site banner
Jump in the discussion.

No email address required.

To give an overview of what I believe is a reasonable bounded-rationality basis to dismiss this objection:

  • I am not equipped to evaluate the claims in Alexandros's post in detail without significant effort and time investment (despite being a working academic in a quantitative field).

  • I'm not particularly worried about COVID and the societal excesses of the response seem to have already died down, so I personally don't see much value in learning about a surprising therapy for it. It seems unlikely to me that even if something like the contents of this post became widely accepted as truth, the societal response next time something COVID-shaped happens would be much beter.

  • Superficially, it seems there is no particular reason why something like Ivermectin (an antiparasitic that apparently works by disrupting the metabolism of fairly complex multicellular parasites) would work against COVID (a virus). I have a strong prior on most medicines claimed to have a minor beneficial effect on popular therapeutic targets actually being completely ineffectual (as this has been my experience).

  • On the other hand, the "parasite load" story seems superficially plausible.

  • Due to the culture-war dimension of Ivermectin, whose efficacy the red tribe in the US has entangled its social status with (no point in recounting the way this happened here), there is an obvious motivation for members of that tribe to produce compelling-looking arguments for its efficacy. Since Alexandros posts around this community, he seems a priori likely to harbour Red sympathies.

  • Moreover, there is a "contrarian" tribe that is motivated by taking down the rationality-orthogonal "trust the science" wing of the blue tribe, and therefore would also derive utility from successfully Eulering in favour of Ivermectin. Many people seem to talk about the abrasiveness of Alexandros's tone. This increases the probability that he's Red or Contrarian and would therefore have the motive to come to his conclusion.

In short, a situation that seems fairly symmetrical to "read this long and extremely compelling essay by a Harvard academic who is also a Twitter superstar using Science and Logic to prove that Blank Slatism is true". If you had unlimited time and resources or a particularly high stake in finding out whether desirable qualities of humans are genetic, sure, by all means you ought to read it and analyse the argument. For most everyone else, it would be more rational to ignore the essay, leave your prior largely unshifted and spend the time it would take to read on something with higher expected utility, like planning tomorrow's healthy breakfast or getting on top of your todo list.

Things that could convince me to take the essay more seriously:

  • Establish that the author does not stand to benefit from Ivermectin working, e.g. has impeccable blue tribe credentials.

  • Establish that rehabilitating Ivermectin would benefit me personally a great deal.

  • Propose a plausible mechanism by which Ivermectin (specifically!) might work against COVID. Some general handwaving like "it modulates the way the immune system operates" won't work; lots of drugs do that, so I don't see why specifically the one that the Blues are raging against and the Reds are swearing will prove once and for all they should actually be in charge should be the one that happens to modulate it just right.

  • Relatedly, but harder, shift my prior regarding medicines that purport to do anything more complex than targeting one particular well-understood metabolic pathway not working.

Bounded rationality is a real field of study, describing optimal behaviour for agents who can't actually reason and obtain information infinitely for free.

Even Yudkowsky concedes that rationality is about winning. It seems pretty straightforward to see that someone who is still busy calculating probabilities to see if some skub paper checks out while the police remove him from the premises as the debt collector wants to foreclose his home is not winning. As a corollary, if the gut feeling strategy consistently gets better outcomes than the "reason and logic" one, it's more rational.

As I said in a parallel comment, I am meaning to explain why people can rationally choose to leave these essays unread and not have their beliefs shifted by them much, to push back against the "rationalists are ignoring this high-quality argument, which proves they are not so rational after all" rhetoric that the essay itself and its proponents are employing. Also, both the addressee of the essay and the backdrop community of this forum use a much more specific definition of "rationality" and "rationalism" than the dictionary one as a core part of their identity. I doubt that Alexandros does not know this, so to suddenly insist on the dictionary definition seems rather disingenuous.

project your distrust on to others as if they must rely their trust upon you more than their own cognitive skills

I think something is still off about your use of words (the "project", "rely their trust") which might result in us talking past each other to some extent, but I'm emphatically not trying to say that anyone should take my word for it. I've already declared that I know little about the field! However, with bounded resources, two rational actors can in fact arrive at different beliefs. I'm asserting that I'm acting rationally in continuing to treat Ivermectin as ineffectual and not reading the essay, and other people who do the same may be doing so as well. Some other people who did not read the essay and continue believing that and acting as if Ivermectin works against COVID may also be acting rational. People who actually read the essay and updated their beliefs to treat Ivermectin as effective may also be acting rational. Therefore, the circumstance that the essay is being dismissed is not prima facie evidence that the community is widely lacking in rationality. (Of course, someone who would stand to benefit a great deal from being convinced that Ivermectin works and can afford understanding the essay would be irrational to ignore it, and conversely someone who should have better things to do with their time may be irrational in reading and updating on the essay.)

The Wikipedia page you linked to makes it clear that "bounded rationality" is a theoretical concept used in posthoc decision-making modeling (such as in economics) and is not something one uses, as your comment does, to buttress a rationalizing of their subjective-intuitive belief giving an impression of it being more factual that it actually is.

Does it? Provide a citation saying that it's only used in post-hoc modelling. I'm not a fan of waving around real-life credentials in general, but here it's probably worth saying that I've actually published in the field and I can assure you there is a plethora of papers written explicitly from the perspective of planning future actions - and, either way, the separation between post-hoc and non-post-hoc you seem to be postulating does not exist, since anything that can be used to evaluate an outcome can also be used to estimate optimal actions, if you know anything about the relationship between actions and expected outcomes at all.

That's the kind of sophistry only rationalists can come up with!

By "rationalists", do you now mean people who swear by "reason and logic", or are you back to using the more specific definition?