Rawls on the embedded self: Liberalism as an affective regime

Introduction

As part of a broader turn within the discipline, political theorists have increasingly come to recognize the central role of affect in social and political life; indeed, many scholars, coming from a number of distinct theoretical traditions, have drawn our attention to the importance of the emotions to the tradition of the history of political thought as well as to normative political theory. ¹ This attentiveness to affect is often cast as a break with earlier, primarily Enlightenment-inspired liberal approaches towards politics. On such a view, liberalism and liberal theorists have a deeply ingrained tendency to either neglect or marginalize the emotions, dismissing affect as inconsequential or as potentially dangerous. ² The view of liberalism as a project committed to distancing politics from the passions is often characterized as anchoring political life in deliberative public reason and a neutralist state. Consequently, according to such a portrayal of contemporary liberal theory, emotions are said to have no place in the public sphere, while proceduralist institutions abstract away from the affective attachments, now cast as private preferences, of individuals qua citizens. Emotions, so the argument goes, ought not to have any place in the consideration of liberal principles of justice or in the design of political institutions. Here we challenge this prevalent view. We argue that no less a liberal theorist than John Rawls is deeply attentive to the place of emotions in his account of liberalism. In doing so, we hope to contribute to upending the conventional view of liberalism as affect-blind, in order to encourage a more nuanced reading of the liberal tradition more generally.

Rawls' unreasonable rationalism?

Casting Rawls as a theorist sensitive to affect and emotional experience may strike many readers as counterintuitive. Indeed, Rawls' work has been frequently criticized for epitomizing some of the deepest problems of contemporary liberal theory, as a result of the emphasis on rationalism and reasonableness in his account of liberal justice. Whether a residual of his earlier attempt to bring rationality and justice together, or a consequence of the explicitly neo-Kantian foundations of his political thought, Rawls has frequently been taken to be guilty of the propensity of contemporary theory to ignore or marginalize the emotions. ³ Far from attending to the embedded and affective dimensions of human nature, so this argument goes, Rawls is an exemplar of the misguided liberal propensity to articulate a politics of illusory neutrality grounded in a deracinated and ‘unencumbered’ understanding of the individual.

Indeed, it is precisely this perspective on Rawls that has informed a number of prominent critiques of Rawls' account of justice. In order to situate our own reading, it is helpful to sketch this prevalent interpretation of the Rawlsian project, from its beginnings in the initial critical response to A Theory of Justice (hereafter TJ) to more recent challenges from communitarian, feminist, Marxist, and realist perspectives. This important, and by no means marginal, understanding of Rawls' project, which focused primarily on the conceptual architecture of the first two parts of TJ, characterized much of the early reception of Rawls' work, on the sides of both critics and defenders. In this reading, the main task of TJ was to offer a conception of justice that ‘generalizes and carries to a higher level of abstraction’ the central insights of the liberal contractual tradition, in order to justify fundamental principles to govern the institutions and basic structure of society. ⁴

To generate such principles, Rawls asks us to imagine an original position of rational and mutually disinterested individuals tasked with choosing principles of justice, under the condition of a veil of ignorance requiring them to bracket morally arbitrary facts, such as their individual characteristics and talents, as well as knowledge of their own actual position and status within society. Under such circumstances, Rawls suggests that the parties would select two principles of justice: legitimate political institutions must secure the basic liberties of persons in the most extensive manner possible, and they must arrange social and economic inequalities so that they are conducive to both fair equality of opportunity and to the maximum improvement of the socioeconomic condition of society’s least advantaged. ⁵ The latter principle, known as the difference principle, specifically tied TJ’s commitment to relative economic egalitarianism to the rationalist assumptions of decision theory, suggesting that under the conditions of uncertainty meant to model an appropriate moral situation, individuals would choose a distributive paradigm that would severely curtail inequalities, only allowing them if they benefited the standing of the representative least well-off individual. Indeed, Rawls appeared to claim that the general project of TJ was itself ‘a part, perhaps the most important part, of the theory of rational choice’, suggesting that a central aspect of his project was to marry rational self-interest maximization and justice. ⁶

Picking up on Rawls' emphasis on rationality in his attempt to articulate a decision procedure for principles of liberal justice, a number of works in the early critical literature focused on whether Rawls principles of justice were in fact congruent with the assumptions of rational choice. Scholars, such as the utilitarian John Harsanyi, as well as more sympathetic writers such as James Fishkin and Allan Buchanan, criticized various aspects of Rawls' rationality assumptions, asking whether they in fact would lead to the adoption of Rawls' conception of justice. ⁷ Would the difference principle in fact be the outcome of a rational decision procedure under the original position’s conditions of uncertainty? Did Rawls' argument rely upon an implausible model of rational choice, or did his account smuggle in dubious assumptions about the significance of the diminishing marginal value of Rawlsian primary basic goods? Key features of this broader perspective and critical reception of Rawls' project concern his purported insistence on the need to abstract away from the particular features and social positions of individuals, as well as his attempt to graft aspects of rational choice theory to his account of the original position.

It was precisely this focus on rationality and abstraction that motivated some of the strongest critiques of those who dissented from the Rawlsian project. Early on, the libertarian theorist Robert Nozick raised doubts about the implications of the original position, suggesting that the difference principle generated by this device of representation violated the project’s purportedly Kantian commitment to the integrity of the person. ⁸ This was because it appeared to take a person’s individual attributes and talents as a collective good, thereby seeming to treat individuals as mere means. In a similar vein, communitarian theorists, such as Michael Sandel, objected to what they took as an unrealistic and highly problematic concept of the human subject at the centre of Rawls' approach. ⁹ According to Sandel, as a result of the project’s problematic Kantian legacy, Rawls' attempt to develop a deontological liberalism presupposed an unencumbered ‘noumenal’ moral agent and with that a conception of the individual that deeply misrepresents our status as socially embedded and embodied beings.

Feminist critics, such as Susan Okin, Iris Young, and Amy Baehr, saw cause for concern in what they saw as Rawls' tendency to privilege reason and abstraction in his account of justice. ¹⁰ If the original position as a device of representation was supposed to model features relevant to the subject of justice in liberal societies, the emotions, gender, and the family were curiously absent from the Rawlsian picture. For Okin in particular, the difficulties apparent in Rawls' project were largely a result of his Kantian assumptions regarding rationality and autonomy, leading him to neglect the role of the ‘human qualities of empathy and benevolence’ in establishing principles of justice. ¹¹

Drawing inspiration from the feminist critique of Rawls, G.A. Cohen criticized Rawls' project from a Marxist perspective, arguing that there is a tension between the sense of justice that would characterize a just society and the appeal to self-interest implicit in Rawls' account of the rationality of his principles of justice. Rather, pace Rawls, according to Cohen a just society requires an ethos of justice that supports equality enabling choices and such an ethos must go beyond narrow self-interest. ¹² Underlying Cohen’s critique were deeper worries regarding the motivational resources of Rawls' project and whether a truly just society would need to rely upon rational incentives – such as the permissible inequalities of the difference principle – to motivate just action.

Most recently, scholars self-identifying themselves with a ‘realist’ undercurrent in political theory have raised similar criticisms regarding the content and legacy of Rawls’ project. Thus, William Galston has attacked the apparent inattentiveness to the emotions in the ‘high liberalism’ that he claims has come to dominate much of contemporary theory. ¹³ For Galston, Rawls’ project as a whole exemplifies this pervasive lack of psychological and motivational realism that he finds endemic to the political moralism characteristic of such approaches. This is because, as Galston asserts, Rawls and his fellow ‘high liberals’ generally ‘leave out an entire dimension of the human psyche – namely, the passions and emotions’ from their accounts of political justice. ¹⁴ According to the realist perspective, such an omission constitutes a serious and enduring drawback of the Rawlsian project, one that can only have lead to a highly distorted vision of politics.

But is Rawls quite the arch-rationalist that these prevalent readings seem to suggest? Does Rawls in fact provide us with an account of justice grounded in an abstract, unencumbered self, in which the embodied and affective dimensions of human experience have no place? Contrary to the prevalent view implicit in these critiques, we argue that Rawls is in fact attentive to the role of emotions in the political lives of citizens and that affect has a central place in his understanding of the liberal regime.

Rawlsian sentimentalism

In line with the reading that we propose, in recent years there has been a notable increase in scholarly attention to the presence of the emotions in Rawls' political philosophy. ¹⁵ As we have already noted, this is emblematic of a more general turn towards affect in political theory. Michael Frazer and Sharon Krause, in particular, are on the vanguard of this reappraisal of Rawls. In this section, we will focus our attention on an important contribution by each thinker: Frazer’s ‘John Rawls: Between Two Enlightenments’ and Krause’s ‘Desiring Justice: Motivation and Justification in Rawls and Habermas’. ¹⁶ While both authors are exactly right to emphasize the centrality of affect in the landscape of justice as fairness, it is our contention that Frazer and Krause only tell a part (albeit an important, under-appreciated part) of the story. In what follows, we hope to give a more thorough and complete picture of Rawls' thinking on the role of affect in political philosophy and in political life.

What, then, are the merits and limitations of this more recent account? At first glance, it seems to be the case, according to Krause and Frazer, that Rawls treats the emotions as a kind of post hoc justificatory mechanism: citizens look back upon the principles (and institutions) that govern them, hopeful that these social structures are worthy of their assent as free and equal persons. Reflection, in other words, is not exclusively a matter of rational reflection, of the hierarchical privileging of our true, noumenal self over our contingent emotions and attachments. On this sentimentalist account, our chosen principles of justice have normative authority because of their compatibility with both the exercise of reason and the complex emotional psychology characteristic of human beings; again, our concern for justice is both a ‘display [of our] independence from the accidental circumstances of our world’ and of our ‘natural sympathy with other persons and an innate susceptibility to the pleasures of fellow feeling’. ¹⁷ Frazer and Krause are thus (rightly) engaged in the attempt to ‘re-embody’ Rawls: the principles of justice are justified (in part) by virtue of their compatibility with the kinds of emotions citizens actually happen to have. There is no sense, in other words, that the possibility of justice depends on the transcendence of our empirical, contingent, affective selves.

Of course, there are important, insuperable limits to the justificatory work that the emotions can do. This, according to Frazer, is a conspicuously Humean moment in Rawls' political philosophy:

The fact that we can have higher order moral sentiments – that we can approve or disapprove of our own of our approval and disapproval – allows for a process of reflection in which the mind as a whole repeatedly turns on itself as a whole, and winnows out those sentiments which cannot pass the test of reflection. ¹⁸

This reason-emotion dialectic is captured by the Rawlsian notion of reflective equilibrium: the elaboration and adaptation of justice principles is, according to this sentimentalist re-reading of Rawls, the product of the mutual adjustment and enlargement of reason and affect as we collectively argue and reflect upon what a just political community ought to be like. ¹⁹ The outcome of this process must be acceptable to our thoroughly interrogated, and therefore rightly authoritative, reasons and emotions. Not any reason can function as the justificatory basis of a given principle, law, or policy: reasons must respect the freedom and equality of all citizens. And the same goes for the emotions: only those politically constructive emotions, such as our natural sympathy for, and empathy with, our fellow citizens, are the legitimate basis of political obligation. This is an important point worth emphasizing: on this sentimentalist re-reading, it is our natural sense of empathy that motivates our attachment to legitimate, other-regarding principles of justice. We are attached to principles of justice, in other words, because they reflect our natural empathetic inclinations towards our fellows. Such sentiments then receive institutionalized expression in the basic structure of society, that is, in the principle-guided institutional structure that protects the freedom and equality of all citizens. This is why the sense of justice itself is rightly thought of as an affective attachment.

Citizens’ commitment to the principles of justice (i.e. the sense of justice itself) is thus conceived (in part) as the ‘reflective outgrowth of basic human emotions’. ²⁰ This is the main thrust of Krause’s argument too: ‘To have a conception of the good […] is to have an affective attachment to it or a desire to realize it; hence when we are rational we are also desiring’. ²¹ Elsewhere, Krause describes the sense of justice itself as an ‘affective attachment’ to the idea of being a just person and to the good of justice. ²² According to this reading of Rawls, citizens in a just liberal regime are thus appropriately understood as desirous of justice. What both Frazer and Krause thus recognize is that Rawls' account of justice is implicated in the politics of affect, and that the emotions always play a central role in providing the motivational resources necessary in any stable, legitimate political community. This reading therefore goes beyond earlier engagements with Rawls' project, which emphasized the exclusive place of reason and rationality in the choice of justice principles, thereby neglecting the central role of the emotions in Theory. We are reminded of Hume: ‘Human nature [is] compos’d of two principal parts, which are requisite in all its actions, the affections and understanding […] these two component parts of the mind […] [are] uncompounded and inseparable’. ²³ This is a very important point: what Frazer and Krause both hope to show is that this emphasis on affect is present in Rawls' thought from the very beginning precisely because of the ultimate inextricability of reason and affect. After all, the desire to live with others on just terms, says Rawls in TJ, ‘exercises a natural attraction upon our affection’ and so the principles of justice themselves are conscientiously designed to be ‘continuous with our natural sentiments’. ²⁴ This ought to correct our initial (post hoc justificatory) impression of the role of affect in Rawls' political philosophy: far from offering a retrospective defence of the (mere) compatibility of Rawls' project with the affect turn in political theory, Frazer and Krause rightly emphasize how an attentiveness to the role of emotion is present at the start of justice as fairness. In the section that follows we will take this important insight even further.

It is patently the case, then, that the principles of justice are designed with explicit reference to feeling, sentiment, and affect – with emotions that citizens actually happen to have, such as sympathy, empathy, care, trust, fellow-feeling, and even love. Indeed, in a revealing moment in TJ, Rawls goes so far as to call justice as fairness a ‘theory of the moral sentiments’. ²⁵ This lends Rawls' theory an additional basis of legitimacy and (by extension) of stability (for the right reasons). Of course, stability occupies a central place in the landscape of Rawls' thought: beginning with TJ, Rawls stresses that stability is an essential part of justice – that choosing a conception of justice that is not stable is not justice. ²⁶ In fact, this concern with stability is an under-appreciated part of Rawls' argument against utilitarianism: utilitarian principles involve strains of commitment that exceed the capacity of human nature. That is, Rawls argues that we cannot accept a lesser share of primary goods in the name of the good of the whole: human nature itself is incompatible with utilitarian distributive outcomes. ²⁷

And so, we need to come up with a set of principles that beings like us, in circumstances like ours, can adhere to in perpetuity. We must only consider principles and institutions that are practically possible and therefore stable given our moral and psychological nature. Hence, Rawls' concern with affect – both as a limit on our capacity to obey justice principles, and as a source of our motivation to comply with them – is present from the very beginning. We care about the principles of justice, and are motivated to act according to their dictates, precisely because they take into account and reflect our care for fellow citizens and our affective attachment to the ideal of a just political community. ²⁸

Of course, this concern with stability is precisely what motives Rawls to revise justice as fairness: the later doctrine, political liberalism, is necessary, so Rawls' thinking goes, precisely because of the instability to which justice as fairness would be subject in a comprehensively plural society. ²⁹ Given the inevitable fact of irrevocable pluralism – and the problems of legitimate coercion that arise as a result – Rawls wants us to think of his political conception of justice as ‘freestanding’ – not as derived from a particular comprehensive point of view, but rather as embedded within citizens’ reasonable religious, moral, and philosophical doctrines. ³⁰ Of course, our purpose, here, is not to decide whether this move is convincing or philosophically coherent; many suspect that it is not. ³¹ Rather, given our purposes, it is important to note that, in the later doctrine, Rawls still wants to harness the affective resources of citizens. Indeed, the fact that the political conception of justice is now embedded within citizens’ deeply held (but ultimately reasonable) comprehensive doctrines constitutes another affective basis of motivation and of stability; that is, we care about the principles of justice precisely because obedience to them is a constitutive expression of our commitment to much more existentially robust comprehensive doctrines. ³²

However, as we have already indicated, this is only a part of the story about the role and place of the emotions in Rawls' political philosophy. On our view, the emotions do not merely function as a source of motivation to comply with justice principles. Rather, what we hope to show in the next section is that a liberal regime is always already concerned with the affective dimension of citizenship, and that justice as fairness in particular is better understood when we incorporate this concern for the relationship between politics and affect in an even more robust manner.

Institutions, affect, and the embedded self

While the recent turn towards the affective dimensions of Rawls' project has focused on how Rawls does recognize the need for desire, sentiment, and emotion as motivational resources in his account of liberal justice, our reading of Rawls wishes to push further in suggesting that at the core of Rawls' project is an even thicker conception of an embedded and affective subject. To demonstrate this, we draw on the extra-Kantian resources of Rawls' project. As we argue, the story about Kant is not the whole story about Rawls: when we look at Rawls through the lens of Rousseau and Hegel as well, we get a deeper and more accurate picture of Rawls' moral and political philosophy. ³³ Supplementing Rawls' Kantian inspired account of liberal autonomy is a richer relational conception of the human subject. Indeed, far from being blind or indifferent to the passions and emotions, Rawls' broader project is deeply attentive to the need to attend to the affective dimensions of human beings. Not only do certain affective dispositions play a central role as the legitimate bases and motivators of political obligations but more importantly, the emotions play a constitutive part in the conception of the subject at the centre of Rawls account. This moves the question of the place of affect in Rawls' theory beyond the motivational issues posed by a reason-driven justificatory liberalism. Rather, as we indicate, Rawls' conception of the just liberal regime presupposes an inter-subjectively constituted and affectively attuned individual.

What we find in Rawls, then, is a theorist sensitive to the embedded nature of the individual and the extent to which subjectivity is the by-product of political and social relations; that is, for Rawls, subjectivity is always already inter-subjectivity. In the end, we hope to decisively rebut those communitarian critics who disregard justice as fairness because (in their view) it is based on a philosophically, psychologically, and metaphysically incoherent conception of the person. ³⁴ What is the basis of this critical view? Contracting parties in the original position, recall, are denied any information about their ends, values, or conceptions of the good; they are, in other words, radically detached from any sense of particularity and from their constitutive attachments. And so, according to his critics, Rawls takes for granted or rather, he fully neglects, the extent to which individuals are fundamentally shaped by their social and political milieu; he subscribes to a kind of unencumbered, asocial individualism. How can contracting parties thus conceived adequately reason about social justice?

This image of the asocial and unencumbered chooser of justice principles – which, as we have already seen, is so pervasive in the secondary literature on TJ and Rawls more generally – is, quite simply, wrong. In this spirit, we hope to show two things in the remainder of this section and in the paper more generally: (1) the extent to which Rawls regards the self as deeply embedded in – as importantly determined by – their social and political-institutional milieu and (2) the extent to which Rawls regards these selves as deeply affective, as desirous of a certain kind of recognition from the institutions that coerce them and from the fellows with whom they are engaged in social cooperation. In the end, justice as fairness is best characterized by its attempt to secure the self-respect of citizens; and, for Rawls, proper self-respect is not just a quality of individuals, it is a quality of the configuration of societies.

In turning to the Rawlsian conception of the embedded self, this thicker conception of the human subject can be brought into sharper relief if we attend to Rawls' own critical engagement with the Kantian approach with which he is so frequently identified; as we shall see, Rawls' criticism of Kant opens up the interpretive space for the influence of Rousseau and Hegel. Here, a crucial resource for reconstructing Rawls' view is his ‘Kantian Constructivism in Moral Theory’; there, Rawls both highlights Kant’s influence on his thought and, more importantly given our purposes here, attempts to distinguish the nature of his project from that found in Kant. Whereas, on Rawls' view, Kant is primarily concerned with moral individuals – with the conscientious decision-making of individual moral agents – Rawls' own theory of justice is concerned first with the institutional contours of social life: ‘Justice as fairness assigns a certain primacy to the social. By contrast, Kant’s account of the categorical imperative applies to the personal maxims of sincere individuals in everyday life’. ³⁵ In another telling passage, Rawls reiterates his conscientious departure from this Kantian perspective:

Kant proceeds from the particular, even personal case of everyday life; he assumed that this process carried out correctly would eventually yield a coherent and sufficiently complete system of principles, including principles of social justice. Justice as fairness moves in quite the reverse fashion. ³⁶

It is of crucial importance, in Rawls' account, that public justice – justice at the level of institutions – comes first. And Rawls ultimately justifies the primacy of the political with reference to the educative function fulfilled by principle-guided institutions; that is, we begin by focusing on institutions because of the educative and socializing purposes fulfilled by them. Rawls simply doesn’t see this in Kant: in his Lectures on the History of Moral Philosophy, Rawls often draws attention to the insufficiently political character of Kant’s thought, that is, the extent to which moral consciousness – our sense of moral obligation – transcends the contingencies of social and political circumstance. ³⁷ And so, Rawls turns to Hegel, for whom the first principle of political philosophy is the ‘deep social rootedness of people within an established framework of their political and social institutions’. ³⁸ This is precisely the appeal of Hegel’s moral and political philosophy for Rawls: ‘TJ follows Hegel when it takes the basic structure of society as the first subject of justice’. ³⁹ Unlike Kant, ⁴⁰ and like Hegel, Rawls is deeply concerned with the educative role of institutions – their constitutive role in forming both the background conditions of social life and liberal citizens.

Indeed, Hegel’s view of freedom is ‘distinctively institutional’. ⁴¹ This connects, says Rawls, ‘with [Hegel’s] view of persons as rooted in and fashioned by the system of political and social institutions under which they live’. ⁴² Of course, this Hegelian view is a response to – it is a direct critique of – Kant’s transcendentalism: freedom is realized not in the pangs of the abstract and isolated conscience but in the world, through political and social institutions at a particular historical moment. ⁴³ In other words, Hegel rejects the Kantian view that one is capable of genuine moral freedom by virtue exclusively of one’s humanity and the concomitant possession of conscience; moral freedom is indivisible from the political institutional conditions in which it develops and evolves. Principle-guided institutions lead us to an idea of ourselves and of our fellows as capable of certain kinds of action and as worthy of certain kinds of treatment. We have freedom, in other words, because political institutions recognize (and concomitantly institutionalize) ‘our dignity as persons who are free’. ⁴⁴

This is where the Hegelian influence on justice as fairness emerges in its clearest light. Rawls follows Hegel in the belief that the ‘basic structure’ is the first subject of justice because of ‘the profound effects of these institutions on the kinds of persons we are’. ⁴⁵ What Rawls recognizes is the powerful and pervasive effects of institutions, which always already deeply shape the moral psychology of citizens and so necessarily have a transformative effect on the character and ethos of those subject to their coercive effects. Rawls' most explicit statement of the socializing function of principle-guided institutions is to be found in Book I of Political Liberalism:

Think of the principles of justice [and the political institutions established in light of them] as designed to form the social world in which our character and our conception of ourselves as persons are first acquired. These principles must give priority to those basic freedoms and opportunities in background institutions of civil society that enable us to become free and equal citizens in the first place, and to understand our role as persons with that status […] We have no prior identity before being in society. ⁴⁶

Political institutions (for Rawls as for Hegel) have ‘decisive long-term social effects and importantly shape the character and aims of the members of society, the kinds of persons they are and want to be’. ⁴⁷ Rawls and Hegel share the foundational belief that institutions (and their guiding principles) determine the kind of people we become. Citizenship is the outcome of an educative process. If, as Charles Taylor succinctly puts it, ‘the doctrine of Sittlichkeit is that morality reaches its completion in a community’, then Rawls is better understood in light of his engagement with Hegel. ⁴⁸

In a certain sense, then, the communitarian critics are right: the original position is based on an unrealistically unencumbered conception of the person. But, unlike the communitarians, we must avoid falling into a familiar trap in Rawls scholarship: treating the original position as representing some sort of ‘philosophical anthropology’ or even as a kind of founding moment – and not as a (mere) hypothetical thought experiment designed to represent, or ‘model’, our intuitions about justice as they happen to have evolved over time. ⁴⁹ The centrality of the original position in TJ is not in the end a claim about the possibility or desirability of stripping away the social identity and attachments of citizens; neither does its inclusion strip Rawls' theory of its emphasis on the social-institutional constitution of identity. While the abstract capacity for agency is, for Rawls, a central feature of our moral life, it is not the only such feature: for Rawls, the relationship of citizens is intrinsically social and other-oriented. It follows that the political problem is the proper institutional support for salutary inter-subjective relations – what Rawls calls the social basis of self-respect. After all, for Rawls, self-respect is not only a matter of having a secure sense of self, a secure sense that one’s life has meaning, and that one is well suited to pursue and revise one’s system of ends. Self-respect also has an inter-subjective or relational dimension: genuine self-respect requires that ‘our person and deeds [are] appreciated and confirmed by others who are likewise esteemed […] unless our endeavors are appreciated by our associates it is impossible for us to maintain the conviction that they are worthwhile’. ⁵⁰

As we have tried to indicate in the forgoing account, far from embracing an implausibly disembodied and affectless subject of political life, Rawls not only acknowledges these aspects of individuals but also insists that political and social institutions harness these affective resources. Of course, the central claim of the previous discussion – that for Rawls individuals are constituted or determined by principle-guided institutions – presents a conspicuously one-sided image of political life: it is easy to get the impression from it that egalitarian institutions are somehow in place, and that these institutions are then able to fulfil their educative or socializing function. But Rawls does not believe that political life begins in media res: we must move beyond the constitution paradigm, according to which individuals are spontaneously constituted (as liberal egalitarians) by their prevailing institutional milieu, towards co-constitution. What, exactly, do we mean by co-constitution? We hope to show in this section is that Rawls does not merely think of institutions as educating citizens to a particular ideal of recognition and reciprocity (although he certainly does think this). Rawls also thinks of persons as fundamentally subject to the psychological need for recognition and for self-respect, which itself possesses a powerful affective dimension in Rawls' account: we desire to be recognized as equals by the institutions that coerce us and by the fellows with whom we are engaged in political cooperation, and it is these ideals of mutual recognition and universal self-respect that inform and guide our shared political labours. In other words, proper self-respect is configured through our relations with others and through the sort of community in which we live.

Justice as fairness is therefore deeply implicated in the politics of recognition; the ideal of self-respect at its core is characterized by an inter-subjective and affective quality. Indeed, the extent to which justice as fairness is motivated by psychological concerns – namely, the need for recognition and for self-respect – has not been fully appreciated by Rawls' interpreters and critics. And, in our view, this is the by-product of the under-appreciated influence of Rousseau on Rawls' thought. Indeed, consciously following Rousseau, ⁵¹ Rawls starts from the human proclivity for interpersonal comparisons, accepts the negative consequences of unfavourable comparisons and (still following Rousseau) includes these psychological tendencies – to compare and to feel shame – as a fundamental consideration for constitutional and institutional design. It is in this spirit that Rawls claims that self-respect is ‘perhaps the most important primary good’. ⁵² And, as we shall see in a moment, like Rousseau, Rawls' solution to the lack of recognition – and, concomitantly, to the lack of self-respect – is to institutionalize the rights of citizenship in the basic institutional structure of society.

This is a very important point: Rawls is not only deeply interested in the mental states of citizens, he also recognizes the political salience of socially destructive psychological sentiments, such as envy and shame. Such feelings are often associated with inequalities of political and social status. In other words, envy-producing arrangements undermine the empathetic identifications necessary for stable, productive social cooperation; in Rawls' mind, envy is the obverse (and deeply counterproductive) psychic phenomenon of empathy. It follows that a constitutive aim of justice as fairness is to negotiate the emergence of such sentiments. ⁵³

How are these destructive emotions mediated or prevented? On Rawls' view, the solution to the lack of recognition – and to the concomitant lack of self-respect among those disadvantaged members of society – is to institutionalize political and relative economic equality in the basic structure of society. ⁵⁴ Put otherwise, equality at the highest level – at the level of citizenship – is the social basis of self-respect. ⁵⁵ Social and political conditions, in other words, are the fundamental determinant of a person’s self-respect: to be a citizen is to be secure about one’s place in society. Again, Rawls follows Rousseau: he gives us an institutional-basic structure solution to the problem of socially destructive envy; he is permanently sensitive to all the ways in which self-respect is potentially undermined by inegalitarian political and economic institutions, and of the ways in which empathetic identification is undermined by envy-generating arrangements. ⁵⁶

It follows that a central purpose of justice as fairness is to mitigate the natural human propensity to envy, which undermines the stability of the political community. Envy, on Rawls' view, is ‘a reaction to the loss of self-respect in circumstances where it would be unreasonable to expect someone to feel differently’. ⁵⁷ In this vein, the interpretation of justice as fairness presented here emphasizes the principles of justice – and the political institutions and distributive arrangements established in light of them – as the essential source not only of autonomy but of inter-subjectively grounded self-respect too, understood here as in part an affective disposition. Indeed, an essential dimension of Rawls' project is to use principle-guided institutions to combat the socially destructive forms of envy that arise due to the lack of self-respect felt by society’s least advantaged members. After all, according to Rawls, the problem of envy cannot be permanently ignored in any theory of justice: such sentiments do exist in society, as opposed to the information-deficient original position where envy has no basis. And, for Rawls, political institutions are often the basic instigating cause of these sentiments. In this sense, Rawls views liberalism as in part constituting an affective economy for its citizens – living under a just liberal regime leads to the cultivation of emotions and affective attachments appropriate to the psychology of mutual recognition.

Indeed, if the basic structure of society gives rise to feelings of pervasive envy – on account of the inequalities permitted by our principles of redistribution, say – this gives us reason to question those guiding principles. For feelings of envy, to reiterate, may lead to mutually destructive policies and actions: ‘The individual who envies another is prepared to do things that make them both worse off, if only the discrepancy between them is sufficiently reduced’. ⁵⁸ And so it is with the spite of the advantaged members of society also subject to the redistributive difference principle: ‘The spiteful man is willing to give up something to maintain the distance between himself and others’. ⁵⁹ Envy, in other words, obscures the mutual (economic) advantageousness of social cooperation (when governed by fair principles of justice). But Rawls does not spend much time contemplating the economic or distributive consequences of widespread envy. Instead, his main concern is the psychological consequences of unfavourable interpersonal comparisons. ⁶⁰ Again, an essential function of political institutions is to support the self-esteem of citizens, as grounded in egalitarian relations of mutual respect. And when individual self-esteem is secure, the pleasures of community are apparent: ‘One who is confident in himself is not grudging in the appreciation of others’. ⁶¹

There is, in Rawls' thought, a kind of dialectic in operation (albeit one that will never come to a complete resolution) between the natural psychological needs of persons and the design of the basic structure of the political community. Hence, the necessity of going beyond (mere) constitution towards co-constitution: institutions educate, but the design of those institutions is fundamentally informed by – it is a response to – the psychological needs of those subject to institutional coercion. We are, simply put, naturally desirous of egalitarian recognition and of self-respect; we need only recall the Rousseauian inspiration for Rawls' characterization of self-respect and self-esteem as basic primary goods. The more important point, however, is that both Rawls and Rousseau believe that self-respect and self-esteem are sentiments that are most effectively satisfied by well-designed political and economic institutions. A just liberal society thus necessarily has an affective economy, one that aims at securing the conditions of mutual respect, which are in turn the necessary grounding for self-respect and self-esteem. This leads to a new, holistic image of the historical evolution of this or that political community: our institutional milieu evolves from being a source of destabilizing envy – on account of political and radical economic inequalities – to being the essential guarantor of deeply sought after recognition and self-respect. ⁶² Only the latter milieu is the legitimate and likely object of citizens’ affective attachment. ⁶³ This insight has important implications for the place of what we might call a liberal character or ethos of justice in Rawls' project.

As we’ve suggested earlier, far from enjoining us to embrace an unrealistically simplistic conception of the human subject, at the centre of Rawls' project is a vision of individuals as both embodied as well as inter-subjectively and institutionally constituted. Moreover, Rawls not only acknowledges these aspects of individuals, but also repeatedly insists that political and social institutions harness these affective resources. Consequently, his project is highly attentive to the necessary role of affect in enabling appropriate relations of reciprocity and mutual respect among liberal citizens. What is more, these fundamental psychological needs of citizens themselves possess an affective dimension; proper self-esteem and self-respect are sentiments cultivated and supported by just liberal institutions. And what is the mechanism for cultivating these dispositions – for creating the conditions necessary to avoid the socially destructive forces of envy and for cultivating appropriate relations of self-love and self-esteem? As we’ve noted, Rawls' answer is institutional: Rawls appeals to the basic structure of society precisely because of the profound socializing capacity he sees it playing in social life, precisely because of the profound effects it has on the values, choices, and motivations of affected citizens. The rules of a just basic structure, in other words, cannot help but transform the psychology of individuals subject to those rules. ⁶⁴ The basic structure of a just liberal society not only promotes egalitarian relations that minimize the socially destructive force of envy, but it also promotes the necessary inter-subjective conditions of equality necessary for the affective dispositions of proper self-respect and self-esteem; after all, the latter rest on satisfying individual citizens’ desire for a particular kind of recognition – recognition as equals. Liberalism, on Rawls' view, therefore explicitly depends on the creation of institutions that satisfy fundamental human desires – most fundamentally our desire for recognition and respect. But it is equally necessary that such institutions play a role in developing the affective dispositions and attachments conducive to just and stable social cooperation. This highlights the sense in which Rawls' project describes an ongoing historical process, a process by which liberal institutions and liberal citizens are mutually transformed, with the latter developing what we might call a liberal character or ethos of justice.

A liberal soul-craft?

Our reconstruction of Rawls' political project, pace prevalent although ultimately misleading readings, is now complete. As we have shown, affect does have a central, fundamental, and yet unappreciated place in Rawls' attempt to constitute and defend a liberal theory of justice. The emotions are present in Rawls' thought from the very beginning, both in his account of the natural constellation of psychological needs and in the role that the emotions must play in the articulation of justice principles and in the institutional arrangements characteristic of a just liberal order. In the end, Rawls does have a rather capacious understanding of the self, and he does see that such an understanding is an essential element in the articulation of any compelling theory of justice.

We must concede, however, that there are deep tensions within Rawls' project and that Rawls' incorporation of affect into his understanding of the liberal subject is far from uncomplicated. Although here we can only do so in a cursory manner, by way of conclusion we wish to draw attention to some of these potential issues. For one, Rawls' incorporation of the emotions into his account of liberal justice certainly must trouble our understanding of Rawls as a key representative of traditional liberal, and especially Kantian inspired, visions of autonomy. As we have shown, Rawls anchors his account of liberal society in an inter-subjectively and affectively constituted conception of the self, as signalled in his self-avowed departure from the (purportedly) Kantian conception of the person. The individual can no longer rightly be conceptualized as the purely rational, unencumbered chooser of ends; citizens’ ends (and the value systems used to evaluate and choose those ends) must now be thought of as fundamentally constituted by their institutional milieu. All of this suggests that we must dispense with the image of the original position as capturing Rawls' conception of the autonomous self. Instead, Rawls follows Rousseau and Hegel: he views autonomy as an accomplishment, the contingent by-product of institutional circumstances that are themselves part of a larger historical process.

What is more, this suggests that the general liberal repugnance towards soul-craft simply isn’t present in Rawls: he clearly recognizes that institutions cannot avoid having a decisive effect upon our moral and emotional development as individuals and as citizens. Indeed, this is precisely why the basic structure must be governed by principles of justice: the basic structure of a just society not only facilitates self-development; liberal institutions also determine the nature of self-development. Liberalism is therefore fundamentally implicated in the cultivation of a liberal character or ethos. Of course, this puts Rawls in tension with the liberal aspiration to neutrality, and may demand that we move away from the typical liberal aversion to viewing political life as always already implicated in the project of character formation. And given the centrality of the ideas of impartiality and neutrality between completing conceptions of the good in the later iteration of justice as fairness, political liberalism, our view raises potential questions about the coherence of Rawls' later political philosophy.

However, it is only once we first acknowledge the place of affect in Rawls' thought that we can begin to grapple with these matters. Far from banishing the emotions from the domain of politics, Rawls' account of liberalism seeks to harness these embodied aspects of liberal citizens to bring about and sustain the stability of just liberal institutions. And while we may ultimately find ourselves dissenting from the vision of an affective economy that emerges in Rawls' project, with its implications of a distinctively liberal soul-craft, surely the starting point for such an evaluation must be a serious reckoning with the place of the emotions in his thought. Moreover, while recognizing this dimension of Rawls' project does raise deeper questions, it is worth considering that often the hallmark of a great thinker consists in part in these very tensions and ambiguities. Indeed, when read in this light, Rawls can be seen as representative of the complicated and ongoing engagement of liberal theory with the question of the place of the emotions in political life.