This whole thing falls down around the word "creative" and its synonyms.

Weatherby’s core claims, then, are that to understand generative AI, we need to accept that linguistic creativity can be completely distinct from intelligence, and also that text does not have to refer to the physical world; it is to some considerable extent its own thing. This all flows from Cultural Theory properly understood. Its original goal was, and should have remained, the understanding of language as a system, in something like the way that Jakobson and his colleagues outlined.

"Linguistic creativity" is used here in terms of words that are arranged in a novel fashion, not in terms of new ways to describe or say things. This is the basic argument against LLMs and always has been: they are engines of Markov processes whose entire purpose is to stochastically provide points on a well-characterized plane. LLMs are "linguistically creative" if language is dead and unchanging. If language was dead and unchanging, however, there would be no slang. There would be no jargon. There would be no teenagers embarrassed by their parents' fruitless attempts to speak in their register. There would be no cringe compilations of brands attempting to be hip. "Linguistic creativity" as understood by humans is "coming up with new ways to say things" while "linguistic creativity" as understood by the author is synonyms.

As Weatherby suggests, high era cultural theory was demonstrably right about the death of the author (or at least; the capacity of semiotic systems to produce written products independent of direct human intentionality). It just came to this conclusion a few decades earlier than it ideally should have.

Barthe argued that literary theory shouldn't contextualize literature. The argument is obscure now because human culture broadly disagrees.

Ultimately, the question society has always asked is "why are they saying that" while the Deconstructionists were all about "it doesn't matter." The whole core of their argument is that meaning can be derived solely from the writing while culture, in general, moved on. This essay is broadly arguing that since LLMs don't have the first fucking clue what they're talking about, they fit perfectly into Deconstructionist theory which, fine, but that doesn't necessarily make either Deconstructionism or LLMs useful.

LLM, then, should stand for “large literary machine.” LLMs prove a broad platform that literary theory has long held about language, that it is first generative and only second communicative and referential. This is what justifies the question of “form”—not individual forms or genres but the formal aspect of language itself—in these systems. Indeed, this is why literary theory is conjured by the LLM, which seems to isolate, capture, and generate from what has long been called the “literary” aspect of language, the quality that language has before it is turned to some external use.

That theory has been largely rejected, though, because people disagree that language doesn't derive from meaning. The fact that LLMs match a theory that most of the world has rejected is not a feather in the cap of LLMs.

Yet the extraordinary, uncanny thing about LLMs is that without any material grounding, recognizable language emerges from them.

BUT CONTEXT DOESN'T. And the world works on context.

Large language models, then, show that there is practical value to bringing the study of signs and statistical cybernetics together in a single intellectual framework. There has to be, since you can’t even begin to understand their workings without grasping both.

Holy shit it's like the symbolic AI camp that's been completely sidelined was right all along

A poem “is an intentional arrangement resulting from some action,” something knit together and realized from the background of potential poetry in language: the poem “unites poetry with an intention.” So yes, a language model can indeed (and can only) write poetry, but only a person can write a poem

So the argument is that only a person can write a poem but LLMs can write poetry; language doesn't need context but art does?

That book fucking sucks because it argues that there is no meaning. The meandering read is the point of the meandering read. The argument for LLMs is they'll do your homework for you. The argument against LLMs is the purpose of homework is to provide fucking understanding.

And perhaps for much longer. As I read Weatherby, he is suggesting that there isn’t any fundamental human essence to be eroded, and there cannot reasonably be. The machines whose gears we are trapped in don’t just include capitalism and bureaucracy, but (if I am reading Weatherby right), language and culture too. We can’t escape these systems via an understanding of what is human that is negatively defined in contrast to the systems that surround us.

Right - if you're not looking for meaning in the first place you never have to worry about finding it.

What we can do is to better map and understand these systems, and use new technologies to capture the ideologies that these systems generate, and perhaps to some limited extent, shape them. On the one hand, large language models can create ideologies that are likely more seamless and more natural seeming than the ideologies of the past. Sexy murder poetry and basically pleasant bureaucracy emerge from the same process, and may merge into becoming much the same thing. On the other, they can be used to study and understand how these ideologies are generated (see also).

So an LLM can't create a poem but it can create an ideology?

This is an exciting book. Figuring out the heat maps of poetics has visible practical application in ways that AGI speculation does not.

This is... a low bar.

This was a good long read. I'm going to have to come back to it. Thanks for sharing.