#hubert dreyfus

[memoriesofthingspast]

🌊🐋✨🌈🦊🐦‍🔥🌱🌻🌌

[passionthatmoves]

“[T]he choice to experience the world as sacred and meaningful–to do so by dint of effort and will–is a choice that is within our power to make. It is a choice that takes strength and courage and persistence, of course; perhaps it even takes a kind of heroism. But it is possible.”

Hubert Dreyfus & Sean Dorrance Kelly, All Things Shining

[sopstvena]

In a post-Enlightenment world we have tasked ourselves to identify what’s meaningful and what’s not, an exercise that can seem arbitrary and induce a creeping nihilism. “The Enlightenment’s metaphysical embrace of the autonomous individual leads not just to a boring life,” Dreyfus and Kelly worry; “it leads almost inevitably to a nearly unlivable one.”

Craftsmanship, Dreyfus and Kelly argue in their book’s conclusion, provides a key to reopening a sense of sacredness in a responsible manner.

The task of a craftsman, they conclude, “is not to generate meaning, but rather to cultivate in himself the skill of discerning the meanings that are already there.” This frees the craftsman of the nihilism of autonomous individualism, providing an ordered world of meaning. - Cal Newport, Deep Work

[perkwunos]

Then take the materialist turn. We deny Cartesian dualism by denying one of its terms. There is no “mental substance,” everything is matter, and thinking itself arises out of matter. This is the kind of position which Quine espoused, for instance. And yet Quine re-created a similar structure in the new metaphysical context. Our knowledge comes to us through “surface irritations,” the points in our receptors where the various stimuli from the environment impinge. It is these which are the basis of our knowledge. Alternatively, he sometimes takes the immediate description of what is impinging, observation sentences, as basic, and he sees the edifice of science as built under the requirement that shows how (most of) these hold. In either variant, there is a mediational, or “only through,” structure here. The proof of the indeterminacy of translation, of the uncertainty of reference, of the plurality of scientific accounts comes from considerations that the choice between different ontological or scientific postulates will always remain not fully determined by these basic starting points.

“Inner” is being given a materialist sense here, in this “naturalized epistemology.” Our knowledge of the external world comes in “through” the receptors, and so they define the boundary, only in a “scientific,” and not a “metaphysical,” way. Similarly, we see the Cartesian structure repeated in various conjectures about a brain in a vat, which might be fooled into thinking that it was really in an embodied agent in a world, as long as a fiendish scientist was giving it the right input. Just as the old epistemology worried that as long as the contents of our minds remain the same, some evil demon might be controlling the input so that the world could change without our being any the wiser, so contemporaries re-edit a structurally similar nightmare concerning the brain. This has become the material replacement of the mind, supposedly because it is what causally underlies thinking. The mediational structure, and the mediating interface of inputs (now controlled by the fiendish scientist), and hence a parallel “only through” claim, all survive the “materialist” transposition.

If one asks the proponent of the brain-in-the-vat hypothesis why he focuses on the brain, he will reply something to the effect that thinking “supervenes on” the brain. But how does he know this? How do we know that you don’t need more than the brain, maybe the brain and nervous system, or maybe even the whole organism, or (more likely) the whole organism in its environment, in order to get what we understand as perception and thinking? The answer is that no one knows. The brain-in-the-vat hypothesis only looks plausible because of the force of the mediational structure, our captivity in the picture implicit in modern epistemology, which requires something to play the role of “inside.”

Hubert Dreyfus and Charles Taylor, Retrieving Realism

[esoteric-codes]

Lessons From Early AI: On Cognition and Reasoning

“It is by no means obvious that in order to be intelligent human beings have solved or needed to solve the large data base problem” - Hubert Dreyfus

In the late 60s, AI had failed to meet many of the predictions made a decade before. There were no programs discovering new mathematical theorems, or playing chess at more than a dopey amateur level, or processing more than the most rudimentary natural language -- most importantly, it had not become self-learning. AI researchers remained enthusiastic, blaming the slow growth on hardware. Hubert Dreyfus, however, predicted that AI, as it was conceived then, had already hit its limits, that it was grounded in a flawed model of intelligence. 

According to Dreyfus, computer scientists had unwittingly adopted a Rationalistic view of the mind dating back to Plato. A key strategy of "Good Old fashioned AI” (GOFAI), as Dreyfus called it, was to build trees of information and use search algorithms to scan across the tree to collect information: a dog is a mammal, a mammal is an animal, etc. This mimics the concept of mind that Plato had put forth: the mind holding a map of the world as we understand it, which we use to rationalize and come to an understanding of a given situation. 

In Dreyfus’s "What Computers Can't Do" (1972), he encouraged AI researchers to consider other models of the mind, including Heidegger's concept of “thrownness.” When we want to hammer a nail, we don't recall data about hammers, analyzing their history or associated facts; instead, we pick up an object that we barely register in a linguistic form, thinking of it only in terms of its current utility (a "driver-of-nails"). This is the difference between cognition and reasoning. Likewise, when playing a game of chess, we don’t mentally run through the 20,000 possible outcomes from a given scenario. Instead, we focus on parts of the board that feel wrong, based on our experience of playing the game -- expertise lets us hone in on what’s important, rather than considering every possibility. While his critique was seen as ungenerous at the time, he was more of less proven correct: Connectionism, the statistically-based competing model of AI (then in early stages), now dominates the field.

One of the early successes of AI (in the GOFAI era) was Terry Winograd's SHRDLU, a program where one orders an mechanical arm to manipulate differently shaped blocks in an artificial space (all of this simulated through text). SHRDLU could understand what blocks you mean by what you were saying previously, despite the fact that many blocks are identical -- in other words, it could understand context. However, like many of the other early AI successes, it would not scale; adding new elements to that contained world quickly ran into a database too large to manage on hardware of the time.

By the 1980s, Winograd was one of the AI pioneers turning away from GOFAI, and he (along with fellow Stanford engineer Fernando Flores) wrote "Understanding Computers and Cognition.” A key point of Winograd and Flores is a lesson from the Speech Act Theory of Austin and Searle: language is not only (or perhaps even primarily) about the exchange of information; we speak for many reasons other than to trade data with other nodes. While this may sound obvious, it was not clear to computer scientists in the 1970s. 

The classic example from Austin is the performative utterance: we can can make a promise, name a ship, etc.: make something occur in the world, rather than state something with a truth value. While this seems friendly to code (we've looked at Speech Act Theory previously on this blog, in terms of the performativity of the text of code), it also pushes the "exchange of information" quality of text to a secondary attribute. Drawing from Habermas, he shows how most statements are not true or false but rather felicitous or misleading, depending on the shared context of speaker and listener in terms of culture, personal history, and other factors that are not so easy to represent or evaluate in AI. 

This problem of context is still relevant in AI: an “intelligent” personal assistant like Alexa can understand phrases in many different voices (a problem more easily solvable through statistical techniques) but is worse than a four-year-old child in understanding what it is we’re talking about. We have learned to speak in an absurdly specific (and often patronizing) way to get such a system to respond appropriately. If we could get Alexa to understand context better, it would perhaps overly humanize her, creating a creepy, a verbal uncanny valley.

So how is this relevant to esolangs?

As designers of programming languages, we’re building interfaces between person and machine at much more raw level than an Alexa. Here, we are even more squarely on the machine’s turf, translating our intent into discreet, logical steps. Looking at the missteps of designers working the other way, we can see the implicit assumptions about language -- how we use it, why we use it, where its ambiguities lie -- and it could give great material for exploring that chasm of understanding in languages that mediate between person and machine.

[protoslacker]

The third part, because it is the deepest and the most pioneering, is simultaneously the most difficult and the least developed-and, accordingly, I think, also the most rewarding to reconsider. Its principal theses are: that human intelligence is essentially embodied; that intelligent bodies are essentially situated (embedded in the world); and that the relevant situation (world) is essentially human. And these, I would like to argue, all come to the same thing: namely, to understand the possibility of intelligence is not to understand a property of some possibly isolable system, such as an “intellect”, or a “mind” (- intellect + affect), or even an “agent” (-intellect + affect + body). Rather, it is to understand a larger whole comprising a number of cultured, embodied individuals living together in an already meaningful world.

John Haugeland in Artificial Inetlligence 80 (1996) 119-128 (PDF). Body and world: a review of What Computers Still Can’t Do: A Critique of Artifcial Reason

What Computers Still Can't Do

[rachelle-makes-stuff]

Update on the science jerks embroidery; )

[passionthatmoves]

“The intense and meaningful world of Homer’s Greeks evidently shone with sacred force. Our technological world, by contrast seems impoverished and dull. We cannot return to Homer’s world, and we should not hope to do so. but we can become receptive to a modern pantheon of gods–the ways in which Gehrig and Federer shone, the ways in which Marilyn Monroe and Albert Einstein changed how we see the world in which we live. 

We can lure back the gods of old–the great works that were venerated once before and now can be re-experienced in their sacred worth. To do this requires more than simply canonising this works on reading lists  and classroom syllabi. It requires developing the skills for responding to the manifold sense of the sacred that still linger unappreciated at the margins of our disenchanted world.”

– Hubert Dreyfus & Sean Dorrance Kelly, All Things Shining