No tears for spilt milk : temporal neutrality and the rationality of future-bias

Abstract

Temporal neutrality has become widely accepted as a rational requirement for agents, due to recent arguments from Sullivan (2018), Greene and Sullivan (2015), Dougherty (2011, 2015) and Brink (2010), which build on older remarks from Rawls (1971), Parfit (1984), and Sidgwick (1874). According to Brink, an agent is temporally neutral if she does not prefer an event over another solely based on temporal location and gives equal significance to all parts of life. Temporally neutral agents are required not to be time biased.

There are two forms of time-biases in the debate, near-bias and future-bias. An agent is near-biased if she prefers positive events to be closer to her and negative events to be further away in the future. An agent is future-biased if she prefers positive events to be future rather than past, and negative events to be past rather than future. Both forms of time-biases are seen as irrational by proponents of temporal neutrality due to (1) concerns of arbitrariness, (2) concerns around pragmatic loss, and (3) irrelevant influences.

While I accept near-bias as rationally impermissible, I develop a systematic defence of the rationality of future-bias in this thesis. My thesis provides defences against all types of arguments from temporal neutralists and aims to set out rational grounds for being future-biased based on rational agency, control-asymmetry, and emotional prudence.

Additionally, I explore implications of future-bias for some moral theories. argue that we should understand future-bias as a comprehensive preference pattern that includes all evaluative aspects of life. If we accept a comprehensive reading of future-bias, this undermines some moral theories by making it permissible to focus moral evaluation on the present and future.

In short, my thesis argues that we are permitted to, and sometimes should be future-biased, and should reassess our moral theories accordingly.