I have started reading The Cultural Logic of Computation, by David Golumbia.
This, from Chapter 1:
«
Computers come with powerful belief systems that serve to obscure their real functions, even when we say we are acutely aware of the consequences of our technologies. The thought surrounding issues like global climate change and genetics (and, in an earlier time, research into atomic physics) suggests that technologies have strong inherent destructive potentials, even when we don't see them.
The fact that computers empower users is not in doubt; what is in question is what power it gives which users, how, and why. In a world where corporations already inhabit an ideal personhood (exported uniquely from an Anglo-American model) that obscures what we understand as human being (not least because the humans who inhabit them are rarely held accountable for a corporation's actions), and as with commercials for detrimental technologies like always-on wireless connectivity, it is all the more necessary to articulate the ideological operations of the computational tropes as they come into being, rather than afterwards.
We need to find a way to generate critical praxis even of what appears as an inarguable good. What historicist and poststructuralist writers like Foucault and technological skeptics like Ellul and Innis share is the view that social transformations emerge anywhere other than political movements, even when their overriding trope is technological. The lesson from that work that this book deploys is that we have to learn how to critique even that which helps us (much as computers help us to write books like this one, among many other things). It would be better not to have computers, in that sense, than to live in a world where many more people come to believe that computers by themselves can "save us," can "solve our problems."
»
p. 19
« Of course, from the beginning of computation as a practice and in fact as part of it, writers have proposed that language itself might be subsumed by formal systems, eradicating the ambiguity that so troubles human society.
As physical computers themselves came into being, scarcely a year has gone by when several corporate or governmental entities have failed to generate multiple press stories about computers that are about to speak--and only recently, in no small part because of the heated attention they receive, have we seen fewer claims that computers are about to start thinking. But it is a core commitment of this book that neither of these events is about to happen, soon if ever.
The reason is not because we and our thought and language are magical entities, beyond the science of computers; it is instead because we are material beings embedded in the physical and historical contexts of our ex-perience, and it turns out that what we refer to as "thought" and "language" and "self" emerge from those physical materialities.
Yes, we can easily build codes that are independent of our bodies; but we don't even know how to conceive of what we call speaking and what we call thinking as independent of our bodies and selves. We can't conceive that the destruction of identity that accompanies the "uploading self" fantasies of so much computer fiction has already and can always happen, because there is no self there to realize it or to feel it. Our selves can only stay where they are, in a singularity that has already happened--and is no nearer than it has ever been. »
(for context, this book was published in 2009)
p. 22, important distinction between analog machines and digital
« While human beings can surely engage in activities that resemble or are even equivalent to digital ones, it is their capacity to engage in analog activities-their propensity so far in history to engage most of the time in such activities-that are of signal concern in this study. Famously, Deleuze and Guattari write at length in the two volumes of Capitalism and Schizo-phrenia (I983, 1987) that much of human life and human society can be characterized in terms of machines; they go so far as to include much of everything we recognize as part of a "machinic phylum."
There is much to recommend this view, and it is not my purpose to put it under scrutiny here. But what is notable for our purposes is that these machines are generally analog: like most of our world, they are machines built for one or more specific functions, sometimes able to be repurposed for other uses, inexact, rough, fuzzy.
They don't choose between 1 and o to build up symbolic operations; rather, the machine of the animal elbow moves at any number of stretchable angles, which no part of the body needs to decompose into numeric approximations. While enough frames-per-second can make digital animations appear as smooth as analog ones, there is still a translation occurring inside the computer that the animal body does not need to make.
There is no mystery here; analog machines are at least as old as digital ones and pose no conceptual obstacles (that they might is arguably a symptom of exactly the computational mania with which this book is concerned). Lawn mowers, toasters, drills, typewriters, elbow joints, pianos, and jaws may be mechanical, but there is no reason to suspect them of being digital (Derrida [1993] offers an excellent account of the machinic qualities of the organic world that nevertheless remain different from digital representation). That digital media can approximate their function should raise this suspicion no more than the existence of baseball or golf simulations makes us suspect that these games are digital. »
p. 22 on virtual reality
« Few theorists addressed the interconnection of politics, culture, and technology more closely than Deleuze and Guattari. In A Thousand Plateaus, Deleuze and Guattari develop a concept, striation, that arguably emerges in part from the growing emphasis on computerization that was evident even in the 1970s, but that has sometimes been overlooked by media theorists in favor of what is clearly a misreading of Deleuze and Guattari's discussion of virtuality (see especially Lévy 2001).
The term virtual reality emerged in wide use (popularized in particular by the computer evangelist Jaron Lanier) after Deleuze and Guattari's pathbreaking work, and it is clear that Deleuze and Guattari intended the virtual to refer to a generic use of the term rather than to a computer-based phenomena (see De Landa 2002; Massumi 2002; Shields 2003; and Wark 2004 for more accurate discussions of what Deleuze and Guattari mean by the virtual and how it relates to the computer).
The idea that computers represent a better instantiation of «virtuality" than do the human brain or human society is a curious and curiously computer-centric notion, one that bespeaks the tremendous cultural power of computation itself. »
p. 26 on enslavement, and enslavement by
« It can be no coincidence that the computer emerges at just a moment when the public ideology of human enslavement has been changed by intense social effort. We address computers as our slaves, and never think of the power and satisfaction we feel precisely in knowing how perfectly the machine bends to our will. We exercise and intensify mastery over the machine at the individual and the social levels; we experience frustration when the real world fails to live up to the striated and rigid computational model.
Yet we continue to look to the computer for solutions to this problem, itself largely created and intensified by the computer. We don't see people who use computers extensively (modern Americans and others around the world) breaking out everywhere in new forms of democratic action that disrupt effectively the institutional power of capital (see Dahlberg and Siapera 2007, Jenkins and Thorburn 2003, and Simon, Corrales, and Wolfensberger 2002 for close analysis of some of the more radical claims about democriti-zation), yet our discourse says this is what computers bring. Our own society has displayed strong tendencies toward authoritarianism and perhaps even corporate fascism, two ideologies strongly associated with rationalism, and yet we continue to endorse even further tilts in the rationalist di-rection.
This book is written in the hope that this historical imbalance between rationalism and " anti-rationalism" has gone about as far as it can go it the rationalist direction. Perhaps, despite appearances, there is a possible future in which computers are more powerful, more widespread, cheaper, and easier to use and at the same time have much less influence over our lives and our thoughts. »
p. 41 on obsession with automating language
"The promise of a theory of language that could eliminate exactly what had troubled almost every student of language to that day namely, the inability to fully formalize linguistic practice, or to come up with fully automatic schemes for language-to-language translation- -spread through the intellectual community like wildfire. Notably, the thinkers who were most struck by these theories were almost exactly the same ones who were so possessed by computers: they were generally white, highly educated males, and rarely female or people of color. They were also, for the most part, not linguists."
p. 43 on erasing culture and disempowered communities
« To some linguists, the Chomskyan revolution represents the greatest disaster that had happened to the study of language in nearly two hundred years. Despite Chomsky's overt leftist politics, Chomsky's effect on linguistics was to take a field that had been especially aware of cultural difference and the political situations of disempowered groups and, in some ways, to simply dismiss out of hand the question of whether their practices might have much to offer intellectual investigation. One of Chomsky's earliest and most devoted followers, who went on to become one of his most ardent critics, Paul Postal, recalls that his early 1960s attempt to apply generative principles to Mohawk in his dissertation, which might have been thought especially interesting since it would demonstrate the applicability of generative grammar to a language that does not, on the surface, look much like English, was "considered slightly comic for some reason" by other generativists. In other words, Chomsky took one of the few actually leftward-leaning academic fields in U.S. culture and, arguably, swung it far to the right. By arguing strenuously that linguistic phenomena could be separable into form and content, essentially out of his own intuitions rather than any particular empirical demonstration, Chomsky fit linguistics into the rationalist tradition from which it had spent nearly a hundred years extricating itself. »
p. 43 on the bias toward English
« The early generativists did not merely display a bias toward English; they presumed (no doubt to some degree out of lack of exposure to other lan-guages) that its basic structures must reflect the important ones out of which the language organ operated. The arrogance and dismissal of gener-ativists toward other approaches, especially in the 196os, can hardly be over-stated. This was a scientific revolution; it was making clear and straight what until that point had been fuzzy and multivariate; it was cleaning up what had been dirty. »
p. 48 on iterability (important to understanding constraints of large language models)
« After all, there is little doubt that human languages can realize logical forms, or that some parts of linguistic practice appear logical on the sur. face: this is the observation that licenses almost all linguistic approaches that are interested in form. But human language can be used to work mathematical and overtly logical systems like the formal logic taught in contemporary philosophy classes, and even programming languages them. selves: the fact that language is capable of simulating these systems cannot be taken as strong evidence that language is such a system. The parts of language that escape formalization are well understood and have long been recognized as problems for pure and purely autonomous theories of form; these are exactly the phenomena that Derrida calls iterability, and the presence of idioms. Iterability in this narrow sense points to the fact that any linguistic object can apparently be repurposed for uses that diverge from what appear to be the syntactic compositional elements of a given ob-ject. Thus while the phrase "snow is white," so often invoked by logicians, appears to have a single, stable worldly denotation, in practice it can be used to mean an almost infinite number of other things, and in fact such a bald declarative statement would rarely, in human language practice, be used simply to express its apparent denotative meaning. Even obvious and repeated phrases like "hello" and "goodbye," via processes of iteration and citation, often bear much more meaning than they would seem to do from the syntactic view, and can be iterated for other purposes, if we can even say what "the meaning" of words and phrases are in any given context. »
p. 54 on equating brains with computers
« This moment of ferment gives birth not just to cognitive science as a discipline but to the doctrine that would come to be known as functionalism, which by the 1980s had become "the prevailing view in philosophy, psy-chology, and artificial intelligence . .. which emphasizes the analogies between the functioning of the human brain and the functioning of digital computers" (Searle 1984, 28).
Whether or not one finds this idea credible, it is notable that despite the general objections of some of the closest former workers in the field, by the 1970s the view that the brain itself must be something like a digital computer had become widely adopted throughout the academy. It is not even always clear what was meant by the comparison so much as that it had to be true, had to be that in some way the machine we had created was also a model of a more originary creation still in some ways beyond our under-standing. But it is still remarkable the degree to which philosophers in particular took hold of this idea and ran with it, and perhaps even more remarkable is the degree to which, just as Chomsky's ideas became a cultural lightning rod in linguistics -on my argument, exactly because of the political forces to which they were and are tied--their extension in philosophy perhaps even more clearly came to define the boundaries of the field, and in a sense, thought itself. »
p. 59 on functionalism
ONE OF THE MOST striking developments in the cultural politics of mid-to-late twentieth-century intellectual practice in the West, and particularly in the U.S., was the rise and at least partial fall of a philosophical doctrine known as functionalism. A term with application in nearly every academic discourse, functionalism has a specific meaning within contemporary analytic philosophy: as proposed by Hilary Putnam and subsequently adopted by other writers, functionalism is a "model of the mind" according to which «psychological states (believing that p,' "desiring that p,' 'consider-ing whether p,' etc.) are simply 'computational states of the brain. The proper way to think of the brain is as a digital computer." This is not simply a metaphor: "according to the version of functionalism that I originally proposed, mental states can be defined in terms of Turing machine states and loadings of the memory (the paper tape of the Turing machine)" (Putnam 1988, 73). Many of its advocates give this view the straightforward name
"the computer model of the mind" (e.g., Block 1990; Schank 1973). According to functionalism, the brain just is a digital computer, or something similar enough to one such that if we could discern its physical structure in sufficient detail, we would discover a binary mechanism, probably electrochemical in nature, that constitutes mental representations, exactly as a computer can be said to create representations.
Today, what we might call orthodox functionalism no longer holds sway in analytic philosophy, although its influence has also not vanished.
p. 60 on cognitive science
"Functionalism emerges, explicitly and in public, along with a new academic discipline called cognitive science, a discipline connected much more directly to computerization than is widely understood. Put most clearly: in the 1950s both the military and U.S. industry explicitly advocated a messianic understanding of computing, in which computation was the underlying matter of everything in the social world, and could therefore be brought under state-capitalist-military control-centralized, hierarchical control. The intellectuals who saw the promise of computational views did not understand that they were tapping into a vibrant cultural current, an ideological pathway that had at its end something we have never seen: computers that really could speak, write, and think like human beings, and therefore would provide governmental-commercial-military access to these operations for surveillance and control."