Liberalism: Keynesian, Cold War, Rawlsian…and Cosmopolitan

At the U.S. Intellectual History blog there’s a post by Andrew Hartman prompted by Kwame Anthony Appiah’s essay in the New York Review of Books on Rawls’ now published senior thesis. Hartman concludes in part as follows:

“Appiah, famous in his own right for his work, among others, Cosmopolitanism: Ethics in a World of Strangers, sees Rawls as a leading theorist of cosmopolitanism. I had never understood Rawls as such. I was more inclined to think of Rawls as my colleague Mike O’Connor does in his manuscript-in-progress on democratic capitalism—as perhaps the last leading Cold War liberal thinker. By way of conclusion, I would like to pose questions. Does it make sense to think of Rawls as a cosmopolitan thinker?” [….]

Herewith my comment to the post:

I don’t think it’s very helpful to categorize Rawls as “the last leading Cold War liberal thinker,” or even an exemplary “Cold War liberal thinker.” The secondary literature on Rawls is enormous, and much of it reflects the biases and ideologies (not always in a fatal way, mind you) of its authors. And Rawls’ works are themselves in part responsible for the wide range of interpretive positions owing to occasional inconsistencies, ambiguities, contradictions, and so forth, in spite of the brilliance of the works in toto. Here we might agree with Gerald F. Gaus in his invaluable analytical survey, Contemporary Theories of Liberalism (2003):

“The interpretation of Rawls’s texts is notoriously difficult [as is often the case with any decent philosopher]; at different points Rawls appears to reinterpret his earlier statements, and at times apparently affirms competing interpretations of his views. It is not difficult to find in Rawls’s work passages that support widely different interpretations.”

For the moment, let me say that it is true that to the extent there exists (or that one can find) “cosmopolitanism” strains in Rawls’s thought (and I believe there to be some) it reflects the influence of Kant, nonetheless, as Onora O’Neill has made plain,* in some respects at least, Kant was more of a cosmopolitan than Rawls, at least the Rawls of Political Liberalism (1993) and perhaps more clearly of The Law of Peoples (1999) as well. The “Kantian constructivism” of the former work “identified the reasonable with the public reason of fellow citizens in a given, bounded, democratic society [i.e., like ‘ours’].” Thus, here at least, Kant is more cosmopolitan than Rawls, for Kant’s ethical method was indeed cosmopolitan while Rawls’ is “implicitly statist.” Now we might also bear in mind the audience or rhetorical character of Political Liberalism insofar as it appears to deliberately restrict itself to members of “actually existing democratic societies” so to speak, and this certainly represents an endeavor to respond if not appeal to communitarian critics like Sandel, perhaps reflecting, as some have argued, a needless concession to the communitarian argument generally. However, things may not be quite as clearcut as O’Neill makes them out to be if we focus on both the rhetorical and rational character of the argument. After all, Rawls aims for a political conception of justice that is “freestanding,” that is, not (deductively) dependent in its presuppositions, axiomatic assumptions or premises upon a commitment to any specific comprehensive worldview, and in so doing it aims to ground itself in reason as such, be it as “the rational” or the “reasonable.” Now Rawls does appeal here to the public culture of a constitutional regime because he believes its fundamental principles, say, autonomy, liberty, equality, fairness, or welfare, are core Liberal political principles and values and thus, as such, are capable of transcending their possible justification in any particular comprehensive view, in other words, they need not depend on any such worldview, indeed, these differing comprehensive views may in fact possess the intellectual resources from which one might (or better: should) reason to an endorsement of the aforementioned principles or at least come to an appreciation of their indispensable and basic political value or worth (by analogy: consider the manner in which people of differing worldviews or ‘cultures’ across the globe have come to endorse at least some human rights, or the notion of human dignity, even if they’ve come to that conclusion, that is, even if they’ve reasoned from different conceptual premises or values and beliefs seemingly unique to their specific worldviews in a manner that permits or encourages them to endorse the inviolable dignity or inherent worth of the individual human being).

Now it will not do to dismiss Rawls’ argument here because the political salience or significance of the above values and principles were historically and politically generated from or common to Liberalism, for that would be to commit the genetic (or ‘reductive’) fallacy, if only because these principles are in fact generalizable or universalizable** regardless of their association with various thinkers and theories in the tradition of Liberalism. Or, with Gaus, we can point out that principles like (negative and positive) liberty, equality, welfare, and (moral) autonomy “are not necessarily embedded in any comprehensive doctrine [or ‘worldview,’ the term I prefer],” yet those of differing comprehensive doctrines are capable, presumably, of agreeing to them (or at least they should insofar as they share a commitment to the value of rationality and the need to be reasonable). Insofar as Rawls is speaking to citizens of constitutional democratic regimes, the rhetorical purpose of his argument is “consensual,” i.e., he aims to show that everyone “has reason R to accept belief X” (we share a reason for endorsing X). And insofar as his argument is capable of appealing to those of different comprehensive doctrines or worldviews, that is, irregardless of whether or not one is a member of a democratic society, it makes use of a “convergence” argument which seeks to demonstrate that “we have different reasons for endorsing X, though we all have some reason for endorsing it” (I’m relying here on Gaus who, in turn, is drawing upon Fred D’Agostino for the distinction between ‘consensus’ and ‘convergence’ arguments). O’Neill appears to be focusing on the former argument to the neglect of the plausibility or possibility of Rawls making this latter argument. Those who reject the conception of persons as being free and equal (in at least the Kantian moral if not the metaphysical sense) and, no less importantly, lack a commitment to being reasonable and rational, are outside the rhetorical scope and attraction of Rawls’ argument. As Gaus explains,

“Rawls argues that justice as fairness is a justified political conception because it articulates the requirements of the person and society that all reasonable citizens in our democratic societies share. However, Rawls does not believe that this exhausts ‘full’ and ‘public’ justification [and here we get closer to Kant and his notion of practical and public reason in the sense that the scope of ethical reasoning encompasses an unbounded plurality of agents]—citizens draw upon their full range of beliefs and values and find further reasons for endorsing the political conception. Thus ‘overlapping consensus’ constitutes a convergent public justification, drawing on our various ‘comprehensive doctrines.’”

The Rawls of A Theory of Justice (1999 ed.) does not make ethical arguments or even assumptions that can be dismissed as “statist,” and the extent to which it relies on Liberal principles in the broadest sense (in conceiving ‘justice as fairness’), these are universalizable or generalizable, as they are in Kant’s case, and thus Rawls’ argument can be seen as having cosmopolitan implications, even if he himself did not tease these out in any sustained manner. To the degree, in other words and for example, that he relies on the Kantian notion of “free and equal” persons he falls within the ambit of the cosmopolitan tradition going back to the Stoics (as does Kant himself for that matter). Now I’m not here going to attempt any critique (as Gaus himself does) motivated by the fact that Rawls concedes his notion of “justice as fairness” is only one possible liberal conception, granting, in other words and as Gaus points out, “that there are diverse interpretations of the basic concept of a liberal political order,” a plurality that at bottom is inevitable if only because of the abstract character and level of the fundamental principles and values found in the Liberal tradition, indeed, this also accounts for the different “kinds” of Liberalism: (to some extent) Hobbesian, Lockean, Millean, Nozickian, Rawlsian, etc.

In The Evolution of Rights in Liberal Theory (1986), Ian Shapiro correctly notes that Rawls’ ideological impact, while more complex than that of Robert Nozick,

“can be summed up by saying that his is the natural response of a liberal who has read Pigou and Keynes seriously. He appeals to those who believe in the desirability, efficiency, and justice of capitalist markets, recognize that they may not always function well and may generate serious inequities for some, and want to find efficient ways of addressing those inequities without altering the essential nature of the system. The ambiguous moral status of Keynesianism and welfare economics has always inhered in the fact that they appeal to the short-term interests of the disadvantaged (such as unemployed workers and firms on the edge of bankruptcy during recessions) by ensuring subsistence, creating employment, and expanding credit, yet these policies are geared in the medium term to sustaining the system which generate those very disadvantages—hence the ironic force of Joan Robinson’s quip that the one thing worse than being exploited is not being exploited at all.”

Now while this may account for the ideological appeal of Rawls, it does not do his argument full justice in as much as he was capable of envisioning, like not a few contemporary democratic socialists and Social Democrats, and even some (especially self-described ‘analytical’) Marxists (e.g., Roemer), the possibility of non-capitalist or socialist markets (hence the debate on the Left about ‘market socialism’). Rawls speaks (in his Lectures on the History of Political Philosophy (2007)), for instance, of the “illuminating and worthwhile view” of “liberal socialism,”*** enumerating its four basic elements and referring in a note to John Roemer’s “Liberal Socialism (1994),” which is not, it turns out, a title of any of Roemer’s books (if I recall correctly, this error was confirmed in correspondence with Samuel Freeman). The volume Rawls probably intended to cite is Roemer’s A Future for Socialism published the same year (it is no less a delightful Freudian slip for all that!). Here once more we might appreciate why it may be more than a tad misleading to view Rawls as a “Cold War liberal” (there are, however, not a few of that species: for a discussion of the ‘intersection between the various approaches to rational choice liberalism and Rawls’s theory of justice,’ please see S.M. Amadae’s Rationalizing Capitalist Democracy: The Cold War Origins of Rational Choice Liberalism (2003)).

With regard to political cosmopolitanism proper, particularly insofar as it includes a theory of distributive global justice, there are at least several theories of global justice that are avowedly inspired by Rawls even if they depart from him in some or significant respects (e.g., Charles Beitz’s Political Theory and International Relations (1979) Gillian Brock’s Global Justice: A Cosmopolitan Account (2009)). Martha Nussbaum’s cosmopolitanism, which is intellectually adverse to the social contract tradition, remains in some ways beholden to Rawls as she herself has acknowledged. This alone should provide sufficient presumptive reason for refusing to exclude Rawls from the rubric of cosmopolitanism, whatever the shortcomings from vantage points provided by more vigorous—or “robust” as we say today—theories of same (e.g., Simon Caney’s Justice Beyond Borders: A Global Political Theory (2005) and Nussbaum’s Frontiers of Justice: Disability, Nationality, Species Membership (2006), and several books by Thomas Pogge).

*In her essay, “Constructivism in Rawls and Kant,” in Samuel Freeman, ed., The Cambridge Companion to Rawls (2003): 346-367.

**It might help to recall, again with O’Neill, that abstract, “universal principles need not mandate uniform treatment; indeed, they mandate differentiated treatment. [....] Even principles that do not specifically mandate differentiated treatment will be indeterminate, so leave room for differentiated application.” Uniformity is, in other words and in the first place, a matter of content: “For example, principles such as ‘Each should be taxed in proportion to ability to pay,’ or ‘Good teachers should set work that is adjusted to each child’s level of ability’ respectively prescribe and recommend universally, but both will require varied rather than uniform implementation in a world of varying cases.” In addition, “even when universal principles specifically prescribe some degree or aspect of uniformity of action or treatment, they underdetermine action, so must permit varied implementation.” Finally, universal principles of action “hold universally only relative to some domain of agents.” In sum, it follows that, as Kant himself appreciated and many of his critics have not (among those critics O’Neill includes contemporary ‘particularists’ like ‘communitarians, virtue ethicists, and some “radical” feminist writers’), “the application of principles to cases involves judgement and deliberation.” Thus principles serve as “side-constraints (not algorithms) and can only guide (not make) decisions.” (For a fuller treatment, see her book, Towards Justice and Virtue: A Constructive Account of Practical Reasoning (1996)). One might compare the manner in which (indeterminate) Platonic Forms, as abstract and general ideas, are intuitively and dialectically realized (at least according to what T.K. Seung terms ‘bedrock’ Platonism) “in many different ways.” Constructivism is therefore essential to ethical reasoning in both Plato and Kant, assuming the form and function of utopian thought in the sense outlined by William A. Galston in Justice and the Human Good (1980): “First, it guides our deliberation, whether in devising courses of action or in choosing among exogenously defined alternatives with which we are confronted. Second, it justifies our actions; the grounds of action are reasons that others ought to accept and—given openness and the freedom to reflect—can be led to accept. Third, it serves as the basis for the evaluation of existing institutions and practices.”

***A work that intriguingly combines Rawlsian ethical conceptions and reasoning with Marxism that is a must-read, and further evidence of the wide appeal and implications of Rawls’ work beyond the label of “liberalism” (Cold War, Keynesian, what have you), is Rodney G. Peffer’s (rather neglected) Marxism, Morality, and Social Justice (1990).