Transcendental Idealism, Part II

by Jasper Gilley

I recently stumbled across a Twitter thread centered around a controversy surrounding a famous actor. This actor was accused by some of an ideological crime particularly heinous to the liberal Western world (by which I mean virtually everybody – in the US, at least.) This Twitter thread contained a myriad of vehement responses alternately indicting and defending said actor. The utter confidence with which these internet people were debating controversy made me realize something – they weren’t talking about the same thing at all.

Last summer, I wrote about transcendental idealism, which I interpret as the recognition that humans must always perceive the universe through the lens of their brain (not directly.) I still stand by what was written in that post, but I think it actually didn’t sufficiently generalize the implications of transcendental idealism (obviously, it’s going to be pretty hard to fully generalize such a macro-scale idea, but I’ll try nonetheless.)

I think that there is something like a universal human experience – for all we know, it’s 100% identical for everyone – but this is impossible to verify because of the inherent limitations of language. That is, language (especially written language) is a crude mapping to mental representations, which is all that actually exists (see also: cogito ergo sum.) This phenomenon is especially pronounced on Twitter, because written language is all Twitter has. Since the same written words mean subtly (or potentially not subtly) different things for different people, it can appear to someone unwilling to dissect things deeply that some stranger has outrageous ideas, when in fact that stranger has the same ideas you do, just mapped differently from mental representations to language.

Perhaps a visual representation would be more intuitive. 10-year-old me derived great quantities of interest from the notion that for the same object flipped over itself, each object’s respective left and right sides would be reversed. Like so:

In similar fashion, I’m postulating that there’s effectively a fixed spectrum of experiences/mental representations that humans can have. It’s probably not discrete – like the above left/right arrows – but, like the above graph, different experience spectra (and their associated word mappings) are inevitably oriented in different ways. (This is the possibly the mistake that the hippies mentioned in Transcendental Idealism, Part I made: failing to acknowledge the disparate orientations of experience spectra.) If experience spectra are oriented in complete juxtaposition, they could be represented graphically as:

An example of such juxtaposition: it’s probably a truism that everyone – literally everyone who has ever existed – has at almost all times in their life had a gnawing hunger for something they do not have which (they think) will make them genuinely, permanently happy should they obtain it. (This is just a psycho-evolutionary quirk about which it’s very easy to be in denial.) Naturally, different people will have and not have different things, so the happily-married, poor person will think that finally having financial stability will make them permanently happy, whereas the single, wealthy person will think that finally finding love will make them permanently happy. Critically, there’s probably no difference to the way each desire is represented in each person’s mind, but of course language is going to be utterly hopeless at helping them realize this. (This is, I believe, much of the empirical justification for Buddhism, and a large part of what makes Breaking Bad compelling.)

When you’re living in a tribe of 100 hunter-gatherers, or even an agricultural village of 1000, the limited ability of language to merge disparate experience spectrum orientations is probably largely masked, due to the high-bandwidth (potentially non-lingual) communication that such environments inevitably foster. It probably also doesn’t hurt that experience spectra for different people wouldn’t be too disparate in these environments. As a result, the same words would mean virtually the same thing for different people. However, on Twitter (such environments’ polar opposite by the metric of communication bandwidth), it would probably take a back-and-forth of many replies to have any hope at merging genuinely different experience spectra and their corresponding words. Understandably, this isn’t going to happen very frequently.

To add to this, there are subjects on which there’s a strong incentive not to attempt any synchronization of words and representations. Take, for example, another Twitter debacle I was recently witness to: a “conversation” surrounding “thin privilege” – the fat/skinny analog of white privilege. Naturally, any attempt to discuss this in a forum such as Twitter is going to be dead on arrival. There’s enough of an equilibrium built up around discussions of white privilege such that people are only going to hear thoughts on it – or lack thereof – from people for whom the mapping of representations to words is virtually identical. Not so with thin privilege, apparently. As a result, you’ll get several different groups coming to the discussion with several different mappings of representations to the word privilege: progressives for whom it’s an in-group slur, genuinely marginalized people for whom it’s nothing more than a voicing of shitty life experiences, and conservatives by whom its very usage is perceived as an assault on their way of life.

Clearly, this isn’t a problem that just Twitter has – any time you’re attempting to communicate in a digital forum with low information bandwidth, it’s going to be very difficult to merge word meanings.

Lately, I’ve come increasingly to the conclusion that the human brain (and its associated communication methods) is the last Terran frontier – that is, the final problem we can solve as a species that lives trapped on a small speck of unusually sophisticated chemicals. It’s understandable that we’d bump up against the limits of language (written or spoken) as a communication vector sooner or later, of course. After all, spoken language (by which more information can be conveyed) is really just a dog bark that we’ve hacked to carry several orders of magnitude more information. I think the limitations of language become a lot more intuitive when you think about how/why it developed. If you’re two cavemen trying to build language for the first time, the way to go about it is to point at various objects (rock, tree, fire, etc.) and agree on custom, distinguishable grunts for each of them. Voilà, you have a language. But this only works precisely because your experience of rocks, trees, and fire is the same! Still, you’re nonetheless mapping words to the external perceptions of rocks, trees, and fire, not the internal mental representations of them. This is related to another age-old question known to fascinate 12-year-olds: if Alice perceives the color blue as Bob perceives the color red, and vice versa, the two have absolutely no way of figuring this out, because there’s absolutely no mapping from the experience of the color blue to a word, just a mapping from the external perception of blue to a word (that is, blue.) The problems start to arise when you leave the realm of things for which there’s absolutely no lingual representation and enter the realm of things for which the lingual representation is fuzzy, or inconsistent.

Once brain-computer interfaces are a thing, this problem will (presumably) vanish overnight. However, until then, we’re stuck in a shitty situation where large numbers of people dedicate significant quantities of time to rebutting views that they erroneously think others have. Worse yet, my suspicion is that the range of possible ways to nefariously harness this power has yet to be fully explored. Elon Musk, Neuralink, and co. had better come through with another humanity-saving invention.