Regulation Enables: Corporate Agency and Practices of Responsibility

Both advocates of corporate regulation and its opponents tend to depict regulation as restrictive—a policy option that limits freedom in the name of welfare or other social goods. Against this framing, I suggest we can understand regulation in enabling terms. If well designed and properly enforced, regulation enables companies to operate in ways that are acceptable to society as a whole. This paper argues for this enabling character by considering some wider questions about responsibility and the sharing of responsibility. Agents who are less able or willing to act well are obviously more likely to face criticism, mistrust, and adverse responses. It will be more difficult to hold those agents responsible, especially so when there are many who fail in their responsibilities or where there are wide-reaching disagreements about those responsibilities. Regulatory standards, like other norms and ways of defining responsibilities, address these problems: by restricting, they also enable social cooperation. Like other forms of holding responsible, ways of enforcing those standards against recalcitrant agents, or encouraging conformity to them, may also seem restrictive. Again, however, these practices play an important role in enabling responsible agency. This is partly because they can bolster readiness to act well in agents who experience or witness such responses. It is also because they free other agents to exercise initiative and commitment in defining their individual responsibilities in line with higher standards.


Introduction
Corporate regulation is a central issue of our time. Some voices deplore regulation as a limitation on freedom and enterprise. Companies and lobbyists talk about regulation as 'red tape' and extol the virtues of self-regulation. These claims have had some impact in removing regulations or enforcement mechanisms. Many other constituencies criticise this weakening of regulatory regimes or call for closer regulation of corporate activities. The concerns are well known and wide ranging: public health risks, environmental damage, harmful financial systems, the commercialisation of childhood, heavy market concentration, poor employment conditions, the confusion of information with propaganda, and much more besides.
Both opponents and advocates of regulation usually depict regulation in restrictive terms-a policy measure that limits freedom in the name of welfare or social goods. 1 In this paper, I will argue that this framing is mistaken and places the burden of proof too heavily on advocates of various regulatory measures. Instead, I believe that we should understand regulation in enabling terms. If well designed and fairly enforced, regulation can enable companies to do what they generally say they want to do: to operate in ways that are acceptable to society as a whole. This attempt to reframe regulation is open to obvious counter-argument. Like all rules and norms, regulations are bound to have restrictive aspects-they may cost money or limit the scope of specific markets; they may be badly framed or have unwanted effects (for example, discouraging new market entrants). As with other rules and norms, regulation raises 1 For an influential statement of this view, see Nuffield Council on Bioethics (2007). Although it eschews the language of paternalism (or even 'nanny state') in favour of 'stewardship', this report therefore repeats the framing of such interventions used by industry front groups such as the US Center for Consumer Freedom (e.g. 2012). I also note one referee's comment, to the effect that regulations in particular markets operate as conditions rather than restrictions since there is, after all, no obligation on any corporation to enter a particular area of business.
1 3 questions of responsibility. Compliance cannot be taken for granted; enforcement may prove hard or impossible, especially where a powerful actor disputes the validity or applicability of a given norm. Above all, regulations will only seem enabling to an organisation that is ready and willing to act on the standards that they lay down.
To defend a more positive, enabling view of corporate regulation, this paper considers some fundamental philosophical questions about responsibility and the sharing of responsibility. In particular, I will highlight the collective endeavour that responsible agents can and must engage in, to allocate and divide responsibilities between themselves. (Throughout, I use the term 'agents' to refer to both individual persons and to collective actors like corporations, regulatory agencies, and legislative bodies.) There is no doubt that this endeavour involves limits and restrictionsthe definition of specific responsibilities and attempts to hold responsible where agents fail to fulfil these. My claim will be that these restrictions also have an enabling character, in the sense of enabling cooperation and also in fostering and constituting abilities to act responsibly.
In highlighting this collective enterprise, I mean to strike a different path to that taken in most philosophical accounts of responsibility. The past decades have seen important discussions of collective responsibility, exploring the ways in which we can compare the capacities and actions of collective actors-such as business corporations-to those of individual persons, and the ways in which it may be appropriate to hold such collectives responsible. 2 However, there is another way in which collective aspects might figure in a theory of responsibility. 3 Philosophical accounts tend to focus on the capacities and actions and liabilities of a specific agent, be it individual or collective. This means that they sideline the ways in which agents enable one another and share responsibility by treating one another as responsible agents. In this paper, I suggest that we can see these forms of sharing and enablement when both individual and collective agents interact. As well as being especially urgent, the corporate case is illuminating because the standards that corporate actors ought to fulfil are so disputed. This makes it harder to bracket difficult questions about the norms that agents ought to live up to; hence it invites us to consider how those norms should be decided and enforced, so too the actual practices involved in holding responsible and the ways in which those practices must relate to the capacities of both accuser and culprit. When we consider business corporations, two vital background conditions of responsibility become obvious: the allocation, division, and sharing of responsibilities between agents; and the fostering of capacities to fulfil these responsibilities. Both conditions reveal how setting and upholding standards has an enabling character. 4 Before setting out the structure of the paper, let me say a word on the relevant meanings of 'responsibility', since these are so central to my argument. 5 I will be concerned only with normative uses of the term, where 'normative' covers a variety of standards: moral, legal, and regulatory. This closes off purely causal usages, such as 'The mutation of the virus was responsible for the virulence of the epidemic', but still leaves a lot open. First, we can speak of responsibility in retrospective and prospective senses. 6 Retrospectively, we may hold an agent responsible for various actions or omissions, and I will underline the variety of practices involved when we do this. Prospectively, responsibilities can refer to duties or roles or standards: for example, responsibilities to comply with regulations, to look after a particular child, or to honour a contract. There is a simple analytic connection between these: it will only be justified to hold an agent responsible (retrospective) where he, she, or it has somehow failed to fulfil his, her, or its (prospective) responsibilities. As I will stress, however, this connection is complicated in practice: if there is dispute about the relevant standards (prospective), attempts to hold responsible (retrospective) are bound to prove contentious. Second, agents can bear, take, and share responsibility. These things can be done both prospectively and retrospectively. A regulatory authority is created to bear or take responsibility for deciding or enforcing standards in a particular area (prospective); should it fail to regulate well, it might also bear responsibility, in the sense of being open to blame or sanctions, or take responsibility, in the sense of striving to make amends (retrospective). Several companies might share responsibility in the sense of collaborating on a self-regulatory initiative (prospective); they might also bear or share responsibility, in the sense of all being partly to blame for the collapse of certain standards in their business sector (retrospective). Third, note that the adjective 'responsible' can be used in three ways: (i) A responsible agent may refer to an agent with certain capacities that make it appropriate to regard the agent as bound to fulfil some (prospective) responsibilities and liable to be held responsible (retrospective) if it does not-in this case, the contrary is 'non-responsible', like the virus mentioned above. (ii) We may refer to an agent as responsible when assigning blame or liability or credit (retrospective, e.g. 'The CEO was partly responsible (that is, culpable or blameworthy) for the company's failure')-in this case, the contrary is 'not responsible' in the sense of innocent or uninvolved. (iii) We may praise an agent as 'responsible', as having particular virtues (Williams 2008)-in this case, the contrary is to criticise an agent as 'irresponsible'. This sheer variety of meanings may raise doubt as to whether 'responsibility' is a term fit to bear theoretical weight. On the account offered here, however, the uses are importantly related, and the paper's final paragraph summarises the connections highlighted by my argument.
The paper is in five sections. The first section highlights a philosophical perplexity about responsibility that arises from one prominent way of understanding the connection between an agent's normative capacities and his, her, or its culpability. I will suggest that this apparently theoretical perplexity would be better approached in practical terms-that is, in terms of difficulties in sharing responsibility and holding responsible. The second section considers one side of this difficulty-the problems and practices bound up with holding responsible when agents are less able or willing to take responsibility (in both prospective and retrospective senses). The third section considers the other side of the problem: how to arrive at defensible agreements concerning the standards different agents ought to fulfil and how responsibilities should be distributed. In particular, I suggest that the less responsible the agents are (in the sense of lacking the virtue of responsibility), the less able they are to appreciate the responsibilities (prospective) that they ought to fulfil. As a result, others must step in to define these for them. In this case, regulations and responsibilities appear in the guise of external impositions and restrictions, thus obscuring their enabling role. In a short fourth section, I draw some lessons for the relation between capacities and retrospective responsibility, before returning to the enabling role of corporate regulation in a fifth section. This underlines the ways in which defining and upholding the roles and responsibilities of different actors facilitates cooperative relations and fosters capacities for responsibility.

A Philosophical Perplexity About Responsibility
It is now quite common in the philosophical literature to consider responsibility in terms of an agent's responsiveness to reasons. The thought is simple: an agent that cannot respond to moral reasons is not responsible-thus animals, infants, insane persons, crowds, or perhaps even corporations (if these are understood, for example, in Bakan's (2004) terms as psychopaths blind to all reasons apart from self-interest). Responsible agents can understand and act on (normative) reasons, whereas non-responsible agents cannot. Such thoughts are often bolstered by the maxim 'ought implies can' 7 : there cannot be duties without a capacity to fulfil them-or at least it is unfair to hold an agent to duties if she lacks the ability to fulfil them. If we think of holding responsible as imposing a cost or harm, a limitation or a restriction, then an agent ought to have a fair opportunity to avoid this. We find this sort of view in many writers advocating compatibilist accounts of responsibility, such as Wolf (1990), Wallace (1994), Fischer and Ravizza (1998), Watson (2004), and, most recently, Nelkin (2011). These accounts overcome an obvious perplexity of metaphysical free will (how is responsibility meant to follow simply from being uncaused to act in any particular way?) by putting weight on capacities of obvious practical importance: the abilities to consider reasons and to act on them. Such accounts also accommodate the developmental nature of responsibilityas children develop these capacities, we are more inclined to treat them as responsible for their actions-and the aberrations of agency that impair the responsibility of otherwise normal adults-forms of compulsion and addiction, for example. Most important for the present paper, they suggest a promising approach to corporate responsibility (Hess 2013). Some collectives are structured so as to respond to reasons-courts, for example, carefully weigh reasons to judge whether the accused has transgressed a legal requirement; equally, it makes more sense to hold a court accountable than a crowd, say, which lacks such a structure.
I hope this sounds fair enough. But it leads to a strange result. Assuming that agents lie along a continuum in terms of capacities to respond to reasons, then less able agents will count as less responsible. If an agent consistently acts badly, then we lack any empirical basis on which to attribute an ability to respond to the relevant reasons; therefore that agent is non-responsible or less responsible for its wrongdoing. 8 In other words, it is precisely those agents who act worse-and who we might normally think of as needing to be held responsible more often and more stringently-who count as less responsible for their misdeeds. To use a corporate example: a badly managed company, with second-rate executives dominated by short-term shareholders, in a poor regulatory environment-well, no surprise that it commits all sorts of malfeasance and proves unable to appreciate or act on reasons to do better. If culpability and capability go hand in hand, then such incorrigible, intractable wrongdoers seem not to be responsible. 9 By the same token, more virtuous agents seem to be more culpable should they, on occasion, commit comparable misdeeds. 10 I believe that the conceptual puzzle arises because this framing-and, in particular, this way of using the adjective 'responsible'-brackets the interactional aspects of responsibility. In seminar discussions, I have found it striking how often the question is posed: Is a corporation, or some of its members or officers, really responsible for its wrongdoing? As others have remarked, the tendency to add an emphatic really to an abstract but everyday term (for example, real, true, or good) often betrays a philosophical misstep: we move onto metaphysical terrain, without necessarily appreciating the structure of use and practice that we are leaving behind. In this case, a question about how agents should interact has been framed in terms of a single agent's attributes. 11 As noted in my introduction, some entities, such as viruses or infants, are plainly non-responsible: it would be absurd to ascribe duties to them or to hold them responsible. 12 But the question of how to deal with agents who persistently inflict harms and wrongs-be they vicious individuals or irresponsible corporations-is not so easily decided.
My proposal, then, is to step back and emphasise the practical aspects of attributing responsibility. The corporate case is especially illuminating because the practical difficulty (that is, how to get to grips with the irresponsible corporation) is more compelling than the theoretical one (that is, whether it is really responsible for its misconduct). The practical difficulty has two aspects, both prominent in debates about corporate regulation. First, can any other agent meaningfully hold it responsible? To answer this, we need to consider the variety of responses that we might make when an agent acts wrongly. Second, how can or should the various parties reach agreement on what counts as right-or wrongdoing? To focus on whether the agent is 'really responsible' obscures this problem, not least because it neglects one of our most important forms of collective action: the allocation of responsibilities to specific agents. Again, the corporate case makes clear what we should have noticed in the individual case. Sharing out, judging, and revising responsibilities is primarily a matter of collective deliberation and mutual accommodation. Individual reasoning could never, on its own, decide the matter. 13

Practical Implications of Limited Abilities to Act Well (I): Holding Responsible
Let me start with the question of what is involved in holding an agent responsible. First, notice that if we conceive of this as something simply done to an agent-such as a prison sentence or a fine or a removal of privileges-then there needs to be a large power inequality between accuser and culprit. Otherwise, there is a good chance that the culprit can evade whatever penalties the accuser imposes, or pass these penalties onto others, or take counter-measures against the accuser, or hide future bad behaviour, or all of these (Rubenstein 2007). The history of corporate regulation provides many examples of these problems.
Second, even given a marked power inequality, it is barely possible to hold a culprit to his (or her, or its) responsibilities-by which I mean, to ensure that the agent actually acts and responds as he (or she, or it) is meant to. Notwithstanding the difficulties just mentioned, a powerful and credible threat of sanctions certainly represents an incentive to change behaviour. However, altering some incentives facing an actor does not, in itself, alter other factors that generate an agent's readiness to act badly. Notoriously, many corporate actors seem to view fines as a cost of doing business. In other words, the underlying regulatory norm is not internalised as having any validity in its own right; the fine may have some effect, but the norm remains external to the actual policies and procedures of the organisation. In such cases, accusations tend to generate ill-will, obfuscation, and counterallegations; sanctions rarely lead to improvement and often have unwanted effects, not least in terms of the resources that the accuser must use to impose them; more intensive measures, like sending in regulatory staff to monitor behaviour 13 See the section titled, 'Practical Implications of Limited Abilities to Act Well (II): Reasons Versus (Prospective) Responsibilities', which argues for the collective dimension of these tasks and that individual reasoning relies on this. 9 Watson notes the difficulty and contends that we must respond to such agents broadly as if they were responsible, condemning the acts and acting to limit the damage they might cause; but, he argues, these responses 'lack their normal expressive function ' (2004, p. 281). See also Smith (2008) and Hieronymi (2007) against the idea that inability implies non-culpability. 10 In the final section, I will offer a different explanation as to why we may respond with more vigour to the wrongdoing of more virtuous agents. 11 In another context (that of moral motivation), Herman suggests that it can be helpful to approach a conceptual puzzle as a practical one (2007, p. 266). I think the wisdom of this maxim owes to the fact that practical difficulties can reveal a mistaken theoretical frame. In my context: what looks like a theoretical difficulty (are these incorrigible wrongdoers responsible?) is more aptly framed as a practical one (how are we to interact with those persons and organisations?). 12 Compare the oft-cited historical example of criminal trials against animals, from rats to pigs to insects (Girgen 2003). on a daily basis, require considerable energy and attention, and their results are unlikely to be uplifting. Rewards for approved forms of activity-such as tax breaks or less stringent monitoring-may seem more constructive, insofar as they are less likely to cause ill-will or evasion. Again, however, unless the agent actually endorses the norm that motivates others to offer these benefits, improved conduct will always remain contingent on the threat of their withdrawal; we can expect only the required minimum, or even feigned compliance.
However tasty the carrot and however big the stick, then, it is surprisingly difficult to find measures that exceed the effectiveness of mere words-be they exhortations to do better or words of blame and protest. Yet 'mere words' have their risks too, since every agent has reasons to be jealous of its standing and reputation. Hence even the most powerful regulatory agencies face continual frustrations-difficulties in detecting and confronting wrongdoing, the ineffectiveness or perverse effects of available penalties, the encounter with aggressive counter-tactics that challenge or undermine their authority, the feeling of playing Whac-a-Mole-that is, the sense that if one tackles one sort of malfeasance, it is sadly predictable that another sort will pop up elsewhere. As individuals in everyday relationships or within organisations, we should be equally familiar with these problems. Even more than regulators, who after all are somewhat insulated from the effects of companies' malfeasance, we often feel the need to hold someone responsible and the impossibility of doing so. 'Least said, soonest mended', we mutter to ourselves, as we let slide an infraction by someone we care about or have no choice but to associate with. As most managers can testify, formal superiority does little to alter the basic dynamics, at least where one relies on ongoing cooperation and cannot simply banish the offending actor and costlessly replace him. The problem is: how to really make contact with the basic policies, motivations, and moral sense of an agent who, thus far, has proved unwilling to live up to the responsibilities that one believes he (or she, or it) should accept?
Notwithstanding the difficulties and frustrations bound up with the various ways we attempt to 'hold responsible', we should also note the more fruitful possibilities. In philosophers' language, these depend on our achieving some shared sense of the moral and practical reasons involved. For a rare example of blame that proves entirely constructive, think of Jane Austen's Emma, admonished by Mr Knightley for her cruelty toward Miss Bates at that unforgettable picnic on Box Hill. 14 Emma immediately feels the force and fairness of Mr Knightley's rebuke. The relationships between all those involved are deepened; moral considerations are better appreciated. But we should also remember that Emma is a heroine-capable of self-deceit but basically honest with herself, spirited but fundamentally concerned for others, willing to question Mr Knightley's judgment but fully able to appreciate his character and integrity. I say this not to express doubt about the capacity of some corporations to respond to accusations and adverse responses in an analogous way, and much important work has been done on the appropriate strategies (e.g. Braithwaite 2011Braithwaite , 2015Bovens 1998;Fisse and Braithwaite 1993). Nonetheless, once things have gone wrong enough for a company to create significant harms or breach important regulatory standards, I think it is fair to say that it will require as much luck as judgment to elicit a constructive response. After all, we know that organisational change tends to be a fraught process, and that policies and commitments do not change overnight. 15 In more general terms, the point is that holding responsible tends to prove an exercise in frustration and even futility unless we make contact with an agent's ability and willingness to take responsibility. The less responsible an agent is, in the sense of lacking the virtues required to appreciate and act on the relevant considerations, the harder this will prove. But when there is some sort of success, what actually occurs is a form of joint action. The accuser takes responsibility by confronting the culprit, as opposed to merely ignoring the wrong or putting up with the problem. 16 The culprit takes responsibility by acknowledging its fault, as opposed (for example) to merely paying the fine and hoping to get away with similar (mis)conduct in future. If the culprit offers compensation, apology, or commitments not to repeat the offence, it then falls to the accuser to accept these-perhaps with some further response, conditions, or concessions. In 14 Jane Austen, Emma [1815], ch. 43 (or Vol. III, ch. VII depending on the edition). 15 To illustrate the point, note two recent examples of companies' responses to grave problems in their supply chain. Apple and H&M both have many virtues; both have acknowledged their responsibility to address the issues; and both have been praised for their responses. But in the years since scandals first came to public notice, the companies have made only partial progress-not least, because they continue to place incompatible demands on suppliers (on the one hand, rapid, lowest-cost production; on the other, safe and humane working conditions). See, e.g., China Labor Watch (2017), Clean Clothes Campaign et al. (2016), and Preston and Leffler (2016). 16 This aspect of accusation is well stressed by Goodhart. He refers to 'anti-sweatshop and other global justice activists [as] taking responsibility for injustice even though they do not, in either the philosophical sense or the prevailing conventional sense, have it [that is, they are not to blame for the problems they are highlighting-GW]. The very point of their activism is to educate people about the implications of their choices and habits, to try to change popular thinking so that consumers begin to regard themselves and others as responsible for the effects of their behavior. They are seeking to change conventions, not as a way of discharging responsibility [that is, their own guilt or culpability-GW] but in hopes of pinning the blame where they believe it belongs ' (2017, p. 191). these ways, accuser and culprit share responsibility: they act together to craft a worthwhile response to the original wrong and find ways to put it behind them. 17 I have emphasised that the process is rarely as uncomplicated as Emma's acceptance of Mr Knightley's rebuke, and resolutions rarely as quick and uncontentious. Nonetheless, to prove constructive 'holding responsible' cannot be something merely done to the culprit: the parties must act together, in a shared taking of responsibility.
To put the same point in terms of my central concern with enablements and restrictions: practices of holding responsible may well seem restrictive-after all, they mean that agents cannot simply act as the moment takes them, without liability to adverse responses from others. At their best, however, they enable agents to share responsibility: to create a shared acknowledgement that some mode of conduct is unacceptable; to foster the will and the means to avoid its recurrence; to make reparations and repair relationships; and even to strengthen schemes of cooperation.

Practical Implications of Limited Abilities to Act Well (II): Reasons Versus (Prospective) Responsibilities
Underlying the question of what it means, in practice, to hold an agent responsible is a second difficulty: how to decide the standards an actor should be judged by. As pointed out in the introduction, any assignment of retrospective responsibility, be it blame, sanctions, or some other measure, must invoke a claim about prospective responsibilities-that is, it will rest on assumptions or assertions about what the actor ought to do and to have done. Among other problems, I have been pointing out that we should not assume a shared sense of right and wrong between accuser and culprit. Such disagreements represent a major practical obstacle for any attempt to hold responsible that goes beyond mere hard treatment by a more powerful agent. 18 Disputes about corporate responsibilities and appropriate regulation illustrate the point all too well.
Contemporary philosophical talk of (responsiveness to) reasons tends to obscure this problem. The language of reasons is highly abstract; moreover, reasons are meant to be out there, waiting to be discovered, apparent to any rational mind. But as is obvious when we think about corporate regulation, consensus is a fragile achievement and disagreement to be expected. I suggest it is better to frame the matter in terms of responsibilities instead. This suggestion may seem unfortunate, since it brings another cognate of 'responsible' into play. 19 However, I see two arguments for making this shift. Both bring us closer to the problems and realities of corporate regulation, as well as to the specific duties that individuals bear, as citizens or employees or other roleholders. And both help us to see the mutual enablement and sharing of responsibility involved when agents arrive at a common sense of the standards and roles that they should live up to.
One reason for speaking of responsibilities rather than reasons is that it reminds us that the relevant normative considerations vary between agents, be they individual persons or organisations. Responsibilities must be divided, allocated, and taken up. We expect all agents 20 to abide by certain minimum standards-law, regulation, and everyday morality all have a crucial role in stipulating these. But a central aspect of human freedom lies in our ability to choose which roles and responsibilities to take on, and even to carve these out as our personal vocation. This applies equally to institutions. Economic and political freedom consist partly in the endeavour to create, maintain, and reconfigure various forms of collective action. Decent organisations, like decent people, exercise initiative and commitment not just in carrying out their responsibilities, but also in defining them. A successful business corporation will build up a distinctive internal culture and external reputation; it will specialise in particular products or services, developing ways to deliver these reliably and competitively. The organisation carves out a particular role for itself, defining its responsibilities vis-àvis customers and suppliers and employees. As I will stress shortly, these responsibilities are not set in stone. But they still represent meaningful commitments, which both insiders and outsiders expect the organisation to live up to.
As a result, no one supposes that any actual or ideal division of responsibilities could be specified by Reason, capital R. (One might read Plato's Republic as illustrating the 17 Having said that this interactive aspect of responsibility rarely features in philosophical accounts, I should also note important exceptions. Walker (2006) has stressed the variety of possible responses and the range of parties who might cooperate to create these. McGeer's recent work (2012, 2015a) is also noteworthy, as is McKenna's (2012) image of holding responsible as a 'conversation' between accuser and accused. 18 Apart from moral objections to hard treatment, simply as such, we might recall the familiar difficulties of inflicting it on collective agents. 'No soul to damn and no body to kick', in the maxim made famous by Coffee Jr. (1981). While there may be 'assets to seize', we know how easily such penalties are passed on to those with no part in the original wrongdoing. See also Fisse and Braithwaite's classic discussion (1993, esp. 41 ff, 82 ff, 140 ff), which highlights many options that avoid or minimise such difficulties. 19 Or to put the point in more positive terms: it can help us to see how the different meanings of responsibility are related, as I will summarise in conclusion. 20 That is, all agents of a particular type: obviously the basic norms differ, depending on whether we consider a state or a church, a business corporation or an individual person. despotism inherent in deciding such matters by solitary reasoning (Arendt 1968) or, alternatively, as an ironical reductio of the attempt.) These are social and political matters, involving imagination, freedom, negotiation, contestation, compromise, and adaptation to circumstances. As such, they have an essentially historical aspect. Whatever timeless Reason pronounces, human beings can only create a division of responsibilities by adjusting the one that they already have. No doubt, there are well-reasoned criteria to steer our deliberations as we try to judge what might count as an improvement, such as justice, stability, or welfare. But what allows more abstract ideals to guide action-and hence enables constructive interaction-is a broadly accepted sense of the responsibilities and freedoms belonging to different actors. We see this within large organisations, where an accepted sense of the powers and duties of different role-holders and departments empowers them to play their parts-and, in turn, constitutes a functioning collective actor. 21 And we see it at the wider level of interaction between corporations and other actors. A critical task of regulation, just like that of other forms of law and social norms and customs, is to specify the scope of legitimate action and initiative for different actors. When agents act in line with shared knowledge of who is entitled to do what, a whole host of potential frustrations is eliminated; trust and cooperation are deepened.
Differently put: what is required of a responsible agent is not responsiveness to reasons in the abstract; instead, a responsible agent must show responsiveness to this framework of responsibilities and the ways that other agents act and respond within it. 22 A second point follows from the fact that such frameworks are sustained out of action and initiative. Like the wider circumstances in which they are situated, they are never fixed or final. Many responsibilities, for individuals, for role-holders, for business corporations, and for other organisations, have been worked out over a long period. So they represent fairly stable reference points for practical judgment, providing an anchor in terms of each agent's specific purposes or goals, and the criteria by which actions and outcomes should be judged. But these responsibilities invariably leave problems to be addressed and are open to contest in various ways. As I have noted, reason implies something an agent can undertake alone, or at least the tidy consensus of rational minds, marred only by the errors of the irrational. But responsible agents know that they must live with disagreement, uncertainty, and change; they know how events can easily disrupt even the best-established division of roles and responsibilities; they know that mistakes can be expected, and also that it often proves unclear whose mistake it was. A business corporation may find its market disappearing as new technologies emerge, for example, and have to redefine its offerings and role in the marketplace. Or a company may find that it has become reliant on suppliers whose practices are unsustainable or unjust, and face fierce criticism (whether from inside or outside) for this. Resolving such problems and the disagreements that often ensue tends to be a matter of negotiating and creating a more satisfying complementarity of roles and expectations, rather than deciding who got it wrong. 23 Some readers may baulk at my talk of complementarity and responsibilities in the context of business corporations, where we hear so much about competition and self-interest. I am stressing that responsibilities belong to broader schemes of cooperation, which need to be developed and agreed, and thereby enable the different actors involved. But this does not mean that different agents must pursue a shared goal or even deliberately cooperate, let alone reach a miraculous consensus. It means only that their responsibilities should mesh within a larger framework. I already mentioned the complementarity of roles within a reasonably functional organisation. Directors' and employees' parts must add up to a whole. But they will also involve opposing priorities: those of a safety officer may well conflict with those of the financial accountant, for example. This point applies equally to market competition. Market actors have a responsibility to compete, in fact, and not just an interest in acting competitively. Indeed, in another sense they have no interest in competing, since the freedom of suppliers and buyers to go elsewhere constitutes a major constraint upon a corporation's ability to set terms and prices. 24 We have seen the result of this whenever 'anti-trust' regimes have been weakened, in the US and elsewhere: competitor firms are liable to merge or buy one another out, leaving a market dominated by a very few actors, so that suppliers or buyers lack meaningful options. This sort of market concentration may serve corporate interests all too well, as may other forms of cartelisation 21 At the risk of overusing the term 'constitute', let me note the political parallel: a constitution lays down the responsibilities of legislature, executive, and judiciary, and thereby constitutes the state. Interestingly, this lesson stems from the early modern experience of corporations, where the charter played a similar role (Ciepley 2017). 22 Similarly for an agent who judges or responds to another. In one sense, this is a matter of her (or its) individual agency and reasoning. (I blame him; the regulator criticises the company, and so on.) I am emphasising the context in which such judgments make sense, and how the underlying reasoning of the agents must begin from some existing social division of responsibilities, even if it takes critical distance from them. and collusion. But the basic structure of meaningful competition has been lost. In other words, we should remember the parallel with other rule-governed forms of competitionthus many games, or the legal contest between defence and prosecution, or constitutional checks and balances. Although it would be strange to speak of these actors cooperating, simply as such, we cannot understand their responsibilities without reference to a wider scheme of cooperation, which renders their competing activities complementary and not merely conflictual.

Against the Question, 'Is the Agent (Really) Responsible?'
In the last two sections, I have highlighted two achievements that are central to our practices of responsibility. On the one hand, agents need to arrive at a working division of responsibilities. This will include some common responsibilities, such as regulation lays down for actors who engage in specific modes of economic activity, or law and everyday morality lay down for individual persons. It also involves differentiation: the specific responsibilities that individuals and organisations bear, in terms of particular roles (as a business corporation, as a lawyer, and so on) and in terms of particular commitments (as this corporation, with these employees and this track record; as this lawyer, with these clients; and so forth).
On the other, responsible agents must find modes of joint action to address failings-including ways to hold one another responsible. Such responses will certainly include punitive measures, at least so far as power relations permit, in order to insist on certain basic standards and to hinder those actors who are unable or unwilling to play even a minimal role of their own accord. I have stressed how difficult these practices are-how much friction they involve, how often 'mere words' prove costly, how limited carrot and stick are bound to be. More constructive possibilities, by contrast, rest on cooperation that clarifies responsibilities and fosters their acceptance by all involved. There must be agents who bear, or who are willing to take, responsibility for pointing out when others fail to do their parts (albeit with the awareness that it is sometimes hard to decide whether the agent is at fault or whether supposed responsibilities have been wrongly judged or ascribed). If the overall scheme of cooperation is not to falter, there must emerge a new or renewed resolve to divide responsibilities in a particular way that fosters active commitment and requires only sparing use of sanctions. 25 To suppose that an agent can literally be held to her responsibilities-that is, be somehow forced to fulfil what some other agent sees as her responsibilities-is to be misled by language, or even tempted to tyranny. Agents must take responsibility and make those responsibilities their own.
In terms of the philosophical puzzle about capacities, I have suggested that degrees of culpability do not relate to degrees of capability in any simple way, and certainly not as if less able agents were less culpable for their misdeeds. One problem is the mistaken abstraction involved in treating culpability as a scalar property. This makes too much of the fact that there can be more or less harsh responses and not enough of the variety of responses open to us. As individuals, we may blame or avoid or alter relations or demand compensation or even punish, depending not only on the wrong but also on our assessment of the wrongdoer. Regulators and policymakers, not to mention consumers and NGOs, may similarly reach for a whole range of different measures, depending on their powers and assessment of the culprit, case, and context. 26 A related problem arises from the failure to appreciate the enabling character of attempts to ensure that normative standards govern other agents' conduct, as well as our own. Other philosophers have recently made similar points: both Vargas (2013) and McGeer (2015b), for example, have argued that practices of responsibility provide a 'moral scaffolding' or 'moral ecology' for our agency. That is, capacities to act decently are bolstered when those around us endorse defensible norms, and are willing to remind us of them and to respond critically to us if we fail to abide by them-or, more happily, to reward and enable us when we act well. Instead of picturing responsible agents as possessing an independent capacity to respond to reasons, the emphasis shifts to abilities to participate and respond to others' responses, and to the ways in which these interactions can develop abilities to act responsibly. Of course, 25 This is merely to telegraph a point that has been made with great thoroughness by Braithwaite in his account of 'responsive' or 'restorative' regulation. (For summary, see his 2011.) This advocates graduated or 'pyramidal' sanctions in the aim to bolstering organisational capacities to act well, rather than merely deterring. Just putting an agent 'on notice' may well be enough, in the better cases, to lead it to actively take responsibility. If this fails, and increasingly stronger measures fail too, this implies that an agent does not really accept, or is otherwise incapable, of meeting a particular duty. As such, it is not fit to participate in a scheme of cooperation that requires such standards-so it may be apt to withdraw its licence for particular business activities, or even to consider the 'corporate death penalty.' Footnote 25 (continued) for various purposes we may still wish to distinguish agential capacities from surrounding conditions. But corporate agency, like other artificial forms of agency, demonstrates how profoundly agents' capacities are bound up with the affordances and constraints surrounding them. (For example, as noted in the previous section, the vital role of competition to the good functioning of most corporate actors and sectors.) Hence my suggestion that capacities for responsible conduct are constitutively tied to practices of responsibility, by which I mean both the collective enterprise of deciding which agent bears which responsibilities and the sharing of responsibility involved when agents hold one another responsible. 27

Corporate Regulation as Enabling
The previous sections recall some familiar problems of regulation and its enforcement alongside some more abstract questions about responsibility. In this final section, I consider what they can tell us about how corporate regulation can play an enabling role, and not merely a restrictive one. The largest problem for this suggestion lies in the fact that restriction and enablement are relative quantities: they can only be assessed in terms of wishes and intentions, purposes and capacities. Readers of this journal will need no reminding how vexed it is, to ascribe these to business corporations. I will not try to scale this peak in a few paragraphs, but I do think some helpful pointers follow from the previous discussion.
One is that business activities, however competitive, also belong to a cooperative framework: this can be more or less hospitable or hostile to corporations' activities, interests, and purposes, however those are finally judged. Surrounded by many other actors who may decide whether to trust it, to buy or sell from it, to compete with it, or to publicise or praise or protest its activities, a corporation depends on its reputation and its capacities to sustain credible commitments. A similar point applies to the members of these complex organisations, who are bound to seek a range of commitments from it and to take different views of the priorities and interests that it ought to serve. 28 In other words, unitary ascriptions of purposes or interests-profitability or shareholder value, for example-are misleading, since many different criteria will be deployed by different agents. In one sense, this means that my question permits no a priori answer: a regulatory requirement is almost inevitably bound to be constraining in terms of some interests or purposes and enabling in terms of others. 29 In another sense, it means that any judgment about enablements or restrictions is inherently normative-that is, it must invoke claims about the proper purposes that business corporations should serve, how their interests should be defined, and the relative priority of those interests to those of other agents. Again, those questions are largely beyond my scope, except to underline two points.
First, constructive forms of interaction depend on some degree of agreement about the respective responsibilities of different agents. As I have emphasised, attempts to hold responsible always invoke views on the responsibilities that an agent ought to bear. Disagreements about those responsibilities are bound to spill over into contests about blame, reputation, and penalties. What regulation provides, in this context, is an anchor point for judgment and responsesthough not, of course, the final word on corporate responsibilities. This is true both from the point of view of other agents who are deciding how to interact with a corporation and from the point of view of the corporation itself. Insofar as it abides by publicly recognised rules, a corporation is well placed to rebut some demands as unreasonable. 30 Or to put the same point more positively, it is entitled to the rewards of recognition and cooperation; it wins a 'licence to operate'. Of course, those standards may still be disputed and (as I will stress in a moment) more responsible agents may wish to go beyond them in all sorts of ways. But debating and revising those standards is a responsibility shared with other bodies, such as legislatures, regulatory agencies, and NGOs. In short, a sound regulatory context means that no corporation is expected to act or judge entirely on its own account or to respond as such.
To see how much of an enablement this is, compare our personal experiences of blame. We do not defend ourselves against an accusation by reasoning from first principles 28 In saying this, I do not mean to imply that the intentional states we attribute to collective agents need to be the same as those of all or some members, just to point out that the collectivity will serve a range of different purposes so far as those members are concerned. 29 Cf note 24 on the ways in which competition enables and restricts. 30 At a more abstract level, we can express this point in terms of Searle's well-known distinction between regulative and constitutive rules (1995, pp. 43-48). Some rules govern activities that are conceivable without them; other rules constitute new modes of activity, just as the rules of chess actually define the game or rules of exchange constitute trade. One might suppose that corporate regulation, as the word suggests, merely regulates-standards for product safety regulate manufacturing and sales, hence presuppose abilities to make and sell. However, like other legal norms, regulations also have an important constitutive dimension. That is, they constitute some actions as publically unacceptable, hence meriting authoritative intervention, and some as publically acceptable, hence entitled to non-interference or even approval. that it rests on an unjustified view of our duties. Instead, we invoke familiar social expectations, the responsibilities belonging to our roles, or well-recognised qualities of character and action. Those everyday standards may not be perfectly defined (we know sometimes societies have gone very badly wrong in defining them; we know that changing circumstances and conflicting initiatives are always liable to demand revisions). Indeed, we often learn about their imperfections when we encounter accusations from others and realise, perhaps quite painfully, that something is awry. But I have been suggesting that a key form of social cooperation is involved in reaching at least a working definition. This enables us to go about our business with some assurance that we are entitled to others' forbearance and cooperation; it provides sufficient everyday agreement for us to raise and deal with particular disagreements.
Second, as I noted at the end of the previous section, our agency and purposes are profoundly shaped by these expectations. For a well-socialised human being, most moral demands do not feel especially restrictive, because our agency has a certain degree of unity: our desires and inclinations are not at war with one another; they cohere with our own moral sense and our sense of what we should be doing with our lives. In other words, our agency is partly constituted by our responsibilities. The same is true for an organisation that has learned to accept certain standards: it takes these up into its structures and procedures and culture. No doubt, it will still experience incentives to break some of these, just as even the best of us is still sometimes tempted to do wrong. Likewise, it is quite possible that a virtuous agent (be it a person or a corporation) will make mistakes-in which case, however, the agent is in a reasonable position to recognise the justice of accusations, to make amends and put things back on track. In such cases, other agents' attempts to hold it responsible will be accepted, perhaps with some misgivings and friction, as enabling it act more responsibly.
Compare this with the continual difficulties faced by an irresponsible agent, one that makes only token acknowledgement of some responsibilities and standards (half-hearted procedures, a weak compliance department, mission statements that are more PR than policy). In such cases, the enabling aspect falters: responsible conduct will always be a struggle; the organisation will always be liable to the opposition and mistrust that irresponsible conduct rightly attracts. For such an organisation, some regulations are bound to seem like restrictions and some will prove ineffective in guiding its activities. For other agents to enforce the letter of the law requires constant and exhausting vigilance, not to mention a power position that regulators often lack. To recall the philosophical puzzle about capability and culpability: this is not to say that attempts to hold responsible will be out of order. Quite the reverse: regulatory sanctions, pressure by consumers and consumer organisations, protests by activists and NGO's-all will count as important contributions to upholding reasonable claims about corporations' responsibilities and enabling companies to do better and to become the sorts of agents that can really call themselves socially responsible. They create material and reputational incentives; they strengthen the hand of more responsible members and departments within the company; they increase the salience of responsibilities that might otherwise be ignored. But the enabling character of these efforts will be harder to make out.
It is also worth underscoring the collective dimension of this problem. Evoking the image of rule-governed competition, regulation is sometimes pictured as creating a 'levelplaying field'. This points to a straightforward way in which regulation enables. For corporate agents who want to act responsibly, codified regulations and corresponding sanctions represent an important assurance mechanism. They lessen the risks that a business will become uncompetitive if it lives up to certain standards, insofar as other companies cannot undercut it by taking shortcuts or externalising costs. The other side of this point is that widespread irresponsibility and frequent misconduct can trap all parties. Fouls and gamesmanship can hardly be addressed where low standards are the rule and infringements are the order of the day. To shame a whole business sector requires strong and concerted outrage, across publics and media and political fora. This is bound to be hard to sustain when many other problems compete for attention, as our news cycles testify daily. By contrast, when most corporations in a market sector live up to reasonable standards, efforts to hold miscreants responsible can be more focussed and are more likely to prove effective. Debates about corporate social responsibility often demonstrate the difficulty of judging whether abiding by certain standards would really be in the interests of a company, or indeed of defining those interests when short-and long-term incentives conflict. The shape of those calculations shifts considerably, though, where reasonable regulatory standards and widespread compliance are the norm-so that spaces between short and long term, between conduct and reputation, do not open into outright gulfs.
As well as enabling more responsible conduct across the field, widely shared standards open up another important dimension of freedom: the ability to define and shape one's responsibilities. As Mayer (2013) argues, there is and should be scope for more responsible corporations to define their own roles and responsibilities, in ways that go beyond regulatory standards. Apart from a business context that ensures such efforts will not be penalised, the other precondition for this is that such agents really must be more responsible: with abilities to make commitments, to define these in socially acceptable ways, and to consistently live up to them. Again, we can compare this point with the artificial perplexity about capability and culpability. It is indeed true that more capable agents will judge themselves against higher standards. But I would argue that this is simply part of what it is to be more able. Only where an agent has grasped the point and value of a responsibility will it be fully capable of instituting the measures needed to live up to it and meaningfully evaluating its own performance. (Compare the tick-box approach that a merely adequate company might take to non-discrimination law or fair trade standards.) It is also part of what it is to enjoy a good reputation. When we hear of malfeasance by some corporations, we may feel dismay but not surprise. For those that have built up a reputation for reliability and responsibility, similar infractions may attract more outrage. From one point of view, this may seem unfair (why should similar wrongs attract different responses?). From another, however, it just marks the standing of a trustworthy organisation-that it will act to sustain its reputation; that closer cooperation poses fewer risks; that stakeholders of every stripe can count on it ('expect better', as we sometimes say); that it is less risky or costly to criticise it; and, above all, that it will respond quickly and appropriately if it gets things wrong.
In other words, regulatory standards provide an important background against which more responsible corporations can define their own roles and gain the benefits-at all sorts of levels-of more active cooperation that come with that. Doubtless, regulation will pose some frustrations even for these actors-who do not need to be told about and checked for basic standards, for whom the good (or the acceptable minimum) may sometimes be an enemy of the best. Nonetheless, to the extent that regulation can hinder their undercutting by less responsible actors, to the extent that they can endorse its spirit even if not its every letter, it still counts as a genuine enablement for them. Equally, they can play a constructive role in debates about regulatory standards, and indeed self-regulation, in a way that less responsible actors cannot. 31 Last-but perhaps most important of all-my argument suggests that regulations governing specific forms of business activity must be set alongside provisions that shape the capacities and responsiveness of business corporations, simply as such. The extent to which companies are able to offer more than the most grudging compliance with regulations, like the extent to which they are able to meaningfully judge or debate their responsibilities, depends on how far they are trustworthy organisations. 32 A whole range of different measures are relevant here, and I mention just a few 33 : legal provisions for incorporation can foster more responsive governance structures (compare the oft-cited example of employee representation within German corporations); various measures can discourage executive autocracy (for instance, by enabling staff to protest policies and initiatives that they believe are ill-advised, illegal, or immoral (Bovens 1998)); there are many ways to promote self-knowledge and counteract misinformation, including reporting and audit requirements and trustworthy organisations to monitor these (so that different parts of the organisation are transparent to one another, and executives are aware of the actions and policies of various departments or subsidiaries 34 ); not least, measures to prevent market concentration and cartelisation ensure that both suppliers and customers have fair choice, so that companies are less able to dictate terms. 35 Such provisions may look like restrictions and have sometimes been opposed in the name of market freedom. 36 Indeed, it is no part of my argument to deny that they have restrictive aspects. Even where they do not specifically constitute a new form of activity, however, they also represent enablements: 34 As Braithwaite remarks, employers often 'have good strategies for protecting themselves from day-to-day knowledge of their organization's failures to meet its legal obligations. The primary function of restorative inspection is therefore simply to squarely draw the attention of chief executives to these failures... Deterrence is only needed in cases where this restorative approach repeatedly fails' (2015, p. 165). 35 One might also mention measures to ensure that directors do not face excessive pressure from single constituencies. 'Accountability' sounds like a good thing, but if just one set of stakeholders can insist on its claims, then these claims can easily become unreasonable, jeopardise other goals, and undermine capacities for responsibility. Restrictions on trade union protests were once justified in this way. Of more contemporary relevance is the doctrine of shareholder value, which has often led to short-termism and undermined the corporation as a long-term contributor to prosperity (Stout 2012). 36 With regard to the language of free markets, it is worth noting that corporate markets are not free in any straightforward sense. As organisations and legal actors, business corporations are constituted by their charters and the legal provisions that allow them to come into being; these impose constraints on their members as well as on everyone who may be affected by corporate activities. Like other rules and regulations mentioned here, these restrictions also enable: they constitute a new form of business enterprise and facilitate its activities, creating freedoms, opportunities, and powers that go much beyond the classical liberal rights to property and contract. See further North et al. (2009) andCiepley (2013). 31 Lane (2005) nicely stresses the importance of corporations engaging in wider debate about the regulations and norms by which they should be guided-as well as how difficult this task may be for them. 32 As Pettit has commented, we would be-and indeed have beenunwise to permit such entities to come into being: '[I]t would seem to be a serious design fault, at least from the viewpoint of society as a whole, to allow any group agents to avoid making evaluations [about the relative value of the options they face]. Why should any group of individuals be allowed to incorporate under a constitution that deprives it of the ability to evaluate its options and that ensures, therefore, that it will not be fit to be held responsible for what it chooses? ' (2007, p. 187). I have suggested that Pettit's final 'therefore' should be read as a matter of practical implications, rather than a strictly conceptual entailment.
Footnote 32 (continued) 33 For a careful treatment of this vital topic, see Bovens (1998, espe-cially the chapters in Part III). they create or foster abilities that we have no right to count on in their absence.

Conclusion
With our more practical hats on, such as we must wear when we confront corporations and their formidable powers, I have suggested that some important aspects of responsibility come to the fore-aspects that philosophers tend to lose sight of when they focus on the capacities and actions of the wrongdoer, or culpability as an abstract quality. Disagreements about the role and importance of regulation remind us of the profoundly social nature of moral and practical demands, and the many forms of action and cooperation involved in practices of holding responsible. To understand responsibility, we should not focus only on the question of what a particular agent can do, be it an individual or collective actor, or on its culpability, considered in the abstract. We also need to consider what a plurality of agents can achieve together. Regulatory standards, like other ways of defining and upholding norms and responsibilities, represent a crucial way in which agents can act together to enable cooperation, responsibility, and initiative.
We know that dealing with many of our corporations is incredibly hard work: they are, one might say, rather like overgrown children whose capacities for responsibility still need to be developed-capacities for self-control, self-knowledge, and long-term planning; abilities to handle many responsibilities at once, to appreciate the demands of their role amid a complex division of labour, to actively reshape that role with initiative and commitment. This irresponsibility has only been encouraged by the framing of regulation as red tape-restrictions and hindrances that need to be cut through, removed, or avoided in the name of efficiency or enterprise. (There may be many places where we could do with better regulation, but 'better' is no synonym for 'less'.) The problem we face is not the idle theoretical one of whether some corporations are 'really' responsible for their irresponsible conduct. 37 The problem, rather, is just their irresponsibility-that is, the relative inability of some corporations to act well and the corollary of this inability, their tendency to act badly. I have not dwelt, here, on the most obvious difficulties that this causes, such as damage to public health, communities, or environments. But I have emphasised a more timeless problem, one that seems to me unduly neglected in the philosophy of responsibility: the task of upholding and adjudicating a reasonable framework of cooperation. Above all, I have sought to emphasise the enabling power of such a framework, including the enabling role of regulation in the corporate context.
As a final note, let me summarise how the different senses of responsibility come together in this argument. For more responsible (virtue) agents, some 'restrictions' merge into the background and their enabling character comes to the fore. They facilitate cooperation and trust; in corporate terms, they provide or strengthen 'social licenses to operate' (Demuijnck and Fasterling 2016). As part of the responsibilities (prospective) that an agent endorses, they become second nature. Since the agent takes responsibility (prospective) for them, there is less need to hold the agent responsible (retrospective). If the agent does go wrong, we may register greater disappointment. At the same time, holding responsible will involve less effort and costs, since the agent is better able to take responsibility (retrospective). By contrast, less responsible (vice!) agents are less able or willing to live up to their responsibilities; they will be more likely to take a partial or unreasonable view of their duties; they will chafe under reasonable regulations and other norms; they will try to deny legitimate criticism and evade adverse responses. Other agents must bear the ongoing responsibility (prospective) of reminding them of their responsibilities and holding them responsible. Although these attempts are bound to feel restrictive and hard-going for all concerned, the price of not holding these agents responsible is to give up on the relevant standards and to sacrifice the trust and cooperation that they might have fostered. Practices of responsibility thus play a vital enabling role. They may foster abilities to take responsibility, that is, to acknowledge past wrongs and do better in future-for example, as a company learns from reputational damage and makes the organisational changes needed to prevent future infractions. They alter the structure of incentives, discouraging further misconduct and reassuring other actors who might otherwise have been tempted to go wrong. At the very least, they help to keep responsibilities and norms from being lost to view. mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.