Introduction
There has been much recent discussion of a ‘practice turn’ in international theory.Footnote 1 Building on the work of social theorists as well as constructivist International Relations (IR) scholars, Adler and Pouliot aim to compliment existing approaches to the study of international politics with a focus on what practitioners do.Footnote 2 This article takes a side-step from that agenda and makes a contribution from a different angle to the idea of practice in international theory. Without implying the need to endorse or to advocate a new ‘turn’ of any sort, this article differentiates ‘practice’ from ‘a practice’. What follows is primarily an inquiry about the latter, although an enquiry about the latter might ultimately raise questions and challenges for the study of the former.
Adler and Pouliot claim that ‘one cannot fully explain strategic interaction from a practice without first asking where practices come from and how they have been established’.Footnote 3 This article's first, methodological contribution is to argue that this cannot be done validly without first asking what's in, and what's out, of the practice in question. One cannot ask where a practice comes from without a prior account of what that practice is.
Scholars could simply study what everyone ‘does’ and call it ‘practice’; however, this is not good enough. International studies researchers, those both inside and outside of the proposed practice turn, often aim to study the work of particular practitioners rather than the idea of practice in general. For example, one might want to study what human rights practitioners, or humanitarian practitioners, or practitioners of security, or diplomacy, and so on, do. Who and what counts as acting within a certain practice, and who and what does not? Without a method, or set of methods, to answer this question, there is a risk of selection bias. If a researcher studies Doctors Without Borders or the International Committee of the Red Cross – rather than, for example, the US NavyFootnote 4 – in order to arrive at conclusions about what humanitarian practice is, based on what humanitarian practitioners do, then the accuracy of the results generated will depend in part on whether, why, and to what extent those non-governmental organisations (NGOs) and their work are actually cases of humanitarian practice. More broadly, questions such as ‘what is security?’ or ‘what is diplomacy?’ are the sorts of issues that IR scholars frequently ask their students to consider. A focus on practice provides one possible answer to exactly this kind of question, which can then be evaluated against others.
This article's second, substantive contribution is about human rights practice. The article suggests how to resolve the lack of clarity that currently exists about how best to define human rights practice, both currently, and also looking forward toward its ongoing and future construction. The article will develop and use one method, which I call the ‘purpose’ method, in order to locate human rights practice. This method will be based on Rawls has called ‘constructivism in moral theory’.Footnote 5 The article uses an interpretation of the point and purpose of human rights practice to challenge the view that human rights practice is necessarily state-centric, in the sense of being inextricably bound up with states and practices of sovereignty.
The argument will proceed as follows. The first section briefly introduces the single-authored work of Pouliot as well as his co-authored work with Adler.Footnote 6 It critiques Pouliot's view that practice-based logics of action can be distinguished from the logics of consequences and appropriateness (on the specific grounds that the former, according to him, does not involve the conscious representation of reasons for action, whereas the latter two supposedly do).Footnote 7 In fact, agents use concepts as labels to represent practices, and in this specific sense, representation and practice are compatible. Practitioners themselves can distinguish between the different practices within which they act – for example, practising parenting, medicine, religion, security, diplomacy, or human rights – and researchers should be able to make these distinctions as well. The use of a unique concept to describe a cluster of human activity is a prime facie indicator that a unique practice exists, but this indicator is not necessarily the best or final answer about what a practice is.
The second section presents four methods that can be used to locate a practice: the ‘agreement’ method, the ‘contextual’ method, the ‘value’ method, and the ‘purpose’ method. I shall apply each method to the question of how to locate human rights practice. In the process, I shall provide an overview of the most significant strengths and weaknesses of each one. According to the agreement method, practices can be located by referring to speech acts, for example, treaties that refer to a practice and provide rules for defining what it is. According to the contextual method, practices can be located by exploring the intentions and priorities of practitioners who use political concepts to advance projects that are contemporary in their time. According to the value method, practices can be located by looking for an underlying value or principle that a cluster of human activity aims at instantiating or achieving. According to the purpose method, practices can be located by taking the perspective of an idealised practitioner who has reasons to construct a practice in order to fill a social or political gap, and one asks, from that person's perspective, what to do. There are reasons to think that the purpose method responds well to some of the key weaknesses of the others.
The third section applies the purpose method to the question of whether assigning responsibilities to non-state actors can occur meaningfully within human rights practice. The purpose of practices of sovereignty is to decide on limits and rules that agents in the international system must accept in order to enable the possibility of international relations in a sovereign-states system. It might be the case that human rights standards represent some of these limits. However, even if this is true, all of these questions are about practices of sovereignty. Human rights practices have a different purpose. The purpose of human rights practice is to assign specific responsibilities to the right agents, for the right reasons, to protect and to provide for individuals' human rights, such that every human is covered under the scope of some agent's human rights responsibility. The reasons that practitioners might have had to assign human rights responsibilities to sovereign states can be reconsidered, in today's circumstances, to see if those same reasons would lead practitioners, of the right kind, to identify non-state actors as the bearers of responsibility to protect and to provide for human rights. This application of the purpose method provides a good way to ask whether non-state actors can be responsibility-bearers within human rights practice.
‘Political’ theories of human rights, such as those advanced by Rawls, Beitz, and Raz, begin with the practice of human rights, rather than with abstract philosophical accounts of objects to which all people are entitled.Footnote 8 They then argue that the central purpose of human rights practice, in light of which its location can be determined, is tied to states and limits to their sovereignty. This article provides a different and original interpretation, which considers the meaningful possibility of assigning and attributing duties to protect human rights to non-state actors in some political contexts. Finally, the article provides suggestions for future research into the location and nature of human rights practice, based on taking the perspective of practitioners who have certain reasons to construct specific practices. The framework can easily be adapted and applied to answer questions about the location of practices other than human rights.
Practice and practices
In order to locate the contribution of this article within the existing IR literature on practice and practices, there are many places that one could feasibly begin. These include: constructivist and post-structuralist perspectives that draw from the social theory of Foucault, Bourdieu, and Schatski; MacIntyre's Aristotelian notion of practice; Searle's philosophy of social reality; or Oakeshott's account, which by now has been incorporated into the English School of IR, of practical associations.Footnote 9 This article begins with the recent work of Adler and Pouliot. This is because their co-authored work is an attempt to synthesise their own single-authored contributions, as well as those several of the others just cited, into a larger framework: one which has, in a relatively short period of time, already made a significant scholarly impact.Footnote 10 They provide the following usefully broad definition:
Practices are socially meaningful patterns of action, which, in being performed more or less competently, simultaneously embody, act out, and possibly reify background knowledge and discourse in and on the material world.… [They] are patterned actions that are embedded in particular organized contexts and, as such, are articulated into specific types of action and are socially developed through learning and training.Footnote 11
This is slightly different from the familiar Anglo-American conception of practice as elaborated by Rawls, for example, in the first footnote of ‘Two Concepts of Rules’ (1955).Footnote 12 But there is still enough core overlap – practices are activities that structure and organise social action for participants – to suggest one concept of practice as a plausible starting-point, from which to begin to answer the question of what practices are.
This is fine as an abstract definition, but it does not say much – nor does it aim to say much – about how to locate the specific, named practices that the scholars who suggest a ‘practice turn’ aim to discuss. There are two problems with the interpretation of practice currently on offer. The first problem, which is clarified by analysing Pouliot's single-authored work, centres on his claim that practitioners do not ‘represent’ their actions consciously.Footnote 13 This overlooks that we have conceptual labels that do represent practices, and that these labels matter. The second problem, which becomes apparent when examining the co-authored work of Adler and Pouliot, is that practitioners themselves are not necessarily the best or final arbiters on how to define and to locate the practices that they practise.Footnote 14
Pouliot contrasts the ‘logic of practicality’, as he calls it – I prefer the expression ‘logic of practice’ – with the two logics of social action most familiar to IR scholars and political scientists: the logics of consequences and appropriateness.Footnote 15 The latter are, respectively, at the root of rationalist and constructivist explanations of social action.Footnote 16 According to Pouliot, agents' action can be explained through these two logics by looking for a set of detached interests/preferences or identities/norms. Agents themselves might refer to directly their interests, preferences, identities, or norms when taking decisions. Otherwise, social scientists can look to these in order to explain action as if agents themselves had referred to them.Footnote 17 For example, consider how best to explain the actions of a medical doctor.Footnote 18 According to the logic of consequences, the way of explaining a doctor's actions would be to posit that there is a set of outcomes – the health of the patient – that the agent wants to maximise. The doctor's actions can be explained and predicted by assuming that the doctor will rationally consider all available information, and use that information in order to determine how to advance that outcome to the greatest extent possible. According to the logic of appropriateness, the way of explaining a doctor's actions would be to posit that the doctor has a social role – ‘doctor’ – to which attaches a set of normative criteria about how to act well in one's role (what counts as being a ‘good doctor’: for example, maintaining doctor-patient confidentiality, even if doing so fails to maximise the patient's overall health). The doctor's actions can then be explained and predicted by assuming that the doctor will engage in logically valid interpretation about how best to act out the requirements of her role.
The logic of practice jettisons the idea that agents take decisions, or can be treated ‘as if’ they take decisions, in such a reflexive, logical, and analytical way. According to the logic of practice, as Pouliot develops it, it is the doctor's circumstance-based intuitive knowledge – based in the experience of practising medicine in the times and places, and in the communities that the doctor has practised it so far – that will explain what the doctor does.Footnote 19 This practice-based logic of action is offered to compliment, rather than to replace, existing theories, which hold that agents engage in conscious deliberation about consequences and/or obligations.
At this point, the first problem crystallises. Within practices, human agents assign labels to what they do. If one asks practitioners what they do, it is rare to find one who simply says ‘I practise’. Doctors practise medicine. Diplomats practise diplomacy. Islamic people practise religion: more specifically, Islam. Practitioners practise something. They do not just ‘practise’, but rather, they act within a practice. The indefinite article is a small change that makes a big difference. Within the logic of practice, agents use meaningful conceptual labels to define and to delimit practices. That a conceptual label exists, and is employed by agents, to distinguish different practices from one another – for example, Islam from Christianity, and Protestantism from Catholicism, Sunni from Shi'ia, reform Judaism from orthodox JudaismFootnote 20 – is a strong prima facie indicator that a distinct practice, going under this conceptual label, exists, and is a coherent object of analysis.
Calling a practice by its own name rather than by another name, representing it in this specific sense, helps both agents and the social scientists who study them to distinguish which experiences and knowledge-sets are most usefully employed in order to decide what to do. Practical knowledge of fatherhood, for example, does not necessarily translate into practical knowledge of diplomacy, or policing, or medicine. Agents themselves can make these distinctions between the different practices within which they act and decide, and so should scholars.Footnote 21 In IR specifically, we should have an account of how to decide whether and when agents are acting within practices of diplomacy, or security, or environmentalism, or humanitarianism, or human rights, in order to discuss and to defend the validity of research that aims to study these practices.
There are three possibilities about the relationship between practice-concepts and practices themselves. Firstly, and most straightforwardly, one concept could pick out one practice. Secondly, if there is no difference whatsoever, either in kind or in category, between one putative practice and another, then some of the conceptual labels are extraneous, do not represent distinct practices, and can be usefully removed from our social and political vocabulary.Footnote 22 Thirdly, and by contrast, if we are using one concept to refer to two distinct kinds of activity – see Berlin's famous analysis of ‘Two Concepts of Liberty’ – then this suggests that we need additional labels in order to capture and to represent practices that are actually distinct from one another.Footnote 23
This discussion leads to the second problem. Individual practitioners themselves, who act within practices, and use a concept – a label for a practice – to define and to understand what they do, are not the final arbiters on whether they are actually acting within that practice. There is a distinction between whether agents practise well (which is not what I have in mind), and whether they are acting in a practice at all, despite thinking that they are. A doctor who performs surgery after a good night of sleep is usually able to practise medicine better than a doctor who stays up all night partying before going to work. This example highlights the appropriateness-based logic of what it means to act well within a role, identity or practice. By comparison, one could argue that letting blood in order to cure fever does not count as practising medicine, because part of practising medicine is the ability to distinguish (through practice) between procedures that work and procedures that do not. To use another example, the individuals who flew planes into the two towers of New York's World Trade Center on 11 September 2001, might have thought that by doing so they were practising Islam. It is available, however, for others to argue that not only were the individuals not acting well as Muslims, but moreover – and more significantly for the present purpose – that intentionally murdering others is not an act that can occur meaningfully within the practice of Islam, properly understood, at all. These are strong reasons to reject the claim, advanced by Adler and Pouliot that ‘defining what counts as an international practice and what does not is best left to practitioners themselves in their actual performance of world politics’.Footnote 24
So far, we lack a set of methods that can be used in order to settle these questions about the location of practices – what's in and what's out? – one way or the other. How, and according to which method, could one go about deciding who is right, and why, about how to locate the boundaries of a specific practice? When one wants to study what human rights practitioners do, how does one know that one is actually studying human rights practitioners rather than agents who just ‘practise’? Beyond human rights, the same question can be asked of any international practice. The fact that their activities have come to be clustered under the label ‘human rights practice’ is a crucial starting-point: a clue that there might actually be a distinct and identifiable practice that goes under that concept. However, this starting-point is not necessarily the best and final answer. One needs a method, or set of methods, in order to answer such questions in a more rigorous way. The next section provides this set of methods.
Locating practices
There are four ways to test the prima facie assumption that the existence and widespread use of a political concept to represent a practice actually points to the existence of a unique social practice that correctly goes under that label. I call these the ‘agreement’ method, the ‘contextual’ method, the ‘value’ method, and the ‘purpose’ method. This section uses the question of how best to locate human rights practice as the example to illustrate and to discuss the merits and drawbacks of each method. The section makes a case for the purpose method. But it is a certain kind of case: one based on my evaluation of the strengths and weaknesses of each method for the sake of locating human rights practice. I encourage other scholars to consider all four of the methods in their own research projects, but to do so more reflexively, and more cognisant of the need to respond to potential weaknesses. All of this lays the groundwork for the subsequent section to apply the purpose method in more detail to the question of how to locate human rights practice.
Facts, values, and practices
It might seem at first glance that the agreement and contextual methods will locate practices empirically, whereas the value and purpose methods will ‘merely’ allow one to answer normative questions about what a practice requires participants to do. However, such an interpretation would obscure the actual analysis. Each of the four methods presented in this section responds to this same constitutive question: what is ‘the practice’?Footnote 25 They all provide ways of including or excluding people and activities from the definition of something called a practice. Each method offers a very different kind of answer, which includes some things while excluding others. This is precisely the point.
Separately, it is important to note that it would be a prima facie strength of any of the four methods if it can account well for a practice's normativity.Footnote 26 Recall the discussion in the previous section. If one's concern were just to locate practice – about who is doing activity in a general sense – then it would be possible to think about the location of this purely empirically. However, the same does not apply to specific practices. Part of what defines a practice according to both Bourdieusian social theory and Rawlsian political theory is that it conditions and enables judgements about what participants need to do in order to act successfully within it, ‘as’ a practitioner of this or that. What a practice is is therefore bound up with the action that seems to be called for within it in particular cases. A method that does not account for this internal normativity might have identified ‘practice’, but it is unlikely to have successfully located a specific practice.Footnote 27 The respective answers that each method offers to the question of what a specific practice is can therefore be assessed, in part, as stronger or weaker, based on how well they capture this internally normative feature.
In addition to this internal normativity, practices can also have external normativity. Human beings, although we can be ‘programmed’ by logics of action in ways that are still being explored and understood, still have the capacity to act autonomously on the world.Footnote 28 We can act in ways that run contrary to rules or habits. More specifically, we can offer, challenge, and reflect on our reasons for engaging in certain practices.Footnote 29 Those reasons are not always sound from an objective normative perspective, but this is beside the current point. Beyond an intuitive sense that specific actions ought to be taken within a practice (internal normativity), practitioners can, on reflection, offer reasons for why they engage in their specific practices. They think that they ought to practise (external normativity). I should quickly say that there are bound to be exceptions to this claim; however, the most paradigmatic international practices – security, diplomacy, human rights, religion, strategic bargaining, even terrorism – are not obviously among those exceptions. It is therefore a prima facie strength of a method, though not necessarily an overriding strength, if it can locate practices that practitioners seem to have reason(s) to practise. By contrast, theorists who locate practices in a way that can make no sense of the reasons offered by practitioners for engaging in their practices seem less well connected to the actual features of the world of practice than the language of a ‘practice turn’ would initially suggest.
Agreement, contextual, and value Methods
The first method is the ‘agreement’ method. The agreement method is one way to locate the boundaries of a practice. In order to find a social practice, according to this method, one needs to look for, and to interpret, written and/or verbal speech acts, which explicitly delimit a particular practice. Beitz offers an account of the idea of human rights that relies on the agreement method.Footnote 30 He defines human rights as ‘an emergent discursive practice consisting of a set of norms for the regulation of the conduct of governments and a range of actions open to various agents for which a government's failure to abide by these norms supplies reasons’.Footnote 31 The rationale for why human rights practice is directed at the conduct of governments, rather than the conduct of other (non-state) actors, is based in what was agreed in international law in the form of treaties, covenants, and enforcement mechanisms, most clearly crystallised within the United Declaration of Human Rights (UDHR), the International Covenant on Civil and Political Rights (ICCPR) and the International Covenant on Social, Economic, and Cultural Rights (ICESCR).Footnote 32 To be clear, he is explicitly against the idea that agreement is the best way to determine what human rights there are, that is, what belongs on the list of objects that human rights practice is concerned to protect.Footnote 33 It is at a different level, the level of what constitutes human rights practice itself – and why, for example, it should be thought of as directed against governments – that his argument is based in an account of what international actors came together and agreed.
The agreement method is associated to some extent with positivism in legal theory. Hart, in a seminal text on legal positivism, argues that the law – a social practice – can be identified by looking for a combination of authoritative primary rules, which tell agents how they ought to act, and authoritative secondary rules, which tell agents how to assess, to recognise, and to change the primary rules.Footnote 34 The agreement method, often to the exclusion of all others, is used by most international lawyers to locate human rights practice. International lawyers look at texts, mainly in the form of treaties, in order to determine what human rights and their associated responsibilities are.Footnote 35 On this view, one can know that humanitarianism, human rights, and environmentalism are distinct practices, because each has its own branch of international law. This method, in turn, can be used to determine whether and when agents' actions fall inside or outside of practices. For example, the ‘business and human rights’ activism of NGOs such as Amnesty International, the Center for Constitutional Rights, and EarthRights International – aimed at holding business actors to account for violations of the normative standards associated with human rights – might be defined, by the agreement method, as operating outside of human rights practice. This is because, according to this method, the practice is constituted, through agreement, by a focus on states as the primary human rights responsibility-bearers. This would remain so unless, firstly, the subjects of international law can agree, in a way that successfully becomes a part of international law, that the human rights regime should be extended to include businesses as responsibility-bearers; and/or, secondly, it can be demonstrated that non-state actors already have international-legal human rights obligations.Footnote 36 A move toward agreement on these issues is arguably in process at the UN Human Rights Council, in which the business and human rights policy agenda has been led by John Ruggie since 2005.Footnote 37
One advantage of the agreement method is that it offers a clear reason, based on the idea of consent, as to why actors ought to follow the action- and policy-guidance that falls under a specific practice. In other words, it identifies an object as ‘the practice’ that already has a claim to external normativity. There is a bridge between analytical statements such as ‘human rights practice requires action A or policy P’ and normative statements such as ‘agents actually have reason to do A or to implement P’. There are important objections to the idea that the agreement or consent of a majority successfully creates normativity. The agreement of some might be detrimental to others, especially when those others cannot offer meaningful consent.Footnote 38 But, in the agreement method, at least there is an obvious bridge between the analytical and the normative that can be assessed according to the well known benefits and drawbacks of the idea of consent as a basis of obligation.Footnote 39
Overall, this article does not endorse the agreement method. It can be criticised on the following grounds. Either: It needs to limit itself to making claims such as: only those texts that make explicit claims about primary and secondary ‘human rights’ rules and standards (that is, beginning with the publication of the UDHR) are part of human rights practice. This would exclude earlier natural-rights texts such as the American and French constitutions, and positive-rights texts such as the body of English common law beginning with the Magna Carta, from the scope of human rights practice.Footnote 40 This narrow focus on the exact phrase ‘human rights’ in texts that meet legal-positivist standards of recognition also would exclude pre-World War II agreements to protect minority rights and labour rights; these were central concerns of the interwar period.Footnote 41 In short, this method can offer a clear account of what counts as part of the practice, but in a way that fails to capture elements that seem, at first glance, to be at the practice's core. Or: If those who rely on the agreement method of locating a practice choose to include some of these other texts, despite the lack of explicit and exact reference to ‘human rights’, then the method lacks a principled basis upon which to select which, if any, of those texts belong within the same practice as the UDHR. Accepting even a handful of texts that do not have the exact phrase ‘human rights’ in them starts users of the agreement method down a slippery slope that would make every historical legal text aimed at advancing human well-being a ‘human rights text’. A similar point can be made about other, non-written kinds of speech acts. Different methods are therefore needed in order to find an answer to this dilemma that does not rely on pure, unfounded intuition about what human rights practice is.
The second method is the ‘contextual’ method. The contextual method rejects a narrow focus on what has been agreed in written and verbal speech acts. It focuses, rather, on the meaning of concepts in the political and historical contexts in which they arise and operate. More specifically, the method requires the study of the background knowledge and intentions of the actors who use ideas, in order to advance political projects that are contemporary in their time. This way of approaching the construction and meaning of political ideas is frequently associated with the early work of Quentin Skinner, as well as other founders of the Cambridge School of political thought such as Pocock.Footnote 42 It can also be associated with the IR constructivist focus on the work of social activists, whose entrepreneurship – including attempts at socialisation, persuasion, and framing – creates the norms that only result in interstate agreement at a later stage in a complex and potentially long process or ‘lifecycle’.Footnote 43
This method is the most directly relevant of the four to the practice turn's proposal that the study of International Relations is intertwined with the study of what practitioners do. Barnett, for example, has done influential work that studies humanitarianism by closely examining the work of organisations such as Doctors Without Borders and the International Committee of the Red Cross. He looks at the ways in which the nature of humanitarianism has ‘transformed’ (for better or worse) due to the shifting beliefs, values, and priorities of these actors.Footnote 44 Similarly, Hopgood has produced a fascinating study of Amnesty International. In it, he describes the organisation's angst over the question of whether human rights practice is to be defined and implemented in terms of the production of transcendental authority – a quasi-religious faith in absolute values – or whether it is a consequentialistic and pragmatic attempt to secure, for individuals, the objects of their rights.Footnote 45 The point is that what these actors do – the way that they use concepts, and assign meanings to those concepts, in order to advance political projects – can be viewed as constituting (respectively) humanitarian practice and human rights practice. Furthermore, the definition and boundaries of such practices can change over time, as practitioners make adjustments to their decisions and actions.
The prima facie assumption that a unique concept picks out a unique practice is, in this way, testable by a history-of-ideas analysis of when a particular concept actually came into widespread political use in order to advance social and political projects that are contemporary in their time. The agreement method would locate the beginning of human rights practice in the period from 1945 to 1950, because it would define the boundaries of that practice in terms of speech acts such as the UDHR and subsequent international agreements in the twentieth century. By contrast, the contextual method would look for the beginning of the adoption and use, by social activists, of the specific compound concept ‘human rights’ in order to describe and to cluster their activity. The origins of this entrepreneurship can be found most prominently not in the late 1940s, but rather, in the early 1940s. For example: H. G. Wells explicitly organised his political activism around the idea of ‘human rights’; US President Franklin D. Roosevelt introduced the term into official political discourse in 1941 in connection with his ideas about the four freedoms; the idea's use became more widespread in 1943, up to and including the release of a US Government document on ‘Human Rights and the World Order’.Footnote 46 UNESCO then commissioned and published a volume illuminating the way that political thinkers interpreted this concept – which in some ways, was linked to the earlier traditions of ‘natural rights’ and the ‘rights of man’, but which was understood, by those thinkers in their own time, to be genuinely innovative – around the time of its textual codification in 1948.Footnote 47
I shall now examine strengths and weaknesses that one might attribute to the contextual method. One potential strength is that, as with the agreement method, it provides a clear, empirically grounded way to locate an object as ‘the practice’. Many scholars seek to justify social-scientific claims to knowledge about any given topic, such as the location of a practice, by developing testable hypotheses, which can then be proven as right or wrong through research.Footnote 48 The agreement method generates clear, falsifiable claims: either agreement happened or it did not. This is arguably one of the agreement method's prima facie strengths, even though this article did not endorse it due to its overriding weaknesses. The contextual method seems at first glance to duplicate this strength, while at the same time overcoming the agreement method's blind spot for the non-inclusion of the social activity that leads up to agreement within the definition of the practice. Careful readers might already have anticipated the problem with this argument. The contextual method, when it is applied to locate practices, works through a (discredited) verificationist method of hypothesis-testing rather than through falsification.Footnote 49 Take, for example, the claim that human rights practices involve the international community's assignment of responsibilities primarily to states and not to non-state actors. This would be easy to verify by assembling a body of contextual evidence that supports this interpretation of activists' intentions in constructing a practice, associated with the concept of human rights, at a certain point in history. The claim is also remarkably easy to falsify by finding other activists whose intentions and use of the concept are different. I can think of no interesting and meaningful contextual-method-based claims about the location of human rights practice that could withstand a falsificationist standard of justification. There will always be at least some activists with different understandings and different intentions. Scholars who are drawn to purely empirical methods for locating practices are often the same ones who tend to think that the contextual method is the strongest, especially vis-à-vis other explicitly non-positivist examples such as the purpose method. However, the contextual method does not then fit with a basic standard of justification for positivist research.
There is a potential response to this, based on the fact that only some norms and practices become widely adopted whereas others do not, and that the widely adopted norms and practices are likely to be the right ones because they have gone through a quasi-democratic, intersubjective selection process.Footnote 50 This, however, is clearly a collapse back into the agreement method.Footnote 51 In the final analysis, the contextual method is a helpful way to locate ‘practice’, when that term is understood (without the indefinite article) as the description of social action in terms of activity that practitioners do. However, it becomes less helpful if one wants to locate and to study a specific international practice, such as human rights, because we come straight back to the same problem of which putative practitioners are right, and which are not, about their understanding of the location of that practice. An explanation of how a practice came about is not necessarily the strongest way to locate a practice for the sake of further research into it.
One potential weakness of the contextual method is that it cannot account for the normativity of the objects that it identifies as practices. The agreement method can, at the very least, rely on the notion of consent in order to account for why agents ought to follow whatever practice-specific guidance the method locates. But with the contextual method, there is no apparent bridge all – save for the quasi-democratic collapse into the agreement method just mooted – between analytical and normative statements. That certain networks of people engage in, and collectively define, a particular kind of social activity is not a sufficient normative reason for them, or for anyone else, actually to do it. This article does not, of course, dispute the fact that there can be practices that no one should be engaged in. My point, however, is a slightly different one. The contextual method can be used to identify all kinds of human activities and habits, much of which no one would seem to have any reason to follow as a structured practice. By reading out the normativity that international practitioners themselves understand to be at the core of their practices, the contextual method seems insufficiently connected to the reality of the objects that it aims to identify.
The third method is the ‘value’ method. According to the value method, a practice can be located by identifying a unique value or principle that a cluster of social activity is aimed at advancing or achieving. If this can be found successfully, then any human action – regardless of the historical context in which this action occurs – aimed at advancing or instantiating this same value or principle can be identified as part of the practice in question. One's original conception of the practice can then be revised in light of this new information about how best to define it. This method can be associated with Dworkin's interpretivism, which contains a three-step method to identify, interpret, and reconstruct a practice in light of its fit with a specific value.Footnote 52 It can also be unpacked in light of Berlin's perhaps more widely known work on the concept of liberty. Berlin argued that ‘liberty’ is two concepts rather than one, because ‘negative liberty’ aims at advancing autonomy and non-interference, whereas ‘positive liberty’ aims at advancing self-rule and self-actualisation.Footnote 53 Different values imply different concepts, and this can often require different practices that might conflict with each other, both at the level of action-guidance and at the level of principle. Dworkin essentially provides a value-monist version of the same method, according to which different practices and values, for example, security and human rights, must be made to cohere both with each other and also with higher-order values.Footnote 54 A key point for this article is that, according to the value method, the practice of human rights can exist irrespective of whether practitioners have a concept of human rights. In this respect, the value method is quite different from the other two methods examined so far.
Starting with the prima facie assumption that the widespread use of the ‘human rights’ concept during the World War II era and in its aftermath locates a distinct practice, one could then test this assumption by trying to identify some value such as the equal moral worth of individuals (Dworkin's view) or respect for the values that constitute personhood (Griffin's view) as the best possible normative justification for that cluster of social and political activity.Footnote 55 If one can successfully locate a value of this kind, then one can use the value to broaden one's conception of human rights practice to include other kinds of historical and contemporary socio-political activity, which aims at advancing that same value – regardless of the non-use of the specific compound concept ‘human rights’ by practitioners and political thinkers to describe and to cluster their action. The eighteenth-century concern with the ‘rights of man’ and/or the broader historical tradition of ‘liberal rights’ could be read back through history, in a way that would be anathema to the contextual method, and included within one's conception of human rights practice, due to the way that the activities associated with these ideas advance a single specific value.Footnote 56
The value method unmistakably provides the bridge between the analytical and the normative that the contextual method lacks. However, the way in which this bridge is provided raises a new problem. Assume, for the sake of argument that the value method ends up successfully identifying the equal moral worth of individuals as the value that allows one to locate human rights practice. It might then be advisable for each agent to ignore the practical guidance that comes specifically out of human rights practice, preferring instead to go directly to the value, and to interpret for him- or herself what equality requires. If this method of identifying a particular object as ‘the practice’ in question is sound, then once one gets to the value, the practice itself can drop out of the picture. It becomes unclear why agents would follow the ‘logic of practicality’ rather than a reflexive logic of doing what best advances the value that helps to define and to locate the practice in the first place.
Furthermore, there are actual and conceivable practices, other than human rights that advance the value of the equal moral worth of individuals. Humanitarianism, charity, and development assistance are all candidates. The value method therefore runs the risk of blurring together practices that, in practice, are notably distinct. International human rights law, for example, assigns duties to states to protect a broad range of civil-political and socioeconomic normative standards, whereas international humanitarian law places special focus on limits to action by combatants toward each other and toward non-combatants in the context of armed conflict. (Humanitarianism itself can, of course, be viewed as much broader than its international-legal instantiation.) Otherwise, the value method runs the converse risk of accepting that various practices, such as human rights and humanitarianism, can be associated with the same value, but without giving any importance to those practices. Normative theorists who believe in the practice-independence of moral principles would be happy with this picture.Footnote 57 However, taking a broader view, there is a certain counter-intuitiveness about beginning by asking how to locate a practice, and then moving to the conclusion that the practice itself can be ultimately disregarded in favour of asking normative questions about how to act in line with abstract values.Footnote 58
The purpose method
The fourth method is the ‘purpose’ method. This method is an adaptation of what Rawls has called ‘Kantian constructivism in moral theory’.Footnote 59 I shall therefore outline the most basic contours of moral constructivism, as Rawls understood it, before proceeding. Consider this quotation:
What distinguishes the Kantian form of constructivism is essentially this: it specifies a particular conception of the person as an element in a reasonable procedure of construction, the outcome of which determines the content of the first principles of justice.Footnote 60
Rawls has characterised the operative conception of the person, for the purpose of constructing the principles of justice for the basic structure of a society, as the ‘free and equal moral person’ behind a veil of ignorance.Footnote 61 It is possible to interpret this approach in at least two different ways, and I shall adopt the second. Firstly, one could focus on the plain fact that Rawls requires idealised agents, who do not actually exist in our world – in which we are fairly attached to our own personal circumstances and perspectives – in order to provide the perspective from which justice can be constructed. Indeed, Aaron James has listed, as one of the steps of Rawls' constructivist method: ‘draw a veil of ignorance’.Footnote 62 One recent example of this interpretation in IR literature is provided by Renee Jeffery, who draws from the idea of the veil of ignorance in order to characterise Rawls's approach as ‘extreme rationalism’.Footnote 63 Actual people have considerable difficulty abstracting away their identities in the way that Rawls's approach seems, on this interpretation, to require. This leads to the objection that Rawls stretches his approach beyond the bounds of practical possibility and usefulness.Footnote 64 Followers of Habermas, for example, often suggest that we can construct just (and justified) practices by idealising the structure within which people decide, but not the people themselves.Footnote 65
However, even though the veil of ignorance is an important feature of Rawlsian approaches to justice, it is not necessarily one of the steps that one must take to extend Rawls's constructivist method to other practices. There is a second interpretation, as follows. The conception of the person who would have the authority to take moral and political decisions within a practice is subject to the reasonable agreement of actual people. According to Rawls' constructivism, as I interpret it, actual people in actual conditions are required to appoint a non-real person, conceptualised in a particular way, as an epistemic authority. Then, in order to take substantive decisions about what to do, those same (actual) people can then ask themselves ‘what would a person, conceived in such a way, decide to do?’ The free and equal moral person, behind the veil of ignorance, whose purpose is to discover principles to which all other people conceptualised in this way could reasonably agree, is the conception of the person for constructing the practice of justice. This is due to the specific contours and purpose of that practice. Practices other than justice might require different conceptions of the person from whose standpoint that practice can be located and constructed. In fact, difference should be expected. There is no a priori reason to think that the free and equal moral person behind a veil of ignorance is the appropriate authority on decisions about the location of human rights. The key point is that the description of the person from whose standpoint a practice is constructed does the heavy lifting, according to this method.
A famous episode from Plato's Republic helps to illuminate this point.Footnote 66 Prompted by Socrates, the characters set out to answer the question ‘what is justice?’ Before they can do this in a substantive way, they first must establish that they are discussing the same subject matter: the same sphere of human life. Thrasymachus offers the view that justice is best located (as the starting-point) by looking for a practice of the strong dominating the weak. However, in saying this, Thrasymachus reveals himself as an unsuitable candidate to participate in a conversation about how to locate justice. The substantive conclusions that he would arrive at, if any, would be about something other than what everyone else is talking about. He therefore needs to leave the table before the discussion can fruitfully begin. Plato's substantive conclusions often overshadow this method, which has some elements that are echoed in Rawls: only certain parties are admitted into a discussion about the nature of justice, for reasons having to do with their successful recognition of the initial location, point, and purpose of the practice under discussion. As characters in a dialogue, these are non-real people. But actual people in the actual world can still have meaningful debates about what Glaucon, Socrates, or Thrasymachus would decide, given who they are, and given their purposes in constructing a practice. Think about the kind of debate that occurs all the time about what a parent, or particular friend, or retired colleague, with certain concerns, commitments, experiences, and character traits, would do about a given issue, if only they were here. There is nothing ‘extremely rationalistic’ nor unrealistic about this process.
According to the purpose method, practices are defined and located not by the practical judgements of actual practitioners. Rather, they are located by the judgments that we think that idealised practitioners, with certain purposes in constructing a specific practice, would arrive at. A substantive account of who these practitioners are is bound to be contestable, but that is not an excuse not to provide one, for others to consider and to debate. The next section of this article does this, using the example of human rights practice.
In the following ways, the purpose method does not contain the weaknesses of the agreement method (that consent, by itself, is a problematic normative foundation for practical action-guidance; and that elements of the practice are either uncritically excluded or otherwise included in a way that introduces a slippery-slope issue), the contextual method (which identifies objects that, unlike our actual practices, have no clear link to normativity), and the value method (according to which practices themselves drop out of the picture, in favour of agents pursuing values directly). The idea of a purpose robustly makes sense of practitioners' reasons to practice. However, the original practitioners might – under the circumstances just described – decide to set up authoritative rules, guidelines, and limits that practitioners in today's world have to work within. This enables the advancement of a purpose through the establishment of a practice, rather than relying on agents' reflexive and more direct pursuit of values or outcomes.Footnote 67 Although not the end of the story, this seems to be an important step, and an improvement on other methods for locating human rights practice.
The purpose of human rights practice
This section applies the purpose method in order to show how it can adjudicate competing claims about the nature of human rights practice. It distinguishes the point of the practice of sovereignty from the point of the practice of human rights, in order to critique the widespread view, in IR and international law, that human rights practice necessarily places primary duties on states. The idea that sovereignty and human rights are bound together in this way has been recently reaffirmed at the international level, in the form of the Responsibility to Protect policy framework, which outlines how and why sovereignty should be viewed as conditional on states' treatment of residents.Footnote 68 This putative link between human rights and states' sovereignty also drives most so-called ‘political’ theories of human rights.Footnote 69 An analysis of this is significant, because of the amount of contemporary policy work and practitioner effort, as introduced above in the analysis of the agreement method, to bring non-state actors in to an international-legal human rights framework that seems to be designed to assign duties to protect human rights to states.
Is the ‘business and human rights’ activity of human rights practitioners necessarily a part of human rights practice just because those practitioners do it? Do responsibilities to protect human rights actually fall on non-actors such as private military and security companies, especially in zones of weak or mixed governance?Footnote 70 The textual method might uncritically say ‘no’ to both questions. In trying to expand the scope of duties to protect human rights to non-state actors, a self-identified human rights practitioner cannot, on this account, be acting successfully within human rights practice, because of the centrality of states to that practice according to most interpretations of positive international law. The contextual method might uncritically say ‘yes’ to both questions. To the extent that practitioners have changed the targets of their actions to include non-state actors, this straightforwardly means, on this account, that human rights practice now includes the duties of those actors. However, the reason for the answer to the second question would rest, in a problematic way, on the answer to the first. The value method might say ‘no’ to the first question but ‘yes’ to the second. From a normative perspective, we all have moral duties not to cause serious and avoidable harm to others. However, an application of the value method in this way begs the question of whether these universal moral duties really constitute the location of a particular international practice called ‘human rights’. This section applies the purpose method to suggest that actual practitioners are not the final authorities on whether they are acting successfully within human rights practice. The answers to questions about the location of human rights practice lie in the judgement of idealised (rather than actual) practitioners, defined in such a way that they have particular reasons to which they refer when constructing and reconstructing a practice.
The purpose method has three steps. Firstly, one needs a lock on the people who have reason to fill a gap in socio-political life by creating something called human rights practice. Actual people in today's world need to try to think in terms of what those constructed people would judge and do. Secondly, one needs to establish and defend a view on what their reasons and purposes are in constructing a practice. Thirdly, one needs to establish and defend a view on what those people, with those reasons, would decide in particular cases, with regard to the ongoing construction and reconstruction of a practice, according to the logic of practice. The overriding aim of this section is to illustrate this method.
In order to distinguish between sovereignty and human rights practices, according to the purpose method, one needs to see if one can draw a sound distinction between two different kinds of practitioners who have different purposes in constructing practices linked to those concepts. Practices of sovereignty involve placing limits on what individual participants (states) can do in pursuit of their values, goals and ends, in order to create the basic conditions of further interaction in an international system made up of more than one such unit.Footnote 71 It is possible, as suggested by Rawls, Beitz, and Raz that human rights standards should play a role in determining the constitutive rules of sovereignty.Footnote 72 In order to answer this kind of question, one may take the perspective of the idealised sovereignty practitioner (with the capacity for context-based judgement as identified by the logic of practice) whose purpose is to determine the shared rules and understandings that make it possible for the main units in the international system to interact with each other. One needs to ask, from that person's perspective, whether there is good reason to include respect for human rights standards in this set of basic limits. It is important, however, to recognise that this discussion is about is about the location and (re)construction practices of sovereignty, rather than practices of human rights. The starting-point is the perspective of the idealised sovereignty practitioner.
In order to use the purpose method to check whether it is necessarily true that the primary purpose of human rights practice itself is to constrain practices of sovereignty, one must start from the other end: from the perspective of a different idealised practitioner, with a different set of reasons and purposes. A brief historical interpretation of the distinctive gap in international-political life that a practice called ‘human rights practice’ has the potential to fill is a starting-point that can help to develop an account of the reasons for and purpose of this practice. During the interwar period and World War II, pre-existing universal ‘rights’ concepts, such as ‘liberal rights’, ‘natural rights’, and the ‘rights of man’ – if understood simply as a normative list of material and abstract objects to which all people are entitled – did nothing in practice to stop widespread humanitarian atrocities.Footnote 73 States breached responsibilities to protect and to provide for the basic rights of all of the insiders within their territories. In large part, they did so by redefining insider identity in a way that allowed European governments, including many of the Allies, who would not accept nationality-stripped refugees, to justify protecting certain humans' basic rights and not others'. Human rights, the specific compound concept, as a contemporary political idea (distinct from other ‘rights’ concepts in the same family), came about in the context of the breakdown of modern states into war machines, which did not protect or provide for the rights of all of the individuals within their sovereign territory.Footnote 74 The idea also came about in the context of the international community's unwillingness to accept responsibility to individuals and groups beyond borders.Footnote 75 Finally, it came about in the context of a normative rejection of European colonialism, at least in official public discourse, if not always in fact.Footnote 76 Long-standing practices of declaring universal rights and outlining responsibilities that we all have to others, based, for example, on Kant's categorical imperative or Mill's utilitarianism, did nothing to prevent the atrocities that occurred in the context of the twentieth century.Footnote 77
What is missing from this picture is a practice of assigning responsibility to the right agents, and holding these agents to account for breaches of their responsibility, such that all individuals are covered under the scope of some specifiable agent's sphere of human rights responsibility, even if – perhaps especially if – the states within which individuals reside act irresponsibly toward them. This is the gap that human rights practice needs to be able to fill. It is what makes it distinct from other, related, rights practices.
In this context, and throughout the second half of the twentieth century, it might have seemed appropriate, to idealised practitioners, to assign human rights responsibility primarily to states for at least two reasons. Both reasons are based on Weber's definition of the state as the entity that ‘successfully claims the monopoly on the legitimate use of physical force within a given territory’, but each reason focuses on a different aspect of this definition.Footnote 78 The first reason is the capacity to cause harm that follows from the monopolisation of violence. According to this line of thought, the scale of the humanitarian atrocities committed during the interwar period, World War II, and the era of colonialism, was possible to achieve only because of the ability of modern states, with modern technology, to cause harm to others in a systematic and brutally efficient fashion.Footnote 79 Human rights responsibility can be linked, for this reason, to the enhanced capacity of particular actors, in the modern context, to cause harm to individuals in a systematic way.Footnote 80 The second reason focuses instead on the fact that the state's monopoly on the right to use force is defined as legitimate. According to a Weberian picture, states use violence in the name of the public: for the public good rather in the name of individual or private interests. An actor's public nature or role can be a reason for practitioners to think that those (public) actors are the right actors to identify as the protectors and providers of human rights.Footnote 81
In order to use the purpose method to answer the question of whether non-state actors can be bear responsibilities to protect human rights within today's international human rights practice, one needs to take the perspective of the idealised human rights practitioner, and to ask what he or she would judge as falling within the scope of human rights practice, and why. These same two reasons, capacity and publicness, which would have been available to practitioners who wanted to construct human rights practice in a context where it did not exist, can be considered in today's context, to assess whether and to what extent it is actually the case that sovereign states are the sole appropriate bearers of human rights responsibility for those reasons.Footnote 82
In terms of the first criterion, ‘capacity’, it has by now become clear that, especially weak states or weak governance zones – what Robert Jackson has called ‘quasi-states’ – states can lack the capacity to protect and to provide for a full range of human rights.Footnote 83 By contrast, especially in the numerous contexts where the provision of services ranging from security and policing to health and education by the private sector is the norm rather than the exception, non-state actors do have a capacity both to cause harm to individuals in a systematic and organised fashion, and also to protect and to provide for the objects of many human rights.Footnote 84 Therefore, if ‘capacity to cause harm to individuals in an organised, systemic and distinctly modern fashion’ is a valid reason to assign human rights responsibility to a particular actor, then an analysis of many of today's empirical contexts reveals that non-state actors meet this criterion, in addition to states.
In terms of the second criterion, ‘publicness’, the state's monopoly on violence is not necessarily legitimate just because violence has been monopolised. As Charles Tilly, on the one hand, and critical-security-studies scholars, on the other, have noted, many states resemble ‘racketeer’ outfits – which cause threats and then justify their use of force on the grounds that those threats need to be eliminated – more so than they resemble virtuous providers of peace, order, and security for a public.Footnote 85 Furthermore, it is possible to supplement the Weberian picture with other approaches to social theory in order to assess whether and in which circumstances non-state actors can ever be relevantly public. Robert Cox, for example, looks at the exploited class and its leadership, rather than at the state, in order to find actors that truly represent the public.Footnote 86 This alternative would provide a basis to identify some non-state actors as ‘public’ in the sense relevant to the idealised practitioner's judgement about who is the right agent to bear human rights responsibility such that every individual is covered.
These questions about the role of non-state actors within human rights practice – and some potential answers to them – have been used as an illustration of how the purpose method works, in a way that provides a substantive challenge to the idea that human rights practices are inextricably bound up with practices of sovereignty. The purpose method provides a methodological framework that allows one to develop an account of what the purpose of human rights practice actually is. The (contestable) conception of the ideal practitioner, and the judgements that that practitioner would arrive at given his or her purpose in constructing a practice, is the foundation of the approach. This is a significant advancement on existing ‘political’ theories of human rights, which have tended to use, as their foundation for defining the purpose of human rights practice, a putative historical/social fact that the purpose of the practice of human rights is to assign duties to states and to place conditions on practices of sovereignty.
Conclusion
It is one thing to say that we need to study ‘practice’. This project is well under way. It is another thing to claim that one is studying a practice by studying specific practitioners, such as human rights practitioners, security practitioners, or diplomacy practitioners. In order to study specific practicioners, it is important to have a method that allows one to locate a practice's boundaries, in order to determine what (and who) is in, and what (and who) is out. One cannot justifiably say things such as ‘human rights practice is like this’ or ‘security practices are like that’ if one has helped oneself to an account of what and whom to study in the first place. Adler and Pouliot tend to privilege practical sense, in a way that leaves questions about the actual location of specific practices under-problematised. This article enables scholars to be more explicit, in future research about practices and practitioners, about the methods that are being used in order to define and to locate practices.
The article exposed weaknesses in the agreement, contextual, and value methods for locating an international practice. Firstly, the agreement method either has too narrow and legalistic a conception of what counts as part of a practice, or otherwise needs to be supplemented with other methods – for example, the contextual method or the value method – in order to explain why speech acts that do not include exact concepts (such as ‘human rights’) should count as part of a practice. According to the agreement method, the normativity of practices is also reduced to the notion of consent. This provides a clear bridge between analytical statements about what practices are, and the normative reality of practitioners' reasons to practise, but this clarity comes at a price: consent is a problematic basis for normativity. Secondly, and by comparison, the contextual method lacks any bridge – even a problematic one – between the analytical and the normative. It therefore lacks a response, for practice-based logics of action, to the same problem that others have identified in the logic of appropriateness: human agents often show a knack for overriding their rule-based or habit-based programming in order to do what, in their judgement, is right. It is a feature of our world that practitioners think that they have reasons to practice. It is a prima facie weakness of a method for locating a practice if it has trouble locating social objects that fit with this feature of reality. Thirdly, the value method results in practices dropping out of the picture. Practitioners would no longer have a reason to practice, because they can assess for themselves how to pursue values and principles directly. Other scholars might think that one of these methods is most suitable for their questions, and for their research into practices. If so, they would need to engage critically and constructively with these conclusions, and in particular with the question of why the weaknesses that this article has identified might be less significant for the sake of particular kinds of research into other international practices.
Neither a practice, on the one hand, nor an individual agent's judgement about what to do, on the other, is necessarily authoritative. One strength of the purpose method is that it operates at the intersection of the two by combining reasoned judgements with the construction of social practices – in a way that rejects the conservative acceptance of an easy and obvious alignment between a practice and an actual practitioner's judgement. One potential weakness of the purpose method, which was indicated in this article, is that it rests on a substantive conception of the person from whose standpoint human rights practice can be appropriately constructed. This conception of the person is contestable. It is not possible to prove empirically that a certain conception of the person is appropriate, in the same way that it is more plausibly possible to prove that agreement about the location of a practice occurred, or that a political idea arose in a certain context, amongst certain actors, with certain intentions. However, this is not a significant weakness, and in my judgement, it is not even a weakness at all. There is room for reasonable disagreement about the conception of the idealised human rights practitioner, and about what such a person would decide when constructing or reconstructing a practice. Having such a clear locus for debate – having clarity on which specific, replicable steps must be taken and justified in order to reach a conclusion – is an improvement on an egoistic or relativistic picture, whereby the concept of human rights is taken to mean whatever a particular individual, group, culture, or society thinks that it means at any given moment in time. It is also improvement on a picture of social ontology in which there are only values and principles, but no rules or practices.
By asking and answering questions about which intuitive and circumstance-based judgements an idealised practitioner – whose purpose it is to fill a specific practical gap in international-political life – would arrive at in today's context, I arrived at the suggestion that modern actors who, firstly, due to a degree of control over the means of organised violence, have the capacity to cause harm to individuals, and/or, secondly, are relevantly ‘public’ in a particular context, are candidate bearers of responsibility for human rights. This remains so regardless of whether those actors have been recognised as sovereign. This suggestion was provided as an illustration of the way that the purpose method works. More detailed scholarship is needed and called for in order to expand on this assessment of whether and how non-state actors fit as responsibility-bearers within contemporary human rights practice. Most importantly, future research can now draw from a clear set of methods that allow one to locate the boundaries of any international practice.
