2. Robustly Public Justification and Discursive Standing

Let me start with a generic notion of robustly public justification. I will first consider publicness and then robustness.

Roughly, to say that φ (e.g., a value judgement, principle, conception of justice, etc.) is justifiable ‘publicly’ is to say that φ is authoritatively (e.g., reasonably, rationally, coherently) acceptable by relevant people (e.g., actual or possible people, affected others, fellow citizens). Alternatively, it is to say that φ is justifiable by ‘public’ reasons, or reasons that are authoritatively acceptable by relevant people. This ‘acceptability’ standard can take many different forms. For example, there are positive and negative variants—we might require authoritative acceptability, or authoritative non-rejectability. And, if ‘acceptability’ is interpreted in thin modal terms, so as to count φ as ‘acceptable’ by people when it is possible that they accept φ, the standard might amount to little more than an O’Neill-type followability requirement.^{Footnote 4} Next, if the standard is applied only to φ, and not also to Betty’s and Paul’s reasons to accept φ, it is compatible with convergence requirements that allow φ to count as publicly justifiable even if Betty and Paul cannot agree at the level of their reasons to accept φ. Alternatively, the standard might apply also at the level of Betty’s and Paul’s reasons to accept φ, but not also to their views as to why their respective reasons are good: this would allow for convergence at a deeper level of thought and argument. Not least, the standard can be applied at different levels of thought, argument, or decision making—as the example just provided already illustrates—and it can be applied with varying scopes, or constituencies. For example, the standard might seek acceptability by all or only some affected actual people, or perhaps only by non-existent, ideal agents.

Less uncontested amongst theorists of public justification is robustness. I follow Gerald J. Postema’s account of the notion. Robustness concerns the features through which robustly public justification confers on its conclusion whatever merit it confers on them. Robustly public justification takes it that the interpersonal authoritative acceptability of φ, or reasons to accept φ, confers, or is a key part of what confers, that merit—or that constitutes, or is a key part of what constitutes, φ’s authority, validity, rightness, or reasonableness.^{Footnote 5} This makes robustly public justification intersubjective at heart and constructivist in a weak, justificatory sense.^{Footnote 6}

Of course, not all conceptions of public justification are robustly public in this sense. For example, public justification is sometimes understood as public in an extremely thin, promulgatory sense as justification-in-public, or as a matter of laying out for others the reasons that, by the speaker’s light, justify φ. Evidently, there is no implication here that these reasons, or their goodness, are subject to an interpersonal acceptability constraint. On a slightly more robust view, public justification is justification by reasons that are both good and subject to a (weak) publicity constraint. On one recent view, for instance, to justify φ publicly is to justify φ by a subset of good (or ‘valid’) reasons, namely, those good reasons that at the same time are such that relevant people could, at a relevant level of idealization, come around to grasping their goodness.^{Footnote 7} This, too, does not entail that the interpersonal acceptability of these (allegedly) good reasons is what constitutes, or even contributes to, their goodness. Robustly public justification goes beyond these weak senses of the publicness of public justification. Stephen Macedo, a first-generation political liberal, aptly captures robust publicness:

The last sentence makes the relevant point: robustly public justification is robustly public in the sense that it takes the goodness of good reasons to entirely be a function of their authoritative acceptability by relevant people. Arguably, first-generation political liberals like Macedo, Charles Larmore, and John Rawls, amongst others, take public justification to be robustly public in this sense.

Next, given its robustly public character, robustly public justification allocates a strong form of discursive standing to its fully enfranchised participants. To bring out what this standing is, let me assume that, where we are reasonable and others are concerned, we pursue at least two aims. First, we aim to act toward them on grounds, widely conceived, that are acceptable by them—that is, at least in principle, ideally, or given the (perhaps counterfactual) assumption that they look at things in the right way—thus, we accord them discursive standing. Second, however, we also aim to act toward them on grounds that, at least as far as we can tell, are good. This entails that reasonable people can accord others at least two kinds of discursive standing, depending on how they relate these two aims to one another.^{Footnote 9}

First, we might accord others derivative discursive standing. When we accord this standing, we aim to act toward others on grounds that are good and acceptable, but we take the acceptability of these grounds, or valuable forms of it, to derive from a proper appreciation of their goodness. Thus, we seek “ideal” unanimity (Thomas Nagel) or “normative” consent (David Estlund).^{Footnote 10} Second, we might accord others a stronger, constitutive form of discursive standing. Where we accord others this standing—or, as I shall also say, were we accord them discursive respect—we aim to act on grounds that are good and acceptable, but we take the goodness of these reasons to (at least partly) depend on, or be constituted by, their acceptability. As the last passage quoted above reflects, Rawls-type public justification allocates to its fully enfranchised participants standing of this second, stronger kind.

We can describe conceptions of robust public justification as allocations of discursive respect: they prescribe that some form of constitutive discursive standing be allocated to each fully enfranchised participant, and lesser discursive standing, if any, to other people. As I argue in more detail elsewhere, allocation profiles of discursive respect can differ in various (interdependent, as we shall see later) dimensions, including what we might refer to as their ‘scope,’ ‘depth,’ and ‘discursive purchase’—we might call this the ‘matrix of discursive inclusion.’^{Footnote 11} The following provides a working notion of these three dimensions.

To begin with scope, we ask questions about scope when we ask questions such as: ‘Who counts, or should count, as a fully enfranchised participant in public justification, or as a full member of its constituency?’ or ‘Who or what merits discursive respect?’. Questions about the scope of discursive are questions about the range of actual or possible others to whom we accord, or should accord, constitutive discursive standing. Next, take depth. We ask questions about the depth of discursive respect when we ask, e.g., how fundamental in the order of justification the things are that (allegedly) depend for their authority on their (qualified) acceptability by (some) other, actual or possible people. For example, are these first-order practical reasons only? Do they also include more fundamental, general moral or political principles? Is an acceptability requirement in some form applied even as a constitutive requirement of all practical reasoning, as some Kantian moral philosophers would have it?^{Footnote 12}

Especially relevant here is discursive purchase—an often-neglected dimension of discursive respect that is rarely recognized as a relevant dimension of such respect in its own right. Often, it seems, purchase is construed as a mere function of the degree in which justification standards idealize, while its importance is seen as secondary to the importance of identifying suitable levels of idealization. It is true that idealization and purchase are tied together. As I argue elsewhere, high degrees of idealization entail low degrees of purchase, and high degrees of purchase require low degrees of idealization; it does not follow, though, that purchase is systematically secondary to idealization.^{Footnote 13} What we should take to be the systematic ordering between these things depends on the underlying aims of robust public justification—where some candidate aims favour high degrees of idealization, and others high degrees of purchase. But let me set this matter aside now, and treat purchase as a legitimate and relevant dimension of discursive respect. What I shall say suffices for my present purposes; for a more thorough picture, it is best seen in conjunction with what I say elsewhere.^{Footnote 14}

As I understand the notion, we consider the purchase of an agent’s discursive standing when we consider what degree of normative influence in relevant justification practices this standing accords to that agent, given her actual deliberative resources, widely conceived (so as to include whatever views, volitions, skills, attitudes, capacities, and so on, she actually draws on in her practical reasoning). Purchase so construed is not a matter of the normative influence of non-actual, ideal agents in justification practices, nor is it a matter of the influence of actual agents with idealized resources. Instead, it is a matter of the normative influence of actual agents, given their actual deliberative resources, in relevant justification practices. In a sense, therefore, purchase helps to earth the normativity of justification: it helps to describe what normative distance there is, if any, between how actual agents actually go about their practical reasoning, and how they would have to go about their reasoning in order to do so in ways that relevant justification practices construe as authoritative. And, evidently, purchase is especially relevant in relation to emancipatory intuitions—assuming that emancipation (rightly or wrongly, for better or worse) requires that real people such as they are be accorded a meaningful level of normative influence in matters that affect them.

Discursive respect can take many forms that greatly vary in purchase. To see how, recall that such respect involves an aim of acting on ‘acceptable’ grounds. When is φ ‘acceptable’? Many modally different answers are possible. For example, consider:

1. (1) Paul accepts/does not reject φ.

2. (2) Paul is not committed to reject φ in light of his actual views and volitions.

3. (3) Paul would not reject φ in light of views and volitions he would/could have if he was ideally reasonable/rational (as we define this).

Evidently, the purchase that discursive respect can have for its recipients will vary greatly depending on what we take acceptability to require. Take two extreme examples. First, consider, say, actualist discursive respect—a conception of discursive respect that takes it that proper ‘acceptability’ requires something like (1) or (2). On such a conception, Betty’s constitutive discursive standing can have a high degree of purchase. If Betty cannot now accept φ coherently, given her actual views and volitions, this would count as showing that φ is not suitably acceptable by her. This seems to give her a veto: her ‘no,’ it seems, counts for a ‘no.’ Contrast this with counterfactualizing discursive respect—or a conception of discursive respect that takes it that proper ‘acceptability’ requires something more like (3). On such a view, Betty’s discursive standing can have very little purchase. That she cannot now accept φ coherently, or is committed to oppose φ, might not count against φ at all. For her rejection of φ could be countered by the consideration that she would not reject φ if she appreciated matters in a different, and, as others see it, more reasonable or rational light. Thus, Betty’s actual views and volitions here might become irrelevant: her ‘no’ might not count for a ‘no.’

Actualist discursive respect gives greatest normative weight to the actual views and volitions of people, or their actual deliberative resources. Once we start to normatively qualify, constrain, or launder these views and volitions, their impact decreases and a gap opens between the actual acceptability and the normative acceptability of our grounds. This gap widens as these qualifications become normatively more remote, or harder to meet, relative to people’s actual views and volitions, or their actual deliberative resources. And, as it widens, the normative influence of their actual views and volitions decreases—and with this decreases the purchase of their discursive standing. Take again a counterfactualizing view. My discursive respect for you can have fairly little significance for you if you know that it allows me to take my reasons to be suitably ‘acceptable’ by you even if the actual you, in light of your actual views and volitions, could never actually accept them.

Two more remarks are in place. First, I index discursive purchase to an agent’s actual deliberative resources. An alternative is to index it to the ideal resources of ideal agents, or the idealized resources of actual agents. As will emerge in Section 5, below, one reason to set aside these possibilities here is this: a focus on actual deliberative resources helps to keep in view potentially relevant forms of discursive inequality. Second, purchase and scope are linked: constraints on the scope of public justification can often be translated into purchase-determining qualifications of the sort of acceptability that public justification seeks (and vice versa).^{Footnote 15} For example, if acceptability by reasonable people justifies, what justifies is reasonable acceptability. Still, we should not subsume purchase under scope (or vice versa). Purchase is useful in assessing in what way, if any, non-ideal public justification accords to people equal discursive standing (or so I claim). At any rate, it is relevant here as my focus is on views of public justification that, in one way or other, aim at a “basic structure of justification”^{Footnote 16} that make questions of equality and emancipation, and with this of equal discursive purchase, relevant.

3. Political Liberalism

I now turn to Rawls’s political liberalism, or one reading of the idea of public justification at its core. On this reading, which I call the ‘deep view,’ public justification has a special position in the matrix of discursive inclusion. On the deep view, public justification not only fully enfranchises actual reasonable people, given their actual reasonable deliberative resources; it also accords them discursive respect that is deep in political liberalism’s order of justification, rich in discursive purchase, yet problematically limited in its scope of inclusion. I will not here try to defend or even fully lay out the deep view, as this would go beyond what can be done now—I do so elsewhere.^{Footnote 17} Thus, I will employ it heuristically: it serves as one candidate view of the role of public justification in political liberalism that situates one of the longest-standing, recurrent concerns about the nature of that project. This is the concern that some of political liberalism’s core commitments stand in need of a (public) justification that political liberalism seems unable to itself provide^{Footnote 18}—and this in a way that renders political liberalism vulnerable to a charge of public dogma (see below). What matters here is this problem (I shall later address challenges faced by attempts to respond to it). Readers who prefer a different reading of Rawls hence can set aside the specifics of the deep view, so long as they grant that political liberalism, on some relevant reading, faces a problem of that overall type.^{Footnote 19}

I start with the idea of a political liberalism. If a conception of justice, φ, is a political liberalism, φ is liberal in content and political both in its form of justification and domain of application. If φ is liberal in content, φ allocates to citizens basic rights and liberties of special priority, as well as all-purpose means that help citizens to make use of these rights and liberties. If φ is political in form of justification, it does two things. It (i) adopts the constructivist standard that reciprocal, equal acceptability by reasonable people justifies. And, importantly, (ii) it attaches a populist ^{Footnote 20} interpretation to that standard: that is, it interprets the standard such that φ fails to meet the standard, or fails to qualify as reciprocally acceptable by reasonable people, already if φ is the actual subject of reasonable disagreement between these people (where reasonable disagreement is defined as disagreement that can arise between reasonable people).^{Footnote 21} (i) and (ii) are at the core of Rawls’s political constructivism. And, for some political liberals, they are at the heart of an idea of political justification that fully respects reasonable people as free and equal.^{Footnote 22} Not least, if φ is political in its domain of application, φ does two things. φ regulates only the basic structure of a given society, or its most important social, political, and economic institutions, or its domain of the political. And φ regulates this domain exclusively in terms of ‘political values.’ Values are political only if they exclusively apply to this domain. But such values must also be part of the political tradition of the society under consideration. And, as Rawls focuses on the United States and its democratic political tradition, he takes it that political values favour liberal democracy.

As to the deep view of public justification, then, it contrasts with a more familiar ideal theory view. On the ideal theory view, public justification is part of ideal theory only, or supposes the context and normative resources of an ideal, well-ordered society of the sort prescribed by Rawls’s theory of justice, Justice as Fairness (JF). Ideal theory public justification thus includes within its scope only the ideal reasonable citizens of an ideal, well-ordered society.^{Footnote 23} Now, the deep view of public justification does not deny that Rawls makes use of an idea of ideal theory public justification. But the deep view takes it that Rawls invokes an idea of public justification also for other purposes, outside ideal theory.

Consider Rawls’s argument from overlapping consensus.^{Footnote 24} He argues that political legitimacy is possible only if political power accords with a conception of justice that can gain an overlapping consensus between the reasonable comprehensive doctrines of reasonable people. This applies also here and now, outside ideal theory. Now, a reasonable overlapping consensus matters for the purposes of legitimacy because a conception of justice, if it is incompatible with any one of the reasonable doctrines that reasonable people endorse, would not be reciprocally acceptable by these people. To ensure a reasonable overlapping consensus just is a way to ensure reciprocal acceptability by the reasonable. Thus, political legitimacy here and now requires reciprocal acceptability by the reasonable here and now. Next, Rawls (notoriously) takes it that a conception of justice can gain an overlapping consensus, or can meet the standard of reciprocal acceptability by the reasonable, only if it avoids reasonable disagreement between reasonable people. Rawls interprets this in populist terms: actual reasonable disagreement about φ can count as evidence that φ is the subject of reasonable disagreement, and as such is not reciprocally acceptable. Rawls also insists that only a political and liberal conception of justice, if any, avoids reasonable disagreement. And it is for this reasons that he here and now (re)formulates JF as a political liberalism. Taken together, these things suggest that reciprocal acceptability by reasonable people plays a fundamental role. It effectively serves as a standard of theory selection for the domain of the political, as it states a condition any conception of justice (allegedly) must meet in order to be suitable as a basis for a legitimate exercise of political power—and this applies also here and now, outside ideal theory.

To now bring in discursive respect: as anticipated above, Rawls-type public justification allocates discursive respect that is (i) deep in the order of justification, (ii) rich in discursive purchase, but (iii) notoriously limited in scope. The discursive respect it allocates is deep since the standard of reciprocal acceptability by reasonable people effectively serves as a standard of theory-acceptance for the domain of the political. It is rich in discursive purchase as Rawls interprets this standard in populist terms: at least in some cases, the fact that reasonable people actually reasonably disagree about φ, given their actual reasonable deliberative resources, counts as evidence that φ is not reciprocally, equally acceptable by all reasonable people. As to scope, public justification is highly limited in scope. Only reasonable people are included in its scope on equal, fully enfranchised footing, and hence enjoy full constitutive discursive standing. How limited in scope public justification thereby is depends on how much contested normative content goes into the idea of the reasonable on which public justification builds. As a rule of thumb, the more such content goes into this idea, the more exclusionary public justification will be. Like others, I take it that this idea is rich in contested normative content. Let me indicate why, on the deep view, this often-voiced complaint is justified.

Recall that, for Rawls, political liberalism can avoid reasonable disagreement. Yet the label ‘political liberalism’ does not refer to just any suitably acceptable conception of justice, but to a conception that (i) is liberal in content, and (ii) constructivist and populist in form of justification, and that (iii) exclusively applies to the domain of the political, and that does so by (iv) containing political values only. As legions of critics have observed, however, there is disagreement about the merits of these requirements. Now, if Rawls construes these disagreements as reasonable, he needs to concede that political liberalism is not reciprocally acceptable by all reasonable people. Yet Rawls does not draw this conclusion. He takes it that only a political liberalism, if anything, is not the subject of reasonable disagreement. Arguably, then, he cannot construe disagreement about (i)-(iv) as reasonable. And this is tantamount to building a commitment to these requirements into the very idea of reasonableness on which public justification builds.^{Footnote 25} By implication, Rawls needs to suppose, as well, that reasonable people accept that reciprocal acceptability by people like them justifies, and that public justification needs to include within its scope only people like them. This makes political liberalism “insular.”^{Footnote 26}

The upshot: deep public justification here is justification between people who endorse the defining requirements of political liberalism. A commitment to these requirements is part of the doxastic bedrock that public justification supposes, and that everyone included within its scope is expected to share. But, again, in the actual world, outside JF’s ideal, well-ordered society of people who are reasonable in political liberalism’s sense, there are many relevant people who relevantly contest these requirements and who hence do not pass political liberalism’s threshold tests of reasonableness.^{Footnote 27} This is why the complaint that public justification here is exclusionary is justified. In fact, in light of the above considerations, political liberalism looks like a stretch of public dogma.^{Footnote 28}

4. Scope-Purchase Interdependence

What to make of this? Suppose we do not reject that a conception of justice, in order to be able to here and now provide a basis for political legitimacy, must here and now be justifiable in robustly public terms. A standard line of response, then, is to opt for a more inclusive form of robustly public justification—one that builds the standard of reciprocal, equal acceptability on thinner threshold tests of reasonableness, if any, and that hence extends discursive standing of the strong, constitutive kind to more people in moral or political matters that affect them.

At first pass, this seems compelling especially in light of emancipatory and egalitarian intuitions. As Rawls notes, we regard ourselves as “self-authenticating sources of valid claims”^{Footnote 29}—but not only in relation to the claims we make on institutions in pursuit of conceptions of the good.^{Footnote 30} We desire to be recognized by others as having an entitlement to make claims on them in relation to what they see as good reasons, at least in matters that affect us. One important part of this recognition is that they regard our non-acceptance of what they see as good reasons as something that can call into question, or put in need of justification, the goodness of these reasons. It may or may not be the case that this desire can be met only if we are being accorded constitutive discursive standing of much purchase. But fully enfranchised inclusion in practices of robustly public justification evidently speaks to this desire. And, if a justification practice allocates this strong standing to anyone, it should allocate it equally to equal people. To say the least, not endorsing political liberalism’s defining requirements does not seem to make people relevantly unequal. If anyone is to have discursive standing that is rich in purchase, then not only people who are reasonable in political liberalism’s sense should have it.

In what follows, I assume that this line of response is plausible—other things being equal and assuming that a conception of political justice should be robustly publicly justifiable in the actual world, and not only in ideal theory. What I want to do now is to turn to the two problems referred to in the introduction. The first problem reflects the fact that the scope of robustly public justification and the purchase of the discursive standing that such justification can allocate are interdependent. The second problem reflects the complex nature of discursive equality in practices of justificatory reciprocity. Both problems call into question whether a more inclusive form of robustly public justification really can sit well with emancipatory and egalitarian intuitions. I shall now attend to the first problem.

Let me suppose that fully enfranchised discursive inclusion is an important good. Still, we will need to balance this good against other, perhaps no less important goods when it comes into competition with them. And this it does: if discursive respect is rich in purchase, robustly public justification is a mixed blessing. The more inclusive in scope public justification is, the more doctrinal diversity is included in its scope—at least in the real world, outside ideal theory. But the richer in purchase the discursive standing is that each fully enfranchised person enjoys within practices of robustly public justification, the stronger is the normative impact of this diversity on what these practices may count as suitably acceptable, or justifiable. Thus, there is scope-purchase interdependence:

For example, suppose liberal political principles must be equally acceptable by relevant citizens, while we read the phrase ‘equal acceptability’ in high-purchase, actualist terms: say, we count φ as ‘equally acceptable’ by others only if φ can coherently be accepted by them in light of their actual views and volitions. Plainly, then, liberal principles—or, by extension of argument, any other, morally interesting normative principles—cannot pass such an actualist threshold if the group of relevant citizens includes people whose moral, political, ethical, or other outlooks are inconsistent with them (and there almost always are such people). The point generalizes: the interdependence of scope and purchase can have problematic implications at any level of argument at which a standard of reciprocal acceptability applies as a standard of justification. It hence can have these implications also in relation to the justification status of the commitment to robustly public justification, or of the view that discursive respect is important—namely, if these things must be justified by the standard of reciprocal acceptability.

As has often been noted, then, inclusive robustly public justification is at risk of being both self-defeating and unable to secure substantively plausible results^{Footnote 32}—that is, so long as it accords discursive standing that is rich in purchase. This leaves robustly public justification (RPJ) with several options:

1. (i) RPJ exempts from the requirement of reciprocal acceptability all normative content that no relevant person should reject (whatever content this is, but assuming it includes values that commit to RPJ and discursive respect).

2. (ii) RPJ accords high-purchase discursive standing exclusively to people who do not reject normative content that no relevant person should reject.

If robustly public justification accords to people the kind of standing that (ii) refers to, it accords them a standing such that, if they cannot coherently accept φ in light of their actual views and volitions (given basic criticality and reasonability), this counts as evidence that φ is not suitably acceptable by them. A third option, then, is to reject an actualist notion of discursive standing altogether and to premise robustly public justification on counterfactualizing discursive standing (David Enoch loosely refers to this as a variety of ‘idealization’):^{Footnote 33}

1. (iii) In examining φ’s public justifiability, RPJ examines whether relevant people can (could, would) suitably accept φ provided they do not reject normative content that no relevant person should reject.

For example, suppose we want to protect φ from likely opposition, and so decree that φ is reciprocally acceptable by people if φ is such that relevant people would not reject φ if they were reasonable—where reasonableness is then defined in terms that are suitably geared toward the non-rejection of φ. Thus, if Betty-when-‘reasonable’ would not reject φ, then actual Betty’s actual inability to coherently accept φ would not count as evidence against φ’s reciprocal acceptability (rather, it would count as evidence of Betty’s unreasonableness).

It is not clear what these three options would amount to in practice. But none seems desirable from an emancipatory and egalitarian point of view. Prior to further argument, the second option abandons the idea of equal discursive respect within inclusive scopes. And prior to further argument, the third option hollows out discursive respect and with it the emancipatory value of robustness in public justification: what worth and use, if any, can Betty’s constitutive discursive standing have for her if φ counts as ‘equally acceptable’ by her even if she cannot accept φ coherently in light of her actual views and volitions (given basic criticality and reasonability)? Not least, all three options call for an odd kind of government-house public justification: a practice of public justification the terms of which are from the outset tailored to insulate and protect pre-selected moral, political, or other content from likely opposition. And so long as we lack reassurance that this content really is such that no relevant person should reject it—rather than content that, say, reflects the parochial convictions of some dominant group or other—this seems to return us to a variant of public dogma.

The first problem, then, is this. In light of emancipatory and egalitarian intuitions, robustly public justification should allocate to people equal discursive respect that is inclusive in scope and meaningfully rich in purchase. But robustly public justification seems unable to do so while ensuring plausible conclusions and avoiding self-defeat. At any rate, it seems that if such justification must both be inclusive in scope and allocate high-purchase discursive standing, it has to custom-tailor its agenda or abandon its commitment to equal discursive standing.^{Footnote 34}

5. Reciprocity and Discursive Purchase

To now turn to the second problem, let me slightly shift the focus. Justification by a standard of reciprocal acceptability—or justificatory reciprocity, or “reciprocity of reasons”^{Footnote 35}—seems egalitarian at heart. In requiring reciprocal, equal acceptability of the views it applies to, the standard seems to call for justification practices that accord equal discursive standing to their fully enfranchised participants—even if this standing is being accorded only within limited scopes and with respect to limited agendas. Yet it is not clear in what way this standing really is ‘equal.’ I will now elaborate on aspects of the issue that have to do with the purchase of discursive standing.

To begin with, two assumptions should be put into place. The first is this: in a justificatory capacity, a standard of reciprocal, equal acceptability cannot require just any kind of equal acceptability. It must require equal authoritative acceptability. For example, equally incoherent, ill-informed, or unreasonable acceptance cannot positively contribute to the justification status of a view (or its validity, correctness, or reasonableness), nor can similarly deficient rejections negatively contribute to it. The need for authoritativeness constraints complicates matters. For it entails that the question of whether some view, φ, is equally acceptable by relevant people must be crossed with considerations as to whether meeting the underlying authoritativeness constraint is suitably available to them. (Given my present context, I shall often put things in terms of a reasonableness threshold.)

A second, initially plausible assumption is this: the ‘equal’ discursive standing that practices of justificatory reciprocity allocate should be equal not just in some abstract sense, but concretely in terms of its discursive purchase. Consider a justification practice, JP1, that allocates to Betty and Paul constitutive discursive standing, and hence discursive standing of the same abstract kind, but applies to them different thresholds of authoritativeness such that their discursive standing has unequal purchase. Say, Paul enjoys actualist discursive respect such that, if Paul cannot accept φ in light of his actual views and volitions, the practice recognizes this as counting against φ, or as showing that φ fails to be suitably acceptable by Paul. Yet Betty is being accorded counterfactualizing discursive respect such that, if she cannot accept φ in light of her actual views and volitions, this is not recognized as counting against φ, but as counting against her—say, it is counted as evidence that she is insufficiently reasonable, rational, right-minded, and so forth. Surely, this would instantiate relevant discursive inequality: Paul has a veto, but Betty does not. Their discursive standing is equal in abstract kind, and so is up to a point equal, but it is relevantly unequal in purchase. (We shall see below that this assumption, despite its appeal, is problematic.)

These two assumptions suggest that the standard of justificatory reciprocity calls for justification practices that do two things (amongst many others): they apply authoritativeness constraints—or, given my focus: some reasonableness threshold—to the members of their constituency, or their deliberative behaviour, while simultaneously according them discursive standing of equal discursive purchase. However, real-life justification practices may not be able to do both, or be known to do both.

Consider another justification practice, JP2, that adopts a counterfactualizing conception of discursive respect and applies it consistently to all participants: say, one that defines deliberations, objections, acceptances, rejections, and so on, as authoritative, or as counting toward the justification status of views, only to the extent that they are reasonable—while it construes reasonableness in normatively demanding terms. And suppose that these terms are demanding enough to make it relevantly harder for some participants than others to count as ‘reasonable’—or to register ‘reasonable’ objections, rejections, acceptances, and so forth. Say, Outstanding Betty meets these terms with ease so that her voice usually counts as reasonable anyway, while Limited Paul, given his limited deliberative resources, does not meet them with ease so that his voice usually counts as reasonable only after his views and volitions have undergone (sometimes even considerable) second guessing, filtering, laundering, or revisions. I take it that Betty and Paul here have standing of unequal purchase. But this purchase differential is now not owed to the fact that actualist discursive respect is applied to one but not the other. It is owed to the fact that Betty and Paul bring different resources to the task: given her superior resources, Betty qualifies as more ‘reasonable’ than Paul.

The contrast between JP1 and JP2 suggests a distinction between formal and substantive equality of discursive standing. Let me take it that a justification practice accords to relevant people discursive standing that is formally equal only if (i) it accords them discursive standing of the same abstract type (e.g., constitutive discursive standing), and (ii) applies to each the same threshold of authoritative acceptability. A justification practice accords to people discursive standing that is substantively equal, however, only if their respective discursive standing has equal discursive purchase. Thus, JP1 seems further removed from the egalitarian commitments of justificatory reciprocity than JP2: JP1 accords to Betty and Paul discursive standing of the same abstract kind, but that standing is neither formally nor substantively equal. By contrast, JP2 accords them discursive standing that is formally equal, although it is not also substantively equal.

Alas, purchase differentials that result from resource differentials are unavoidable in any form of larger-scale real-life public justification. Who counts as an Outstanding Betty or a Limited Paul in a given practice of public justification depends on how the practice defines reasonableness (assuming it adopts some reasonableness threshold), but it also depends on what deliberative resources people bring to the task of public reasoning. And there always are differences between our deliberative resources—or whatever capacities, skills, commitments, attitudes, views, volitions, and so on, our practical reasoning draws on. People’s deliberative resources are unlikely to fall below some baseline that comes with intelligent agency. But it is equally plain that their resources often vary greatly in ways that affect how readily available it is for them to count as ‘reasonable’—especially when ‘reasonableness’ is taken to require not just basic criticality and reasonability, but, say, substantive commitments not everyone shares anyway. And we usually do not know just how readily available this is for them at least when distant strangers are concerned—if only because we do not know just what their deliberative resources are. Thus: if the standard of justificatory reciprocity calls for justification practices that allocate substantively equal discursive standing, we would not know when our attempts at justificatory reciprocity succeed where large and diverse, society-scale groups are concerned. But such just are the constituencies of the standard in its alleged main role as a standard of moral or political justification. If the standard is to be a serious candidate for this role, then something must give way.

This point is less trivial than it might seem at first sight. It suggests we reject a maximalist conception of justificatory reciprocity—call it ‘deep reciprocity’:

Deep reciprocity requires not simply that justification practices allocate discursive standing that is equal in some purchase-sensitive way. Rather, it requires substantive equality. This may have intuitive appeal. As we have just seen, however, it would push justificatory reciprocity out of its alleged main role. As we do not know with relevant specificity what the deliberative resources of relevant people are, we would rarely, if ever, be able to tell whether a given reasonableness threshold is suitably available by all relevant people. In the world as we know it, then, we have reasons to set aside DR.

We should not fall into the opposite extreme and opt for a purchase-insensitive, minimalist conception of justificatory reciprocity—call it ‘nominal reciprocity’:

Nominal reciprocity is consistent with objectionable forms of discursive inequality. Consider a justification practice, JP3, which accords to all relevant people discursive standing of the same abstract type and applies to all relevant people the same reasonableness threshold. But JP3 defines that threshold in terms that give trumping weight to moral conceptions, C, that members of Dominant Group typically endorse, and members of Marginal Group typically reject. Say, JP3 counts acceptances, or rejections, as authoritative, or reasonable, only if they cohere with C. JP3 instantiates formal equality. Yet this would merely mask that JP3 gives full discursive standing only to members of Dominant Group—thus cementing their dominating influence at the level of practices of reason giving. It hence seems that NR would not provide a plausible conception of justificatory reciprocity.

If NR is unsuitable, a more familiar conception of justificatory reciprocity is not quite suitable either—call it ‘strong reciprocity’:

Strong reciprocity leads us back to the problem of nominal reciprocity, albeit at a deeper level of thought. For when are reasonableness thresholds equally acceptable by Betty and Paul? Surely, what matters is not their equal brute acceptability, but their equal authoritative acceptability. Yet that Betty and Paul can accept some reasonableness threshold ‘authoritatively’ does not mean that this is an option equally open to both. Thus, SR, too, can mask relevant discursive inequality.

The second challenge, then, is to specify the nature of discursive equality in justificatory reciprocity in a way that maintains purchase-sensitivity, but that does not push the standard of justificatory reciprocity out of its alleged main role. The above suggests that a conception of justificatory reciprocity must be located in the normative space between deep reciprocity and nominal reciprocity. If purchase-sensitivity matters—as it seems to, not least given emancipatory intuitions—we should reject purchase-insensitive conceptions of justificatory reciprocity such as nominal reciprocity and with it, I suggested, strong reciprocity. However, purchase-sensitivity cannot take the maximalist form of deep reciprocity. The standard of justificatory reciprocity is a viable candidate for the role of a standard of robustly public justification only if either (i) it does not after all require discursive standing to be equal, or (ii) discursive standing can count as equal even when it is unequal in purchase. This is an uneasy result. If we set aside (i)—given egalitarian intuitions—then if equal discursive standing is not (also) a matter of discursive standing that is equal in purchase, what is it a matter of? How much purchase inequality is compatible with justificatory reciprocity? How much is permissible, or justifiable? The target of these questions is not the importance of discursive equality, but the view that justificatory reciprocity is a natural ally of egalitarianism at the level of discursive inclusion. For it is not clear what kind of discursive equality, if any, justificatory reciprocity can require.^{Footnote 37}

Let me add one more point. One manoeuver that is possible here surfaced earlier: we might index the purchase of an agent’s discursive standing not to the actual deliberative resources of that actual agent, but construe purchase in idealized or normalized terms. For example, we might index it to the recourses that the agent should have, or would have after relevant improvement; or to ideal resources of ideal agents; or to average resources or minimal resources—amongst other possibilities. Purchase may need to be construed in such terms if we want to maintain (i) that the standard of justificatory reciprocity is viable as a standard of public justification and (ii) that the standard in that role calls for discursive standing that is equal in purchase. After all, idealization and normalization enable us to identify discursive standing as ‘equal’ in purchase even when it is relevantly unequal in actual purchase. However, this is a spurious achievement. If discursive standing that is equal in idealized or normalized purchase can be relevantly unequal in real purchase, equality of idealized or normalized purchase can still mask relevant discursive inequality. And so the question still is how much (actual) purchase inequality is compatible with justificatory reciprocity, and how much is permissible or justifiable.