Stephen Chilton

Department of Political Science
University of Minnesota - Duluth
Duluth, MN 55812-2496 (U.S.A.)

Tel: 218-726-8162 FAX: 218-726-6386

email: schilton@d.umn.edu
WWW: http://www.d.umn.edu/~schilton/


A larger and more muddled version of this work, "The Two Moments of Discourse Ethics", was delivered at the International Symposium on Justice: Kant, Rawls, Habermas, held at the Federal University of Santa Catarina, Florianopolis, Brazil, August 18-22, 1997. A still earlier version, "The Role of Emotion in Discourse Ethics", was delivered at the "First Reading XVII" conference on April 18, 1997. I am indebted to Beth Bartlett, Reginald Chevillon, Maria Wyant Stalzer Cuzzo, Nythamar de Oliveira, Lois Erickson, Sônia Felipe, Harry Hellenbrand, Dick Hudelson, Scott Johnson, Kirk Kersten, Paulo Krischke, Ken MacKendrick, Noelle McAfee, Dale Miller, Michael Neblo, Steve Ostovich, Kate Pearson, Vic Peterson, Steve Ropp, Rob Schaap, John Wilson Schiemann, Paul Sharp, Tracy Strong, Eileen Theimer, Tim Torgerson, John Walker, Saul Wax, and Janelle Wilson for encouragement and intelligent commentary. I was also fortunate both in my "Contemporary Political Theory" students, who courageously faced questions without answers, and in the members of the Habermas and Frankfurt School e-lists, whose pertinent concerns and active discussions helped me develop and structure this argument.

© Stephen Chilton 1998. This work was published in Sônia T. Felipe (ed) Justiça como Eqüidade: Fundamentação e interlocuções polêmicas (Kant, Rawls, Habermas) [Proceedings of the International Symposium on Justice: Kant, Rawls, Habermas] (Florianopolis, Brazil: Editora Insular, 1998).


Discourse ethics' demand for universal free agreement to norms seems incapable of satisfaction in the real world, despite a hypostatized moment (an "Ideal Speech Situation") believed capable of producing such agreement. This work proposes to add a new moment to discourse ethics to handle situations where a decision must be made and free agreement is not forthcoming. This second moment shifts our concern with agreement away from the specific norm(s) being considered and toward the underlying Relationship from which discourse ethics arises. This shift in concern brings to light a number of "transition conditions": normative criteria and socio-political practices that enable us to establish norms in a manner consistent with discourse ethics, norms that are just insofar as we have any clear sense of justice.

KEY WORDS: Jürgen Habermas; (D); discourse ethics; (U); universalization criterion; agreement problem; two-moment perspective; ideal speech condition


This work arose out of a difficulty I faced in understanding Jürgen Habermas's discourse ethics. As is commonly known, Habermas uses his universalization principle (U) to solve a number of philosophical problems. Here is one formulation of (U), taken from Habermas's essay on discourse ethics in Moral Consciousness and Communicative Action:
. . . every valid norm has to fulfill the following condition:
(U) All affected can accept the consequences and the side effects its general observance can be anticipated to have for the satisfaction of everyone's interests (and these consequences are preferred to those of known alternative possibilities for regulation) (Habermas 1990a:65).(1)
Note Habermas's emphasis on agreement. This hypostatized agreement allows discourse ethics to bypass both the attacks of noncognitivists (Habermas 1990a:55-56) and post-modern objections to normative systems (since agreement requires that no objections - including postmodern ones - be lodged), and of critics like Seyla Benhabib (1986) and Nancy Fraser (1986) who reject a morality that has no room for people's individuality (since each person agrees individually).

But where does the agreement comes from? The essay (and book) from which the above quotation is taken does not address the issue. We are left, then, with the problem, what gives us the right to assume the existence of any agreement? An initial, naive position is that agreement is unproblematic in principle, if not necessarily in practice: the people affected meet, they propose and discuss norms in some yet-to-be-specified open, equal forum - a so-called "Ideal Speech Situation" -, and they come to an agreement on what norms will bind them. This position sees agreement as blocked only by the communicative distortions arising from systemic factors or by people's emotional retreats into strategic action. People of good will, speaking openly, listening carefully, and reasoning calmly, will be able to reach agreement. In short, this "single-moment" view of discourse ethics contemplates an inevitable final agreement, at least in principle - perhaps requiring special rules of discourse, perhaps requiring effort and time, but achievable nevertheless. We may not yet recognize all the ways communication can be blocked, and we may not yet know how to overcome all blockages, but these practical problems can gradually be overcome.(2)

This answer is not satisfactory, however. What evidence is there that agreement will finally occur? Our actual experience seems to be quite the opposite: we are still arguing over the same issues that Plato and Aristotle raised.(3) We frequently find ourselves in the situation in which discussion has gone on for weeks or years, the participants have each clarified their own positions and carefully listened to those of the others, they have collapsed in exhaustion and ill humor, . . . but they now finally have to decide, and no agreement is in sight. Certainly we can describe forums that maximize the possibility of agreement or that minimize the severity of any final disagreement. These are useful, desirable accomplishments. However, it seems most reasonable to expect that complex issues, tapping profoundly different images of the Right and the Good, limited in the time and resources available, will not be settled by agreement.

We are left, then, with a real puzzle. In such circumstances, by what magic is agreement supposed to arise? Is discourse ethics merely idealistic, merely hypothetical? What could Habermas mean by saying (U) solves problems when it is almost never satisfied in practice? Does he intend discourse to continue until almost everyone agrees, and then ignore the remaining dissidents? Is he content with a merely theoretical, hypothetical solution? Does he have in mind some Ideal Speech Situation that would solve this problem? These options all seem either undesirable or unlikely. Some new approach seems necessary to handle lack of agreement.

There is a structural parallelism - what Jean Piaget calls a "vertical decalage" - in the development of moral reasoning, so that problems of dyadic relationships appeared later in structurally similar form as problems of social organization (Chilton 1988). We can therefore work backwards, asking how the problem of agreement is handled in marriage.(4) Spouses do disagree, and yet this seems to have created no terrible philosophical problems. Certainly there are abusive marriages and unequal marriages, but these do not cause us to call the institution of marriage itself into question. Disagreements - sometimes never-settled disagreements - occur even in good marriages. What can we learn from that?

The specific lesson drawn here is that spouses are (or should be) in a Relationship, a Relationship that takes precedence over any specific norm, so that it does not disappear because of any single disagreement.(5) So agreement is secondary to the existence of a Relationship that commits one to continued engagement despite disagreement. Applying this to discourse ethics, it implies that once we have run out of time and/or ideas about how to coordinate action, (U)'s injunction to seek agreement becomes a command to agree. If we are to take (U) and (D) seriously, and if we have reached a point where a decision is necessary - where failure to decide is itself a decision -, then we all agree that our primary task is agreement regardless of our free preferences.(6) In effect, (U) reverses course: "if the mountain won't come to Mohammed, Mohammed must go to the mountain"; if people's free preferences do not freely result in agreement, then (U)'s requirement of agreement recoils against their free preferences. (U)'s requirement changes the situation such that the central point of agreement becomes our Relationship with one another, not the specifics of any particular norm (within some limits, discussed below).(7) Even though our Relationship is strained, it still exists, and this new situation is not a failure of the Relationship but instead a statement of its priority above any specific norm. In other words, when pushed to the wall, (U) ceases to be a request for discourse and becomes a command to agree on the only thing left to agree on: we are still in a Relationship with each other arising from the continuing requirement for coordinating action. Note that this leaves intact the logic of Habermas's (1990a) justification of (U): whether we agree or not, the presuppositions of argumentation remain that we are in a Relationship of trying to come to agreement, a Relationship that exists even in the face of current disagreement.(8) His argument concerns our need to assume we can reach an agreement with each other, not the existence of such an agreement. Even if we can't agree freely in the time and circumstances at hand, our presuppositions do not change, and thus the existence of the Relationship embodying them does not change. The issue therefore becomes how we are to constitute our Relationship, an issue that includes but is larger than any particular norm. Although we must agree, willy-nilly, we must do so in a way that still reflects and supports the nature of our Relationship.

Now the dynamics of our discourse have changed. The goal remains agreement, but different processes and considerations are involved. Discourse ethics now consists not of one moment but two, which I will term M1 and M2. In M1, the goal is to agree on a norm that violates no one's sense of the Right or the Good. That goal is still preferred, obviously, but it may be too restrictive for agreement. So we move into M2, where we seek agreement even though some or all people will feel violated.

O.K. - but so what? Have we gained anything in recognizing M2? Is M2 merely a new name for the unbridled strategic action that already goes on in the face of persistent disagreement? Are we back to politics as usual? No - what we've gained is an understanding that the key issue in M2 is the existence of our Relationship. Even though we must agree, we can do so only in ways that respect the Relationship giving rise to the imperative in the first place. If the existence of this Relationship forces us, willy nilly, into some shotgun agreement with each other, it can only do so to the extent that the norms and practices agreed to are consistent with and support the Relationship creating them.(9)

This gives rise to a number of second-order "transition conditions", second-order because they are not specific norms (zero-order) or specific rule-making processes (first-order), but they are instead conditions reflecting the foundational Relationship giving rise to the processes of making norms. Some of these transition conditions prescribe our fundamental attitude in M2, some dictate the way we engage each other there, some constrain the norms we allow to be established, and some dictate how we behave in the aftermath of a norm having been established.

Before going into their details, however, I note that these rules are not unique to M2; they hold within M1 as well and thus throughout discourse ethics. However, since the universal agreement in M1 makes them unproblematic there, they are not prominent until M2. I also acknowledge that even if these rules constrain norms and guide our discourse, they do not in the end provide specific norms. I haven't forgotten that we need specific norms, however; I am simply deferring their consideration to a subsequent section.

So what are these transition conditions? First, we need to carry into M2 a number of basic attitudes about our situation and each other. In particular, we must accept that someone - and we explicitly accept the possibility that it will be ourselves - will be hurt by any norm we establish. "Acceptance" does not mean we masochistically delight in being hurt but rather that we recognize the risk of being hurt as an inherent part of acknowledging and honoring our underlying Relationship. We accept that even if we cannot all agree on a norm, we do all agree to hold to the Relationship despite the possibility that the norm we establish will feel like a violation to us and the certainty that it will feel like a violation to someone.

We also accept that none of us has a privileged position vis-à-vis others in determining what is Right. We have left behind in M1 the chance to make and support such claims. We each still feel ourselves to be Right, are convinced we are Right, and yet we recognize that there is no longer any means of pressing our feelings and convictions on others who themselves have such feelings and convictions. We are entitled to hold and press for our preferences, and we accept that we can claim no more right to them than anyone else.

These attitudes will seem less idealistic and/or misguided if we remember that all of this takes place within the context of both thorough, fair discussion having been exhausted in M1 and the other transition conditions being in place to get us into M2. Let me clarify this. In an oppressive society, most people's usual experience with public issues is to have their opinions disregarded or disrespected in the first place, so M2 must seem like it means an humble acceptance of whatever norm is then established. But this is not the case. The attitudes I suggest adopting are predicated on one's already having had one's say in M1, on others having listened closely to one's position.(10) I am also insisting that one's situation and opinions be recognized fully in M2, where "recognized fully" means consistent with the transition conditions. These preliminaries are essential to move us beyond mutual suspicion and strategic action.

There are also certain constraints on the norms that M2 can produce, since not all norms are consistent with the Relationship giving birth to them.

The most important and obvious constraint is that no norm can prevent people from expressing their disagreement with existing norms or from continuing the process of M1. For example, a Nazi-proposed norm to exterminate Jews would be a nonstarter. The very existence of M2 is predicated on our accepting disagreement over norms. If silencing people's disagreements was a legitimate way of settling disputes, we would have no need for discourse and agreement in the first place.  I note in this connection Jane Mansbridge's (1996:58) injunction that we need to "recognize and foster enclaves of resistence"..(11)

No one (or no identifiable group collectively) can consistently be subject to greater or more frequent violation than others.(12) Once we have reached M2, there is no longer any way to favor one party over another in the Relationship bringing M2 into existence. One's agreement to the norms emerging from M2 is predicated on one's being treated in terms of the Relationship. If the norms actually adopted consistently violate one person's free will and not another's, then there is some power differential that needs to be corrected. Furthermore, . . .

 . . . lacking a way to rule that any given person must bear a special burden, we constrain our norms to those minimizing the degree to which the free will of any single person or group of people is violated. In M2 no one can claim to have special insight into what norm is Right. Such claims lie in the domain of M1, where they can be tested; in M2, though, we are taking our normative differences as given. Obviously each of us will feel that we have special insight and that our position should be adopted, but in the absence of discursive justification - left behind in M1 - we accept that none of us has a special entitlement. The imposition of a norm violates people's free will to varying degrees. Each of us would like to say that because we are Right, the Wrongness of others implies they must endure the violation that their being forced into our Rightness entails, but we cannot make this claim. We can, however, require that we not be subjected to greater violation than necessary. Recognizing that the necessity for agreement will inevitably result in a sense of violation for some, people will not agree to a Relationship whose norms unnecessarily disadvantage them. And the less necessary the violation appears, the more strenuously they will resist the system of decisions.(13)

The two previous constraints resemble Rawls's (1971) "maximin principle", that inequalities arising from the basic structure must make the worst off in the society as well off as possible. Note, however, that in discourse ethics the maximin principle arises not as the assumed outcome of theoretical consideration in a hypothetical Original Position but rather directly from the need for agreement in practice.(14)

Discourse ethics, particularly in M2, requires a Relationship of mutual confidence in others' good faith and good will. When a norm is established, people agree to it because of the Relationship, not necessarily because it accords with their sense of the Right and the Good. Someone's (maybe everyone's) free preferences will be violated. Someone (maybe everyone) will be hurt. This strains our Relationship with each other, undercutting our mutual confidence. Some attention needs to be paid to this strain; mutual confidence cannot be assumed into existence, even if it "ought" to occur. Within marriage, divorce occurs even though marriages have the advantages of frequent contact, mutual dependence, initial mutual choice, social support, and legal incentives and penalties. Why would we imagine that the enormous diversity of a society can somehow unite without equally significant support? To put discourse ethics into practice, therefore, we need to find emotional supports to carry us through such strains.

M1 presents no special problems, since all have agreed out of their free will. In M2, however, powerful emotions arise as people's free will is violated, creating a sense of "winners" and "losers" according to whose free will has been more violated. The resulting strain needs to be relieved insofar as possible.

U.S. society, at least, provides little support for losers. In our political culture, political discourse is cast as a battle between opponents, not as a search for a decision in which any outcome will be painful to some. When the battle is over, the "winners" get to claim that the Right has been served and the Wrong defeated, so that the pain felt by the "losers" is only the just penalty for their having been Wrong and no further examination of the question is needed. The "losers", on the other hand, are left with only the taste of defeat, the continued conviction of being Right, impotent anger at their inability to have their perceptions understood and engaged, and resentment at their forced compliance with the victorious norm. All of this goes to divide us from one another, regardless of the issue, and it breeds continuing conflict.

Let me give a practical example: I believe that the conflict between the Christian Right and "secular humanists"(15) in the United States continues as a direct result of the continued cultural dominance of the secular humanists. Even though I believe that the remedies proposed by the Christian Right would not be effective in achieving what they desire (and would be morally wrong as well), I also believe that the pain that they experience is real (and in fact stems from real problems in U.S. society caused by capitalism's colonization of their lifeworld). By failing to acknowledge this pain, and even further by calling the Christian Right "fascist" and so on, secular humanists create many difficulties for themselves: they deprive themselves of what value lies in the Christian Right's perceptions, they inhibit the discourse that might enlighten both parties, and they degrade the sense of Relationship on which we all depend: for the coordination of action despite disagreement and for mercy when we are the losers.

Similar things seem to be happening around the issue of gay rights. When a victory for gay rights is achieved, little attention is given to the feelings of the "bigots" and "homophobes" who oppressed gays for so long. Such people may feel genuinely distressed at gay marriage, gay soldiers, gay teachers, gay tenants, gay colleagues, and so on; but "get over it" or "you deserve this for all the pain you inflicted" frequently seems to be the only response to this distress.(16)And lest this discussion appear too one-sided, I note that the same phenomenon in reverse happens when the "forces of righteousness" triumph over the "sodomites".

To deal with such strains on our Relationships with each other, I propose the following as useful political directions:

  1. The establishment of a norm should be accompanied / followed by the invocation of institutions and rituals designed to remind us of our underlying Relationship and call forth the attitudes supporting it. We need to acknowledge that we are still committed to it, not just in our hearts but also publicly, so that the Relationship is established and maintained as the cultural norm.(17)
  2. Among the elements of these institutions and rituals is an explicit, public statement that the norm is not a victory for the winners but only the beginning of an experiment assessing its value.(18) Since there is still no universal acceptance of the norm, M1 continues even when M2 ends; the winners do not attempt to secure their victory as a way of ending the search for better norms.
  3. The obverse element is an explicit, public statement that this experiment has to be given a fair chance. In particular, the losers don't attempt to sabotage the norm, actively or passively.
  4. We continue to engage in a search for systematic power differentials and the associated violation of weaker people / groups,(19) and we revise our system of decision-making in that light. For example, we may discover that certain groups of people - say, women - have difficulty speaking up for themselves, and this difficulty then reinforces an oppression that in turn reinforces the original difficulty. In such circumstances we might consider adopting policies like, say, "All women must speak before any man speaks."
  5. Finally, the winners need to resist their sense (or at least mute their huzzahs) that since their Right has been victorious, they need not acknowledge those who were Wrong, i.e., the losers. Instead, the winners need to explicitly, publicly recognize that in being willing to endure the sense of violation, the losers are in fact contributing to the underlying Relationship. And a beneficial consequence of this would be to support the winners' willingness to lose in subsequent discussions.(20)
I have deferred until now any discussion of how decisions are actually, ultimately made in M2. What can be said about a process in which we seem to have neither justice, reason, nor agreement to guide us but only a sense of Relationship? I will discuss this in terms of decision-making within a marriage. The problems of agreement in M2 are not essentially different from within a marriage to within an entire society,(21) and the exposition of M2 will be simpler for the narrower case. For concreteness I will call the two spouses Sergio and Susanna.

I distinguish two elements of a marital Relationship. One element is the existence of their Relationship in the first place, a Relationship mutually defined and supported by the two partners. "We are, and will continue to be married to each other. We are both in this marriage and responsible for defining it." This is common to all marriages. The other element is the content of the Relationship, which is specific to each marriage. For example, Sergio and Susanna might agree that "in our marriage, . . .

. . . we use pottery rather than china tableware."
. . . the person who doesn't cook, washes up the dishes."
. . . Sergio is responsible for maintaining everything outside the house, and Susanna everything inside."
. . . Susanna gets a two-week vacation by herself every year."
. . . Sergio has the final say on what movies they go to."
. . . we alternate going to Sergio's family and Susanna's family for Christmas."
. . . we unwrap our presents on Christmas Eve, not Christmas Day."
. . . we celebrate birthdays and anniversaries on the actual day, not when merely convenient."
. . . etc.
These and other such specific agreements are what make their marriage theirs. No two marriages will have precisely the same content; each will surely have its own agreements. Sergio and Susanna are certainly free to proselytize for their marital norms, but the mere fact that they have agreed to them does not make them generalizable. They apply to Sergio and Susanna's marriage only, so that their agreement to them carries no moral weight in others' consideration of their own agreements. But all marriages have in common the existence of the Relationship.

These specific agreements are the "lifeworld" of a marriage. This rich concept has many connotations and implications; here are a few that bear on the present argument:

Beyond the transition conditions, then, the actual decision-making depends on the entire complexity of lifeworld considerations unique to each marriage and to the decision at hand. Consider a specific decision: will Sergio and Susanna go to a movie or stay home this afternoon? Despite the apparent simplicity of the issue, many considerations could bear on it, of which the following are just a few: Notice that these considerations, and an infinity of possible others, are unique to Sergio and Susanna's marriage and to the specific issue. In the end, Sergio and Susanna make their decision based on circumstances that are unique to them and perhaps even accidental. Having already applied what reason they can bring to it, they at last have to lay their future at the feet of their lifeworld.(22)

The overall point I am making about the content of a marriage, the actual norms governing it, is that there is no overall point. The general considerations mentioned earlier take us as far as Justice can; beyond them lie only the specifics of the relationship and the accidental ways in which our separate lifeworlds collide and are meshed with one another. It is in this sense that I understand Nietzsche (1966:§153) when he writes, "Whatever is done out of love always occurs beyond good and evil." After M1 and the transition conditions allowing us to enter M2, we are at a point where the decision is made in a loving Relationship with each other but is otherwise beyond any rational sense of good and evil. The ideal of free agreement has not been reached, but this is moot: unless we can produce a norm that in fact commands agreement with less violation of people's free decisions, there is no meaningful sense in which the norm actually established is unjust. Agreement can be said to exist, as discourse ethics demands, when one's sense of violation is embedded within the larger recognition of one's Relationship with others even when they disagree. The violation is real, it still hurts, but the sting is less when the violation has not arisen from the ill will of others.

It is inappropriate to talk about the decision-making process in M2 in terms of power or strategic action. Within discourse ethics, i.e., within the framework of a Relationship that both spouses seek to respect and maintain, the final decision-making does not devolve into a power struggle but into a dialogue that continually creates and re-creates the lifeworld of the Relationship. Although the spouses are in extremis, their interaction is still governed by their maintaining and establishing anew the basic Relationship between them, and its normative content is still subject to the transition conditions.

Many people are likely to see power as inevitably involved, even if only to the extent of one spouse arguing more cleverly than the other. I reiterate that this is not the case. As Lois Erickson (personal communication) expresses it, a true Relationship has no room for ego processes; no "persuasion" should be involved. Even the ability of one spouse to argue better would result in a relationship contradicting the earlier injunction against any party being consistently subject to greater / more frequent violation than others.(23)

In conclusion, I believe that the meaning of "all" in Habermas's universalization principle can be resolved by seeing discourse ethics in terms of two moments, not just one. These two moments arise because people's discursive validity claims always already constitute a Relationship between them, a Relationship that exists prior to and independent of agreement on particular norms. We need to structure our society and conduct our political action in terms of this Relationship.


In this appendix I want to address the discomfort some readers will experience about my analogy between marriage and the State, an analogy the skeptic might hold to be imprecise or - given the number of obvious differences between the two situations - suspect. A few obvious differences:

For some readers this discomfort with the analogy between marriage and the State will be compounded with a discomfort with discourse ethics itself, particularly the requirement of agreement by all.

I feel it necessary to address this discomfort more directly than is possible in the flow of the text's argument. Here is a recapitulation of the argument, intended to relieve the discomfort:

1. I take Habermas's argument at face value. Norms regulating any social entity must be agreed to by all in order to be considered norms. This does not speak to the so-called "motivation question" ("Why be moral?"). It doesn't say that people are forced to be moral, or will inevitably act morally. It simply says that from a normative point of view, all must agree. Habermas's justification does point to a means of solving - or at least addressing - the motivation question by pressing the person to recognize h/her implicit assumptions, but it isn't a magic wand. The action of this pressure can be seen in the long-term development of our species, I believe, but the argument doesn't rest on that. So if you disagree with me on these issues (the need for agreement), go argue with Habermas This work attempts to solve a problem that arises strictly within the framework of discourse ethics; the validity of discourse ethics is a separate issue.

2. The central claims in this work are that the process of discourse ethics should be conceived of in terms of two "moments", not one, and that the two moments are connected and coordinated by a set of "transition conditions". I justify these claims through the following arguments / considerations:a. Agreement on norms is required. According to discourse ethics, this applies to all norms, whether those within a marriage and or those within the State. True, in applying discourse ethics to marriage, we bracket the fact that marriages are embedded in the larger framework of the State. But this bracketing is necessary to the analogy, does not distort the marital situation very much,(24) and does not affect the bare fact that agreement is required in both. What about the fact that divorce is unique to marriage? Well, an offhand response is that one can divorce one's State simply by emigrating and renouncing one's citizenship. But even leaving that aside, notice that the examples and considerations used do not involve divorce; the argument makes the (implicit) assumption that divorce is not an option. This is consistent with the analogy, since Habermas's justification of discourse ethics arises from the observation that even the would-be complete moral skeptic cannot escape the Relationship created by language. Certainly divorce exists, but like any moral theory, discourse ethics does not solve the "compliance problem". Telling people what is required for a norm to be moral is one thing; inducing them to actually follow moral precepts, however right, is another. Like discourse ethics itself, the present argument ignores the compliance problem.

b. Agreement on norms is difficult to attain through open discourse. The considerations involved here apply to both marriage and the state:

i. People's views of the Good will differ, obviously in the State, but also in marriage even though one has chosen one's marriage partner. These differences might arise, but do not essentially arise, from State action.(25)
ii. Such differences in views of the Good may not be overcome within the limits of time and energy available to us. Decisions often need to be made willy-nilly; even failure to make a decision is often a decision in itself; decisions often need to be made without the knowledge of their effects; participants may not have the psychic energy to change their sense of the Good quickly, even if they should. All of these things are true whether one is speaking of marriage or the State.
iii. Differences in views of the Good need not be something to be overcome, at least from a moral point of view. They make cooperative life difficult, but one cannot say that the mere fact of difference means that one view is better than the other. The Good is subordinate to the Right, of course, so that views of the Good that conflict with justice must to that extent give way,(26) but there is no sense in which there is only one final, absolute, correct sense of the Good. Therefore even with an infinite amount of time and resources available for free discourse, we may still not reach free agreement, that is, agreement in which no one's sense of the Good is compromised. The possibility of essential conflict exists whether one is speaking of marital or State norms, i.e., agreement between marital partners or within a society generally.
c. In discourse ethics, or so I argue, the necessity of agreement is prior to the existence of different senses of the Good, meaning that agreement is required even if free agreement is not possible. The considerations producing this conclusion - basically, the logic of the argument advanced in Habermas (1990a) - are the same for both marital and State relations.

d. If agreement is to be legitimately forced, then the mechanism whereby this is accomplished (M2, in my terminology) is legitimate only when it respects the underlying ground of its legitimacy. This respect is named the "transition conditions" in this work. Since the ground is the same for both marriage and the State, as I said previously, then the transition conditions apply to both. The reader should note that they are abstract enough to apply equally to both levels.

In sum, the argument presented in this work applies to both marriage and the State equally. It is presented in terms of marriage for three reasons only: marriage is a simpler situation to present; the agreement problem in marriage is one with which almost all of my readers will be familiar; and the discussion of these issues in terms of marriage serves to remove the issues from the overly-contentious, overly-complex situation of State norms. Obviously this mode of presentation comes at a price - as witness the need for this lengthy appendix -, but overall I believe it is a price worth paying.


1. I am bracketing the question of whether (U) itself is a valid criterion for norms. Habermas's sketch of a justification of (U) seems unproblematic to me. His argument is incomplete, as he acknowledges (and, indeed, as his reconstructive procedure would in any event be in principle), but I see no reason in principle that it couldn't be formalized to the extent necessary here. 

2. There is the additional attraction that regardless of whether final agreement is produced, any country's politics would markedly improve after instituting even the bare bones of an Ideal Speech Situation. Still, let it be noted that whatever its virtues, Habermas no longer regards the Ideal Speech Situation as a philosophically adequate foundation of moral theory. 

3. My colleague Scott Johnson sent me this quotation from Victor Scheff [citation lost]: "How can Quine expect universal consent on anything in any language-using community that allows for the existence of philosophers?" I note also that Habermas seems to acknowledge the problem:

Legal experts have the advantage of discussing normative questions in connection with cases to be decided. Their thinking is oriented to application. Philosophers avoid this decisionist pressure; as contemporaries of classical ideas extending over more than two thousand years, they are not embarrassed to consider themselves participants in a conversation that will go on forever (Habermas 1994:135).
4. The arguments here can be applied to any mutually maintained, dyadic relationship: friendship, same-sex partnerships, etc. I discuss marriage not to privilege it over such other relationships, thus perpetuating the oppression of nonstandard marriages, but rather because it provides a clear, elaborated, and widely understood example - and with which I have some experience. At the dyadic level, at least, the example is unproblematic. I am indebted to Ken MacKendrick for pressing me on this issue. 

5. To clarify an important distinction in the following discussion, I distinguish between "Relationships" and "relationships". Capital-R Relationships are those founded on the presuppositions of argumentation, those in which action is coordinated by means of good reasons, those that conform to the "transition conditions" (defined below). Like "the presuppositions of argument", from which it derives, the nature of Relationships is determined only through a reconstructive science - meaning that no ideal type exists, only contingent "place-holders" for it. Lower-case-r relationships, on the other hand, are those that exist in reality. They do not necessarily conform to the transition conditions, and in the subsequent discussion the term usually refers to relationships that are abusive in nature. 

6. I appreciate Reginald Chevillon (personal communication) for pressing me on this issue. 

7. "If M2 [introduced on the next page] inherits (U) . . . , it is the relationship that is agentic, and not the individuals. American ideology typically constitutes relationships as individuals exercising autonomous agency in each other's company" (Rob Schaap, personal communication).

8. The key point is that communication puts us always already in a relationship with each other based on our implicit promise to redeem the validity claims we implicitly make in communication. 

9. Reginald Chevillon (personal communication) asks how these are to be discovered - a priori or a posteriori? This is the same question as asking Habermas how the presuppositions of argumentation are to be discovered, because the Relationship referred to here is derived from these presuppositions. Choosing either alternative loses the dialectical nature of the reconstructive science by which the presuppositions are discovered. 

10. Does M1 have a point if agreement can never be reached? After all, it makes heavy demands on us, even though it only requires us to search for something we all agree on. It is demanding in that we have to: look for commonalities; generate creative candidate norms; understand the Other, considering ideas and perspectives that may at first be repellant; examine our own values, motives, and so on, "coming clean" about what we really deserve and want, an examination that may prove painful and embarrassing; and overcome (or at least process) distress and anger. But despite these demands, actors should spend as much time as possible in M1 before turning to M2; even if the prospect of our finding a norm in M1 seems remote and no agreement is in fact reached, our efforts in M1 can still accomplish a variety of useful things:

So in proposing a second moment, I am not suggesting that we skip the first. M1 might produce agreement, and even if it doesn't, our efforts in it will almost certainly improve the final agreement. So we continue in M1 as long as possible, pursuing it even after M2 has produced an agreement. I am indebted to Michael Neblo for pressing me on this point and to Maria Wyant Stalzer Cuzzo for reminding me, in the midst of my preoccupation with M2, the continuing importance of M1. 

11. I note in this connection Jane Mansbridge's (1996:58) injunction that we need to "recognize and foster enclaves of resistence". 

12. Ideally, we would equally experience winning and losing. One couple I know makes decisions by tossing a coin when they cannot agree. 

13. This "maximin" criterion only applies strictly to foundational norms, norms affecting people's overall life chances. For more limited norms there is obviously some room for accepting greater violations than necessary in one area in order to achieve less violation in another. 

14. Rawls's "difference [maximin] principle" of justice is subject to the objection that in his Original Position, people might choose a utilitarian (utility maximization) principle instead of maximin principle. Not knowing what social position they would occupy, they could choose to maximize their expected welfare. Rawls advances some good arguments why this would not occur, including some philosophical problems with the concept of utility itself, but while his arguments have weight, they do not seem logically decisive, particularly when compared to the present work's more direct approach. 

15. The term is in quotes here because it is more an ideological term used by the Christian Right than a clear or accurate description of people who believe differently from them. I will not use the quotation marks in further references, however. I follow the same practice for other ideologically loaded terms. 

16. Saul Wax (personal communication) points out that many gay sons and daughters have reached out to their parents, many of whom in turn have responded with loving acceptance. Certainly organizations like PFLAG (Parents and Friends of Lesbians and Gays) do not take a "get over it" attitude. 

17. Murray Edelman (1967) and others are rightly skeptical of such symbolic gestures when they merely disguise the oppressive reality. But however we guard against this problem of misrepresenting reality, the fact is that we cannot get along without such symbols; society is formed and maintained in important ways through such symbolic structuring. 

18. Insofar as possible. Not all decisions are reversible. 

19. See Fraser (1986:425-429) for a more extended description of this. It is in this arena that postmodern critics make such an important contribution. 

20. I am indebted to Eileen Theimer for this observation. 

21. In Chilton (1988:49) I have discussed this similarity with respect to the "vertical decalage" (structural parallelism) between Kohlberg's Stages 1-3 and 3-6. But I need to clarify that this parallelism only obtains for the points made below. The actual process of M2 will be significantly different between marriage and the larger society, and I explicitly note that M2 in the larger society needs to recognize the unique, systemic forces at play. 

22. I have not found a good way to point out here that the struggle between the two spouses will involve some of the issues that Schiemann (1998), drawing on Schelling's (1980) analysis, terms "abstract salience". Such dicta as "split the difference" or "follow precedent" have a power of their own even if not a normatively justified or rational power. As Schiemann argues, it is rational (or at least not irrational or immoral) to use such dicta in coming to agreement and settling disputes even if they are not themselves rational or moral. 

23. Shifting for a moment from philosophy to social commentary, I hold that people have difficulty understanding Relationships without power because we see so few models of them. In a world organized through the attitude of mutual indifference underlying market systems - increasingly so since the fall of Soviet-type state socialist regimes -, we can create and sustain other models of relating only with difficulty. 

24. Recall that the State does not appear in the examples and considerations used (going to the movies; the heat of the afternoon). 

25. Thus forces in the larger society might give women one sense of the Good and men another, and these forces might arise because they serve the desire of those in power (men) to continue in power, but differences in the sense of the Good can arise even if no such social force is involved. A lesbian relationship, for example, would not involve gender-related differences in the partners' sense of the Good, but such differences exist anyway. 

26. Thus one's conviction that the, say, Zoroastrian life is desirable may be a valid choice for oneself, but that conviction does not permit one to force all others to become Zoroastrians.

Page URL: http://www.d.umn.edu/~schilton/Articles/SMDE1998.html
Page Author: Stephen Chilton
Link to Home Page: www.d.umn.edu/~schilton/index.html
Last Modified: December 4, 1999
Mail suggestions and comments, especially re. typos and other errors, to: schilton@d.umn.edu
Honor Roll of Proofreaders and Colleagues: www.d.umn.edu/~schilton/Honor.html
This way to the UMD homepage
This way to the Political Science Department home page

The University of Minnesota is an equal opportunity educator and employer.
Copyright © 1999 Regents of the University of Minnesota. All rights reserved.