On “The Great Decoupling” of Consciousness from Intelligence

THIS IS A PARTIAL CRITIQUE OF PROFESSOR YUVAL NOAH HARARI'S BOOK, "HOMO DEUS" (2017, HARPER'S PUBLISHING). 

 -In his chapter on “The Great Decoupling” of consciousness from intelligence, Harari posits that with the rise of unprecedentedly intelligent algorithms, 'intelligence' as a faculty is no longer intrinsically tied to consciousness or subjective experience. As computers, these algorithms are not at all conscious or experiencing subjectively; they are 'mindless,' and thus require a new and novel description: “mindless intelligence.”

-As a result of the rise of mindless intelligence, the competition between humans and machines comes to regard not only physical abilities and capacities, but psychological abilities and capacities as well. This means that the paradigm of mass automation taking human jobs has, perhaps, reached a point where machines will not be employed simply for the automation of physical work, but also for jobs which until recently would have required a certain cognitive acuity and depth possessed only by humans. As such, Harari asks: “As long as machines competed with us merely in physical abilities, you could always find cognitive tasks that humans do better. So machines took over purely manual jobs, while humans focused on jobs requiring at least some cognitive skills. Yet what will happen once algorithms outperform us in remembering, analysing and recognising patterns?”

-Earlier in “Homo Deus,” Harari points out that science typically understands consciousness (insofar as it does understand it) as akin to a complex set of algorithms as our artificial intelligence is. Winding the clock back to the 19^th century Industrial Revolution, he points out how the general scientific / academic consensus of that time understood consciousness through the comparable analogy of the steam engine. Sigmund Freud's work is a good example of this insofar as he writes of such tactics as a military trying to curb and exploit the 'sexual drives' of individual soldiers by sexually frustrating them, thus building a pressure that is strategically released and utilized for combat. In other words, like a steam engine, pressure is strategically and mechanistically applied and released to propel the engine forward, hence the saying that one needs to “let off some steam.”

-Why were steam engines the analogical focus of the time? Harari says, “[b]ecause that was the leading technology of the day, which powered trains, ships and factories, so when humans tried to explain life, they assumed it must work according to analogous principles. Mind and body are made of pipes, cylinders, valves and pistons that build and release pressure, thereby producing movements and actions.” The same applies to our use of algorithms as the dominant analogical device for consciousness; because algorithms are “the leading technology of the day,” we work our studies of brain, mind, and body quite strictly from said analogy.

-It is my assertion that, though the analogy is very valuable, it should be recognized for what it is: an invented analogical framework with limits intrinsic to how strictly we apply it as a comparative rule of thumb. This seems to get lost on Harari as he delves further into his writing and applies the analogy within strict parameters as if it is now the only analogy possibly relevant to understanding consciousness. Though consciousness does undoubtedly have manifestly algorithmic qualities, to take our theoretical understanding of intelligent algorithms as we have invented and discovered them and then turn to strictly¹ apply it to our studies of the human mind would ultimately be a case of making the tail wag the dog. Though Harari does undermine the strictness of the analogy as applied by himself and others by presenting the parallel between the 19^th century steam engine analogy and our present-day analogy of algorithms, it seems he still holds on to the latter too tightly himself, thus perhaps obfuscating certain insights (and endorsing such obfuscation as a rule) with an implicit doctrinalism.

-Further into “The Great Decoupling,” Harari says there may come a day when algorithms are essentially able to do everything we currently do, but much better. As a result of such an evolution, people will no longer be economically relevant, leading many to say this could be an opportunity for all those who have become productively disenfranchised to focus on creating art. However, employing his rather reductive tone, Harari then goes on to give examples of how algorithmic intelligence is able to compose incredible classical piano, orchestral, and symphonic pieces alongside haiku and other such poetry as well as or even better than average humans can. Thus, he asserts, humans will also become artistically irrelevant. The automation of classical composition, for example, has led to much disgust and fear on the part of traditional classical musical buffs who a) do not think a computer is capable of meaningful artistic composition because humans have a certain 'touch' that cannot be replicated, and b) likely feel threatened at the idea of being relegated to said artistic irrelevance.

-However, just because algorithms are capable of creating beautiful artistic works, so are humans. It seems impossible to me that humans will somehow thus be rendered valueless in their artistic endeavours as a result of artificial intelligence. Likewise, the idea that a computer can create art should not degrade its value to anyone for two interlinking reasons:

1. just as the name of the chapter implies, this 'Great Decoupling” of intelligence from consciousness means that, though the algorithms are irrefutably intelligent, they are not conscious and do not draw their artistic insights and abilities from experience as humans do, but instead from aggregated and processed data; though people may not be able to tell the difference unless explicitly stated, when aware of whether a computer or a human composed a piece, humans will understand and appreciate more highly the piece that is known to be the product of a combination of human intelligence and consciousness, as it roots from the expression of a subjectivity we can and must relate to as said subjective relatability is one of the prime reasons for the existence of art. Art as human self-expression for other humans will not disappear simply because algorithms are now capable of effective creativity. This is the reason people became offended when they discovered a beautiful piano piece that had truly moved them was actually composed by a computer: because now the rug of self-expressive relatability had been pulled out from under them whereas they were under the illusory understanding they were relating to something emanating from a like subjectivity at least as complex as their own, with similar life trials, tribulations, ecstasies, boredom's, and joys. When it was revealed to be the creation of an algorithm, the audience understood this for what it is: the result of the aggregate data-processing and self-learning of a mindless computer—of mindless, as opposed to a relatable mindful, intelligence.
2. And, in the algorithms defense as well as to contextualize, the compositions created by said algorithms are the result of aggregating, processing, and learning the techniques of innumerable human composers from throughout history. As such, it is still from the perch of unconscious intelligence, but as the old medieval saying goes, it is “standing [or, perching] on the shoulders of giants.” Human composers do this as well, but from a necessarily more partial position as regards the limited capacities of an individual human brain. Thus, human compositions will remain valuable even if algorithms are capable of much more technically impressive heights regardless of personal technical ability and prowess as subjective relatability is and will remain a central tenet of artistic expression and reception. However, the algorithm's music can also still be appreciated for what it is, as it obviously would retain a human beauty and relatability as a result of it ultimately being the product of nothing but. In other words, human creativity will be joined by human-generated algorithmic creativity and perhaps even fused with it in novel ways, but it will not be steamrolled out of existence.

1: In some ways, 'literally' applies in a format similar to tired monotheist dogmas, as in the singular 'mono' implying 'this and only this' analogy as it once implied 'this and only this' God or doctrine.