SECTION

Which Way is Forward: Value Differences, Drift, & Convergence

Lorem Ipsum

On What Matters - Derek Parfit. The successor to Reasons and Persons provides an overview of classic moral theories, such as deontology, consequentialism, contractualism, and of major moral issues, such as free will and responsibility.

Meta Ethics Sequence - Eliezer Yudkowsky. Especially The Moral Void, Whither Moral Progress, Existential Angst Factory, Morality as Fixed Computation, Value is Fragile, Could Anything Be Right, and Changing Your Meta Ethics.

Report on Consciousness and Moral Personhood - Luke Muehlhauser. For figuring out which consciousnesses to include in your ethics and how.

Crucial Considerations - Nick Bostrom. By thinking a little harder we may often come to opposite conclusions about the desirability of our actions, so it’s worth thinking hard about which crucial considerations should be guiding our actions.

Facing the Unknown, Infinite Ethics - Nick Bostrom. Our epistemic limitations about the (long-term) consequences of our actions are problematic for making the right decision. In Problems And Solutions In Infinite Ethics, Ben West lays out a few strategies that may help us counter some of those epistemic limitations.

Normative Uncertainty - Will MacAskill. Argues that we should treat moral uncertainty and empirical uncertainty analogously, and use expected utility theory as framework to decide between our differing moral convictions. 

The Moral Parliament - Nick Bostrom, Toby Ord. Suggests to approach moral uncertainty across mutually exclusive moral theories by assigning each of them some probability and letting the theories send a respective number of delegates to a moral parliament.

Moral Trade - Toby Ord. Argues that we can trade our values similar to how we exchange goods: Just like people with different tastes or needs can exchange goods or services such that they each feel they have been made better off, people with different moral views may be able to exchange those when trying to improve the world.

Reflective Equilibrium - John Rawls. Another method for handling normative uncertainty. It suggests continuously working back and forth among our moral intuitions about actions, the principles that govern them, and the theoretical considerations behind them, revising them when necessary to achieve coherence among them. The Wide Reflective Equilibrium by Norman Daniels adds that we should actively seek out objections that may refute our convictions.

In The Balance - Scott Alexander. A tongue in cheek story on handling infinite regress in moral updating: “And you will tell them the story of how once you found the Artifact that gave you mastery of the universe, and you refused to take more than about three minutes figuring out what to use it for, because that would have been annoying.”

Fundamental Value Differences Are Not That Fundamental, The Whole City is Center, Value Differences as Differently Crystallized Metaphysical Heuristics - Scott Alexander. Argues that human value differences are more shallow than we commonly think and may track the same universal core values that could help us reconstruct a common crude human morality.

On Value Drift - Robin Hanson. Shows how much values may drift over time. In Let Values Drift, G. Gordon Worley III argues that this is not a cause for concern but only to be expected under the correct (Friston-aligned) conception of valuing.

Moral Progress vs. the Simple Passage of Time - Holden Karnofsky. How to potentially distinguish value drift from moral progress.

My Outlook - Paul Christiano. On the probability of civilizational survival, and the relative influence, distribution and entrenchment of human values.

Three Worlds Collide - Eliezer Yudkowsky. A novella about how much values can differ across different mind-architectures.

Artificial Intelligence, Values, and Alignment - Iason Gabriel. On AI Alignment but discusses the problem of different people with different values, and how Contractualism, Rights, Rawl’s Veil of Ignorance, or Social Choice Theory may help us reach overlapping consensus.

Read more