Of the many developments in business ethics that Thomas Donaldson has helped pioneer, one is the application of social contract theory to address questions about the responsibilities of business actors (Donaldson Reference Donaldson1982; Donaldson and Dunfee Reference Donaldson and Dunfee1995, Reference Donaldson and Dunfee1999). This approach has attracted a great deal of attention among scholars (Wempe Reference Wempe and Smith2009a).Footnote 1 Although much of the scholarship centers on Donaldson’s work with Thomas Dunfee on Integrative Social Contracts Theory (ISCT) as an account of global business ethics, the approach originates with Donaldson’s account of the purpose and responsibilities of for-profit corporations in Corporations and Morality (1982).Footnote 2 In Corporations and Morality, Donaldson develops one of the most sustained and comprehensive accounts about the purpose and responsibilities of for-profit corporations. In this paper, I argue we would do well to continue engaging with Donaldson’s account, but that many of the insights to be gained come from recasting the role played by social contract theory.
Central to Donaldson’s account are two aims: 1) to justify the existence of for-profit corporations and 2) to specify and ground their responsibilities. To justify the existence of for-profit corporations, Donaldson argues it is not enough to point to a right on the part of individuals to incorporate. “Even if there were a right to incorporate,” he writes, “it would fail to justify corporate existence in the sense of showing why corporations ought to exist” (1982, 39). Nor is it sufficient justification to point to corporate productivity. As Donaldson writes, “We must consider not just its capacity to produce wealth, but rather the full range of its effects upon society” (1982, 38). Instead, Donaldson aims to provide a full account of the values realized by corporations along with any negative impact they may have such that their existence and operation would be morally desirable. In trying to specify and ground the responsibilities that attach to corporations, Donaldson’s focus is on “indirect obligations” (1982, 33-34). These are obligations not formally specified through existing agreements and institutions. That is, they are obligations that corporations ought to respect even if not institutionally required.Footnote 3
To justify the existence of for-profit corporations and to specify and ground their responsibilities, Donaldson takes the for-profit corporation as a member of the more general class of productive organizations, and draws upon the social contract tradition from political theory to construct a hypothetical social contract between members of society and productive organizations. The thought is that just as the idea of the social contract has been used to justify the existence of government and to specify limits to its power, so too can the idea of the social contract be used to justify the existence of for-profit corporations and to specify their responsibilities. As Donaldson makes clear, the version of the social contract tradition he follows is not that of a contract among members of society to justify their obedience to the state along the lines of Thomas Hobbes and Jean-Jacques Rousseau. Rather, the version of the social contract he has in mind is along the lines of that between the citizens and the state. Specifically, Donaldson has in mind a contract between individual members of society (not some supra-individual entity) and productive organizations (1982, 42).Footnote 4
To date, most of the critical discussion has centered on the question of whether Donaldson’s account succeeds in justifying the existence of for-profit business corporations and the responsibilities associated with them. Paul Hodapp (Reference Hodapp1990), John Kultgen (Reference Kultgen1986), and Gordon Sollars (Reference Sollars2002), for example, argue that Donaldson’s application of social contract theory largely fails in its aims. Others argue that the general approach holds promise (Conry Reference Conry1995, Levitt Reference Levitt1986), and that with suitable modification, such as explicit recognition of the role of the state and background institutions (Brock Reference Brock1998), the approach provides justification for the moral defensibility of corporations.
In order to further our understanding about the purpose and responsibilities of productive organizations, and as a contribution to the scholarship on Donaldson’s thought, this paper gathers together the critical responses to Donaldson’s account along with Donaldson’s replies to his critics (1986, 1990). The first section of this paper is a brief summary of the steps involved in constructing the hypothetical social contract between society and productive organizations, and the resulting rights and responsibilities specified by the contract. In the second section, I examine objections that the hypothetical contract itself gives no reason to members of society to permit the existence of productive organizations and that it fails to ground any obligations on their part. In the third section, I examine the extent to which proposed amendments to Donaldson’s account help address these objections. In the fourth section of the paper, I explore making the hypothetical contract central to Donaldson’s account by looking to T. M. Scanlon’s contractualist account of morality. In the fifth section, I conclude by taking stock of the role of social contract theory in realizing the two aims of 1) justifying the existence of productive organizations and 2) specifying and grounding their responsibilities.
Three broad claims emerge from this discussion. First, there is reason to doubt that the device of the social contract justifies the existence of productive organizations or plays a role in specifying and grounding their responsibilities. Second, if the idea of the social contract does play a role in Donaldson’s account, the most plausible role it plays is as a heuristic or test for the claim that what justifies the existence of productive organizations are considerations of social welfare. Third, despite the case for this limited role, the social contract in Donaldson’s account embodies a distinctive conception of corporate purpose and responsibility and an approach to theorizing about them that remains underexplored and is worth developing.
1. CONSTRUCTING THE SOCIAL CONTRACT
On Donaldson’s account, constructing the hypothetical social contract between productive organizations and society involves three steps (1982, 44). The first is to image a society without productive organizations—i.e., “a state of individual production” in which individuals “produce and work alone” (1982, 45). The second step is to specify what problems would be remedied by productive organizations such that society would allow their existence. By society, Donaldson has in mind individual members of society, rather than taking society as a single entity. He focuses on the benefits to two groups: consumers and employees. From the perspective of consumers, productive organizations would “enhance the satisfaction of economic interests” on their part, for example, by improving productive efficiency and stabilizing production and distribution (45). Productive organizations also would benefit employees, for instance by increasing their income and capacity for social contribution (1982, 48). The third step is to use the “reasons generated in the second step as a basis for specifying a social contract between society and its productive organizations” (1982, 44).
The resulting contract specifies rights and responsibilities of productive organizations. In addition to the right to exist, productive organizations are assigned “recognition as a single agent, especially in the eyes of the law,” as well as authority “to use land and natural resources” and “to hire employees” (1982, 43). These rights are needed for productive organizations to benefit members of society. At the same time, Donaldson argues that members of society would assign responsibilities to productive organizations to minimize the drawbacks associated with their introduction. From the standpoint of consumers, productive organizations should minimize 1) “pollution and the depletion of natural resources,” 2) the lack of personal accountability on the part of employees to consumers, and 3) the “misuse of political power” for personal gain or in ways that harm the general public (1982, 51). From the standpoint of employees, productive organizations should minimize 1) the alienation that may arise from greater separation between workers and their final product, 2) the “lack of worker control over work conditions,” and 3) the “monotony and dehumanization” frequently associated with highly specialized and automated work (1982, 52). The contract also specifies the balance to be struck across various consumer and employee interests when trade-offs need to be made, for example, as in the case of the trade-off between higher salaries and lower consumer prices. In addition, Donaldson argues that members of society would enact the contract “on the condition that it did not violate certain minimum standards of justice.” Recognizing the debate surrounding standards of justice, Donaldson holds that, at a minimum, they imply that “productive organizations avoid deception or fraud, that they show respect for their workers as human beings, and that they avoid any practice that systematically worsens the situation of a given group in society” (1982, 53).
Donaldson presents this account as an alternative to the shareholder primacy view famously associated with Milton Friedman—that is, the view that the responsibility of corporate managers is to maximize share price or shareholder wealth subject to the constraints of market mechanisms and the law.Footnote 5 According to Donaldson, his account places constraints on the sorts of contracts that individual economic actors may enter (1982, 55). For example, he writes, “the social contract’s requirement that productive organizations serve consumer interests would outweigh the rights of a stockholder and manager to agree to market an inherently dangerous product” (1982, 55).Footnote 6 More generally, underlying the social contract account is the thought that allowing for productive organizations is to the benefit of all members of society and not just the organizations’ shareholders.Footnote 7
Before continuing, one point of clarification needs to be made about the use of the term, “social contract.” In Donaldson’s account, the social contract is between productive organizations and members of society. It takes the form, “We (the members of society) agree to do X, and you (the productive organizations) agree to do Y” (Donaldson Reference Donaldson1982, 42). In many applications of social contract theory, however, the parties to the contract are individual members of society and the object of their agreement is the institutional arrangement under which they are to live. The choice for the parties is often framed as between living without those institutions (i.e., “the state of nature”) and living under them. In using the idea of the “state of individual production,” and asking whether members of society would find it rational to allow for productive organizations, Donaldson draws on this other understanding of the social contract—that is, as a hypothetical agreement among members of society to a certain set of institutional arrangements. Much of the critical discussion of Donaldson’s account concerns his use of this aspect of social contract theory. Accordingly, in what follows, use of the term, “social contract,” is often in reference to the more general methods of social contract theory, and not only the terms of agreement between productive organizations and members of society.
2. THE SOCIAL CONTRACT AS A HUERISTIC DEVICE
In this section, I examine objections that the social contract method, as employed by Donaldson, fails 1) to justify the existence of productive organizations and 2) to specify and ground responsibilities on their part. Central to these objections is the claim that the social contract method—which involves the hypothetical act of parties agreeing to the terms governing the formation and conduct of productive organizations—does not aid in specifying or justifying those terms.Footnote 8 I also examine Donaldson’s responses to these objections. What emerges in Donaldson’s responses is the view that the social contract is not meant to play so much a justificatory role as a heuristic role, a role frequently attributed to the social contract in political theory. As part of this analysis, I distinguish various ways in which the social contract can serve as a heuristic device, and conclude that the most plausible role it plays is as a heuristic for Donaldson’s claim that what justifies the existence of productive organizations are considerations of social welfare.Footnote 9
2.1. Justifying the Existence of Productive Organizations
In discussing Donaldson’s account, Paul Hodapp objects that, “Donaldson’s social contract theory as a methodology is circular, presupposing the information which it is supposed to generate” (1990, 128). The problem, according to Hodapp, is that “we do not know how to imagine a state of nature without productive organizations without already knowing the purposes of such an organization” (1990, 128). Imagine someone who views productive organizations as having a purpose other than one specified by Donaldson or no purpose at all. According to Hodapp, there is nothing in the idea of the hypothetical social contract that would give that person reason to arrive at the conclusions advanced by Donaldson. Rather than a social contract theory, the way in which to read Donaldson’s argument is as a functionalist theory. That is, it is an account of the benefits that members of society have reason to receive by introducing productive organizations, and in specifying these benefits, there is no need to introduce the idea of a social contract.
In response to Hodapp, Donaldson distinguishes between the purpose of a specific productive organization—“to produce tractors, cigarettes, advice, nuclear bombs, space shuttles, accounting services, knowledge, children’s toys and super computers” (1990, 134-135)—and the underlying reason that members of society may have to allow for the existence of productive organizations in general. Donaldson argues that the purpose of specific organizations need not be known in order to arrive at the outcomes of the social contract. However, he does acknowledge there is a sense in which the underlying reason or “why” must be “known in advance before we begin moral analysis by way of the social contract” (1990, 134). Donaldson writes, “the idea of a social contract is not a set of normative prescriptions; it is, rather, a device or method for clarifying normative concepts, one that embodies surprising plasiticity of purpose” (1990, 134). The idea of the social contract helps to clarify the purpose of productive organizations because “rational members of society will legitimize such organizations only insofar as they stand to advance society’s aggregate interests, and hence from society’s perspective the advancement of social welfare is the ‘purpose’ of the productive organization” (1990, 135).
By clarifying the purpose of productive organizations in this manner, the device of the social contract in Donaldson’s account plays a heuristic role that is similar to the role attributed to it in other contemporary (as opposed to traditional) applications of social contract theory. For traditional social contract theorists—i.e., writers such as Thomas Hobbes, John Locke, and Jean Jacques Rousseau—the central concern is normally taken to be the problem of political obligation—that is, the justification of a duty to obey the state and its laws. The social contract enters as a way to generate such an obligation from the act of consent, on the part of members of society, to the terms of the contract. For contemporary social contract theory, the concern no longer is taken to be the problem of political obligation, but rather the justification of political and social institutions. On this approach, consent is not the salient feature of the parties’ agreement to the social contract, and agreement need not entail an obligation to obey. Instead, the parties’ endorsement of a set of political and social institutions is “a ‘test’ or a heuristic” of the reasons given for having that set of institutions (D’Agostino et al. Reference D’Agostino, Gaus, Thrasher and Zalta2014). The thought is that by asking whether parties would agree to a set of political and social institutions, we can test whether those institutions can be justified with reference to the sorts of reasons attributable to the parties.
Understood in this manner, Donaldson’s use of the social contract method need not be “circular” as Hodapp objects. In its role as a heuristic, the idea of a social contract provides a test for the reasons that members of society might give for endorsing a set of institutions—in this case, the rules governing the formation of productive organizations. On this reading of Donaldson’s account, one reason under consideration is that productive organizations serve “to advance society’s aggregate interests” (1990, 135). Insofar as we can imagine the parties in Donaldson’s account agreeing to rules permitting the formation of productive organizations for this reason, then, the role of the social contract in Donaldson’s account is to serve as something of a heuristic or test of the claim that advancing social welfare is the rationale for existence of productive organizations. On this interpretation of the role of social contract theory, stipulating the purpose of productive organizations in advance is not a problem for Donaldson’s account. The device of the social contract is to test whether the stipulated purpose can plausibly be attributed to productive organizations.
2.2. Specifying and Grounding the Responsibilities of Productive Organizations
I now turn to consider the role of the social contract in the second objective of Donaldson’s account, which is to specify and ground the responsibilities attributed to productive organizations. To do so, it will help to consider two objections raised by John Kultgen (Reference Kultgen1986) against Donaldson’s account and Donaldson’s responses to them (1986).
One objection Kultgen raises is that “an imaginary agreement is not an actual one and qua agreement binds no one, however, it may disclose obligations based on some other ground” (Kultgen Reference Kultgen1986, 31). Donaldson interprets Kultgen’s objection to be that “the social contract may serve as a clue in discovering such rights and obligations, but can boast itself of no genuine existential status,” and as such cannot ground the responsibilities assigned to it (Donaldson Reference Donaldson1986, 42). In response, Donaldson asks about the existential status of rights and obligations in general. Suppose, he asks, there were no law to protect rights of freedom and tolerance. He continues, “would it be an objection to say that since it does not exist it can be nothing but a ‘merely heuristic device,’ of use in tracking down what the real rights or real obligations … are?” For Donaldson, this would be “nonsense” because “fundamental rights, obligations, duties, … and a host of other fundamentally prescriptive concepts are not descriptive of existing states of affairs at all” (Donaldson Reference Donaldson1986, 42-43). Donaldson’s response to Kultgen is that the social contract is no less real than other normative concepts, and as such, it need not be merely a heuristic device for identifying the real rights or responsibilities.
If the social contract is to serve as more than a heuristic device, what role might it play? A natural candidate, in the light of Donaldson’s aims for his account, is to help generate responsibilities on the part of productive organizations toward workers and consumers. The difficulty, however, is that Kultgen’s criticism seems to be independent of the existential status of the social contract. That is, even if the social contract is no less real than other normative concepts, such as rights and obligations, the point is that because consent to the contract is hypothetical, it cannot generate these obligations. This is a point frequently raised about the social contract tradition in political theory. Hypothetical consent, as it is pointed out, is to be distinguished from actual consent. Whereas the latter, under the right conditions, is grounds for an obligation, the former is not, at least in the way normally associated with consent (Holton Reference Holton1992). The emphasis in contemporary applications of social contract theory on justifying institutional arrangements rather than on the problem of political obligation is consistent with this view. Accordingly, there is reason to doubt that the idea of the social contract plays a role in generating the responsibilities assigned to productive organizations in Donaldson’s account.
A second objection raised by Kultgen would appear to deny that the social contract plays even a heuristic role with respect to identifying these responsibilities. To be clear, the way in which the social contract serves as a heuristic device in this case is somewhat different than the role attributed to it in justifying the existence of productive organizations. In section 2.1, the social contract was said to serve as a heuristic or test for evaluating the rationale stipulated for the existence of productive organizations. The device of the social contract—that is, the agreement to move from a state of individual production to one that involves productive organizations—is a way to represent the reasoning of the parties. In this case, the sense in which the social contract is meant to be a heuristic device seems more general. As described by Donaldson in his reply to Kultgen, “the point of the method is to engage persons in a thought-experiment, one which will utilize their powers of reason and moral intuitions in a manner calculated to achieve moral truth” (1986, 45). If the social contract is to be understood as a heuristic device, its role, it seems, is to help identify the responsibilities of productive organizations.
Kultgen’s second objection is that the device of the social contract fails to play such a heuristic role. Kultgen writes, “corporations are artifacts, so they have only the nature that is bestowed on them by their designers. … The question is precisely what do the obligations of these natural person become when they assume their roles in the corporation. … Their role responsibilities are precisely those assigned by those who have designed the corporation and now support it, including the public insofar as its representative have chartered it and its members deal with it” (Kultgen Reference Kultgen1986, 31). Kultgen concludes there are no responsibilities that attach to corporations or their managers apart from those that are already specified by those who designed them in the first place.Footnote 10
In response to Kultgen’s second objection, Donaldson writes, “guns, hydroelectric dams, prisons, monarchies, nuclear weapons, neighborhood associations—at least those that are human organizations—have specific role responsibilities as part of their design. But it would be ludicrous to say that those organizations have no general obligations deriving from the kind of organization they are; no obligations, in other words, apart from the role responsibilities designers happened to institute. The leaders of the Nazi government had responsibilities qua leaders of government, regardless of the role responsibilities designed into the Nazi regime” (1986, 44-45). Donaldson’s point seems to be that the role responsibilities that are in fact attributed to productive organizations are not exhaustive of the moral responsibilities that attach to the organizations or their managers.
All of this seems correct. The question that arises is whether the idea of the social contract is helpful in identifying these additional moral responsibilities. Consider the example of the leaders of the Nazi government. The point seems to be that no matter what their role responsibilities, there are general duties about what we owe to others that apply to them and that make their actions wrong. If so, these duties are not the sort of duties for which a social contract is needed as a heuristic device to help identify them. Rather, they are the sorts of duties that attach to natural persons and are sufficiently general to stand independently of any social contract. Otherwise, it is difficult to imagine that they could be identified independently of a given set of role responsibilities. The question then is, what sorts of moral responsibilities are independent of role responsibilities but are insufficiently universal such that they cannot be identified without the aid of the device of the social contract.
One response is that the responsibilities relating to workers in Donaldson’s account are examples of such responsibilities. With regard to workers, recall that the responsibilities of productive organizations are to minimize 1) the alienation that may arise from greater separation between workers and their final product, 2) the “lack of worker control over work conditions,” and 3) the “monotony and dehumanization” frequently associated with highly specialized and automated work (1982, 52). These responsibilities, it may be said, are distinct from other responsibilities Donaldson attributes to productive organizations—e.g., minimizing pollution or not misusing political power. In the case of minimizing pollution or not misusing political power, for example, there is reason to hold the corresponding responsibilities apply just as much to economic actors in the state of individual production. Accordingly, there is no need for the social contract to serve as a heuristic device in identifying them. However, it may be said that responsibilities regarding workers are different, because the conditions that make them relevant arise precisely because workers leave the state of individual production. That is, they are responsibilities that are distinctive to a society with productive organizations. In the case of these responsibilities, the social contract serves a heuristic role according to this response.
Other parts of Donaldson’s account, however, call into question the extent to which such a response establishes a heuristic role for the social contract in identifying the responsibilities of productive organizations toward workers. In describing the responsibilities enumerated in the social contract, Donaldson raises the possibility that “the inhabitants might believe that, on balance, people as workers stand to lose from the introduction of productive organizations, and that potential alienation, loss of control, and other drawbacks make the overall conditions of the worker worse than before.” He continues, “But if the benefits to people as consumers fully overshadowed those drawbacks, we should still expect the contract to be enacted” (1982, 53).Footnote 11 Later in Corporations and Morality, Donaldson details an extensive set of employee rights (1982, 129-157). In defending these rights, however, Donaldson writes that, “unfortunately, the language of the social contract is too imprecise (at least as formulated) to bring the issue of employee rights into relief (1982, 137). Instead, he turns to other bases for grounding such rights, including perfect duties, basic needs and interests, the right to equal freedom, and the right to behave responsibly (1982, 139). All of this gives us reason to doubt the extent to which the social contract serves as a heuristic device in identifying the responsibilities of productive organizations.
At this point, I turn to consider the possibility that the social contract serves as a heuristic device in the sense attributed to it in section 2.1—that is, as a test or heuristic of the reasons given for justifying a specific institutional arrangement. On this interpretation of the role of the social contract device, the idea would be to take the responsibilities assigned to productive organizations as part of the rules governing the formation and operation of productive organizations and ask whether the parties would agree to these rules on the basis of the reasons given for them. As in the case of testing the reasons given for the existence of productive organizations, the idea here would be to test the reasons given for the specific responsibilities assigned to productive organizations.
In specifying the responsibilities of productive organizations with respect to customers and workers, Donaldson asks, “If people in the state of individual production must agree upon the terms of the social contract, and if these terms directly relate to the task of enhancing society’s welfare, then why stop with maximizing prima facie benefits? Why not also minimize prima facie drawbacks?” (1982, 49). As in the case of justifying the existence of productive organizations, the reason stipulated for the responsibilities of productive organizations toward customers and workers is the consideration of social welfare. Consider, for example, the obligation not to misuse political power. “Such power,” Donaldson writes, “sometimes is used to secure favors from government which damage both consumer interest and the interests of the general public” (1982, 50). Another example is the obligation to minimize worker alienation on grounds that workers in productive organizations “are typically removed from the product … in a way that may block their very capacity for self-expression” (1982, 51). In the light of these drawbacks, Donaldson writes, “in order maximally to enhance welfare, productive organizations should both pursue positive goals and minimize negative ones” (1982, 51). The role of the social contract as a heuristic device is to test whether maximizing social welfare serves as the reason or justification for assigning these sorts of responsibilities to productive organizations.
The question that arises is whether the agreement of parties provides support for the maximization of social welfare as the reason to justify inclusion of these obligations among the rules governing the formation and conduct of productive organizations. At one level, given the way in which Donaldson models the reasoning of the parties, their agreement supports taking social welfare as a reason for assigning these obligations to productive organizations. The parties, in Donaldson’s account, are imagined to have “‘economic interests,’ i.e., as people for whom it is desirable to have some things or services produced by human labor” and the sorts of benefits they derive from permitting the existence of productive organizations are basically economic in nature: e.g., products of greater quality at lower price, more stable output, or greater income potential (Donaldson Reference Donaldson1982, 45-49). In turn, the drawbacks associated with introducing productive organizations can be characterized as drawbacks in that they are welfare-reducing.
At the same time, there is reason to doubt that the agreement of the parties counts in favor of social welfare as the rationale for assigning responsibilities to productive organizations, at least in the same way that social welfare is the rationale for the existence of productive organizations. While it is the case that the drawbacks from introducing productive organizations can be framed in terms of their negative impact on the economic interests of members of society, recall that Donaldson models the parties as also deciding for reasons of justice. He writes that, “at a minimum, … the application of the concept of justice to productive organizations appears to imply that productive organizations avoid deception or fraud, that they show respect for their workers as human beings, and that they avoid any practice that systematically worsens the situation of a given group in society” (1982, 53).Footnote 12 Given that the parties also introduce obligations for productive organizations on grounds of justice, the question may be asked why the obligations grounded in considerations of social welfare are not better understood as grounded in other considerations, such as justice. Consider again the obligation not to abuse political power. Although the abuse of political power can reduce welfare and harm consumer interests, one may ask why considerations of welfare count against the abuse of political power rather than considerations of justice, or even more directly, the wrong in abusing political power. Similarly, consider the obligation to minimize worker alienation. If the contract requires that productive organizations respect workers as human beings as a matter of justice, and the obligation against worker alienation is to avoid workers being blocked “in their very capacity for self-expression,” it seems reasonable to ask why the agreement of the parties to such an obligation is not grounded in considerations of justice rather than welfare.
At this point, it may be objected that considerations of social welfare also need not be taken as the reason for the existence of productive organizations. For example, it may be objected that the benefits of greater income associated with the existence of productive organizations need not be understood as a consideration of social welfare, but rather as a consideration about personal autonomy or the ability to fulfill one’s life plans. In one sense, this is correct. The rationale for greater income may be understood in terms of autonomy or the ability to fulfill one’s life plans. At the same time, there is reason to hold that the existence of productive organizations can be justified with respect to considerations of social welfare without such considerations also serving as the rationale for the responsibilities attributed to productive organizations. In Donaldson’s account, the parties are modeled as acting on economic interests in the state of individual production. The benefits associated with productive organizations count as benefits for the same reason as engaging in individual production. In contrast, although the responsibilities attributed to productive organizations are meant to counteract the drawbacks that arise from their existence, what makes them drawbacks need not be the same as the considerations for introducing productive organizations. Put another way, just because these drawbacks arise from the introduction of productive organizations, they need not count as drawbacks in terms of the reasons for introducing them.
This is not to deny there may be good reasons to reject framing the benefit of productive organizations in terms of social welfare.Footnote 13 Rather, the point is that we can accept Donaldson’s justification for the existence of productive organizations in terms of social welfare without accepting that social welfare is the rationale for assigning responsibilities to productive organizations. In turn, we can reject the idea that social contract generates the responsibilities of productive organizations or serves as a heuristic device in specifying those responsibilities or their grounds, without having to reject the more limited claim that the social contract serves as a heuristic device in understanding the rationale for productive organizations.
3. AMENDING THE ACCOUNT
In this section, I examine the extent to which proposed amendments to Donaldson’s account give us reason to reconsider the argument in the preceding section. The starting point for this examination is an objection to Donaldson’s account raised by Edward Conry (Reference Conry1995). Conry writes, “As Donaldson adapts social contracts to business, he moves away from a somewhat non-determinative human nature to an even more speculative and ephemeral interaction between the ‘state of pre-organization nature,’ and the benefits and detriments of organized life. Because this is even more uncertain input, the output is therefore characterized by greater uncertainty” (1995, 201). Conry argues that the social contract method could just as easily ground something like shareholder primacy as it does Donaldson’s account of corporate purpose and responsibility (1995, 202).Footnote 14 The problem, according to Conry, is that “the ‘pre-organization world’ lacks the illuminating power which characterizes Hobbesian human nature. It is at least arguable that Donaldson’s conclusions do not follow from the analysis of the pre-productive organization state of nature; they follow instead from his correct intuitions” (1995, 202).Footnote 15 Let us call Conry’s objection, “the problem of indeterminacy.”
Conry’s objection suggests a line of response to the argument put forward in the previous section. Conry’s objection is that Donaldson’s description of the state of individual production is sufficiently general to support a wide range of conceptions about the purpose and responsibilities of productive organizations. When considered in relation to the argument of the previous section, Conry’s objection suggests that the central challenge for Donaldson’s account is not in invoking the social contract, but rather with its description of the state of individual production. That is to say, if the description of the state of individual production were suitably amended, then the social contract might serve to generate the responsibilities of productive organizations or as a heuristic device in specifying those responsibilities or their grounds. Accordingly, in this section I examine the plausibility of accounts that aim to amend Donaldson’s description of the state of individual production. These accounts rely upon the work of John Rawls (Reference Rawls1971, Reference Rawls1999, Reference Rawls and Kelly2001).
3.1. Invoking Rawls
Gillian Brock (Reference Brock1998) makes a similar observation to Conry and offers an amendment to address the problem of indeterminacy. According to Brock, whether or not members of society find it rational to allow for productive organizations requires further specification about the resulting state of affairs and the background institutions in place. For example, in the case of a minimal “night watchman” state, Brock writes, “we all might want powerful productive organizations to act as welfare providers of some kind” (1998, 708). A further challenge Brock raises is that Donaldson assumes a shared sense of fairness for how to resolve trade-offs among competing interests (1998, 709). As a way to resolve these challenges, Brock argues for adopting an approach that follows the work of John Rawls. Adapting Rawls’s idea of the “veil of ignorance,” Brock asks what rational, self-interested parties would agree to from behind a veil of ignorance regarding their place under the new economic arrangements. The advantage of the Rawlsian approach, according to Brock, is that it allows us to imagine better what individuals would agree to under limited information and it makes explicit the choice of background institutions. Brock argues for a three-way conditional contract according to which members of society agree to allow productive organizations under the assumption that the state provide various protections and guarantees, including protecting civil rights, providing unemployment benefits, and access to educational benefits (1998, 716).
A second contractarian alternative to Donaldson’s account is put forward by John Bishop (Reference Bishop2008). Like Brock, Bishop aims to bring his account in line with John Rawls’s approach. Bishop does so by specifying that the social contract to be imagined is not between society and productive organizations, but rather among members of society with respect to the rules governing the formation and conduct of corporations (2008, 195-196). These rules, on Bishop’s account, are to be determined in a manner consistent with the principles of justice from behind the veil of ignorance. Consider for example, the first principle of justice, which states that, “each person is to have an equal right to the most extensive total system of equal basic liberties compatible with a similar system of liberty for all” (Bishop Reference Bishop2008, 195, quoting Rawls Reference Rawls1971).Footnote 16 Bishop argues that because allowing for the creation of corporations represents an expansion of liberty, recognizing such a freedom would be consistent with the first principle so long as corporate actions did not interfere with the liberty of others (2008, 199). Proceeding in this manner with respect to other dimensions of justice, Bishop concludes that, “for-profit corporations can, and indeed should, exist in just societies” (2008, 205).
By appealing to Rawls’s account of justice, a difficulty arises for Brock and Bishop in their accounts. The problem roughly is this. On Rawls’s account, the principles of justice, which are to be specified from behind the veil of ignorance, apply only to the basic structure. Rawls defines the basic structure as “the way in which the main political and social institutions of a society fit together into one system of social cooperation, and the way they assign basic rights and duties and regulate the division of advantages that arise from social cooperation over time.” He continues, “the political constitution within an independent judiciary, the legally recognized forms of property, and the structure of the economy (for example, as a system of competitive markets with private property in the means of production), as well as the family in some form, all belong to the basic structure” (Rawls Reference Rawls and Kelly2001, 10). Productive organizations, and other associations, such as religious organizations, are not part of the basic structure. Instead, the rules governing productive organizations are to be determined at the legislative stage, which is subsequent to the process of specifying the principles of justice from behind the veil of ignorance (Rawls Reference Rawls and Kelly2001, 114).Footnote 17 At the legislative stage, the principles of justice are “to be applied directly by legislators and regulators as they make decisions about the rules that govern the many complicated institutions within which economic production, trade, and consumption take place” (Freeman Reference Freeman2007, 100). While the requirements of justice will inform the formulation of the rules governing the formation and conduct of productive organizations, the point is that the rules themselves are not the object of the social contract.
At this point, it may be objected that even though Rawls does not include the rules governing productive organizations within the basic structure, there is nothing to preclude Donaldson from adapting Rawls’s theory to his own purposes in order to respond to Conry’s objection of indeterminacy.Footnote 18 Conry objects that because Donaldson’s characterization of the state of individual production lacks specificity, parties will find it rational to agree to any one of a number of alternative economic arrangements. Following Brock and Bishop, Donaldson might try to avoid this problem by adapting Rawls’s account of the original position to include, among the principles to be chosen from behind the veil of ignorance, the rules governing the formation and conduct of productive organizations. In his account, Rawls goes to great lengths to specify the conditions under which parties in the original position are to choose the principles governing the basic structure of society, and some scholars argue that even on Rawls’s own terms, rules governing the formation and conduct of productive organizations ought to be included within the basic structure.Footnote 19 In what follows, I examine the extent to which adapting Rawls’s account in this manner provides a way for Donaldson to respond to the objection of indeterminacy.
One issue concerns the degree of specificity about economic arrangements that can be realized by the choice of parties in the original position. Although the basic structure, as defined by Rawls, does not include the rules governing the formation and conduct of productive organizations, it does include the “structure of the economy (for example, as a system of competitive markets with private property in the means of production)” (Rawls Reference Rawls and Kelly2001, 10). In his writings, Rawls is clear that not all economic regimes with private property in the means of production meet the requirements of justice. To make this point, Rawls distinguishes between “welfare-state capitalism,” which “permits a small class to have a near monopoly of the means of production,” and a “property-owning democracy” that aims to ensure “widespread ownership of productive assets and human capital (that is, education and trained skills) at the beginning of each period, all this against a background of fair equality of opportunity.” Of these two regimes, only the latter meets the requirements of justice (Rawls Reference Rawls and Kelly2001, 139).Footnote 20 On the choice between these two regimes, the device of the social contract allows for specificity. At the same time, Rawls writes that economic regimes with public ownership of the means of production (i.e., “liberal socialism”) also meet the requirements of justice (2001, 138). On the choice between public and private ownership of the means of the production, Rawls’s account is indeterminate.
If such a fundamental choice regarding economic arrangements is left unspecified in the original position, there is reason to doubt that questions about the rules governing productive organizations will result in much more specificity. To be certain, as discussed above, there is reason to hold that an economic regime that allows for the formation of productive organizations can be consistent with the requirements of justice. In this respect, corporations need not be incompatible with the requirements of justice. However, an account that gives us no prima facie reason to rule out corporations is not the same as an account that provides a positive account for why parties will choose an economic regime that permits the formation of corporations.
A second question is whether having the rules governing productive organizations chosen from behind the veil of ignorance risks undermining key features that are thought to make Rawls’s account of justice attractive in the first place. A defining feature of Rawls’s account is that it is a liberal account “in that it protects and gives priority to certain equal basic liberties, which enable individuals to freely exercise their consciences, decide their values, and live their chosen way of life” (Freeman Reference Freeman2007, 44).Footnote 21 This is captured in the priority accorded to basic liberties over considerations of social or economic welfare in Rawls’s account. For example, Rawls makes clear that the basic liberties include civil and political liberties, but not rights regarding the ownership or control of the means of production (2001, 114), and that the claims of liberty are to be satisfied before other social needs (Freeman Reference Freeman2007, 64).Footnote 22 If the case for permitting corporations is that doing so enhances the basic liberties, as Bishop (Reference Bishop2008) argues, then such an argument risks undermining a defining feature of Rawls’s account. Another key feature of Rawls’s account is that the parties “put aside reliance on knowledge of all particular facts about themselves and their social and historical circumstances, including their particular conceptions of the good, and even including their comprehensive religious, philosophical and moral convictions” (Freeman Reference Freeman2007, 154). They do not have information about economic interests and contingencies. What information the parties do have is a concern for “primary social goods” which are “essential to individual’s freedom and self-respect: rights and liberties, powers and opportunities, income and wealth, and the bases of self-respect” (Freeman Reference Freeman2007, 63). All of these restrictions are to preserve the impartiality associated with the original position. If the case for an economic regime that permits the formation of corporations is that such a regime will further the economic interests of consumers and workers, then making such a decision from behind the veil of ignorance will require changing the nature of the information available to the parties and the relevant standards for impartiality.
None of this is to deny that parties behind the veil of ignorance will acknowledge the need to introduce requirements for economic activity that will apply to productive organizations, should they exist. For example, to ensure the enjoyment of the basic political liberties, parties from behind the veil of ignorance may acknowledge the need for restrictions on corporate political activity should corporations be allowed to exist and operate. In this sense, some rules governing the formation and conduct of productive organizations may be said to be part of the rules and institutional arrangements to be chosen in the original position. This is not, however, because questions about the formation and conduct of productive organizations are to be addressed in the original position. Instead, it is because answers to questions about the political liberties, which are the proper subject for deliberation from behind the veil of ignorance, have implications for the formation and conduct of productive organizations.Footnote 23
3.2. A Rawlsian Amendment
The accounts discussed above aim to address the problem of indeterminacy by incorporating the contract about productive organizations into the contract made by parties from behind the veil of ignorance in Rawls’s account of justice. As discussed above, doing so involves significant challenges given the way in which questions about the purpose and responsibilities of productive organizations are addressed within Rawls’s account. These challenges, however, need not rule out reference to Rawls’s framework altogether. In Rawls’s framework, members of society do address questions about the purpose and responsibilities of productive organizations, but only at a later stage—namely, the legislative stage. One way in which to draw upon Rawls’s framework then is to frame the questions Donaldson raises about the existence of for-profit corporations and their responsibilities as questions addressed to legislators in Rawls’s account.Footnote 24
There is something to be said for amending Donaldson’s account along these lines. The legislators in Rawls’s account occupy a similar situation to economic agents described in Donaldson’s account. Both groups are considering what sorts of rules to put into place regarding the formation and conduct of productive organizations, including the possibility of permitting corporations. Both groups also have in the background a conception of justice that places constraints on what sorts of organizational forms and conduct are permissible. Despite these similarities, however, there is reason to doubt that amending Donaldson’s account along these lines adequately addresses the problem of indeterminacy.
To begin, there are a few key differences in the approach to be taken by each group. In the case of Donaldson’s account, the parties are motivated, it seems, by economic interests (Donaldson Reference Donaldson1982, 42, 45). In the case of Rawls’s account, the legislators are to make decisions based upon what justice requires and permits (Rawls Reference Rawls and Kelly2001, 48). This does not mean that legislators are to ignore the impact of legislation on the ability of citizens to satisfy their economic interests. Rather, the impact of legislation on economic interests is to be considered in ways that relate to considerations of justice—namely, “maximizing the long-term expectations of the least advantaged under conditions of fair equality of opportunity” (Rawls Reference Rawls1999, 174-175). Such considerations may very well give rise to legislation that permits the formation and operation of corporations. However, the case for such legislation is not, as in Donaldson’s account, that economic interests are better satisfied. Instead, legislators look to how the least advantaged fare across a variety of economic regimes, some of which are likely to allow for the formation and operation of corporations. Although the satisfaction of economic interests is likely to be relevant (e.g., the long-term expectations of employees are likely to depend on their wages), it is not the primary rationale for their existence. It is an indirect consideration, and in a more general sense, it may not matter at all. That is, individuals may have any one of a number of reasons to form corporations. What matters is how they fare under an economic regime that allows their existence.Footnote 25
Furthermore, even if one were to imagine that the parties in Donaldson’s account were similarly motivated to the legislators in Rawls’s account, there is reason to doubt they would arrive at the same set of responsibilities enumerated by Donaldson. Recall that for Donaldson, productive organizations should minimize 1) the alienation that may arise from greater separation between workers and their final product, 2) the “lack of worker control over work conditions,” and 3) the “monotony and dehumanization” frequently associated with highly specialized and automated work (1982, 52). Work that is structured to meet these responsibilities is often termed “meaningful work”—i.e., “work that is interesting, that calls for intelligence and initiative, and that is attached to a job that gives the worker considerable freedom to decide how the work is to be done and a democratic say over the character of the work process and the policies pursued by the employing enterprise” (Arneson Reference Arneson1987, 522). It is a matter of considerable debate, however, whether Rawls’s account of justice—or a liberal egalitarian account of justice more generally—would require institutional guarantees for meaningful work. One of the central objections to requiring such guarantees is that the value of meaningful work is grounded in a specific conception of human good, and liberal egalitarian accounts of justice, including Rawls’s theory, must not rely upon a specific conception of the good life in determining policies and institutional arrangements.Footnote 26
This is not to deny there may be other ways to amend Donaldson’s account so that the device of the social contract serves as more than a heuristic device to test the rationale for productive organizations. Recall that in section 2.2, two additional roles for the device of the social contract were considered. The first was as a heuristic device to help identify responsibilities that would not be recognized without the device of the social contract. The second was to serve as a heuristic or test for the rationale given for these responsibilities, which on Donaldson’s account, is social welfare. The above discussion concerning the responsibilities regarding workers suggests one way to amend Donaldson’s account so that the device of the social contract serves these roles. This would be to introduce a much more specific conception of welfare, one that incorporated the idea of meaningful work and that grounded these responsibilities—along with the additional responsibilities toward workers enumerated in Corporations and Morality (1982, 129-157) and discussed in section 2.2—without reference to considerations such as justice or equal freedom.
At the same time, the above discussion points to a difficulty in adopting such an approach, which is that it involves a potentially controversial conception of welfare. Insofar as Donaldson intends for his account to have broad appeal and applicability, it remains a question whether such an approach is compatible with his account. Furthermore, as discussed in section 2.2, Donaldson himself invokes considerations other than welfare to specify and defend the rights of workers within productive organizations. In the end, there is reason to doubt that amending Donaldson’s account represents a plausible way to address the problem of indeterminacy and ensure that the social contract plays a heuristic role in identifying the responsibilities of productive organizations or the reasoning for them.
4. A CONTRACTUALIST INTERPRETATION
In the light of the previous discussion, I now turn to explore another way to interpret Donaldson’s account to respond to the objection that the social contract plays no role in specifying and grounding the responsibilities of productive organizations. This interpretation draws inspiration from T. M. Scanlon’s contractualist account of morality (1998) and the way in which it makes central the role of agreement in specifying and grounding obligations. If Donaldson’s account lends itself to such an interpretation, then we have reason to accept Donaldson’s claim that the device of the social contract helps to specify and ground the responsibilities of productive organizations.
Like Donaldson and other social contract theorists, Scanlon invokes a kind of hypothetical agreement in his account of morality. The rough idea is that the principles we ought to follow are those that reasonable people would accept. Put more precisely, “an act is wrong if its performance under the circumstances would be disallowed by any set of principles for the general regulation of behavior that no one could reasonably reject as a basis for informed, unforced general agreement” (Scanlon Reference Scanlon1998, 153). In response, one objection that has been raised against Scanlon’s account is by Simon Blackburn (Reference Blackburn1999). Blackburn asks, “Suppose it is reasonable to reject my principles because, for instance, they lead to vast inequalities of wealth. Why then isn’t that the very feature that makes my principles wrong? Why go through the detour of dragging in the hypothetical agreement with others?” (1999). The appeal to a hypothetical contract, according to Blackburn, becomes redundant.
Michael Ridge (Reference Ridge2001) argues that Blackburn and other critics, such as Colin McGinn (Reference McGinn1999) and Philip Pettit (Reference Pettit1999), overlook a key feature of the sorts of reasons that count toward rejecting a principle. Consider another example, this time from McGinn (Reference McGinn1999). McGinn writes, “It is wrong to drop radioactive debris … over the population below … if it may be criticized on moral grounds—namely, that it has needlessly caused the suffering and death of thousands of innocent people—then that is the reason the action is wrong …. The reference to interpersonal justifiability adds nothing to the simple claim that the action was wrong because of the suffering and death it caused” (1999, 35). Inequality, suffering, and death are all moral considerations that count against an action or principle in a way that apply to all persons. They are reasons that are agent-neutral.Footnote 27 As Ridge points out, however, for Scanlon, “impersonal reasons do not, themselves, provide grounds for reasonably rejecting a principle” (475, quoting Scanlon, 220). Instead, reasons must be personal to count as grounds for reasonably rejecting a principle, and once we recognize this, the fact of reasonable rejection is no longer a “detour” or “unhelpful epicycle” as it is when considering only reasons that are agent-neutral (2001, 472).
The idea is roughly as follows.Footnote 28 Suppose your project is to become a world-class chess player, and I do not care for chess or care about your chess career. If I am considering an action that would prevent you from completing your project and would have no negative impact on my projects, you would have an agent-relative reason to reject any principle that allowed such an action. This is an agent-relative reason in that it is not a reason I share. Whether the reason qualifies as grounds for reasonable rejection remains to be determined. However, the point is that “insofar as your agent-relative reasons could ground a reasonable rejection to any principle allowing my action, that fact provides me with moral reason not to perform the action” (Ridge Reference Ridge2001, 478). The moral reason for me not to commit the action is not simply the frustration of your project. As an agent-relative reason for you, that is not a reason for me. Rather, the moral reason is the fact that your agent-relative reason could ground a reasonable rejection of a principle that would permit my action. In this manner, consideration of the possibility of reasonable rejection matters.
Ridge (Reference Ridge2001) outlines some of the ideas about morality that Scanlon’s approach aims to capture. In trying to appreciate the personal projects of others, one idea that Scanlon’s approach captures is the idea of putting ourselves in the shoes of others. This is different, for example, from appreciating the “agent-neutral disvalue of pain” (2001, 479). Another idea captured by Scanlon’s approach is that a very small benefit to a large group of people does not justify an enormous harm to a particular individual. The thought here is that the particular individual’s agent-relative reasons can be taken into account in a way that they cannot under agent-neutral theories, such as utilitarianism. Scanlon’s approach also captures the idea that actions can wrong particular persons rather than “from the point of view of the universe” (2001, 479).
Central to Scanlon’s contractualist account is the idea that an individual’s complaint is itself what characterizes the wrongness of an action. On the contractualist view, “the wrongness of an action is not to be equated with the properties that make it unjustifiable” (Ashford and Mulgan Reference Ashford, Mulgan and Zalta2012). Instead, the wrongness of an action “is to be equated with its being unjustifiable” (Ashford and Mulgan Reference Ashford, Mulgan and Zalta2012). Whether a principle can be reasonably rejected is not merely a heuristic or test; it is itself constitutive of what makes an act or rule right or wrong. The requirement that a principle cannot be reasonably rejected is not the same as the requirement that all parties agree to a principle. However, both requirements have at their basis the idea of unforced agreement. In turn, one may ask whether there is a plausible contractualist interpretation that can be given to Donaldson’s account that would make the hypothetical agreement central to specifying and grounding the responsibilities of productive organizations.Footnote 29
On such an interpretation, the idea would be to relate the responsibilities of productive organizations to principles that no one could reasonably reject on the basis of agent-relative reasons. As an illustration, consider the responsibility on the part of productive organizations to minimize worker alienation (Donaldson Reference Donaldson1982, 52). Following the discussion by Ridge (Reference Ridge2001), there would be such a responsibility if someone’s agent-relative reason could ground a reasonable rejection of any principle that did not require productive organizations to minimize worker alienation. The agent-relative reason is a reason that not all agents need to share as a reason to reject any such principle. What might such a reason be? An example of an agent-relative reason in this case might be the lack of self-expression and inability to exercise her craft that a highly skilled woodworker experiences when she no longer constructs individual pieces of furniture, as she would in the state of individual production, but instead attaches machine-milled legs to mass-produced chairs in a furniture factory. If her identity is deeply rooted in her ability to exercise her craft as a woodworker, then the alienation she experiences in the factory is a reason for her to reject any principle that does not minimize worker alienation. Insofar as other factory workers need not share her experience and hence her reason for rejecting any such principle, then her reason is agent-relative.Footnote 30
At this point, a number of questions arise if one is to complete this account. One question is whether the frustrated woodworker’s agent-relative reason can ground a reasonable rejection of any principle that does not require productive organizations to minimize worker alienation. The answer will depend, in part, on various considerations, including the impact on others and the trade-offs involved in minimizing worker alienation. As stated, the responsibility on the part of productive organizations is to minimize worker alienation. In this respect, the case for grounding a reasonable rejection in the first instance seems more plausible than in the case, say, of any principle that required productive organizations to eliminate worker alienation or to maximize worker fulfillment. Another question is just what counts as an agent-relative reason. As Ridge (Reference Ridge and Zalta2011) points out, there is debate within the literature about how to draw the distinction between agent-relative and agent-neutral reasons. For our purposes, however, the more pressing question is whether framing the responsibilities of productive organizations in terms of agent-relative reasons is consistent with what Donaldson has in mind for his account of the responsibilities of business organizations.Footnote 31
In some ways, contractualism can accommodate Donaldson’s account. For example, there is no veil of ignorance in contractualism as there is in Rawls’s theory of justice. Also as in Donaldson’s account, agents are not assumed to be purely self-interested under contractualism, a feature that distinguishes it from many social contract theories (Ashford and Mulgan Reference Ashford, Mulgan and Zalta2012). However, if one considers the reasoning process that Donaldson envisions for members of society, it seems to involve agent-neutral reasons. Although there are disagreements, as noted above, on how to characterize the distinction between agent-relative and agent-neutral reasons, in the most general sense, agent-neutral reasons are reasons that apply to anybody and not only to particular individuals. The standard example of an agent-neutral reason is something like the promotion of overall welfare. What matters is not the promotion of any specific individual’s welfare, but rather welfare in an impersonal sense. On Donaldson’s account, the reasoning of the parties involves considerations such as increasing social welfare and minimizing the drawbacks that arise from general features of the production process (Donaldson Reference Donaldson1982, 51). There is reason to doubt that contractualism provides a way to specify and ground the responsibilities Donaldson assigns to productive organizations, at least if we are to keep to the sorts of reasons he advances as grounds for those responsibilities.
5. THE ROLE OF THE SOCIAL CONTRACT
In Corporations and Morality, Donaldson draws upon the social contract tradition in order 1) to justify the existence of productive organizations and 2) to specify and ground their responsibilities. In this paper, I argue there is reason to doubt that the device of the social contract justifies the existence of productive organizations or plays a role in specifying and grounding their responsibilities. I argue that if the idea of the social contract does play a role in Donaldson’s account, the most plausible role it plays is as a heuristic or test for the claim that what justifies the existence of productive organizations are considerations of social welfare.
To be clear, in arguing for a limited role for the social contract in Donaldson’s account, I do not mean to suggest that the criticism in this paper extends to all of social contract theory.Footnote 32 For one thing, I do not argue that social contract theory has no role to play. In the case of justifying the existence of productive organizations, I argue that the device of the social contract plays a heuristic role that is similar to the role attributed to it in other contemporary applications of social contract theory. In Donaldson’s account, the agreement of the parties to move from a state of individual production provides a test of the reasons for introducing productive organizations, reasons that relate to the ways in which productive organizations help satisfy economic interests and improve social welfare. Where I do argue that the application of social contract theory is less successful is with respect to specifying and grounding the responsibilities of productive organizations. On this matter, Donaldson’s account differs from other applications of social contract theory in a way that criticism of his account need not extend to these other applications.
The difference roughly is this. In most applications of social contract theory, whether traditional or contemporary, the institutional arrangements under consideration are fairly foundational in the role they play in society and in structuring the lives of its members. For Hobbes, for example, the institutional arrangement under consideration is one in which the state has absolute authority with limited rights on the part of citizens to resist. For Rawls, the subject of agreement is the basic structure of society. Social contract theory captures the intuition that these arrangements require agreement or justification in a way that relies minimally on outside standards or authority, and appeals as much as possible to the fact of agreement or justifiability of the arrangements to the parties directly affected. This intuition is strongest, it seems, when institutional arrangements under consideration are foundational and extensive in their impact. They stand in greater need of justification, and at the same time, there are fewer independent standards or authority to which to appeal.
In contrast, in Donaldson’s account, the institutional arrangements under consideration are not as foundational or as extensive in their scope. In the state of individual production, members of society, it seems, already have a conception of justice, a system of property rights, and mechanisms for trade. The choice under consideration is whether to introduce a set of rules that allow the formation of productive organizations and if so, what responsibilities to assign to them in their conduct. While such a choice may have an extensive and pervasive impact on members of society, it is not as though members of society lack independent standards or outside authority in making this choice. In this light, the conditions that help to make plausible the application of social contract theory—as either a justificatory or a heuristic device—do not obtain in the case of justifying the existence and responsibilities of productive organizations.
In his response to Kultgen, Donaldson puts forward another way to understand the role of the social contract that is distinct from either a justificatory or a heuristic device. This additional role is as an “analogical” concept—that is, “an ‘as if’ moral construct the logical form of which is ‘One should behave as if such and such were the case.’” Donaldson continues, “in the present context, this means that the social contract advises individuals and corporations to act as if the members of society had agreed to define the rights and obligations of productive organizations in a certain manner” (1986, 43). Furthermore, according to Donaldson, one need not move from an analogical concept to uncover the rights and obligations of productive organizations. Instead, “we might instead first ask what specific persons’ rights or obligations are, and then use our findings to generate the outline of an analogical concept, for example, of a social contract” (1986, 44). On this reading, the social contract need not play a role in justifying the existence of productive organization or in specifying and grounding the rights and obligations of productive organizations. Instead, parties are to act as if there were such a contract. As an analogical concept, the social contract should be understood as embodying a conception of the purpose and responsibilities of productive organizations.
In concluding this paper, I outline the case for interpreting the social contract in Donaldson’s account as an analogical concept and two key themes it embodies that serve to distinguish Donaldson’s conception of the purpose and responsibilities of productive organizations. In doing so, it will help to address the question of how the social contract can serve as an analogical device if it is limited in its role as a heuristic device in specifying and grounding the responsibilities of productive organizations.Footnote 33
As noted at the end of section 1, there are two ways in which the idea of the social contract enters into Donaldson’s account. The first is as an agreement between productive organizations and members of society that lays out the rationale for the existence of productive organizations and the rights and responsibilities assigned to them. The second way in which the idea enters is more general and refers to the methods of social contract theory to answer questions about what justifies the existence of productive organizations and what are their responsibilities and the grounds for them. In framing the social contract as an analogical device, Donaldson states that the social contract, understood in the first sense, can stand independently of any role it might play in the second sense. The idea seems to be that the terms of the social contract can be specified and grounded without reference to the device of the social contract. This idea is consistent with what has been argued in this paper. As I have argued, whereas the social contract is best understood as heuristic for identifying the purpose of productive organizations, many of their responsibilities (e.g. not abusing political power or minimizing dehumanization) are more plausibly specified and grounded with respect to considerations that are independent of the hypothetical agreement, such as a background conception of justice. There is nothing to preclude including terms in a contract that derive their normative force from considerations independently of the fact of agreement to the contract.
If the purpose and responsibilities of productive organizations are largely identified and justified independently of the device of the social contract, it may be asked what is gained by framing them as if there were a social contract between society and productive organizations. In sketching an answer to this question, two features emerge from Donaldson’s account that, together, reflect a distinctive conception about the purpose and responsibilities of productive organizations and an approach to theorizing about them that remains underappreciated.Footnote 34
The first is the idea that the ability of productive organizations to operate is dependent on their operation benefiting members of society as a whole. Acting as if there were a social contract between productive organizations and society captures this idea. On this view, productive organizations are not merely voluntary associations whose existence is justified with reference to their individual members’ right to associate.Footnote 35 Nor is the idea that productive organizations are somehow extensions of public institutions.Footnote 36 The idea is similar to the view that productive organizations require legal permission and social legitimacy to operate, so that if they are to receive permission or a “license” from society in order to operate, they must meet certain conditions, among which is benefiting society.Footnote 37 Recall that Donaldson characterizes the form of the social contract as, “We (the members of society) agree to do X, and you (the productive organizations) agree to do Y” (Donaldson Reference Donaldson1982, 42). At the same time, the sense in which the operation of productive organizations is dependent on benefiting society is stronger than simply in the sense of a quid pro quo. This is the sense in which the contract is not a contract among equals. Rather, the rules that govern the formation and conduct of productive organizations are the product of human design, which in turn suggests that productive organizations through their operation must serve the interests of society taken as a whole.
A second feature of Donaldson’s account is to place economic production at the center of an account of corporate purpose and responsibility. Production, in this sense, is understood as more than exchange. It involves the transformation of various inputs, such as raw materials or knowledge, into goods and services that can be used or enjoyed by people in ways that the original inputs, on their own, could not.Footnote 38 At first, focusing attention on economic production may not seem to be that distinctive a feature of an account of corporate purpose and responsibility. After all, it may be said, if the focus is not on economic production, what else is being evaluated?
In response, I want to suggest two ways in which contemporary accounts of corporate purpose and responsibility tend to overlook the relevance of economic production or take it for granted. The first is a tendency in many accounts to focus on the achievement of an outcome that is somewhat removed from the proximate control of members of productive organizations (e.g., social welfare).Footnote 39 While it is true that Donaldson also focuses on benefit to society, he characterizes the benefit in terms of the immediate outputs of productive organizations—e.g., satisfaction of consumer interests for goods such as “shelter, food, entertainment, transportation, health care, and clothing” (1982, 45) or “increasing income potential” (1982, 48). From the perspective of providing guidance to members of productive organizations, focusing on immediate outputs as opposed to broader, more distant goals may be more helpful. The thought seems to be that the very fact that individuals are better able to meet their economic wants and needs is of social value without reference to some overarching conception of social good. The second tendency in contemporary accounts of corporate purpose and responsibility is to characterize “social value” or “social impact” in terms of the impact of productive organizations on constituencies beyond the firm, such as the local community.Footnote 40 By justifying the existence of productive organizations in terms of their benefits to workers and consumers, the social contract is a reminder that engaging in economic production in the right way is an important value to society, something that tends to be overlooked in many contemporary approaches to theorizing about corporate responsibility.
Taken together, these two features of Donaldson’s account reflect a distinctive conception about the purpose and responsibilities of productive organizations and an approach to theorizing about them that remains underappreciated. On this conception, productive organizations are neither purely private associations nor extensions of public institutions, and yet they are meant to serve society as a whole. In theorizing about their purpose and responsibilities, the approach is to focus on their role in economic production and the parties who are involved and benefit from this activity. As an analogical device, the social contract in Donaldson’s account reminds us of these features. Donaldson’s account charts a course for further theorizing about the purpose and responsibilities of productive organizations, even if the social contract plays only a limited role.
Acknowledgements
An early version of this paper was presented at “Normative Business Ethics in a Global Economy: New Directions in Donaldsonian Themes,” a conference held by the Wharton School, University of Pennsylvania in honor of Tom Donaldson. I thank the conference organizers, Bill Laufer and Alan Strudler, for the opportunity to participate and audience members for their comments and questions. For their excellent guidance in revising the paper, I am grateful to Ben Wempe, three anonymous reviewers, and the special issue editor Alan Strudler. The final version is much-improved thanks to their many insightful questions and detailed comments. For financial support, I thank Harvard Business School. Finally, I express my immense gratitude to Tom Donaldson. I am deeply indebted to him for his work in establishing and developing the field of business ethics, and for his inspiration, guidance, and generosity.
