By David O'Brien

Community, Equality, and Value Pluralism in G.A. Cohen’s Why Not Socialism? 1

Winner of the Outstanding Graduate Paper Award at the 57th Annual Meeting of the Florida Philosophical Association

David O’Brien, Florida Atlantic University Introduction

Among G.A. Cohen’s many disagreements with John Rawls’s theory of justice is a macro- level dispute about the structure of political values. 2 An important and illustrative instance of that dispute is Cohen’s and Rawls’s differing views about the relationship between the values of community and justice. Rawls regards justice as the primary social value and thinks it a happy consequence of his theory that it is consistent with the value of community. A Rawlsian society, in other words, not only is just but is also an intrinsically valuable kind of community. 3 In Cohen’s different and more pessimistic view, political values do not fit together so neatly: it is possible for a society to be just but not to affirm the value of community. Indeed, Cohen thinks it a general truth that political values present themselves to us “in competitive array.” 4 Upholding one fundamental political value may often entail disaffirming another, distinct value.

In the teeth of that unhappy situation, the correct way to proceed, in Cohen’s view, is to articulate as principles our moral intuitions, acknowledge what conflicts there may be between those principles, and make the best practical trade-offs between them that we can. Cohen (2009) gives such an articulation of two principles that Cohen regards as encapsulating socialist values—a principle of equality (or egalitarian justice) 5 and a principle of community. And, in line with his just- advertised views on the structure of political values, Cohen countenances potential conflicts between the principles of equality and community.

In this paper I argue that Cohen’s formulations of the principles of community and equality in Cohen (2009) are in tension with his views about the structure of political values. More particularly, I claim that, given his formulations of community and equality, Cohen faces the following dilemma: either the values of community and equality are conceptually inconsistent, or else the two values are conceptually compatible. By two values’ conceptual inconsistency, I mean that, as a matter of logic, their demands cannot simultaneously be upheld. The two values of self-determination and slavish obedience to another, for instance, are conceptually inconsistent. By two values’ conceptual compatibility, I mean that the two values are not unrelated or distinct from each other. To take a (trivial) example, equality of political status and equality of economic welfare are not unrelated values; they are particular instances of the more general value of equality.

I say that the just-explained disjunct is a dilemma for Cohen because accepting either of its horns carries some cost for him. If Cohen were to accept the conceptual inconsistency of community and equality, his vision of socialism would be significantly less attractive, because socialism would then suffer not only from practical but also from logical difficulties of realization. If, on the other hand, Cohen were to accept the conceptual compatibility of community and equality, his views about the structure of political values would require revision, because it would be the case that not all political values are mutually distinct. I shall also argue further for a felicitous consequence of accepting the dilemma’s second horn, which is that the suggested conceptual compatibility between community and equality suggests a way in which Cohen’s worries about practical conflicts between the two values might be resolved. So while accepting the dilemma’s second horn is bad news for Cohen’s more general commitment to pluralism about political values, it is good news of a sort for Cohenian socialism.

By way of preliminary to my argument, I discuss in section I and II Cohen’s formulations of the principles of community and equality, and the conflicts he countenances between them. In section III, I show how Cohen’s formulations of community and equality produce the dilemma that is the subject of this essay, and in section IV, I show the felicitous consequences of accepting the second horn of that dilemma. Section V illustrates the interest of the arguments of sections III and IV by applying them to a puzzle about (Cohenian) socialist society that has recently been raised by John Roemer. Section VI concludes the essay with a brief discussion of the limited implications of my argument for value pluralism more generally.

I. In Cohen’s formulation of it, community obtains when two conditions are met. First, the citizens of a socialist community care about their life prospects not varying too greatly from each other: that is communal solidarity. 6 If people’s prospects are different, people will have very different powers to control the course of their lives. When those with less control over their lives face troubles, the knowledge that those with greater control could help them, but do not, precludes community between the two groups—hence the necessity of communal solidarity for community. Inequalities that result in greatly differing life prospects, therefore, will be forbidden by a principle of community which reflects the communal solidarity condition. 7 In particular, Cohen has in mind large-scale lotteries, including markets, since the outcomes of such markets, or such lotteries, are community-threatening inequalities. 8

Communal solidarity alone is not enough to exhaust socialist intuitions about the good of community. The second condition that must be met, in Cohen’s formulation, is that people be motivated in their economic actions, not by a desire for their own gain, but by their valuing of a serve-and-be-served conjunction: that is communal reciprocity. 9 The necessity of communal reciprocity is explained by a need for “human relationships to take a desirable form.” 10  In particular, the existence of markets poses a special problem for the form of people’s relationships. When people interact via markets, those interactions are characterized by motivations of greed and fear; 11 markets both presuppose and foster those twin motivations. If people’s economic interactions are so governed—if they see themselves and each other as the objects of greed and fear—they cannot, Cohen claims, be in community with each other. A norm of communal reciprocity removes the need for the existence and fostering of market motivations, and such a norm is therefore a necessary condition of socialist community. 12

Community as communal solidarity and reciprocity is thus perhaps most easily distinguished from other conceptions of community in its focus on people’s material—or at least their economic—relationships. 13  Not all arrangements of economic life are consistent with community. Some inequalities, for instance, whatever their status at the bar of the socialist principle of equality (or, what amounts roughly to the same thing for Cohen, justice), vitiate community, since such inequalities entail certain results about what people do and do not value. Yet community is not just equality, for community could not be revealed, as equality could, by a purely external examination of people’s holdings. That is, community not only places limits on certain objective facts about people’s holdings but also requires the structure of some elements of their subjective relationships with others to take a definite form; socialist community means people hold certain attitudes about themselves and others. Nevertheless, be it that community is not equality, what will be further explored in section IV should already be somewhat apparent: community and equality are not values that exist in splendid isolation from each other.

Cohen’s views on equality having been much rehearsed and discussed, 14  it is not necessary here to engage in extensive reconstruction of his position. For the purposes of this paper, Will Kymlicka’s useful characterization of Cohen’s so-called luck egalitarianism suffices: the socialist principle of equal opportunity is insensitive to people’s choices and sensitive to their unchosen circumstances. 15  Thus, for example, the luck-egalitarian principle permits inequalities of income where those inequalities reflect nothing but people’s differing work/leisure choices, because such choices do not, or at least do not in the normal case, form part of people’s unchosen circumstances. By contrast, the socialist principle of equal opportunity forbids inequalities of income that reflect some people’s superior innate talents, since such talents are part of people’s unchosen circumstances.

Taken together, then, the values of community and equality are claimed to be sufficient to specify a just, socialist society of an attractive character. I turn now to examine why Cohen is not sanguine about the consistency of the two principles in which those values find expression.


Cohen (2009) contains a rather deflationary speculation about the relationship between community and equality. Since, as was seen in section I, those are the two values necessary to specify a socialist society, Cohen’s worry about their consistency is pressing. If principles that express those values are inconsistent with each other, then the idea of a socialist society appears to lose some of its attractiveness, since the socialist society would then not in general be characterized by, what its simpler camping trip 16  model normally can be, the simultaneous upholding of the two values.

It is important, therefore, to be clear about what kind of conflict between community and equality is envisaged, for there is more than one way to interpret the idea that two values conflict with each other. What, exactly, is the form that conflict between the two values might take? Cohen’s worry is that:

Certain inequalities that cannot be forbidden in the name of [equality] should nevertheless be forbidden in the name of community. But is it an injustice to forbid the transactions that generate those inequalities? Do the relevant prohibitions merely define the terms within which justice will operate, or do they sometimes (justifiably?) contradict justice? I do not know the answer to that question. (It would, of course, be a considerable pity if we had to conclude that community and justice were potentially incompatible moral ideals.) 17

Thus, Cohen’s picture might be recast in the following way. The set of inequalities form a province that is governed by the two principles of equality and community. Conflicts between those two principles can arise when an inequality which the equality principle permits would be forbidden by the principle of community, because the two principles, reflecting different values, might recommend different courses with regard to the same inequality. Because the problem inequality passes muster with the equality principle, it is permitted by that principle; however, because it puts pressure on community, it is forbidden by the community principle. Therefore there is a conflict: which principle should be adjudicate the dispute? Cohen’s value pluralism, which takes both values to be fundamental and distinct, precludes a general solution that privileges one value over the other.

One can analyze a little more precisely the nature of the advertised conflict between community and equality. The two values, we have seen, potentially conflict because the equality principle may leave untouched some inequalities that put pressure on community. But more specifically: those inequalities put pressure on communal solidarity, since inequalities makes life prospects go differently from each other. So the potential conflict between equality and community is more accurately characterized as a conflict between equality and communal solidarity. Cohen makes out no argument that inequalities put pressure on people’s communally reciprocal motivations. 18

Now, there are various forms an answer to Cohen’s worry might take. One solution, biting the bullet, would be to (a) accept the incompatibility, agree that community requires forbidding some justice-permitted transactions, and conclude that no more can be said at the level of ideal theory. However, Cohen also nods in the quoted passage at two solutions that differ from (a) by not accepting the incompatibility of community and equality. One might (b) assign community a lexical priority over equality (or egalitarian justice), or one might (c) assimilate community as a part of justice, and allow that it will conflict with the demands of equality.

Cohen’s value pluralism seems most naturally to fit an (a) solution: values are fundamental and distinct, and only practical adjudications between them may be made. 19  In the Rawls-esque (b) solution, inconsistency between the two values is avoided in practice, since principles of egalitarian justice will only be asked to adjudicate what the principle of community has already allowed. But an argument is then needed, of course, for why community, not egalitarian justice, should occupy the lexically prior position. In the (c) solution, it is accepted that community will require forbidding what equality permits, but suggested that, somehow, overarching considerations of justice will provide a mandate for community’s demands. It is difficult to see what such overarching considerations could be, however, since ex hypothesi they cannot be considerations of either equality or community, and what then is left?

As mentioned earlier, although an (a) solution might be palatable for Cohen, but it would at least somewhat reduce the attractiveness of his socialist vision, as perhaps suggested by the concluding parenthetical remark in the quoted paragraph. Therefore I take it that there is motivation to investigate a fourth kind of solution: a demonstration that community and justice are not potentially inconsistent. Cohen does not countenance such a solution, because he thinks that community and equality are fundamental and distinct values. Thus they cannot, being distinct, be entered into a common calculus, nor, being fundamental, can they be reduced to a third value.

By way of the dilemma I propose in the next section, I hope to outline a solution of the fourth kind. That is, I shall claim that Cohen must either accept that his formulations of community and equality are not even in principle consistent, or that they are compatible in a way which obviates the conflict between them discussed in this section.


The formulations of the values of community and equality, discussed in section I, lead Cohen to worry about the kind of contingent conflict discussed in section II; that is, in some circumstances, the principle of community recommends a different course from the principle of equality, with regard to some candidate inequality. I now begin my argument that Cohen’s formulations invite a worry about conflict between community and equality at a more basic, conceptual level, which conflict results in a dilemma for Cohen.

This dilemma can be seen by considering two situations, the first of which displays a lack of community (from Cohen’s point of view), the second of which does not display such a lack.Situation 1 is Cohen’s own example of people who are paradigmatically out of community. We are asked to imagine a man who ordinarily drives to work in a car, but one day is obliged to take a bus, since his car is unavailable. The driver of the car, Cohen says, might complain to another similarly inconvenienced car driver about how hot, crowded, and unpleasant the bus is. 20 But the car driver cannot so complain to habitual bus passengers. His (commuting) life is so different from the bus passengers’ that his complaint sounds a false note. Why, exactly, can he not complain, given that the bus really is hot, crowded, etc., and that he is experiencing the heat and crowdedness, just as the others are? The reason is that the badness of the bus-riding experience, for him, is its contrast with the usual air-conditioned solitude he enjoys in his car. The badness of the bus trip for the other passengers is against a background of their having constantly to experience the bus trip. Fully stated, their complaint is: “How awful being on this hot, crowded bus is again today.” The car driver’s full complaint is: “How awful being on this hot, crowded bus today.” So the two complaints, when fully stated, are quite different. The car driver cannot make the bus passengers’ complaint; his history of car driving precludes him from having the necessary grounds to advance it. Therefore, since the car driver not only does not but cannotmake the same complaint, he is (to that extent) out of community with the bus passengers.

Now consider a second scenario. By way of introduction to the scenario, recall from section I that Cohen’s principle of socialist equality of opportunity allows for different income/leisure preferences, assuming that those preferences result in a comparable aggregate enjoyment of life. 21 Cohen models the differing outcomes of such preferences via an analogy with a table laden with apples and oranges from which people choose a fixed number of pieces of fruit, of either kind. No one, at the end of the process, has a (comprehensible) grievance on egalitarian grounds against the final distribution. I cannot complain that you have more apples than I, since I had the option of choosing more apples, at the expense of oranges, but chose otherwise. So much for the scenario in terms of equality; I now want to consider a version of that same scenario with an eye to community.

Situation 2: Let us suppose that, in the just-described scenario, you have selected no apples, and I no oranges. 22  I cannot then join in your delight at your oranges’ zest, you cannot join in my complaint about the awkwardness of removing apple skin from between my teeth, and so forth. Our very different choices, within the bounds of the socialist principle of equality, seem to remove us from community with each other. Translated back into real-world terms, my very different life, characterized by a great deal of work where yours is characterized by a great deal of leisure, appears to preclude community between us, since our lives go very differently. This situation thus seems to show that a norm of socialist community is violated even when the norm of socialist equality is fully satisfied. That is, the principles of community and equality now appear not only to face contingent conflict when faced with problematic market-produced inequalities, but also to be in conflict even when there is socialist equality. The values of community and equality, therefore, appear to conflict at a basic, conceptual level.

To avoid the conclusion that community and equality are (not only potentially practically but) conceptually inconsistent values, Cohen would have explain why community is present in situation 2—why, in other words, the apparent lack of community in situation 2 is distinguishable from the real lack of community in situation 1. Let us therefore turn to examine the ways in which life prospects diverge in the two situations. In situation 2, life prospects diverge because socialist equality of opportunity permits different choices which, in effect, add up to equality, all things considered. Now situation 1 might be recast as the outcome of similar choices. My being a bus passenger might reflect my all-things-considered choice to ride the bus but eat expensive foods; your choice to drive a car might reflect your all-things-considered choice to eat less delicious foods but enjoy the air-conditioned solitude of car travel. If that were all the lack of community in situation 1 relied upon, the two scenarios would be equivalent from the point of view of community, and Cohen would be obliged to accept the conceptual inconsistency of community and equality exposed by situation 2.

Now situation 1 is indeed distinguishable from situation 2 on other grounds. Indeed, the polemical point being made relies on the fact that, independently of their choices, the bus passengers cannot drive cars and are obliged to use a bus, whereas the car driver has the habitual freedom to drive a car and also to take a bus as a last resort. Community does not obtain in situation 1 because the differing life prospects in that situation do not merely reflect different choices. Market luck is present in situation 1 as it is not in situation 2, and it is the operation of market luck that has resulted in a community-threatening asymmetry of life prospects. Therefore it is open to Cohen to say that there are ways lives can go differently, innocently from the point of view of community, when, as in situation 2, those differences in life prospects are the result of people’s positive choices. 23

It might be asked what it is about the outcomes of market operations, as modeled in situation 1, that distinguishes them from the differing outcomes of people’s choices, as modeled in situation 2. One’s positive choice reflects only one’s will; my choice to eat an apple is a direct expression of my will. Markets are not like that, because market operations are influenced by factors that are arbitrary from the point of view of the agent—contingent facts about demand, talent, resources, and so forth. Therefore, markets are inimical to community in a way that choices are not: markets are specially inimical to community because they introduce arbitrariness into human relationships in a way that definite choices do not. 24

It was noted in section I that Cohen’s principle of socialist equality of opportunity is also motivated by an aversion to arbitrariness or unchosenness. Therefore, if this reply to the problem posed by situations 1 and 2 is accepted, it entails that equality and community, in its form of communal solidarity, are expressions of the same value: an aversion to unchosenness governing human relationships. Therefore this reply entails that equality and communal solidarity, as Cohen formulates them, are conceptually compatible, being expressions of the same underlying value.

Thus Cohen faces the following dilemma: 25 either community and equality are conceptually inconsistent or they are conceptually compatible. As noted above, neither horn of the dilemma is entirely without cost for Cohen. Accepting the first horn would mean a serious degrading of the attractiveness of Cohen’s picture of socialism; accepting the second means accepting a modified value pluralism about community and equality. I turn now to examine an important, and felicitous, consequence, from Cohen’s point of view, of accepting the dilemma’s second horn.


Recall that it was conflict of a contingent, not conceptual, kind that featured as a worry in the passage from Cohen (2009) quoted at the start of section II. The concern was that the demands of a principle of community might conflict with the demands of a principle of equality, and such possible conflict was a worry because, community and equality being incompatible values, there would be no systematic way to adjudicate such a conflict. I have suggested that Cohen must accept either that the conflict between community and equality occurs at a more damaging (because more basic) conceptual level, or that community and equality are not incompatible values. Supposing that Cohen were disposed to accept that second option, what are the implications for the kind of contingent conflicts between community and equality about which Cohen was seen to be concerned? 26

I believe that the kind of compatibility between community and equality envisaged in the second horn of the dilemma suggests a solution to the problem of contingent conflict. That is, if community (as communal solidarity) and equality are correctly analyzed as emanating from a common intuition—an aversion to what is unchosen governing human affairs—Cohen’s picture of how contingent conflicts between the two values arise may be revised to obviate the conflicts. The solution I suggest is to partition the inequalities that form the domain governed by the two principles. 27  The inequalities that are the proper domain of the equality principle would be adjudicated by that principle, and similarly for those inequalities that are the proper domain of the community principle. 28  This solution is not equivalent to giving one principle priority over the other, which strategy, as explained in section III, is otherwise undesirable for value pluralists and for Cohen in particular. I do not propose, that is, that the community principle pass over the domain of inequalities and forbid some, leaving the balance for the equality principle to adjudicate. Instead, the domain of candidate inequalities is divided, so that the community principle only ever governs some inequalities, and the equality principle only ever governs some inequalities, and there is no intersection between the sets of inequalities governed by each principle.

Now a defensible rule is needed to partition the domain of inequalities such that the partition is not arbitrary; otherwise, it will be mere assertion to say that the principles do not conflict. Section III, by its identification of the distinctive source of unchosenness that puts pressure on community—human-made, as opposed to naturally occurring—provides such a rule. Therefore the domain of inequalities may be partitioned into inequalities the result of human-made sources of unchosenness, which community (as communal solidarity) governs, and inequalities the result of naturally occurring sources of unchosenness, which equality governs. There is no intersection between human-made and naturally occurring sources of unchosenness, and therefore there is no ambiguity about the proper sphere of each principle. Thus the kind of contingent conflict envisaged by Cohen—the possibility of the two principles taking different attitudes towards the same inequality—does not arise, if one accepts the second horn of the dilemma posed in section III, and derives, from the underlying motivation of the principles of community and equality, a rule that allows their range of application to be restricted.


John Roemer has argued recently that Cohen’s principle of equality permits a kind of inequality, unforeseen by Cohen, that is particularly troublesome, because the inequality in question is unacceptable, on socialist grounds, to permit, but unacceptable, on liberal grounds, to forbid. Roemer has in mind the inegalitarian outcomes that result from parents systematically transmitting

values and preferences of a life-influencing nature to their children. Since I think that Roemer’s conundrum may be solved in the terms of the analysis of community, equality, and the relation between them given in sections III and IV, and since I think it is illuminating of my claims to apply them to Roemer’s puzzle, I propose to consider Roemer’s special case in this section.

In effect, Roemer thinks that Cohen’s picture of a socialist society faces a dilemma. He asks how the society’s principles should adjudicate the special case of parents passing to their children certain values and preferences which greatly affect the course of their children’s lives. So, in Roemer’s example, coal miners might pass on to their children a love of coal mining, bankers a love of banking, and so forth. 29  To forbid parents passing such values to their children would, Roemer thinks, constitute an unacceptable interference with parents’ lives. Yet if the values and preferences pass the bar of the socialist principle of equality, they result in lives going very differently, perhaps even across generations. In short, class structures seem to re-emerge in Cohen’s socialist society, a situation which would threaten community (as communal solidarity). Thus the dilemma: either the socialist society makes intrusive demands upon family life or it accepts a greatly reduced communal solidarity. Since Roemer thinks the first option unacceptable, he seems, albeit unhappily, to accept the second. 30

I think that Roemer’s dilemma may be profitably viewed in the light of the analysis I have given of community and equality. Suppose we stipulate that the problematic values and preferences are unchosen, that they cause lives to go differently, and yet are not forbidden for the reasons Roemer advances. Denying the second horn of Roemer’s dilemma, I claim now that they need not be thought to put pressure on community. For, if one accepts the second horn of the dilemma presented in section III, one accepts that lives can go differently, innocently from the point of view of communal solidarity. Differing occupational choices need not be thought to be any more detrimental to communal solidarity than differing work/leisure choices, for instance. Thus, the mere fact of differing choices of occupation—even predictably different choices, even across generations—need not put pressure on communal solidarity. One is only led to the thought that differing choice structures threaten community because those structures often occur in tandem with other troubling features of a capitalistic class structure (e.g., unfreedom) or social structure (e.g., market reciprocity). If Roemer means that there is unfreedom in such class-like entities in a socialist society—dynasties of socialist coal miners, for instance—then one can agree that such unfreedom may be in tension with socialist values without thereby agreeing that community and equality are in tension. Or again, when combined with the effects of markets, such class-like entities might indeed be in tension with community, if dynasties of gold miners, for instance, find themselves the beneficiaries of market outcomes at the expense of dynasties of coal miners. But Cohen has already denied that markets are compatible with community, in its form of communal reciprocity. Since Roemer thinks that markets “will be necessary in any complex economy,” 31 perhaps that is the real ground of his substantive disagreement with Cohen. If that is so, then the analysis of community and equality given above may be helpful in clarifying that markets, not problems associated with occupational choice, mark the crucial difference between Roemer’s and Cohen’s pictures of socialism.


If the argument I have advanced is correct, Cohen need not think that community and equality are incompatible moral ideals, nor need he worry about the potential practical conflicts between them. So this argument carries some limited implications for Cohen’s value pluralism, insofar as it establishes that not all apparently dissimilar values exist in splendid isolation from each other. It should be clear, however, that I have not argued against the value pluralist position in general. It may very well be, as Cohen says, that it is part of the human situation to be obliged to make “discursively indefensible trade-offs” between moral values. 32 I say only that in the case of the two values of community and equality, as those values are formulated by Cohen, that I do not think pluralism is warranted, because of the grounds of compatibility between them argued for in sections III and IV.


Anderson, Elizabeth. “What Is the Point of Equality?” Ethics109 (1999): 287-337. Arneson, Richard. “Justice Is Not Equality.” Ratio21.4 (2008): 371-391.
Cohen, G.A. Rescuing Justice and Equality. Cambridge, MA: Harvard UP, 2008. Cohen, G.A. Why Not Socialism? Princeton UP, 2009.

Knight, Carl. Luck Egalitarianism: Equality, Responsibility, and Justice. Edinburgh UP, 2009.

Kymlicka, Will. Contemporary Political Philosophy: An Introduction. Oxford UP, 2002.

Miller, David. Principles of Social Justice. Cambridge, MA: Harvard UP, 1999.

Rawls, John. A Theory of Justice. Revised edition. Cambridge, MA: Harvard UP, 1999.

Roemer, John. “Jerry Cohen’s Why Not Socialism? Some Thoughts.” Journal of Ethics 14.3-4 (2010): 255-262.

Vrousalis, Nicholas. “G.A. Cohen’s Vision of Socialism.” Journal of Ethics 14.3-4 (2010): 185- 216. Walzer, Michael. Spheres of Justice: A Defense of Pluralism and Equality. Oxford: Martin Robertson, 1983.









  1. This paper benefited greatly from lengthy conversations with Jari Niemi of Florida Atlantic University.
  2. See, for instance, G.A. Cohen, Rescuing Justice and Equality (Cambridge, MA: Harvard UP, 2008) 4-5.
  3. Or in Rawls’s parlance, a “social union.” For more on Rawls’s arguments about how justice delivers the good of community, see John Rawls, A Theory of Justice, Revised ed., Cambridge, MA: Harvard UP, 1999, pp. 456-464.
  4. Cohen, Rescuing Justice and Equality, 4.
  5. For criticism of Cohen’s idea that justice amounts to equality, see Richard Arneson, “Justice Is Not Equality,” Ratio 21.4 (2008): 371-391.
  6. G.A. Cohen, Why Not Socialism? (Princeton UP, 2009) 35-6.
  7. The communal solidarity sub-principle does not entail that what produces such inequalities—the market— ought to be forbidden. It only supports the weaker conclusion that the effects of such inequalities ought not to be allowed persist. Only the second condition of socialist community condemns markets themselves, as well as their resultant inequalities.
  8. It might be asked how markets are like lotteries, or objected that they are not, since lotteries ostensibly involve an element of chance in a way that markets do not. Lotteries select an arbitrary person to receive an advantage that others, not being the arbitrarily selected person, do not receive. Markets, by contrast, seem no more than neutral rules about exchange. The arbitrariness involved in markets, I take it, is that market rewards are a function of what happens to be desired by a relatively larger number of people at a given time. If I happen to possess an asset or talent that is in demand, my rewards are greater, not because of any specially pressing need nor any extra burden that my asset or talent imposes upon me, but because of the chance conglomeration of a great many individual desires. Thus markets introduce arbitrariness into human affairs in a relevantly similar way to the arbitrariness involved in the selection of a lottery winner.
  9. Cohen, Why Not Socialism? 35.
  10. Cohen, Why Not Socialism? 39.
  11. Cohen, Why Not Socialism? 40.
  12. It might be thought that there is a logical caesura at this point of Cohen’s argument. Could one, for instance, agree that the motivations characteristic of market reciprocity are destructive of (socialist) community, and yet disagree that a norm of communal reciprocity is thereby shown to be necessary for community? I do not know the answer to that question, and a general defense of Cohen’s formulation of community goes beyond the scope of this paper, which is a purely internal investigation of the relationship between the two socialist principles of community and equality. However, I think Cohen’s position is at least plausible, given the not inexhaustible range over which economic interactions might vary. If one is not to be motivated either by self- or other-regarding interest, what remains?  
  13. Cohen is also associated with another notion related to community—justificatory community— which features prominently in Cohen’s Rescuing Justice and Equality but not Cohen’s Why Not Socialism? The concept arises in the context of Cohen’s polemic against Rawls as a way of showing what Rawlsian fraternity lacks, since Rawlsian fraternity can obtain without justificatory community obtaining: this is said to mean that Rawlsian fraternity contravenes an “elementary condition” (Cohen, Rescuing Justice and Equality, 32) of community. Thus I take it that justificatory community does not exhaust or describe the value of socialist community but is rather a necessary but basic condition for community—or, better, a basic aspect of community, since it is not just causally necessary for but expresses part of the meaning of community. See Nicholas Vrousalis, “G.A. Cohen’s Vision of Socialism,” Journal of Ethics 14, 3-4 (2010), for a suggestion that justificatory community might be thought a positive criterion of community tout court.
  14. See, for example, Carl Knight, Luck Egalitarianism: Equality, Responsibility, and Justice (Edinburgh UP, 2009) and Elizabeth Anderson, “What is the Point of Equality?” Ethics 109 (1999) for detailed treatment and criticism of luck egalitarian views.
  15. Will Kymlicka, Contemporary Political Philosophy: An Introduction (Oxford UP, 2002) 59.
  16. See Cohen, Why Not Socialism? Chapter 1.
  17. Cohen, Why Not Socialism? 37.
  18. It might be thought that large inequalities do put pressure on communal reciprocity, insofar as those who benefit from the inequalities will have fewer needs that require to be served by others. But even if that is so, such pressure is derivative of the direct pressure that large inequalities put on communal solidarity, so I take it that the equality/communal solidarity conflict is nonetheless primary.
  19. For more on Cohen’s views on the necessity of deferring to practical adjudications (and the possibility thereof), see Cohen, Rescuing Justice and Equality, 5-6.
  20. Cohen, Why Not Socialism? 36.
  21. Cohen, Why Not Socialism? 19-21.
  22. Do not say that this is impossible since everyone must engage in some work and someleisure, and so therefore we can never, in the terms of bus passengers/car driver example, be completely out of the community to the extent that the text to this note suggests. For it might be true that bus and car passengers also still have some community, when considered alongside magnates of industry who take helicopters to work. Total absence or total presence of community, therefore, is necessary neither for Cohen’s nor for my argument.
  23. On this topic, see further section V on John Roemer’s argument about transmission of values and preferences from parents to children.
  24. It might be objected at this point that people can choose to enter markets, and therefore that the outcomes of markets, even if markets are like lotteries in the relevant sense, is not unchosenness that need be of concern from the point of view of community, no more so than the different apples/oranges choices that I have just argued put no pressure on community. This objection says, in effect, that people can choose unchosenness and that, because it is the result of choice, such unchosenness ought not to be considered a source of dis-community. The answer to this objection is that market choices are unlike apples/oranges choices because apples/oranges choices are bounded by a condition of rough all-things-considered equality. No such condition governs the outcome of the choice to engage in market transactions. Again, market choices are specially inimical to community, because, unlike other kinds of choices, which have definite objects, market choices, by their nature, introduce unpredictable unchosenness into human affairs.
  25. Of course, Cohen would not face a dilemma if an alternative, third account of the distinction between situations 1 and 2, from the point of view of community, were possible. I say “dilemma” only because I do not believe there is another resolution of the problem that does not vary Cohen’s formulations of community and equality. It may be an interesting sub-question whether, in the face of the kind of conflict I have exposed, one’s intuitions about socialist or other moral values should be yield to one’s higher-level commitment to value pluralism, or vice versa. Here I am assuming that one’s moral intuitions take priority over the more general value pluralist thesis.
  26. It should not be thought compatibility between values immediately entails a lack of conflict. Even if community and equality are both informed by an aversion to unchosenness, it might be that in trying to counteract one source of unchosenness one could cause another source of unchosenness to flourish. One might think, for example, of trying to remove creases from a shirt. Ironing out the creases in one shirt panel might cause new creases in another panel, even though there is no internal inconsistency in the injunction to iron out creases from all a shirt’s panels.
  27. I am here thinking of the principles as analogous to two functions mapping a domain of candidate inequalities to a range comprising two elements, p (“permitted”) and f (“forbidden”). The alleged inconsistency between the two principles is represented in this analogy by the two functions mapping the same inequality to different range elements. In quasi-mathematical notation: C(x) = p; E(x) = f; p ≠f :. C ≠E, where x is a candidate inquality, Cthe community principle and E the equality principle.
  28. The solution I propose bears perhaps some similarity to the proposal in David Miller, Principles of Social Justice (Cambridge, MA: Harvard UP, 1999), after Michael Walzer, Spheres of Justice: A Defense of Pluralism and Equality (Oxford: Martin Robertson, 1983), to investigate whether “[the range of benefits and burdens] can be categorized in such a way that each category carries with it its own principle of distribution” (25). Miller’s pluralist notion (not here endorsed) of “spheres” of justicehas some parallel in the idea here advanced that community and justice have their proper spheres of application.
  29. John Roemer, “Jerry Cohen’s Why Not Socialism? Some Thoughts,” Journal of Ethics 14, 3-4 (2010): 257.
  30. An alternative way of avoiding the dilemma is to deny its first horn by claiming that children, in acceding to their parents’ values, thereby accept responsibility for them, a solution Roemer attributes to Ronald Dworkin (Roemer, “Jerry Cohen’s Why Not Socialism?” 256). Roemer and Cohen (Roemer thinks) would be uncomfortable with that solution. I do not here take a stand on the question of whether children become responsible for values into which they are born, since I wish to examine whether my analysis provides a new way of viewing the associated conundrum so as to dissolve it.
  31. Roemer, “Jerry Cohen’s Why Not Socialism?” 259.
  32. Cohen,Rescuing Justice and Equality, 4. In deference to other political or moral values, for instance, people might sometimes engage in community-threatening lotteries or market activity. My thesis does not say that community can never come under pressure, nor that egalitarian justice need always be binding. What it does say is that, were such situations to arise, they would be conflicts not between community and equality, but between, on the one hand, community equality, and on the other, some third value.

David O'Brien