Recently there has been an interesting debate concerning the status of immigration controls between David Miller and Arash Abizadeh in the pages of the journal Political Theory. The debate involves a number of interlocking questions that are too involved to adequately summarize here but at the heart of it there is a dispute concerning whether it is correct to view immigration controls as being "coercive" with Abizadeh presenting them as not only being coercive but, as such, in violation of "autonomy" whilst Miller, by contrast, denies that they are intrinsically "coercive" and, hence, that they are in principle "coercive" (something Miller defends by distinguishing "coercion" from "prevention").
Rather than try to in any sense arbitrate between the positions in this dispute I would prefer instead to raise a question concerning the degree of agreement there is between the parties to it concerning the understanding of "coercion". Abizadeh explicitly follows a view of coercion that was first set out in the form he adopts by Robert Nozick and that understands coercion primarily as an imposition on the one who is coerced, viewing coercion primarily as an attack on autonomy that requires special justifications. On the basis of this view Abizadeh has motivated a view of the state that sees it as having to answer to a general demos beyond its own declared borders. In response Miller has presented a parallel view of coercion that sees it as imposing a negative on others whilst "prevention", by contrast, is something that simply blocks one's following a certain kind of path to one's end (but not the end itself or as such). So, for Miller, the analogue is that if someone refuses to marry you then they have not "coerced" you but merely prevented you as you can still marry but not the person in question. On these grounds Miller hopes to provide a basis for states not being answerable to an extensive demos for acts of prevention whilst still apparently accepting that "coercion", as such, does primarily have to be understood as a violation of autonomy that requires special justification.
I'm going to here leave aside the example over which the argument between Abizadeh and Miller is taking place, namely, the example of immigration controls. It is not that debate over the status and limit of them is not interesting but only that the question concerning the understanding of coercion and its relationship to autonomy is a conceptually prior question and that on this prior question the disputants seem to have a great deal of common ground.
The common ground is that the Nozickian paradigm of understanding "coercion" receives a lot of assent from both (despite Miller's central point being that the notion of it in Nozick is too broad). This is in the respect that both tend to see coercion as problematic in itself and in need of very special justification due to their joint commitment to a kind of liberal political philosophy that begins from the standpoint of the individual and relates to state intervention as something that is imposed on that individual and hence as always needing to be viewed in such a way that it requires specific justification.
A more republican understanding of the state does not begin by viewing it with suspicion due to a kind of pre-political commitment to "autonomy". This can be seen clearly in the case of Kant's view of coercion. Rather than start from a position that sees coercion as something entirely extrinsic to the status of the person and thus as compromising of "autonomy" Kant begins with a view of "external freedom" that sees it as a product of reciprocal interaction and mutuality. Kant begins from a notion of universal law that is the basis of discussion of political freedom. So, Kant speaks about a "reciprocal relation of choice", and this leads him to a general account of right as "the sum of the conditions under which the choice of one can be united with the choice of another in accordance with a universal law of freedom" (Ak. 6: 230).
The formulation of this "universal law" defines the conditions of what I have elsewhere termed "effective choice", choice that can be exercised in such a way that there can be mutual interaction between individuals such that autonomy can be given content. So, rather than viewing autonomy as a "primitive" that is accepted as a pre-political value and then requiring special justification for any violation of it, Kant rather interprets "autonomy" in such a way that its effective existence is comprehended through the reciprocal operation of coercion. This is why the universal principle of right is formulated in the following manner:
"Any action is right if it can coexist with everyone's freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone's freedom in accordance with a universal law". (Ak. 6: 230)
So, any action that is capable of coexistence with the freedom of others would be an action that is right, and any attempt to restrict this action would be wrong but, similarly, if action is formulated in this way then the reference to coexistence is written into the social contract from the beginning, not added to it later in such a way as to always require additional justification. On this conception, then, right is intrinsically connected, as Kant explicitly states, with an authorization to use coercion so coercion is at the heart of right itself. In this respect whatever position one takes on the right of states to restrict immigration it cannot be correct to view this "coercive" act (if it is that) as one that requires some new element of justification to be added in to our understanding of the state as, rather, the state needs to be seen as the coercive enterprise that it is due to its being the basis of right.