While I do think there are parts of Buddhist practice and thought that could be useful to rationalists, I think this post is fundamentally in the wrong direction. Instead of trying to identify and isolate these useful parts you simply try to mash together all the fundamental principles of Bayesianism and Buddhism in a single swoop, which makes you come across as the victim of an affective death spiral.
Unfortunately, this post will probably make people somewhat less likely to consider more fruitful ideas in this direction later on.
Seeing as its your first post this should have been submitted to the Discussion section first.
...While I do think there are parts of Buddhist practice and thought that could be useful to rationalists, I think this post is fundamentally in the wrong direction. Instead of trying to identify and isolate these useful parts you simply try to mash together all the fundamental principles of Bayesianism and Buddhism in a single swoop, which makes you come across as the victim of an affective death spiral.
Unfortunately, this post will probably make people somewhat less likely to consider more fruitful ideas in this direction later on.
Seeing as its your first
I see epistemic rationality as a connection between Bayesian inference and Buddhism. Bayesian inference focuses on the mathematics for updating belief, Bayes' theorem. Buddhism focuses on the human factors of belief, how belief can lead to dukkha (suffering, discontent), and how dukkha can be avoided.
Combining these ideas could give us a practical path to achieving bohdi (enlightenment, optimal belief) from within the context of our human mind (emotional and physical). Call it the optimal enlightenment hypothesis. I am opening this concept up for LW discussion.
Related questions:
From recent posts by Luke_Grecki and Will_Newsome, it seems that others are thinking along the same line.
Epistemic Rationality
To be rational we need to be willing to see the world as it is and not as we want it to be. This requires avoiding emotional attachment to our beliefs and adopting an even approach to evaluating evidence.
See Also:
Epistemic rationality as defined by Eliezer Yudkowsky:
Bayesian inference is a method for updating belief within epistemic rationality:
Buddhism encourages people to view reality as it really is and to constantly seek enlightenment.
In Buddhism's Four Noble Truths, as stated here:
Buddhism's concept of rebirth can be seen as an openness to updating beliefs.
We are not limited by our current sense of self, here:
The Context Principle
The context principle is integral to both Bayesian inference and Buddhism.
Context Principle: Context creates meaning, and in its absence there is no meaning.
When considering a belief the context must be considered. Outside of its intended context a belief may be wrong or even meaningless.
In Bayesian probability the meaning of the evidence, and the prior probability distribution, depend on the context. When different observers use the same method and evidence, they can come to different conclusions if they have adopted different contexts.
For example:
In Buddhism, the context principle can be found in the three marks of existence:
Anicca
From here:
If our beliefs are conditional on our context, and our context is in a constant state of flux, then we must be mindful and be ready to observe these changes, and we must be constantly willing to change our beliefs based on the evidence.
Dukka
From here:
A rough summary, "suffering is context dependent, if you believe you are suffering, then you are".
Accepting the world as it is and not as we want to be is not simply about becoming comfortable with the status quo. It is about gaining the awareness we need to transform our context. This is explained well in Radical Buddhism and the Paradox of Acceptance.
Anatta
From here:
From here:
The concept of self is a convention, not an absolute, it refers to a constantly changing composite. The meaning of self depends on context. Clinging to an inappropriate concept of self can lead to dukka.
Challenges
Buddhism is not inherently rational
Although there are hints of rationality in Buddhism, it was not created to be a philosophy of rationality. There is a lot of mysticism, and the different schools of thought appear to split primarily on metaphysics.
It is tempting to discard the mysticism outright, but I suspect that there are ideas encoded in the mysticism that help people understand and adopt the beneficial beliefs of Buddhism. Reincarnation/rebirth for example could help people accept Annica (impermanence). I'm not proposing that the mysticism be kept; I'm suggesting that it may be useful to understand its context specific value.
Bayesian inference is not really a philosophy
There are aspects of Bayesian inference that sound like a philosophy, for example changing beliefs based on evidence, but it is really a method of statistical inference. Extending these ideas to domains where it is difficult or impossible to calculate probabilities will be difficult.