May 14, 2024
if you interact with people for long enough you begin to see the light dull behind their eyes; some topic you bring up activates a sleeper-cell activation code sending them hurtling down a well-etched path of thought which was not dug by themselves as they begin to parrot lines you’ve read before in various forms. maybe it’s their position on the minimum wage; maybe there’s something behind their eyes pulling the strings.
just kidding. if you genuinely believe that ideology has rotted out someone’s skull turning them into a puppet enslaved to some nebulous concept in the noosphere you may be failing one of the most basic cognitive tests. not to say that i don’t feel sympathy for this view – after the thousandth time running into the same dialogue in texas politics it begins to feel like the restaurant scene from being john malkovich. and if you take the view of egregores possessing those who disagree to its logical conclusion it is hard to avoid the thought that you and everyone you agree with may be colonized by an egregore adversarial to the one that you react to. nobody is conscious and all swept silt through noospherian currents by entities much larger than oneself. though there are much simpler explanations for the homogenization of a group’s view.
some beliefs are more stable than others; the reasons for their stability a web of other forces holding them in tension with other factors such that one may not point to a single action as to identify the reason why someone is a certain way. there may be risk factors – if one has read nyx land, their ideology could be guessed with decent reliability, but this obscures that gender accelerationism was already an stable attractor for that individual.
an opponent of this view may cite the rising/cresting/falling tide of “wokeness”, an idea which did not feature as prominently in the ideological realm twenty years ago. ignoring its similarities to the phenomenon of “political correctness”, this is a nonstarter. first, the parts of “le wokisme” already existed; some examples are the aforementioned political correctness, euphemistic treadmills, the wishy-washy leftism of the sixties that accomplished little, boomer environmentalism, etc. second, i am not a noosphere denier; rather i do not believe that tendrils extend from it into our world hijacking our nervous systems (usually). new ideas emerge and recombine in that realm; this is the natural course of events. third, everything is connected. the reasons why an ideology would rise to prominence have nothing to do with demonic possession and everything to do with the circumstances of one’s time. the reasons why capitalism may not be compelling to gen Z are that they haven’t been able to look at alternatives. prior generations had good examples of monarchy or state communism to look at and decide where they stand. with the fall of the ussr america has caught its own tail and now flails for a struggle to define itself by. this is not the only reason – nothing is monocausal and everything is related – but this is a major factor. others include the conditions in which gen z was raised, etc.
i avoid talking about llms because i frankly do not care much for them. i do not use them because i like how i write and like being able to see how my tone changes over time and these tools tend to obscure the little details. they can be useful: studying with them seems to work well if one is a computer science major (though their chemical engineering knowledge is lacking); their summarizations are good; and GPT-4V has a lot of use cases. however, i do not think that they are 10x multipliers for people and that the progression of their skills will level off as a sigmoid function. that being said i think that they are interesting for exploring the noosphere.
when one analyzes (whatever amount of input tokens which i will not look up), patterns inevitably emerge! this is an artifact of what jaynes discusses as metaphor in the origin of consciousness in the breakdown of the bicameral mind; language is naturally ill-suited for description, else, metaphor would not exist. thus, we substitute other words for what we mean, and words for those words, and so on, creating a chain of meaning. people don’t conceptualize this. however, gpt4 can map gpt2’s concepts! now, a claim that these neurons map 1:1 onto the noosphere would be silly, but they’re probably more accurate than chance.
i don’t really have a thesis here, i just think it’s nifty! a practical example supporting my positive view of wittig would be this neuron. interesting how the gender-neuron mainly cares about male issues and doesn’t mention women. its almost like women are primarily defined as not-men. there are also interesting ones, like wrestling being upstream of sexually-explicit language, supporting the hypothesis that wrestling is inherently homoerotic.
naturally some beliefs are connected to others. if you’re complaining about a new idea catching on and finding stable equilibrium in a demographic’s head, you probably cannot change the idea itself but rather its relation to other ones. or, introduce a new concept, though this one is more difficult. the demons in the brain? there are very real ways to exorcize them.