1 Introduction

Discussing the fact/value separation – that long-standing nugget of philosophical controversy – is apt to offer much frustration to any scholar eager to go beyond it. The difficulty is that making distinctions between facts and values can sometimes be very helpful, while at others times it seems to be precisely the obstacle one needs to overcome. The pragmatic solution, also adopted in this paper, is to avoid characterising the distinction as having any metaphysical bite, and treating it simply as a contingently relevant distinction that can play its own modest role in inquiry, thereby avoiding any strict dichotomy between facts and values. Indeed, the pragmatic solution goes further, for it looks for devices that we may explain as having both a factual and an evaluative dimension. These are devices that themselves are best thought of, and referred to in this paper, as ‘factual-evaluative complexes’, but ones that can be profitably analysed, and explained, as having on the one hand a factual side, and on the other, an evaluative side. Arguably, vulnerability is one such device: to characterise someone as vulnerable is to take an evaluative stance, i.e. to think that someone who is in danger of harm or suffering is worthy of being protected against such a danger being actualised. This is the evaluative side of the device. However, to characterise something as worthy of protection is also to consider that it is worthy of protection in light of certain circumstances that make it susceptible to that danger of harm or suffering. That is its factual side. Of course, it could be said – and this paper agrees with the assertion – that ‘harm’ or ‘suffering’ are themselves factual-evaluative complexes. As deep as we dig, we will not reach a foundation, but rather another device that is best thought of as having both factual and evaluative dimensions. All the same, we must choose devices we wish to use in our investigations. In this paper, the device being chosen is vulnerability, and it is claimed that this device, though itself ultimately a factual-evaluative complex, can be analysed as having a factual side – i.e. the identification of circumstances in which someone or something is susceptible to the danger of harm or suffering – and an evaluative side – i.e. the taking of an attitude that that danger of harm or suffering is worthy of protection.

At this point, one might very well ask: what has this got to do with law, and even more so with the methodology of any inquiry into law? To see the relevance of the device of vulnerability one must take a step back and look at the specific relational contexts in which it arises. Any modern society can be said to be composed of the following five relational contexts: (1) relations between individuals; (2) relations between individuals and communities; (3) relations between communities; (4) relations between individuals or communities on the one hand, and institutions on the other; and (5) relations between institutions. Situating the study of law in such relational contexts is here called ‘relational jurisprudence’.

Further, relational jurisprudence not only situates law in such relational contexts; it also considers how law affects or influences the quality of those relations. Here we have to be immediately careful: insofar as we are investigating the quality of relations, we must recognise that there is a diverse range of specific ways in which relations can go better or worse, depending on the relational context. For instance, what characterises good relations between persons is likely to be different from what characterises good relations between communities or between institutions. As we shall see later, a good way of seeing these differences is by considering what specific vulnerabilities arise in different relational contexts. Given that what characterises good relations in the different relations changes, and that there are different vulnerabilities at play, so it follows that law takes on different forms and plays different functions in these relational contexts. Situating law in the varying quality of different relational contexts, then, allows us to be more detailed in our understanding of the complexity of law, especially in modern societies. Arguably, and especially insofar as we use the factual-evaluative complex of vulnerability, we are likely to be more adequately detailed in our study of law than if we used more abstract devices, such as, for instance, considering whether law deserves our fidelity (see Fuller 1969). At the same time, however, the device of vulnerability is arguably more flexible than more concrete devices (e.g. consider the basic goods listed by John Finnis (2011)),¹xThe seven basic goods are: practical reflection, life, knowledge, play, aesthetic experience, sociability (friendship), practical reasonableness, and ‘religion’ (see Finnis 2011). Vulnerability is arguably more flexible because, for instance, one can be vulnerable to death, to being deprived of the opportunity to acquire knowledge, etc. Vulnerability, then, does not pick out distinct values, but also does not exclude specific values that might be cherished in any particular society at any particular time. in that it is a general device applicable to investigating law’s role in different relational contexts. Finally, it is also arguably historical astute – it recognises that what vulnerabilities the law protects, or how it balances that protection, has changed over time, and is likely to continue to change, subject often in any one polity to much debate and disagreement, though also a good deal of overlap. Nevertheless, it insists that law is profitably understood and also evaluated by examining how it manages vulnerability in different relational contexts.

This way of situating the study of law – as featuring in the above-mentioned five different kinds of relational contexts, where at least some of law can be understood as managing vulnerability – can translate into several concrete tasks. For example, we can understand what kinds of vulnerability the law currently regards as worthy of protection, and which it does not, or how it balances the protection of those vulnerabilities. In addition, we can consider how law and legal institutions can be responsive to the emergence of new kinds of vulnerability that might appropriately fall within the scope of law’s protection. Finally, we can be on the lookout for law itself creating and/or exacerbating vulnerabilities. Each of these tasks carries with it the same methodological implication: it requires a mixture of empirical and normative elements. For example, trying to understand what vulnerabilities the law currently protects, and how it balances that protection, requires attention to the usual sources of law, but it also requires the rational reconstruction²xFor more on ‘rational reconstruction’, see MacCormick (1997) where he says that his use of the term is inspired by M.P. Golding’s ‘Kelsen and the Concept of “Legal System”’ (1971). of those sources via the factual-evaluative complex of vulnerability;³xOne could be more radical in one’s argument here and assert that law just is that which manages vulnerability in different relational contexts. This argument is too radical for this paper, for this paper argues that vulnerability is a device that can help us make intelligible and evaluate some, not necessarily all, of law (ultimately, a suite of devices will be needed). In other words, the paper is sympathetic to a conventionalist reading of the sources of law; what it proposes is an object and method for the study of law, given our identification of laws via the usual sources (e.g., in the United Kingdom, legislation passed by the Queen in Parliament, as further developed by judges). Nevertheless, it is hoped that the object and method offered here would not only be applicable to modern legal systems, such as the system in the UK, but would be applicable generally, to other systems at other times, including informal and de-centralised forms of political association. In those other contexts, where, for example, the sources of law are not stable (e.g. where the authority of any one ruler is too fragile, or authority is widely diffused), the object and method offered here could be taken as a way of identifying law, i.e. one looks for that set of practices that manage vulnerability in different relational contexts (with the proviso that it may be that in some informal and de-centralised forms of political association there would be fewer relational contexts to consider). similarly, investigating the emergence of new kinds of vulnerability requires attention to circumstances in which someone or something is in danger of harm or suffering, but also requires a judgement as to which of those dangers are worthy of law’s protection. The point, then, is that the use of such factual-evaluative complexes to study law requires a combination of empirical and normative elements in one’s methodology.

The structure of this paper is as follows: first, the fact/value separation is considered, and the idea of a factual-evaluative complex is explained; and second, relational jurisprudence is introduced, with a focus on showing how profitable it can be to study at least some laws in different relational contexts as managing vulnerability.

2 The distinction, but not the dichotomy

One of the pernicious effects of an approach to inquiry that insists on its metaphysical credentials, as distinct from philosophical anthropology with no such aspirations, is that as a result of adopting certain purported metaphysical dichotomies – such as the alleged dichotomy between facts and values – we force ourselves to choose between a method supposedly adapted to one side of the dichotomy (empirical for the factual side, and normative for the value side). This is especially unfortunate in the case of the study of law, which demands a combination of the virtues of both a sensitivity to the world (including the way it changes) on the one hand, and a sensitivity to the attitudes (including histories of attitudes) we have taken to the world on the other hand. This paper argues that we can avoid such pernicious effects on our method if we discard the dichotomy, but keep the distinction, i.e. we do not ascribe any metaphysical importance to facts or values, but we acknowledge the contingent usefulness of the distinction. In order to do this, we need to identify and use in our inquiries into law devices that are factual-evaluative complexes, i.e. that we understand as having both factual and evaluative sides. In doing so, we will allow ourselves to use a method that combines both empirical and normative components.

The general strategy of avoiding the dichotomy while keeping the distinction is by no means new. A similar strategy was advocated by Hilary Putnam in his critique of the fact/value separation. In his The Collapse of the Fact/Value Dichotomy (2002), Putnam suggested we drop the ‘dualism’, but retain the distinction between facts and values. In making this suggestion, Putnam was following – as he acknowledges – John Dewey. As Putnam says, ‘Dewey’s target was not the idea that, for certain purposes, it might help to draw a distinction (say, between “facts” and “values”); rather, his target was what he called the fact/value “dualism”’ (Putnam 2002, p. 9). The dualism is pernicious not merely because it ‘typically gets accompanied by a highly contentious set of metaphysical claims’ (Putnam 2002, p. 61), but also, if not more so for Putnam, because as a result of a particular way of understanding ‘facts’, it denigrates values and value judgements to mere expressions or subjective preferences. Putnam traces the dichotomy back to David Hume, adding that it reaches its apogee with logical positivism. In its most mature form, the logical positivists insisted on a dichotomy ‘between cognitively meaningless judgements, which included but were not limited to value judgements, and cognitively meaningful judgements’ (Putnam 2002, p. 61). Cognitively meaningful statements were of two kinds, either analytic or synthetic, such that a statement was either a ‘tautology’ (as were certain logical statements or mathematics) or ‘a description of some “fact” or possible fact’ (Putnam 2002, p. 61). A lot turned here on just what a ‘fact’ was – indeed, problems with where to ‘place’⁴xSee Price 2011 for discussion of the ‘placement problem’. terms obviously useful in scientific practice like ‘atom’ or ‘bacterium’ led to making the above-mentioned division between two kinds of cognitively meaningful statements, e.g. in the case of Rudolf Carnap, a division was made between ‘observation terms’ (such as ‘blue’, ‘hot’, ‘larger than’ etc.) that are observable and the test procedure for which is simple, and ‘theoretical terms’ (precisely like ‘atom’, ‘bacterium’ etc) that are cognitively meaningful but only because they enable us to derive the more basic observation terms (see Putnam 2002, p. 22-25).

The resistance to the dichotomy has played on various fronts. For instance, it has been claimed, and Putnam also endorses the claim, that no description of facts in science – e.g. no use of observation terms, no interpretation of any test procedure (no matter how simple) – is value free, for all such descriptions are guided by certain ‘epistemic values’, such as simplicity or elegance.⁵xW.V.O. Quine was the leading voice here, though he referred less to values than conventions of scientific practice; see Putnam 2002, p. 135-7. Putnam’s wife, Ruth Anna Putnam, also contributed to this debate (see Putnam 1985; 1998). Turning to everyday rather than scientific practice, it has also been claimed that we have no direct, immediate access (as the empiricists believed we did) to the way things are (at least in a way that can count as knowledge), but that, instead, we are able to engage with and respond to the world only because of and thanks to our conceptual apparatus. Insofar as we rely on our conceptual apparatus, in turn, we are responsible beings who are capable of being correct and incorrect (and thus, on this view, having knowledge) by either applying or misapplying, respectively, our concepts. This is a claim made in Kantian spirit, and one that has continued to flourish in contemporary philosophy, especially, in the English world, via Wilfred Sellars, and more recently, John McDowell, Robert Brandom and others (see, e.g. Putnam 2002, p. 118-9).

Putnam’s defence of values – as precisely not mere subjective preferences, unable to be made subject to rational discussion and justification – appears to be a version of this conceptualism (at least as he presents it in the 2002 book). Putnam is not, however, easy to pin down. His general strategy is to focus on the presence in our language, and the importance and ubiquity of our use, of so-called ‘thick terms’ or what he also calls ‘entangled concepts’, e.g. ‘cruelty’ or ‘kindness’. This paper – in advocating the use of factual-evaluative complexes, such as vulnerability, in the study of law – is sympathetic to this strategy of identifying ‘entanglement’ (though without relying on conceptualism). However, some of what Putnam says about this ‘entanglement’ is not entirely helpful for present purposes. When Putnam says that the concepts in question are both descriptive and evaluative at the same time (Putnam 2002, p. 34), that is helpful, though of course we still need to explain how that is so. When, on the other hand, he says that such concepts reveal entanglement because the term can be used either ‘normatively’ or ‘descriptively’ (Putnam 2002, p. 35), he is in danger of re-introducing the dichotomy at the level of ‘uses’. Furthermore, Putnam often mixes two kinds of distinctions, i.e. the evaluative-descriptive, and the prescriptive-descriptive, e.g. he says that one cannot ‘separate thick ethical terms into a descriptive component and a prescriptive (or imperative) component’ (Putnam 2002, p. 62). It is, however, one thing to argue that any evaluation is, in part, a description, and quite another to say that any use of a certain (thick) term also implies a prescription. This is so because the notion of prescription is more intimately tied to action than evaluation: certainly, some if not many evaluations may be linked to action quite intimately (thinking of someone as cruel might imply that we recommend s/he stops what s/he is doing, or that s/he be stopped), but others less so (e.g. we might evaluate someone’s behaviour as jealous, but without implying that one ought not to be jealous in that context, or that someone ought to stop them being jealous – the point being that we evaluate, but without prescribing).

It is also not entirely clear what Putnam might mean by the idea that one cannot break down a thick ethical term into a descriptive component and (let us read him as saying) an evaluative (rather than prescriptive) component. The closest he comes to elaborating on this is when he says that if we tried to break down ‘cruel’ into an allegedly purely descriptive component of ‘causing deep suffering’, we would end up recognising that ‘causing deep suffering’ is not itself ‘free of evaluative force’, e.g. ‘“suffering” does not just mean “pain”, nor does “deep” just mean “a lot of”’ (Putnam 2002, p. 38). This is a helpful point, and one endorsed in this paper. At other times, Putnam puts the focus on learning, i.e. that we cannot learn to use thick ethical concepts (including, say, distinguish ‘courageous’ from ‘rash’ behaviour: see Putnam 2002, p. 62) without ‘identifying imaginatively with an evaluative point of view’ (Putnam 2002, p. 39). This, too, is a potentially useful point, but we need to dig deeper into what Putnam might be helpfully understood to mean, for it will help us see how vulnerability is a factual-evaluative complex. In doing so, we can also clarify why we need not rely on the ‘conceptualism’ that Putnam seems to rely on in order to make a case for factual-evaluative complexes.

It is true that when we make an evaluation that someone is vulnerable, it is potentially misleading to say that there is, on the one hand, the fact of someone being in danger of being harmed or suffering, and on the other, an attitude expressing that that is a bad thing. To insist rigidly on such a division would return us precisely to the fact-value dichotomy, and to the very problems that gave rise to classical empiricism and especially logical positivism, i.e. problems to do with how to make sense of the idea that there could be moral facts, e.g. the fact that someone is in danger of being harmed or is suffering, which then tend to lead precisely to the idea that the only thing that moral terms could be are ‘expressions’ of the ‘boo-hooray’ variety, and the related claim that they are thus entirely subjective. However, if in order to turn away from those problems we come to promote conceptualism, then (1) we risk not seeing how our valuations are related with and connect to the world (there is a form of idealism that seems to breathe down the neck of conceptualism); and (2) we risk characterising our abilities to evaluate that someone is in danger of being harmed or is suffering too rigidly, i.e. we come to define being in danger of harm or suffering as a concept with certain rules that determine when it is correct to say that someone is in danger of being harmed or is suffering.⁶xIf a phrase is needed here, we might say that what we are after is a ‘pragmatism without normative determinism’ (and thereby precisely a pragmatism not dependent on conceptualism). Of course, whether conceptualism can be characterised in a way that does not result in normative determinism is a very complex question that falls outside the scope of this paper. A good beginning for any such inquiry would be an intimate analysis of the works of John McDowell (e.g. 1998) and Robert Brandom (e.g., 1994 and most recently 2011).

It follows that we need to say that there is a descriptive and an evaluative side to any exercise of our ability to evaluate someone as vulnerable, but then also insist that this has no metaphysical implications (i.e. no claim is being made that there is a fact of being in danger of harm or suffering on one side, and a mere expression on the other). Notice that not making any metaphysical implications means that we are in part agreeing with Putnam, for we would agree that ‘harm’ or ‘suffering’ are themselves factual-evaluative complexes – it is, if you like, factual-evaluative complexes all the way down. In other words, the point is that for the purposes of revealing the entanglement we need to point to a descriptive side as well as an evaluative one. But the crucial step here still awaits, i.e. how to characterise the link, or precisely the entanglement, between the two sides (and to do so in a way that is not deterministic, normatively or causally). Here is where we can connect with Putnam’s idea that we learn to use such terms by taking an evaluative point of view, and also with his recurring references (though without elaboration) to Iris Murdoch. Murdoch, it should be recalled, in her Sovereignty of Good (1970) insisted that we learn to use ethical terms in specific contexts, adding that in doing so we develop our moral imagination. In the burgeoning literature on moral imagination since Murdoch (for instance in the work of Martha Nussbaum), this has lead theorists to recognise that we can enrich and develop our understanding of these terms by seeing how they are used in contexts unfamiliar to us, e.g. as portrayed in literature and film.⁷xSee, e.g., Nussbaum 1990, especially the essay ‘Finely Aware and Richly Responsible: Literature and the Moral Imagination’. For example, we enrich our understanding of the term ‘cruelty’ or ‘kindness’ by reading, and watching adaptations of, Dickens’s novels. The point here is this: the connection between the descriptive and the evaluative in factual-evaluative complexes is dynamic; it is loose in character and it changes over time. When we have an ability to evaluate someone as cruel or kind, or as vulnerable, we have an ability that associates the term with certain kinds of circumstances (where we have learnt the term used), but any such association is a tentative and loose one, i.e. it allows us to develop our understanding of the term by increasing the set of circumstances we associate with it (for instance, by analogy as a result, say, of exposure to the way others, including authors and film-makers, use the term). In certain communities, the associations we have of vulnerability with certain circumstances might overlap, i.e. we might all roughly associate similar circumstances in which persons are in danger of being harmed or suffering as worthy of protection, but the overlap is never such that we could say there is one, and only one, concept of vulnerability that is correctly applied when such-and-such circumstances obtain. The overlap is loose, for it is always possible for us to be persuaded by someone’s novel use of the term vulnerable, e.g. to describe the environment as in danger of being harmed or suffering by our polluting practices, or to describe stay-at-home mothers as in danger of economic and emotional exploitation by their husbands.⁸xIn other words, the central problem for conceptualism is change: is there a separate ability that explains how we are able to enrich our use of certain terms, or is it part of the same, but loose and tentative, ability? This paper argues for the second view, and thus distances itself from conceptualism.

It is important to see that this way of characterising the entanglement also allows us to avoid both subjectivism and objectivism. Subjectivism comes into play when we completely disassociate evaluative terms from the world, i.e. when we think of evaluative terms as but expressions that we can use whenever it suits us or works for us (precisely ‘boo-hooray’ terms about which there can be no rational disagreement). We avoid this in the characterisation above because we insist on their being an association – overlapping roughly in certain communities – between the uses made of certain terms and sets of circumstances (in that respect, we require a connection between these terms and the world, but just not a neat, correspondence-like matching connection). It is vital to see that the use of evaluative terms is responsive to the world: it engages, connects with, and relates to the world. But this is an engagement that can also be surprised by the world, e.g. one of us, a group of us, can extend our use of the term vulnerable to a new set of circumstances (of course, whether this becomes widespread and affects the use many of us then make of the term vulnerable is another question). It is precisely also because of this flexibility that we avoid objectivism, i.e. we do not claim that there are sets of circumstances in which, independent of what anyone thinks of them, someone is vulnerable. To say that someone is vulnerable is to make an evaluation that the danger of the harm or suffering in question is worthy of protection. But we can deny objectivism in this sense and still insist that we engage with the world when we use the term and that we can have rational disagreement over the use of the term. There can be rational disagreement in the absence of fixing some facts that determine what is or is not a state of vulnerability and, equally, in the absence of constructing a concept that determines when it is correct or incorrect to use it. As members of a community, we share overlapping uses. These are uses that can be enlarged as a result of engagement with the world, such that, furthermore, the spread of any such enlargement from the imagination of any one of us to the community as a whole is something that can be the subject of rational argument, i.e. in the pragmatist spirit, an argument over whether such enlargement helps us to cope better with problems that arise (e.g. take care of certain persons more effectively than we have done previously).

We are left, then, with a distinction, but not a dichotomy between facts and values. The distinction, perhaps somewhat paradoxically, is needed for it helps us to see the entanglement between fact and value. But the distinction carries no metaphysical commitments; it is but a contingently useful reminder – a bulwark against our tendency to drift into the dichotomy.⁹xAs this paper is not a survey of views, mention has not been made of many other theorists – aside from Putnam and Murdoch – who have advocated avoiding the dichotomy between facts and values. For a recent discussion of some of these contributions, and an application of it to the methodology of the social sciences, see Sayer 2011.

3 Vulnerability in relational contexts

Before we can turn to pointing to some of the implications and virtues of the use of vulnerability in legal research, we need to provide an object of inquiry – a structure within which we can employ the method of drawing on factual-evaluative complexes like vulnerability. It is the task of this section to sketch out such a structure that, it is claimed, is particularly well-suited to the use of the device of vulnerability. Having said that, it is not being asserted that the two are so strongly linked that one could not use vulnerability to study law without this structure – just that, once again, the structure is a particularly useful one for the use of vulnerability as a device for studying law.

It has been common in legal theory, perhaps especially but certainly not only in its contemporary guise, to focus on the first-person, individual experience of law. The roots of this focus may well be religious with the emphasis in the Christian tradition on ‘conscience’. Certainly, there is an argument to be made that ever since Aquinas’s attempt to Christianise Aristotle, legal theorists have theorised about law by examining how law features in the practical reason of individual agents. The approach is familiar to contemporary students of legal theory, and it is one, furthermore, that unites theorists otherwise with disparate theoretical viewpoints, e.g. H.L.A. Hart and John Finnis. Of course, there are disagreements even at this level – Hart and Finnis both emphasise agency, but Hart focuses on the internal point of view exercised by officials, while Finnis (like Aquinas) examines the conscience of citizens – but the overall approach is clear. It must be said that one sees this tendency to favour agency not only in legal theory, but also in social science (in this respect, it is worth remembering that Max Weber’s Verstehen was probably an influence on Hart’s notion of and emphasis on the internal point of view) and in philosophies of normativity generally (one need only be reminded of, for example, the very influential work of Christine Korsgaard). No doubt an intellectual history of this focus on first-person agency would need to take into account the political project of the Enlightenment – the freeing of the individual from the bounds of tradition, and then the fight against an encroachment of this liberty by the state. Needless to say, such an intellectual history is outside the scope of this paper.

There is no doubt that there is a lot to learn about law – as the above-mentioned theorists have shown us – by focusing our analysis on first-person agency. As with any object of inquiry, however, such an approach is bound to have its limitations. Again, it is not proposed that these limitations be investigated in this paper, but if the presence of limitations is accepted in principle, the following obvious question arises: what alternatives might there be? One option, and the one sketched in this paper, is to focus not on the exercise of individual agency, but rather on relations, including investigating how they can go better or less well (i.e. on their quality). In other words, the suggestion is that we study law by situating it as one of the factors that influences and affects the quality of different kinds of relations. There will be many other factors, of course – including everything from language to fashion – but law can safely be said to be one of them. The first step, then, is a gestalt switch: away from the focus on individual agency and its emphasis on improving the quality of the reasoning of agents, and towards relations and what makes them better or worse, including how law might improve (or of course lower) the quality of such relations.

Relations, however, can be of various kinds. Similarly, the quality of relations differs depending on who the relations are between. The most familiar kind of relation is the relation between individuals: two persons can relate to each other in a wide range of personal (husband/wife; child/parent) and professional ways (as business partners or business rivals). There are, however, many other kinds of relations, e.g. relations between individuals on the one hand, and communities on the other (e.g. relations between a member of a religious group and that group as a whole); relations between communities (e.g. between different kinds of religious groups); relations between individuals or communities on one side, and institutions on the other (e.g. individual members of a religious group or the group as a whole relating with institutions such as courts); and relations between institutions (e.g. between courts, legislatures, hospitals or prisons).

All these relations can clearly go better or worse: we know, generally speaking, that in many relations between persons, the quality of those relations is improved by persons caring for each other, looking out for each other, helping each other out when the other needs it. Of course, matters are not always that simple, and in many kinds of relations between individuals we make a great deal of room for conflict (e.g. between business rivals – indeed, in that context we often require conflict, prohibiting various kinds of collusion or co-operation between business rivals). To study relations, including their quality, is not to preach generalities about perfect harmony – any such study must be richly contextual, recognising the enormous variety of ways in which relations can go more or less well, even within any particular relational context, and thereby acknowledging that what is generally thought to improve relations in any one kind of relation can be very complex. For example, as much as we might wish to prohibit various kinds of collusion or co-operation between business rivals, we nevertheless also encourage some forms of mutual respect and even trust.¹⁰xFor an insightful account of different kinds of trust that characterise different communities, including business communities, see Cotterrell 1995. The examples can be multiplied: the quality of relations between communities is improved by agreements enabling the sharing of valuable resources; relations between individuals and communities on one side, and institutions on the other, are generally improved when those institutions enable participation in their decision-making process by those individuals and communities; and relations between institutions are improved when those institutions make their reasoning transparent to each other, and even more so when their reasoning processes take account of the other’s reasoning processes. None of these claims about improvement are infallible; none are a priori – all are matters we have learnt over time, and are generally committed to, but can change if we decide the circumstances require it.

So far we have made two moves: first, to focus on relations rather than individual agency; and second, to study the quality of those relations, not only in different relational contexts, but even within those contexts to focus on particular kinds of relations. The third move is then to zoom in on one of the factors that influence and affect the quality of those relations. For relational jurisprudence, the relevant factor is law, but putting things this way might allow us to put the practice of jurisprudence in context, i.e. it can and ought to sit aside other inquiries into the quality of relations that focus on other factors. Ultimately, a holistic approach is to be favoured – studying the various factors that influence and affect the quality of different kinds of relations in different relational contexts. For the purposes of this paper, the focus shall be squarely on relational jurisprudence.

We have now clarified the object of inquiry, but we have hardly begun to study how law influences and affects the quality of relations. In order to get going, we must decide on which devices will help us ascertain law’s role in such relations. Here is where this paper makes a plea for the use of factual-evaluative complexes – like vulnerability – for it is such devices that not only enable, but positively require us to employ a mix of empirical and normative methods. Another way to put this point is that if we want to seriously study the quality of relations, and law’s role in that, then we should choose as our devices factual-evaluative complexes, for it is such devices that will allow us to be sufficiently respectful of the variety of ways in which different kinds of relations in different relational contexts can go more or less well. We can, then, use vulnerability as a device to study law’s role in the quality of relations, thereby enabling both understanding and criticism of law. For instance, we can understand law as a means of protecting vulnerability, and thus study how law currently protects vulnerability in particular relations in different relational contexts. Similarly, we can criticise law (1) by examining what vulnerabilities we think the law ought to protect that it does not currently, including vulnerabilities that have either already been recognised (except not by law) or by proposing new vulnerabilities as worthy of law’s attention; and (2) by being on the lookout for ways in which the law might itself create and/or exacerbate vulnerabilities, and thus reduce the quality of relations in different relational contexts.¹¹xFor a related discussion of how law can legitimate suffering, see Veitch 2007.

We have so far been speaking at a high level of generality: in order to see how vulnerability might be useful as a device for studying law’s role in the quality of relations, let us turn to some relational contexts: first, relations between individuals; and second, relations between individuals and communities.

3.1 Vulnerability in relations between individuals

Before we can consider some examples of how at least some of law can be profitably understood via the device of vulnerability in relations between individuals, it will be important to show how helpful this device is in general for understanding the quality of relations between individuals.

The first general point to make is that it is important not to limit the device of vulnerability in this context to identifying physical vulnerabilities of individuals. Indeed, acknowledging that vulnerability is now being used as often, if not more so, to identify harms or forms of suffering that are psychological or even moral is an illustration of the way in which the evaluative and the factual sides of this factual-evaluative complex are dynamically linked. Thus, as Bryan Turner (2006) observes, it may be the case that vulnerability is derived from the Latin ‘vulnus’, meaning ‘wound’, thereby clearly being associated in its early days with physical harm or suffering, but, as Turner adds, in ‘modern usage, the notion (…) increasingly (…) refers to our ability to suffer psychologically, morally and spiritually’ (Turner 2006, p. 28). Thus, and this will be particularly important in making use of the device of vulnerability when studying how law manages it in this context, we must overcome the theoretical bias to associate vulnerability with purely physical terms (such as pain, sickness, malnutrition and death: see Turner 2006, p. 28).¹²xAn example of this bias is Hart’s take on vulnerability in the context of his discussion of the minimum content of natural law (Hart 1994, p. 194-5).

The second general point is that, having tried, with the focus on relations, to move away from a focus on individual agency, we must be careful not to fall back into the tendency to focus on individual autonomy. Vulnerability, insofar as we connect it, as we should, to the mutual dependency of persons, ought to help us resist this return to individual autonomy. This point has been made numerous times in moral philosophy, though it has still to catch wide currency. Alasdair MacIntyre, for instance, has emphasised the link between vulnerability and mutual dependency, observing at the same time that in the history of moral philosophy there are ‘only passing references to human vulnerability and affliction and to the connections between them and our dependence on others’ (MacIntyre 1999, p. 1). Significantly for present purposes, MacIntyre does not limit the study of vulnerability to conditions where dependence is perhaps most obvious, e.g. early childhood and old age, but instead insists that all relations are, to some extent, characterised by mutual dependency. Others have associated vulnerability not only, or not centrally, with mutual dependency, but with mutual trust, thereby also simultaneously emphasising its ubiquity.¹³xThus, for example, quite recently J.M. Bernstein, building on earlier work by Annette Baier (see Baier 1995), says of trust that it ‘is the ethical foundation of everyday life. Trust is trust in others before whom we are unconditionally vulnerable that they will not take advantage of our vulnerability’ (Bernstein 2011, p. 395).

The third general point, and of particular relevance to law’s role, is that we need to avoid over-idealising relations – this also means, as noted above, realising that managing vulnerability can differ in different kinds of relations, even within the one relational context, and thus acknowledging that we may tolerate more risk in certain kinds of relations than others, i.e. that we may tolerate greater exposure to what we might in another kind of relation characterise as an unacceptable danger of harm or suffering. Thus, although it may be interesting and important to engage in a task of describing what considerateness between persons entails,¹⁴xConsider, for example, Baier’s recent description of how the ‘considerate person is appropriately aware of how her attitudes and actions affect those around her, and if necessary she alters them so as not to cause fear, hurt, annoyance, insult, or disappointment in others, particularly in those who hoped for cooperation or help’ (Baier 2010, p. 176). we may, especially (but perhaps not only) when considering the role of law, be wary of picturing relations between individuals as ideally risk free. The point is not that there may not come a time (as difficult as it is to imagine at the moment) of regulating, in some formal way, considerateness; it is that when using the device of vulnerability, we must be careful not to think that any form of risk in a relation between individuals, including even putting others in danger of some kinds of harm or suffering, cannot be part of the quality of that relation. In other words, it might sometimes help a relation between individuals to express annoyance of or disappointment in others, perhaps even in the case of those who hoped for cooperation or help (imagine, for instance, that the demands for cooperation are simply excessive, with the other person becoming too dependent on the assistance of others). Vulnerability, then, is not a device for over-idealising relations. As noted above, this is important, for it makes the device more amenable to the study of law.

Finally, as a fourth general point, we should add that the very idea of protecting vulnerability is at least in part a proactive rather than a wholly reactive one. In other words, it is important to see that we are not speaking here of protecting against harm or suffering directly; instead, vulnerability points to a state in which one is in danger of being harmed or in danger of suffering. In effect, this widens the scope of what we can see with the device of vulnerability: we are not looking at states in which we can ascertain (subject to all the provisos noted above that harm and suffering itself are probably better thought of as factual-evaluative complexes themselves) that persons are actually being harmed or are actually suffering. Instead, we are looking at the quality of relations between individuals from the perspective of the different ways in which individuals are vulnerable to each other’s actions – actions that may, but need not actually, result in harm or suffering. This should help us in particular with understanding law’s role, which, at least in part, and perhaps especially so in modern societies, is prophylactic, i.e. it creates proactive structures that protect persons against the possibility of harm, and not merely reactive structures that compensate persons for harm or suffering done. This is an important point also because this kind of proactive stance of the law alleviates persons from having to be ceaselessly aware of the different ways in which they might be vulnerable in any situation.

With these provisos in place, we can finally begin to see how we might make intelligible and also evaluate law’s role. First, we can take all relations between individuals, in the everyday, as relations the quality of which depend, at least in part, on persons not taking advantage of each other’s vulnerabilities and also taking reasonable care with respect to those vulnerabilities. Arguably, a good deal of criminal law, contract law and tort law are profitably understood as creating conditions for the management of vulnerabilities that arise in relations between individuals, e.g. there are criminal laws that enable us, as a community, to punish those that place the community’s members at risk of being physically harmed; similarly, the law recognises that when one person relies on the promise of another, they are made vulnerable in various ways, and arguably, a good deal of contract law is profitably understood as managing these vulnerabilities (e.g. consider the doctrines of undue influence or duress, which protect us against the vulnerability of our will); and, further, the law recognises that when we are neighbours, our very proximity puts us in danger of certain kinds of harm or suffering (including, for instance, damage to the enjoyment of our property) and so much of tort law may be profitably understood in this way.

It is also important to see, as briefly noted above, that the law does not protect all the various kinds of vulnerabilities that we might recognise, e.g. vulnerabilities to being disappointed by or being annoyed with one another. It might also suspend the protection of certain vulnerabilities in certain contexts, e.g. if a police officer is chasing a dangerous criminal, whom the police believe to be sufficiently dangerous to the public, then the protection against the usual danger of harm to the body may not apply.¹⁵xI am grateful to one of the anonymous referees for this example. Indeed, vulnerability may be a useful device with which to understand how law features in the quality of relations between individuals precisely because it is limited in what it protects, it being interesting to examine why the law protects certain vulnerabilities and not others, and also how it has developed over time to increase or indeed decrease protection of certain vulnerabilities (e.g. consider, in contract law, the rise of the doctrine of economic duress, where the law has increasingly come to recognise the sometimes very sophisticated and clandestine imposition of illegitimate economic pressure). In that respect, it is more accurate to say that the law manages vulnerability, rather than simply protects it. That being so, one can once again see how a mixture of empirical and normative methods is necessary: one needs, first, to understand what kinds of vulnerabilities there are in relations between individuals (this, in itself, requires understanding what we have recognised as vulnerability so far, as well as considering what we might go on to recognise as vulnerability); and second, one needs to understand what vulnerabilities law protects and which ones it does not, including evaluating whether the law has achieved a welcome balance, i.e. whether it manages vulnerability well. It is impossible to do these tasks well without a mixture of empirical and normative methods.

The point of choosing a device that is as thick as vulnerability is to encourage closer attention to the specific kind of dangers that arise in specific kinds of relations in this specific relational context. Saying ‘specific’ here does not mean we reach rock-bottom, or foundational facts, or anything of the sort; instead, the point is that vulnerability is a device that enables us to dig down more than other devices (e.g. such as simply considering whether the law is morally acceptable, or worthy of our fidelity), and thus also to engage with some level of detail of these relations, and also with some level of detail with the law, which often needs to make fine-balanced judgements about how to manage these vulnerabilities (and, equally, how far to generalise them into, say, a general principle of tort law, or a more circumscribed rule pertaining to relations between doctors and patients). Equally, however, getting dirty with detail, and thus engaging in empirical inquiry, is not at odds with evaluating – on the contrary, at each point we are doing both: we are choosing a certain level of description, and, in describing, we are choosing what we (given our experience, our research into the views of others etc.) take to be important to, for instance, the quality of relations between persons and law’s role in improving that quality. Indeed, the plea here for using factual-evaluative complexes in studying law is simultaneously a plea for recognising the benefits of mixing empirical and normative methods in inquiry.

3.2 Vulnerability in relations between individuals and communities

We have spent considerable time on one relational context, i.e. the relation between individuals. This has been done on purpose, for that relational context is arguably more intuitive or experientially accessible than some of the others – after all, it is a context in which we participate very regularly. This being so, that context is also one that we may often find ourselves drawing on in seeking to understand what might also affect the quality of relations in other relational contexts. In other words, we extend, by analogy, certain experiences we have in relations between individuals to other kinds of relations. There is nothing inherently bad about that – but we do need to temper the reliance on this kind of anthropomorphising, i.e. having used the ladder of experiences familiar to us from that context, we need to try to find ways of characterising the specific ways in which other relations can go more or less well.

There is also another point to be made about the importance of considering other relational contexts, and thus moving beyond the context most familiar to most of us. Arguably, law is particularly well developed in managing vulnerability in relations between individuals. But is it as well-developed in managing, for instance, vulnerabilities in relations between individuals and communities? Here, we see that the device of vulnerability is one that not only allows us to investigate, in some detail, and thereby also make intelligible, the diversity of law in different contexts, but that it is also one that enables us to be progressive, i.e. to show where law falls short in managing vulnerability in certain relational contexts, and to consider whether it should be law’s role to contribute to improving the quality of relations in those contexts by protecting vulnerabilities it has thus far not recognised.

Noting these two preliminaries, let us now very briefly consider how the device of vulnerability might help us in investigating and assessing law’s role in another relational context, i.e. the relation between individuals and communities.

The actions of individuals certainly have an impact on community interests, and vice versa. Individuals, as consumers, for example, have an interest in low prices. Companies, if we think of them as ‘individuals’ for this purpose, have an interest in maximising profit. Arguably, however, the good of communities depends on more long-term access to resources – otherwise, the community risks a very short life span indeed. The point, then, is that the interests of certain individuals – e.g. their relentless pursuit of lower prices by, for instance, exploiting the environment or cheap labour – might place the community at risk of harm or suffering, e.g. in running out of resources, especially for future generations. Of course, this can also work the other way round: a community might control access to resources to such an extent that individuals are placed at risk of either not being able to purchase goods (because they are too expensive) or simply not having access to enough goods. Arguably, we can understand at least some of environmental law, social welfare law (and the older poor laws), tax law, social and economic human rights, labour law, and others, as influencing and affecting the quality of relations between individuals and communities.

To make this a little more concrete, consider the evocative example discussed by Iris Marion Young in her last book (published posthumously), Responsibility for Justice (2011). The example involves Sandy, a single mother who ends up in a position of vulnerability to housing deprivation. What is unique, and also challenging, about Sandy’s story is that her vulnerable position is not the result of any particular bad government policy or any bad faith exercised by any particular individual. In fact, to the contrary, Young constructs the case in such a way that (at least some) individuals go out of their way to assist Sandy, and Sandy herself does nothing that we would not reasonably expect of her. The case, then, is one where the actions of the community, taken collectively and cumulatively, come to place one of its members – an individual – at risk of harm or suffering (in this case, vulnerability to housing deprivation).¹⁶xIn outline, the case is as follows (see Young 2011, p. 43-44). Sandy and her two children live in a central-city apartment building that has recently been bought by a developer who plans to convert the apartments into condominiums. Sandy works as a sales clerk about two hours from her current location, so the planned development gives Sandy an opportunity to look for an apartment closer to her work. As Sandy looks around (and she is assisted in this endeavour by a real estate agent acting beyond his duty), she discovers rental prices nearby to her work are beyond her reach, and that the few closer ones that are available are in neighbourhoods where she would be reluctant to raise her children. Given that it looks likely she will need to live further away, she invests in car payments. She also applies for a housing subsidy program, but is put on a two-year waiting list. As the eviction deadline looms, Sandy eventually decides upon an apartment that is smaller and further away than she wanted. However, as she has spent some of her money on car payments, she no longer has the three-month deposit required by the landlords. Accordingly, Sandy now faces the prospect of homelessness. Young calls Sandy’s predicament a case of ‘structural injustice’, defining that to exist

‘when social processes put large groups of persons under systematic threat of domination or deprivation of the means to develop and exercise their capacities, at the same time that these processes enable others to dominate or to have a wide range of opportunities for developing and exercising capacities available to them.’ (Young 2011, p. 52)

Young’s project here is certainly progressive: she is identifying a certain kind of vulnerability that clearly lowers the quality of relations between the community and the individual. We can deal with this in a reactive manner, e.g. by asserting that Sandy’s vulnerability is a result of the community’s actions, we make it justifiable to draw on the community’s resources to subsidise Sandy’s access to housing. We might also make it proactive or prophylactic, e.g. we could impose duties on real estate developers who wish to knock down or upgrade a building to find reasonable alternative accommodation for poor persons currently living in that building. In both cases, such changes might require the reform of the law. It is important to see that such reforms might not even come to the table without an inquiry into the specific kinds of vulnerabilities that arise in certain relational contexts. In the above case of Sandy, no one individual did anything that would qualify as making Sandy vulnerable in the relational context of relations between individuals – indeed, as noted above, individuals went out of their way to assist Sandy. Of course, we might find, upon making such an investigating, that some of law can already be made intelligible by understanding it as aiming to protect such vulnerabilities, or indeed having decided not to (e.g. having decided that, in the end, to place such heavy duties on real-estate developers would be to place the real estate market at too much of a risk, thereby once again managing vulnerabilities). But we might very well find that the law currently does not recognise such vulnerabilities, and then we, as theorists and scholars, can go on to make a case for law to do so. Relational jurisprudence is, then, a deeply engaged, progressive practice, but, simultaneously, by no means one that operates in the high altitude of utopias or ideal worlds. This is a jurisprudence that has one ear to the ground and one eye on the horizon.

4 Conclusion

This paper has attempted to provide both an object and, to some extent, a method for inquiry into law. In terms of the object of the inquiry, this paper has advocated situating law’s role in a series of different relational contexts. In terms of method, it has called for investigating law’s role in such relational contexts by employing factual-evaluative complexes – such as vulnerability – as devices for both making intelligible and assessing the way law affects and influences those relations going more or less well. The use of such devices requires commitment to utilising a mix of empirical and normative methods. In other words, the task of investigating how law manages vulnerability – which vulnerabilities it protects, and which it does not, or how it balances their mutual protection – is a task that cannot be performed well without intermingling empirical and normative methods. In the result, the proposal is for a jurisprudence that is engaged, but carefully and realistically so, paying attention to the ways in which we have thus far recognised vulnerability and also managed vulnerability with law, but going on to identifying vulnerabilities we have missed, or noting the emergence of new vulnerabilities, and considering whether law is a good vehicle for protecting them. In addition, this is a jurisprudence that recognises law’s ability to itself create and/or exacerbate vulnerabilities. In each of those tasks we see echoes of what some might characterise as precisely the fact/value separation – after all, someone could say that such echoes are evident in the very talk of ‘empirical’ or ‘normative’ methods. But this is nothing to be scared of: ‘facts’ and ‘values’, the ‘empirical’ and the ‘normative’, are terms that are useful; they help us, for instance, to see the complexity of the task that awaits us. What we must not let the usefulness of the distinction do is convince us that there is something metaphysical at stake. For, when facts and values are sold on the metaphysical market, they make it more likely that we will crawl like little petrified crabs into methodological sand castles – either empirical or normative – never venturing out. When we do that, we blunt our inquiry into law.