Posts

Sorted by New

Wiki Contributions

Comments

Sorted by
VaRuna42

I think this is a great outline of how these strategies form. A very similar idea is described in The Elephant in the Brain, but this is straightforwardly written and more visceral in a way I felt the book (and most other attempts to describe it) lacked. Kudos!

The drive to be "perfectly rational" and push all slivers of self-deception out with force is, I think, one of the core psychological errors made in rationalist circles (including the writing) for exactly the reasons you lined out. Well explained!

Honesty, and specifically self-honesty, is held as one of the highest virtues. I think it's even cited as the only example of an instrumental good that can and should just be treated as an inherent good. Considering that, I would have loved to see you engage a little more in the discussion of how what you propose interferes with this and the tradeoff there.

Your proposed solution resembles IFS therapy, focusing on not forcing the part of you that tries to do something you might consciously disapprove of, and instead being understanding and accepting of that mechanism as well. The difference is that the intention is to find out what is happening, and not to try to solve it from a distance.

The last point touches on my central issue with your suggestion. The intention of really letting the self-deception sit does make it easier to notice when it happens, but so does investigation when it is done gently and without force. The problem with building up the habit to let self-deception run is that you will notice it more, but it will also continue to happen habitually, or the propensity for it might even grow stronger.

If there were actually plenty of hostile telepaths running around, that might not be a bad thing, but I think it is actually a rare occurrence in adulthood. To be specific, what is rare is a situation in which you really do want to self-deceive in that way and not see the avoided problem instead. Almost all of these situations are cases in which the solution will be "investigating whether the telepaths are in fact hostile and discovering they're not," as you wrote. There are obvious situations in which another person might be dismayed by your true mental state, but keeping that hidden only serves you in situations where the other person does not actually care about your well-being (in the form of your true mental state) anyway, in which case a detectable lie or insincere appeasement will work just as well and does not require the Newcomb-like deception in the first place. Having better information about your own internal state and preferences is a net good in all these situations.

Of course, there are situations in which you really would want to lie, but I would argue that you have to make the distinction between whom you can be honest with and whom to lie to anyway, subconsciously. So if you do, then you can also do that consciously, and rather learn to face the discomfort of confronting conflict. And, in these rare cases, realize and integrate that if someone is truly hostile enough to not just disregard your true preferences, but will even use knowledge of them against you, then you should be consciously aware of that (not just subconsciously, as you have to be in order to pull off Newcomb-like deception in the first place) and treat them appropriately, which includes lying to them with impunity. And if you do not make the distinction between whom you can really be honest with about everything, then that itself poses a significant issue, and bringing that to attention is invaluable in itself.

Another practical problem with the method is that most of the situations in which this would apply in fact do not have a practical solution, not even a concrete emotional one, because the problem is the avoidance of the feeling itself. There can be practical things you can do to make an instance of shame or procrastination less overwhelming, but when the badness of the problem is inherent, then no strategy can ease it enough to make the self-deception superfluous. In most cases, that means that the problem just festers until the urgency forces it out, in which case nothing is learned or gained, just a solution postponed. Or, without a timeline, the deception just stays unaddressed, you never learning what is behind it, which is tragic, considering that there might be something positive in there, and the negative is in fact only in the anticipation, since there are no hostile telepaths raining consequences on you as a result of your self-knowledge. In my experience, being in a safe environment suffices as a prerequisite to the safety of lifting any self-deception, with time and patience. Another person who can assure you that it's okay and safe can be helpful, but also not strictly needed.

I would be curious to hear your thoughts, most of which comes down to the discussion of tradeoffs between solving the problem and the high value of self-honesty and having accurate models over all mentioned above.