1. Different Questions, Differing Structures
Kant, Rawls and human rights thinking are at one in seeing justice as different from (the rest of) ethics. Justice is for cities and for states, for polities and for the public domain. Only where there is an organized public sphere, whose structures support and enforce a degree of order, will certain types of action by individuals or by institutions count as just or unjust. However, accounts of the scope, the justifications, and the interpretation of standards of justice proposed by these three liberal approaches to justice differ in marked and sometimes surprising ways. Questions that are central for one approach are often marginal or ignored in another. Justifications of standards that are supported by one approach may be unavailable in others. Approaches to interpreting and enacting justice that are seen as basic by some approaches may be inadequate or even impossible in others.
Many traditional accounts of justice were coupled with, and often integrated with, accounts of other ethical standards. They addressed the first personal agent’s question ‘what ought we (or I) do?’ They set out accounts of requirements both to act in specific ways and to cultivate specific traits of character, so proposed accounts both of duty and of virtue. Today duty and virtue remain familiar ethical terms, but both are often seen as old-fashioned, and sometimes as suspect or obsolete. The classical agent’s question has increasingly been set aside in favour of narrower questions about required standards.Footnote 1
Although some contemporary discussions of ethical standards and justice continue to address the agent’s question, a good deal of contemporary work in political philosophy takes a perspective-neutral view of the basic question to be addressed. It seeks to identify and to justify standards that just institutions should meet, without prioritizing the perspective either of agents or of recipients, and is silent about standards that are not relevant to justice. Yet other contemporary discussions of required standards and justice, and above all the vast range of discussions of human rights, prioritize the recipient’s perspective, by focusing on aspects of justice that individuals can claim as their rights.
Each of these questions addresses a range of practical questions about required standards, but there are also numerous other discussions of ethical standards that entirely avoid practical or normative questions by taking a third-party or spectator perspective. Their accounts of ethical standards, including standards of justice, report on the values (or supposed values) actually adopted or preferred by certain individuals, social groups or indeed by entire cultures and communities. Discussions that take a spectator perspective typically ignore questions about the justification of the (supposed) values they discuss, often seeing them simply as reflecting individual choices or preferences. Some third-party accounts of ethics and of justice suggest (implausibly) that action is justified provided it expresses ‘my values’ or ‘our values’ (or ‘my’ or ‘our’ principles or preferences). Although this substitute for normative justification was ridiculed a long time ago by Groucho Marx, with the comment ‘Those are my principles, and if you don’t like them … well, I have others’, it remains persistently popular. Here I shall ignore discussions of (supposed) values that take a spectator perspective in order to concentrate on some of the differences between those approaches that take (a varying range of) questions about justifying practical or normative standards seriously.
Starting from differing questions about normative standards yields systematically different accounts of justice (and in some cases also of other ethical standards). What is at stake is not just a choice between differing substantive claims about justice (or about other ethical standards), but a deeper divergence in views about what matters and why it matters. Approaches that address one range of questions are sometimes silent about questions that are central in other approaches, or render them invisible. So, I shall begin by considering the broadest and most inclusive approaches, and will then discuss positions that take narrower views of the basic questions and consequently of the scope of normative reasoning.
I therefore start with a sketch of the scope and structure of positions that address the agent’s questions, taking Kant’s broad and systematic account of justice and of ethics as exemplary. Other Aristotelian and Kantian approaches to ethical standards (including standards of justice) also address the agent’s question, so could provide starting points for a similarly broad range of questions about justice and about ethics. However, I shall focus on Kant’s arguments, rather than on other Kantian approaches or on Aristotelian approaches, because his work both addresses the full range of the agent’s question and relies on a stringent view of normative justification. I shall then turn to the leading example of contemporary (quasi) Kantian writing on justice developed by John Rawls, which also supports a broad account of justice, but is silent about other ethical standards. Finally, I shall consider the range of standards put forward in discussions of justice that prioritize human rights.
2. Questions, Scope and Structure
Kant: Justice and Ethics
Kant’s discussions of justice formed part of a broad account of duties that addresses the classical agent’s question ‘what ought I (or we) do?’ He argued for a systematic and complex account of human duties, using a range of distinctions that had been widely used by early modern natural law theorists,Footnote 2 including by those who took perfectionist approaches to justification. He distinguished perfect duties that are owed without exception (e.g. duties to keep promises; G, 4: 421ff. – one of many passages) from imperfect duties that leave agents discretion on whether to act in a particular situation (e.g. duties of beneficence, G, 4: 423 – again, one of many passages).Footnote 3 He subdivides perfect duties into duties of justice, that can and should be enforced by public authorities (cf. MM, 6: 230ff.) from perfect duties that are not duties of justice, so should not be enforced by public authorities. Among enforceable duties of justice, he classifies some, but not all, as duties whose performance can be claimed as rights by others. For Kant an account of rights is therefore only part of a theory of justice, which in turn provides only part of an account of perfect duties, which in turn provides only part of an account of duty. A focus on the classical agent’s question allowed him to cover a complex and systematic account of required action that dealt with imperfect as well as perfect duties, that distinguished perfect duties of justice from perfect duties that form part of ethics (but not of justice) and that subdivided duties of justice into those whose performance others can rightfully claim and those that others cannot claim as their right.Footnote 4
Rawls: Justice as Fairness, But without Ethics
Rawls’s theory of justice has a narrower scope. He does not aim to answer the generic agent’s question, ‘what ought I (or we) to do?’, but to address a perspective-neutral question about the ‘basic structure of society’ that identifies the principles that a just society should embody. He claims that this approach ‘generalises and carries to a higher level of abstraction the traditional conception of the social contract’ (TJ, 3; cf. 11), but that it does not provide a basis for an account of ethical duties.Footnote 5
However, while Rawls does not aim for as full an account of required action as the one Kant had proposed, he too offers more than an account of duties that are the counterparts of claimable rights. Only the first of Rawls’s two principles of justice defines standards of justice in terms of rights that each should be accorded, which he groups under the heading of ‘an equal right to the most extensive basic liberty compatible with a similar liberty to others’ (TJ, 60). Rawls’s second principle of justice is a distributive requirement that does not define individual rights. It demands that social and economic inequalities be arranged so that they are reasonably expected to be to everyone’s advantage, and are attached to positions and offices that are open to all (TJ, 60). ‘Justice as fairness’, as Rawls articulated it, therefore demands both a range of rights that can be claimed by individuals and a set of institutions that meet certain structural conditions. Institutions that secure these structural conditions will thereby secure claimable rights: but these will be the positive legal or institutional rights of a specific jurisdiction, rather than fundamental rights with universal scope. Principles of distributive justice often do not define rights for individuals, since they need not show who may claim what from whom.Footnote 6
Human Rights: Justice as Claimable Rights
Human rights approaches to justice address a narrower range of questions than those considered either in Kantian or in Rawlsian accounts of justice. They offer neither a full answer to the agent’s question, nor an account of full range of perspective-neutral claims about the basic structures of just societies. Rather they set out the specific claims right-holders can make to others’ action and forbearance. Since human rights approaches are intended to answer the claimant’s question ‘What am I (or we) entitled to claim?’, they are unavoidably silent not only about imperfect duties (e.g. beneficence), but about perfect duties that cannot or should not be claimed or enforced by public authorities (e.g. many aspects of honesty or promise keeping), as well as about aspects of justice that are not directed to individual claimants and do not define individual entitlements, but prescribe structures or patterns for distributions of certain goods.
Daily discussion of human rights often purports to offer an account not only of distributive standards, but of individuals’ claims under those standards. However, this is secured only by yoking human rights standards to institutional assumptions and interpretations, such as claims about human rights law or other specific institutional arrangements for protecting and securing rights. For example, rights to goods and services, such as a ‘right to food’ or a ‘right to health’ (or to health care), do not identify relevant duty-bearers, let alone prescribing just which food or health care is to be provided for whom. Such claims are incomplete or indeterminate without institutional measures that assign the relevant duties to specified (preferably competent!) duty-bearers. And distributive standards that range across unidentified (sometimes unidentifiable!) ranges of persons (e.g. rights of future generations; rights to fair economic arrangements) require institutional supplementation not only in order to specify the relevant duty-bearers, but also in order to specify the relevant right-holders). They are doubly indeterminate unless and until supplemented with institutional structures that identify those duty-bearers and right-holders. Unavoidably the institutional structures that can implement human rights requirements define positive duties and positive rights, which are not justified merely by offering broader moral justification for the underlying human rights principles.
In the past some of the indeterminate aspects of justice that require supplementation by specific institutional provisions might have been spoken of as ‘provisional right(s)’, indicating that their claims were under-defined and could not be secured unless and until institutional provision determined who ought to do what for whom, and who could claim what from whom. Recent disagreements about whether human rights should be thought of as ‘moral’ or ‘political’Footnote 7 show that deep uncertainty about the scope and status of claims about human rights persists, and indicate just how much institutional ground has to be covered before determinate and effective claims to rights can be made.
3. Patterns of Justification
One might expect these striking differences in the scope of liberal approaches to justice to be reflected in the relative difficulty of the justificatory arguments needed by each approach. More specifically, one might expect that the wider scope of Kant’s account of duties would make its claims particularly hard to justify, and that the narrower scope of human rights thinking would make them particularly easy to justify. However, matters are not that simple, and it is far from obvious that the narrower accounts of justice proposed by accounts of human rights are more readily justified than the accounts proposed in broader theories of liberal justice. Indeed, I shall suggest that Kant’s approach to standards of justice relies on less demanding assumptions than are typically used to justify some liberal approaches that seek to justify less ambitious accounts of justice.
Kant: Justification as Modal Constructivism
Kant’s approach to the justification of practical principles relies on the thought that reasons for action should be fit to reach their intended audiences, hence that any reasons intended to be relevant for everyone, such as accounts of universal standards of justice or universal ethical standards, must have ‘the form of law’. This is not a demand for agreement by all, but a weaker demand for intelligibility to all. Kant offered a weak, modal account of practical reasoning: reasoning fails unless reasoners offer considerations that their audiences could in principle follow in thought, or adopt for action. Anybody who aims to justify principles for all – whether principles of justice or other ethical principles – must therefore, as Kant sees it, reject principles that could not be adopted by all.
This demand is the key to Kant’s account of the constructive method of justification on which his practical philosophy relies, which is articulated in the universal law formulation of the categorical imperative. Although Kant did not use the contemporary term constructivism to describe this approach to justification, there is strong textual warrant for describing his approach as constructivist. At the start of the Doctrine of Method of the Critique of Pure Reason he set out systematic parallels between the activities of reasoning and of constructing. As he sees it, building projects need materials, a workforce and a plan, and the plan must not demand unavailable materials. We must therefore begin with ‘an estimate of the materials, and have determined for what sort, height, and strength of building they will suffice’ (CPR, A707/B735; tr. OO’N). A realistic view of the available materials might show that some over-ambitious constructions are not feasible. Like the ancient builders of the Tower of Babel, we might find that ‘although we had in mind a tower that would reach the heavens, yet the stock of materials was only enough for a dwelling house – just roomy enough for our tasks on the plain of experience and just high enough for us to look across the plain’ (ibid.). Like them we might come up against the reality that some plans are not realizable because ‘the bold undertaking had come to nothing for lack of materials, quite apart from the Babel of tongues that unavoidably set workers against one another and scattered them across the earth, each to build separately following his own plan’. Like Kant we might find ourselves driven to conclude that:
Our problem is not just to do with materials, but even more to do with the plan. Since we have been warned not to risk everything on a favorite but senseless project, which could perhaps exceed our whole means, yet cannot well refrain from building a secure home, we have to plan our building with the supplies we have been given and at the same time to suit our needs. (Ibid.)
Yet for more than two centuries Kant’s commentators have often asserted that he relied on extravagant rather than on minimal premises, and insisted that he did not take an economical, or even a plausible, approach to the justification of fundamental standards. This is undoubtedly a possible reading of the relevant texts, and has been widely repeated and endorsed. But it does not seem to me to offer a convincing reading of Kant’s views on reasoned justification, nor therefore of his accounts of justice or of (the rest of) ethics.
Just as every building project requires materials, a workforce and a plan, so too does the project of reasoning in order to justify either truth claims or normative claims to an unrestricted audience. The materials available to us are those elements of human experience and understanding that Kant discussed in the Doctrine of Elements. Of course, we might wonder whether he offered an adequate inventory of available materials. Are there materials that he overlooks, but which could be used? Does he make inaccurate claims about the materials that are available?
Some philosophical discussions of the justification both of truth claims and of practical claims have suggested that we can and should help ourselves to a far richer inventory of materials, such as specific cultural or social beliefs and conventions (‘social construction’) or metaphysically powerful principles. If we were endowed with powerful enough reasoning capacities, both knowledge claims and normative claims might not have to be ‘constructed’ in the first place. If, as Descartes once supposed, reason were ‘complete and entire in each one of us’Footnote 8 we would presumably need only to inspect the clear and distinct ideas that are immediately present to us. However, for present purposes I shall leave aside difficult questions about the available ‘materials’, in order to focus on Kant’s suggestive comments about the ‘workforce’ and the ‘plan’. These, I believe, are the keys to his justificatory strategy.
As Kant sees it, speaking of a workforce is a way of acknowledging the fact that reasoning is not solipsistic, but seeks to reach and engage with others, but without presupposing agreement. Reasoning cannot presuppose agreement, since others may not share the outlook, the ideology or the beliefs of those who seek to offer them reasons: coordination on specific claims or plans of action is therefore not guaranteed. But since agents can affect one another, failure to put forward considerations that others can in principle follow and consider might lead to disaster. Like the failed builders of the tower of Babel, who ended up dispersed and isolated, without even a common language, we might find that ignoring the conditions for offering reasons to others and for following reasons that they offer us would lead to lack of understanding and of coordination, at worst to catastrophe or annihilation. Kant therefore seeks to identify what can be done in this predicament without assuming a shared culture, without assuming shared (supposed) values and without assuming metaphysically demanding premises. He concludes that we can aim to communicate and coordinate even if common standards and principles for coordinating are not ‘inscribed’ in each of us, only if we rely on principles that could be followed in thought or adopted for action by anybody – and so by everybody, and reject those that could not meet this standard.
An obvious difficulty with this approach to justification is that it may offer too little. Kant does not take the easy route of assuming that human beings are naturally or instinctively coordinated, like animals, or that agreement will emerge spontaneously. He had concluded fairly early in his work that ‘since human beings in their endeavours do not behave merely instinctively, like animals, and yet also not on the whole like rational citizens of the world in accordance with an agreed upon plan, no history of them in conformity to a plan is possible (as e.g. of bees or of beavers) appears to be possible’ (IUH, 8: 17). Coordination with others, and following others’ reasoning, is not naturally guaranteed, but rather a matter of rejecting principles that could not be followed by all.
Yet what can a plurality of embodied beings who can affect one another, but are not antecedently coordinated by any given or inbuilt plan or instinct, do to communicate or to coordinate their action? As a first step, Kant suggests that human agents can reach agreement not on a specific plan or blueprint, but more cautiously on the necessary conditions for having some shareable plan. This requires the rejection of principles or starting points that could not be principles for all. This, he concludes, he argues, is all can be presupposed in seeking to offer reasons to others.
This is a limited strategy, but it rules out certain principles, such as those of injury or coercion, that could not be adopted by all because their adoption even by some would undermine agency for some others, and thereby the adoption of like principles by those whose agency was undermined. This spare formulation is the key to Kant’s modal constructivist approach to justification. Putting the point in more familiar terms, reasoning with others is not a matter of presupposing agreement on the relevant matters, but of relying only on principles that can be thought of as principles for all, that is to say, on principles that are fit to be universal laws.
So, Kant’s constructive approach to justification is at least not empty, but it is distinctive and minimal. It takes a modal rather than a hypothetical approach to answering the question ‘what ought we to do?’ This modal approach demands the rejection of principles of action that cannot be principles for all: but no more. It provides reasons to reject principles such as those of coercion, deception, injury or systematic refusal to help others, which cannot be universally adopted. If we try to envisage the universal adoption of such principles, we inevitably fail since anyone who becomes a victim of their adoption by someone else cannot act on those very principles. Slaves cannot enslave others; victims of coercion cannot themselves coerce; victims of murder cannot themselves commit murder; those who suffer systematic lack of help from others do not survive to deny others all help. Anyone who is committed to acting on principles that are fit to be principles for all must reject these and many other principles of action that are not fit to be principles for all. This approach to justification has substantial implications, but may not offer enough to justify all the standards of justice (or all the other ethical standards) that we might hope to establish.
Rawlsian Justifications: Hypothetical Contractualism to Political Liberalism
Rawls takes a quite different approach to the justification of standards of justice. Although his political philosophy is in some ways deeply Kantian, both in A Theory of Justice (where he described it as contractualist) and in Political Liberalism and later work (where he spoke of it as constructivist), Rawls develops a line of argument that differs fundamentally from Kant’s. He does not favour either the agent’s or the recipient’s perspective. He wrote at the beginning of A Theory of Justice ‘My aim is to present a conception of justice which generalises and carries to a higher level of abstraction the familiar theory of the Social Contract as found, say, in Locke, Rousseau and Kant.’Footnote 9 He prioritizes neither agents nor recipients, neither duties nor rights, but structural principles, which, he argues, can be justified as setting out the terms of a distinctive form of hypothetical contract. In anchoring justification in hypothetical agreement, Rawls relies on more demanding premises than Kant had used to justify principles of justice and (other) ethical principles. His appeal to the Social Contract tradition seeks to identify standards that would hypothetically be agreed to under specified conditions.
An appeal to hypothetical agreement evidently relies on less exacting assumptions than an appeal to actual agreement, but it assumes more than the appeal to possible agreement set out in Kant’s universalizability test. The specific form of hypothetical agreement that Rawls thinks relevant for identifying principles of justice posits individuals who find themselves in a hypothetical ‘Original Position’ in which they have general knowledge of their society, but are ignorant of their own position, advantages and disadvantages. The thought experiment of the Original Position articulates a hypothetical standard that would rule out agreements that (for example) depend on deception, manipulation, abuse, conspiracy, partisan influence or various other forms of wrongful action. However, what would be hypothetically agreed can offer a robust basis for justice only if the standards for hypothetical agreement provide a sufficient basis for identifying principles of justice, and it is far from clear that they would do so.
In A Theory of Justice Rawls argued that agents whose own position in society and its advantages and disadvantages were obscured by a (hypothetical) ‘veil of ignorance’ would agree on two specific principles of justice. But he did not claim that an appeal to agreement under conditions of hypothetical ignorance offered sufficient justification.
In Theory of Justice he underpinned his appeal to agreement in the Original Position with a further claim that that the principles hypothetically agreed to under those hypothetical conditions would also be supported by a deeper, coherentist justification. The two principles of justice could, he thought, be brought into reflective equilibrium –into agreement – with ‘our considered moral judgements’ by an iterative process of mutual adjustment. However, he later came to think that this coherentist approach to the justification of principles of justice was unavailable. Considered moral judgements vary hugely, and there is no common set of considered moral judgements to be brought into reflective equilibrium with the two principles of justice. Agreement on considered moral judgements may indeed be found within homogeneous social groups and cultures, but that could provide at most a limited, communitarian justification. Social constructivism cannot offer an adequate basis for justice in pluralistic societies.
In his later writings Rawls therefore reconfigured his approach to the justification of principles of justice. He replaced the appeal to reflective equilibrium with ‘our considered moral judgements’ with an appeal to agreement reached in a specifically political conception of the context of justice. This revised justificatory strategy offered a political rather than a moral justification, and saw principles as just if they would be agreed among fellow-citizens of a bounded, liberal, democratic society. In effect, Rawls’s later work assumes a quite specific sort of liberal and democratic political order, in which there may be no shared ‘considered moral judgements’ and argues that members of such political societies will converge on the two principles of justice (which he did not alter when he revised his approach to their justification). The revised, political starting point that Rawls adopted in his later work does not establish conclusions about wider ethical standards (as Kant had done, and which Rawls himself had assumed were available in his earlier work) but only about justice, and that only for certain sorts of societies. Political liberalism proposes a theory of justice for liberal societies, and is silent about other sorts of society and about the rest of ethics. Its relevance for justice beyond bounded, liberal, democratic societies, including international and global justice, is therefore limited. While Rawls’s later approach to the justification of principles of justice appeals to reasoned agreement under specified conditions, it is explicitly premised on the context of bounded, liberal, democratic societies that defend their boundaries. In his late The Law of Peoples he also argued for a limited conception of international justice that covered many of the aspects of international (i.e. interstatal) justice that had been agreed in the post-Second World War settlement, but again limits the scope of international justice by assuming the political context of societiesFootnote 10 that meet certain standards.
The Justification of Human Rights: Moral or Political?
Contemporary discussions of human rights are also broadly liberal in content, but rely neither on Kantian nor on Rawlsian approaches to justification. Although assumptions that (all or some) human rights standards can be justified are commonly made, and may be correct, much discussion of human rights says very little about justification. Although some philosophical approaches to human rights seek to ground them in conceptions of human dignity or of human interests, the vast majority of everyday and institutional discussions of human rights anchor them in appeals to the authority of specific international instruments, and so indirectly in the authority of states that are party to those instruments.
This reliance on ‘positive’ justifications of human rights standards had particular cogency during and after the Second World War, when the leading human rights Declarations and Conventions were drafted. Securing the most basic standards of justice, that had been so widely and disastrously breached, was of paramount importance, and questions about justification may have seemed less important – or perhaps easy to answer. Today there is still widespread ambivalence about seeking deeper, non-positivist forms of justification for human rights standards. Human rights thought and practice generally converge on a specific list of rights not because there is agreement or clarity about any justificatory arguments for those rights, but because the central human rights Declarations and Conventions, and in particular the Universal Declaration of Human Rights and the European Convention of Human Rights, and numerous derivative instruments, are taken as authoritative, or at least as authoritative once ratified by states. In effect, most discussions that appeal to human rights standards rely (overtly or tacitly) on arguments from authority. While many of those who take human rights seriously assume that they could also be supported by some deeper, reasoned justification, appeals to human rights Conventions and Declarations are regularly substituted for any deeper justificatory arguments.
In relying on arguments from authority, human rights thinking takes standards that are incorporated in state and interstate agreements as the basis for rights, and brackets questions about deeper justification, or treats them as dispensable. One consequence of this persisting reliance on positive or ‘political’ justifications is that claims about human rights standards often make assumptions about the interpretation of rights that are neither explicit nor complete.
All too often human rights claims are asserted without any certainty about who holds the relevant counterpart duties, and in some cases without identifying who holds the asserted rights. The first problem was noted from the early days of human rights, and was sometimes taken to show that human rights standards were ‘merely aspirational’, because they did not show who held the relevant counterpart duties, so failed to show who should do what for whom.
The fact that the UDHR rights were not adequately linked to a clear account of the allocation of the necessary counterpart duties was initially seen as unproblematic for liberty rights, where the counterpart duties must be held by all persons and all institutions. However, this did not show who held the duties that corresponded to other human rights. Moreover, providing a complete account of the duties that correspond even to liberty rights raised problems. For example, while rights not to be tortured or not to be enslaved indeed require all others to respect the counterpart ‘negative’ duties not to torture or not to enslave, these rights also make further demands that cannot be carried by individuals. Duties to support these and other liberty rights, such as duties to enforce or to protect those rights, must be assigned to specified agents and cannot be adequately discharged by all agents. And duties to ‘realize’ social, economic and cultural rights must evidently be allocated to specified (again, preferably competent!) agents if it is to be clear who ought to do what for whom to secure these rights.Footnote 11 Proclaiming rights without specifying which agents or institutions hold the necessary counterpart duties, and without allocating those duties that need allocation to competent duty-bearers, left it unclear what action human rights require of whom.
This problem was addressed in part in the two UN Covenants that aimed to tie down the duties that corresponded to human rights. The Covenant on Civil and Political Rights (CCPR) (1966) and the Covenant on Economic, Social and Cultural Rights (CESCR) (1967) were intended to clarify and to strengthen human rights, ostensibly by assigning certain duties to states. The Covenants addressed a genuine problem, but may not entirely have resolved it. For they did not assign the most significant duties that have to be met to secure the UDHR rights to the signatory states. Rather they assigned those states second order duties to allocate and enforce some configuration of duties that would ensure respect for the rights in CCPR and some configuration of duties that would support the realization of the rights in CESCR. This was made entirely plain in Article 2 of CESCR, which runs:
Each State Party to the present Covenant undertakes to take steps, individually and through international assistance and co-operation, especially economic and technical, to the maximum of its available resources, with a view to achieving progressively the full realisation of the rights recognized in the present Covenant by all appropriate means, including particularly the adoption of legislative measures.
‘Achieving progressively the full realisation of … rights … by all appropriate means’ is a matter of ensuring that unspecified others – such as unidentified individuals or institutions that may not (yet) exist – discharge the complex range of duties needed to secure those rights. Article 2 in effect requires states to construct institutions and to delegate and allocate tasks in order to secure some effective allocation of the counterpart duties. The practical implications of human rights standards remain indeterminate unless and until an allocation of the counterpart duties and adequate identification of the relevant right-holders is articulated.
So, it is hardly surprising that so many discussions of human rights slide from seeing them as fundamental, universal standards of justice to seeing them as depending on institutional authorization. The effect can be both reassuring and worrying. Whereas discussions of justice that address the agent’s question, or questions about just structures, offer relatively clear views of required action, discussions that centre on human rights have often been unclear about the allocation of duties. Human rights offer an indirect perspective on duties that assigns second order duties to states to bring about some allocation of first order duties that would, if observed, secure the human rights that had been declared. But it may or may not have been a good plan to assign these complex second order duties to states. On one view, assigning them to states means that they are held by bodies that (often) have adequate powers to secure respect for and realization of rights. On another view, assigning this task to states is risky, since all too many states are active and persistent violators of human rights.
4. Realities: Interpreting, Enacting and Enforcing Standards of Justice
Justifications matter, but are not all that matters if justice is to be secured. Standards of justice must also be capable of enactment and enforcement in the world as it is, and both enactment and enforcement depend on specific, sometimes contentious, ways of interpreting the standards that are to be enacted and enforced. Critics of liberal accounts of justice sometimes complain about their supposed lack of realism, and in particular about their supposed lack of realism about the effective use of coercive and other power to enforce standards of justice. A popular – in some cases populist – version of this criticism bluntly accuses liberal thinking of mistakenly rejecting Realpolitik. This over-broad criticism misses the point. Liberal political thinking of all varieties aims to justify standards of justice so must reject Realpolitik. However, liberal accounts of justice cannot ignore the need for a realistic account of the interpretation, the implementation and the enforcement of standards of justice. Yet it is far from easy to show what a realistic approach to liberal accounts of justice demands. Here I shall comment only on a narrow range of questions, focusing on ways in which standards of justice are interpreted and implemented in some versions of liberal thinking about justice. Kantian, Rawlsian and human rights approaches to justice each of them take it that a just society must meet a plurality of standards. Those standards are indeterminate and so allow for complex practical judgements that fit the demands of justice to one another, and to the realities of actual situations (conflicts of obligations may sometimes arise, but I leave that well-recognized problem aside). However, indeterminacy can be resolved only if we can show how a plurality of standards is to be satisfied in actual situations, and what roles practical judgement, interpretation and institutionalization are to carry in doing so. These issues cannot be ignored by any account of justice, including any liberal accounts to justice. However, they are addressed in quite different ways in the three exemplary liberal accounts of justice discussed here. None of these accounts of justice assumes that it is sufficient to identify principles of justice, but they take very different views of acceptable interpretations of standards of justice, and of the proper role of institutional considerations in finding adequate interpretations of requirements of justice in actual situations. Liberal accounts of standards of justice therefore all require supplementation by an adequate account of the use of those standards, and in particular of the standards of interpretation that are to be used in reaching judgements about the accommodation of different demands of justice – and where relevant the demands of other ethical standards. This need for supplementary interpretation is particularly striking in discussions of human rights, where multiple rights are frequently in tension with one another (rights to freedom of expression may conflict with rights to privacy; rights to property with rights to food, and so on … and on). The potential for conflict can be averted only by accepting that all, or virtually all, human rights are qualified rather than unconditional rights, and must be adjusted to one another. However, the specific ways in which rights are to be adjusted to one another are mired in unending contention: of which more shortly.
Interpretation is also needed by other theories of justice. Rawls’s two principles of justice prescribe both liberties for all and institutional structures that secure limits to inequality, thereby unavoidably limiting liberties, and once again requiring clarity about the interpretation of the demands of justice in differing contexts. And Kant acknowledges the same realities when he insists that ‘Right (i.e. justice) is Connected with an Authorization to use Coercion’ (MM, 6: 231), and glosses the claim by arguing that:
coercion is a hindrance or resistance to freedom. Therefore, if a certain use of freedom is itself a hindrance to freedom in accordance with universal laws (i.e. wrong), coercion that is opposed to this (as hindering a hindrance to freedom) is consistent with freedom in accordance with universal laws, that is, it is right. Hence there is connected with right by the principle of contradiction an authorization to coerce someone who infringes it. (MM, 6: 231)
In short, Kantian, Rawlsian and human rights accounts of justice all acknowledge the need to shape enactments of justice by interpreting them in ways that secure (at least) consistency. However, they take very different views of permissible ways of adjusting requirements of justice to one another, and in particular of the role of appeals to institutional authority in interpreting standards of justice. Whichever way we look, matters can seem problematic. Practical judgements that interpret the demands of justice must aim not only to respect multiple claims about justice, but also to take realistic account of the action that is (for example) feasible in a given situation, compatible with other requirements that arise in that situation, affordable for those involved and likely to be accepted by others. The everyday demands of practical judgement are often challenging.
Examples of these difficulties are easy to find. They include disputes about whether the interpretation of human rights standards may or must rely on procedures and principles of interpretation that have been incorporated into the legislation or the court procedures of a given state, or of international courts. Are the standards to which these bodies appeal no more than authorized interpretations that avoid, and may run contrary to questions of deeper justification? Might it be important not to link human rights or their interpretation too closely to state or other institutional action? Does the juridification of principles of interpretation solve problems or does it beg questions and lack authority? Can institutional authorization count as justification in some situations, and if so in which?
Many questions arise in selecting a specific interpretation of any right, and indeed of any standard of justice. Some respected ways of resolving these questions appeal only to authority, giving weight to the past determinations of particular courts or to administrative decisions about particular past cases, which they treat as precedents. However, arguments from authority are often contested, and are not decisive. Court decisions may be subject to appeal, and may be overturned. Constitutional lawyers and jurisprudes debate the principles used in reaching those decisions. What is it to balance considerations? Are appeals to proportionality acceptable, or are they a licence for excessive, covert policy-making by the judiciary? Moreover, both courts and other actors need to take account a broad range of considerations including the feasibility, the likely consequences and the affordability of specific interpretations, and the ethical and social acceptability of giving weight to specific interpretations of human rights standards and other standards of justice. Arguments from authority are, it seems, always open to many challenges, and interpretations of authorized standards themselves need justification.
These realities suggest to me that the task of justification is more open-ended than may seem to be the case if we simply rely on appeals to authorized interpretations, and that disputes and uncertainties cannot be resolved simply by appeals to authoritative determinations. Legal and institutional systems are porous, and among other things porous to considerations about justification, about evidence and indeed about wider ethical considerations. Wider ethical and cultural standards may matter in reaching interpretations of standards of justice – as is all too clear from discussions of (for example) conscientious objection and questions about the recognition of certain ‘identities’. These issues are neither new nor well resolved. Current debates about reliance on appeals to proportionality by judges in the UK and the EU as well as long-standing debates about the construction of the Constitution in US courts illustrate the point. And the malign pressures that corrupt or destructive cultures can exert on the interpretation of rights provide darker illustrations. Standards of interpretation cannot be justified simply by designating the interpretations of specified institutional processes as authoritative.
Not everyone will welcome the conclusions to which I am gesturing. They are that justification cannot be directed solely to relatively abstract standards of justice, and must address ways in which those standards are interpreted. This conclusion will be unwelcome to some because justification, including the justification of human rights standards, is demanding, whereas authorization can seemingly be provided by institutional backing. That, however, rubs the point in: authorization may be easier than justification, but it may also be corrupt or deceptive or inadequate in various ways. My sense of the matter is that, while a justification of interpretive standards may be strenuous, it is not optional if we are to take universal standards that liberal approaches to justice support, including human rights standards, seriously.
