The First Question

Longer term readers may remember that last summer I completed an MPhil in Political Theory here in Oxford. One of the requirements of doing so was the production of a thesis of up to thirty thousand words. Although this was not the topic the title indicated - 'Grounding Neutrality: The Normative Foundations of State Neutrality' - the central concern of the thesis was really the conditions of political legitimacy: what does it take for us to say that the exercise of state power over some group of people is justified or legitimate? Let us begin with a little thought experiment.

Imagine a case where I attempt to justify to some group my demand that they become my slaves. This is a political arrangement... since it involves the imposition of some set of coercive institutions on some group.

Now, intuitively, we feel that slavery is unjustifiable, utterly beyond the pale, so it would count heavily against any ethical or political theory that it did not rule out, absolutely, slavery.

Obviously, however, there is a vanishingly small chance of me being able to justify to those I would enslave their enslavement. The first explanation for this might be that no-one could have any real reason to want to be enslaved.

After all, one of the reasons slavery is beyond the pale is that it is usually fairly remorselessly awful. Everyone knows that, and so they will, all other things being equal, avoid becoming someone else's slave. What if all other things aren't equal though?

However, there are cases in which it could be advantageous to become a slave. If someone would die otherwise, we might well think that they would however unwillingly accept their enslavement as the least bad of a set of horrendous options.

A simple consent theory, then, runs into a problem.

Yet it cannot be the case that simply because all the alternatives for them involved dying in the immediate future, my enslavement of some group is justified. If that were the case, my threat to murder someone unless they became my slave would justify them becoming my slave.

An explanation is required though.

The coercion involved in [this case], the control over the sets of options involved, makes justification impossible: it is only because I am able to manipulate the situation so that slavery becomes attractive that anyone would agree to become a slave.

Thus the first and foundational principle of political legitimacy is derived: that might is not right, that to each according to his threat advantage is not a principle of justice.

Notice that presupposes a particular kind of what the problem of political legitimacy is, though. Here, the worry is about unconstrained violence, about conflict and the fear of conflict destroying the possibility of a tolerably ordered social life. Violence needs to be controlled and placed in the service of some kind of moral framework which lifts or at least amielorates the costs which it, unrestrained, imposes on people. The first question of politics, for this kind of view, is Hobbes': as Bernard Williams puts it in his 'Realism and Moralism in Political Theory', how to secure "order, protection, safety, trust, and the conditions of cooperation", a question which recurs, "can never be presumed to be gone for good".

This, I suppose, is the root of a number of significant differences in political outlook between me and someone like Phil Edwards. For me, the first question is Hobbes', whereas for Phil, I think it is precisely the reverse. Rather than worrying about the violence unleashed in the absence of rules, Phil seems to worry about the violence unleashed in the name of maintaining a set of rules. There is of course, for those of us who depart company from Hobbes in disagreeing with the claim that the "conditions for solving the first problem... [are] so demanding that they [are] sufficient to determine the rest of the political arrangements", a sense in which Phil's first question is the one immediately begged by ours, which turns naturally to the importance of the idea "that the state - the solution - should not become part of the problem". That doesn't get quite to the root of it though: there is a deeper disagreement, related I think to beliefs about where and for what reason resistance to moral progress exists, one I don't intend to do more than draw attention to.

Williams takes Hobbes' question to be fundamental, but in a much more minimal way than most contemporary political theorists do. Although Williams draws out of Hobbes' question something he calls the Basic Legitimation Demand, the requirements of this demand are much less onerous than those of, for example, Rawls. Rather than a single or even fairly limited set of political arrangements satisfying it, any state or similar body which can give an explanation for its existence to its subjects which, hermeneutically, makes sense, which justifies its exercise of power to them at that time, passes it. About the only political arrangement which Williams thinks will always fail this test is the case I used as foundational, as a starting point for cashing out the conditions of legitmacy, that of open slavery.

Further, Williams contrasts the Basic Legitimation Demand with what he calls 'political moralism', an example of which is 'the structural model' which seeks to lay "down the moral conditions of existence under power", the archetype of which is Rawls. This appears, initially, odd. Williams' first question would seem to be the same as that of the structural model, and the only case he absolutely rules out satisfying his condition of legitimacy is one fairly plausible foundational unacceptable case for the structural model, so one would think that his model, whatever it is, would fall fairly close to the structural one. He seeks to quite sharply contrast his view with all forms of political moralism though, calling himself a political realist.

This is because Williams understands the reasons, although not necessarily the conditions, for asking the first question quite differently. Williams disagrees with the second step of the argument I quoted above. For him, slavery is not a political arrangement.

The situation of one lot of people terrorizing another lot of people is not per se a political situation: it is, rather, the situation which the existence of the political is supposed to alleviate (replace).

Because the first political question is Hobbes', for a given arrangement to qualify as political, it must seek somehow to answer that question, and slavery, where, as another's property, there are no limits on what they may permissibly do with you, is no kind of answer to that question: it is a restatement of the problem. Thus, as a distinctively political question the Basic Legitimation Demand arises only in a situation where it is claimed that Hobbes' question is, in some way, being answered:

[i]f the power of one lot of people over another is to represent a solution to the first political question, and not itself be part of the problem, something has to be said to explain (to the less empowered, to concerned bystanders, to children being educated in this structure, etc.) what the difference is between the solution and the problem, and that cannot simply be an account of successful domination.

This is not the same as the Rawlsian concern with "the moral conditions of coexistence under power", because those moral conditions, whatever they may be, exist before some solution or other to the Hobbesian question is claimed. They are not, in Williams' sense, particularly political: they do not respond to any solution to this problem, and the question of its legitimacy, but rather to the bare fact of power, both inside and outside of attempts to solve Hobbes' question. Neither, in Williams' sense, are they really responding to the same question, for in his view Hobbes' question is not about the bare fact of power, but a loosely defined subset of the consequences of the bare fact of power: how to secure "order, protection, safety, trust, and the conditions of cooperation".

That is not quite the same as the structural model's position. Coercion qua coercion is worrying and hence requires legitimation for the Rawlsian, whereas it is rather that coercion claims to be to some worthy end - in particular, the securing of "order, protection, safety, trust, and the conditions of cooperation" - that raises the question of its political legitimation for Williams. In the absence of that claim to be a relief from some burden, and thus in some way justified by that relief, Williams claims the question of coercion's legitimation does not arise, at least as a political concern - which is of course not to say that it does not raise other, equally pressing but nevertheless distinct, moral issues:

[a] condition of there being a genuine demand for justification is this: A coerces B and claims that B would be wrong to fight back: resents it, forbids it, rallies others to oppose it as wrong, and so on. By doing this, A claims that his actions transcend the conditions of warfare, and this gives rise to a demand for justification of what A does. When A is the state, these claims constitute its claim of authority over B.

Thus, for Williams, it is the act of claiming legitimacy that creates the problem of legitimation. Warfare, whatever moral limits it may have, is not the same as, and so does not raise the same questions as, the claim to authoritative rule. This makes the focus of the Basic Legitimation Demand much tighter than that of a Rawlsian account of political legitimacy, because it is only when authority comes in to question that it needs to have an answer. Regimes which rested or rest on myths are therefore legitimate as long as those myths are acceptable, because the demand for legitimacy arises in a specific place at a specific time, and thus requires answers which are appropriate to the particularities of that place and time.

Of course, Williams' position does occupy something of a meta-level: its denial of universalism is distinctly across time, rather than, for us at least, space, as 'here and right now', some form of liberalism is common currency. Neither is it a form of Rorty-esque ironism: the failure of universalism is not reductionist, but rather a somewhat humbling admission of limits, of imperfection. Still, there is something disturbing, disquieting about it: a historicism I am not entirely happy with, a hermeneutical impulse that seems far too concerned with gleefully swallowing its own tail.