Everyone's an (equality) opportunist!
1. Distributive Justice
The concern, scope and task of Distributive Justice
Distributive Justice is a subfield in philosophy concerned with the distribution of benefits and burdens across members in society, more specifically the moral dimension thereof. Works in this field aim to produce a (or a set of) normative principle(s) to govern the distribution of benefits and burdens. Its contributors seek among others to find out what makes given distributions or distributive features morally preferable. They seek to define what the property of distributive justice consists in. Not only do they disagree on what this property consists in, they also disagree on many other dimensions, which we will describe shortly. But what all contributors to this discussion have in common is that they are examining the normative aspect of distributions. To illustrate by contrast, the science of economics is concerned with the positive i.e. descriptive aspect of such distributions. It is concerned with establishing the facts about the causal mechanisms and regularities that actually govern distributions in the world. It tells us that if we pull this lever, then that distribution or distributive feature results, and it does this for various alternative policies. But economics does not tell us which of these outcomes are morally preferable. That is the concern of distributive justice, which seeks to provide normative principles and arguments for which distributive policies are morally preferable. All works fall under the umbrella of distributive justice that have something normative to say about the distribution of benefits and burdens— including those that do not explicitly use the term 'justice' or 'distributive justice'. (Lamont & Favor 2017)
A full theory of justice consists of both distributive and corrective justice, where the first is concerned with the distribution of benefits and burdens, as mentioned, and the latter is concerned with how to respond to past wrongdoings. But the two aspects cannot be separated because on some views, one aspect may be considered a necessary condition for the other. For example, libertarianism claims that a resulting distribution is just only if all the transactions that happened throughout its history of transactions were just. On such a view, past wrongs must be corrected and compensated for i.e. corrective justice must obtain before distributive justice obtains. Thus, to remain substantively neutral, I will not distinguish between distributive and corrective justice in this essay, and I will simply use the term distributive justice or justice tout court interchangeably. There are many conceptions as to what justice even is and means. There are conceptions that define the property of justice itself, directly, i.e. they specify what this property consists in. For example, the conception of justice as comparative fairness requires everyone get the same proportion of whatever it is they are due. The conception of justice as absolute fairness requires each individual get all that they are due, regardless of what others have. The rights-based conception of justice requires that individuals' rights be respected. On this view, justice obtains when there is no violation of anyone's rights. But there are also conceptions which define justice solely by the object or site to which it applies, whilst remaining substantively neutral as to what the property itself consists in. Such conceptions define justice as the property of moral permissibility obtaining in the site to which it applies, for example as the property of social structures and distributions being morally permissible. This conception makes no claim as to what exactly makes a social structure morally permissible, merely that if justice obtains, then this is as a result of the social structure satisfying certain requirements.
Principles and theories of distributive justice disagree on a variety of dimensions. (Lamont & Favor 2017, Vallentyne 2017) As mentioned, they disagree on the definition, the essence, the property of justice itself. They disagree on what object the property of distributive justice should apply to, in other words the appropriate site of distributive justice. Some claim that the property of distributive justice should apply to the actions of individuals, others that it should apply to formal social structures such as legal and political institutions. Others, still, claim it also should apply to the informal social structures of society such as the cultural ethos. (Cohen 1997) "Justice can assess many different kinds of object: actions, the character of agents, social institutions, basic social structures (e.g., constitutions) and distributions of goods." (Vallentyne 2017) They also disagree with regards to the recipients of the distribution, that is, who should be included in considerations of distributive justice, and in what configuration (i.e. as individuals, groups, reference classes or other). They disagree on the appropriate currency of distribution, that is, what to distribute e.g. money, welfare, opportunities, jobs... They disagree on how to distribute the chosen currency among the chosen recipients. Should things be distributed equally, differentially according to merit, to maximize aggregate welfare, or in some other way? Theories of distributive justice also differ in whether they are ideal or non-ideal, whether they theorize without regard for empirical constraints, or whether they do indeed take practical feasibility into consideration and seek only principles of justice that are realistically implementable. (Valentini 2012)
How to assess whether a given theory of Distributive Justice is plausible: Practical Implementability
Choices regarding distributions are inevitable and constantly made. Every individual, group or society inevitably, at all times, instantiates certain distributive principles and thereby performatively endorses and chooses them over others. Even passively continuing with the status quo is an implicit choice, namely that of the current distribution over all possible alternatives. This is why philosophical debates about distributive justice are practically crucial and not merely some theoretical hobby. If we want our choices and performative endorsements to align with our own ideals, we need to clarify and question our normative assumptions and commitments about distribution. Knowing the necessity of this task, how do we adjudicate and choose between the various alternative theories of distributive justice? How do we assess the plausibility of a particular distributive principle? When comparing between alternatives, we need some common standard by which to evaluate them. But the difficulty with this is that different substantive positions often presuppose different evaluative standards for what a plausible theory of distributive justice consists in. The different theories each satisfy their own criteria for a plausible theory, and differ and disagree precisely on these 'metapolitical' criteria of distributive justice. Kuhn (1962) noted this difficulty of incommensurability in philosophy of science, where evaluating across alternative scientific paradigms is in a way theoretically impossible, since each paradigm has its own standards for what a good paradigm consist in, which differ from the standard of other paradigms.
However, despite the differences in evaluative standards between alternative theories, there are at least some criteria for plausibility that do apply to all proposed theories. Internal coherence and consistency is obviously a criterion of plausibility for all theories, in any field. Simplicity and elegance are also desirable, for example Occam’s Razor in philosophy of science. In distributive justice, practical implementability is one such universally desirable criterion— ought implies can. Whatever the distributive ideals we want to behaviorally align ourselves with, they have to be realistic and practically implementable to be relevant to our aforementioned constant and inevitable choices. It makes no difference to our choices to ponder the unchoosable. While the various theories of distributive justice may disagree on many normative-evaluative presuppositions, they arguably agree that practical implementability is an essential criterion for a plausible theory. (Lamont & Favor 2017) So a plausible theory of distributive justice must be non-ideal— it must take empirical feasibility constraints into consideration. Non-ideal theories of course yield less-than-perfect ideals, but this is no objection when the goal is to achieve actual practical improvements, no matter how small. The implication is that any proposed theory of distributive justice that is impossible to implement must be discarded, no matter how good it is 'in theory', and that practical objections are among the most fatal kind of objections to theories of distributive justice. By contrast, theoretical objections are not necessarily fatal to theories of distributive justice, not unless there are better practicable alternatives to the theory being objected to. This is because in seeking practicable solutions, we are seeking the comparatively best solution, i.e. the least of all evils, not the most normatively idealistic principle.
To reiterate, then, practical implementability is an universal criterion for whether a given theory of distributive justice is plausible. It applies to all possible proposed theories of distributive justice, regardless of their differing normative-evaluative presuppositions. In turn, there are various requirements for a theory to be practically implementable. Firstly, the theory's specifications must be clear enough for implementation. Some theories are not yet specified clearly, whilst others do not even seem clearly specifiable in principle. On the other hand, some theories are very clearly specifiable and specified, but the specifications cannot be implemented because they go against the constraints of reality somehow e.g. the laws of physics, economics, human psychology etc. One important subset of the requirement that a theory's specifications actually be implementable is what I will call normative plausibility. Any theory, no matter how clearly specified, is only realistically implementable in a given society if it can gather enough support from the people of that society. After all, people are the principal agents that carry out any given normative framework and system. No political principle, theory or framework can be sustained in a society if there is too much objection to it by its people. Too little consensus and support leads to non-compliance at best, and revolution at worst. In the end, getting the consent and support of the people is what gives a given political regime its authoritative force. So a theory of distributive justice must garner enough support from people, in order to be implemented and sustained (by them) in society. It follows that in assessing any theory of distributive justice for practical implementability (and hence plausibility), one must consider how much support and consensus it can be realistically expected to garner from people.
Normative Plausibility
John Rawls (1985) argued that a political theory can be expected to garner more support from people if it abstains from metaphysical assertions and restricts its axioms to practical rules (that govern social interaction and cooperation). Rawls has written extensively about public authority and why it should have a political rather than metaphysical basis. He argues that a political regime and its authority ought to be based in consensus, rather than in assertions of truths, be they moral, religious, metaphysical or philosophical. Such truths are very hard to change people's minds about, and therefore hard to use as a common (political) foundation in cases where people disagree about them— which is most cases. History is full of metaphysical disagreements that are irreconcileable, enduring and violent. A regime for cooperation cannot be based in irresolvable disagreements. Cooperation requires a bare minimum of agreed upon rules by which to cooperate. Any political theory, if it is to have the slightest chance at succeeding in establishing social cooperation in practice, must sidestep irresolveable disagreements. And this is done by making a consensus of practical rather than metaphysical axioms the basis for political cooperation and system. If people can agree to a minimal set of practical rules for how to interact with each other, their metaphysical disagreements become practically irrelevant. They will consent to live in harmony despite their metaphysical differences, under these shared practical principles.
The practical principles that could make up a minimal working consensus in a given society can be discovered by examining the beliefs and values that people already hold in that society, Rawls argues. Again, it is very hard to change people's minds, beliefs and values. A theory of distributive justice that is very far removed from what people already believe and value will be very hard to garner their support for. By contrast, a conception of distributive justice would be capable of forming a consensus if it aligns sufficiently with the public's existing commitments and assumptions. Its adoption would then be seamless, its persuasive force being anchored in what people already believe. No attempt is made at convincing people of anything— all moral, metaphysical and philosophical arguments and questions are set aside. We do not seek truth. We seek consensus, we seek practical principles that are acceptable to people, regardless of their truth. Of course, the principles cannot only approximate what people already believe and value, as this would be an approach that always sought to preserve the status quo and its prevailing norms. Such an approach would in a sense be purely descriptive (of the status quo) and lack the striving, idealistic and corrective features of a normative theory. A normative theory must be sufficiently anchored in existing norms to remain realistically implementable, but must also strive towards some degree of improvement on the status quo. This is where Rawls proposes the method of reflective equilibrium for balancing between the descriptive and normative, as a method for discovering existing beliefs and commitments, but then improving on them through self-corrective mechanisms of reflection. (Rawls 1971) This method has hitherto been the most clearly defined methodology for evaluating and choosing between alternative normative theories and principles. (Lamont & Favor 2017)
The method begins by extracting people's existing assumptions, moral intuitions, value commitments and understandings from existing political institutions, traditions and culture. Importantly, only considered judgments are included. Considered judgments are moral judgments that are deliberative, complete, and unimpaired by factors that inappropriately compete with moral considerations, such as self-interest, coercion, unwilling ignorance, fear, lust, anger... The considered judgment is made by a person with the "ability, opportunity and desire to reach a correct decision". (Rawls 1971, 48) Attempts are then made to formulate theoretical principles that reflect these preliminary considered judgments. That is, attempts are made to formulate principles the implications of which maximally match the considered moral judgments that people already hold and find plausible. Then begins a reflective loop. The implications of the formulated principles are compared with the moral intuitions that one holds for internal consistency and coherence. If one's principles and intuitions do not match, then either or both must be changed to become compatible. Whichever elements are deemed more justified are kept, discarding or amending the other to become compatible with the ones kept. This process of extracting one's moral intuitions, formulating principles to reflect them, and amending one's intuitions and principles in a reflective loop until they all become consistent, must also be applied to views new and alternative to one's own. That is, one must test one's existing intuitions and principles against alternative intuitions and principles. Here, again, one must decide which options one finds most plausible and justified to keep, and amend all the others to become compatible with the elements one has chosen to keep. Through this process of moral reflection, incremental changes to current beliefs and values are made, which is practicable considering humans' aversion and slowness to change. Reflective equilibrium is the state reached once one has completely ironed out one's inconsistencies and updated one's entire system of normative commitments and beliefs to become coherent and not self-contradictory.1
In summary, any political theory must be practically implementable to be plausible. To be practically implementable, the theory must among others be able to garner enough support from the people of the society in question. A theory can realistically be expected to garner sufficient support for political authority if it is sufficiently close to people's existing moral assumptions and commitments. The theory just needs to apply the reflective equilibrium method to these existing beliefs and values to iron out their internal inconsistencies. This will produce a normative theory that is sufficiently anchored in the status quo to actually be adopted by people, and yet still striving towards improvement enough to be normative. All this applies to theories of distributive justice as well. A given theory of distributive justice is plausible if we can realistically expect it to garner sufficient support from people to be adopted and maintained. We can expect the theory to garner sufficient support if it is sufficiently close to people's existing beliefs and values about distributive justice, and only incrementally improves upon them through the method of reflective equilibrium. I find reflective equilibrium to be a plausible method for arriving at a political theory that is actually authoritative (morally forceful). Many valid arguments exist for and against various alternative theories of distributive justice. But in the end, it comes down to asking ourselves which moral commitments we as a society simply cannot give up (and then ensuring that the rest of our commitments are consistent with these). It is from preexisting commitment and the power of habit and custom that moral force is derived.2
2. Equality of Opportunity & its Variants Defined
The plausibility of Equality of Opportunity from the perspective of Liberal Societies
For the rest of this essay, I will be assessing whether one particular theory of distributive justice is plausible, namely the theory of Equality of Opportunity. In assessing whether Equality of Opportunity is a plausible theory of distributive justice, I will apply the various criteria of plausibility that I discussed in the previous section. I will consider whether it is possible to specify Equality of Opportunity clearly enough for implementation, and whether a clearly specified account of Equality of Opportunity is practically implementable at all. As a subset of practical implementability, I will consider whether Equality of Opportunity is normatively plausible in the Rawlsian sense. I will be assessing whether Equality of Opportunity is normatively plausible as seen from the perspective of liberal societies. Equality of Opportunity would thus be normatively plausible if it is compatible with the considered judgments upon reflective equilibrium of people in liberal societies. Applying the reflective equilibrium method reveals that freedom and equality are the fundamental values broadly shared by people in liberal democratic societies. This result I have borrowed from Rawls’ (1985) historical analysis. Rawls traces the birth of the values of freedom and equality back to the Wars of Religion following the Reformation. The irresolvable metaphysical disagreements of this time produced many violent conflicts, and there was an incentive to stop this constant and pointlessly destructive bloodshed. The solution adopted was to allow people the freedom to their own beliefs, to tolerate differences and to treat everyone equally so that no one felt an injustice or oppression which might incentivize them to war. So the values of freedom and equality— and the various devices implemented to promote them i.e. tolerance, the rule of law (constitutional government), the free market— were practically motivated. They were practical solutions to the irresolveable differences in metaphysical, moral, philosophical, religious beliefs. Ever since then, for the past two hundred years, liberal societies have continued to maintain customs and institutions implicitly committed to and promoting the values of freedom and equality.
This is the consensus of liberal societies from which I begin my examination: Viewing and treating persons as free and equal. The question then becomes narrowed down to: Is Equality of Opportunity a plausible theory of distributive justice, given the practical constraints that exist and given the normative presupposition that persons should be viewed and treated as free and equal? For the latter part, does Equality of Opportunity instantiate the values of freedom and equality, or at least not violate them? In this connection, there is one last complication to mention before we finish with methodological considerations and get to implementation. On the surface, liberals may share the commitment to the slogan of treating people as free and equal. But they actually disagree deeply as to precisely how the values of freedom and equality should be realized. What does it mean, exactly, to treat persons as free and equal? Which substantive practical principles do the values of freedom and equality imply? These very questions are in contention among liberals. A very wide range of substantive positions exist, all of which purport to be committed to and promote freedom and equality. And yet their differences are substantial; they generate political solutions and regimes that are utterly incompatible. (Rawls 1985)
These many different accounts of liberalism, in turn, are what produce the many different accounts of Equality of Opportunity. Equality of Opportunity tout court is very widely endorsed in liberal societies. It seems like everyone is a proponent of Equality of Opportunity. But this consensus is merely apparent. The different variants of Equality of Opportunity are substantially different, and their differences reflect the aforementioned fundamentally different liberalisms which underlie them each. Liberal theories each endorse a variant of Equality of Opportunity that is compatible with their own interpretation freedom and equality, and since these various liberal theories differ so deeply, so must the Equality of Opportunity variants that they each endorse. The existence of all these variants of liberalisms and Equality of Opportunitys implies that in assessing whether Equality of Opportunity is a plausible theory of distributive justice, we have to specify which variant of Equality of Opportunity we are talking about, and which interpretation of freedom and equality, i.e. which liberal theory, we are evaluating it with regards to. In an unqualified sense, adjudicating between different variants of Equality of Opportunity requires adjudicating between the variants of liberalism that underlie them first. That is, the assessment of plausibility must be made at the level of liberalisms before it is made at the level of distributive justice. For example, if certain variants of liberalism have been persuasively refuted, then any variants of Equality of Opportunity that is based on them have also been refuted. But it is beyond the scope of this essay to adjudicate between the liberal theories themselves. My goal in this essay is merely to examine whether the main variants of Equality of Opportunity are normatively plausible as seen from the perspective of the various liberal theories.
The Common Denominators of various accounts of Equality of Opportunity
Before we assess the practical and normative plausibility of the various accounts of Equality of Opportunity, we need to define them first. As mentioned, there are many fundamentally different accounts of Equality of Opportunity. But the different accounts all agree on at least one thing, namely that there exist legitimate and illegitimate determinants of distribution, and that it is possible for unequal outcomes to have been determined by legitimate factors. In such cases i.e. where the outcome was determined by legitimate factors, inequalities are permissible. (Elford 2023) That is, equality of outcome is at the very least not mandated in such cases. (It may be permitted on some variants of Equality of Opportunity, but never mandated.) While there exist self-proclaimed accounts of Equality of Opportunity that identify it with Equality of Outcome, this would be a excessive application of the former term that removes all of its meaning. So we set such accounts aside, and continue on the premise that all accounts of Equality of Opportunity, despite their differences, agree in rejecting Equality of Outcome. To reiterate, the common claim across all variants of Equality of Opportunity is that inequalities in outcome are permissible, so long as they were not influenced by inappropriate factors. Equality of Opportunity obtains if and only if the resulting distribution is free of inappropriate influences. Such factors are considered obstructions— factors that cause unequal opportunity. Obviously, to define Equality of Opportunity in terms of 'inappropriate factors' is so vague that it is basically endorsable by all positions. The fact that almost everyone in liberal societies endorse Equality of Opportunity is a merely apparent consensus. Substantial differences between various accounts of Equality of Opportunity reveal themselves in exactly which factors they each consider appropriate and inappropriate influences on distributions.
There is, however, one factor which they all agree is a legitimate influence on outcomes: agent accountability. (Dworkin 1981, Lamont & Favor 2017, Vallentyne 2017) Accounts of Equality of Opportunity agree that distributive justice requires that the resulting distribution be sensitive to people’s choices somehow. There is a shared ethos that whilst factors 'arbitrary from a moral point of view' (Rawls 1971, 72) ought to be counteracted,3 morally salient factors ought precisely to be preserved, otherwise, Equality of Opportunity collapses into Equality of Outcome. The various accounts disagree on which factors distribution should hold people accountable for. Should the distribution be sensitive to people's involuntary tastes and preferences, voluntary choices, mere efforts and attempts, completed actions, or actually produced outcomes? Let me put all of these views that endorse some form of moral accountability or other under the umbrella of choice-sensitivity, for short. Commitment to choice-sensitivity implies that any distribution that is not choice-sensitive must be considered unjust. For example, redistribution that counteracts people's choices is unjust. The assessment of choice-sensitivity is necessarily historical, since people's past choices have to be considered when assessing the justice of the current distribution. This means that all accounts of Equality of Opportunity, in their commitment to some form of choice-sensitivity, are necessarily historical theories. Regardless of its specifications, any theory that is committed to historical choice-sensitivity necessarily rejects Equality of Outcome. People have never, do not, and will never behave in the exact same way required to warrant equal outcomes given the requirement of choice-sensitivity. Choice-sensitivity is the main objection that accounts of Equal of Opportunity have against Equality of Outcome.
Equality of Outcome may be objectionable to accounts of Equality of Opportunity for a variety of other reasons. Some accounts of Equality of Opportunity, like the welfare-based theories, reject Equality of Outcome because of the leveling down objection: Equality of Outcome would demand equalization, even in cases where this make things better for 'no one'. (Holtug 2015) That is, Equality of Outcome would mandate the distribution (0,0) over the distribution (0,1), even though the latter is better for at least one person.4 But those who are committed to promoting absolute good for people, as opposed to equality for its own sake (strict egalitarianism), would prefer the latter to the former. Such positions claim that treating people as free and equal requires we duly consider and try to promote their absolute good, not their relative good. Other theories would reject Equality of Outcome solely because it violates procedural rules, and not because strict equality is 'better for no one'. According to libertarianism, for example, the injustice of Equality of Outcome consists in the violation of rights that is inevitably required to realize it, such as redistributions of property. (Vallentyne 2017) The reduction in absolute good is irrelevant to the libertarian assessment that Equality of Outcome is unjust. These different reasons for rejecting Equality of Outcome reflect the deeper differences between the different liberal theories that underlie the various accounts of Equality of Opportunity. That is, they reflect the aforementioned different views on how the values of freedom and equality are realized (e.g. whether freedom and equality are best realized by utilitarianism, libertarianism, desert-based principles, luck egalitarianism or other theories).
Procedure-focused versus Outcome-focused Variants
All of this is not to say that accounts of Equality of Opportunity are not committed to equality tout court. All accounts of Equality of Opportunity require that everyone (included in the recipient scope) equally enjoy a certain relation conducive to advantages. I say a ‘certain' relation, because the precise nature of the relation is also a point of contention between the various accounts. Some variants claim that for Equality of Opportunity to obtain, the relation need only be a lack of interference in one's attempted pursuits of advantages. Others require there be "more than a mere possibility" but "less than a guarantee". Some, like Cohen (1989, 916-917), may distinguish between opportunity and access, and argue that there may also be some requirements of justice regarding access, although different from those regarding opportunity. Here, again, there is a point of agreement despite all their differences: The different accounts all agree that said relation should be less than a guarantee to advantages, as equal guarantee to advantages would basically mean Equality of Outcome, which we already mentioned was rejected by all variants of Equality of Opportunity. These discussions, of whether Equality of Opportunity requires people merely not be interfered with, or have outright positive access to advantages (or something in-between) mirrors Isaiah Berlin's discussion of negative versus positive conceptions of liberty. (Berlin 2019) In fact, this and other disagreements between various accounts of Equality of Opportunity reflect a divide that exists more broadly in many topics of practical, moral and political philosophy: the divide between procedure-focused theories and outcome-focused theories.
Briefly put, these two types of theories are distinguished "according to the relative weight they attach to procedures and substantive outcomes." (Miller 2023) Political theories more generally lie on a spectrum as to how procedure- or outcome-focused they are— this is also the case for Equality of Opportunity variants. But let me define the two extremes for elucidative purposes. Strictly procedural theories claim that justice (or rightness, or freedom or whatever the political notion in discussion) consists solely in the satisfaction of procedural rules. That is, the requirements of justice apply only to procedures. Strictly outcome-focused theories claim that justice consists solely in the realization of certain outcomes, i.e. outcomes that satisfy certain specifications. That is, the requirements of justice apply only to (i.e. can be satisfied only by) resulting states of affairs. This same divide pervades different discussions under different names (though of course the procedure versus outcome divide may be but one among many dimensions of contention in these various discussions). It shows up in the negative versus positive liberty discussion— freedom from interference being a procedural rule, and freedom to actually attain being a rule that applies to outcomes. In normative ethics, the divide is clear in the debate between deontological and consequentialist theories, where the former are procedurally-focused whilst the latter are outcome-focused.
In the Equality of Opportunity literature, these camps go by the names of deontic and telic accounts. The deontic/telic distinction refers to the different views as to whether the requirements for Equality of Opportunity should apply to the distributive procedures in society, or whether they should apply to the resulting distribution in society. On deontic accounts, requirements should apply only to procedures, and Equality of Opportunity obtains, regardless of resulting distribution pattern, if these procedural rules are satisfied. Examples of such procedural rules are non-discriminating hiring and selection criteria. On telic accounts, the resulting distribution must satisfy certain requirements for Equality of Opportunity to obtain. Examples include actually equal probabilities for attaining certain positions among equally talented and hard-working individuals. Deontic rules and telic requirements are distinct and independent criteria that do not imply each other (nor do they necessarily exclude each other). If it is the case that they both obtain, this would be a contingent and not a necessary fact. It is possible for deontic rules to obtain in a society's selection procedures, without telic general Equal of Opportunity to obtain in its resulting distribution. An example of this is meritocratic job markets and hiring practices without equal access to education, training and qualification. Vice versa, it is possible for telic general Equal Opportunity to obtain in a society’s resulting distribution, without procedural requirements for Equality of Opportunity being met in its distributive procedures. An example of this is affirmative action in hiring procedures to promote a more equal resulting distribution. Telic accounts are also called patterned theories of distributive justice, in that according to them, distributive justice requires the resulting distribution instantiate a certain distributive pattern. Strictly deontic accounts are non-patterned, since they claim that distributive justice obtains regardless of the resulting pattern, so long as certain procedural rules are respected. (Elford 2023) As I will argue later, however, there is a limit to the extent to which theories of Equality of Opportunity can be patterned, because they are subject to the requirement of choice-sensitivity.
A last important dimension to mention, along which Equality of Opportunity variants differ, is simply the number of requirements, that is, the number of influences deemed inappropriate that must be avoided or counteracted for Equality of Opportunity to obtain. This is called the formal/substantive distinction, where the more formal the account, the fewer (or narrower) requirements there are, and the more substantive the account, the more (or broader) requirements there are. (Elford 2023) This dimension is illustrated through a comparison between Rawls' Difference Principle and luck egalitarianism. While Rawls only prescribes the counteraction of social factors that are arbitrary from a moral point of view, luck egalitarians also require arbitrary natural factors be counteracted. For example, where Rawls only requires the counteraction of arbitrary socioeconomic inequalities, luck egalitarians also require the counteraction of arbitrary genetic inequalities. (Mason 2006) Luck egalitarianism is basically the variant of Equality of Opportunity that takes the choice-sensitivity ethos to its ultimate logical conclusion. Results are to depend only on choices— absolutely no luck-based factors are allowed to influence the distribution whatsoever. Similarly, desert-based principles are a group of accounts that emphasize the moral significance of responsibility, claiming that the distribution should reward people differentially according to what they deserve. But desert-based theories are less substantive than luck egalitarianism in that they evaluate desert based on actual outputs (e.g. actualized efforts, choices, attempts) regardless of the luck-based factors that may have led to differentials in actual output. That is, desert-principles do not find potentials and 'could-have-beens' morally salient, whereas luck egalitarianism does. Though deontic and formal, and telic and substantive, are not conceptually the same, respectively, these features do tend to overlap in variants of Equality of Opportunity.
3. Liberal Theories and their implications for Equality of Opportunity
Having thus roughly defined the various accounts of Equality of Opportunity, their common denominators and differences, let us assess how practically feasible and normatively plausible they are. I will focus on the plausibility of the broad groups og Equality of Opportunity accounts rather than the more specific variants under each group. That is, I will focus on the plausibility of procedure versus outcome-focused accounts, and that of formal versus substantive accounts. As a reminder of our earlier methodological considerations, the assessment of normative plausibility will be done relative to the evaluative standards of various liberal theories. The idea is to show which fundamental liberal commitments, i.e. which interpretations of how freedom and equality are realized, compel one to endorse either a procedural or an outcome-focused account, a formal or substantive account, or some mix in-between. We mentioned in the beginning that theories of distributive justice differ along many dimensions. They differ in the dimension of essence, i.e. what justice even is and means. They differ in what they consider the appropriate object and site of distributive justice. They also differ in the scope of recipients, the currency of distribution and the manner of distribution. The various liberal theories' interpretations of freedom and equality imply certain substantive positions along these dimensions. In turn, these positions along the various dimensions of justice imply, that is, compel the adoption of certain accounts of Equality of Opportunity over others, on pain of inconsistency. Let us briefly review the main competing groups of liberal theories, explore their positions along these dimensions, and examine the implications as to which variant of Equality of Opportunity they are compelled to endorse.
Consequentialism
There are three main groups of theories of justice that have dominated in liberal societies: consequentialism, libertarianism and social contract theories. (Vallentyne 2017) Consequentialist theories are most fundamentally committed to the view that justice requires promoting the moral good, whatever that is, in resulting states of affairs. Treating persons as free and equal, then, requires that one try promote their resulting good. In prioritizing outcomes, consequentialists tend not to restrict the mandate to promote the good to any particular site, e.g. only to agents' actions, or only to formal social structures. Any means, at any site, which successfully promote the moral good are endorsed.5 Some variants, e.g. act consequentialisms, may mandate the maximal promotion of the good in each and every act, while others, e.g. rule consequentialisms, may mandate following rules that tend on average to maximally promote the good. Most relevant to our question is the fact that consequentialist theories are most fundamentally committed to promoting certain outcomes, which obviously compels them to adopt outcome-focused variants of Equality of Opportunity, if they must so choose. But many consequentialist theories are fundamentally incompatible with Equality of Opportunity on multiple points. Many consequentialist theories are committed to maximizing the aggregate good, and make no requirements as to the manner or pattern of distribution. This is fundamentally incompatible with Equality of Opportunity, because while the different variants may differ substantially, they all require that each individual recipient equally enjoy whatever relation to advantages is specified by that account. All variants of Equality of Opportunity thus consider what is called the separateness of persons: individuals' goods are compared, not the aggregate good of alternative choices. And all variants of Equality of Opportunity do have a requirement for the manner or pattern of distribution, namely that it be equal in some manner (as specified by the particular account).
Of course, not all consequentialist theories have nothing to say about distributive pattern. (Vallentyne 2017) Strict egalitarianism, for example, seeks to maximize relative equality, without regard for the absolute level of good. For example, it prefers the (0,0) distribution to the (1,0) distribution, as mentioned in the beginning. Prioritarianism seeks to prioritize the good of the worse off whilst still maximizing aggregate welfare. Mathematically, this would be implemented through differential weights, with higher weights attached to the welfare of the worse off. Sufficientarianism aims to achieve some specified minimum threshold of welfare for everyone. On this view, distributive justice obtains when everyone's minimum threshold has been satisfied, and requires no more than this. These are all consequentialist theories, since they aim to achieve a certain pattern in the resulting distribution. But even consequentialist theories that do have requirements as to the distributive pattern are incompatible with Equality of Opportunity. This is because all accounts of Equality of Opportunity are committed to choice-sensitivity, and choice-sensitivity and pattern realization do not necessarily coincide. That is, if choice-sensitivity is prioritized, then there is no guarantee that the resulting distribution will instantiate any desired pattern. Vice versa, if pattern realization is prioritized, then there is no guarantee that the distribution would be choice-sensitive. So one must inevitably choose which to prioritize between choice-sensitivity and pattern realization, as a decision rule for cases where the two do not coincide. The upshot for our discussion is that all variants of Equality of Opportunity are normatively implausible as seen from the perspective of patterned theories of distributive justice (and vice versa, all patterned theories of justice are implausible as seen from the perspective of Equality of Opportunity).
In addition to prioritizing outcomes, most consequentialist theories are committed to welfare as the most fundamental moral good. On this view, welfare is the only thing with intrinsic value, the only noninstrumental good. Everything else is only valuable insofar as it promotes welfare. This view naturally compels the adoption of welfare as the currency of distributive justice, so that according to welfare-based theories of justice, "all distributive questions should be settled entirely by how the distribution affects welfare."(Lamont & Favor 2017) Welfare-based theories need not be restricted to consequentialisms. Many other theories, such as luck egalitarianism, can adopt welfare as the currency of distribution. (Lippert-Rasmussen 2016) They just differ in how they distribute this same currency, e.g. do they maximize aggregate welfare, perfectly equalize welfare, equalize welfare maximally without compromising absolute welfare, counteract all luck-based factors conducive to welfare, or other? Along the dimension of currency, the main contenders to welfare are resources and capabilities, since, it is argued, these enable people to attain welfare whilst also being less idiosyncratic or more controllable currencies of distribution. Equality of Opportunity, then, would require everyone have equal access to resources or capabilities instead of welfare. What these three currency contenders have in common is that their proponents tend to be theories that lean more in the direction of the positive conception of liberty. This is in contrast to rights-based theories like libertarianism, which tend more towards the negative conception of liberty. Theories of positive freedom, in turn, tend to be outcome-focused, whilst theories of negative freedom tend to be procedure-focused. Conceiving of the good as welfare (or resources or capabilities) puts positive pressure in the direction of promoting it in the outcome. Conceiving of the good as rights puts negative pressure in the direction of preventing rights-violation throughout the entire process (procedure). This is not to say that the rights violations cannot be assessed in outcomes, i.e. whether in the current state of affairs, people's rights are actually being respected. But as a matter of fact and implementation, theories of positive liberty tend to be more telic, and theories of negative liberty tend to be more deontic.
Libertarianism
Libertarians disagree with welfare-based (or resource- or capabilities- based) theories in what they consider to be the most fundamentally important moral good, and hence the appropriate currency of distributive justice. They namely consider rights and negative liberty, as in freedom from interference, to be more fundamentally important than, say, welfare. According to libertarian theories, then, treating persons as free and equal, and justice, consists in following the procedural rules for not violating people's rights or interfering with them. Every individual has two fundamental rights: the right to self-ownership and the right to acquire natural resources in a specified manner. All other rights derive from these fundamental ones. Various libertarian accounts disagree mostly on the rules that should govern the acquisition of natural resources, but I will not get further into that here. Libertarian theories only have procedural requirements for justice. That is, if the history of acquisitions and transactions of a given distribution satisfies specified procedural rules, then the resulting distribution is considered just, regardless of its pattern. Libertarian theories are thus all strictly procedural, non-patterned and historical. The non-patterned and historical features are compatible with all accounts of Equality of Opportunity, as the latter, we have shown, possess these same characteristics. Libertarian theories would naturally be compelled to endorse more procedure-focused and formal variants of Equality of Opportunity, for example an account claiming that Equality of Opportunity obtains so long as the hiring procedures in society be non-discriminatory and meritocratic, regardless of whether people's opportunities to educate and qualify themselves to apply for jobs were equal. Discrimination would be considered interference or a violation of rights. So to answer our question, more procedural and formal accounts of Equality of Opportunity would be normatively plausible, as seen from the perspective of libertarian theories.
Social Contract Theories
Lastly, let us consider social contract theories. Social contract theories derive their normative principles on the basis of a hypothetical contract. They are identified by three defining characteristics: First, they all specify an 'original position', an initial set of conditions for the negotiation of the hypothetical agreement. This original position can itself be hypothetical, or it can be an attempt at describing reality, e.g. the state of nature that humans would experience if we do not succeed in establishing the hypothetical agreement. Secondly, they specify the characteristics of the participants to this negotiation process. Thirdly, social contract theories specify what the negotiators' desires and interests are, that is, what the negotiators value and to what degree. There are three different traditions of social contracts, the Hobbesian, Lockean and Kantian, and they go from more bare-bones and formal in their requirements to more substantive. Hobbesian social contracts restrict their requirements to those that would establish and sustain a minimal peace between people. Lockean contracts add to these minimal requirements the provision of various public goods. To both of these previous components, Kantian contracts add the requirement of impartiality. An example of such impartiality is John Rawls' Veil of Ignorance in his Theory of Justice, whereby participating negotiators are characterized as unaware of their own particular circumstances (be they socioeconomic, natural or other), so that they cannot be biased by them during the negotiation process. (Rawls 1971) So "Kantian contractarianism tends to be more egalitarian [and substantive] than Lockean contractarianism, which in turn tends to be more egalitarian [and substantive] than Hobbesian contractarianism." (Vallentyne 2017)
Naturally, more formal social contract theories also endorse more formal conceptions of Equality of Opportunity, whilst more substantive ones endorse more substantive conceptions of Equality of Opportunity. I invoke again my previous example comparing Rawls' requirement that we counteract (morally arbitrary) social factors to luck egalitarianism's requirement that we also counteract natural factors such as genetic differences. Without the Kantian impartiality requirement, justice would be restricted to how agents treat each other. For example, it would be restricted to how agents should treat each other to maintain a minimal sustained peace on Hobbesian views, and how agents should treat each other to also also promote some public goods on Lockean views. But it would never expand beyond the realm of what agents do and cause. This is a fundamental point of divergence among social contract theories and liberal theories more generally: whether Justice is only concerned with what lies within the human sphere of influence, or extends beyond. The first view automatically deems all factors beyond human action and causation as irrelevant, and excludes them from the theory of justice, claiming, for example, that it belongs not to justice to rectify inequalities in brute luck and natural endowments. Views that restrict justice to how agents treat each other tend to be more procedurally-focused, as they often involve rules for what agents should avoid in their actions and choices, to avoid infringing on each other's rights. (And we already mentioned that theories leaning towards negative liberty tend to be more procedural.) It depends on the particular social contract theory, however, whether it tends to imply the endorsement of more procedurally-focused or outcome-focused accounts of Equality of Opportunity. Rawls' Theory of Justice, for example, has a mix of both procedural requirements and requirements for the outcome in his two Principles of Justice.
4. Practical Plausibility of Equality of Opportunity
Now that we have reviewed the normative plausibility of Equality of Opportunity and its variants from the perspective of various liberal theories, let us briefly review some practical arguments for or against the various accounts of Equality of Opportunity. The first broad group of practical problems are the so-called index problems. These are all the problems related to difficulties in comparing utilities. For welfare-based theories, for example, there are great difficulties involved in comparing welfare levels across persons, and also comparing the welfare produced by widely disparate goods and services. For example, how do we compare the utility enjoyed by one person going to an expensive theater and that enjoyed by another hanging out with their friends at the park for free? The monetary costs and prices can be helpful, but they do not always accurately reflect the subjective value that people place on things. An account of Equality of Opportunity that chooses welfare or some other measure of utility as its currency of distribution suffer the index problem in the specification stage. For luck egalitarianism, it is almost impossibly difficult to measure differences in all luck-based factors between people. For example, how do we measure differences in natural talent? How do we tease out effects of sustained hard effort from effects of natural talent? One difficulty is in acquiring the information, another is how much information needs to be processed to consider all the choice-based and luck-based factors involved in a single decision. Altogether these yield the calculation problems that often plague outcome-focused theories.
For this reason, procedural rules tend to be far more straightforward to both specify and implement than outcome-focused theories. Furthermore, the value pluralism of liberal societies may compel us to adopt more procedurally-focused theories, as these tend to remain more neutral and accommodating towards different values and ends. But strictly procedural theories are not without their practical woes. Rawls, for example, makes the point that inequalities should only tolerated so long as they do not jeopardize people's basic rights. This qualification was necessary because excessive rewards of high social positions, no matter how meritocratic, can still result in an excessive power imbalance that jeopardizes others' rights. Historical theories of justice face the difficulty of having to acquire complete historical information in order to make their historical assessments as to whether justice obtains , e.g. whether a given distribution was indeed sensitive to people's choices throughout. This requires records of people's choices all the way back. Of course, this is a practical requirement taken to its extreme, and we need not reject a given theory of distributive justice for not being perfectly implementable. If the practical problems that I have delineated in this section affect many alternative theories of distributive justice equally, then they are not fatal, since again, we are looking for the least of all evils, i.e. the best practicable solution.
5. Summary & Conclusion
In this paper, I briefly introduced the reader to the philosophical field of distributive justice, its scope of concern and its task, namely that of providing normative principles to govern the distribution of benefits and burdens among members of society. I examined criteria for assessing the plausibility of a given theory of distributive justice. Given the practical aim of distributive justice, the most important and universal criterion for a plausible theory of distributive justice is that it be practically implementable. As a subset of this requirement, a theory is only practically implementable in a society if it can garner enough support from the people of that society. This is the case if the theory is sufficiently anchored in people's existing beliefs and value commitments. With these considerations in mind, I assessed whether the theory of Equality of Opportunity was a plausible theory of distributive justice, specifically for liberal societies. Following Rawls, I took freedom and equality to be the two most fundamental values held by people in liberal societies. My assessment of the normative plausibility of Equality of Opportunity was thus relative to the values of freedom and equality. However, since there are many substantially different views as to how freedom and equality should be realized, the assessment must specify which interpretation, i.e. which liberal theory, is being adopted as the evaluative standard. In turn, there are many substantially different accounts of Equality of Opportunity, and their differences reflect the underlying differences in interpretations of freedom and equality. I only considered the theories of consequentialism, libertarianism and social contract theories. I considered their positions along the various dimensions of distributive justice, i.e. the site, currency, scope and manner of distribution, and whether these positions compel each liberal theory to endorse certain variants of Equality of Opportunity over others. I could have assessed Equality of Opportunity and its variants with regards to infinitely many other theories and values held by people in liberal societies. For example, I have not yet considered whether the various accounts of Equality of Opportunity are normatively compatible with family values. This would be a practically crucial topic, as the structure of the family is definitely a value that people in liberal societies are currently committed to and cannot give up. There seems to be an essential tension between family and Equality of Opportunity, because the structure of the family inevitably leads to inequalities in opportunity, e.g. due to differences in upbringing and socioeconomic starting point. To what degree are family values and Equality of Opportunity incompatible, and how must the one value be subordinated to the other? Many threads remain to be examined, but the humble conclusion of this essay is that whether Equality of Opportunity (or any given theory of distributive justice, for that matter) is normatively plausible is not a question that can be answered unqualifiedly. It must be specified which variant of Equality of Opportunity we are talking about, and which normative framework (presuppositions) we are evaluating it with regards to. The only part of this discussion which is (only slightly) less complicated is the assessment of whether a given distributive theory is practically implementable. But even to this question, entire scientific fields are devoted such as economics.
References
Berlin, Isaiah. 2019. “Two Concepts of Liberty”. In R.E. Goodin and P. Pettit (Eds), Contemporary Political Philosophy: An Anthology. 3rd Edition. John Wiley & Sons, 359-375.
Cohen, Gerald. A. 1997. “Where the Action Is: On the Site of Distributive Justice”. Philosophy and Public Affairs, 26(1), 3-27.
Dworkin, Ronald. 1981. “What is equality? Part 1: Equality of welfare”. Philosophy and Public Affairs, 10(1981a), 185–245.
Elford, Gideon. 2023. "Equality of Opportunity". In Edward N. Zalta & Uri Nodelman (eds.), The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/fall2023/entries/equal-opportunity/
Holtug, Nils. 2015. “Theories of Value Aggregation: Utilitarianism, Egalitarianism, Prioritarianism”. In I. Hirose and J. Olson (Eds.), Oxford Handbook of Value Theory, Oxford. Oxford University Press, 267-84.
Kuhn, Thomas S.. 1962/1996. “The Nature and Necessity of Scientific Revolutions.” In The Structure of Scientific Revolutions, 2d ed. Chicago: University of Chicago Press, 92-110.
Lamont, Julian & Favor, Christi. 2017. "Distributive Justice". In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/win2017/entries/justice-distributive/.
Lippert-Rasmussen, Kasper. 2016. Luck Egalitarianism. London: Bloomsbury,
Mason, Andrew. 2006. Levelling the Playing Field: The Ideal of Equal Opportunity and Its Place in Egalitarian Thought. Oxford: Oxford University Press.
Miller, David. 2023. "Justice". In Edward N. Zalta & Uri Nodelman (eds.), The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/fall2023/entries/justice/
Nozick, Robert. 2013 (1974). Anarchy, State and Utopia. New York: Basic Books.
Parfit, Derek. 2003. “Equality and Priority”. In D. Mattravers and J. Pike (Eds.), Debates in Contemporary Political Philosophy, London: Routledge, 115-132.
Rawls, John. 1971. A Theory of Justice. Oxford: Oxford University Press.
Rawls, John. 1985. “Justice as Fairness: Political not Metaphysical”. Philosophy and Public Affairs, 14(3), pp. 223-251.
Ryan, Alan. 2017. “Liberalism”. In R.E. Goodin, P. Pettit, and T. Pogge (Eds.), Companion to Contemporary Political Philosophy. Blackwell. Second Edition, 360-382.
Valentini, Laura. 2012. “Ideal vs. Non-ideal Theory: A Conceptual Map”, Philosophy Compass, 7/9. 654-664.
Vallentyne, P. 2017. Distributive Justice. In R.E. Goodin, P. Pettit and T. Pogge (Eds.), A Companion to Contemporary Political Philosophy. https://doi.org/10.1002/9781405177245.ch28
Of course, reflective equilibrium is a method and process that must be continuously implemented throughout life, as our intuitions and principles keep changing and new alternatives are presented to us. But equilibria are periodically established.↩︎
Note that I am using the Rawlsian method, but I am not necessarily committed to the substantive principles of justice that Rawls arrived at through his method.↩︎
Albeit still with substantive disagreements as to exactly which factors are morally arbitrary ,of course.↩︎
These sets of two numbers illustrate a distribution of utility among two agents, with the first value showing the utility enjoyed by the first agent, and the second value showing the utility enjoyed by the second agent. In the (0,0) distribution, both agents enjoy 0 utility. In the (1,0) distribution, the first agent enjoys an utility of 1 and the second an utility of 0.↩︎
Particular theories may of course have restrictions as to what means are permitted, but aforementioned is the overarching ethos of consequentialist theories.↩︎