Abstract
Our present moral traits are unable to provide the level of large-scale co-operation necessary to deal with risks such as nuclear proliferation, drastic climate change and pandemics. In order to survive in an environment with powerful and easily available technologies, some authors claim that we need to improve our moral traits with moral enhancement. But this is prone to produce paradoxical effects, be self-reinforcing and harm personal identity. The risks of moral enhancement require the use of a safety framework; such a framework should guarantee practical robustness to moral uncertainty, empirical adequacy, correct balance between dispositions, preservation of identity, and be sensitive to practical considerations such as emergent social effects. A virtue theory can meet all these desiderata. Possible frameworks incorporate them to variable degrees. The social value orientations framework is one of the most promising candidates.
Notes
1 I will define traits as general and stable patterns of behaviour, thought or emotion. Traits are meant to be more robust and general than mere dispositions. One may have the disposition to be frightened by a specific snake appearing inside one’s house, but one might have the trait to be frightened by the sudden appearance of visual patterns resembling consistent evolutionary threats to survival.
2 The good, in turn, might be defined as good consequences or fulfilling norms. Although I will use external good consequences being systematically produced by a trait as a criterion, I will also consider virtues might be connected to internal goods—or at least necessary to avoid undermining certain external goods—when mentioning virtues relation to personal identity.
3 As I defined here, virtue ethics is a normative theory that advocates traits conductive to the good. A popular definition of those traits is that they are the ones possessed by those who are virtuous. Often, a defining feature of the virtuous is being a moral exemplar. The connection with virtue ethics here is likely, but not necessary. Therefore, Crutchfield is not necessarily assuming virtue ethics.
4 Empirical theories, when poorly reasoned, can be incoherent and impractical. These should also be avoided, which I aim to do with my empirical adequacy desideratum.
5 For a more general defence of the idea that we are often better off arriving at a dilemma instead of an outright prescription, see the first chapter of Nussbaum (Citation2001).
6 Roughly, a theory is empirically adequate when its empirical claims are correct. In other terms, a theory is empirically adequate when there is a correspondence (isomorphism) between what it describes (and their relations) and observable phenomena (and their relations; van Fraassen Citation1980).
7 That linguistic patterns reflect real patterns is the central hypothesis of the Big Five framework, called Lexical Hypothesis (de Raad and Mlacic Citation2016). To the extent that this framework has enjoyed empirical success, the hypothesis has been confirmed.
8 Following my earlier definitions, I classify Aristotelian virtue ethics as a virtue theory that further claims to be the correct moral theory.
9 From section 6, Book II of Aristotle’s Nicomachean Ethics: “Virtue, then, is a state of character concerned with choice, lying in a mean, i.e. the mean relative to us, this being determined by a rational principle, and by that principle by which the man of practical wisdom would determine it” (Aristotle, Brown, and Ross Citation2009, 31).
10 Given uncertainty over their importance, they all have the same double weight. Additional work reducing this uncertainty is needed.
11 This means the frameworks are being analysed with an emphasis on safety. If the framework is highly safe but unlikely to be implemented, it will be harmless but useless.