Pragmatist Enlightenment
Actually written DECEMBER 21, 2004
Here I speculate about a phenomenon I call “Pragmatist Enlightenment”, a bizarre reality-changing experience unique to a certain rare group of people who embark on a fantastic philosophical “journey”.
To understand the Pragmatist Enlightenment, we must pay mind to how begins and progresses the journey toward it. And later, we shall also address how finishes the sojourner’s life after it.
The sojourner, whom I’d like to call the “Pragmatist Monk”, begins as an intellectually curious sort seeking justification for his knowledge. Having perhaps become aware of the hazards and harms of going off half-cocked and lording it over others with unfounded opinions, he’s got the urge to link his opinions to the facts of an objective reality. He therefore begins in the philosophical tradition of the foundationalists (or realists, “Objectivists”). “I shall discover the objective foundations for all my knowledge,” he hopes “and therewith find the mutual arbiter for justifying my knowledge among my fellows.”
With great rigor our monk tears into the philosophical texts, like a detective looking for the facts that reveal who, or more precisely what, is the ultimate grounding of real knowledge. But as he goes about this, a trickster begins fucking with him, a trickster that will ultimately turn him around and send him sliding the slippery slope away from foundationalism altogether (and toward anti-foundationalism, aka “pragmatism”). This trickster begins ever so slowly, luring him with the promise of full justification, but at some magical point uses his very craving for that justification to prove no such justification is possible. In a very real sense, the trickster is within him: it is his craving for full proof. And gradually our monk will discover, by way of direct experience, that the pursuit of full proof is a journey containing “the seeds of its own destruction” as it were.
It begins as he discovers that foundationalists have yet to prove that our minds grasp existence the way it really is. If, for example, he surveys the epistemological writings of Ayn Rand, he’ll learn about knowledge-foundational statements called “philosophical axioms” that are beyond proof, and must simply be accepted. And while it can be demonstrated that rejecting these axioms leads to logical contradictions, such demonstrations themselves assume that one already accepts the axioms.
“Seems like some leap of faith to me.” says our monk, unaware that he’ll ultimately end up in a state of mind so rampant with “faith” that accepting these axioms will look like hard science by contrast. But the point for him, at this stage, is that our grasp of the facts, founded in existence, cannot be proven. The point is driven even deeper as he discovers the philosophical significance of Godel’s incompleteness theorem, which states that within any formal system of symbolic statements (such as logical proofs expressed in natural languages), there will be expressions that cannot be proved by other statements. In other words, there will always be unprovable statements, unprovable axioms, in one’s proofs.
Pulled in two directions, he is. Accept some sort of axioms without proof, or resume the quest for a complete proof of all knowledge. Regarding himself as logically rigorous, our monk strengthens his resolve in favor of full proof. “Yes, I want my statements to fully prove a foundation for knowledge.” The key idea at this stage is his realization that proofs are always expressed in language. To demand a proof is to demand a rigorous use of language. Yet, per Godel, he realizes that language is a formal system that itself cannot prove that it refers to objects beyond language. In fact, he cannot prove that any words in a proof actually have meaning. “Damn! I can’t prove the words in my proof have meaning without assuming that the words of my proof have meaning. Ah shit!”
Again, to avoid a “leap of faith” our monk is forced to reconsider the meaning of “meaning” generally. “Fine, if I abandon the usual, yet unprovable ‘correspondence’ notion of meaning, then what kind of thing is a proof in the first place?” And he must conclude: “Well, a proof must simply be a formal, rule-based relation among words, a strictly linguistic affair. And if words have a property called ‘meaning’, such property must come from the rules of how words relate. Meaning comes from how the words are used – from syntax rules only.”
Here our monk is but one step closer to Pragmatist Enlightenment.
Although the enlightenment happens quickly, it is very complex, and must be broken down into simple stages to fully understand. So here are the stages:
STAGE 1
Our monk realizes that he has, up to this point, relied implicitly on the usual, yet unprovable “correspondence” notion of meaning, and that he has done so even in rejecting the correspondence theory. That is, he knows he has formulated the “rule-based theory of word meaning” using the “correspondence theory of word meaning”.
He knows, for example, that he expressed the “rule-based theory of meaning” using words. When he talked or wrote about the theory, he used words like “rule” and “meaning” and “language” and so on. But as he used these words, he had assumed they corresponded to things existing as mere parts of a larger realm beyond language. E.g. he used the word “language” in referring to language the same way he used the word “chair” to refer to a chair. Both things, language and the chair, were things that existed and had words within language that corresponded to them. When a word, being within language, referred to a thing outside of language, such as the chair, it corresponded. And the same could be understood for words that referred to things within language. When a word, being within language, referred to something that was also within language, it did so by corresponding. Broadly put, language (words) could refer to (by correspondence) anything that existed, including itself.
See, the real point of the correspondence theory of meaning, was not only that the words of language could refer to things in a realm beyond language, but that words could refer to anything that existed, including language. It’s misleading to interpret “beyond language” as meaning, “excluding language”. The realm “beyond language” is simply all that exists, the superset that includes both what is and is not language. The correspondence theory proposes that language refers to this “superset” called “existence”. Words (can) correspond to things in existence. And as words are existing things, they can refer to themselves. But the criteria for such correspondence is that the words exist.
Anyway, our monk now realizes that his theory of “rule-based” meaning has so far been expressed in words that he has assumed referred to language and meaning in precisely this sense of “corresponding to themselves” as members of this “existence superset”.
STAGE 2
Our monk realizes that his implicit reliance on the correspondence theory of meaning in STAGE 1 is sloppy, inconsistent, and illogical. For the sake of logical rigor, he knows he must justify his rule-based theory of meaning only by means of the rule-based theory. He must vanquish the last traces of the correspondence theory from his process. He must abandon any sense that there is an “existence superset” through which words can, as members, correspond to themselves.
To repeat: the rule-based theory of meaning must be meaningful (and true) only according to itself.
STAGE 3
Striving for this last bit of logical consistency, our monk begins to apply the “rule-based theory of meaning” to itself. He again reviews a few of the very words used to express the theory.
The word “meaning” is itself a word that has meaning. This being so, the word “meaning” gets its meaning from how we use it, the rules. Likewise, the word “rules” is a word that has meaning. This being so, the word “rules” gets its meaning from how we use it, the rules.
STAGE 4
Our monk discovers a problem when he tries to apply the “rule-based theory of meaning” to the word “rules”.
The word “rules”, remember, cannot correspond to rules that are members of some “existence superset” (that happens to include language). The word “rules” must instead mean whatever it is that the rules make necessary. And how then, are we to know whether the rules make necessary that the word “rules” refers to them?
STAGE 5
Our monk considers again that there must be only one answer to this. The answer is that the word “rules” does not refer to anything. There is no object to which the word “rules” can refer. Reference is correspondence. No proof can be obtained for a word referring to anything beyond itself for the same reason that no proof can be obtained for a word corresponding to anything beyond itself. Same thing.
Meaning is not reference. The word “rules” means the rules, but not because the word “rules” refers to the rules.
All well and good. This is just a reaffirmation of how far he must reject the correspondence theory.
STAGE 6
So again, having re-affirmed his rejection of the correspondence theory of meaning, our monk returns to what meaning is. And in a flash instant he gets it. He views himself as a speech machine whose speech is governed by the rules, a speech machine that does not know or understand what he is talking about, because his words do not talk about anything.
For him, the rules are not an object. The rules are not a thing. The rules are not what he had once naively understood to be the rules. When our monk had claimed that meaning comes from the “rules of language”, he did not refer to the “rules of language”. He was simply making sounds with his mouth in accordance with those very “rules of language”. He cannot talk about these rules, but they “rule” his speech nonetheless. He does not “talk about” anything. He is a speech machine that makes sounds according to rules. And when he uses words to claim he is such a machine, he has no idea what he’s talking about.
STAGE 7
But now our monk protests to himself. “Damn! But I do know what I’m talking about. I do know. That’s how it is that I came to this point. I got here by discovering the role of the rules of language, damn-it!”
STAGE 8
Yet again, he takes a look at himself from the speech-machine perspective. Like this: When he protests against the notion that he doesn’t know what he’s talking about, he’s just making more speech sounds according to some rules, sounds that refer to nothing.
STAGE 9
“Ridiculous!” He protests yet again. “How can I see myself as a speech machine if I can’t refer to (correspond to) myself as a ‘speech machine’?”
STAGE 10
He once again sees himself as a speech machine who just uttered speech sounds according to some rules.
STAGE 11
Having alternated between these irreconcilable perspectives on himself with perhaps a furry of increasing frequency, our monk gives up, mentally exhausted.
STAGE 12
Now he understands that he has taken his quest for logical rigor to the end. There is no logical rigor. Logical rigor destroys itself. There is only unprovable nonsense at the base of everyone’s opinions.
STAGE 13
Having bounced off this failing end of logic, our monk rejects the quest for logical rigor and adopts a more flexible attitude, ignoring or embracing contradictions.
This is the pragmatist enlightenment.
[This piece makes me want to write a piece that speculates: Pragmatism, such as Rorty’s epistemological behaviorism (and maybe Searle’s Chinese room) is the epistemic model for large language model AI like ChatGPT.
And after having wrote the previous sentence, I find my speculation is shared by Jordan Peretson when he said: “I would say that ChatGPT is the ultimate postmodernist, because the postmodernists believe that meaning was to be found only in the relationship between words {the Rules}.”* While Peterson opposes postmodernism, what he said thereafter didn’t so much challenge postmodernism as offer a different kind of Rules, rules that govern the relationship between, not words, but rules that govern the translational relationship between words and images and behavior feedback. It seemed like another kind of functionalist physicalism, reducing consciousness {understanding} to some mystical physical function, like Dennet’s “winning a competition”. To better understand what I’m getting at here, see this book’s {blog's} chapter 09 called “Physicalism”.]
* ChatGPT: The Dawn of Artificial Super-Intelligence | Brian Roemmele | EP 357 https://www.youtube.com/watch?v=S_E4t7tWHUY 24:35
Comments
Post a Comment