Why does Rawls reject meritocracy?
Why does Rawls reject meritocracy?
John Rawls' A Theory of Justice1 is the most significant work of political philosophy published in the last century. Its professional impact has been profound: it continues to define the framework within which other political philosophers must now work. In addition to this, it has made some headway into the world of practical politics, a remarkable achievement for a work of philosophy published merely thirty years ago, and all the more remarkable given the size, depth and complexity of the book. British politicians such as Anthony Crosland, David Owen and Roy Hattersley have all found something in Rawls' philosophy to invoke2. Yet, for all his work's achievements within both professional and political sphere, Rawls has not had the impact of a Hayek or a Friedman, for we seem to be further from Rawls' vision now than we were when his book was first published.
This is particularly true with respect to Rawls' views on meritocracy. A meritocratic society is explicitly rejected by Rawls as a solution to the rational choice problem posed in the original position (§17), yet "Gordon Brown's response to [Rawlsian ideas] is a reformulation of the idea of equal opportunity"3, or a system of Liberal Equality (§12), in Rawls' terms, in which meritocracy is an important element. Thus, given the political saliency of meritocracy, I shall present in this essay an examination of Rawls' treatment of meritocracy. To this end, I shall first clarify the idea of meritocracy, then identify and assess whether the position adopted by Rawls and Rawlsian Persons in the Original Position (henceforth POPs) is meritocratic. Finally, I shall identify the reasoning by which Rawls arrive at his generally anti-meritocratic position, and criticise certain assumptions Rawls makes in so doing.
Meritocracy is not used today in the same sense as similar words like theocracy and aristocracy. Those who wish to bring about a theocracy or an aristocracy wish to install in power a certain group who exemplify certain attributes; namely, exemplifying certain religious values or being a member of a certain class. Those who desire a meritocracy do not necessarily seek to install in power those who are the most meritorious. First, public office in a meritocracy may not place any restrictions on entry to public office, unlike an aristocracy, for example, where entrance to public office is restricted to a specific group of people. In a meritocracy, conditions of formal equality of opportunity exist. Second, those who desire a meritocracy place emphasis not on the ruling class, but on the on the meritocratic nature of the basic structure, and whilst it may be desired and probable that in a meritocratic society those who rule will be the most meritorious, this is not necessarily so4.
Having thus said what a meritocratic society is not, i.e., not analogous to a theocratic or aristocratic society, it is necessary to give an exact definition of what it is. A general definition is that each person should be rewarded according to his or her merit. Differences between meritocrats about meritocracy will therefore concern either the reward schedule or the basis of merit.5 Justification for meritocracy is either that meritocracy, as a partial system of distributive justice, realises society's maximum output, or that merit supports claims of desert[CJH1].
The most famous definition of merit is one used by Michael Young in his book Rise of the Meritocracy6. Young defines merit to be intelligence plus effort. In our society, this definition of merit has some intuitive support. We admire those with intelligence, and we generally reward it. If an example of this is needed, then consider: if two candidates interviewed for a position, one being more intelligent than the other, then ceteris paribus, one would give the job to the candidate with greater intelligence. Similarly, we admire those who make great efforts to achieve their goal, but without the requisite intelligence to perform a task, that effort is stultified. It is the conjunction of these two traits that is rewarded in our society. Of course, this definition of merit is utterly dependent on the nature of society: Sandel remarks that "a hunting society... rewards fleet-footedness over loquaciousness (as prized in a litigious society)"7.
The reward schedule is another area for disagreement. "To each according to his/her merit" does not specify the precise mechanics of accordance, of whether large differences in merit should permit large differences in reward. A minimal definition of a meritocratic reward schedule can be stated in the following ordinal terms: for all persons in society, the most meritorious person should receive more than the next most meritorious person, who should receive more than the next most meritorious person, and so on, for all n persons in society8.
Now that a meritocracy has been defined as a society in which each person is rewarded according to the merit, with merit being defined as a contingent product of the nature of society[CJH2], and the reward schedule being defined so that ordinal merit corresponds to ordinal reward, we may now assess to what extent Rawls' position is meritocratic.
It is when considering different formulations of the second principle that Rawls comes closest to a meritocratic society, or at least a society with strong meritocratic elements. The System of Natural Liberty (where careers are open to talents and ...
This is a preview of the whole essay
Now that a meritocracy has been defined as a society in which each person is rewarded according to the merit, with merit being defined as a contingent product of the nature of society[CJH2], and the reward schedule being defined so that ordinal merit corresponds to ordinal reward, we may now assess to what extent Rawls' position is meritocratic.
It is when considering different formulations of the second principle that Rawls comes closest to a meritocratic society, or at least a society with strong meritocratic elements. The System of Natural Liberty (where careers are open to talents and society is pareto-efficient) is a strongly meritocratic conception of society, since firstly, jobs are open to those with merit (Rawls' usage of talents seems to be equivalent to those with merit), and secondly, there is no redistributive element to upset a distribution patterned according to merit.
Now it is obvious why this interpretation of the second principle is not very useful to those in the original position, since the minimum outcome in this society might be very low indeed. However, that does not preclude this interpretation of the second principle being used as a principle of justice with other principles, as Rawls in fact presents in §21. If the intent is to make this interpretation more palatable to POPs, then we may wish to have the following principles of justice: first, the principle of greatest equal liberty, and second, the principle that careers be open to talents, subject to the constraint that a certain social minimum be maintained. However, Rawls' intent is not to make the Natural Liberty interpretation more palatable, but rather to discredit, and to this end he finds something in the System of Natural Liberty that is objectionable:
"[T]he initial distribution of assets for any period of time is strongly influenced by natural and social contingencies. The existing distribution of income and wealth, say, is the cumulative effect of prior distributions of natural assets- that is, natural talents and abilities- as these have been developed or left unrealized, and their use favored or disfavored over time by social circumstances and such chance contingencies as accident and good fortune. Intuitively, the most obvious injustice of the system of natural liberty is that it permits distributive shares to be improperly influenced by these factors so arbitrary from a moral point of view."9
Thus, that distributive shares are improperly influence by moral arbitrary factors is what is so objectionable about a system of natural liberty. The notion of moral arbitrariness is something I shall return to later, though for the moment I assume that a sufficient understanding of the term will allows us to proceed to Rawls' next construal of the second principle, that of liberal equality.
The system of liberal equality adds to the system of natural liberty the principle of fair equality of opportunity, or that "those who are at the same level of talent and ability, and have the same willingness to use them, should have the same prospects of success regardless of their initial place in the social system"10. Even in the best case, fair equality of opportunity would only preclude a continued accumulation of natural assets. To use a metaphor, fair equality of opportunity would reset the scores after each round of the game had been played, but would not preclude vast disparities between players in each round of the game itself. Thus, as Sandel remarks, "the same reasoning that leads us to favor a 'fair meritocracy' (as in liberal equality) over a purely formal equality of opportunity (as in natural liberty) naturally leads us further to seek what Rawls calls the democratic conception"11. This democratic conception acknowledges that there are natural inequalities in skills and ability, but that since it is practically impossible to even out these inequalities, we should turn them to good use instead, or, to quote from Rawls, we can arrange the basic structure of society "so that these contingencies work for the good of the least fortunate"12. This gives rise to the Difference Principle:
"The higher expectations of those better situated are just if and only if they work as part of a scheme which improves the expectations of the least advantaged members of society."13
Rather than meritocratic, this is a strongly egalitarian principle (though not as egalitarian as ensuring equality of outcome). It is anti-meritocratic in two senses: first, in characterising as unjust certain expectations of those with merit, this principle will give rise to a society that is anti-meritocratic in character. Second, it violates the principles of reward according to merit. Consider the following situation.
State u
State v
State w
Individual a
5
8
1
Individual b
5
8
8
Individual c
5
6
6
Individuals a, b, and c form a society. There are three possible states for a, b and c, and they are states u, v, and w. The numbers in the table represent cardinal values for each individual's expectations under each possible state. a's merit is greater than b's, whose merit is in turn greater than c's. Whilst a change from state u to state v is possible, since it benefits c, the least advantaged member of society, a change from state v to state w is disallowed by the difference principle, since it does not benefit c14. This violates the reward principle, since merit(a) is greater than merit(b), which is greater than merit(c), but reward(a) is not strictly greater than reward(b).
This of course assumes a strong version of the reward principle, wherein reward must be ordered on the basis that each individual reward is strictly greater (less) than the next entry. But the redistributive consequences of Rawls' theory also violate the reward principle. Rawls views as corollaries of the principles of justice "taxation of inheritance and income at progressive rates", partly to secure a "social minimum"15. This redistribution attempts to ameliorate distribution patterned according to merit, and so is anti-meritocratic.
Of course, Rawls' scheme is meritocratic in certain ways[CJH3]. However, it is difficult to see how these meritocratic elements could be avoided within the modern liberal democracy Rawls intends to justify. Rawls never claims to eliminate natural differences, merely to mitigate their effects. "The naturally advantaged are not to gain merely because they are more gifted, but only to cover the costs of training and education and for using their endowments in ways that help the less fortunate as well... no one deserves his greater natural capacity... but... this is no reason to ignore, much less to eliminate these distinctions"16. We can see the difference between the meritocrat and the Rawlsian here: while both acknowledge that inequalities in distribution may be patterned largely in accordance with merit, one claims that this is just because it is deserved, the other claims that it is just because it benefits the least advantaged. Thus, whilst certain extensionally meritocratic principles are employed in a Rawlsian society (e.g. recruitment decisions, increased remuneration for more demanding jobs), their justificatory basis is so radically different that they are intensionally anti-meritocratic principles, and Rawlsian society is thus strongly anti-meritocratic.
This same anti-meritocratic society is the result of principles chosen in the original position from a list of alternatives, one of which was Rawls' two principles of justice, the second principle interpreted as meaning that there be fair equality of opportunity, and distribution regulated according to the difference principle. I shall not examine whether more meritocratic principles of justice might have been chosen in the original position. Rather, I shall argue that in unfairly judging an interpretation of the second principle to be unjust, Rawls has thus unfairly restricted the list of alternatives, and so the rational choice problem he presents to participants in the original position is restricted in its possible outcomes. Of course, Rawls does not make the claim that his two principles of justice would be chosen from any list of principles of justice, since it is conceivable that "for each conception of justice there is another that is better"17. However, given that the missing alternative is another interpretation of one of Rawls' own principles, I believe that this omission is a grave one.
As mentioned above, Rawls considers a strongly meritocratic interpretation of the second principle in considering a system of natural liberty. What Rawls finds objectionable about a system of natural liberty is that distributive shares are improperly influenced by factors that are arbitrary from a moral point of view. It is not altogether clear what "arbitrariness from a moral point of view" is supposed to mean. It cannot mean that they have no bearing on moral considerations, for certain characteristics which Rawls might wish to view as morally arbitrary certainly do have a bearing on persons moral considerations. For example, if a person is of above average intelligence and goes to university to study moral philosophy, then his/her intelligence will, one hopes, have a bearing on his/her moral actions. Perhaps what Rawls means is that although these factors influence moral decisions, they are extrinsic to the problem; that, although intelligence, determination or strength might inform a moral decision, a person's natural talents and abilities are irrelevant as to whether we 'do the right thing by him/her'.
This claim for moral arbitrariness is in direct opposition to claims of desert based on merit. That merit should be rewarded on the basis of desert is a far more contentious claim than merit should be rewarded because doing so maximises society's productivity18, but is the most commonly used justification for a meritocracy. Consider the following argument for reward on the basis of desert.
1. People deserve their natural assets
2. If people deserve x, then they deserve any y that follows from x.
3. People's holdings flow from their natural assets.
Therefore,
4. People deserve their holdings
5. If people deserve something then they ought to have it.19
Of course, this argument is not internal to Rawls' argument, since it uses as premises certain principles which have not been agreed to by POPs, specifically premises 1 and 5. However, this is itself is not a reason to discard this reason completely, since these premises, if true and in accordance with our considered judgements, may suggest that we have not yet reached a reflective equilibrium, and so need to recharacterise the original position so as to lead to principles of justice more in accordance with our considered judgements.
Rawls' rebuttal of this argument starts and ends with contesting the first premise, that everyone, or anyone, deserves his or her natural assets. This premise is of course false, since it is indeed true that no one deserves his or her natural assets, for whenever we say that a person deserves something, the question naturally follows, what is the basis for this desert? The bases of desert in our society depend on prior action and publicly recognised rules dealing with that prior action. I deserve to be paid, because I have worked, and my contract stipulates that for the work I have done, I deserve to be paid a certain amount. For non-legal claims of desert, we can still refer to a basis of desert that makes reference to publicly recognised rules which are informal20. Yet in the case of natural assets, there is prior action upon which to base a claim of desert, nor any publicly recognised rule which makes reference to these prior acts.
Rawls' error then is not in contesting the first premise of the above argument, but rather in his insistence that the bases of success are predetermined by natural contingencies. There is no space in Rawls' concept of the person for non-determined characteristics which may form claims of merit or desert[MH4]. This determinism is not solely a genetic determinism, but includes facts of socio-economic and family background as determining factors. Consider the following sentence:
"The extent to which natural capacities develop and reach fruition is affected by all kinds of social conditions and class attitudes. Even the willingness to make an effort, to try, and so to be deserving in the ordinary sense is itself dependent upon happy family and social circumstances"21.
Rawls here ascribes the basis of desert, as I have done above, to prior action, specifically effort used to realise one's natural assets. Yet with the same stroke he denies that that basis of desert can be independent of determining factors, and so truly come from that person, not, as it were, by genetic or socio-economic proxy, but from non-determined characteristics.
Let us then introduce a second argument for reward on the basis of desert.
. People deserve what comes from their hard work and effort
2. To realise one's natural assets requires hard work and effort.
3. People deserve their natural assets as realised.
4. If people deserve X, then they deserve any Y that flows from X.
5. People's holdings flow from their natural assets.
6. People deserve their holdings.
7. If people deserve something, they ought to have it.
Rawls' objection might be as follows: one's natural assets as realised flow from one's natural assets. Yet one's natural assets are not deserved, and so one does not deserve one's natural assets as realised. One may counter this objection with a remark from Nozick, that "it is not true, for example, that a person earns Y (a right to keep a painting he's made, praise for writing A Theory of Justice, and so on), only if he's earned (or otherwise deserves) whatever he used (including natural assets) in the process of earning Y. Some of the things he uses he just may have, not illegitimately. It needn't be that the foundations underlying desert are themselves deserved, all the way down"22. However, this seems open to the objection that if we need not have foundations underlying desert that are deserved all the way down, then how far down must we deserve what underlies our desert?
Instead, I argue that even if we do not deserve our natural efforts as realised, Rawls' arguments do not undermine the counter-argument that I deserve the fruits of my labours. This principle of desert is surely one of our considered judgements, and non-concordance with the two principles of justice surely means that the two principles of justice do not represent a reflective equilibrium. Where Rawls is, I believe, correct in arguing against merit as a basis for reward is that one cannot divide up the fruits of my natural assets as realised and apportion some part to the effort I put in to realise those assets, and so deserve 'in the ordinary sense', and apportion some other part to natural contingencies. Thus, Rawls' anti-meritocracy is based on an understandable concern about the effects of natural contingencies on reward. However, this concern is taken too far, and requires us to accept that there is no such thing as merit, since what we may think to be meritorious is in fact determined.
Rawls, John, A Theory of Justice, originally pub. 1971, Harvard University Press: page references in this essay will refer to the revised edition, pub. 1999, OUP.
2 Rogers, Ben, "Portrait: John Rawls", Prospect Magazine, www.prospect-magazine.co.uk/highlights/portrait_johnrawls/
3 Gray, John, "Goodbye to Rawls", Prospect Magazine, www.prospect-magazine.co.uk/highlights/goodbye_rawls/
4 There may, for instance, be confusion over the criteria of merit. Or, if a process exists designed to ensure offices are filled according to merit, it may be imperfect.
5 This distinction is used by Norman Daniels, in his article "Merit and Meritocracy", Philosophy and Public Affairs, Vol. 7, Iss. 3, p. 208
6 Young, Michael, Rise of the Meritocracy, pub. Thames & Hudson, 1958
7 Sandel, M. Liberalism and the Limits of Justice, p. 76
8 In terminology used by Nozick, this give rises to a patterned distribution. In Anarchy, State & Utopia, pub. 1974, Basil Blackwell, Nozick shows that, unless 'capitalist acts between consenting adults' and gifts are forbidden, then pattern distribution will always require some redistributive element to correct for the effects of transfers not in accordance with the desired pattern. This and Nozick's "Wilt Chamberlin" example give strong reasons why any patterned distribution criteria may never be perfectly satisfied. However, I shall assume for simplicity that transfers between individuals in society will have no effect on whether or not the reward schedule is satisfied.
9 Theory, §12, pp. 62 - 63
0 Theory, §12, p. 63
1 Sandel, p. 69
2 Theory, §17, p. 87
3 Theory, §13, p. 65
4 Such a change would be accepted by Amartya Sen's lexical difference principle: "where two social states are to be compared, rate more highly the one in which the position of the worst-placed is better; if the worst are equal then rate more highly the social state in which the position of the next worst-placed is better; etc,". Sen, A., Collective Choice and Social Welfare, pub. 1970, Oliver & Boyd, quoted in Shaw, P., "Rawls, The Difference Principle and Equality", Philosophical Quarterly, Vol. 42, Iss. 166, p. 74
5 Theory, §43, pp. 243 - 245
6 Theory, §17, p. 87
7 Theory, §21, p. 106
8 Vide Daniels, "Merit and Meritocracy", in which Daniels argues that "claims of merit, in the restricted sense of that term relevant to meritocracies, are derived from considerations of efficiency or productivity and will not support stronger notions of desert."
9 This argument taken from Nozick, R. Anarchy, State & Utopia, pp. 224 -5
20 We might take games as an example of non-legal rule-bound systems. However, we must be careful of claims such as "The losing team deserved to win", since what this statement expresses is not a claim of desert, but rather a comparative evaluation of the two sides with reference to certain standards of good or skilful play.
21 Theory, §12, p. 64, italics added.
22 Nozick, R., Anarchy, State & Utopia, p. 225
[CJH1]this will need to be clarified if the linkage between merit and desert is similarly to be clarified.
[CJH2]doesn't make sense - perhaps merit is related to efficiency - perhaps some productivity reference, like maximizing primary goods. must not presuppose a theory of the good.
[CJH3]Add something about legitimate expectations.
[MH4]Could I reformulate 'desert' in terms of 'merit' - does merit refer to a particular type of desert? Explain the connection, because it's rather late in the essay to be introducing a completely new concept.
3282 WORDS 455.DOC
