Democratic Respect and Compromise

Forthcoming in Critical Review of International Social and Political Philosophy Rostbøll

2





May (2005, pp. 318-19) similarly distinguishes between reasons for correction and reasons for compromise.
Or they ought to do so (Gutmann and Thompson 2012).
After the election in September 2014, for example, the major Swedish political parties refused to cooperate with the far-right Sverigedemokraterna, which had attained 12.9% of the vote. This led to resentment among the voters of the latter and might have strengthened its electoral support.
Cf. Jones and O'Flynn's (2013, p. 120) definition of principled compromise.
In his 2011 paper, May speaks of "principled or non-instrumental" reasons for compromise (May 2011, p. 581).
Jones and O'Flynn follow May in calling this a principled compromise, while I call it an intrinsic reason for compromise.
It seems to me that democratic instrumentalism is parasitic upon a non-instrumental conception of democracy, but this is a large issue that I cannot go into here. I discuss and reject democratic instrumentalism in Rostbøll 2015a, 2015b, and 2016.
Margalit's (2010) concern is a very different one than mine, and I only refer to his notion of rotten compromise in a rough manner to denote unacceptable compromises.
Democratic Respect and Compromise
Christian F. Rostbøll

Much contemporary political theory examines how people who disagree on fundamental issues of good and right can nevertheless live together. This question is central when we recognize "the fact of pluralism." In democratic theory, recognizing pluralism raises the question whether and why we ought to accommodate cultural and ideological minorities in political outcomes when this is not needed to reach a majority. The latter, I see as the question of the moral significance of political compromise. Do majorities have any reasons of fundamental democratic principle to make concessions to minorities, if minorities have been properly included in and listened to in the political process? Why would there be a democratic obligation to make concessions to one's political opponents when this is not required to create a majority and these concessions would involve a departure from justice? Is it sufficient to include all citizens in the process of decision making via equal rights of participation, or do our democratic ideals also require that all positions be accommodated in policy? These questions do not concern whether compromises are necessary for the well-functioning of democratic politics. No one would deny the latter; that is, no one denies that there are pragmatic reasons for compromise. The question regarding the moral significance of compromise is whether there are reasons inherent in the democratic ideal that requires us to regard compromise as good in itself and not merely a regrettable necessity.
Despite its omnipresence in actual politics, discussions of compromise have until recently been largely absent in political theory and philosophy. The last few years, though, have witnessed a rise in political theoretical reflections on compromise (Bellamy 1999, Fumurescu 2013, Gutmann and Thomson 2012, Lepora and Goodin 2013, Margalit 2010, May 2005, 2011, Weinstock 2013). In some of the normative discussions of compromise, however, it is unclear if compromise is seen as a necessity because of the political institutions we happen to have in particular democracies, or if compromise should be awarded a more fundamental democratic significance. Often, compromise is cherished because of its connection to a fundamental norm of mutual respect under conditions of pluralism and disagreement (Bellamy, Kornprobst and Reh 2012, p. 280, Bird 1996, p. 92, Gutmann and Thompson 2012, pp. 34, 109-17), but the nature of this connection is underdeveloped. In particular, it is unclear if the imperative to compromise can be derived from a norm of mutual respect and thus has the same moral significance as the equal right to vote, to run for office, and to participate as an equal in public deliberation. The question of whether there are not merely pragmatic but also principled reasons for compromise has been discussed by Simon Căbulea May (2005, 2011) and Daniel Weinstock (2013). While May and Weinstock examine the notion of respect and the democratic credentials of compromise, they fail to consider the idea that we need a specifically democratic conception of respect and that the latter can supply an intrinsic reason for compromise.
This article argues that there are, indeed, reasons for compromise that are inherent in the democratic ideal. Under some conditions, compromise can give greater legitimacy to public policy beyond what can be achieved by a mere majority decision, and not merely because of the consequences but because of the very fact that the decision was a compromise. The argument is made via a discussion and elaboration of a norm of democratic respect. I argue that we cannot explain the value of political compromise with some generic idea of respect but need to specify a democratic conception of respect. I defend a conception of respect for fellow citizens as co-rulers and contrast this notion to, first, respect for equality simpliciter and, second, respect for fellow citizens as members of the same community. I argue that only the former is both sufficient to explain the moral importance of democratic procedures, including compromise, and an inherently democratic ideal. My argument, however, does not entail that compromise should be awarded the same moral significance and standing in a democracy as the right to vote, to run for office, and to participate in deliberation. The imperative to compromise is not unconditional, as are the rights of participation among citizens in established democracies.
The argument proceeds in several steps. I begin by offering a definition of compromise and distinguish it from consensus. Then I propose a typology of different kinds of reasons for compromise and explain what I mean by a reason inherent in the democratic ideal. I proceed by analyzing how a norm of mutual and equal respect has been used to justify democracy in general and specific democratic procedures such as majority rule in particular. The question is posed whether the same norm of respect that is used to justify democratic procedures of voting and majority rule can be invoked also to justify an imperative to compromise in politics. I argue that we must go beyond a general idea of respect and specify a notion of democratic respect. Two notions of respect, respect for fellow citizens as equals and respect for fellow citizens as members of the same community are discussed but both found wanting in terms of explaining the moral importance of specifically democratic procedures. I suggest that understanding democratic respect in terms of respect for fellow citizens as co-rulers has advantages compared to the equality and community conceptions. Moreover, including compromise as a key feature of collective self-legislation can help respond to some persistent objections to the notion that democracy is grounded in an ideal of collective self-legislation. Finally, I contend that even if compromise has non-instrumental value, it should not be awarded the same unconditional value as rights of participation in the democratic process. Compromise is a form of representation not only in the process but also in the outcome and as such the value of compromise has limits.

The meaning of compromise
My focus in this article is domestic, legislative compromise in representative democracies, that is, compromise among political opponents or different political parties with the aim of making law and policy. In a representative democracy, how the representatives relate to each other and reach agreements matters not only for themselves but also for the citizens in whose name they speak and act. For example, the type of respect the legislators display or fail to display towards each other mirrors the respect that is shown for the opinions of the citizenry. When representatives compromise they do so on the citizens' behalf; they include or accommodate the plurality of views among citizens in their agreement and policy. In societies characterized by social differences and cultural diversity, compromise is a way of recognizing these in political outcomes.
I understand a political compromise as an agreement in which all sides make concessions in order to be able to act together, and in which the concessions are motivated by the presence of disagreement (Gutmann and Thompson 2012, p. 10, May 2011, p. 583, Margalit 2010, p. 20). There are three aspects to note about this definition of compromise. First, compromise involves mutual concessions; all the parties to a compromise must modify their positions and sacrifice something for the sake of reaching an agreement. Compromise is not surrender. Second, even if the compromise position is seen as inferior to one's preferred option, it is still regarded as better than doing nothing by all parties. Political compromise is a response to the shared perception of the need for common political action within a certain area. Third, compromise should be distinguished from a consensus in which all parties have been convinced of the rightness of a different option than the one they preferred ex ante. In a compromise solution, the parties keep preferring another option than the one they could agree on but accept the latter option because of the presence of disagreement.
The distinction between compromise and consensus is important. Consensus implies that that there is no (longer) disagreement and that the parties are convinced of that the agreed to outcome or policy is the best one. Consensus on a certain issue might simply exist but in democratic theory it is often seen as the product of (or at least the aim of) a deliberative process in which the strongest argument wins. Compromise, by contrast, does not do away with disagreement; it embodies disagreement. A compromise need not by itself entail learning or correction regarding the issue at hand, as does consensus-oriented deliberation. This means also that the reasons to accept a consensus solution and a compromise solution differ. In the first case, one is motivated by the force of the better argument (Habermas), while in the second case one is motivated by the presence of disagreement. Thus, the compromise solution is not accepted by the parties as true or better than what they brought to the table but is accepted as legitimate for a different reason related to the fact of disagreement (Bird 1996, pp. 92f). It is the content and force of this latter reason that is the focus of this article.
Legislative compromises are made all the time. In multiparty systems, they are made among the parties of the governing coalition in order to keep the coalition intact, and minority governments make compromises to keep the backing of government-supporting parties. In political systems with a strong division of power like the USA, compromises are made among the different branches of government to avoid gridlock. However, sometimes compromises are also made beyond what is formally required for passing a law; these compromises are made because the parties think it is good to have broad support for legislation and public policy, which creates stability and gives legislation time to work (Christiansen 2008). Both cases in which compromise is formally necessary to pass legislation and cases of formally unnecessary compromise raise the question whether compromising is the right thing to do. Even if a compromise is necessary in order to make a decision or pass a law, it is not necessary to pass a law: legislators can choose to do nothing. Representatives sometimes have to choose between making the system work through compromise and blocking all legislation for the sake of principle. Democratic legislatures are majoritarian institutions, and normally only a majority vote is required in order to pass legislation. Sometimes – particularly in multi-party systems – legislators must compromise with political opponents even to reach a majority. However, the interesting test case for assessing the moral significance and democratic pedigree of compromise is whether compromises should be made even when they are not needed to create a majority.
Compromising means getting less or something different from what you would prefer or think is just. From the partisan perspective, it means sacrificing the ideally best option for the sake of the second-best option. When and why should one accept such inferior outcomes? My question is whether compromising, despite this form of sacrifice, still under some conditions might be the right thing to do; more specifically, whether it, sometimes, is the democratically right thing to do. Do our democratic principles require that we reach compromises with our political opponents? Do we owe it to our fellow citizens, as a matter of fundamental principle, to concede something to their position? Even if the answer to these questions is affirmative, this does not mean that we should discard the ideals of learning and correction involved in the aim of reaching a consensus. The moral and democratic importance of compromise might come to the fore when disagreement is intractable and cannot be overcome with the force of the better argument, which is often the case in societies characterized by cultural diversity.

Reasons for compromise
There are different kinds of reasons for compromise. A political party might compromise, because it is necessary to reach a majority needed to get some of its program through or to gain or to stay in power. A different case is when a party includes an opponent in a compromise in order to avoid antagonizing it or its voters. Refusing to compromise with certain parties may lead to resentment and strengthen these parties, or even lead to unrest or revolution. These are pragmatic reasons for compromise, because the compromise is made for the sake of some end that is external to the act of compromise itself. The opposite of pragmatic reasons for compromise are intrinsic reasons for compromise. The latter refers to cases in which the compromise is valued in itself rather than because of its consequences, that is, because the very fact of compromise expresses or embodies something that is of value in itself. When some say that compromise expresses mutual respect and concern, for example, this is an intrinsic reason for compromise.
Simon May has highlighted the difference between pragmatic and what he calls principled reasons for compromise, and he has argued, "that moral compromise in political life is only ever warranted for pragmatic reasons" (May 2005, p. 317). Note that pragmatic reasons for compromise can be moral reasons. That reasons are pragmatic means that they do not posit compromise as good in itself, but recommends it because of its consequences. If these consequences are morally valuable, the pragmatic reasons for compromise are also moral reasons for compromise. May agrees with this (May 2005, p. 323), but he still contrasts pragmatic reasons for compromise with "principled reasons," which might create confusion. For one might be "driven by principle" when one compromises for the sake of its good consequences (Weinstock 2013, pp. 543, 552-54). It is for this reason that I suggest that we contrast pragmatic reasons for compromise to intrinsic or non-instrumental reasons for compromise rather than to principled reasons. Pragmatic reasons for compromise are goal-based, while intrinsic reasons for compromise are based on a "belief that we owe it to others to concede something to their position," independently of the consequences (Jones and O'Flynn 2013, p. 120). If compromise is justified pragmatically or instrumentally, the consequences rather than the compromise are doing the normative work (Rossi 2013, p. 562). Only intrinsic reasons for compromise give value to the fact of compromise itself.
The reason why it is important to distinguish between pragmatic and intrinsic reasons for compromise is that the questions of when and why to compromise are connected. If there are only pragmatic reasons for compromise, one should only compromise when one has good reasons to believe that the end one seeks depends on compromise. Pragmatic reasons for compromise are conditional upon certain causal relationship to hold true or on the existence of specific institutions. Intrinsic or non-instrumental reasons for compromise, by contrast, are independent of causal and institutional factors as far as the demand to compromise is based on a principle that is honored by the very act of compromising, independently of the consequences.
As part of his argument that compromise in politics is never warranted for principled reasons, May writes also that a "policy is not more democratic because it emerges as a moral compromise between opposing parties if this also means that it diverges from the fair and just policy that a majority has voted for after public deliberations in which all parties were accommodated" (May 2005, p. 343, emphasis added). However, to assess this conclusion we need more clearly to explain what counts as a democratic reason for compromise than May does. First, not all intrinsic reasons for compromise are democratic reasons. The values and norms expressed in compromise may or may not be of democratic nature. For example, I shall suggest below that the value of community might be an intrinsic reason for compromise but it does not amount to a democratic reason for compromise. Moreover, not all forms of respect are democratic forms of respect. Second, "democratic reasons for compromise" may be either pragmatic or intrinsic. A democratic reason for compromise is a pragmatic reason for compromise if it justifies compromise as a means to further democratic values. A democratic reason for compromise is intrinsic or non-instrumental, by contrast, if compromise in and by itself is justified as expressing or constituting a democratic value or norm. I reserve the notion of "reasons inherent in the democratic ideal" to intrinsic democratic reasons.
It might be noted here that some political theorists believe that there are only pragmatic or instrumental reasons for democracy, that is, that the value of the equal right to vote, majority rule, and the other central features of democracy can be explained only because of their good consequences (Arneson 2009, pp. 197-212, Van Parijs 2011, p. 16). However, this form of democratic instrumentalism cannot tell us which procedures are inherently democratic and which are not – and this is required in order to answer whether or not compromise is inherently democratic. Democratic instrumentalism simply assumes that some procedures count as democratic and argues that the value of these procedures depends on their results. From this position, we cannot determine if only the equal right to vote and majority rule is inherent in the concept of democracy or if compromise is, too. How democratic different decision-making methods are cannot be assessed purely in terms of their consequences. For example, we cannot say that majority rule is more democratic than compromise because it furthers economic growth, peace, or the protection of individual rights. In short, the consequences praised by democratic instrumentalists cannot themselves be definitive of democracy. Therefore, in order to determine whether there are democratic reasons for compromise, the instrumentalist approach is not of much help, and we cannot avoid speaking of which procedures are intrinsically democratic.

Mutual respect, democracy, and compromise
It is commonly agreed among political theorists that democracy is the only form of government that respects the equality of citizens and that this fact is central to the normative justification of democracy (Christiano 2008, Dahl 1989, pp. 84ff, 97ff, Waldron 1999, pp. 108-18). The equal vote, equality of opportunity to run for office, and equal opportunities to participate in public deliberation are all ways of treating citizens as equals. Jeremy Waldron celebrates the democratic majority principle as a respectful decision procedure for citizens who disagree but needs to come to terms with each other. Thus, "majority-decision respects the individuals whose votes it aggregates. It does so in two ways. First, it respects the differences of opinion about justice and the common good: it does not require anyone's sincerely held view to be played down or hushed up because of the fancied importance of consensus. Second, it embodies a principle of respect for each person in the processes by which we settle on a view to be adopted as ours even in the face of disagreement" (Waldron 1999, p. 109).
In this way, many theorists of democracy regard equal respect as the central norm expressed in equal opportunities to vote and run for office, to participate in public deliberation, as well as in the majority principle. Hence, they see respect for equality as demanding equal inclusion or representation in political procedures. My question is if we can give the same justification for compromise in democratic politics. Does the norm of equal respect require inclusion or representation in political outcomes through compromise for identical reasons, as it demands equal inclusion and representation in the political process or in political institutions? Note here that I distinguish between representation in the process (e.g. through the equal vote and majority-decision) and representation in the outcome (through compromise) (Weinstock 2013, p. 543).
In the literature on compromise, the norm of equal and mutual respect is often seen as explaining the moral importance of compromise. Under certain conditions, it is argued, making a compromise is morally required as a way of showing respect for your political opponents' views and the plurality of society. Bellamy, Kornprobst and Reh, for example, praise compromises, because they "show equal concern and respect" to the views of political opponents (Bellamy, Kornprobst and Reh 2012, p. 280). Colin Bird suggests that the imperative to compromise can be derived from a principle of mutual respect. He writes, "we ought to compromise because doing so is the appropriate way to display respect to citizens with a variety of views about the relevant public policy issue. The imperative to compromise thus derives from the moral norm of mutual respect, not from a prudential calculation" (Bird 1996, p. 92).
In Amy Gutmann and Dennis Thompson's recent book The Spirit of Compromise, the notion of mutual respect is given a somewhat different place. They write, "Resistance to compromise undermines practices of mutual respect that are essential for a robust democratic process" (Gutmann and Thompson 2012, p. 34). And continue, "relationships of respect … are necessary to sustain any morally justifiable democracy under the modern conditions of deep and persistent disagreement" (Gutmann and Thompson 2012, p. 35). For Gutmann and Thompson, mutual respect is seen on par with mutual trust, as a virtue that can benefit democratic politics (Gutmann and Thompson 2012, pp. 109-117). Thus, while, for example, Bird regards mutual respect as a fundamental principle from which the imperative to compromise can be derived, Gutmann and Thompson regard mutual respect in more instrumental terms, as a means to a well-functioning democratic process. It is not that Gutmann and Thompson derive the value of compromise from a prudential calculation, but they also do not award it non-instrumental value in the way done by Bellamy, Bird, and others. This is an important difference when we want to understand how fundamental compromise is, morally and democratically.

Respect for citizens as what?
Our question, remember, is whether our democratic ideals require compromise. I have singled out a discussion of respect for equality as a core democratic principle, because of its prominence in the normative literature on both the justification of democracy and on compromise. However, I find the understanding of respect for equality in the literature underspecified. To get closer to an answer to our question of whether respect for equality requires compromise, we need to clarify the meaning of this norm and specifically to develop a notion of democratic respect.
May (2005, p. 342) argues, "there are many ways to express respect without resorting to compromise"; for example, by listening to and responding to others in public deliberation. This is true, but my question is if these other ways of respecting one's political opponents are sufficient for democratic respect. When we are concerned about democratic reasons for compromise, we cannot rest satisfied with the political process showing respect for citizens in some general sense. I show respect for other political theorists, for example, by listening to and responding to their arguments, but this does not amount to a form of democratic respect. Democratic respect is special, and what it entails cannot be explained by some generic idea of what it means to respect others. Democratic respect is tied to the political necessity of acting together and the democratic norm of including all citizens in this acting together. To answer the question of whether the democratic norm of equal respect requires compromise, we must answer the question, "Respect for fellow citizens as what?"

In the literature on democracy and compromise, two conceptions of respect are prominent, the equality conception and the community conception. According to the equality conception, compromise is justified, simply because it treats citizens as equals. In the community conception, compromise is regarded as justified, because it respects other citizens as members of the same community. I think both of these conceptions are inadequate, each for different reasons, and shall argue why before I turn to and defend a third conception, the co-ruler conception.
The equality conception, then, says that democracy and compromise are justified, because they treat citizens as equals. In this way, the norm of mutual respect is satisfied by equal treatment. The trouble with this defense of democracy and of compromise is that not only democratic procedures satisfy a norm of equal treatment; a lottery or a coin flip also treats and respects the involved parties as equals (Estlund 2008, chap. 4). Of course, not all forms of government are consistent with the principle of equal respect; autocratic forms of government are obviously not. The point is that some forms of decision-making that we normally do not regard as democratic are consistent with equal treatment, and, thus, respect for equality when understood in a generic or minimal sense cannot explain the moral importance of democracy or of compromise. Even if being treated as an equal is necessary for an institution to be called democratic; it is not sufficient. The idea of equal treatment or equal respect doesn't explain the value or need for compromise, but it also does not show the need for elections, responsiveness, and so on, which we regard as integral to representative democracy. This objection doesn't entail that democratic respect is not a form of respect for equality, but that it is a specific conception of respect for equality. Thus, we must further specify the meaning of equal standing that is inherent in democracy (Rostbøll 2015a).
Daniel Weinstock appeals to the community conception of respect when he argues, in response to May, that there are, in fact, principled reasons for compromise. According to Weinstock, compromise shows a commitment to a conception of community in which the members show an attitude of "basic respect" to each other. This respect includes "a willingness to make the ends of other members of the group one's own" and a rejection of a "winner-take-all society" (Weinstock 2013, pp. 543, 552). For Weinstock a commitment to such a conception of community is a principled, moral reason for compromise.
From the perspective of our question – whether compromise is inherent in the democratic ideal – the main drawback with the appeal to community is that the latter is "not fully endogenous to a concept of democracy" (Beerbohm 2014, p. 25). The invocation of community finds the value of democracy outside the concept of democracy itself or tries to reduce it to some other value. In other words, the value of community is not inherently democratic. If the family and friendship are paradigms of community (Weinstock 2013, p. 543), one can realize this value without democracy. In addition, when it is suggested that "democracy requires a strong, widely shared sense of community and that such a sense of community can be maintained only by a commitment to compromise" (Carens 1979, p. 135), this also does not show compromise to be integral to democracy. Here compromise and community are seen merely as instrumental to democracy, in a similar way as Gutmann and Thompson regard mutual respect and trust as instrumental to a well-functioning democracy. These may be weighty reasons for compromise, but they are not intrinsically democratic reasons for compromise.

Respect for fellow citizens as co-rulers
If the shortcoming of the conceptions of respect for equality simpliciter and of respect for fellow citizens as members of the same community is that they are either insufficient to justify democratic procedures or external to the concept of democracy, it seems obvious to turn to an idea that is indisputably integral to democracy, namely the idea of collective self-government (cf. Beerbohm 2014, p. 25). This idea goes beyond mere respect for equality to respect for everyone's freedom or autonomy and requires more than a lottery but the type of responsiveness involved in democratic procedures, as we know them (Rostbøll 2015a). And in contrast to the idea of community, collective self-government is endogenous to the concept of democracy. Thus, I propose that the sought for answer to the question, "Respect for fellow citizens as what?" is respect for fellow citizens as co-rulers. This conception of respect alone amounts to a form of democratic respect, and it alone can explain why compromise is integral to the ideal of democracy.
However, many political theorists are suspicious of the idea that democracy actually realizes collective self-government. Beginning with John Stuart Mill, they have pointed out the implausibility of the view that obeying the will of the people is to obey oneself: "The 'people' who exercise the power are not always the same people as those over whom power is exercised; and the "self-government" spoken of is not the government of each by himself, but of each by all the rest" (Mill 1989, pp. 7-8). The difficulty arises especially when the collectivity is understood in a way that does not respect the diversity of opinions among citizens. Thus, it is claimed, in the ideal of collective self-legislation, no respect is shown for the unique views of the individual but only for "the people" as a collective. This objection has been raised specifically to Rousseau and his notion of the general will (Rostbøll 2015b).
Often, the rejection of democracy as self-government focuses on majority decision-making. Thus Allen Buchanan writes, "it is simply false to say that an individual who participates in a democratic decision-making process is self-governing; he or she is governed by the majority … an individual can be self-governing only if he or she dictates political decisions" (Buchanan 2003, pp. 17–18). There are two remarkable aspects to this formulation. First, it reduces democracy to the act of decision-making and ignores the process of deliberation and negotiation that precedes it. Second, it assumes a dichotomy, according to which either you dictate political decisions or you are without political power. The idea of democratic politics as compromise stands in contrast to this view of collective self-government. It makes apparent the possibility and value of a democratic process of mutual accommodation and of having political power without being able to dictate political decisions. Thus, the notion of compromise has something to contribute to the discussion of the viability of the idea of democracy as collective self-government.
The suggestion, then, is that if we see not just majority decision making but also compromise as integral to the democratic process, this can meet or at least mitigate the objections raised against the idea of collective self-government. What is appealing about legislative compromise is that it, on the one hand, does not purport to represent a Rousseauian general will and that it, on the other hand, gives us an alternative to the reduction of democracy to majority rule. A compromise solution is, by definition, not seen by the parties as the true or best solution, or as representing the general will. It should be accepted as legitimate, I suggest, because it respects the parties as joint participants in collective self-legislation, or, in short, as co-rulers.
Compromise gives a special meaning to the idea of respecting citizens as participants in self-government. It does so in three ways. First, compromise does not require that we see "the people" as constituting an undifferentiated collective will. The quest for compromise takes outset in difference and disagreement as circumstances of politics and in respect for the plurality of opinions among citizens as a fundamental norm. As noted, compromise does not eradicate disagreements; rather, "the disagreements among the parties are embodied in the compromise itself" (Gutmann and Thompson 2012, p. 12). In this way, compromise should be seen in contrast to Rousseau's general will, which, as Hannah Arendt puts it, "bound the many into one" (Arendt 1990, p. 77). Compromise does not make us one but supplies a way of governing together that is respectful of our disagreements.
Second, respect for disagreement is important not because we should value diversity as such, but because we should respect citizens as persons with opinions of their own. The opinions of everyone should be given some weight of their own in political decision-making, independently of their content, simply because they are opinions of fellow citizens. As mentioned earlier, in making political compromises, we do not merely tolerate different opinions about what constitutes a good life for each; rather, we respect that citizens have different opinions about justice and the public good – and these disagreement are accommodated in political decisions. When the opinions of each citizen is given weight of its own not merely in the equal right to vote but also in the outcome through compromise, all citizens are more fully respected as joint participants in collective self-legislation.
Third, compromise brings us further toward the ideal of respecting everyone as a participant in self-legislation than do deliberation and voting. Rights of participations in deliberation and voting only entail inclusion or representation in the political process; they do not involve that one's judgment is represented in the outcome. It is exactly the latter that happens through compromise. In a good compromise, the opinions of all parties have not just been heard and counted; they have had a direct impact on the outcome. My contention is that compromise can contribute to approaching the ideal that everyone be participants in collective self-legislation in a way beyond mere majority rule.
Note that I do not claim that the ideal being realized through compromise is that citizens are collectively self-legislating in the sense that everyone get what they prefer. Rather, the ideal to be realized is that everyone can be a participant in self-legislation by having some impact not only in the process but also on the outcome (Rostbøll 2008, pp. 104f). Compromise does not secure, of course, that everyone is self-governing in the sense that they can dictate political decisions, but in a good compromise, all parties can see part of what they wanted reflected in the common decision. The fact that compromise can exhibit respect for citizens as joint participants in collective self-legislation is an intrinsic democratic reason for compromise and not merely a pragmatic reason for compromise. Thus, there are, indeed, reasons of respect – of democratic respect – for compromise.

Limits to compromise
If successful, the preceding argument demonstrates that a democratic conception of respect – respect for fellow citizens as co-rulers – supplies intrinsic or non-instrumental reasons for compromise in politics. But does it also entail that compromise should be awarded the same moral standing as the right to vote and to participate in deliberation? I think not. There are limits to compromise and its value is conditional in a way rights of participation in the political process are not. Indeed, it is exactly because compromise entails representation not merely in the process but also in the outcome that its value is conditional in a way the rights of participation in the democratic process are not. Rights of participation in the process are content-independent in a manner that representation in policy through compromise cannot be.
The right to speaks one's mind, to participate in public deliberation, and to vote in elections is something all citizens have, simply because they are citizens. If a citizen is denied the right to vote, she can complain that she has not been recognized for what she is, a citizen. She does not have to argue that she deserves this respect, and she need not prove anything about the merits of her opinion; she can demand respect and the right to deliberate and vote simply because of her standing as a citizen. Thus, rights of participation in the process are entirely independent of the opinions one has and the policies one supports.
Things look differently when we move from representation in the political process to representation in the outcome. It would seem odd not to evaluate one's opponent's opinions when one considers whether to compromise with her. No one I know of holds that we should compromise with any and all political opponents just because they are fellow citizens or fellow legislators. Some positive appraisal is involved when we compromise with others. All theorists of compromise hold that compromise only has value under certain conditions. As Avishai Margalit puts it, some compromises are rotten compromises. There are disagreements as to exactly what makes a compromise rotten or unacceptable, but it seems an absurd position to hold that compromises that accommodate any and all points of view are valuable.
I would suggest that in a democratic context, one should not compromise with parties that do no not accept the norm of democratic respect that explains the non-instrumental value of compromise in the first place. To be sure, exactly who accepts this norm and who does not will be a matter of judgment, but some parties clearly reject this norm. And no one can complain of being excluded from a compromise on the basis of a norm that they do not share. As John Rawls puts it, "A person's right to complain is limited to violations of principles he acknowledges himself" (Rawls 1999, p. 190). Much more needs to be said to justify and specify this limit to compromise than I can do here. For now the important point is that there are limits to compromise and not exactly where they lie. If not all compromises are valuable, and we should compromise only with person's whose opinions fall within a certain range of reasonableness, then the imperative to compromise cannot be unconditional but must entail an appraisal of the opponent's political views.
What I have just said only tells us when we are freed from the obligation to compromise with someone. It doesn't tell us when and with whom we are obliged to compromise. I believe that in a democracy, citizens have a pro tanto obligation to seek compromises with each other when they cannot agree (Rostbøll, forthcoming). They have this obligation because this is necessary to make democracy work under normal circumstances of disagreement, and because it is required by the principle of democratic respect. Apart from assessing the views of a minority, one may ask how big a minority must be in order to be in a position to demand being included in political compromise making. While this is an important question, also in order to clarify the idea of being respected as a co-ruler, I can only flag the relevance of the question here.
It might be objected that the right to vote is also conditional upon some form of appraisal. Thus, Robert Dahl writes, "Democracy … can be justified only on the assumption that ordinary people are, in general, qualified to govern themselves" (Dahl 1989, p. 97). However, Dahl speaks of people "in general," which means that there is no need of assessing the merits of co-citizens on a one by one basis, as there is when considering whether to compromise. When democracy first has been established, and we live in a democracy, we ought not to consider whether to grant citizens a vote based on their qualifications.
The preceding argument does not entail that compromise is not intrinsically valuable after all, that is, that its value is only instrumental in the sense that the imperative to compromise is contingent on it having better consequences than not compromising. Rather, the point is that the intrinsic value of compromise has conditions and within these conditions and limits, compromise carries a normative weight of its own that is independent of its consequences. A value can be non-instrumental even if it is conditional. Thus, compromise has non-instrumental but conditional value.

Conclusion
The question raised in this article is what moral and democratic significance we should award compromise. The great challenge in defending the non-instrumental, democratic value of compromise is to explain why one ought to concede something to one's opponents when this is not needed to create a majority. In other words, compromise does not seem to be democratically required in and by itself but only warranted as a pragmatic necessity to secure a majority or to create stability. I have argued that this challenge can be met by a democratic conception of respect for fellow citizens as co-rulers. The imperative to compromise has often been defended with reference to a notion of mutual respect, but this article shows that earlier attempts to derive compromise from a notion of respect are inadequate and has not given a specifically democratic and non-instrumental reason for compromise. While I argue that there are reasons inherent in the democratic ideal for compromise, this conclusion does not entail that compromise formation has the same significance and standing in a democracy, as do the right to participate in deliberation and voting. Compromise can make decisions more democratic but only under some conditions.


Acknowledgments

I would like to thank Aaron Ben-Ze'ev, Anders Berg-Sørensen, Hans Dabelsteen, Guo Dingping, Michael Freeden, Simon May, Tore V. Olsen, Theresa Scavenius, Daniel Weinstock, and Lea Ypi for comments on earlier versions of this paper. I would also like to thank Enrico Biale, Anna Elisabetta Galeotti and Federica Liveriero for editorial comments and my anonymous reviewers for their helpful feedback. Previous versions of this paper were presented at the Association for Social and Political Philosophy 2015 Annual Conference, University of Amsterdam, June 25-26, 2015; "Compromise and Disagreement", conference at University of Copenhagen, May 27-29, 2015; and Fudan University, Shanghai, China, May 21, 2015. I wish to thank the organizers and participants.

References

Arendt, H., 1990. On Revolution. Harmondsworth : Penguin.

Arneson, R., 2009. The Supposed Right to a Democratic Say. In: T. Christiano and J. Christman, eds. Contemporary debates in political philosophy. Malden, MA: Wiley-Blackwell.

Beerbohm, E., 2014. The Problem of Clean Hands: Legislative Compromise and Obstructionism. Paper presented at the Conference of the American Society for Political and Legal Philosophy, Chicago, February 27-28..

Bellamy, R., 1999. Liberalism and Pluralism: Towards a Politics of Compromise. New York: Routledge.

Bellamy, R., M. Kornprobst and C. Reh, 2012. Introduction: Meeting in the Middle. Government and Opposition, 47 (3), 275-295.

Bird, C., 1996. Mutual Respect and Neutral Justification. Ethics, 107 (1), 62-96.

Buchanan, A., 2003. Democracy and Secession. In M. Moore, ed. National Self-Determination and Secession, Oxford Scholarship Online.

Carens, J.H., 1979. Compromises in Politics. In J. R. Pennock & J. W. Chapman (eds.).
Compromise in ethics, law, and politics. New York: New York University Press, pp. 123-141.

Christiano, T., 2008. The Constitution of Equality: Democratic Authority and Its Limits. Oxford:
Oxford University Press.

Christiansen, F.J., 2008. Politisk forlig i Folketinget: Partikonkurrence og samarbejde. Aarhus: Politica.

Dahl, R., 1989. Democracy and Its Critics. New Haven: Yale University Press.

Estlund, D., 2008. Democratic Authority: A Philosophical Framework. Princeton: Princeton
University Press.

Fumurescu, A., 2013. Compromise: A Political and Philosophical History. New York: Cambridge University Press.

Gutmann, A. and D. Thompson, 2012. The Spirit of Compromise: Why Governing Demands It and
Campaigning Undermines It. Princeton: Princeton University Press.

Jones, P. and I. O'Flynn, 2013. Can a Compromise be Fair?. Politics, Philosophy & Economic, 12 (2), 115-135.

Lepora, C. and R. Goodin, 2013. On Complicity and Compromise. Oxford: Oxford University Press.

Margalit, A., 2010. On Compromise and Rotten Compromises. Princeton: Princeton University Press.

May, S.C., 2005. Principled Compromise and the Abortion Controversy. Philosophy and Public Affairs, 33 (4), 317-48.

May, S.C., 2011. Moral Compromise, Civic Friendship, and Political Reconciliation. Critical Review of International Social and Political Philosophy, 14 (5), 581-602.

Mill, J.S., 1989. On Liberty and Other Writings. Cambridge: Cambridge University Press.

van Parijs, P., 2011. Just Democracy: The Rawls-Machiavelli Programme. Colchester: ECPR Press.

Rawls, J., 1999. A Theory of Justice. Rev. Ed. Cambridge, MA: Harvard University Press.

Rossi, E., 2013. Consensus, Compromise, Justice, and Legitimacy. Critical Review of International Social and Political Philosophy, 16 (4), 557-572.

Rostbøll, C.F. 2008. Deliberative Freedom: Deliberative Democracy as Critical Theory. Albany: State University of New York Press.

Rostbøll, C.F. 2015a. The Non-Instrumental Value of Democracy: The Freedom Argument. Constellations, 22 (2), 267-278.


Rostbøll, C.F. 2015b. Non-domination and Democratic Legitimacy. Critical Review of International Social and Political Philosophy, 18 (4), 424-439.

Rostbøll, C.F. 2016. Kant, Freedom as Independence, and Democracy. The Journal of Politics, 78 (3), 792-805.

Rostbøll, C.F. Forthcoming. Compromise and Toleration. In C.F. Rostbøll & T. Scavenius (eds.). Compromise and Disagreement in Contemporary Political Theory. New York & London: Routledge.

Waldron, J., 1999. Law and Disagreement. New York: Oxford University Press.

Weinstock, D., 2013. On the Possibility of Principled Moral Compromise. Critical Review of International Social and Political Philosophy, 16 (4), 537-556.