I. Introduction

In Democracy without Shortcuts: A Participatory Conception of Deliberative Democracy, Cristina Lafont moves to the forefront of democratic theory the goal that citizens should ‘own and identify with the institutions, laws and policies’ (Lafont 2019: 3) that coerce them. This is an important move, particularly today, when so many populist movements draw support from people who feel, often correctly, that they have not been ‘heard’ in the system that produces the laws that coerce them, and that consequently their interests do not count. These citizens, like many others, feel far from owning and identifying with the institutions, laws and policies that coerce them. Their goal in electing populist leaders is to reverse this situation by reducing the coercion of government and/or replacing the current political institutions with ones in which they can own and identify with the ensuing laws and policies.

Lafont’s analysis directs those who would build or change democratic institutions to the goal of reducing ‘political alienation or estrangement,’ that is, ‘being alienated from laws that one is bound to obey but cannot reflectively endorse’ (19; emphasis in the original, as in subsequent quotations, except as noted). To the problem that in reality not all citizens can reflectively endorse all laws, she provides a citizen-centered institutional approach. This approach directs us to create and celebrate institutions through which citizens can both contest and contribute to lawmaking. It provides original answers to the questions raised by majoritarianism, judicial review, minipublics based on random selection, and the role in the public sphere of arguments based on religious commitment.

I will suggest in this comment that the core of Lafont’s approach might be furthered by a theory of legitimacy that a) explicitly endorses plural sources of democratic legitimacy, b) acknowledges the aspirational quality of the many democratic ideals that make up this legitimacy, and c) recognizes consequently that democratic legitimacy is a matter not of either/or but of degree. All democratic legitimacy is partial.

The utility of a plural approach to legitimacy appears as early as the first substantive section of the book. The first paragraph of this section (‘1.1’) begins by stating the ‘ideal that one should not be subject to laws that one cannot see oneself as an author of’ (18). Lafont then quickly rejects the simple equation of democracy with political equality: ‘Political equality is necessary but not sufficient for democratic government’ (19, my emphases). Democratic government also requires reflective endorsement and active participation: plural sources of legitimacy.

Lafont traces her concept of citizens’ reflective endorsement to Habermas:

My participatory interpretation of deliberative democracy follows Habermas’s conception of democratic control, which requires an ongoing feedback loop between processes of opinion- and will-formation in the public sphere and political decisions taken by the political system. This dynamic model makes it possible to conceive democratic control as a matter of responsiveness not to actual public opinion, as reflected at a given movement, but to considered public opinion, as it forms and evolves over time (Lafont 2019: 24).1

The concept of a dynamic feedback loop is an important contribution both to deliberative theory and to a more general theory of democratic legitimacy. So too is the idea that considered public opinion has a far greater claim to democratic legitimacy than most current political opinion. Yet even considered public opinion, despite forming and evolving over time, may be wrong. Lafont explicitly agrees that epistemic claims – claims about factual correctness – produce some legitimacy (166), although she argues rightly against some epistemic theorists that epistemic correctness ought not to be the sole criterion for democratic legitimacy. If epistemic claims have legitimacy in themselves, then normative legitimacy will be partially diminished whenever a set of democratic procedures systematically produces epistemically flawed outcomes, even if at the same time those procedures increase normative legitimacy in other ways. Acknowledging both the plural sources of legitimacy and the inevitability of only partial legitimacy would produce theory that considered a law more or less legitimate depending, among other things, on the quality of the feedback loop between public opinion and political decision, with quality defined in many ways.

Lafont’s concept of participation also has a Habermasian lineage because it denotes many forms of contribution to opinion- and will-formation in the public sphere, in addition to voting, organizing, protesting, running for office, or political activism per se. Lafont writes that:

only a democratic political system in which citizens can participate in shaping the laws and policies to which they are subject, can ensure that these laws and policies conform to their judgments about justice … Democratic participation in decision-making is essential to prevent an alienating disconnect between the political decisions to which citizens are subject and their political opinions and will (22–23).

Plurally-sourced legitimacy appears subtly here in the formulation that participation ‘can’ ensure rather than ‘does’ ensure the conformance of laws to citizen judgement; participation is ‘essential to,’ but does not guarantee, non-alienation.

Rightly in my view, Lafont includes in her ‘diffuse’ (26) and ‘capacious’ (27) understanding of participation any citizen contribution to the opinion side of the feedback loop of opinion and decision that has even the smallest influence over political decisions. Such contributions could include even ‘the extent that everyone’s views are reflected in aggregated opinion polls,’ so long as politicians take those polls ‘into account when making political decisions’ (26). In her analysis, the citizen contribution to the feedback loop also includes what I have called ‘everyday talk’ (Mansbridge 1999). Lafont thus urges democratic theorists to think less about political activism or specific one-off events of participation and more about ‘the deliberative system as a whole’ (138).2

This broad, diffuse, and capacious understanding of political participation applies well to normative legitimacy. The most diffuse forms of participation are unlikely to increase perceived legitimacy. Citizens’ perceptions of being heard, let alone their deeper perceptions of owning the eventual law, will probably not increase when a survey organization asks the opinion of another person with their demographic characteristics, even if that person’s opinion marginally affects the behavior of an elected or administrative representative in a direction that supports the interests of the person not interviewed. Even citizens who are themselves directly surveyed – by some organization phoning them around dinnertime and asking a set of often disconnected questions – will probably not feel heard through this experience or change their perceptions of owning the laws or the legitimacy of their government.3 Yet, without increasing perceived legitimacy, if the results of these surveys accurately reflect public opinion and if they change the representatives’ behavior accordingly, these brief contacts should marginally increase normative legitimacy. As with ‘considered’ public opinion, a plural and partial conception of normative legitimacy would consider a law more or less legitimate depending on the quality of the participation within the feedback loop that Lafont stresses between public opinion and political decision.

In my own analysis, that feedback loop is never perfect. We should see it as an aspirational ideal that can never be fully achieved. Both considered public opinion and citizen participation thus lie on a spectrum of quality (and perhaps quantity) that affects a polity’s legitimacy. Because the ideals within the feedback loop can never be fully achieved, the resulting legitimacy is always partial. One implication of stressing the spectrum of legitimacy is to suggest, in turn, a spectrum of deference. Here, although I agree with the main points in Lafont’s analyses of strong majoritarianism, judicial review, and minipublics, I disagree with the implied absolutism in her concept of ‘blind deference.’ Citizens themselves can and should be able to decide to defer to a majority, even knowing that the majority espouses views that they themselves cannot endorse. They can and should be able to defer responsibility to another institution, such as the judiciary. They can and should be able even to insulate that institution to some degree against their own immediate desires. When citizens (or – an important difference – prior citizens) make such decisions, their subsequent deference is not blind. The deference is not blind in its origin. It is also rarely fully blind as it plays out after the institutions have been created, at least if ‘blind deference’ means deference without resources to contest. That deference can, however, be stronger or weaker, and more or less defensible, with the more defensible forms contributing more to a democracy’s legitimacy. Spectrum thinking directs theoretical attention to the quality and degree of deference, some instances more justifiable than others, but few entirely blind.

II. The Central Principle: Owning, Identifying with, and Endorsing State Coercion

It is hard to exaggerate the importance today of Lafont’s identification and exploration of the central goal of dispelling alienation – helping citizens to own their own laws, identify with those laws, and endorse them. I have argued elsewhere that as nations internally and externally become more interdependent, they will face increasingly more free-rider (‘collective action’) problems, which arise when good outcomes require producing goods that anyone can use without contributing to producing them. (These ‘free-use’ goods range from unlittered sidewalks to common defense and a stable climate.) In these cases, each individual has self-interested incentives not to contribute but instead free-ride on others’ contributions. Many of these problems may be overcome by encouraging solidarity (including mutual empathy) and a sense of duty toward others – motives that lead citizens to contribute freely toward the common good. But that solidarity and duty will usually unravel if those who contribute see those who refuse to contribute benefitting consistently and significantly by free-riding on the others’ contributions. Seeing some free-ride successfully begets free-riding in others. In those instances, a little coercion on the periphery, making it unprofitable to free-ride, can provide an ecological niche within which the core motives of duty and solidarity among the cooperators can survive and thrive. In today’s societies of strangers, where interpersonal reputation and informal social sanctions can seldom produce sufficient coercion to manage this free-riding, some state coercion is required (Mansbridge 2014).

As the number and importance of collective free-rider problems, such as those involved in combating climate change, increase, the amount of state coercion needs to increase as well. In order to be both worthy of our trust and prudentially effective, that coercion must be both relatively normatively legitimate (resting on good reasons) and perceived to be relatively legitimate. Understanding the legitimacy of state coercion is thus one of the most pressing questions of our time. That understanding begins with a willingness to speak the word ‘coercion’ and recognize the burdens of justification that the fact of state coercion requires from the voters and representatives who benefit from, or think the polity as a whole will benefit from, the imposition of coercion on those who would otherwise act differently. This is Lafont’s conscious, central task. Her willingness to address directly the need for the justification of state coercion is what makes this book so important.

The democratic response to the challenge of legitimate coercion is, as Lafont points out, to make that coercion in some way self-given. Lafont borrows from Phillip Pettit the apt language of ‘own and identify with’ the laws, while moving from his stress on domination to a focus on alienation (Lafont 2019: 18–19).4 Her anchor in this fundamental principle responds to widespread and relatively justified feelings of alienation and not being heard in many democracies around the world, along with declining trust in government. It is a citizen-centered response.

III. Legitimacy: What to Do with Unattainable Ideals?

A. Institutions and opportunities

Lafont takes seriously the democratic ideal that citizens should ‘own and identify with the institutions, laws and policies’ that coerce them. As she points out, however, this is a demanding and not fully attainable ideal. Much of her book struggles with this problem of non-full-attainability. Lafont introduces the issue by writing, in the second paragraph of the first substantive chapter in the book, in a section worth quoting at some length:

Under the thickest, most demanding interpretation, the ideal of self-government could be understood as requiring that literally all those subject to the law would simultaneously be the authors of the law as well. Demanding that all members of the polity directly participate in making all political decisions to which they are subject would render the ideal incompatible with representative government and unsuitable for complex societies. This is not to deny that authorship in political decision-making is an inextricable component of the democratic ideal. … But … if the ideal of self-government does not literally require citizens to participate in making all political decisions, then, quite apart from authorship, we need to identify some other aspect of citizens’ participation that can illuminate what the ideal of self-government requires of representative democracies (17–18).

She continues,

The ideal that one should not be subject to laws that one cannot see oneself as an author of … does not require that one literally be an author of the laws, but it does require that one can obey them based upon insights into their reasonableness. One has to be able to identify with the laws or to reflectively endorse them (18).

As I read this passage, the key words in this particular escape from the over-demanding ideal of authorship are ‘be able to’ in the last sentence. That is, Lafont’s escape from pure authorship requires only the (genuine) opportunity, through appropriate institutions, to endorse the laws reflectively. It does not require actual identification or endorsement. Her later argument for judicial review, for example, rests on the opportunity it gives citizens to contest the laws ‘whenever they think the priority of public reason has been violated’ (210) and thus set the stage for mutual public justification and reflective endorsement.

Lafont does not make the concept of ‘opportunity’ central to her analysis.5 On the contrary, her words occasionally seem to suggest that legitimacy rests on each citizen’s actually endorsing the law, as in fact justified to them. For example, she writes, ‘According to the ideal of deliberative democracy, the legitimate exercise of political authority requires political decisions to be justified to those who are bound by them’ (132). The words ‘be justified to’ could be interpreted to require that each citizen accept the justification offered. Yet this is not her message. In Lafont’s ‘institutional approach,’

democratic legitimacy does not require every single person to agree on the reasonableness of each coercive law to which they are subject at any given time. Instead, such an approach requires institutions to be in place such that citizens can contest any laws and policies that they cannot reasonably accept by asking that either proper reasons be offered for them or that they be changed. To the extent that such institutions are available to all citizens, even to those who happen to find themselves in the minority, they can see themselves as equal members of a collective political project of self-government (12, emphasis mine).6

Although Lafont does not explicitly use the word ‘opportunity’ in this passage, her words, ‘can contest’ and ‘available to’ seem to denote the opportunity to contest and also to ask for proper reasons or change. This (presumably equal) opportunity should then allow citizens to see themselves as equal members of a collective political project of self-government. They ‘can’ see themselves this way. Lafont does not say that they ‘do.’

What happens, however, if, given this opportunity, citizens do not in fact see themselves as equal members of the collective project of self-government? We might say that the democratic institutions have done all they could by providing the opportunity and may claim full normative legitimacy without doing more. I conclude instead that what I would call Lafont’s ‘institutional opportunity’ approach produces only partial legitimacy.

B. A plurally-sourced, aspirational, and spectrum-oriented approach

In the book, Lafont does not take on directly the question of plural sources of legitimacy. Yet her approach is clearly compatible with plural sources. She persuasively takes to task those ‘deep pluralist’ and epistemic thinkers who insist that their source of legitimacy is the only valid source. She also frequently employs language implying a plural approach. She writes, for example, that ‘mutual justification … is a condition for democratic legitimacy’ (12). A specific practice such as public deliberation, judicial review, or even everyday political talk ‘contributes to’ legitimacy by enhancing the process of mutual justification (19, 29, 168; also 198, 213). The priority of public reason is ‘a necessary component of any plausible account of the legitimacy of the institutions of constitutional democracy’ (204; also 232). Certain ‘processes of political opinion- and will-formation in which citizens participate can lend legitimacy to political decisions’ (32; see also 75, 102, 135). The ‘legitimacy of many political decisions can be undermined not only for procedural but also for substantive reasons’ (75; also 76). ‘Political equality is necessary but not sufficient for democratic government’ (19, quoted above; all emphases mine).7

A pluralist approach does not entail a theory of partial legitimacy, but is highly compatible with it. When an institution delivers on only one value among many, it has only partial legitimacy. When it delivers on two values, it has greater legitimacy. If it is hard (and I would argue it is usually impossible) to achieve full legitimacy on even one dimension, it will be even harder to achieve full legitimacy on many.8

Building on the idea of partial legitimacy, one response to the over-demanding ideal of authorship itself is to understand such authorship as an aspirational ideal, that is, a regulative ideal that is impossible or almost impossible to reach, yet toward which democratic systems should strive. In this approach, degrees of democratic legitimacy rest in part on how closely a democratic system approximates this ideal.9

Lafont opposes such an aspirational approach in a later section of the book. This section begins by addressing Valier’s ‘Public Justification Principle (PJP): A coercive law (L) is justified in a public P if and only if each member i of P has sufficient reason(s) Ri to endorse L.’ (Lafont 2019: 180; Valier 2018). Lafont points out that this principle ‘makes the ideal so overdemanding as to fail to be action-guiding’ (180), just as does a strong interpretation of the democratic ideal of self-government. In response, she argues for her own institutional approach rather than either hypothetical or aspirational approaches. She spends four pages disposing of the hypothetical approach (as evidenced in Rawls) through the Habermasian path of pointing out that it avoids both the actual act of accepting and the actual acts of public deliberation that might lead to such acceptance (180–184). She then disposes of the aspirational approach in two paragraphs, which, following Rawls, create a straw man in which aspirational theorists are taken to argue that ‘when public justifications fail … there is no other acceptable option but to fall back on the legitimacy of a purely procedural solution such as majority rule’ (185). Therefore a citizen ‘does not have an obligation to withdraw support from a coercive law to which she cannot provide the requisite public justification’ and when disagreement occurs, such a citizen finds herself ‘abandoning public reason … altogether’ (idem). Neither I nor most sensible readers would espouse this form of aspirational theory. Why not simply posit that a democratic polity is more or less legitimate to the degree that it approaches a set of aspirational ideals? Citizens would be obligated to support and obey such partially legitimate laws under the general rubrics of mutual commitment and a common sense acceptance of reality. Their obligations would, however, decline as legitimacy declined dramatically. With such an aspirational, plurally-sourced, and spectrum-oriented understanding of legitimacy, no one need abandon a decent ideal simply because it cannot be fully achieved.

If Lafont were to adopt explicitly a plural and partial theory of legitimacy, that theory would generate throughout language slightly different from the language Lafont uses in this book. It would, for example, recast her central message to say that ‘a primary [or possibly the primary] goal in designing democratic institutions should be to increase the degree to which citizens can own and identify with the institutions, laws and policies that coerce them.’

IV. Blind Deference

Making the goal that citizens should ‘own and identify with’ the institutions, laws, and policies that coerce them gives Lafont good answers to theorists of both strong majoritarianism and agonism. It gives her original insights into the process of judicial review. It gives her an original and critical take on a central democratic innovation of our time: minipublics. But it leads her to a stance on ‘blind deference’ that a more spectrum-focused, or partial, theory would avoid. Just as it is impossible for any actually existing democracy to meet fully all democratic ideals, so it is extremely unlikely that any democracy would adopt institutions that incorporate full blind deference, with no capacity for citizen control. The questions for each institution are how much deference is required, how well is that deference justified, how democratically was the institution adopted, and what controls and avenues for contest remain after it is established.

In defining ‘blind deference,’ Lafont writes,

to the extent that citizens maintain some capacity for control over [the ‘representatives, officials, and so on’ to whom they delegate decisions], they are not doing so blindly. By contrast, deference is blind if there is no such capacity for control. The difference between the two can be explained as follows. In the first case, one has some (defeasible) reason to assume that the political decisions endorsed by the agent to whom one is deferring are those that one would have endorsed if one had thought through the issue with access to the relevant information. By contrast, in the second case, one has no reason to make this assumption. This is not to deny that we may have good reasons to blindly defer to the decisions of others. It is simply to point out that, whenever we do so, we are no longer engaging in a democratic project of self-government regarding those decisions. To the contrary, what we have determined is that these decisions should track their considered judgments instead of ours and that we will blindly follow them, whatever they happen to be. An expectation of blind deference is quintessentially incompatible with the democratic ideal of self-government (Lafont 2019: 219).

In my reading, this definition has two components, the procedural and substantive. I see both components of non-blindness as aspirational and the resulting legitimacy as therefore always partial. In the following analysis, I stress the procedural feature of capacity for control rather than the substantive feature of having reason to assume that the decisions of the agents to whom one is deferring will be such that one would endorse them oneself in the right conditions.10 I adopt this stress because of Lafont’s own point that ‘democratic legitimacy does not require every single person to agree on the reasonableness of each coercive law to which they are subject at any given time. Instead, such an approach requires institutions to be in place such that citizens can contest any laws and policies that they cannot reasonably accept by asking that either proper reasons be offered for them or that they be changed’ (12, quoted above p. 6). These sentences mark a shift from a substantive demand for agreement on the reasonableness of each coercive law to a procedural demand for institutional opportunities to contest. I also stress the procedural component because, almost by definition, in a democracy (demokratia), normatively the people underlyingly should rule (have control). They do not have to rule actively in every case. But in order to make the source of the coercion they experience from their government their own, they have to have authorized, explicitly or far more often tacitly, the institutions, such as Supreme Courts, central banks, administrative bodies, and perhaps in the future empowered minipublics, that coerce them without any direct accountability. If the people had no power to revoke or change these institutions, the institutions would not be compatible with democracy. Yet it is not contrary to democracy more deeply understood for the people to put some institutions beyond their own immediate direct control. Human beings are learning machines. We have learned over time that our goals – our own goals – are better achieved by putting some of our institutions beyond our immediate control. In a democracy, we the people take this step and we must be able to revoke it, even with self-imposed difficulty. When we take this step, we do not do so blindly.

Lafont argues convincingly that we can continue without blind deference,

[s]o long as there are effective and ongoing possibilities for citizens to shape the political process as well as to prevent and contest significant misalignments between the policies they are bound to obey and their interests, ideas, and policy objectives (23).

With these possibilities (i.e., procedural opportunities) in place, citizens ‘can continue to see themselves as participants in a democratic project of self-government’ (23). Without these possibilities, deference is ‘blind.’ The problem is that none of the theorists whose analyses Lafont finds unsatisfactory would, as I read them, argue against the importance of maintaining these effective and ongoing procedural possibilities.

A. Deference to a majority

Agonists and majoritarians sometimes suggest that proponents of deliberative democracy assume that all political questions in a democracy can be settled by citizens deliberating to a consensus. Such suggestions are incorrect. Today’s deliberative democrats almost all see deliberation as rightly leading to consensus in some contexts and clarifying conflict in others (see Bächtiger et al. 2018: Table 1). Lafont’s valuable response to these misinterpretations distinguishes between contexts. She insists that, over time, some conflicts can be settled substantively through discussion and something close to final agreement (for example, the conclusion that slavery is morally abhorrent). I would add that even some shallower conflicts can be settled substantively through discussion when the discussion unearths relevant facts and insights of which some were unaware before discussion. Many other conflicts, however, must be ‘settled’ on a more or less temporary basis ‘by compromise, majority decisions, bargaining, negotiation, mutual accommodation, etc.’ ‘Indeed,’ she writes, ‘whenever there is a spectrum of equally reasonable answers to political questions, then decisions by compromise, negotiation, and bargaining are just as adequate as a consensus upon a single answer’ (38).11

This easy acceptance of bargaining marks a distinct evolution within the Habermasian tradition. Jürgen Habermas himself began this evolution. In his early work, Habermas strongly disapproved of ‘bargaining.’ As he began to develop his ideas on the public sphere and the legitimating force of the better argument, his 1962 Transformation of the Bourgeois Public Sphere deplored the disintegration of that legitimating force. In one of many examples, he wrote that ‘the creation of collective bargaining regulations … shatters the forms of the old style public sphere (founded on trust in the power of reason).’ (Habermas [1962] 1989: 199; see also Mansbridge 2012). Habermas did not see how a politics founded only on ‘trust in the power of reason’ could apply once the franchise was extended to include classes outside of the assumed homogeneity of interests within the specific and limited ‘bourgeois’ public sphere. By 1992, however, his thoughts had changed. He still claimed that ‘only those statutes may claim legitimacy that can meet with the assent (Zustimmung) of all citizens in a discursive process of legislation that in turn has been legally constituted.’ (Habermas [1992] 1996: 110; emphasis mine).12 Yet now he addressed more positively those situations in which no ‘generalizable interest or clear priority of some one value’ is ‘able to vindicate itself’ (165). In these situations, he allowed ‘the alternative of bargaining, that is, negotiation between success-oriented parties who are willing to cooperate’ (165). The parties might be oriented toward winning, not toward understanding. They might also rely on ‘threats and promises, thereby introducing … power into their interaction’ (166). Nevertheless, a principle by which ‘just those action norms are valid to which all possibly affected persons could agree as participants in rational discourses’13 (the ‘discourse principle’) can ground ‘procedures that regulate bargaining from the standpoint of fairness’ (166, emphasis in original).

Lafont moves this line of reasoning further in several ways. First, she explains how the ‘unforced force of the better argument’ can work in today’s public sphere through the communicative power of citizens to shape public opinion. Second she shows in persuasive detail how, in cases such as the civil rights movement in the United States and movements for women’s and gay rights in the US and elsewhere, the power of uninvolved third parties in the public sphere who are convinced by better arguments, ‘can provide politically disempowered parties with additional leverage in their political struggles’ (68). Third and relatedly, against agonists who argue that a goal of ‘“deliberation aimed at agreement” fails to properly attend to background power relations and power inequalities that surround political struggles’ (64), she shows how the judiciary, randomly selected minipublics, and other democratic institutions that promote political deliberation can empower ‘those on the weak side of power inequalities’ in these political struggles (65). Deliberation can affect power. It is not the case that deliberation (in its ideal form uncontaminated by power) is legitimate and bargaining (which includes power) is not.14

Most importantly, in this section Lafont argues against the claims of strong majoritarians/‘deep pluralists’ that ‘persistent disagreement on political questions cannot be reasonably settled’ (Lafont 2019: 43 citing Besson 2005: 527). She writes instead15 – that a political question can be said to be ‘settled’ in many different senses of the word. It can be settled through substantive agreement (as discussed in the first paragraph of this section). It can be ‘settled’ temporarily by applying a procedure such as majority rule to settle, say, the question of the next president of a polity (42). It can also be settled through arriving at a ‘shared view on [the question’s] proper answer,’ even when that proper answer is ‘whatever is the fairest compromise’ in the light of some considerations (43). That proper answer cannot, however, be simply ‘whatever majority rule decides,’ because there are both good arguments for and settled citizen support for the ‘modified majoritarian’ position that a set of rights should limit majority rule, at least temporarily.16 Lafont argues that even if there is no full agreement on what rights apply in what cases, for citizens to make the laws their ‘own’ requires that claimed violations of rights be reviewed, evaluated, and given ‘a proper process of public scrutiny’ (61), in a ‘recursive’ process that takes away the ‘finality’ of majority rule (62). This process may even ‘over time enable a settled view on rights to come about’ (63). She reminds readers of the ‘good old hermeneutic platitude that disagreement presupposes agreement’: citizens may disagree on whether hate speech should be protected under free speech rights, but that very disagreement presupposes some relatively settled agreement on protecting free speech; they may disagree on the right not to be raped in marriage, but that very disagreement presupposes some relatively settled agreement on a right not to be raped (64, 65).

By stressing a distinction between questions of fundamental rights and other political issues, Lafont makes it clear why we should not engage in ‘blind deference’ to majority rule. She also sets the stage for her defense of the judiciary. My only disagreement with these arguments lies in my uncertainty as to whether the strong majoritarians against whom she argues actually advocate full blind deference. They would surely argue that majority rule itself provides ‘effective and ongoing possibilities for citizens to shape the political process’ by allowing new majorities to form and overrule the old.

B. Deference to a judiciary

Lafont argues tellingly that judicial review can play an important and citizen-initiated role in opening up nation-wide and occasionally global processes of mutual justification in the crucial realm of fundamental rights.

She begins her defense of judicial review by pointing out that it is ‘inaccurate’ to characterize ‘judges as having final authority to determine questions of rights’ (220). In the United States and elsewhere, citizens can amend the constitution. Moreover, and crucially for Lafont’s argument, judicial decisions, coupled with the opportunity to bring a case involving one’s conception of rights before the judiciary, can provoke and encourage widespread citizen deliberation on those rights. The Irish High Court’s 2002 decision defining marriage as only between a man and a woman exemplifies a judicial act that, far from being taken as final, helped prompt and played a focal role in a recursive process of citizen deliberation in the public sphere, resulting in changed public opinion and then policy change through a majority-vote referendum (220–221). No blind deference here. Instead, the decision was followed by a lively, contested, participatory process in which citizens discussed with and explained their views to one another in the public sphere, asking one another to, so to speak, ‘put on their robes’ to think about the issues as matters of fundamental rights (240). In what I believe is completely original move, Lafont sees judicial review as a ‘conversation initiator’ (228–233).

Lafont is under no illusions as to the equality of access of all citizens to the processes of judicial review. She points out, however, that in the US and elsewhere social movements from below have sometimes had considerable success in taking cases to the courts based on an infringement of their members’ rights well before public opinion, often helped along by judicial decision, evolved in their direction. Moreover, as she also points out, in almost every democratic polity the most powerful citizens have significantly shaped almost all ‘democratic’ institutions to benefit themselves. The wealthy heavily influence elections as well as judicial review, central banks, and the like. The greater influence of the wealthy significantly reduces the democratic legitimacy of all of these institutions. But that is their great democratic flaw, not the fact that some of these institutions are, with different degrees of revocability, not under the citizens’ immediate control.

C. Deference to minipublics

Regarding randomly selected minipublics, Lafont agrees with many other deliberative democrats that, at present, democratic polities should confine themselves almost entirely to convening advisory, not decisional or empowered, minipublics. We do not yet have enough experience, knowledge, or public comfort with the process of random selection to make it prudent to delegate much decision-making power to such groups. Unlike Lafont, however, I would not rule out giving such groups more decision-making power in the future. I consider such a move completely within the parameters of democratic action. I would therefore not consider the result ‘blind deference.’

If the people or its representatives (not the same thing) at some point in the future decide to delegate some decisions to randomly selected minipublics (after we collectively figure out what these minipublics are best at), as long as the citizenry can monitor them and keep the residual power to revoke them, this would not be blind deference. At the moment in all democracies we delegate immense powers to administrations. We must do so, in order to produce effectively the huge amounts of coercion that we need to sustain the duty and solidarity that keep our large complex apparatus of social cooperation going (Mansbridge 2001). Adding randomly selected minipublics to administrative systems currently makes those systems more, not less, participatory. If the administrators were to edge toward pledging to accept the minipublics’ recommendations, thus making the minipublics more empowered, this would not, in my view, make the citizens’ deference to the result more ‘blind.’

Currently, almost all minipublics of randomly selected citizens who meet and deliberate for a weekend, two weekends, or even a year, produce only advisory opinions. They advise administrations, citizens, and occasionally legislatures. Lafont stresses the positive effects such advisory minipublics have and might have on the larger projects of strengthening the citizen-representative feedback loop and increasing the chances of citizens’ being able to ‘own’ the law. Having the minipublics’ assessments available to citizens would be, she writes, ‘a tremendous improvement over the status quo.’ Institutionalizing minipublics on a regular basis would ‘help improve the quality of public deliberation and thus contribute to the formation of considered public opinion.’ It would also ‘strengthen the link between public deliberation and actual political decisions thereby enhancing the responsiveness of the political system to the considered opinions of the citizenry, as the democratic ideal of self-government requires’ (156). Moreover, for ‘novel issues that the public is not yet familiar with, minipublics could contribute to the process of framing public debate by articulating discourses that are most suited to evaluate the issues at hand but which have not yet been developed in the public sphere’ (159). The minipublics could also help sort through the available ‘information, reasons, and arguments’ (159) to determine which are most relevant for assessing public policies.

In the future, if democracies were to decide to dedicate resources to creating far more minipublics than today, almost every citizen could have the experience of advising on the law. The increased use of advisory minipublics could have a large positive influence on the public experience of both mutual justification and owning the law. Empowering a few such minipublics would probably increase that experience. If some minipublics were given decision powers, then some citizens might have the empowered experience of, as Aristotle put it, ‘ruling and being ruled in turn.’

If some minipublics were to become relatively empowered, of course, the world around them would change, as Christopher Karpowitz has noted.17 Social movements, interest groups, political parties, and the wealthy would quickly develop interests in influencing these minipublics. We cannot know now what protections these minipublics would need, but just as the world of mutual justification would be enlarged, so would the opportunities for corrupting that process. These considerations counsel moving slowly and incrementally. They do not rule out full empowerment on the grounds that it would produce blind deference.

V. Conclusion

Cristina Lafont’s timely analysis in Democracy without Shortcuts stresses the importance to normative democratic legitimacy of citizens owning and identifying with the institutions, laws, and policies that coerce them. Human organization will depend in the future on increasing amounts of state coercion. Yet in many of today’s democracies the vast majority of citizens (often rightly) do not own or identify with their governing institutions and the laws they promulgate. It is critical, therefore, to invest significant human effort in the project of creating legitimate coercion.

Democracy without Shortcuts responds to this need. It argues persuasively for recentering the discussion of democracy on the goal of citizen ownership of the laws. Seeing old institutions anew through this sharply citizen-centered lens, Lafont gives the institution of judicial review, formerly seen as exemplifying distance from the citizens, a citizen-initiated and citizen-centered use. By broadening the field of citizen participation, she explains how citizens actively make the public sphere. She shows how public justification, no longer appropriate only for the ‘bourgeois public sphere,’ can work in every citizen’s life. She pens refutations of agonism and strong majoritarianism that start with what citizens themselves think they need.

I have argued for an approach to legitimacy somewhat different from Lafont’s, but have shown that it is compatible with much of Lafont’s language and almost all of her analysis. The plural, aspirational, and partial approach I advocate may not be compatible with all of the details of her critique of ‘blind deference,’ but it is completely compatible with her stress on mutual justification. Today that stress is more important than ever, as many cosmopolitans dismiss the concerns of populists rather than taking the responsibility for mutual justification seriously, distinguishing between demands that violate fundamental rights and other concerns. Lafont’s analysis is extremely valuable for today and for the future. It puts the citizen at the center and takes seriously the citizens’ capacities for reflectively endorsing the laws that coerce them.


  1. Lafont cites Jürgen Habermas, Between Facts and Norms ([1992] 1996), chapter 7 and ‘Political Communication in Media Society’ (2009). For another version of the desirability of a feedback loop, see Mansbridge (2019). (Because my work intersects so frequently with Lafont’s, I will cite several of my articles in this comment.)
  2. Citing Habermas ([1992] 1996: 329) and Parkinson and Mansbridge (2012: 138 and 176).
  3. By contrast, government workers in Demark survey citizens on the actual site of government service delivery, asking for their perceptions and satisfaction regarding various aspects of the service (Mark Warren, personal communication), with those surveys then presumably being used to improve service. This experience is likely to produce a warranted sense of being heard.
  4. Citing Pettit (1997: 184). ‘Making the law one’s own’ can similarly serve as a first principle in democratic theory and practice (see Mansbridge 2020).
  5. Nevertheless, mentions of ‘opportunity’ appear on, e.g., pp. 57, 75, 110, 139, 144, 230.
  6. Lafont also adopts several alternate phrasings that are not opportunity-based. One is ‘the criterion of democratic legitimacy that deliberative democrats endorse, according to which citizens ought to justify the imposition of coercive policies on one another with reasons that everyone can reasonably accept’ (194, 198). Another, more substantively different, is: ‘the democratic legitimacy of political decisions depends upon their ability to track the perceived interest and ideas of those subject to them’ (136), a phrasing that goes back to Pettit (1997) as quoted on p. 18 (see also p. 201). Both criteria lend themselves to partial fulfillment.
  7. In a review of Nadia Urbinati’s (2014) Democracy Disfigured: Opinion, Truth, and the People, Lafont explicitly endorsed a ‘pluralist strategy’ for legitimacy that values both epistemic and non-epistemic (e.g., procedural) justifications. She concluded that when institutions can ‘deliver on both values (that is, reasonable outcomes and political equality) their legitimacy is superior to that of political systems that sacrifice one for the sake of the other’ (Lafont 2015: 327). I thank Cristina Lafont for directing me to this review.
  8. Lafont does not explore explicitly the many possible sources of legitimacy (although she suggests the sources of political equality, reflective endorsement, active participation, and epistemic correctness) or the relation of these plural sources to authorship. One might argue that authorship is the singular source of legitimacy but plural streams contribute to full and legitimate authorship. Any proposed source of legitimacy that does not run through authorship would then not be a genuine source of legitimacy. This is not my own stance, as I accept sources of legitimacy that are relatively non-participatory, such as the legitimacy of delegated, insulated institutions. I also accept sources of normative legitimacy outside the democratic tradition, including trust in bureaucracies that reliably deliver good services (I thank one of the anonymous reviewers for this wording) as well as the merit and good will of powerful actors above and beyond their delegated legitimacy from the people (in Confucian philosophy their ‘virtue,’ as described in Chan, 2014). I urge further thought within democratic theory on the plural sources of legitimacy.
  9. I adopt this approach in ‘Recursive Representation’ (Mansbridge 2019) and ‘Representation Failure’ (Mansbridge 2020).
  10. The wording of the following section responds to some of Lafont’s (2020) comments in ‘Replies,’ an unorthodox move in a symposium such as this, but one that I hope will be acceptable in the service of our common project of clarifying the issues.
  11. Looking to ‘interests’ as well as to ‘reasoned answers,’ one might also make a contingent argument. Contexts of relatively common interests are more suited to political approaches that try to settle substantive disputes through deliberation to consensus. Contexts of relatively conflicting interests are more suited to ‘adversary’ political approaches in which deliberation both clarifies conflict and produces agreement on a procedure for handling that conflict by, for example, legitimating power through making it equal and deciding by majority rule. Effective and normatively approvable political institutions find ways of moving between approaches depending on the context (Mansbridge 1980).
  12. This ‘democratic principle’ might be parsed as a hypothetical (Estlund 2009: 184), were it not for Habermas’s careful use of können (‘can’) here in contrast to könnten (‘could’) in the ‘discourse principle’ (see note 13 below) and his explicit aim to avoid the hypothetical: ‘[T]he justification of norms and commands requires that a real discourse is carried out and thus cannot occur … in the form of a hypothetical process of argumentation occurring in the individual mind’ (Habermas 1990: 68; cited in Lafont 2019: 183). Rosen (1999), however, suggests that Habermas’s attempt to avoid the hypothetical is untenable.
  13. Ibid., 107 (‘could’/könnten). I thank Isaac Nakhimosvsky for help on these meanings.
  14. Warren and Mansbridge with colleagues (2015) suggest a spectrum within legitimate democratic interaction, running from pure deliberation through fully integrative negotiation, partially integrative negotiation, and fair compromise to pure power-based bargaining (154, Table 5.1). Power-based bargaining, like the power in elections, is democratically legitimate only to the extent that the power is equal among citizens.
  15. As Lafont and others have pointed out (44), majority rule cannot be the definitive answer to deep disagreement because some will always disagree about the fairness of that procedure (in comparison to, say, proportional outcomes when some group is in a permanent minority).
  16. The term ‘modified majoritarianism’ comes from Jeremy Waldron (1999). Waldron is the foremost ‘deep pluralist’ and opponent of judicial review to whom Lafont addresses her arguments.
  17. Comment, American Political Science Association annual meeting, September 2018.

Competing Interests

The author has no competing interests to declare.


1 Bächtiger, A., Dryzek, J., Mansbridge, J., & Warren, M. E. (2018). Introduction. In A. Bächtiger, J. Dryzek, J. Mansbridge, & M. E. Warren (Eds.), The Oxford handbook of deliberative democracy. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oxfordhb/9780198747369.001.0001

2 Besson, S. (2005). The morality of conflict: Reasonable disagreement and the law. Oxford: Hart Publishing.

3 Chan, J. (2014). Confucian perfectionism: A political philosophy for modern times. Princeton, NJ: Princeton University Press. DOI:  http://doi.org/10.23943/princeton/9780691158617.001.0001

4 Estlund, D. (2009). Democratic authority: A philosophical framework. Princeton, NJ: Princeton University Press. DOI:  http://doi.org/10.1515/9781400831548

5 Habermas, J. ([1962] 1989). The structural transformation of the public sphere, trans. T. Burger & F. Lattimore. Cambridge, MA: MIT Press.

6 Habermas, J. (1990). Moral consciousness and communicative action, trans C. Lenhardt & S. W. Nicholsen. Cambridge, MA: MIT Press.

7 Habermas, J. ([1992] 1996). Between facts and norms, trans. W. Rehg. Cambridge, MA: MIT Press. DOI:  http://doi.org/10.7551/mitpress/1564.001.0001

8 Habermas, J. (2009). Political communication in media society in Europe: The faltering project. Cambridge: Polity.

9 Lafont, C. (2015). Review of Democracy Disfigured: Opinion, Truth, and the People by Nadia Urbinati. Constellations, 22(2), 326–328. DOI:  http://doi.org/10.1111/1467-8675.12165

10 Lafont, C. (2019). Democracy without shortcuts. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oso/9780198848189.001.0001

11 Lafont, C. (2020). Against anti-democratic shortcuts: A few replies to critics. Journal of Deliberative Democracy, 16(2), 96–109. DOI:  http://doi.org/10.16997/jdd.367

12 Mansbridge, J. (1999). Everyday talk in the deliberative system. In S. Macedo (Ed.), Deliberative politics (pp. 211–239). Oxford: Oxford University Press.

13 Mansbridge, J. (2001). A ‘moral core’ solution to the prisoners’ dilemma. In J. W. Scott & D. Keates (Eds.), Schools of thought: Twenty-five years of interpretive social science (pp. 330–347). Princeton, NJ: Princeton University Press.

14 Mansbridge, J. (2012). Conflict and commonality in Habermas’s Structural Transformation of the Public Sphere. Political Theory, 40(6), 792–804. DOI:  http://doi.org/10.1177/0090591712457661

15 Mansbridge, J. (2014). What is political science for? Perspectives on Politics, 12(1), 8–17. DOI:  http://doi.org/10.1017/S153759271300368X

16 Mansbridge, J. (2019). Recursive representation. In D. Castiglione & J. Pollak (Eds.), Creating political presence (pp. 298–337). Chicago, IL: University of Chicago Press.

17 Mansbridge, J. (2020). Representation Failure. In M. Schwartzberg & D. Viehoff (Eds.) NOMOS LXIII: Democratic failure. New York: New York University Press.

18 Parkinson, J., & Mansbridge, J. (Eds.) (2012). Deliberative systems: Deliberative democracy at the large scale. Cambridge: Cambridge University Press.

19 Pettit, P. (1997). Republicanism: A theory of freedom and government. Oxford: Oxford University Press.

20 Rosen, M. (1999, October 8). Utopia in Frankfurt: Realism and optimism in the life’s work of Jürgen Habermas. Times Literary Supplement, no. 5036, 3–4.

21 Valier, K. (2018). Public Justification. In Stanford encyclopedia of philosophy. Retrieved from https://plato.stanford.edu/entries/justification-public/.

22 Waldron, J. (1999). Law and disagreement. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/acprof:oso/9780198262138.001.0001

23 Warren, M. E., & Mansbridge, J., et al. (2015). Deliberative Negotiation. In J. Mansbridge & C. J. Martin (Eds.) Political negotiation (pp. 141–198). Washington, DC: Brookings.