The Vexing Evanescence of Verisimilitude

Alternatively: why Vulcans are the perfect metaphor for military-grade autism.

I. How does it work? And why does it even matter?

I want to briefly recap: The Descartography of Mythopoesis: The Simulation Theory of Stories.

I started my original inquiry with a question: “what is a story?” It seems like a question with a very simple and obvious answer: It’s a sequence of events. But this answer never felt satisfying to me. It’s one thing to ask for a mere dictionary definition, and another to probe its platonic essence. I couldn’t shake this feeling that, had I genuinely felt like I understood the essence of stories, I’d be able to answer the corollories (at least at an abstract level): A) why are stories so interesting; and B) what distinguishes a good story from a bad story? For a while, I had no answers. For one thing, stories should not be so interesting prima facie. And the other thing, is that if stories are just sequences of events, then why are some stories so much better than others?

I’m reiterating this because I want to emphasize a distinction between definitions and models. A mere definition (e.g. stories are sequences of events) is not conducive to understanding a phenomenon at a deep level. All stores are “just” sequences of events. Yet somehow, some stories are more storylike than others.

I believe a similar thing is going on with the concept of Truth. People can define it to some extent, if asked. And people use the concept regularly in everyday life. But somehow, a deep understanding seems elusive. I do believe that The Binary Classification Theory of Knowledge is correct, but also incomplete. So what is truth, really?

I have a solution, and it has some ramifications some might find counterintuitive.

II. true vs THE TRUTH

One oddity that I’ve noticed is that there appears to be some distinction between asking if something is merely “true” (in the sense of truth-functional logic), vs demanding “the truth, the whole truth, and nothing but the truth” (as in sworn testimony).

In the first case, “true” is being used merely in the same way as described in The Binary Classification Theory. I.e. as asking if the map corresponds to the territory. It’s merely an inoccuous question of accuracy. Notice that “map” and territory” constitute two objects.

But in the second case, “THE TRUTH” seemingly takes on cosmic, quasi-religious significance. It’s spoken of with gravitas and solemnity. Why does “THE TRUTH” take on special significance when the general concept of truth is deployed in this way? Notice that in this case, there’s at least three objects: the territory, the accurate map, and the deceptive map.

III. Map Hacks

I tried Starcraft2, once. The game is very hard, and very stressful. I never climbed very high, and quit after a while. But I stuck with it long enough to learn some more general lessons from it.

One lesson is the importance of vision. At the start of a match, each of the two players starts in opposite corners of the map. And the map is covered in a fog of war. Army units and buildings, however, grant vision within a certain radius. It’s considered standard practice to “scout” the opponent periodically, by sending a single, highly mobile unit to get vision of what the opponent is up to. At certain timestamps, players choose to cut corners and not scout, as they think they’ll get a very minor advantage. This is also a risk.

“But why is it a risk? How do I calculate the value of getting vision, so I can know when to cut corners? How exactly does scouting help win the game?” I’d ask myself.

“Well fance, if you don’t scout, you risk getting rushed or cheesed. And even if they’re not rushing or cheesing, scouting is also useful for determining what the opponent’s army composition is, so you can counter it effectively. It’s also useful for checking how greedy they’re being, so that maybe you can attack them while they’re relatively vulnerable.”

Which is all well and good, but these are quite specific to SC2. I was also interested in a general theory of what scouting was, in some deep platonic sense. Many players in the community just copy what the best players do, but without actually understanding the purpose of their actions, which leads to players going through the motions but not actually achieving the desired effect. I, for one, didn’t want to just blindly ape the best players. I wanted to understand.

And after some contemplation, I’d concluded that scouting was mostly a matter of resource-aqcuisition and threat-detection. And I figure this is also why vision evolved biologically. In other words, vision is useful because it helps organisims find food, mates, tools, and helps avoid predators or rivals (in other words, adversaries).

The reason why THE TRUTH takes on special significance, is because it’s your North Star and guiding light amidst a hostile, adversarial environment. An adversary will hide, camoflauge, or otherwise deceive. The adversary will try to make truth-seeking maximally difficult. They will do this in order to sabotage your decisions. In sum, THE TRUTH takes on special significance when truth is scarce, and the stakes are high.

IV. Epistemic Import

There’s three reasons why framing vision in terms of resource-acquisition and threat-detection is important.

1. Truth is necessarily relevant.
   
   Naturally, the concept of vision and truth are tightly related. Humans rely quite heavily on vision, which allows us to draw accurate correspondences between our mind and reality. But what our meditation of SC2 shows is that mere correspondence, per se, is only half of what makes something “truthy”. The other half is relevance.
   
   Many people seem to have this idea that the only measure of truth is accuracy. But that’s not actually the case! E.g. suppose a judge is presiding over a homicide case. The judge suspects the witness is lying about the defendent's testimony, and demands "the Truth". The witness responds that "for every 60 seconds, a minute passes in Africa". If we were to ask the analytic philosphers, the defendent's proposition could reasonably be construed as "true". It accurately describes reality, does it not? And yet... somehow it's not what the judge was asking for. The judge will naturally be upset, because it’s expected that the witness is will say something relevant to the homicide.

2. Truth is necessarily instrumental.

   
   Many people have this idea that THE TRUTH is the most important thing in the world. Or very close to it. E.g. when you give someone a dilemma between “choosing to do a good deed, which can only be accomplished by lying” vs “choosing to be honest, but with negative consequences”, the latter is often seen as the more ethical path. To which I say… yes and no. I’ll return to this dilemma in a moment. For now, I just want to establish in broad strokes that The Truth is the servant of The Good, not that The Good is the servant of The Truth.

3. Truth is necessarily robust.

   
   I.e. the adversarial nature of lying implies when you find THE TRUTH, you can stop looking. But until you find it, you must keep looking for cracks in the false narrative. But how does one know when one has actually found THE TRUTH?

IV. The Roads Not Taken

In Magic Runes and Sand Dunes, I claimed that truth is a correspondence between a mental model and reality. In Compression and Comprehension, I claimed that reality amounts to perspectival invariance. Therefore, “finding the truth” is often not so much discovered, as it is approximated. As per Karl Popper. But vexingly, the approximation will never be perfect. Because it’s always possible that some wider perspective will show that the phenomenon was, in some way, illusory after all. No matter how large you expand the scope of your perspective, it’s always possible that a wider perspective will prove that you’re in The Matrix, and everything you know was a lie. Nonetheless, it’s often useful to expand one’s perspective as wide as possible. This is especially true of an adversarial environment. Because the bigger the lie, the more energy is expended for the illusion to hold within the larger frame of reference.

What this tells us about truth-seeking, is that: A) models must be forged in the crucible of experimentation; and B) truth-seeking isn’t so much seeking the truth, as it is exploring branching worldlines for nasty surprises. Models which give off signals that don’t align with the worldlines are like impurities, and they must be removed.

V. The Ends and the Means

Many people have this idea that THE TRUTH is the most important thing in the world. Or very close to it. E.g. when you give someone a dilemma between “choosing to do a good deed, which can only be accomplished by lying” vs “choosing to be honest, but with negative consequences”, the latter is often seen as the more ethical path. To which I say… yes and no. I’ll return to this dilemma later. For now, I just want to establish in broad strokes that The Truth is the servant of The Good, not that The Good is the servant of The Truth.

Hot take: truth is not the most important thing in the world. It only seems this way in adversarial environments. Most of the time, people are willing to lie to themselves with convenient fictions. Many of which were detailed in Game Theory: The Force that Binds Us. E.g. social status is a convenient fiction, fiat money is a convenient fiction, etc. They have little basis in objective reality, but people often pretend like they do, because it’s useful for coordinating behavior. If you point this out, people will maybe nod, and then continue behaving the way they’ve always behaved. Because the fictions are ultimately in their interest to pursue.

Falsehood becomes adversarial when person A tries to covertly manipulate person B’s decisions (especially in ways which reduce person B’s agency or otherwise countervene their interests). Furthermore, truth is not free. Truth-seeking can be expensive and time-consuming. So at some point, we just have to accept operating in noisy environments.