BASIC LIBERTIES, THE MORAL POWERS AND WORKPLACE DEMOCRACY

The article responds to previous work, by Martin O’Neill, about the Rawlsian case for an entitlement to an element of workplace democracy. Of the three arguments for such an entitlement that O’Neill discusses, this article focuses mainly on the one he rejects (on the grounds of its having an implausible premise): the Fundamental Liberties Argument, according to which the right to an element of workplace democracy is a basic liberty. This article argues that while the argument can be improved to withstand O’Neill’s objection, it is invalid. The article sets out a new argument, the Argument from Risk to the Moral Powers. While inspired by the Fundamental Liberties Argument, it is valid. Moreover, its premises are at least as plausible as those of one of the two arguments upon which O’Neill builds his Rawlsian case for an entitlement to an element of workplace democracy.


Introduction
indicates that "democracy in the workplace", more specifically, and strongly, in the form of "Mill's idea of worker-managed firms" is "fully compatible" with his conception of justice. 1 Rawls (2001, p. 179) raises, but does not attempt to answer, the question of whether the "long-run prospects of a just constitutional regime may depend", in part, upon the establishment of workplace democracy. A positive answer to that question implies more than the mere compatibility of workplace democracy with a Rawlsian theory of justice.
Such an answer suggests that either Rawlsian principles of justice themselves, or those principles in conjunction with certain social and/or psychological facts, entail that there is a right to an element of workplace democracy (and, therefore, that there is a moral requirement to institute democracy in the workplace).
Three questions about the normative and logical connections between a broadly Rawlsian theory of justice and workplace democracy are relevant here.
First, do Rawlsian considerations allow for an element of workplace democracy? This is the question of consistency. Second, do these considerations favour an element of workplace democracy? This is the question of whether there is, by Rawlsian lights, (moral) reason to institute democracy in the workplace. Third, do these considerations require an element of workplace democracy? That is, does Rawlsian justice entail that there is a right (whether basic or non-basic) to 1 Rawls (2001, p. 114) considers the theory of justice to be neutral, at the level of fundamental rights, with respect to capitalist and socialist modes of production. The specific claim of Rawls (2001, p. 178) is about the compatibility of the worker-managed firm with "property-owning democracy", which is the form of social order that, assuming capitalism, and in contrast, for example, with the welfare state, Rawls favours. On the differences between these two forms of social order, with references to Rawls and others, see, e.g., Hsieh (2005, pp. 129-30;2009). Hsieh (2005, p. 115) begins by quoting more fully and commenting upon remarks from Rawls (2001, p. 178) about, in Hsieh's words, "worker participation in the governance of economic enterprises". Cf. Hsieh (2008, p. 397).
an element of workplace democracy, or decisive moral reason to institute it? 2 It is easier to establish that there is moral reason to institute workplace democracy than it is to establish, and, from a strategic point of view, to convince the opposition, that there is a right to an element of democracy in the workplace.
Thus, it would be regrettable to concentrate too much -whether here or at largeon the third question. This article assumes, like the work with which it mainly engages, and the passing remarks of Rawls himself on the topic, that the question of consistency is to be answered affirmatively. Over the article's course, the main focus shifts from the third question onto the second. This shift has the additional strategic and theoretical advantage of allowing for the possibility that democratic rights are only, in the end, appropriate in some workplaces rather than others. 3 The powerful will be more likely to concede that there is moral reason to share some of their power than that their power belongs, largely, and by right, to others.
The argument that there is such moral reason, short of an inalienable right, favours gradualism and reform over revolution. It is likely to have greater practical influence, and unifying force, than the appeal to a putatively inalienable right. 4 Perhaps you are already thinking: to what does workplace democracy amount and what forms might it take? 5 It is not the purpose of this article to specify any putatively Rawlsian answers to these questions. Rather, it settles for working with the notion of a "measure", "degree" or "element" of workplace democracy: it is concerned with whether there should be some democracy in the workplace rather than none. It is one thing to answer that question affirmatively and another to work out the details, both normative and practical, of measures 2 The distinction made here between consistency, favouring and entitling (as a matter of right) builds on Vega (2000, p. 35) and Hsieh (2008, pp. 72-73). A Rawlsian account of justice recognizes that consequentialist considerations can qualify as morally decisive reasons when acting on them violates no rights. To cut down on clutter, in what follows the word "right" will be used as shorthand, except when the context excludes this, for anything to which all citizens are, as a matter of Rawlsian justice, entitled for decisive moral reasons. Vela's term "call for", quoted in the epigram above, accommodates both entitlement claims and favouring claims. 3 Thus, it is consistent with the view of Christie (1984, p. 116), according to which not all small businesses should be required to recognize such rights. 4 Norton (2003) has influenced these thoughts. In relation to debates about environmental conservation, Norton (2003, p. 469) distinguishes between the "intellectual debate" about what "the correct moral stance towards nature is" (including on the question of whether non-human species and ecosystems have inherent value) and the "strategic debate" about which arguments are "likely to be effective in saving wild species and natural ecosystems". My point is that even if the correct moral stance on workplace democracy should turn out to be more demanding that the stance I defend (e.g., in that the former demands inalienable democratic rights for all employees), the stance I defend seems more likely to be strategically effective in advancing workplace democracy as a progressive cause and practical project. 5 You might also be asking: what counts as a workplace? This question must here be left aside. On a separate issue, Vela (2000, pp. 144-145) argues that when it comes to democratizing production, employees should be at the front of the queue. I agree, but the issue has no bearing on the arguments below.
for instituting democracy in the workplace. Nevertheless, some brief observations may be in order to help forestall some over-estimations of what "workplace democracy" might mean here. In this connection, consider some shifts in the usage of "workplace democracy" over the course of a series of articles by Nien-hê Hsieh. The defence of an element of workplace democracy is a more modest project than defence of that which Hsieh (2005, pp. 115-116) describes as follows (while not setting out to defend it): Workplace democracy, as I understand it, is an economic regime in which workers have a right to participate in the governance of economic enterprises on [the] grounds that they have a right to exercise control over the means of production.
The presence of an entitlement to some degree or other of workplace democracy entails neither workers' (full or partial) ownership of, nor their (full or partial) control over, the means of production. 6 On this matter, the conception of workplace democracy with which this article will work converges with the "workplace republicanism" defended by Hsieh. According to it, writes Hsieh (2005, p. 137) "what matters is that the decision-making procedure [in the workplace] incorporates the voice of workers as part of the process". The view defended here adds that the incorporation of voice should be democratic, rather than, for example, done merely via such mechanisms as consultation, focus groups, and one-to-one discussions between employer and employee.
In a subsequent article, Hsieh (2008, p. 82) works with a less demanding definition under which a guarantee of the worker's right to participate in governance is independent of any requirement of ownership on the part of workers. In addition, (...) no constraints [are placed] on whether worker participation is direct or representative. Furthermore (...) workers and providers of capital [might] share control of the enterprise as in a system of co-determination. For a regime to count as one of workplace democracy, workers must be guaranteed at a minimum a say equal to that of the providers of capital. 7 On this definition, workplace democracy has three elements: (i) the workers have, as a matter of right, voice in respect of governance; 8 (ii) their voice is expressed by democratic means; (iii) their voice is equal to that of the owners of the means of production. This article is concerned only with the first two elements. The question is not whether Rawlsian considerations require, or favour, the possession by workers of equal voice, democratically expressed, to that of owners or bosses. Rather, it is whether Rawlsian considerations, require, or favour, that workers should have some democratically expressed voice (rather than none). This is a relatively modest question. The proposal that Rawlsian justice favours, or requires, the institutionalization of some measure or other of workplace democracy is also relatively modest. Therefore, the proposal is apt for philosophical evaluation in advance of any more demanding proposal that entails it.
Martin O'Neill (2008) has set out three Rawlsian arguments, two of which he endorses and the third of which we will shortly examine in detail, for the conclusion that a measure of workplace democracy is a requirement of Rawlsian justice (and therefore morally required). The relative modesty of O'Neill's goal, shared here, enables his arguments to avoid what Hsieh (2005, pp. 116-17) calls "the objection most often [levelled] against workplace democracy, which is that it pays insufficient attention to the need for managerial decision-making in largescale economic enterprises". 9 O'Neill (2008, p. 31) calls these arguments the "Fundamental Liberties Argument", the "Democratic Equality Argument" and the "Democratic Character Argument". 10 These arguments differ mainly in the terms of the sub-conclusion 8 A putative right that Hsieh (2005, pp. 117, 134-40), discusses is "the right to contest managerial decisions". This involves a form of voice, but need not be democratic. To see that voice in the workplace does not entail democracy in the workplace note, for example, that individuals could have the right, as individuals, to contest managerial decisions that directly affect them without having any organized role in determining the organization's explicit and documented values, strategy, tactics and policies. Moreover, their right to express their views on such decisions might permissibly, at least in respect of a right to voice that is so weak, elicit nothing more than the managers' classic response: "We hear what you're saying, but we'll be doing it anyway". 9 Christie (1984, p. 115) mentions the more general objection that calls for workplace democracy are, in practical terms, impossible to implement. Since there already are enterprises in which workers have some institutionalized democratic rights (and even economies in which a measure of democratic involvement by workers in decision-making is legally required), this more general objection is also here circumvented.
10 I have capitalized "Argument". Throughout, O'Neill uses both the expression "economic democracy" and the expression "workplace democracy". The former might be thought to encompass not only "democratization of economic enterprises" (O'Neill, 2008, pp. 30-31), at the level of the firm or agency of production, but also democratic political control, at a state or federal level, over production. Christie (1984, p. 113) regards economic democracy as something that involves "a serious attempt (…) to democratize the economic sphere, including workplaces". This suggests that workplace democracy is a form of, but not identical with, economic democracy. It is the implications of a broadly Rawlsian account of justice for workplace democracy, rather than for other forms or aspects of economic democracy, that is of interest here. For references to some other work, dating from 1978 to 2000, that argues, on meant to support the conclusion that justice requires a degree of workplace democracy.
The Fundamental Liberties Argument moves from the sub-conclusion that a degree of workplace democracy is a Rawlsian "basic liberty" to the conclusion that justice requires it. 11 The Democratic Equality Argument moves from the subconclusion that implementation of the difference principle, when understood as a principle that encompasses all social primary goods, requires that "an economic system" should allocate "decision-making powers in a broadly dispersed way" to the conclusion that a measure of workplace democracy is a requirement of justice (O'Neill, 2008, p. 31). 12 According to the Democratic Character Argument, the institutional recognition of a right to a measure of workplace democracy is appropriate because "a society which institutes forms of economic democracy will be more likely to preserve a stable and just basic structure over time, by virtue of the effects of economic democratization on the development of an active, democratic character among citizens" (O'Neill, 2008, p. 30 Argument. This article aims to show, first, that we can circumvent O'Neill's objection to the Fundamental Liberties Argument. Secondly, that, even so, the argument fails to establish the conclusion that workers have, on Rawlsian grounds, an inalienable right to a measure of workplace democracy. Thirdly, that the argument obscures the distinction between two types of requirements of justice: deontological requirements (which follow from the principles of justice alone) and practical requirements (which follow from the principles of justice only alongside certain facts about actual social conditions that are independent of the principles of justice). Fourthly, that there is nevertheless a strong case for saying that considerations of Rawlsian justice favour (and even, in light of social circumstances, morally require) the institutionalization of an entitlement (albeit alienable) to an element of workplace democracy for employees.
Section 1 explains the background to the Fundamental Liberties Argument and provides a detailed reconstruction of the argument.
Section 2 discusses the most significant Rawlsian objection, previously Section 4 explains why, rather than being an ad hoc manoeuvre designed to save the argument, this weakening is independently justified. This argument is on a par with the Democratic Character Argument. Both arguments suggest that Rawlsian considerations about justice favour a measure of workplace democracy: given certain actual social conditions (about which parties in the original position are ignorant) its presence would help enable a just social order to be reached and, once established, to be stable (given the continuation of some of those social conditions) over time. There is no suggestion that the practical requirements of justice cannot thereby be moral (and thus deontic) requirements. They are nevertheless distinct from, and more derivative than, those deontological requirements that are either: (i) inalienable rights of citizens (such as the right not to be discriminated against under the law on the grounds of race or sex); or, (ii) otherwise (e.g., via the difference principle) directly entailed by the principles of justice (independently of facts about social conditions that differ across the societies, actual or possible, to which the principles of justice apply).

Section 5 raises an objection to the Reformulated Fundamental Liberties
Section 6 is a concluding summary.

The Fundamental Liberties Argument
Before setting out the argument, let us clarify some of its background. The conception of justice as fairness regards citizens as persons engaged in social cooperation who have "what we may call 'the two moral powers'": "the capacity for a sense of justice" and "a capacity for a conception of the good" (Rawls, 2001, pp. 18-19;O'Neill, 2008, p. 34). The principles of justice concern the design of the basic structure of society. This basic structure is 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 co-operation over time (…). (Rawls, 2001, p. 10) The fundamental case in which the capacity for a sense of justice is exercised is in "the application of the principles of justice to the basic structure and its social policies" (Rawls, 2001, p. 112;O'Neill, 2008, p. 34). The fundamental case in which the capacity for a conception of the good is exercised is in "forming, revising, and rationally pursuing such a conception over a complete life" (Rawls, reconstruct the Fundamental Liberties Argument as follows.

Fundamental Liberties Argument
14 Though O'Neill seems to interpret what Rawls writes here as, in effect, providing a definition of the basic liberties, this is not the role that the quotation actually plays in Rawls's own work. Rather, it concerns the question of how the basic liberties should be weighted, in a scheme of liberties, once a list of them is already in place. See further McLeod and Tanyi (2017).

1.
A liberty is basic (and thus its institutionalization is a requirement of justice) if and only if it is necessary to the provision of "the social conditions essential for the adequate development and the full and informed exercise of [people's] two moral powers (...) in the two fundamental cases" (Rawls, 2001, p. 112;O'Neill, 2008, pp. 35-36). 14 (Premise) 2.
The "freedom to take part in decisions about economic production" is necessary to the provision of these conditions (O'Neill, 2008, p. 35). This freedom is a basic liberty (and thus its institutionalization is a requirement of justice).

5.
Justice requires a degree of democracy in the workplace.

(From 3, 4)
According to the argument, as O'Neill (2008, pp. 31, 33) states it, "a right to participate in the democratic determination of decision-making within the social and economic institutions to which citizens belong" is one of the fundamental liberties that must be constitutionally protected in order to enable citizens to exercise the capacity for a sense of justice and the capacity to have a conception of the good. As a consequence, the institutionalization of a degree of workplace democracy is a requirement of justice.
Premise 1 is, at least as it features in the argument, definitional and it goes unquestioned in O'Neill's discussion. However, as we shall see, there are good and independent reasons for taking the definition embodied in Premise 1 to be in need of weakening. Premises 2 and 3 are debatable. In O'Neill's formulation of the argument, Premise 3 is hidden. It may be, in effect, that O'Neill does not spell it out as a separate premise because he takes it already to be entailed by, or included in, Premise 2. However, it is helpful to distinguish between Premises 2 and 3 because the freedom to take part in decisions about economic production could arguably be embodied in a form of state socialism under which there is no element of workplace democracy within enterprises or industries. Workplace democracy involves each worker in a firm having democratic rights within that firm. While this is a form of economic democracy, defined as the freedom to participate in decision-making about production, it is not exhaustive of it. For example, in a form of state socialism citizens who do not work could, via the political process, have such decision-making powers.
In setting out the arguments that underlie the non-definitional premises of the argument, O'Neill (2008, pp. 35-36) appeals, as follows, to the fundamental case in which the capacity for a sense of justice is exercised, namely in making judgements about the justice of the basic structure and its social policies: (…) with regard to the sense of justice (…) unless individuals have some first-hand experience in the deliberative direction of some collective enterprise (such as a firm), then they will lack the skills that will be needed in order to participate fully in the "free use of public reason" in democratic politics. (…) citizens who are used to the regular exercise of their deliberative capacities with regard to the direction of economic enterprises will be best placed to exercise those capacities in a full and informed manner in the political sphere.

O'Neill's objection to the Fundamental Liberties Argument
Before we turn to the objection that leads O'Neill to reject the argument, it is worth noting, in the interests of completeness, some of his other critical observations about the argument. O'Neill (2008, pp. 37-41) notes that Rawls himself would have rejected the Fundamental Liberties Argument, though for a reason that O'Neill finds indecisive. This reason can be summarized as follows.
Rodney Peffer proposed an amendment to Rawls's principles of justice. Under the amendment, a right to workplace democracy, while not a basic liberty, is a requirement of justice (Peffer, 1990, p. 14;O'Neill, 2008, p. 37). According to Peffer's first principle, persons are entitled to have their "basic security and subsistence rights" met (Peffer 1990, p. 14). According to his second principle, here is to be a maximum system of basic liberties" (Peffer, 1990, p. 14). Peffer's third principle is: (3) There is to be (a) equal opportunity to attain social positions and offices, and (b) an equal right to participate in all social decision-making processes within the institutions of which one is part. (Peffer, 1990, p. 14) Rawls (1993, pp. 7-8, note 7) rejected (3)(b) on the grounds that it appeared to him "to require a socialistic form of economic organization" that should not, in his view, have been a requirement of "the first principles of justice" (that is to say, those principles that are chosen in the original position under the veil of ignorance). Thus, although Peffer's proposal does not entail that a right to workplace democracy is a basic liberty, Rawls's general reason for rejecting Peffer's proposal also applies to any proposal that entails that such a right is a basic liberty (for here it would be included in the very first principle, on Rawls's account, of justice).
For the following reasons, O'Neill does not find this Rawlsian line of the objection to the Fundamental Liberties argument convincing. First, it might be maintained that principles of justice need not be neutral about modes of socioeconomic organization and that, accordingly, the theory of justice may quite aptly have a socialistic, rather than a liberal, orientation from the start. As O'Neill (2008, p. 39;cf. Mayer, 2000, p. 304) puts the point, "we should not reject an otherwise normatively attractive account of social justice simply because it can only be realized under some particular system of ownership". Secondly, Rawls, in his response to Peffer, mistakenly bundled up ownership rights over the means of production with the right to workplace democracy. Therefore, the Rawlsian line of objection to the Fundamental Liberties Argument also suffers from this flaw. While ownership rights are often, de facto, bundled up with rights of control, there is no essential moral connection between them. 15 In relation to economic production, full ownership of the means of production need not be a necessary condition for the exercise of some control over how production is to take place. For example, one could allow that productive capital could be owned by particular private individuals, whilst nevertheless allowing that the use of that productive capital could legitimately be constrained by a requirement that workers (…) were entitled to some degree of participation in decision-making about production. (O'Neill, 2008, p. 39;cf. Vela, 2000, pp. 156-158) For this reason, provided that by "socialism" Rawls intended a system involving workers' ownership of the means of production, Rawls's objection to

Argument
Even if Premise 2 is false, it would be a natural piece of advocacy on behalf of the argument to look for a way of weakening it whilst still securing the argument's conclusion. One way of doing so would be to replace Premise 2 itself with material about mitigating against contingent risk (above a certain threshold) to the moral powers and their exercise in the two fundamental cases. Given the elliptical nature of Premise 2, its weakening can be brought about by weakening the wording of Premise 1. The reformulation of Premise 1 can be accomplished, as Section 4 argues, in a manner that is independently justified rather than ad hoc. Moreover, valid inference from Premises 1 and 2 to line 4 of the argument only requires that Premise 1 should state a sufficient condition for a liberty's being basic: a full definition of the basic liberties, stating their necessary and sufficient conditions, is not required. As we shall also see in Section 4, this change is also independently justified: even if the necessity of an entitlement to the provision of The "freedom to take part in decisions about economic production" is necessary to the provision of these conditions. This freedom is a basic liberty (and thus a requirement of justice).
Justice requires a degree of democracy in the workplace. (From 3, 4) There are some capacities that are unlikely to atrophy or be stunted just because the person who has them cannot exercise them at work (rather than because, say, of the nature of the work or the length of the working day or week).
For example, if one is an office worker working normal hours within reasonable commuting distance of home, then one's job does not interfere with one's capacity to be an excellent cook. Likewise, if I am unable to learn a foreign language at or on the way to work, I might still be able to do it in my leisure time. The workplace, per se, and working life in general, pose no threat to capacities and excellences like these.
In the original version of the Fundamental Liberties Argument, the suggestion was that, in a society in which there are firms, workplace democracy is necessary to the provision of "the social conditions essential for the adequate development and the full and informed exercise of the two moral powers".
O'Neill's objection was that the moral powers can be developed and exercised, in a manner that fosters their full and informed exercise in relation to the justice of the basic structure and its social policies, outside of the context of making economic decisions about production: for example, in voluntary associations. The objection suggests that the two moral powers are capacities, like being an excellent cook or a good cyclist, that people can adequately gain and maintain outside the workplace. No arrangements in the workplace are therefore necessary to their development or their exercise. This analogy is implicit in the objection to the Fundamental Liberties Argument. The moral powers, however, are relevantly unlike other some other capacities, in the following respects. First, the moral powers are intimately related to habits of thought, attitude and action that colour our relations and everyday interactions with our fellow persons: some of these habits can only be sustained if practised, more or less consistently, throughout one's social dealings with others. 16 Those who occupy working situations in which unjust practices are encouraged, or worse, necessary to career success, are to some (perhaps significant) degree likely to have their sense of justice corrupted in a manner that risks spillover into areas of life outside the workplace and which might prejudice their capacity to exercise the sense of justice in a full and informed manner in the fundamental case. To put the point another way, thinking, feeling and acting justly are habits that can be lost not only if one does not have the opportunity to exercise them, but if one is regularly put into a situation in which one is under pressure to participate in, or to remain silent about, actions that are contrary to the requirements of justice. Secondly, and relatedly, the hierarchical nature of the typical workplace means that it is a theatre of special threats to the two moral powers. Thirdly, the hierarchical workplace, as a coercive environment, is one in which people are not always free to act with impunity in accordance with their consciences and in which conscientious objections to particular workplace demands are unlikely to be institutionally tolerated or accommodated (particularly when not based on grounds protected by anti-discriminatory legislation). Fourthly, for working people in full-time jobs, work is the predominant mode of social co-operation, It is presumably not enough, if a just basic structure is to be stable, that some people should be able regularly to engage in activities that provide the opportunity to develop and engage the two moral powers while others do not. On Rawls's account, maintaining a stable and just basic structure requires that citizens should all be able to develop and engage the two moral powers. Thus, despite O'Neill's objection to his formulation of the argument, there seems, at this stage, to be plenty of life left in the argument yet.

Weakening Premise 4 of the Fundamental Liberties Argument is not ad hoc
In order to show that the above weakening of Premise 1 is not ad hoc, this section offers two arguments.
The first begins with the theoretically conservative presumption that it is a condition on the extensional adequacy of a definition of the Rawlsian basic liberties that it should include all the rights and liberties that Rawls included in his lists of the basic liberties. A consequence of this presumption is that the definition used in Premise 1 must, independently of its bearing on the Fundamental Liberties Argument, be abandoned. This point can be illustrated by appeal to the case of freedom of movement. For Rawls (1993, p. 335), "the liberty and integrity of the person" is "violated (...) by denial of freedom of movement"; thus, freedom of movement is a basic liberty. Samuel Arnold (2017, §4.2; cf. Pogge, 2007, p. 87) argues that if a law were enacted that restricted people's freedom of movement to within their metropolitan areas, this would not make it impossible that every citizen should possess, and exercise in a full and informed way, the moral powers in the two fundamental cases. Even if it is very unlikely that, under such conditions, every citizen could be so fortunate as to be able to do this, it does not seem to be impossible. A way to remedy the definition so that it includes freedom of movement as a basic liberty is to weaken it precisely along the lines suggested above: i.e., so that it appeals, instead of necessity, to probability. 17 Given that it is not ad hoc, the putative critic of Premise 1 of the Reformulated Fundamental Liberties Argument must show that the sufficient condition on a liberty's being basic that it employs is too permissive.
The second argument begins by noting that Rawls jumbles up modality and probability when he writes that: a liberty is more or less significant depending on whether it is more or less essentially involved in, or is a more or less necessary institutional means to protect, the full and informed exercise of the moral powers in one (or both) of the two fundamental cases. (Rawls, 2001, p. 113) Necessity, which is a modal notion, does not admit of degrees. Neither does essentiality. For any given liberty, either it is essentially involved in, or a necessary institutional means to protect, the full and informed exercise of the moral powers in the two fundamental cases or it is not. In other words, this is a binary property of the liberty, not a scalar property. The relevant notion that does admit of degrees is probability. To retain Rawls's distinction between those liberties that are essential to citizens considered as free and equal and those that relate to the institutional means for protecting the moral powers in the relevant way, while at the same time respecting the analytical point that necessity does not admit of degrees, something like the following amendment to Rawls's wording would be required: a liberty meets the threshold of significance for being basic if and only if either (i) it is essentially involved in the full and informed exercise of the moral powers in one (or both) of the two fundamental cases (as it is an essential feature of persons, considered as free and equal, that they have this liberty), or (ii) its absence is, other things being equal, and for some people, more likely than not to jeopardize the full and informed exercise of the moral powers in one (or both) of the two fundamental cases.

The Fundamental Liberties Argument is invalid
O'Neill's objection to the Fundamental Liberties Argument is a content objection that the Reformulated Fundamental Liberties Argument, in view of its weaker and more plausible content, avoids. Unfortunately, however, both versions of the argument fall to a formal objection.
The inference from Premises 1 and 2 to the sub-conclusion at line 4 of the (Reformulated) Fundamental Liberties Argument is valid. The other inference, from lines 3 and 4 to the conclusion, embodies a fallacy. If justice requires that P and P, in turn, requires that Q it does not necessarily follow that justice requires that Q. To take an example from retributive justice, suppose that justice requires that some crimes should be punishable by imprisonment. Imprisoning criminals requires that there should be jailors. Now it does not follow that justice requires that there should be jailors: for jailors are required only if criminal acts are actually committed that are serious enough for imprisonment to be an appropriate retributive measure. That there should be jailors is not strictly a deontological requirement of justice: rather, it is a practical requirement that follows from a requirement of justice under certain contingent conditions. There is no direct analogy between this example and the case of distributive justice, for it is the circumstances of justice that are the relevant contingent conditions in the case of distributive justice. 18 Nevertheless, the example shows that the form of inference employed in deriving the conclusion of the (Reformulated) Fundamental Liberties Argument from lines 3 and 4 is invalid. Moreover, the argument obscures the distinction between requirements of justice that are deontological and those that (though they can be moral) are merely practical and it can be seen, thereby, as embodying a fallacy of equivocation. On the assumption that an entitlement is a core basic liberty only if it is a deontological requirement of justice that it be extended to all citizens in every society in which the circumstances of justice obtain, the (Reformulated) Fundamental Liberties Argument does not secure its intended conclusion.
The moral powers are not requirements of justice in the sense, for example, that equality before the law, or the distribution of social primary goods according to the difference principle, are supposed to be requirements of justice. If a certain person never happens adequately to develop them, or adequately to be able to exercise them, then this is not necessarily a matter of injustice, for it need not be due to any social disadvantage. Rather, the moral powers are prerequisites for justice. A just constitution can only be drawn-up, and can only be sustained, thanks to the moral powers. Their continuance is necessary to the stability of that constitution. Given that we are not yet close to living in what Rawls would consider to be a well-ordered society, we require that the moral powers should, if necessary, be developed, and should, certainly, be exercised to take us closer. So, full and informed exercise of the moral powers, while necessary to the continuation, and thus to the stability, of a well-ordered society is also necessary to the formation of such a society. On Rawls's account, persons in a well-ordered society should be able to formulate for themselves the principles of justice; the ability to do so rests upon, or is a special case of, the ability to exercise the moral powers in a full and informed way in the two fundamental cases.
Rawls uses the expression "basic liberties" to encompass two forms of entitlement: inalienable freedoms (such as freedom of thought) and inalienable rights (such as the right to equality before the law). An entitlement is inalienable only if it is: (i) due to all persons, considered as free and equal citizens, in societies in which the circumstances of justice obtain; (ii) not apt for trade-offs, made either by its owner or by anyone else, against other goods. We can see that a right to a measure of workplace democracy is not a basic right if we consider a society in which the circumstances of justice obtain but in which the economic circumstances are somewhat different to those of the societies with which we are familiar. For example, we can image a society in which the working week is significantly shorter than is the case for the typical full-time worker in today's advanced industrial or post-industrial societies. In such a society, a lack of democratic rights in the workplace might not meet the threshold of posing a significant risk of jeopardy to the moral powers or their full and informed exercise in the two fundamental cases. Suppose, alternatively, that legislation granting democratic rights in the workplace was introduced in a current society, but that it came with a proviso allowing employees to trade-in those rights in return for a radical reduction in the duration of the working week, with no corresponding reduction in their remuneration. Such a proviso would, by the lights of the Reformulated Fundamental Liberties Argument, have to be regarded as breaching no requirement of justice provided that the trade-in also effected a sufficient reduction in jeopardy, taking the risk below the threshold, to the full and informed exercise of the moral powers in the two fundamental cases. 19 19 Cf. Vega (2000, pp. 31 40, 43-44). According to Mayer (2001, p. 301), "the decisive issue upon which the debate" about workplace democracy "ought to turn", which has "largely neglected by both advocates and critics" of the right to workplace democracy, is that of whether the putative right is alienable. Even if it were valid, the Reformulated Fundamental Liberties Argument could not secure an inalienable right to workplace democracy as long as factors other than an entitlement to workplace democracy might mitigate against the risk posed, by the undemocratic workplace, to the moral powers and their exercise in the two fundamental cases. Mayer (2001, p. 302) argues that "employee claims to a share of power in the workplace (...) are not inherently inalienable". Mayer (2001, p. 301) is concerned with a more substantial right than O'Neill, namely the right to an equal say in the running of the firm on the part of all workers, but many of the points he makes are of more general relevance, including to arguments intended to establish only an element of workplace democracy (rather than an equal say for workers).
The moral powers and their full and informed exercise in the two fundamental cases are essential aspects of the conception of citizens as free and equal that is integral to Rawls's theory of justice. It is a deontological requirement of justice that no-one who is naturally capable of having and exercising these powers in these ways should, through morally contingent social conditions, be impeded from doing so. It is useful to distinguish between two sorts of social impediments. On the one hand, there are impediments that are intrinsic it is a matter of moral contingency that the undemocratic workplace poses a special threat to the moral powers and their exercise in the two fundamental cases. 20 If humans were immune to having the moral powers and their exercise in the two fundamental cases impeded by undemocratic workplaces, then there would be no need for a right to a measure of workplace democracy. The distinction that is being made here is similar to the deontologist's distinction between the intrinsic features of an act (e.g., that it is an act of promising or of lying) and its extrinsic features (e.g., that it secures, but is not constitutive of, a business contract). The effects of an act are among its extrinsic features. So far as arguments for an entitlement to a measure of workplace democracy that appeal to the moral powers are concerned, the existence of permissible worlds 20 Also, forms of social co-operation that are undemocratic need not, merely in virtue of this fact, be unjust. For example, if the members of an amateur sports team defer to their coach, and give the coach the lion's share of decision-making rights in respect of the team's activities off the field, then this does not, at least on a Rawlsian account of justice, entail that they have attempted to hand to the coach some of their inalienable rights. I owe the example to Iñigo González Ricoy. Social co-operation within economic production need not always be an exception to this. For example, the apprentice may defer, and give over decision-making rights concerning production, to the accomplished practitioner for whom the apprentice works, as sole employee. On a Rawlsian account of justice, this does not entail that the apprentice has thereby attempted to forsake an inalienable right. If it is a practical requirement, in order for a just and stable social order to come about, that possession of the two moral powers and the effective ability to exercise them in a full and informed manner, should be prevalent in society, or (more weakly) if it is practically unlikely that such a social order will come about without this, then this is a good (though not necessarily decisive) moral reason to organize society in ways that will bring this about. Our practical starting position is not the original position, and in order for us to approximate to what Rawls would consider a well-ordered society, the ability for full and informed exercise of the moral powers in the two fundamental cases must be protected and promoted. Existing institutional arrangements, which commonly operate in ways that stifle the moral powers, must be reformed so that the risk of such stifling falls below a certain threshold of significance. 21 Factional interests, for example, hinder and stifle the full development and exercise of the ability to be impartial.
Hence, they hinder the sense of justice. Partly because of conflicting factional 21 Even if it is not true, contrary to Schwartz (1982, p. 639), that "persons' autonomous developments are stunted when their jobs severely restrict their opportunities for rationally framing, pursuing, and adjusting their own plans", it might still be the case that there is a tendency towards, or a significant risk of, such stunting. While it is, at least partly, a matter for empirical investigation, it is plausible that in the hierarchical workplace there is a tendency towards, or a significant risk of, the stunting of the moral powers.
interests, the workplace is frequently a place of strife. Within contemporary societies, the development of workplace democracy, whilst taking some decisionmaking power away from senior executives and shareholders in order to redistribute it to employees, would be likely to result in a decrease in factional interests in production. These interests have traditionally been at the fulcrum of political life, including to the point of civil disorder, in industrialized societies.
They remain relevant, even if less pronounced, in post-industrial, service-led economies. It is a reasonable conjecture, open to empirical investigation by social scientists, that a levelling out of control in decision-making in the workplace correlates with a decrease in factional tensions in both the economic sphere and the more narrowly political sphere, with a corresponding increase in the extent to which people are able to view policy issues in an impartial, rather than factional, manner.
We have seen that while the Reformulated Fundamental Liberties Argument, with its appeal to probability, is not vulnerable to O'Neill's objection to the Fundamental Liberties Argument, neither argument is in fact sound. While the original argument contains a false premise, the reformulated version is, alongside its predecessor, invalid. Nevertheless, the foregoing observations enable the construction of a new argument, as follows, that is an adaptation, designed to support a weaker conclusion, of the Reformulated Fundamental Liberties Argument. Under prevailing social circumstances, ensuring an entitlement to "freedom to take part in decisions about economic production" is necessary to the provision of the conditions under which that risk is kept below the threshold. We have moral reason, under prevailing social circumstances, to ensure an entitlement to "freedom to take part in decisions about economic production".

Argument from Risk to the Moral Powers
(From 1, 2) 5.
We have moral reason, under prevailing social circumstances, to ensure a degree of workplace democracy.

(From 3, 4)
The Argument from Risk to the Moral Powers can be seen as claiming that, other things being equal, the full and informed exercise of the moral powers in the two fundamental cases is put into too much jeopardy in the absence of economic-democratic entitlements, including the entitlement to a degree of workplace democracy. Unlike each version of the Fundamental Liberties Argument, the argument is valid. Critical discussion of its premises would be likely to focus on the issue of where the line is to be drawn regarding what constitutes a reasonable threshold of risk to the moral powers and their exercise in the two fundamental cases and on the plausibility of Premises 2 and 3. The Argument represents a significant advance on the Fundamental Liberties Argument, for, in addition to being valid, its controversial premises are logically weaker and more plausible than those of each version of that argument. Notably, the premises are even more plausible than those of the Reformulated When a just society is one in which citizens can play their full roles as citizens, that society is likely (other things being equal) to be more stable over time than a just society that is less developed in this respect. If an entitlement has positive effects on the stability, over time, of a just society, then there is moral reason to secure that entitlement (as a matter of public policy).
(Hidde n Premise) 22 The reconstruction paraphrases O'Neill (2008, p. 42), makes the content and structure of the argument more explicit and weakens its conclusion. 23 O'Neill couches his premises in terms of democratic participation itself, rather than the entitlement to it. Since the argument's conclusion is about the entitlement to it, in the reconstruction the premises explicitly mention the entitlement (so as to secure the argument's validity).

5.
An entitlement to democratic participation in workplace decision-making enables citizens to play their full roles as citizens.
A just society in which there is an entitlement to democratic participation in workplace decision-making is, thereby, likely (other things being equal) to be more stable, over time, than a just society that is less developed with respect to the ability of citizens to play their full roles as citizens.
There is moral reason to secure an entitlement to democratic participation in workplace decision-making (as a matter of public policy). 24 (From 4,6) The Argument from Risk to the Moral Powers and the Democratic Character argument are two sides of the same coin. While the Democratic Character argument emphasizes the benefits, to moral character, of an entitlement to workplace democracy, the Argument from Risk to the Moral Powers emphasizes the risk of harm to moral character in the absence of such an entitlement. The first emphasizes a positive correlation between two factors and the second a negative correlation between those same factors.
Both arguments establish, at best, that a measure of workplace democracy is a practical requirement of Rawlsian justice. That is, given certain actual social conditions (about which parties in the original position are ignorant), it is 24 At the point at which he presents the substance of the argument, O'Neill (2008, p. 42) does not actually formulate its conclusion. Instead, the conclusion, that an element of workplace democracy is "a requirement of justice", is stated in his abstract (O'Neill, 2008, p. 30), though the element of workplace democracy appears under the moniker "economic democratization". The conclusion in the above reconstruction is weaker so as to render the argument valid and to protect it from the ambiguity of "requirement of justice" that was discussed earlier in this section. required if a just social order is to be reached and, once established, to be stable over time (assuming the continuation of some of those social conditions). In so far as these arguments support an entitlement to workplace democracy it is only one that is apt to be secured at the legislative stage in the four-stage sequence. 25 Thus, this entitlement is not on a par with such core basic liberties as freedom of thought and equality under the law. Nevertheless, Rawlsian considerations of justice, along with certain seemingly plausible empirical hypotheses about the interplay between institutional arrangements in the workplace and the full and informed exercise of the moral powers in the two fundamental cases, provide a strong case for favouring an element of workplace democracy over none.
Moreover, Rawls's theory of justice allows, at the legislative stage of the four-stage sequence, and at subsequent stages, for entitlements that are merely practical requirements of justice to serve the functional role, within a just social order, of being rights apt for protection under the law. While such rights are of lesser moral and political significance than are the core basic liberties, their significance may still be profound.

Conclusion
The arguments considered in this article do not show that a degree of workplace democracy is a deontological requirement of justice in a well-ordered society. However, there seems to be a strong Rawlsian case for the introduction of a degree of workplace democracy into current societies: this is plausibly a practical, and even moral, requirement if a just social order is to come about and be stable over time. Under prevailing social conditions, and alongside some seemingly plausible empirical hypotheses, considerations of Rawlsian justice certainly favour the institutionalization of a measure of workplace democracy and might well require it too, albeit not as a matter of inalienable right. 25 In the case of the Democratic Character Argument, O'Neill (2008, pp. 31-32) acknowledges this. The four-stage sequence consists of (in this order) the original position, the constitutional stage, the legislative stage and the judicial stage. For details and discussion, see Rawls (1971, §31;2001, §13), Wenar (2017, §4.9) and McLeod and Tanyi (2017).