Conceptual analysis of justice has fallen out of favour in political philosophy since the publication of John Rawls’s seminal text, A Theory of Justice. Influential figures such as Rawls and Ronald Dworkin ushered in a methodological shift that drew attention to substantive normative questions and away from conceptual questions about the usage of words. In his book, A Conceptual Investigation of Justice, Kyle Johannsen provocatively raises the worry that turning our backs on conceptual analysis means that contemporary political philosophy lacks a shared understanding of the concept of justice, which is why he calls for renewed attention to the manner in which the word ‘justice’ is and should be used. While in contemporary political philosophy Rawls’s shadow looms large, Johannsen’s book is a testament to the enduring legacy of G.A. Cohen, for whom justice, as a conceptual matter, is a fundamental value that should be balanced against other normative considerations when we turn our attention to questions about how to design institutions, all things considered.

As someone sympathetic to the Rawlsian project, my commentary is presented in the spirit of a family squabble. In spite of all that divides us, Johannsen and I are approaching the same problems concerning the nature of justice, albeit from the perspective of opposing factions. Both of us, for instance, agree on the importance of the distinction between justice (proper) and rules of regulation, while disagreeing about what constitutes justice. And Johannsen is absolutely right to insist on this being more than merely a verbal dispute. If our dispute were merely verbal, it would be of little consequence; our disagreement would be, as Johannsen points out, “merely semantic and thus of not of any philosophical interest.”^{Footnote 1}

To avoid the worry that Johannsen and I are talking past one another, when contextualists and fundamentalists about justice are engaged in a conceptual dispute over the nature of justice, in part, what we are engaged in is a dispute over the function of the term. A concept, after all, refers to how a word functions, and in our disputes about justice, the precise nature of the disagreement is over whether justice is, as fundamentalists like Johannsen argue, an input in our practical reasoning, or, as contextualists like myself argue, an output of our practical reasoning concerning how we ought to design society’s major institutions.^{Footnote 2} When the concept of justice is understood as an input, as Johannsen urges us to do, justice is a singular value out of a plurality of values, which must then be balanced against these other values. Our dispute is properly conceptual, and of deep philosophical interest, because it concerns a conceptual disagreement over the role of justice. On the one hand, we have the likes of Johannsen and Cohen who identify justice as a primarily evaluative criterion of distributive fairness that only later contributes to our practical reasoning about how to design institutions. On the other hand, we have the likes of myself and Rawls, who conceive of the role of justice differently, making it an output of our practical reasoning given the kind of problems or questions that many of us think any viable concept of justice is meant to resolve.

In what follows, I respond to Johannsen’s implication that Rawls and other contextualists are typically engaged in an enterprise that is less than justice by theorizing about institutional rightness, instead of justice (proper). My intention is to revive the contextualist endeavour and to persuade Johannsen, and those sympathetic to his project, that complex problems, like those of justice, may require a complex value, like Rawls’s understanding of justice.

Johannsen opens his book with a distinction drawn from Aristotle between broad and narrow justice.^{Footnote 3} Broad justice concerns itself with what we today traditionally conceive of as moral philosophy, and so encompasses ideas related to moral rightness in general. Narrow justice, on the other hand, is related to, but distinct from, moral rightness. As its name suggests, narrow justice concerns itself with a part of morality, covering a narrower subject matter, specifically that part of morality that contemporary political philosophers normally refer to as ‘justice,’ and that pertains to matters of fairness, with a particular focus on matters of distribution and rectification. While Johannsen traces the roots of this idea back to Aristotle, contemporary political philosophers continue to insist upon the distinction between broad and narrow justice,^{Footnote 4} with the most systematic articulation being given by Rawls in his paper, “The Independence of Moral Theory.”^{Footnote 5} Using different terminology, Rawls charts a similar distinction between moral philosophy (i.e., broad justice, and moral rightness in general) and moral theory (i.e., narrow justice), which for Rawls means the comparative study of substantive moral conceptions, or “how the basic notion of right, the good, and moral worth are arranged to form different moral structures.”^{Footnote 6} Whereas Rawls offers a broader metaethical thesis regarding how progress in moral and political theorizing does not depend on metaphysical or epistemological beliefs, the relevant similarity between Johannsen and Rawls is the restricted scope of inquiry by focusing on a (narrower) part of moral philosophy. Nevertheless, the similarities tend to breakdown here, and, in what follows, I map some of the key distinction between justice fundamentalists like Johannsen and Cohen, and contextualists, like myself and Rawls.

The conceptual dividing line between contextualists and fundamentalists about justice can be charted along a number of dimensions. Contextualists maintain that (a) justice is an output of our practical deliberations; (b) justice focuses on institutional rightness; (c) justice is a complex value made up of a combination of values and normative considerations; which entails that (d) values and normative considerations—such as liberty, stability, legitimacy, etc.—are properly internal to justice; and so (e) the role of justice is said to be action-guiding. Fundamentalists, on the other hand, insist that: (a*) justice is an input in our practical deliberations, which must then be balanced against other kinds of normative considerations; (b*) justice is about comprehensive morality; (c*) justice is a simple, and thus a singular and unified value; which entails that (d*) all other normative considerations are external to justice (proper); and so (e*) the role of justice is said to identify a defeasible value.

Whereas the dominant trend in contemporary political philosophy is to theorize about justice in broadly contextualist terms, Johannsen urges us to theorize about justice by adopting a fundamentalist understanding of justice, and proceeding on the basis of a purely conceptual analysis of justice. However, before we join Johannsen in his purely conceptual analysis of justice, we should reflect upon what the point and purpose of a concept of (narrow) justice is, and why a clarified contextualist (or Rawls inspired) position may be the best understanding of justice for the task at hand.

To begin then, for Johannsen, the point and purpose of justice is to identify an evaluative criterion of distributive fairness, importantly, in a way that is distinct from the exigencies of political practice. Since other commentators have touched upon the scope and content of Johannsen’s favoured distributive principle, instead I shift my attention toward two methodological issues, as a way of raising some concerns over whether Johannsen is indeed operating with the ‘correct’ understanding of what constitutes justice.

The first methodological issue concerns the narrowness of Johannsen’s concept of justice, both as a matter of scope, and his reliance on a single defeasible value. Johannsen’s emphasis throughout the book on distributive fairness is indicative of an underlying theme running through his work, which is that justice (or more specifically, narrow justice) is coextensive with distributive justice. And Johannsen is in good company here thinking that justice is first and foremost a distributive concept. After all, when we think about the concept of justice, apart from specific conceptions of justice, we tend think of dividing the benefits and burdens that stem from social cooperation. Here, Johannsen’s favoured principle of luck egalitarianism, supplies us with the evaluative criterion by putting forward that all and only those unchosen circumstances stand in need of compensation.

To put things somewhat crudely, justice, on Johannsen’s understanding of the term with its exclusive focus on distribution, is analogous to distributing ‘bundles of stuff.’ Of course, I don’t mean to imply that distributive justice is only concerned with distributing ‘stuff’ in a material sense, as if distributive justice is only concerned with the distribution of income and wealth. Included as well, we may presume, are also bundles of rights, so that people may have, for instance, a right to freedom of conscience. Nevertheless, we may ask whether questions about the distribution of stuff (broadly construed to include income, wealth, rights, and opportunities) exhausts what we may take to be salient questions of justice. And so, Rawls warns us that “We cannot, in general, assess a conception of justice by its distributive role alone, however useful this role may be in identifying the concept of justice.” And so, in spite of the primacy of justice (understood in a distributive sense) as the first virtue of institutions, “other things equal, one conception of justice is preferred to another when its broader consequences are more desirable.”^{Footnote 7}

So even conceding Johannsen’s central point that conceptual analysis of justice has fallen out of fashion, and has been superseded by questions about institutional rightness, perhaps we have good reasons to study what is in fashion, given the role of justice in organizing a society and our interactions with one another. Or at least that is a point of which I shall attempt to persuade Johannsen in the remainder of my commentary.

My point is that there are a number of vital problems and questions that at least a sizeable number of liberal political philosophers take to be internal to justice, and that are not reducible to questions about the proper distribution of the benefits and burdens of cooperation. Questions, say, about the place of religion in a liberal society, the boundaries of citizenship, the status of cultural minority groups, or the permissibility of using public funds to support the arts or environmental protection, may not fall within the purview of what we traditionally take to be matters of distributive justice. Some of these questions can, and have been, subsumed under the rubric of ‘rights talk,’ which may blunt the force of my objection here (despite my own personal reservations about ‘rights’ being able to do all the conceptually heavy lifting in these instances). Nevertheless, even if we take it as read that a principle of distributive fairness can supply us with answers to what we take to be questions pertaining to justice, a further question is: if we understand justice to be a simple (singular and unified) value, are we left with a satisfactory answer to these questions of justice?

Underlying Johannsen’s justice fundamentalism is a broadly luck egalitarian understanding of distributive justice, wherein we should compensate people for various unchosen circumstances. A paradigmatic instance of unchosen circumstances, and a pressing problem of justice, involves cases of cultural minorities. Here, it is possible to contrast Johannsen’s approach with a fellow luck egalitarian, Will Kymlicka. Membership in a particular culture, according to Kymlicka, falls within the scope of ‘unchosen circumstances’ that makes cultural membership a prime candidate for compensation via special group rights.^{Footnote 8} So far, so good for Johannsen’s understanding of justice. Except Kymlicka’s grounding for cultural minority group rights crucially depends on justice being a complex value. For Kymlicka and other liberal political philosophers, there’s a reason (or value), which is importantly internal to justice, for why we should compensate people based on unchosen circumstances beyond the fact that these are merely unchosen circumstances.

What’s unique about culture is how it relates to other preeminent liberal values, such as freedom and autonomy. As Kymlicka makes clear, “freedom involves making choices amongst various options, and our societal culture not only provides these options, but also makes them meaningful for us.”^{Footnote 9} This means that the reason why we should compensate people on the basis of their cultural membership is that one’s culture provides a meaningful context of choice, which is a precondition for the effective exercise of freedom and autonomy. According to Kymlicka, culture provides individuals with a shared vocabulary that makes meaningful choices possible by supplying individuals with access to information and the capacity to reflectively evaluate various option sets that may or may not be available to them to pursue what he or she takes to be a worthwhile and meaningful plan of life.

Although, a person’s membership in a particular cultural group, is, as we know, among a myriad of unchosen circumstances that people may face, it is far from the only one. According to a number of prominent luck egalitarians, including Cohen, unchosen circumstances include expensive tastes, such as a love of opera, photography, or plover’s eggs. Yet many other theorists (and I suspect most non-philosophers at-large) justifiably think that, while we should compensate people for the burdens associated with being part of a minority culture, we are less inclined to compensate people for their love of plover’s eggs, or the latest Nikon DSLR with its assorted lenses. To repeat: what’s unique about culture that makes it a prime candidate for compensation, unlike other expensive tastes, is the connection between culture and other values that many liberal political philosophers think is properly internal to justice, such as freedom and autonomy.

By focusing on the value of distributive fairness alone, Johannsen’s justice fundamentalism must treat ‘expensive tastes’ and matters relating to culture and religion on an equal moral footing in terms of what stands in need of compensation (as a matter of justice proper), or at least provides no principled basis (as a matter of justice proper) for why claims of culture and religion merit weightier claims for compensation. To what extent we should be troubled by this possible consequence depends on what kind of importance we place on aspects of people’s lives such as religion and culture. For some who embrace the promise of a secular age, these matters may be of little to no importance; for others, whose religion and culture represent a fundamental aspect of their identities, to lose access to their culture or religion would be a serious blow. And whatever our personal feelings about the importance of culture or religion, to diminish the importance of culture or religion because neither fits with what we find to give life meaning would be antithetical to the grand liberal vision of creating a just society where each can sincerely pursue a plan of life that gives a person’s life meaning—at least in cases where the pursuit of that plan of life doesn’t infringe upon the rights of others.

If Johannsen is at all troubled by this issue and agrees that matters of culture and religion are part of the domain of justice, then he finds himself on the horns of a dilemma: either he must concede that justice requires more than just distributive fairness, and may include values such as freedom, autonomy, and integrity, or he must conclude that political judgements relating to religion and culture are left to be settled by rules of regulation. If Johannsen takes the second horn of the dilemma, he opens himself up to unwelcome—and potentially unnerving—trade-offs between justice and the whole score of other kinds of values (both moral and not), in addition to the calculus of social interests.

What makes the present dilemma all the more pressing for Johannsen is that he wishes to reconcile, as he puts it himself, “two seemingly conflicting ideas”: his understanding of narrow justice as one value among many, with Rawls’s key insight that justice enjoys a special kind of primacy, being the first virtue of institutions.^{Footnote 10} In his efforts to rescue justice from the conceptual confusion allegedly associated with treating it as a complex value, Johannsen’s argumentative strategy is to introduce justice into our regulatory framework (i.e., constructing rules of regulation) at both the procedural and value trade-off levels, so that justice—as Johannsen understands the concept—plays a key role in shaping the hypothetical contract situation. In order to assess the success of Johannsen’s argumentative strategy, we must assess the moral desirability of the outcomes.

To see why Johannsen’s argumentative strategy may be a worrisome route to take, consider a case from Rawls wherein some large segment of the population regards certain sexual practices as abominable. Rawls further supposes that it is insufficient that these practices are being kept from public view, since the very thought that these practices are occurring is enough to arouse feelings of anger and hatred, and so these practices must be condemned.^{Footnote 11} As a justice fundamentalist with luck egalitarian leanings, Johannsen may no doubt respond that one’s sexual orientation is an unchosen circumstance, and so merits in this case special protections. However, we may posit that the group who abhors these practices also finds that their reactions too are based on unchosen circumstances; perhaps they are simply hard-wired to be repulsed by certain sexual practices, or these practices are an affront to their religious beliefs, and so must be admonished. That is, even if we find the first explanation for why these abhorrent views are unchosen circumstances to rest on a psychologically dubious foundation, we must still confront the notion that, for some particularly devout citizens who subscribe to certain kinds of orthodox religious beliefs, they are experiencing a psychological harm by sharing a community with—what these devout members of the public consider to be—sinners.^{Footnote 12} It would appear, at first glance, that Johannsen may lack the appropriate justice-based grounds to prevent restrictive measures against various kind of sexual practices, which for those who don’t share these sentiments, cause no social injury.

Johannsen’s framework seems unable to resolve this dispute as a matter of justice: if both sets of circumstances are unchosen, we appear to have reached a stalemate on how to proceed. All Johannsen can do at this point is rely on rules of regulation to protect people’s liberty to partake in whatever sexual practices they see fit (on the provision that these involve consenting adults, and various other stipulations), however even this appeal must contend with the whole realm of values that are relevant to how we should guide political practice. Let’s assume these feelings of anger and hatred are sufficiently widespread that if we allow certain types of sexual practices, they may threaten the stability of the society, and so for the sake of balancing the values of justice, with say, stability, it would be prudent to enforce repressive measures against these sexual practices.

Despite his ardent attempt to preserve the so-called primacy of justice as the first virtue of institutions, by turning justice into one value among many that must be balanced against other values risks losing what makes justice the first virtue of human activities: namely, the uncompromising nature of justice. We may, as Rawls concedes, have to bend to the convictions and passions of the majority when these sentiments make liberty impossible to maintain. However, Rawls adds that if or when we must bow to these practical necessities that is a far cry from “accepting the justification that if these feelings are strong enough and outweigh in intensity any feelings that might replace them.”^{Footnote 13} What justice requires, and what Johannsen’s understanding of justice may not be able to achieve, is the impulse to move speedily toward a set of just institutions that protects the value of liberty for all as soon as circumstances permit in spite of existing sentiments to the contrary. My contention is that, as a conceptual matter, and so as a conceptual (and logical) consequent of Johannsen’s position, we may be forced to compromise justice against other values, and as Rawls warns, this risks subjecting the rights secured by justice “to political bargaining or the calculus of social interests.”^{Footnote 14}

In contrast to Johannsen’s justice fundamentalism, on a contextualist’s understanding about justice, we may rely on values that are—by a contextualist’s own lights—internal to justice, such as liberty and equal citizenship, in order to conclude that restricting people’s freedom to engage in certain sexual practices is unjust. For instance, proposed pieces of legislation that would allow proprietors to discriminate based on sexual orientation for the purposes of excluding people from their establishments (e.g., a B&B, or a bakery) can be said to be antithetical to our standing as equal citizens, and people should have the liberty to engage in whatever sexual practices they wish. Rawls does concede that, as a matter of political practice or rules of regulation, the convictions of the majority make it impossible to maintain liberty. So, as a matter of practical necessity, we accept restrictions on certain sexual practices, or at least until social circumstances evolve to the point where a society‘s norms allow for greater (sexual) liberty. What’s important here is that a practical concession doesn’t entail possible restrictions on liberty as justified (or ‘just’), and this distinction between what is just (or publicly justified) and what’s a practical concession provides us with the impetus, in the sense of a principled basis, to move toward just institutions as speedily as possible. Pitting values which other theorists conceive of as internal to justice, against practical considerations leaves us with little to no principled basis to moral/justice-based values in cases of conflict with non-moral practical considerations (e.g., what’s feasible given prevailing sentiments on-the-ground).

This brings me to the second methodological issue, which is Johannsen’s insistence that contextualists like myself and Rawls, by focusing on issues of institutional design, are not actually talking ‘about justice,’ and instead are merely talking about rules of regulation. There’s a sense in which Johannsen is implying that contextualists are doing something ‘less than justice.’ Bruised egos aside, I think we lose something important by saying that we are either talking about ‘justice’ as a fundamental value or else we find ourselves in the realm of action-guidance with all the trials and tribulations of effectively implementing what justice requires.

Contextualists should fully endorse Johannsen’s distinction between justice (proper) and rules of regulation. Deciding what to do, in an all-things-considered sense, requires taking into account a whole host of moral and non-moral considerations, some of which Johannsen and contextualists will agree are properly external to justice. After all, we live in a finite world, with finite resources, and cognitive, institutional, practical and motivational limitations, that will inevitably constrain what we can achieve in the real world, with its warts and all. However, Johannsen attributes to theories of justice a false dichotomy between justice understood as a single) fundamental value, and action guidance. Even though most contextualists (perhaps including Rawls), share a desire for our theories to have some practical important, at bottom, when contextualists theorize about justice, we often do so at the level of what has come to be known as ideal theory. For contextualists like myself and Rawls, we want to know what a just society could look like, with these concerns gaining momentum after his political turn, where Rawls became deeply interested in the question of how a just and stable social world, characterized by deep diversity, is possible. Importantly, contextualists want to know what a just (and stable) society might look like before we take into account the necessary concessions and calculus of trade-offs that must be taken into account when we attempt to implement these requirements.

What contextualists’ understanding of justice allows us to do, according to Rawls, is to take some of the most divisive political issues off the political agenda.^{Footnote 15} Of course, Rawls makes these statements in the context of achieving stability, which for some is merely a practical matter, except there’s a deeper philosophical reason to take some issues off the political agenda that don’t bear on stability considerations. Nevertheless, for Rawls and those who share his understanding of justice, some issues are simply too important to be left up to our intuitive weighing of various normative considerations because the interests at stake are too important to be traded off to achieve, for instance, a more efficient distribution of the benefits and burdens of cooperation.

The dispute between justice fundamentalists and contextualists is not, as Johannsen correctly points out, a mere verbal dispute. For both sides of the dispute, we’re arguing about the same thing: namely, what is the best understanding of justice, importantly, before we start down the path of thinking about how to implement justice, or guide action in our less-than-ideal, or real-world societies? For both sides, we’re addressing questions, which at bottom, are deeply theoretical, while disagreeing sharply about what is properly internal to justice. Contextualists insist upon a more robust sense of what’s internal to justice because if justice is coextensive with a distributive fairness, cashed out conception of luck egalitarianism, we risk losing a principled basis to adjudicate between competing values that fall outside the scope of distributive fairness. What contextualists want to contribute to our theorizing about justice is a systemic weighting of our normative considerations and our considered convictions in an effort to decide, as a matter of justice on theoretical level, what is just and unjust. The assignment of weights to various normative criteria is, as Rawls tells us, an essential, and not a minor, component of a conception of justice.

Johannsen may respond that he too is concerned with how we weigh and balance competing normative considerations, and could even make the case that justice (as a fundamental value) has a normative primacy in our practical deliberations. What’s troublesome about Johannsen’s approach is that how we balance values that fall outside the scope of distributive fairness must rely on our intuitions. Many, if not most of us, may think that freedom, liberty, autonomy, integrity, etc. are important values in our practical deliberations about what constitutes justice, and perhaps even that they should be accorded an additional weight in relation to other, more practical, considerations.

The hope among contextualists is that, by assigning weights to various normative considerations that we (contextualists) believe are internal to justice, we may shape our practical deliberations and political discourse on-the-ground, and perhaps, optimistically, provide us with some necessary critical leverage to the problems we face in societies today. Including values in addition to distributive fairness means that the kinds of values contextualists think are internal to justice all share a particular kind of primacy. Johannsen may want to claim that liberty and autonomy are significant and weighty values, but he may only make that case according to his own ‘intuitions,’ and for others who do not share his intuitions, the values of liberty and autonomy may be in trouble, as these values are subject to the inevitable trade-offs against other values in actual, practical, and political deliberations about how we should design society’s major institutions.

In closing, Johannsen’s A Conceptual Analysis of Justice offers a spirited and novel defence of luck egalitarianism, which, if his conceptual analysis of justice is correct, supplies his critics with powerful responses. The worry I have been raising in this commentary is that Johannsen’s conceptual analysis of justice may be in trouble because, when we classify justice only as an input in our practical deliberation, we pit what justice requires against a score of other values and the calculus of social interest. Johannsen’s search for an evaluative criterion of distributive fairness entails that he demarcates the boundaries between what’s internal and external to justice (proper) in way that may jeopardize what people take to be their most cherished commitments and convictions. Both fundamentalists and contextualists about justice agree that at one level we’re theorizing about justice, while questions of implementation are another matter, so we construct rules of regulation. What’s at stake is how we define justice, and whether that definition can be stated as a singular value, or as a complex amalgamation of values that together construct what we take to be justice.