venn Diagram would go in the "inconsistency" pile), and just like natural languages, we use these formal languages quite reliably.

      Notice, too, that this does not require some bloated ontology, but just human brains, comprehensible and patterned codes, and the assumption that human brains can imagine states of affairs that they don't directly observe.

      Delete
  5. sorry if I sent multiple times

    formal and natural languages aren't categorically different. Peirce could have believed in brains and signs, he is great!: 'the logical purport of any sign is the sum of its empirically measurable consequences.' So the two kinds of languages are just different kinds of semiotic syntax, natural languages having a syntax of use that can perhaps be given a naturalistic basis, assuming a basis of events is possible.

    ReplyDelete