Centre for Ethics in Public Life

In the face of widespread uncertainty about how best to characterize the project of neo- Aristotelian naturalism, this paper presents a novel way to understand the distinction John McDowell draws in 'Two Sorts of Naturalism' between a crass empirical naturalism and a more satisfactory alternative. Empirical naturalism is attacked using the representation problem as presented by Bernard Williams. The alternative, excellence naturalism, understands the reasons for action we draw from nature as the application of the characteristic functions of a being of our type towards its characteristic ends. They are excellences because they allow the overcoming of obstacles towards the good life for a being of that type. Finally, a problem for excellence naturalism is indicated, regarding how to handle new or changing forms of life.

Difference democrats object to the deliberative theory of democracy because it might not give the appropriate weight to claims made by members of some cultures in multicultural societies. This objection to the claim of deliberative democracy as a theory of justice can be understood in two ways: as a ‘communicative challenge’, which states that the reliance upon a single deliberative procedure is a shortcoming which might privilege some cultures by the decision-making process being in their political idiom; or as a ‘representative challenge’, which holds that the goods that some cultures might want to secure through the deliberative process might not be recognised as important aims by other cultures. I argue that the communicative challenge fails, because it is mistaken about what parts of political communication are pertinent. Recognising this failure allows us to see more clearly that the real challenge of multiculturalism to deliberative democracy is the representative challenge: the difficulty of getting one culture to acknowledge the importance of the aims of another culture which it itself does not share.

In this chapter I present and defend limited conventions, a version of Lewisian conventions where the result of a convention is limited by general principles (or rules, standards, etc.), such that the conventions supplement rather than replace principles. My substantive claim here is a hypothetical: for any moral theory, if that theory faces a certain class of problem cases, any method that allows people subscribing to that theory to navigate through such problems amounts to a limited convention. This only covers the kind of problem case I will discuss, but I will argue that these are so common and troublesome that we need the kind of solution I offer here. The kind of problem case in question are instances of what I call the strategic underdetermination problem. I start from the observation that, unless every moral issue could be settled from general principles, underdetermination will arise, meaning there are multiple courses of action that satisfy your principles as well as any other but are mutually exclusive. In such a case it would matter which of these options you decide upon, but your principles give you no way to choose. My focus here is on the social dimension of this problem: there are situations where what you should do depends on what other people do in these cases. To be more specific, I concentrate on what is called strategic cases,. In a strategic case, if it is uncertain what the other parties will do, then it is uncertain what you should do as well; if underdetermination is in effect, you don't know which of the available options the other parties will take; thus, given the uncertainty about their actions, you are uncertain about yours as well. That is the strategic underdetermination problem (SUP for short), and it undermines your ability to reason towards your moral ends. That is the problem for which I argue limited conventions are the unique solution. This paper has three sections. In the first section I introduce limited conventions by applying Lewisian conventions to the hypothetical case where principles underdetermine what we should do. In the second section I give a range of examples of the kind of phenomena I intend limited conventions to cover, to illustrate my claim that every moral theory that is vulnerable to the SUP has a need for them. In the third part I give a defence of the normativity of such conventions, and also respond to various objections that have been raised against similar positions.

I defend the view that a sufficient condition for some command to be authoritative is if the command were to produce a particular kind of convention if it were followed. The convention in question is what I call a 'limited convention', a Lewisian convention where possible outcomes are ranked not according to individual preferences but instead to the degree that they conform to the community's shared principles. If all the subjects conform to the command, the community coordinates towards an outcome that is consistent with their principles, whereas if they act their own way has unreliable results at best. Commands with conventional authority should be followed because to do so allows the members of the community to handle instances where they both need to know what each other are doing in order to not work at cross-purposes and the principles underdetermine what they should do. Failure to follow the command undermines your fellows' ability to follow your shared principles, because it strips away the coordination required to do so, and thus you have a pre-emptive reason to conform. I go on to show how parental authority, something that is not in general taken to be established by convention, is often an instance of conventional authority.

As the use of algorithmic decision-making becomes more commonplace, so too does the worry that these algorithms are often inscrutable and our use of them is a threat to our agency. Since we do not understand why an inscrutable process recommends one option another, we lose our ability to judge whether the guidance is appropriate and are vulnerable to being led astray. In response, I claim that guidance being inscrutable does not automatically make its guidance inappropriate. This phenomenon is not restricted to algorithms, and there are many social processes which we should conform to but are similarly unable to judge for ourselves. I provide a framework for how we can depend on inscrutable processes by introducing a distinction between knowing conformity (where I understand what justifies the guidance) from mere conformity (where I merely do what I am told), and showing how mere conformity is often positively valuable, because it allows for extended processes that in turn allow us to accomplish more than we could on our own. This is in effect a division of deliberative labour, which I argue is something commonplace but not often recognised, of which algorithmic guidance can be an example.