more studying on liberalism ...
Kant on political freedom
This essay looks at Kant's political philosophy through his theory of freedom. In the Groundwork of the Metaphysics of Morals, Kant develops a famous conception of "practical freedom" as independence of the will from determination by "sensuous impulses"; or, in its positive characterisation, the "autonomy" of the rational agent. When Kant later comes to develop a formal theory of justice or "right" in The Metaphysics of Morals, an idea of freedom is again central. But is this juridical/political idea of freedom the same conception as that developed in the foundational moral philosophy? Pursuing freedom through these two works not only helps us to understand the relation between Kant's moral and political thought, but elucidates the political project -- a project of liberalism -- which works through the theory of right.
1. moral freedom
We will need at least a thumbnail sketch of Kantian psychology in order to work on these questions. A useful starting point is the distinction which Kant makes, particularly in the Metaphysics of Morals, between Willkuer or (the faculty of) "choice" from Wille or the "will". As Henry Allison (1990) explains, Wille has both a broad and a narrow sense. We can think of Wille, narrow sense, as the legislative and Willkuer as the executive functions of the overall "faculty of volition" -- or will in the broad sense. Willkuer, choice, makes the final decisions to act. Non-rational animals, Kant believes, have only this lower faculty of choice, and the desires on which their choices are based are always sensuous "impulses" or inclinations (Neigungen) which enmesh them in the causal chains of the sensual world. But humans also have the higher faculty of will, which can command Willkuer and cause it to follow reasoned desires rather than impulses.
The issue of causal determination, or its absence in the case of a free noumenal human, is central to Kant's thinking. Choice (Willkuer) is always determined by another force; will (narrow sense) is never determined, but determining. Kant says that for humans, choice is "affected but not determined by impulses"; instead, choice can be "determined to action by pure will" -- and this pure will is identified with "pure practical reason". There is perhaps an ambiguity here: does will always operate and command choice, or are there at least some occasions when human beings respond directly to impulses, i.e., act (or simply move) non-rationally, like animals? If so, we might say: there may be occasions where human choice is caused by impulse, but it is always at least possible for the will to intervene, overriding impulse and instead directing choice according to a rational principle.
Where Wille (narrow sense) is active, it determines willkuer by setting it laws or principles; willkuer then chooses particular actions (or, more accurately, maxims) which follow these principles. Thus for Wille, the legislator, principles are freely chosen; for Willkuer, they are imperatives. There are two kinds of principles or imperatives, hypothetical and categorical. A hypothetical imperative is a principle which tells Willkuer to act in a certain way in order to pursue an impulsive desire. That is, in the case of acting following a hypothetical imperative the ultimate guiding desire is still a sensuous impulse, and in this sense the human agent is still causally linked to the world of sense: affected, but not determined.
The human agent is only fully free when she follows a categorical imperative. In the Groundwork Kant gives both a negative and a positive characterisation of this "practical" freedom. Positively, freedom means that the will is autonomous, self-legislating. Kant famously holds that "Everything in nature works in accordance with laws." (4:412). But a free rational will makes its own laws. It is "a law to itself (independently of any property of the objects of volition)." (4:440). In this essay, however, we will work with the negative characterisation of freedom as independence from determination by sensuous impulses, from being causally bound to the "world of sense". So the observation above that human choice is not determined by sensuous causation can be restated thus: it is always, at any moment, possible for a human being to be morally free.
2. morals and politics
The relationship between Kant's moral philosophy and his political theory is a debating topic. Is Kant's political theory, in what Marcus Willaschek (2009) calls the "traditional approach", a "mere application of his moral theory, developed in the Groundwork and the Second Critique, to the juridical sphere"? Or, on the other hand, does Kant's discussion of right introduce wholly new principles which, as Stuart M. Brown (1962: 36) alleges, "have no discernible logical relationship to the Categorical Imperative"?
The first position is taken by, for example, Howard Williams (1983), one of the first of the recent English language writers to revive interest in Kantian political philosophy. Williams begins with the point that Kant himself expounds his theory of right as a division of the Metaphysics of Morals. Right and Ethics (Ethik) are presented as two sub-domains of morality; in both cases the subject matter is the application of a priori principles of morals to the empirical world. The Doctrine of Right opens with the Introduction to the Metaphysics of Morals, in which Kant restates the preliminary concepts developed in the Groundwork. Foremost amongst these, of course, is the Categorical Imperative, the "supreme principle of the doctrine of morals", which is here given pithily in the "Universal Law" formulation:
(CI): "Act on a maxim that can also hold as a universal law". (6:226)
Within morality as a whole, the particular sphere of Right concerns the "external relations" of people through their actions (6:230). Whereas ethics looks at the world of inner intentions for actions, what is right or wrongful in the sphere of Right is simply the action itself (or its maxim). Williams explains: "The theory of right outlines ... those maxims of moral philosophy which are capable of being made into external laws, whereas the theory of virtue outlines those maxims of moral philosophy which it is not possible to make into external laws." (60-1)
As we shall see, the fact that these "external laws" dictate only external conduct means that they can be enforced not only through inner promptings of the will but also through external applications of force to other people. So can say: the political or juridical domain is that area where moral principles can be enforced by lawful coercion. Note that this external enforcement complements the internal obligation which duties of right also carry as moral duties: Kant believes that it is always virtuous, as well as right, to obey the law -- to the point that it is morally obligatory to obey even unjust laws made by a legal sovereign (though I won't cover that well trodden ground here).
However, in contrast to the very careful treatment given to the derivation of the categorical imperative in the Groundwork, the principles in the Doctrine of Right are introduced quickly and sparsely. This is the universal principle of right:
(UPR): "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." (6:230)
For Christine Korsgaard (2008: 237), UPR is a "restricted version of the categorical imperative"; and clearly there is, at the very least, a strong resemblance to the universal law formulation. However, Kant does not presents a deduction of UPR from CI, or even discuss the relation of the two. In fact, as Willaschek or Arthur Ripstein (2009) point out in arguing against the "traditional approach" and for the "independence" of the doctrine of right from Kantian moral theory, Kant instead calls UPR "a postulate incapable of further proof". This opens the way to the position held by Willaschek that principles of right are "independent expressions of rational autonomy, on a par with, or at least not derivable from the Categorical Imperative."
One theme from Willaschek's discussion could be useful for setting out our question here. He notes that Kant himself at some points used the universal law formulation of CI to deduce grounds of legitimate coercion (he maintains that Kant later moved away from this approach). Thus in a 1793 lecture transcript Kant argues that a "moral coercive act" is one whose maxim qualifies as a universal law (27:524f; Willaschek 2009:61). For example, one such maxim could be ‘I will coerce others into fulfilling their duties from a promise where this is necessary.’ But, notes Willaschek: "The problem with this is obvious. If the maxim in question would in fact pass the test of the universal law formula, it would be morally admissible to coerce others into observing their promises even in non-juridical cases."
More generally: if the principle of right is derived directly from CI, with the same scope of application, then all of morality is enforceable: the juridical domain is simply the whole moral realm, and there's nothing left for ethics. This leaves us with two alternatives. Either: (a) as Willaschek argues, the principles of right are not deduced from the categorical imperative at all; or (b) they are deduced from not from CI alone, but from CI plus "further postulates" which serve to delimit the scope of the juridical. In other words: if the universal principle of right is indeed a "restricted version" of the categorical imperative, we need to see where the restriction comes from.
3. the law of freedom in a maxim
John Rawls (2000) analyses the "natural law" formulation of the Categorical Imperative as a staged "CI procedure" for testing potential moral laws. Similarly, we might get clearer on the principles of right by looking at UPR as a staged process for lawmakers to test potential "external laws".
(1) frame the maxim under consideration: "I am to do X in circumstances C"
(2) consider it as a "universal precept": "Everyone is to do X in circumstances C"
(3) the test: if this universal precept became an external (and thus -- enforceable) law, could everyone's freedom of choice remain intact?
An external law is rightful if it passes this test. It seems that the content is carried by the notion of freedom applied in the test, which is where we must now turn.
4. external freedom
In the Introduction to the Metaphysics of Morals, Kant's begins with a definition of freedom (in its negative form) which is directly carried over from the Groundwork: "independence from being determined by sensuous impulses" (6:213). Though there is one qualification: we are only concerned now with external relations, with how one person's choice affects the choice of another, rather than their "mere wish", desire or need (6:230). Thus what is at stake in this context is what Kant calls "freedom in the external use of choice" (6:214).
But further in to The Doctrine of Right , in the definition of the one innate or "original right" of freedom, we find a new definition:
"Freedom (independence from being constrained by another's choice), insofar as it can coexist with the freedom of every other in accordance with a universal law, is the only original right belonging to every human being by virtue of her humanity." (6:237)
I will use the label "external freedom" to refer to the second characterisation of freedom as "independence from being constrained by the choices of others". But are these just two different characterisations of the same idea of freedom; or are there actually two distinct ideas of freedom at work in the Doctrine of Right?
To see this we need to understand what it means, in terms of Kantian psychology, for one individual's choice (Willkuer) to be constrained (genoetigt) by the choice of another. Things become clearer here if we relate the language of constraint back to Kant's understanding of causation. I will propose the following gloss. To affect choice or action, as when sensuous impulses affect but do not determine human choice, is to provide not a cause but an influence or, in contemporary philosophical usage, a causal factor. At the other end of the scale, determination (bestimmung) of an action means straightforwardly to cause it.
Between these two, constraint seems closer to determination. It might in fact be the case that they are equivalent, at least in terms of how they relate to causation. Or if we don't want to go that far, I would suggest the following: if a constraint does not in fact cause an action, it at least prevents the choice of certain actions. For example, if you have a choice between three possible actions, and an action of mine removes one of them, leaving you only two available options, then I have constrained, though not fully determined, your choice. If I leave you only one possible action (if such a thing is possible), then my constraint is also a full determination. (In this sense I will approach a determination as a special case of a constraint.)
We can think of a number of ways in which someone's Willkuer may be constrained. For one thing, when Wille determines Willkuer in an action that stems from pure practical reason to act morally this can be seen as an internal constraint. Kant holds that the Willkuer of animals is determined -- hence, constrained -- by their sensuous impulses. But, as we well know, human choice is not so determined. I will assume here (without arguing for it) that in fact Kantian moral freedom can also be understood of freedom from constraint (in the sense above) as well as from determination by sensible impulses.
As we saw above, Kant believes that it is always open, at any time, for a human being to be morally free. An obvious but important corollary is that no other person can ever prevent me from being morally free. Or, to put it another way: no action or choice of another person can cause me to be constrained by sensuous impulses. That is: while another person's choice may affect my choice via influencing my sensuous impulses, it can never constrain my choice in this way.
Another way to make this same point is: no other person can choose my ends for me. But Kant understands a purposive action as involving choosing both an end, and means to achieve that end. Someone's actions can then be influenced not only through affecting their ends, but through affecting the means at their disposal.
In his Force and Freedom, Arthur Ripstein explains Kant's political freedom as: "you are independent if you are the one who decides what ends you will use your means to pursue, as opposed to having someone else decide for you." (2009:33). Strictly speaking, given what we said above, no one else is able to "decide what ends" another pursues; but what others can do is to interfere with your means, or use them for their own ends. For example, if someone takes away a tool that you need to perform a certain action, you are no longer able to choose (rather than simply wish) that action. That is, in the sense defined above, by interfering with your means they have constrained your choice.
We can now clear up the question about the two characterisations of freedom. If this account is right then external freedom is indeed something distinct from moral freedom. I am externally free if I am independent from others interfering with my means. Others can take away my external freedom by interfering with my means; but they cannot take away my moral freedom, which is always only in my own hands.
5. legitimate means
But now there's another question: what does the "my" mean in "my means"? For example, if freedom were simply the ability to use any means at all without interference, then, e.g., it would be an infringement of my freedom for someone to interfere with my using stolen goods -- which seems a rather un-Kantian conclusion.
We can outline two possible responses here; it remains to be seen which is compatible with Kant's position. On the one hand: (a) we might indeed define (external) freedom simply as independence from interference with any means at all. Freedom here means being able to act however I choose, making use of any means available to me. We can still say that it is wrong for me to use stolen goods, and right for someone to constrain me from doing so: only it is not wrong because it restricts my freedom. Rightfulness is not tied to freedom: there can be wrongful uses of freedom, and rightful constraints of freedom.
Alternatively: (b) we might define (external) freedom as independence from interference only with legitimately held means. We will then have to say what makes means legitimately held.
To see the implications of these alternatives, we can try and apply them in the universal principle of right. It's easier to start with the second case:
(b) an action is right if, were its maxim to become a universal external law, no one's legitimately held means would be interfered with.
To formulate UPR in the first case forces us to think a bit more about what we can mean by the means available for an action. For example, we might say: freedom is being able to use, without interference, any things (tools, objects, etc.) that I can physically take up. But in fact this is already a limitation of my available means. For example, as Kant emphasises in his theory of property, we can use things as means in our actions without having any physical connection with them. For example, a merchant can use and profit from goods she owns which are being shipped far away; or make contractual arrangements for future delivery of goods that might not even exist yet. The only general formulation of UPR which didn't already involve a restriction on peoples' use of things as means would be something like this:
(a) an action is right if, were its maxim to become a universal external law, no one's use of any thing at all as a means would be interfered with;
But on that formulation, any action that made use of any thing, unless that thing were a truly non-depletable (non-rival) resource, would be wrong. This is the point that Ripstein (2009:12) makes in observing that the principles of right are about "governing persons represented as occupying space"; or Willaschek (2009:63), somewhat more generally, with what he calls the "principle of impossibility of non-conflicting rights". The subject matter of right is the application of morality to external relations in the empirical world; means of action are things in the empirical world (or agreements based on empirical things), with all the properties of finite things. We get the possibility of conflict over resources: if I am able to use a thing as my means, in many cases (though perhaps not as many as Kant thought: he seems to ignore the possibility of public goods), that means that someone else is not.
This means that freedom in the sense of an absolute licence to use any thing in the world is an impossibility. Or more precisely: it is a possibility for only one person in the world, who is either the sole inhabitant, or keeps everyone else as slaves. It is only possible for more than one person to be free -- and, specifically, for all people to be free, for there to be freedom "in accordance with a universal law" -- if we take freedom in a more restricted sense as the ability to use certain means without interference.
Thus we come to position (b): the definition of external freedom needs to involve some way of specifying what means count as available for my use, that is, what means are legitimately available for me. For Kant, this is a question of what means are rightfully mine: that is, the idea of external freedom, from the beginning, involves property.
6. mine and yours
If this account is right, in order to fully characterise the idea of external freedom we need to jump forward now to the next section of the Doctrine of Right, where Kant expounds his theory of property. Here I will just pick out a few points.
Kant's starting point in the section on private right is a general idea of rightful ownership: "That is rightfully mine (meum iuris) with which I am so connected that another's use of it without my consent would wrong me." (6:243) He then goes on to look at how this concept can be applied to external objects, beginning with a "nominal definition of what is externally mine" in the terms of UPR:
"That outside me is externally mine which it would be a wrong (an infringement upon my freedom which can coexist with the freedom of everyone in accordance with a universal law) to prevent me from using as I please." (6:248)
On the account above, these propositions -- or definitions -- are tautologous. Something is rightfully mine if it would infringe my freedom for another to interfere with its use; but freedom has to be defined just as independence from interference with what is rightfully mine. If Kant is not making a circular argument here, there must be additional principles which ground the idea of rightful ownership. While his account is not easy to follow, we can identify or extrapolate a few such basic principles, some of which are argued for, others assumed.
First of all, following Ripstein (chap. 2), we can interpret the innate right of freedom as self-mastery or ownership of one's person, where "your right to your person is your right to your body". In the first place, then, my rightful means include the powers and capacities of my own body. Note that Kant gives an illuminating characterisation of innate right as "the 'internal mine and thine' (meum vel tuum internum)" (6:237).
I can then supplement these means by acquiring external objects. An infringement of my freedom can be interference with either my own body or with my external property. As Ripstein puts it (43): "your powers can be interfered with in two basic ways, by usurping them or by destroying them." Usurpation means using means without the rightful owner's consent: e.g., non-consensual use of a person's body, or stealing their property. Destruction can involve, e.g., bodily injury or property destruction. Rightful property can be acquired through "original acquisition" in an act of "taking control (occupatio)" of unowned things (6:259); or through consensual transfer. (Both, for Kant, are possible only in civil society, i.e., once a system of law is already in place.) As well as rights over "corporeal objects", Kant also discusses two further categories of external rights involving contracts and "relations of status." Kant devotes some time to arguing that property must be understood in a strong sense of ownership, with full rights to dispose of a thing as the owner sees fit, rather than some form of "usufruct".
7. property and freedom
A quick recap. There are two main elements in the universal principle of right: the idea of a universal law -- so a universalisation procedure for testing potential external laws; and the idea of external freedom. The first element clearly comes from the moral theory of the Groundwork. The question of the relation between Kant's moral and political theories then concerns the second element, the idea of external freedom. We saw that the idea of external freedom is distinct from the moral freedom of the Groundwork; but can it be understood as derivable from the idea of moral freedom under the specific conditions of the domain of right, i.e., of external relations?
Up to a point. First, we saw that external freedom must involve independence in the use of means, as no one can constrain another's ends. Second, we saw that the properties of things in the empirical world already place a limit on the notion of external freedom. But this limit in itself is not enough to determine the definition of external freedom: for that we need further principles to define the legitimate availability of means. In Kant's theory, these further principles are those that identify rightful property. Some of these (there may be more) are principles of bodily self-ownership, original acquisition, and the necessity of absolute ownership. A particular conception of property ownership is at the heart of Kant's idea of political freedom.
I think that none of these ownership principles derive from wither the categorical imperative or elsewhere in the moral theory of the Groundwork. As I can't go into detail here on Kant's property theory, unfortunately I have to leave this claim with no full demonstration but only the following sketchy remarks. The criterion of universalisability in the universal principle of right is not sufficient to lead us to Kant's ownership principles: quite different property systems (for example, systems of usufruct ownership, where no one has rights of alienation) could also be instituted as universal laws. The idea of a universal law tells us that, (a) however our system of right defines what a person can legitimately use as means, it must be a principle that applies to all people alike. And that (b) given the nature of the empirical world, of finite spatial objects, our system of right must account for the possibility of conflicting demands for use of a thing. Neither of these demands uniquely identify Kant's system of property, or his idea of freedom.
8. Coercion
We have defined external freedom as independence from interference with one's legitimate means. However, interference with bodies or things through immediate usurpation or destruction -- that is, through force -- is only a small part of the mechanics of unfreedom. Can Kant's theory account for coercion in a broader sense?
We can distinguish a few senses of coercion. Just two are: (a) coercion as force -- actual interference with a person's legitimate means; and (b) coercion as threat -- of future acts of forceful interference conditional on a person's present actions. Leaving aside whether these two categories describe coercion in all its forms, is there a Kantian account of threat?
In the recent literature on coercion starting with Robert Nozick (1969), a threat is a proposal in which: A proposes an action X to B; if B refuses to do X, A performs an action which makes B worse off than she would be "in the normal and expected course of events". A good deal of the subsequent discussion on this theme then focuses on the definition of the comparative "baseline" (Nozick's "normal and expected course of events"). We cannot simply define a threat as involving making someone worse off than they are at present, as events might "naturally" have led to the situation deteriorating, even without any intervention. An assessment of threat then needs to suppose a hypothetical baseline state against which outcomes are measured.
To follow Alan Wertheimer's analysis, it makes sense to recognise that there is no one universal definition of the appropriate baseline for assessing all threats. There may, e.g., be "phenomenological" baselines defined according to the expectations of the different parties; or "moralised" baselines defined according to whether the proposal makes the coercee worse off than what is considered the morally right state of affairs. On a rights-based moral theory this becomes: "we set B's baseline by what A has a right to do". (1987: 215)
With some adaptation, a Kantian account of coercion can be approached as such a moralised account of threat. The appropriate baseline can be defined in terms of the concept of external freedom. There is a necessary adaptation: Kantian theory of right is not about what makes a person better or worse off, but whether her external freedom stays intact. Incorporating that, we can analyse a (Kantian) threat as follows: A proposes an action X to B; if B does not carry out X, A will carry out the threatened action T; T is an action which interferes with B's legitimately held means.
9. the circle of right
Kant holds that people can acquire new rights to external things through consensual agreements: for example, through transfers of property. But if we define coercion as above then there seems to be a danger of circularity: in order to assess coercion we need to know what means people rightfully own; but we may not be able to know that without assessing coercion.
In contemporary liberal theory Nozick (1974) employs one type of solution to this problem: we might suppose that a person's rightful powers must have been rightfully acquired at every historical step; in theory we could trace back a full history of each transfer to primitive origins where there were only innate powers and "original acquisitions" of unclaimed things. Of course, this story is very much "in theory": not only does no one suppose it can ever actually be told, but no one really thinks that its supposed truth or otherwise makes much difference to justice claims in the present.
Does Kant take an approach of this kind? It is compatible with the theory of contract and consent developed in The Doctrine of Right. As Ripstein puts it (p116): "In the case of a transfer of property ... there is a right ... which is transferred ... and the transfer itself does nothing to alter either the form or the matter of right in any way." There is no way that the legitimacy of a property holding can change through transfer: if I am a rightful owner and transfer my property to you, your ownership is now rightful, whoever you are. Nor does there seem any other way, in Kant's theory, that the legitimacy status of a property right can change over time: for example, Kant's absolute sense of property as alienable ownership includes the right to neglect or damage one's property. Then there seems nothing more to say about the legitimacy of property accumulated through transfers than whether the historical chains involved were rightful.
On the other hand, Christine Korsgaard (2008: 244) reads Kant's position as: "we simply take it for granted that, generally speaking, what people now have is their property. And we try to ensure that, from here on in, transactions will be legitimate and just."
Kant himself is elusive on this point. For example, in the essay "Theory and Practice" he famously pays attention to the lack of any historical "original contract" for the establishment of a state, which leads him to a "hypothetical" version of social contract theory. He also notes the parallel question of the absence of a history of legitimate property transfers. Then he quickly drops it: "we leave aside the question of how anyone can have rightfully acquired more land than he can cultivate with his own hands (for acquisition by military seizure is not primary acquisition), and how it came about that numerous people who might otherwise have acquired permanent property were thereby reduced to serving someone else in order to live at all." (Reiss 1970:78)
10. liberalism
We can understand Kant's political theory as a form of liberalism. There is an important distinction to be made with, say, a Lockean theory based on natural property rights, in which individuals transfer to the state a coercive authority which they already possessed in the "state of nature". For Kant, external property rights are impossible without a juridical order, and the need for human beings to realise their freedom through property ownership in fact presents them with a duty to institute government. Without government, we could have ownership only of our own bodies, and then "freedom would be depriving itself" of legitimately available means. (6:246).
At the same time, the outcome of Kant's position is similar to Locke's. Here too, legitimate coercion is restricted to the defence of property; and legitimate coercion is the province of the state; thus the domain of law and government is limited to the protection of individuals' "mine and thine". The core of Kant's political position is thus a classical form of liberalism. But the way this delimitation of the domain of right is derived is specifically Kantian, involving the theory of the human will and its relation to sensual causation, and of the meaning of freedom in a finite world. While Kant never makes all of the links explicit, we can construct the following argument:
(1) Individuals are always able to be morally free; that is, it is always open for the pure will to override sensuous impulses.
(2) This means that, while other people can affect our desires and ends through our impulses, they cannot constrain our choice in that way.
(3) The only way other people can constrain our choice is by interfering with the means we have legitimately available, in our bodies or in the external world, to pursue our ends.
(4) To coerce someone involves constraining their choice.
(5) Thus coercion involves interference with another's legitimate means.
(6) Coercion is only legitimate where it resists the coercion of another: is a "hindrance of a hindrance". That is: protects individuals' means from interference by others, by interfering with their means.
(7) The juridical realm is that area where coercion is legitimate.
(8) Thus, the juridical realm -- the domain of the state -- is the protection of individuals' rightful means: their bodies, and their property.
Works by Kant: Citations to the Gesammelte Schriften (Berlin: Walter de Gruyter, 1902 --). Translations used are:
Groundwork of the Metaphysics of Morals, trans. by Mary Gregor, 1997, (Cambridge: Cambridge University Press)
Metaphysics of Morals, trans. by Mary Gregor, 1991, (Cambridge: Cambridge University Press)
"On the Common Saying: This May Be True In Theory, But It Does Not Apply In Practice"; and "What Is Enlightenment?"; in Political Writings, trans. by H.B. Nisbet, ed. by Hans Reiss, 1970, (Cambridge: Cambridge University Press)
Secondary writings:
Allison, Henry. 1990. Kant's Theory of Freedom (Cambridge: Cambridge University Press)
Brown, Stuart M., Jr. 1962. "Has Kant a Philosophy of Law?", Philosophical Review 71: 1, 33-48
Korsgaard, Christine M. 2008. "Taking the Law into Our Own Hands", in The Constitution of Agency - Essays on Practical Reason and Moral Psychology (Oxford: Oxford University Press)
------ 2009. Self-Constitution (Oxford: Oxford University Press)
Nozick, Robert. 1969. “Coercion”, in Philosophy, Science, and Method: Essays in Honor of Ernest Nagel, ed. by Sidney Morgenbesser, Patrick Suppes, and Morton White (New York: St. Martin's Press)
------ 1974. Anarchy, State, Utopia (New York: Basic Books)
Rawls, John. 2000. Lectures on the History of Moral Philosophy (Cambridge, Ma.: Harvard University Press)
Reath, Andrews. 2006. "Legislating the Moral Law", in Agency & Autonomy in Kant's Moral Theory (Oxford: Clarendon Press)
Ripstein, Arthur. 2009. Force and Freedom (Cambridge, Mass.: Harvard University Press)
Wertheimer, Alan. 1987. Coercion (Princeton: Princeton University Press)
Williams, Howard. 1983. Kant's Political Philosophy (Oxford: Blackwell)
Willaschek, Marcus. 2009. "Right and Coercion: Can Kant's Conception of Right be Derived from his Moral Theory?", International Journal of Philosophical Studies, 17: 1, 49-70
http://www.youtube.com/watch?v=AkK2lrJDNeg
