Self-Respect and the Importance of Basic Liberties

1 Introduction

A common idea in the literature on self-respect and recognition is that self-respect is an important good and that social arrangements should therefore be evaluated, at least in part, by how they affect people’s self-respect. As Robin S. Dillon (2003) puts it: “Self-respect is frequently appealed to as a means of justifying a wide variety of philosophical claims or positions, generally in arguments of the form: x promotes (or undermines) self-respect; therefore, x is to that extent to be morally approved (or objected to)” (23). In particular, various accounts of self-respect are used to defend the importance of an equal distribution of basic liberal rights and liberties. I will call this “the self-respect argument.” It is that (a) self-respect is highly important to autonomy and/or welfare and (b) best supported by basic liberties and their equal distribution, and that (c) such a distribution is therefore important. The idea is to give a reason for prioritizing the equal distribution of basic liberties above other distributive principles.

We find different versions of the self-respect argument in the literature on recognition and self-respect, where it is debated how self-respect must be conceptualized for the argument to work, and whether it can succeed at all. Two of the most known, widely applied and discussed versions of this argument are those of John Rawls and Axel Honneth. The appeal to self-respect plays an important role in both of their theories. In Rawls’s theory, the importance of self-respect is used, among other considerations, to support the contention that the “liberty principle” should enjoy a lexical priority^[1] above other distributive principles. The liberty principle is, roughly, that standard basic liberal rights and liberties are to be distributed equally (Rawls 1993).^[2] In Honneth’s early framework, before Freedom’s Right (2014), the argument from self-respect is arguably the most central reason for defending a similar position, a system of equal and at least highly prioritized legal rights (Honneth 1995, ch. 5). Several authors have taken up Rawls’s and Honneth’s approaches and followed similar lines of reasoning. However, there is also an ongoing discussion as to whether considerations of self-respect actually lend any support to the liberty principle and its special status (Brennan 2016, ch. 5; Eyal 2005; Massey 1995; Schemmel 2018).

This paper examines the self-respect argument, specifically versions of it that contend that self-respect is best supported by the liberty principle for reasons related to the recognition that this principle conveys. In short, I argue that (a) the two standard approaches loosely associated with Rawls and Honneth fail to give us a reason to prefer central parts of the traditional liberty principle above certain alternatives worth taking seriously, and that (b) there is a version of the argument that at least shows that the liberty principle is more important to autonomy than what we might otherwise have reason to believe. I begin (Section 2) by examining the approach associated with Rawls (or, rather, the parts of his argument), which is based on an evaluative conception of self-respect. Here self-respect is based on an evaluation of your character, traits and conduct.^[3] According to Rawls (1971, 440), to have self-respect is to be convinced that your conception of the good is worth pursuing and that your character is fit to pursue it. Self-respect has prudential value because it is a condition for enjoying other goods and for being motivated to pursue your aims. Furthermore, Rawls (1971, 546) argues that self-respect is best supported by the liberty principle and its priority because this signifies our equal status. I reject this approach because even though this evaluative form of self-respect is valuable, we have no strong reasons to think that it is better supported by the liberty principle than by certain alternatives worth taking seriously.

The other approach (Section 3), more associated more with Honneth and recognition theory, is based on a “personhood view” (Dillon 1995, 29) of self-respect, according to which having self-respect means respecting your moral status as a person. Such self-respect is not based on an evaluation of your character or conduct, but rather on the mere fact that you are a person. Honneth (1995, ch. 5) argues that self-respect thus understood is a condition for autonomy and that it is dependent on respect in the form of equal legal rights because this signifies our moral status as persons. I reject this approach because it is plausible that personhood self-respect is better supported by the liberty principle than by certain reasonable alternatives only if it is interpreted in such a narrow way that it is not a condition for autonomy (or welfare for that matter) in any plausible sense.

In the last two sections, I provide a version of the self-respect argument based on a third kind of self-relation that does not quite fit either of the above views of self-respect. Drawing on the evaluative and personhood view of self-respect and the literature on relational autonomy, I isolate a form of self-regard that is necessary for exercising autonomy (Section 4). I call it “a sense of competence.” To have a sense of competence is to trust your own capacities for valuation and reflection, and your ability to act in accordance with them. This differs from self-respect based on moral personhood because it does not entail any specific moral beliefs. It is therefore a more plausible condition for autonomy. Moreover, a sense of competence differs from evaluative self-respect because it does not entail the conviction that your aims are particularly worthwhile or that your character is particularly fit to pursue them. In the end (Section 5), I suggest that the best political social basis for this self-relation is the liberty principle because it is the principle that best signifies recognition of the capacity of citizens to make sound decisions. This shows that the liberty principle is more important to autonomy than we might otherwise have reason to believe.

The motivation for this paper is that while there are already good reasons in the literature to doubt standard self-respect arguments for basic liberties, more needs to be said, given that the merit of such arguments is far from settled. This paper identifies some problems that, to wit, have not been clearly and thoroughly exposed before. In addition, the argument from a sense of competence shows that basic intuition is not completely without merit. Indeed, it can be understood as a part of the standard arguments since there are many connections and overlaps between the various self-relations.

2 Evaluative Self-Respect and Basic Liberties—Rawls

Rawls’s account of what he interchangeably calls “self-respect” and “self-esteem” has two aspects in A Theory of Justice. It includes, firstly, “a person’s sense of his own value, his secure conviction that his conception of his good, his plan of life, is worth carrying out. And secondly, self-respect implies a confidence in one’s ability, so far as it is within one’s power, to fulfill one’s intentions” (Rawls 1971, 440).

In Political Liberalism, Rawls complements this account of self-respect by explaining its connection to his political ideal of personhood. In Rawls’s ideal of personhood, what characterizes such beings, and makes them free and equal, are two fundamental moral powers: rationality and reasonability. To be rational means to have “the capacity to form, revise and pursue a rational plan of life” (Rawls 1993, 318). To be reasonable means to be “ready to propose principles … as fair terms of cooperation and to abide by them willingly, given … that others will likewise do so” (Rawls 1993, 49).^[4] Moreover, persons are also understood as being engaged in a determinate life plan suited to their particular character. On the basis of this political conception of persons, Rawls derives three “higher-order interests” of citizens: the interest in developing and exercising each of the two moral powers and the interest we have in pursuing our determinate life plan. From these interests Rawls (1993, 307) derives an account of “primary goods,” which are social goods that are “generally necessary” to satisfy our higher-order interests. His principles of justice concern the distribution of these goods. The main primary goods are basic liberties, income, opportunities, powers and prerogatives of social power, and the social bases of self-respect (Rawls 1993, chs. 3, 5).

In relation to this ideal, Rawls describes self-respect as having two elements. The first element is “our self-confidence as a fully cooperating member of society rooted in the development and exercise of the two moral powers” (Rawls 1993, 319). The second is “our secure sense of our own value rooted in the conviction that we can carry out a worthwhile plan of life” (Rawls 1993, 319).

Rawls (1971, 440, 68, 178–79; 1993, 318) argues that the social basis of self-respect is “perhaps the most important primary good” because it is perhaps the most important means for pursuing a rational life plan. The reason for this is that without self-respect we lack the necessary motivation to pursue our life plan. As Rawls (1971) says, “Without [self-respect] nothing may seem worth doing, or if some things have value for us, we lack the will to strive for them” (440).

There is a considerable debate on what exactly Rawls means by “self-respect.” For the purposes of this section, I follow a standard interpretation according to which Rawls operates with an evaluative form of self-respect, albeit it a broad one that also includes what some would say falls under self-esteem (Dillon 1995, 32; Doppelt 2009; Eyal 2005). As Dillon (1995) formulates it “For Rawls self-respect … [consists of] a) confidence in the worth of one’s ‘plan of life’ – the aims and ideals one seeks to attain … b) the assessment of one’s character and abilities as well suited to successfully living that life” (27–28). This strikes me as a plausible interpretation of the two aspects of self-respect as Rawls most often describes it in A Theory of Justice and in Political Liberalism. Here self-respect is grounded in your character, traits and conduct in the sense that it concerns the merit of what you are doing and your ability to pursue your aims and ideals.^[5]

According to Rawls (1971), the parties in the original position would “avoid at almost any cost the social conditions that undermine self-respect” (440). Therefore, “the fact that justice as fairness gives more support to [self-respect] than other principles is a strong reason … to adopt it” (Rawls 1971, 440). While the exact status of self-respect, including how it ought to be distributed, is somewhat unclear in Rawls, it is common to interpret him as claiming that the social bases of self-respect must be distributed equally (Doppelt 2009, 128; Eyal 2005, 197; Rawls 1971, 546; Zaino 1998, 742). By social bases, Rawls arguably means social conditions that either affirm people’s self-respect or guarantee them opportunities to achieve self-respect, together with their public justification (Doppelt 2009, 135–37; Eyal 2005, 207; Rawls 1971, 544–46). The social bases of self-respect are, according to Rawls, primarily equal basic rights and liberties. He therefore claims that his two principles of justice provide citizens with an optimal basis for self-respect, at least to the extent that principles of justice can do so. “It is argued that self-respect is most effectively encouraged and supported by the two principles of justice, again precisely because of the insistence on the equal basic liberties and the priority assigned to them, although self-respect is further strengthened and supported by the fair value of the political liberties and the difference principle” (1993, 318).

Rawls’s (1971) main argument for this is that what matters most to our self-respect is our status, the worth we have in the eyes of others: “Our self-respect normally depends upon the respect of others. Unless we feel that our endeavors are honored by them, it is difficult if not impossible for us to maintain the conviction that our ends are worth advancing” (178). And the liberty principle—both the content and priority—is the principle that best conveys our full and equal worth, at least on some of Rawls’s formulations (1999, 155–58, 477; 1993, 318–20). Therefore, the liberty principle serves as the main social basis of self-respect. Note however that the second principle of justice requires fair equality of opportunity and social and economic inequality must benefit the least advantaged members of society—both of which also serve to affirm our status as citizens.^[6] I will call claims to the effect that certain distributive principles affirm our worth rather than provide us with means of attaining self-respect “status arguments.” It should be added that, in addition to signaling our worth, basic liberties also serve as means for attaining self-respect. This is because, firstly, they enable one to participate in associations in which one’s worth is recognized, and, secondly, they are necessary, according to Rawls (1971, para. 67; 1993, 318), to guarantee an adequate development and exercise of our moral powers which the second element of our self-respect is rooted in.

These considerations of how our evaluative self-respect is most effectively supported by society is one of Rawls’s arguments for both the content and lexical ordering of his principles of justice. In the following, I will discuss this argument. My focus will be on the claim that the liberty principle affirms our status (henceforth the “status argument”) in the evaluative sense.^[7] However, I shall not try to interpret what exactly Rawls means by the claim that the liberty principle offers recognitional support to evaluative self-respect. Rather, I shall argue that there are some general problems with this kind of reasoning. I will not say much about the value of evaluative self-respect. It is hard to deny that it is important for humans to have evaluative self-respect in Rawls’s sense, that is, a sense that their life plan is valuable and that their character is suited to pursuing it. We need not establish more than this to say that if the liberty principle is important to evaluative self-respect, then this gives us a pro tanto reason to favor it.^[8]

One problem with the claim that the liberty principle affirms our evaluative self-respect is that it cannot affirm the value or merit of our particular aims and ideals. To see this more clearly, it might be helpful to distinguish between two ways in which a determinate life plan can be recognized. The first is respect for the mere fact that I have a determinate life plan. Such respect simply consists in taking into account that we have a determinate life plan that we wish to pursue. When I recognize your life plan in this way, I simply take into account that it is important to you to pursue it when considering my own actions. This implies no evaluation of the plan’s worthwhileness, apart from the value it has for you. According to Rawls (1993, ch. 5), what is to be recognized by principles of justice is that pursuing some determinate life plan suited to our character is one of our “higher-order interests,” not the objective worthwhileness of this or that plan. Drawing on Stephen L. Darwall (1995), we can describe this as “recognition respect” for the fact that you have life plan x. I take this kind of recognition to be part of respect for citizen’s moral status as persons, not their particular activities, conduct or character (see Section 3).

The second way of recognizing a life plan is by recognizing its worthwhileness. We may call this appraisal respect or esteem, again following Darwall (1995). This is primarily the kind of recognition of your life plan that affirms your evaluative self-respect. Suppose my life plan is to be a professional football player so that I respect my character and traits to the extent that I am convinced that this pursuit has merit and feel confident in my ability to pursue it. Now, mere recognition respect for the fact that I want to be a professional football player—that others take into account that this is important to me because they respect my right to live as I want—does not do much to affirm my conviction that being a football player is worthwhile. It rather affirms my sense that my own interests have weight vis-à-vis those of others, which, again, falls under the personhood view of self-respect.

With this distinction in mind, I think it is clear that the liberty principle cannot affirm the merit of our particular aims and ideals. Nir Eyal (2005, 203–4, 207–10) argues that the liberty principle, but also wealth egalitarianism, “rationally” affirms the worthwhileness and feasibility of our determinate plans of life, no matter their content. I think this is mistaken. Firstly, no distributive principle affirms the merit or worthwhileness of my plan regardless of what it is. The claim that some principles do is not intelligible. For I can only think of one rational reason another person or institution can give for thinking that my aims have merit regardless of what they are: namely, the mere fact that their merit is acknowledged by others who I in turn regard as competent in judging the merit of my life plan and/or my ability to pursue it. How this recognition can be intelligibly expressed in a principle of distribution surely depends on the content of my plan.

Secondly, it is hard to see how the liberty principle can affirm the value of any of our determinate and particular ends or our ability to pursue them. For when I cannot fail in being entitled to, say, the same basic liberties that everyone else has, how can these affirm that I have succeeded in forming and pursuing a worthwhile determinate life plan? That is, how can I take it as an affirmation that my particular career, hobbies, etc. have merit?

That being said, in Rawls the liberty principle, in conjunction with its public justification, plausibly affirms the merit of some aims and ideals. Specifically, I think that it credibly and intelligibly affirms the merit of developing and exercising the moral powers and pursuing a rational life plan: in short, the merit of satisfying Rawls’s ideal of personhood. For an important part of Rawls’s (1993, chs. 3, 5) public justification of the liberty principle is that satisfying the ideal of personhood is important and that the liberty principle makes it possible. In that sense, the principle and its public justification signal that a life plan that is rational and which includes developing and exercising the moral powers adequately has merit.

However, there is a more basic problem with the status argument, one that also applies to the specifically moral part of our evaluative self-respect. This is that the liberty principle cannot affirm that we satisfy any standards of merit to a significant extent. For this is a question of how we act, and the liberty principle does not say that liberties are to be distributed based on how people act. When I cannot fail in being entitled to the same basic liberties as everyone else, how can they affirm my sense of succeeding in developing and exercising my moral powers and being a fully cooperating citizen? To put the point another way, by respecting your rights as determined by justice as fairness, I do not express any evaluation of your character and conduct. Arguably, I simply recognize the fact that you are a person and treat you accordingly. Intuitively, this impersonal form of respect does not do much to support your evaluative self-respect.

So, the liberty principle cannot do much to directly affirm the merit or feasibility of our plans and projects. And mere respect for the right to live as one want—which is arguably closer to what Rawls has in mind—does not do much to support people’s evaluative self-respect (at least not the impersonal respect of rights that the liberty principle prescribes). Therefore, I do not think that the liberty principle offers much positive recognitional support to evaluative self-respect.

It might be objected that an unequal distribution of the basic liberties would nonetheless involve a negative evaluation of the citizens who are denied the liberties in question. For example, Jason Brennan’s argument for epistocracy is that some citizens are less politically competent than others and therefore should have less political power. This is a negative evaluation of the citizens in question. But, as Brennan (2016, 123) argues, it is unclear why it could reasonably be experienced as denigrating in contrast to, say, the negative evaluation that prohibiting incompetent people from performing surgery entails. Indeed, while we can expect offices and positions that call for generally desirable talents and are associated with social goods (e.g., being a surgeon) to enjoy a relatively high status (Scanlon 2018, 32–33), this point does not apply to the right to vote in an epistocracy of the kind that Brennan defends. For similar reasons, it is even less clear why an individual (or general) restriction on the right to vote for the sake of other distributive principles would necessarily threaten the evaluative self-respect of the citizenry—though much depends on the precise justification. I do not have the space to discuss other liberties, but though there may be other alternatives to the liberty principle that would not undermine evaluative self-respect, there are also many that clearly do.^[9] Nonetheless, the appeal to evaluative self-respect fails to give a reason for preferring central parts of the liberty principle above alternatives worth taking seriously.

3 The Personhood Account of Self-respect and Basic Liberties—Honneth

In this section, I discuss whether the moral personhood concept of self-respect can ground an argument for the importance of equal basic liberties. While, as previously mentioned, you can reconstruct an account of personhood self-respect in Rawls, I shall mainly use Honneth as a proponent of the personhood view of self-respect. This is simply because Honneth’s personhood view is more developed and because I believe the discussion in this section highlights some general problems with using the personhood account to ground a case for the liberty principle. I focus on Honneth’s early theory of recognition as it is stated in The Struggle for Recognition (1995) and in some of his subsequent works, where the emphasis is on the psychological conditions for “autonomy” and “self-realization.” The main idea is that autonomy requires a specific sense of your moral status as a person, which in turn requires recognition in the form of equal legal rights. I begin by outlining Honneth’s argument in more detail before turning to some problems with using any personhood concept of self-respect to defend the value of legal rights and liberties.

According to Honneth, having self-respect means to respect oneself as a person, an autonomous being, a responsible agent morally equal to all other persons and a bearer of legal rights. This involves having certain moral expectations of being treated as an equal and taking oneself seriously as a competent judge of one’s own choices and actions. In contrast to Rawls, Honneth clearly and consistently views self-respect as grounded in personhood, not in character and conduct. He categorizes the self-relation that concerns the value of one’s particular traits and one’s abilities and achievements as “self-esteem.” Self-respect, in contrast, concerns traits that are common to all persons (“person-making features”). Moreover, while self-esteem comes in degrees and must be earned, self-respect is a matter of being conscious of and respecting your unconditional and nongradual moral status as a person. If you are a person, you will always have sufficient grounds for full self-respect (Honneth 1995, ch. 5). Honneth (1995, 113) here explicitly draws upon Darwall’s (1995) influental notion of “recognition respect” for persons as persons, that as a person you have importance and act accordingly. The crucial point here is that recognition self-respect concerns the moral status you have simply by virtue of being a person. It is not conditional upon whether you are a good person, morally or otherwise.

The relationship between self-respect in Honneth’s sense and autonomy is seemingly straightforward. As previously mentioned, having self-respect involves viewing oneself as a subject who is competent and worthy of making decisions. And as Joel Anderson and Honneth (2005) write: “If one cannot think of oneself as a competent deliberator and legitimate co-author of decisions, it is hard to see how one can take oneself seriously in one’s own practical reasoning about what to do. Those with diminished self-respect—with less of a sense of their personal authority—thus are less in a position to see themselves as fully the author of their own lives” (132).

Self-respect is in turn dependent on recognition in the form of respect, which for Honneth means recognition of our status as moral equals capable of autonomy (Anderson and Honneth 2005, 132; Honneth 1995, ch. 5). The idea here, as with Rawls, is that your positive self-attitudes are dependent on confirmation from others. You thus need to be recognized as autonomous in order to be autonomous. “The agentic competencies that comprise autonomy require that one be able to sustain certain attitudes toward oneself … and … these affectively laden self-conceptions … are dependent, in turn, on the sustaining attitudes of others. [Self-respect, self-trust and self-esteem] can be viewed as being acquired and maintained only through being recognized by those whom one also recognizes” (Anderson and Honneth 2005, 131).

In particular, Anderson and Honneth (2005, 132–133) argue that we need to have the standard liberal individual freedoms and rights recognized by others in order to develop and maintain self-respect. Note that this is a status argument. Basic liberties support self-respect because they convey certain attitudes, not because they are means of achieving any of its requisites (autonomy, moral status). To sum up, the claim is that to be autonomous we must respect our moral status as autonomous persons, which involves identifying with this ability. This, in turn, is dependent on living in a liberal democracy where our basic legal rights are protected.

I have now outlined Honneth’s self-respect argument for the liberty principle based on a moral form of self-respect. Can such an account work? I shall argue that Honneth’s account of self-respect fails in providing a case for central parts of the liberty principle. The reason for this is that we cannot establish both that personhood self-respect is best supported by the liberty principle and, at the same time, that it is important for any plausible and relevant kind of autonomy.

To make this case, I shall rely on a distinction between personhood self-respect simpliciter and specific forms of such self-respect. By the former I mean a sense of being a morally equal person entitled to equal moral concern based on a reasonable conception of personhood and moral equality. The “reasonable” clause means deserving of serious attention as a conception of moral equality and justice. One example is epistocracy of the kind that Brennan (2016) defends, which is close to democracy but where “people do not, by default, have an equal right to vote or run for office” (208). Another example is a conception of justice that mandates a high but not lexical priority of the liberty principle. The distinction between different forms of personhood self-respect corresponds to the different conceptions of moral equality and personhood they are based upon. I shall argue that we have no strong reasons to believe that personhood self-respect simpliciter is supported better by the liberty principle than, at the very least, these reasonable alternatives. My case thus primarily concerns two central aspects of this principle: the equal distribution of the basic liberties and their priority.

Though my case primarily targets its distribution, it is useful to start by considering an argument for why the right to vote per se affirms personhood self-respect. Chris Melenovsky (2018) argues that all the items on Rawls’s list are necessary to the development of the moral powers, powers Honneth would say, I believe, that at least partly constitute our autonomy. If Melenovsky is correct, you might argue that all the items are necessary to convey respect for persons qua persons understood in terms of these powers. Indeed, Meena Khrisnamurty (2013, 188) makes this argument about voting rights in particular.

But suppose I am a utilitarian and do not think of myself as a moral equal in this sense. I do have a form of personhood self-respect, but it is based on my welfare counting as much as that of everyone else and not on my rationality and reasonability (Labukt 2009, 215–17). Suppose further that I do not believe universal suffrage is necessary to maximize welfare.^[10] Why, in this case, would granting me the right to vote be necessary to provide optimal support for my personhood self-respect? It seems plausible that which specific principles that affirm my sense of being a moral equal are, at least largely, determined by my view of justice—specifically the kind of treatment I think I am, as a moral equal, entitled to (Schemmel 2018, 8–10; Labukt 2009, 212–17; Brennan 2016, ch. 5). We should also note that Honneth, at least after The Struggle for Recognition (1995), holds a similar position when he contends that the need for recognition is historically contingent. By this he means that personhood self-respect simpliciter is determined by the basic underlying norms of recognition (e.g., norms of respect) in the society we have been socialized into, and not by some invariable aspect of our human nature (Honneth 2002, 501–2, 508–10).

One might argue that the equal distribution of the basic liberties is nonetheless important. In the case of voting rights, the idea would be that even if having this right is not essential per se, being denied it would still undermine personhood self-respect if others have it—at least in general. Not being shown the same degree of personhood respect as others signals that you have a lower moral status in the eyes of others. This, in turn, is arguably generally damaging to personhood self-respect, even if one thinks that this conception of respect is misunderstood.^[11] However, this argument is only convincing if the unequal distribution is indeed justified by the fact that I have a lower moral status. Brennan’s (2016, 117–118) position implies no such judgement but is merely based on a conception of moral equality where we should choose the kind of government that reliably produces the best results. So, why would I take the fact that I lack voting rights while others enjoy them to signal that my interests count for less in the eyes of society?

Most of the basic liberties on the traditional list endorsed by Rawls and Honneth, and their equal distribution, still seem important. They appear important in the sense that, in most situations, it is hard to justify their restriction in favor of other distributive principles in a way that credibly conveys personhood respect and not a lack thereof. But this does not mean that it is never possible.

This takes us to the question of priority. By lexical priority Rawls (1993, 294–97) means that under “reasonably favorable conditions” (as opposed to times of war and other nonideal conditions) the basic liberties have absolute weight vis-à-vis other social goods. “For example, the equal political liberties cannot be denied to certain social groups on the grounds that their having these liberties may enable them to block policies needed for economic … growth” (Rawls 1993, 294–95). Now I do not think we have good reasons to endorse the claim that no such restrictions on liberty can ever be given a justification that all citizens can be expected to accept as just, or as just enough. At least not without presupposing at the outset that, all things considered, we should prefer lexical priority even ignore considerations of self-respect as such, in which case the argument from self-respect is superfluous. However, a Rawlsian could respond that strict lexical priority is not essential to self-respect even under reasonably favorable conditions. Also, it is not clear whether Honneth (2007a, 141) advocates such a priority. As already indicated, I can agree with this weaker claim. Yet, I shall argue in the final section that the case from a sense of competence may give a reason for an even higher priority of the liberty principle.

So much for personhood self-respect simpliciter. I now turn to a line of reasoning centered on the value of a specific form of personhood self-respect that is arguably better supported by the liberty principle than by the reasonable alternatives I have sketched. This form is based on a conception of personhood and moral equality that implies that the liberty principle is what justice requires. This kind is closer to what Honneth is talking about than epistocratic and utilitarian self-respect. It is the form of self-respect that Honneth (1995, 107–21; 2005, 133) would say has become prevalent in contemporary liberal democracies. As previously mentioned, it is based—according to Honneth (1995, 119)—on an understanding of individual autonomy as the “respect-worthy core” of persons. For the sake of simplicity, I will henceforth refer to this form as “basic-liberties self-respect.” Arguably, citizens growing up in a society where the liberty principle is implemented would tend to respect themselves in this sense.

Why should our focus be on this form of self-respect? In what follows, I discuss what I take to be the two most central answers in the literature: (a) People that have basic-liberties self-respect are more autonomous than people with a different, but reasonable, form of personhood self-respect. This, I take it, would be Honneth’s answer.^[12] And (b), the moral beliefs that basic-liberties self-respect involves are correct. Therefore, such self-respect has moral value (e.g. Meyers 1995).

Consider (a), the idea that people with basic-liberties self-respect are more autonomous than people with other reasonable forms of personhood self-respect. Why should we think that such a demanding form of self-respect is a condition for autonomy? To fully understand and discuss what exactly Honneth is claiming about autonomy and its connection to self-respect (and what can be claimed in this regard), it is necessary to take a brief look at some distinctions between different theories of personal autonomy.

Personal autonomy concerns the agent’s own aims and personal values, her own view of the good life. To be autonomous, your preferences, values, etc. must be your own in some deep sense. Liberals typically advocate a procedural account of autonomy, which are content-neutral in the sense that they say nothing about which values and beliefs a free agent must have or not have. The question is only whether they are affirmed in the appropriate way. “Procedural conceptions characterize autonomous agents—agents whose preferences and desires are genuinely their own—as those who critically reflect in the appropriate way to evaluate their preferences, motives, and desires” (Stoljar 2013, sec. 4, emphasis in original). Most procedural accounts require only some sort of reflective endorsement in order for a preference, value or action to be autonomous (Christman 2003, sec. 1.2). Hence, “in a thoroughly liberal manner, this shift to formal, procedural conditions allows this model to accommodate a diversity of desires and ways of life as autonomous” (Christman and Anderson 2005, 3). Procedural accounts of autonomy contrast with substantive accounts. Substantive accounts are not content-neutral but require that autonomous agents have specific values and/or beliefs (Christman 2003, sec. 1.2).

It may seem that we presuppose a substantive conception of autonomy if we claim that it depends on basic-liberties self-respect (which involves having specific moral beliefs). However, following Antti Kauppinen (2011), we can read Honneth’s account as a theory of (some) of the social conditions for exercising procedural autonomy rather than as a definition of freedom as such. This also strikes me as the most plausible interpretation of Honneth’s (2002; 2005) argument. Following Kauppinen then, basic-liberties self-respect is perhaps best understood as a “second-order autonomy capacity” (2011, 266). What he means is that self-respect is not itself an autonomy capacity such as the ability to reflectively endorse your desires and the like, but a condition for exercising them. The idea here is that you must identify with your abilities to make effective use of them. In other words, self-respect is a condition for living an autonomous life: “roughly put, an autonomous person is one who has the capacities that are exercised in autonomous choice and action, and an autonomous life is one led by an agent who successfully exercises these capacities to a significant extent over time” (Westlund 2009, 28, emphasis in original). If this distinction is viable, you might say that there is no conflict between liberal theories of autonomy, which are typically procedural, and Honneth’s account (Kauppinen 2011, 274–75). In Kauppinen’s (2011, 264–275) view, the concept of second-order autonomy capacities is perfectly compatible with procedural accounts of autonomy. Moreover, as far as people are interested in the possibility of living autonomous lives and not merely being autonomous, they have an interest in this form of self-respect (Kauppinen 2011, 274–75).

We have at least two ways then of understanding the position that autonomy depends on basic-liberties self-respect: we can understand it as a substantive theory of autonomy, or we can view it as a claim about the conditions for exercising procedural autonomy. Consider first the idea that the exercise of procedural autonomy depends on basic-liberties self-respect. Why would a procedural form of freedom—however we specify it—presuppose such specific moral beliefs? As far as I can see, it is perfectly possible to not have basic-liberties self-respect and still be able to exercise procedural autonomy. To take an example, think again of those utilitarians who do not believe that the liberty principle does the best job of promoting welfare. They do not have basic-liberties self-respect, but still respect their own moral status as beings with interests that count as much as everyone else’s. And having this moral view is clearly not an obstacle to exercising procedural autonomy. I therefore do not see why these utilitarians would not be able to reflectively endorse and pursue their conception of the good.

Basic-liberties self-respect is thus not important to procedural autonomy. Therefore, let us consider, secondly, abandoning the focus on procedural autonomy and defending basic-liberties self-respect as part of a substantive account of autonomy. However, this defense also seems highly implausible. Again, consider the implications. It implies that no one who does not favor the traditional liberty principle can be autonomous, or quite as autonomous as those who do. Even if we put the issue of state-neutrality versus perfectionism aside, this still does not strike me as a defensible view of personal autonomy. It is hardly reasonable to say that people who favor somewhat different views of justice cannot be autonomous, or quite as autonomous, as those who respect themselves in this specific way ipso facto. To say that deeply submissive individuals may be free is perhaps something that violates our intuitions about freedom (e.g., Oshana 1998). However, to reiterate, to say that those utilitarians who do not believe in the special status of basic liberties can be as personally autonomous as anyone else, does not violate our intuitions. In short, I do not see why lacking basic-liberties self-respect should make you less autonomous (or worse off in terms of welfare for that matter) in any plausible and relevant sense of that word.

Let us now consider (b). So much for the importance of basic-liberties self-respect with regard to autonomy. It is worth considering another value that such self-respect may have. We may argue that the liberty principle is the correct answer to what treating people as moral equals means and that it is therefore important for people to have basic-liberties self-respect. In the words of Dillon (1995): “maintaining an explicit congruence between self and value is appropriate and good for beings such as ourselves” (35). However, while this may very well be the case, it cannot establish the kind of argument that we are looking for. The problem is that most if not all modern theories of justice can be viewed as interpretations of the norm of moral equality, of what it means to treat people as moral equals (Dworkin 1987, 7–8; Kymlicka 2002, 1–5). That treating people equally requires that we give the liberty principle a special status is precisely what the argument from self-respect is supposed to support. As Stephen J. Massey (1995, 207–10) points out, if we defend a form of self-respect by arguing that it is morally correct, then we cannot point to this form of self-respect to defend the moral notion in terms of which it is correct. Say that we defend a notion of self-respect as respecting yourself in accordance with a conception of moral equality that implies that the liberty principle is what justice requires by arguing that this conception is correct. In that case, we cannot defend the importance of the liberty principle by arguing that it is conducive to self-respect thus understood. That would be a circular argument.

4 A Sense of Competence

So far, I have tried to show that the two standard approaches to self-respect fail to give a reason for preferring central parts of the liberty principle above alternatives worth taking seriously. Note that the problems and strengths of these approaches mirror each other. While the evaluative form of self-respect is clearly valuable, the connection between it and the liberty principle is too weak. Basic-liberties self-respect, on the other hand, is clearly supported by the liberty principle, but it is too comprehensive. The idea that we are somehow less autonomous or worse off if we do not think that the liberty principle is exactly what justice requires is not plausible. What we need then is, at the very least, a self-relation that is not too tied up either with our character and conduct, or with our moral status and sense of justice. In this section, I will try to clarify what I take to be such a self-relation. What I suggest is that both of these concepts involve a self-relation that is necessary for exercising procedural autonomy and which is possible to distinguish from their problematic elements. In the next section, I use this self-relation to make a case for the liberty principle by arguing that this principle provides an optimal basis for it.

Recall Honneth’s basic idea (as interpreted by Kauppinen) that the effective and regular exercise of autonomy depends on some self-relation and that this self-relation requires some form of recognition. Though I have criticized the specific way that Honneth develops this point, I still find it to be an intuitive idea that is worth building upon. It is a main point not only in recognition theory, but also in several relational theories of autonomy.^[13] Rather than suggesting a new relational theory of autonomy or of self-respect, I want to focus on an essential, if not the most basic component in self-relation required to exercise autonomy as understood in standard procedural conceptions.

To elaborate, typically both authenticity and competency requirements play a role in conceptions of personal autonomy. “To govern oneself, one must be in a position to act competently and from desires (values, conditions and so on) that are in some sense one’s own” (Christman and Anderson 2005, 3). These two types of requirements correspond to two types of autonomy capacities: capacities for valuation and capacities for reasoning or rationality. “Authenticity conditions are typically built on the capacity to reflect on and endorse (or identify with) one’s desires, values, and so on” (Christman and Anderson 2005, 3). A few examples of competency requirements are self-control, self-understanding, the capacity for care, empathy, lack of mental illness, the ability to follow norms of reasoning, imagination, determination and the like (Christman and Anderson 2005, 3). As previously mentioned, most procedural accounts of autonomy specify it as the ability to reflectively endorse preferences, values and choices in some sense.

My aim is not to defend a specific view of personal autonomy. My point is rather that in order to exercise our capacity for valuation and reflection regularly and effectively, we need to trust our own competence to make sound judgements and act on them. Again, this is a common idea in the literature on self-respect and relational autonomy.^[14] I will call this self-relation “a sense of competence.” As I have said, a variant of a sense of competence is part of or implied in most standard concepts of self-respect. However, it is important to point out that a sense of competence also differs from those self-relations in ways that make it more suited to ground the self-respect argument. In the rest of this section, I will present my arguments for why this is the case.

A sense of competence is included in Honneth’s notion of basic-liberties self-respect. We are morally equal by virtue of our common capacity for autonomy. It follows that respecting yourself in Honneth’s sense implies a certain level of confidence in your autonomy. Indeed, as previously mentioned, he says a self-respecting person views herself as a “competent deliberator and legitimate co-author of decisions” (Anderson and Honneth 2005, 132). I think this accounts for much of the intuitive plausibility of Honneth’s theory of how self-respect is important to autonomy. However, this is a one-way implication. Humans are commonly thought to have equal moral worth due to a shared capacity for rational thought and action, but from a psychological and conceptual point of view these are distinct notions. I do not see why you would need to hold specific ethical and/or political views in order to be confident in your ability to make sound judgements. A sense of competence is therefore conceptually distinct from personhood self-respect: it does not essentially involve respecting yourself in accordance with some view on personhood. In contrast to basic-liberties self-respect, it is also compatible with many different views on justice. This makes it more plausible, by default, as a condition for autonomy.

A sense of competence is also included in Rawls’s evaluative view of self-respect—the conviction that your life plan is worthwhile and that your character is fit to pursue it. Most importantly, you need a sense of competence to the extent that the standards by which you evaluate the merit of your character and actions require that you exercise your capacities for autonomy. For example, depending on how we understand Rawls’s account of the moral powers, perhaps his formulation of self-respect as rooted in the development and exercise of these powers requires the degree of a sense of competence that you need to be autonomous in the procedural sense. If so, his argument, or parts of it, might not be that different from my own.

Having evaluative self-respect requires more, however, than having a sense of competence. Firstly, that you trust your own capacity for autonomy does not mean that you are secure in your disposition to act autonomously. More precisely, it does not entail that you satisfy any standard of autonomous behavior such as we find in, say, Rawls’s ideal of personhood. Secondly, having a sense of competence does not entail that you hold an ideal of autonomy in the first place. It does not mean that you count exercising your capacities for autonomy regularly and effectively as a moral or personal standard. Indeed, it does not imply anything about your evaluative standards. Thirdly, a disposition to exercise procedural autonomy does not—in any case—exhaust any plausible conception of a good or decent character. It says nothing about whether you are, say, compassionate, honest, reasonable, etc. In short, that you have a good character, at least good enough to be respected, means more than that you are able to act autonomously or satisfy other important standards. Therefore, far more is required to have evaluative self-respect than a sense of competence. Consequently, I think a sense of competence avoids the problem I noted in Section 1 about how it is hard to see how the liberty principle can do much to affirm that people’s character and conduct have merit.

5 A Sense of Competence and Basic Liberties

So far, I have discussed the importance of a sense of competence for autonomy and tried to clarify the meaning of this concept. In this section, I argue that this account may yield a reason for thinking that the liberty principle should enjoy an even higher priority than we might otherwise have reason to believe.

Why should we think that the liberty principle provides a particularly good support for a sense of competence when this self-relation is compatible with a multitude of views on justice? The core of my argument is that a political system based on the liberty principle provides an important social basis for a sense of competence by virtue of the attitudes it expresses. Recall the idea that your own self-attitudes normally depend on affirmation from others. Others must treat you in accordance with your self-attitudes. This is a basic premise in recognition theory and much of the literature on self-respect. Now, if this premise is true, then to develop and maintain a sense of competence, others must recognize you as autonomous.

Does the liberty principle express recognition of autonomy then? Now it is common sense that a political system wherein the liberty principle is implemented expresses recognition of autonomy in a more clear and credible way than contrary arrangements. Consequently, it may, as a matter of empirical fact, support a sense of competence better than alternative ways of organizing society. Of course, we may say the same about personhood self-respect and the liberty principle. It is common sense that the liberty principle affirms our moral status. Arguably, most people think of themselves as morally equal in broadly Kantian terms, as autonomous beings entitled to rights that protect their autonomy. In other words, most people have basic-liberties self-respect. If so, the liberty principle would affirm most people’s personhood self-respect (excepting some utilitarians and the like).

We cannot contend with merely appealing to popular beliefs about moral equality, however. Though starting from common sense, liberals like Rawls aim to offer principles that could be the object of informed and reasoned consent and not simply those that are most in line with what the citizenry believe is just right now. Very roughly, the aim is a reflective equilibrium between common sense and theoretical considerations (Rawls 2001, 5, 29–31; Honneth 2011, 408). Similarly, we cannot be content with saying it is common sense that the liberty principle expresses recognition of autonomy in a more clear and credible way than alternatives. Consider Brennan’s (2016, ch. 4) claim that voting rights are not important to autonomy. If true, it seems to follow that the belief that voting rights are necessary to offer the most credible recognition of autonomy is mistaken and that citizens should adjust their beliefs accordingly.

To avoid such objections I shall assume that the liberty principle with all its basic liberties does the best job of securing the other conditions for autonomy, the “external” ones we might say (Kauppinen 2011). It enables us to pursue our goals and prevents others from unduly interfering in our lives. In conjunction with a public justification that centers on autonomy, I would say the liberty principle thereby provides the best expression of recognition of autonomy.

The public justification is important. As Gerald Doppelt (2009, 135–38) points out, what rights signify depends on their symbolic meaning. If rights are to give people a reason to be convinced of their autonomy, culture needs to be such that rights are taken to express genuine respect for (or trust in) autonomy. Conversely, if public justification of the liberty principle is centered on autonomy, then deviations from this principle convey distrust of autonomy (or at least entail the absence of a sign of trust). And at least some social groups might for historical or social reasons be vulnerable to public distrust of their autonomy—such as the kind conveyed by restricted voting rights.

However, this raises the question of what exactly the account of a sense of competence and its political basis adds to the more general autonomy-based defense of the liberty principle. How does it give us an additional reason for its high status? If we presuppose that the liberty principle must enjoy a high status to protect our autonomy even when we ignore a sense of competence (because it protects the external conditions for autonomy), then what relevance does it have that a sense of competence is important to autonomy and best supported by the liberty principle? Note that this problem is similar to how the idea that basic-liberties self-respect has moral value ends up with a circular defense of the liberty principle (Section 3). If, say, we argue that the liberty principle rather than some other principle is required to protect autonomy on the basis of a sense of competence, we end up with a circular argument.

My answer is essentially that the argument from a sense of competence shows that basic liberties are more important to autonomy than we might otherwise have reason to believe. Though I assume that the liberty principle is already required to protect our autonomy, the point of the argument from a sense of competence is that deviations from this principle threaten autonomy in a specific way. They not only hinder autonomy by making it practically harder for people to pursue their own projects. They also undermine autonomy by threatening people’s self-relation. This preserves a basic intuition in recognition theory. In the words of Anderson and Honneth (2005 emphasis in original): “on the recognitional approach, guaranteeing rights does not ensure autonomy only directly (in the negative sense of blocking interference) but also supports autonomy via the support for self-respect” (133).

This may give us a reason to think that the liberty principle should enjoy an even higher priority than it ought to if we only considered the external conditions in situations where autonomy was balanced with values such as utility, equality or stability. It may also count in favor of a higher priority than arguments from personhood self-respect because a sense of competence relies on recognition of autonomy per se, whereas personhood self-respect relies on being recognized in accordance with a reasonable conception of justice, that is, recognized as just, all things considered so to speak. This latter form of recognition may only involve a restricted recognition of, or trust in, autonomy.

In short, I have tried to give a reason for thinking that the liberty principle is more important to autonomy than we might otherwise have reason to believe. How much more important? This depends on how vulnerable humans are to misrecognition (see e.g., Colin Bird 2010). I have assumed throughout this paper that our self-attitudes normally depend on the attitudes of others. I do not have the space to discuss the extent to which this claim is true and in what sense. If we are not particularly vulnerable, the account of a sense of competence does not give a reason for thinking that the liberty principle should enjoy a significantly higher priority than we might otherwise have reason to believe. In that case, my argument may have more theoretical than practical interest. Nevertheless, given the premise that our self-attitudes rely—at least to some extent—on affirmation from others, the account of a sense of competence yields an additional reason for thinking that the liberty principle should enjoy a special status.