In last year’s survey, someone likened Less-Wrong rationalism to Ayn Rand’s Objectivism. Rand once summed up her philosophy in the following series of soundbites: “Metaphysics, objective reality. Epistemology, reason. Ethics, self-interest. Politics, capitalism.” What would the analogous summary of the LW philosophy be?
In the end, I found that the simplest way to sum it up was to cite particular thinkers: “Metaphysics, Everett. Epistemology, Bayes. Ethics, Bentham. Politics, Vinge.”
A few comments:
For metaphysics… I considered “multiverse” since it suggests not only Everett, but also the broader scheme of Tegmark, which is clearly popular here. Also, “naturalism”, but that is less informative.
For epistemology… maybe “Jaynes” is an alternative. Or “overcoming cognitive bias”. Or just “science”; except that the Sequences contain a post saying that Bayes trumps Science.
For ethics… “Bentham” is an anodyne choice. I was trying to suggest utilitarianism. If there was a single well-known thinker who exemplifies “effective altruism”, I would have gone with them instead… Originally I said “CEV” here; but CEV is really just a guess at what the ethics of a friendly AI should be.
For politics… Originally, I had “FAI” as the answer here. That may seem odd—friendly AI is not presented as a political doctrine or opinion—but the paradigm is that AGI will determine the future of the world, and FAI is the answer to the challenge of AGI. These are political concerns, even if the ideal for FAI theory would be to arrive at conclusions about what ought to happen, that become as uncontroversial and “nonpartisan” as the equations of gravity. I chose Vernor Vinge as the iconic name to use here; I suppose one could use Kurzweil. Alternatively, one could argue that LW’s cardinal metapolitical framework is “existential risk”, rather than FAI vs UFAI.
I wonder whether more people will think of Julian Jaynes rather than E. T. Jaynes if you just rattle “Everett, Jaynes, Bentham, Vinge” at them. This does seem like a very nice ultra-concise description though.
In this context, this should surely be “Epistemology, Jaynes”—attributing the LessWrong conception of “Bayesianism” to the Rev Bayes is a bit of a stretch. Though it’s unclear what Jaynes would have thought of the claim that Bayes trumps science.
In last year’s survey, someone likened Less-Wrong rationalism to Ayn Rand’s Objectivism. Rand once summed up her philosophy in the following series of soundbites: “Metaphysics, objective reality. Epistemology, reason. Ethics, self-interest. Politics, capitalism.” What would the analogous summary of the LW philosophy be?
In the end, I found that the simplest way to sum it up was to cite particular thinkers: “Metaphysics, Everett. Epistemology, Bayes. Ethics, Bentham. Politics, Vinge.”
A few comments:
For metaphysics… I considered “multiverse” since it suggests not only Everett, but also the broader scheme of Tegmark, which is clearly popular here. Also, “naturalism”, but that is less informative.
For epistemology… maybe “Jaynes” is an alternative. Or “overcoming cognitive bias”. Or just “science”; except that the Sequences contain a post saying that Bayes trumps Science.
For ethics… “Bentham” is an anodyne choice. I was trying to suggest utilitarianism. If there was a single well-known thinker who exemplifies “effective altruism”, I would have gone with them instead… Originally I said “CEV” here; but CEV is really just a guess at what the ethics of a friendly AI should be.
For politics… Originally, I had “FAI” as the answer here. That may seem odd—friendly AI is not presented as a political doctrine or opinion—but the paradigm is that AGI will determine the future of the world, and FAI is the answer to the challenge of AGI. These are political concerns, even if the ideal for FAI theory would be to arrive at conclusions about what ought to happen, that become as uncontroversial and “nonpartisan” as the equations of gravity. I chose Vernor Vinge as the iconic name to use here; I suppose one could use Kurzweil. Alternatively, one could argue that LW’s cardinal metapolitical framework is “existential risk”, rather than FAI vs UFAI.
I wonder whether more people will think of Julian Jaynes rather than E. T. Jaynes if you just rattle “Everett, Jaynes, Bentham, Vinge” at them. This does seem like a very nice ultra-concise description though.
In this context, this should surely be “Epistemology, Jaynes”—attributing the LessWrong conception of “Bayesianism” to the Rev Bayes is a bit of a stretch. Though it’s unclear what Jaynes would have thought of the claim that Bayes trumps science.