Reclaiming Communal Self-Governance in Liberty Rights — Epoche C1
There is a difference between no one interfering with your choices and no one being in a position to interfere with them at will, and the whole of this essay turns on it. A slave with an indulgent master suffers no interference; he nevertheless lives at another's discretion, and the classical republican tradition counted him unfree for that reason alone. Philip Pettit's revival of that tradition under the name freedom as non-domination is the framework used here. The claim defended is that non-domination cannot be delivered once and for all by institutional design, because the capacity of the powerful to act at their own discretion regrows unless citizens continuously contest it — so that participation is a condition of liberty rather than an optional exercise of it. Two claims in the earlier version of this argument are corrected along the way: its account of Joseph Raz, and its suggestion that the communal dimension is something Pettit's theory lacks. Three Ideas of Freedom, and Why the Third Is Not Redundant The terms have to be fixed before the argument can move, because 'liberty' names at least three different things in this literature. Isaiah Berlin's distinction, set out in his 1958 inaugural lecture and reprinted as 'Two Concepts of Liberty' (Berlin 1969), separates negative liberty — the absence of obstacles deliberately placed in one's way by other people — from positive liberty, which concerns self-mastery: being governed by one's own reason rather than by appetite or ignorance. Berlin's warning about the second concept is the pivot of his essay: once freedom is identified with acting as one's 'true' or rational self would act, a ruler who claims to know that self can coerce people while describing the coercion as liberation. The warning matters here because any theory that makes civic participation part of freedom must show that it is not sliding into what Berlin feared. Republican liberty is a third thing. Writing $I(a,x)$ for the relation '$a$ actually interferes with $x$' and $C(a,x)$ for '$a$ has an uncontrolled capacity to interfere with $x$', the two conditions are $$\text{non-interference: } \neg \exists a\; I(a,x), \qquad \text{non-domination: } \neg \exists a\; C(a,x).$$ The difference is modal: the first quantifies over what happens, the second over what could happen at another's discretion. Three consequences follow, and they are what give the republican view its bite. Unexercised power is still unfreedom. The benevolent master's slave satisfies non-interference and fails non-domination. The Roman law that the tradition draws on classified persons as liber or servus by status, not by how they were treated, and it is that status-based judgement the republican account revives (Pettit 1997). Domination shows up as pre-emptive deference. A person who must anticipate the powerful will trim their conduct in advance — flattering, self-censoring, avoiding whole courses of action. No interference is recorded, because the subject has already conceded. A theory that counts only actual interference cannot see this, although it is the most common form the loss of freedom takes. Not all interference reduces freedom. This is the counterintuitive corollary and the most important. If a law is forced to track the interests of those subject to it, and they have effective means of contesting it when it does not, then its interference is not arbitrary — not exercised at anyone's mere discretion — and it does not dominate. Law therefore is not, as Hobbes and Bentham held, a subtraction from liberty in every instance. It can be constitutive of it. What Raz Actually Says The earlier version of this essay treated Joseph Raz as a representative of an atomistic liberalism that neglects the communal preconditions of freedom. That is a misreading, and correcting it strengthens rather than weakens the essay's case. Raz (1986) holds that personal autonomy — being in part the author of one's own life — requires three things jointly: appropriate mental capacities; an adequate range of valuable options; and independence, meaning freedom from coercion and manipulation. Note the second condition. An adequate range of options is not something a person can supply for themselves, and Raz says so: the options available to anyone are constituted by the social forms of their society. One cannot choose to be a barrister, a novelist or a monk unless one's society contains the practices of law, literature and monasticism, complete with the shared understandings that make those roles intelligible. This is Raz's social forms thesis, and it makes autonomy dependent on collective achievement in a direct sense. Raz also rejects the anti-perfectionism his position is often assumed to share. Because autonomy is valuable and requires valuable options, he argues the state may and should act to sustain the social forms that supply them. The result is that Raz's liberalism is not the atomistic doctrine the earlier version attacked. The genuine disagreement lies elsewhere: Raz's account is about the range of options an agent faces, and can be satisfied by a benign authority that provides a rich menu without being answerable to anyone. Non-domination asks a different question — who holds the power to alter the menu, and on whose terms. A society could score well on Raz's criteria and still leave its members dependent on the goodwill of those who supply the options. That, and not neglect of the communal, is the gap the republican account fills. Pettit Went There First: Contestation and Control The earlier version presented the participatory element as a supplement to Pettit, whose theory it described as resting on institutional design and legal safeguards. That understates the source, and the correction matters because the essay's central proposal turns out to have a defender in the very author it was distancing itself from. Already in Republicanism , Pettit (1997) argues that non-domination requires not consent to governm