September 5, 2002


[Chapter 3]

The Agreement Problem: Pragmatic Limits of Empathy, the Persistence of Disagreement, and the Use of Force

[URL: TheAgreementProblem]

SECTIONS (& perhaps subsections):

            I          The Agreement Problem

            II         Why Is It a “Problem”? Consequences of the Agreement Problem

                        A         The use of force and the abandonment of morality

                        B         Case study (or maybe just an anecdote)

                        C         Encouraging stubbornness and game-playing

            III        Existential Origins of the Agreement Problem

                        A         Natural limits of empathy

                        B         Variation in senses of the Good

                        C         Decisions made willy-nilly

                        D         Distress patterns and painful emotion

            IV       Summary


I    The Agreement Problem

My first glimpse of the ways of relating perspective came as I struggled to take seriously Habermas’s recognition that the only final standard of morality is agreement among all affected by a proposed norm. This in itself is not an outlandish concept, because as Habermas notes elsewhere, every moral injunction always already claims to command the assent of all, even while such injunctions can only be made contingently. Even fascists hold that their norms should be agreed to by all, even if fascism doesn’t rest its legitimacy on that agreement but instead sees the agreement as deriving from the legitimacy. The same is true of U.S. society, for that matter, despite our nominal concern with the consent of the governed. “Consent” has two different meanings here: consent in the sense of agreeing to the basic institutions whereby leaders are selected and laws made, and consent in the sense of participation in those institutions. Today, however, the difference between those two consents is conveniently ignored, so that the institutions themselves are deemed legitimized by people’s participation in them – or even by people’s opportunity to participate, regardless of its effectiveness.

            So “agreement” by itself is not what distinguishes Habermas’s moral theory. It is, rather, the idea of agreement in practice that distinguishes it. Habermas (or the ways of relating perspective, for that matter) is not satisfied with theoretical agreement, which is the cover story of even the meanest moral theory. Fascists, classical liberals, Kant – all argue that people ought to agree if they only understood things aright. But whether these arguments are simple or complex, smart or stupid, respectful or disrespectful of others, they are not in fact justified except by the gold standard of people’s actual, uncoerced agreement. There is simply too much room for self-delusion to allow anyone, or even any group, to assume others’ agreement. We need not even blame self-delusion; all of us are unknown to each other to some extent. I will discuss this in more detail in §III; my point here is simply that no norm can really be said to be agreed to unless it is – duh – really agreed to.

            There has been much discussion of the conditions under which agreement can be said to be uncoerced. Some of this discussion concerns the so-called “Ideal Speech Situation” (ISS), Habermas’s initial foray into the question. Footnote And of course the general question – what are the mechanisms of oppression? – extends beyond the specifically Habermasian language and framework of the ISS. But despite the extensive work in this area, we must remember that it is still only about a contingency. Our basic proposition is, “IF we can achieve uncoerced agreement to a norm, then it is morally valid.” The discussions about the ISS and, more broadly, the nature of oppression are concerned only with what the predicate clause means; they are not concerned with whether agreement an in fact be reached. I sense that the belief is that true harmony is blocked primarily by oppression, so that once we understand and overcome oppression, agreement will follow naturally. In the forum of the ISS, the true exchange of perspectives will lead quickly or at least reliably to an agreed-upon norm. Well, perhaps so, but I can also see an opposing argument: the more we ensure that everyone’s views are duly included in our deliberations, and the more we guarantee that everyone’s agreement is autonomous and uncoerced, the slower and harder it will be for us to reach agreement. Oppression gives at least an appearance of agreement; I don’t see any magical force for agreement arising just because we are released from oppression’s lockstep. The achievement of agreement remains a problem.

            This is not the agreement problem, however. We face the agreement problem, in my sense, when we need to act even though disagreement remains. In real life, failure to decide is usually a decision itself, and there is no guarantee that the necessity of deciding will bring agreement. So the agreement problem is, “What are we to do when we are as yet unable to reach agreement? What is morality when we lack the affirmation of its gold standard?” I am reminded of the title of the book, What to Do until the Messiah Comes. Even while we seek agreement, even while we refuse to admit anything less as defining morality, what do we do in the meantime?

            Do not take this as an idle question, or a matter of application instead of theory, or (worse) idle trouble-making. From a pragmatist’s point of view, “by their fruits ye shall know them.” Morality inheres in what we actually do, not in what some counterfactual theory asserts. Footnote If the actual results of, say, discourse ethics is, “People sit around trying to reach agreement while actual decisions get made willy-nilly – through institutional inertia, or default, force, or fraud” –, then it is an ideology of powerlessness, not morality.


II    Why Is It a “Problem”? Consequences of the Agreement Problem

Please note that I am not arguing either against agreement as the criterion of morality or against fair decision processes. All I mean to do here is point out their practical consequences. But what are the practical consequences of the agreement problem? Are they so dire that the ordinary level of disagreement must be termed a Problem? One might argue that discourse ethics need serve only as an ideal version of morality, one we can aspire to and be guided by, so that our actual decisions are merely imperfect rather than Immoral.

            I have some sympathy for that argument. I believe that discourse ethics and other, similar philosophies do in fact help us even if they only serve as ideals. However, I also believe that there are some important ways in which mere idealization creates serious problem – yes, Problems and Immorality. I take up these problems below.


A.  The use of force and the abandonment of morality

The agreement problem asks us how we are able to relate to each other when the only certainty of morality – universal, uncoerced agreement of those concerned – cannot guide us. Guides such as majority rule do exist, but they provide at best shaky ground. It surely is not true that the majority is always right. And if morality has been exhausted when we fail to reach agreement, what moral force remains to persuade the minority to accept its status, particularly if it finds itself more or less a permanent minority, and particularly if it sees its basic values being threatened? And in a larger sense, even the existence of a majority is questionable, since any norm will be “agreed to” in quite varying degrees: only a few people support it wholeheartedly, others see it as acceptable but not idea. Others see it as bad but better than anarchy. Furthermore, it seems clear that there are many possible norms that could produce equally large (and equally divided) majorities. The mechanisms by which the proponents of potential norms identify them, advertise them, and attract support to them – these very mechanisms are subject to inequalities and injustice. Footnote

            And so on. My point is not that majority rule is uniquely bad but rather than once we no longer have the philosophical compass of universal, uncoerced agreement, we appear to be left directionless and adrift. And when we are in this situation, the temptation is to use force. By force I mean any means of getting one’s way, physical force being one means, but deception, game-playing, and the like being included as well. To put this in Habermas’s neat language, force includes all instrumental action.

            This is a rather detached and formal description of the sit6uation. Let me state it as one usually encounters it.

Those asshole Tweedledums are proposing to enforce on us, the Tweedledees, immoral and abusive practices. We’ve repeatedly tried to explain to them why their actions are wrong, pleaded with them, but they don’t listen. They are deaf to logic and blind to plain facts. The selfish bastards care only for their own prejudices and make fun of our concerns. The fact that they outvote us only shows that there happen to be more of them and that their leaders have blinded them to our concerns. Their position has no moral authority, only the force of coercion and strategic advantage, and we will resist it with whatever means we have until they at last see why we’re right.


It goes without saying that the Tweedledums describe the Tweedledees in equally dismissive language.

            I have written the above in language that may make it appear to be a caricature, not reality. But despite this tone, I am serious, because it describes pretty accurately my sense of the cultural conflict in the United States between liberals and conservatives or in Habermasian language (and more accurately) between the proponents of system rationalization and those of the meanings embodied in existing lifeworlds. In their quest to gain the advantages of certain system configurations, system rationalizers are willing to advocate ways of relating that override local, particularistic social-cultural-economic-political arrangements. Indeed, I exaggerate when I say they are “willing to” override local arrangements; my sense and personal experience is that they accord these local arrangements no value whatsoever, taking notice of them only as sources of (irrational) resistence. Thus the system-rationalizers-dominated media’s coverage of the WTO protests in Seattle portrayed them primarily in terms of anarchists and rioters, ignoring the many other groups who had simply come to protest what they anticipated to be WTO’s intent of bulldozing their ways of life through trade liberalization. I don’t want to make this sound overly conspiratorial, however, because there is a far simpler explanation for this pattern of coverage: that the people who own and manage the media cannot see the other protests as anything but anarchists and rioters – if not today, then tomorrow. From their point of view, after all, there is nothing that really separates the other protesters from the anarchists and rioters. Once one accepts the view that people can protect their social arrangements regardless of the consequences for the whole society, then anarchy and riot are the immediate logical consequence and therefore the eventual practical result. If Pamela Protester believes that her individuals beliefs entitle her to defy broader social choices, what is that but anarchy? We cannot make the law subject to people’s individual judgment (or communities’ local judgment) of whether they like it or not. So in the media’s eyes, according to my theory, their focus on anarchists and rioters is completely legitimate as a way of encapsulating [what they see as] the foundation of the protests as a whole. To me it is much more plausible to see the biased coverage as a result of a shared worldview than as a conspiracy. Viewing something as a conspiracy implies that the conspirators understand your views. In the ordinary course of life this is an easy conclusion to reach, given how obvious one feels one’s perspective is and how hard one has tried to communicate it. But however natural this inference may seem, it fails to take into account egocentrism – the egocentrism of the others, who cannot hear what you’re saying except through the filter of their own preconceptions, and one’s own egocentrism, which leads one to believe that others must do [sic] understand one’s perspective.

            I know I am afflicted with that egocentrism. I try consciously to resist it, but it is so automatic a response that I often discover, too late, that I have slipped into it. As a teacher, I am able to do this without my students calling me on it. They recognize it, of course, but most find it too difficult, or too unlikely of success, or too unimportant, to persist in challenging it. Footnote This egocentrism appears outside the classroom, too, because my various sources of status let me exercise it without being called on it. Status: I am white, tall, well-spoken, neatly dressed, and highly educated. Getting away with it: My paradigmatic case occurred in 1971 or 1972, when my (now ex-)wife and I were graduate students living in the attic of a home owned by Mr. & Mrs. Schmidt (not their real name). Mr. Schmidt had had to drop out of school early, and worked in a nearby factory as an unskilled laborer. One day as I started to vacuum our attic, I found this task quite difficult. Our vacuum’s short power cord would reach only about a quarter of the attic, so that I was going to be repeatedly required to shift it from one hard-to-reach socket to another. Mr. Efficiency had the brilliant idea of eliminating all this hooking and unhooking by using a 50' extension cord I happened to have in the car. However, when I attempted this, the vacuum wouldn’t run well. When I next encountered Mr. Schmidt, I complained to him that something was wrong with the power lines: there wasn’t enough power being delivered to the outlets. Footnote He told me that the extension cord was causing the problem, because the voltage drops the further away from the source one goes. I told him this was not so, knowing (or believing, anyway) that the copper wiring in the extension cord was virtually a perfect conductor and therefore could not account for such a marked reduction in power. After all, I had just graduated in applied mathematics from Brown University, with a minor in physics. I knew Faraday’s Laws, and Mr. Schmidt was a school dropout with the uneducateds’ superstitions about how electricity works. Footnote After a few “yes it’s the extension cord” - “no it isn’t” exchanges, Mr. Schmidt got angry and walked off. It was only years later that I learned that he was right and I was wrong.

            These little stories aren’t just meant to purge my conscience but to say in as immediate a way as I can how this egocentrism works. And I hope that they yield something more than the conclusion that I’m an ass. Even if that were true, as my friends assure me, I think that if you are honest with yourself, you will find much of yourself to recognize in my prejudices and experience. My point is not that you’re an ass too but rather that this is a very common, natural, and even in some ways necessary behavior. In the midst of life’s unscripted turbulence we must make thousands of judgments instantly, and preconceptions and assumptions are often all that we feel we have time for. This is not a state of affairs to be sought, and I am not recommending the use of prejudices when the opportunity exists for true knowledge, but it is often simply what we are faced with in the human condition. I will return to this topic of the human condition in the next section.

Perhaps an implicit or intuitive sense of limits, but nothing clear. Footnote


B.  Jumping to force

  Once one reaches the boundary of discourse ethics and its mutual respect, it is a very short step from there to the violence that can erupt when decisions need to be made without the validation of agreement. This was brought home to me at a panel I attended recently at a professional conference. The panel was concerned with understanding how political systems could be structured to prevent oppression. It could have been a panel on Marxism, or communitarianism, or discourse ethics, or classical liberalism, or libertarianism – no matter; all seek to end oppression, though they understand oppression in different ways and look to different sorts of remedies for it. So in this panel I watched a series of well-educated, well-dressed, well-spoken, articulate, good-looking academics talk about the need to respect, solicit, and take cognizance of all points of view. During the question period following the panel, I asked the panel generally what they thought should be done if, despite all their arguments, people didn’t agree with them. After some moments of dead silence, one of the panelists finally said that (paraphrasing) “they would have to be forced. I mean, we’ve given them every opportunity to understand our logic.” Even though I was expecting that general answer, I was taken aback by the baldness of the reply, but if the rest of those present reacted to it, it was only with nodding heads. The Asch effect overcame me, and the discussion moved on.

            But what is objectionable about that reply? Doesn’t it just state the reality of the situation? We (political theorists) spend a great time trying to find ways to be just to everyone. Surely the time to disagree with our conclusions is while we’re developing them, not afterwards? If we’re playing each other in a tennis tournament, isn’t it unfair to play a game and then, if one loses, to say that the game didn’t count? In any case, what’s the alternative to force? More discussion will likely result only in further disagreement, even if the original objections are addressed.

            I understand these objections, and I do have an answer to suggest, which is the point of this chapter. Before I present it, though, I want to motivate Footnote it so you can see what problem I’m actually trying to address. I am not trying to address the problem of finding a way to march inevitably to agreement; the Münchhausen trilemma says no such logically-inevitable agreement can exist, and our thus-mistaken preoccupation with it leads us to neglect what I see as the real problem. This “real problem” is the disrespectfulness with which we treat others when we grant ourselves the authority to force compliance. Suppose I were to post the following in newspapers around the country:

January 15, 2002



It has been shown that Chiltonianism, described in the book “Ways of Relating” (available at cost from the Chilton Institute) is Right. Members of the Chilton Institute, expert political theorists one and all, unanimously agree that its logic is flawless and that disagreement is merely evidence of ignorance or emotional disturbance. This is to notify you that we at the Chilton Institute will therefore be replacing the current U.S. political regime with Chiltonianism. You will cooperate, of course, since you will see that Chiltonianism is Right, but Chiltonians are morally entitled to use force to compel those who cannot understand its Rightness.




Steve Chilton, Maximum Leader


Would you rejoice that the perfect political system had at last been found? Even though you haven’t read the book yet and don’t know what Chiltonianism is, do you accept that its advocates have a right to compel your obedience in the event of your disagreement? You do want to be saved from your own ignorance and emotional disturbances, don’t you?

            O.k., so you wouldn’t rejoice. As a matter of fact, you’d probably think that Maximum Leader Chilton was a nut and that the Institute was a danger to the community, so that both should be suppressed immediately. And yet this is precisely the position the above-mentioned panelists are taking; I’ve merely made its features a little more prominent.

            “Yes, but the difference is that Maximum Leader Chilton is at the very least a dangerous nut and potentially an evil dictator, while we’re moral, considerate people.”

            Well, stated baldly in this way, it’s pretty apparent that this argument is simply dogmatic, not logically valid. “I’m right and you’re wrong” isn’t a very persuasive argument. Footnote Here’s a better response: “The difference is that Maximum Leader Chilton sounds like a Hitler or Stalin, who terrorized, exiled, imprisoned, or even killed dissidents. We, on the other hand, wouldn’t use these terrible methods.” And it is true that violence can have a variety of levels, from physical punishment and compulsion to mere propaganda. But once the barrier between respect and disrespect is broken, it is much less clear what actions can be morally justified. From a practical perspective, the milder the form of violence employed, the cheaper it is to deploy and the less resistance and reaction will arise from its use. Footnote But the question is not what level of violence most efficiently enforces or protects one’s concept of the Right; the question is, rather, how do we decide what level of violence is justifiable to achieve that end?

            I am not claiming that humans become animals once the moral certainty of agreement is left behind. Centuries of common and statutory law have given us some guidelines about the use of force; they represent what might be called our collective intuition about such issues. We recognize that the use of force is something to use carefully, not indiscriminately. Nevertheless, this still remains only an intuition, and a beleaguered, battered, and disputed one at that. I take heart from the existence of the intuition, but I believe that we need to clarify its nature. That’s what I’m attempting to do in this chapter, in Chapter 5, and really in this work as a whole. So to return to my original example, of my colleagues’ willingness to turn to force, my objection is not their contemplation of force itself but rather their treating it without any concern for its perils and for the moral responsibility its use must entail.


C.  The Prisoner’s Dilemma problem: Encouraging stubbornness and game-playing

This section of the essay has focused on what makes the agreement problem a “problem”. The final reason is the possibility that the requirement of agreement can be used strategically to negotiate a better deal for oneself. If agreement is indeed taken as the only criterion of morality, then one can withhold one’s agreement as a bargaining tactic. Even if one actually believes that the proposed norm is perfectly o.k., why not see if one can’t get a bit more by threatening to defect?

            The obvious reply – that if one person uses that tactic, so can everyone else, to the detriment of all – is not effective, at least on the face of it. Adherents of the rational choice model of social decisionmaking can simply point out that while this possibility of general defection is indeed a consideration, one that any rational actor needs to factor into h/her calculations, it is not a morally binding consideration; it imposes no duty on h/her specifically.

            A discourse-ethics rejoinder to that might be that the rational actor has things backwards: Habermas (19xx) has shown that this is a morally binding consideration, not just in some general sense but rather through a method that can prove it to each specific actor. Footnote “Perhaps so”, replies the rational actor, “but thus I refute discourse ethics” – acting as s/he pleases, regardless of what is supposedly moral. Footnote I will not pursue this issue farther here. It leads directly into the motivation problem and the meaning of “moral”. Habermas did address this issue, but not in a way that entirely solved the problem. I’ll address the issue more thoroughly later.


III    Existential Origins of the Agreement Problem

I will be arguing that the Agreement Problem cannot be philosophized away but rather comes from the nature of human existence itself, at least insofar as I am able to see its limits. Establishing this claim is a necessary preface to my subsequent discussion, which involves the nature of morality beyond agreement. If disagreement is not inevitable, then we would spend our time better in figuring out how to achieve it than in saying what we should do in its absence. But if, as I claim, disagreement is and always will be the usual result of moral discourse, then we need to spend our time looking at what to do then.

            The points below are quite brief; my purpose is to list and describe them, not to prove them. In my mind at least no proof is necessary for these obvious points.


A.  Variation in senses of the Good

The first problem is of course that we disagree on what we want to see happen. I want to see a country where everyone worships Gozer, while you want a country with freedom of religion. Disagreement is inevitable. One might settle the issue through some grudging compromise, but this would not be true agreement. It might be, for example, that our beliefs are so firmly rooted that any compromise would vanish at the first hint of a problem – sort of like the famous Molotov-von Ribbentrop non-aggression treaty between Stalin and Hitler, where each found it momentarily inconvenient to attack the other but fully intended to when prepared.

            Even a shared sense of what is desirable can create disagreement when there is a limited supply of it. Stranded on a desert island together, we might both desire that coconut, but that doesn’t mean I’ll let you have it or even agree to share it.


B.  Natural limits of empathy

Morality is a matter of empathy for the other, so that one takes h/her sense of TWIW as seriously as one’s own. Agreement might arise by accident even without such empathy, even with substantial misunderstanding of the other, but only in the sense that a clock that doesn’t run is still right twice a day. I therefore simply detail a couple of the reasons why empathy cannot ever be perfect.


1:  Historical/geographical/cultural

People live in different lifeworlds, with different background assumptions. One could argue that even the most intimate of friends still see the world in different ways, so that their empathy will always be imperfect, even if such imperfection comes to light only rarely. But regardless of whether you share that belief of mine, it seems plain to me that there the differences in lifeworlds and thus in empathy grow rapidly, the more two people are separated by culture, by geographical distance, and by their own personal histories. I mention geographical distance not to say that it creates different lifeworlds (although it seems likely to) but only to say that the less opportunity people have to interact with each other, the less able they will be to come to understand each other’s sense of the Good.


2:  Recognition of concrete but not abstract connections

I am jumping ahead a bit here, since I won’t be discussing this factor until Chapters 4 and 5, but the point is pretty simple. In a complex society like ours, my actions (and, more generally, my ways of relating) will have consequences for people I am not likely to meet. I can understand in the abstract that, say, dumping waste into the river here will mean problems for the people downstream, and so we pass antipollution laws to recognize that general understanding. But that doesn’t mean I can empathize with those others in any detail. To me, the cost savings of dumping my waste into the river might make the difference between survival and bankruptcy, something that seems to me obviously more important than inflicting a smelly river on others. But of course this likely misunderstands the problems downstream, which might be not the smell but (say) having their fishing industry destroyed. So while laws and general customs certainly try to institutionalize what would be our empathic concern for others (were we able to meet them), they are inevitably clumsy tools and cannot create the actual detailed knowledge necessary for real empathy.


C.  Decisions made willy-nilly

Even if none of the previous difficulties were present, finding agreement still takes time: time to develop or clarify our empathy for each other, time to work out the details of what seems right in consequence. Unfortunately, decisions often have to be reached willy-nilly: Footnote Failure to decide will be a decision in itself, one that may be worse for both of us than the adoption of either one’s preference. Thus the story of Buridan’s ass, referred to in Chapter 1. The ass must decide which bale to eat from; failing to decide means starvation. If we’re coming to a T intersection, and I say to go left and you say to go right, failure to decide means running right off the road. If Republicans and Democrats fail to agree on a budget bill, as happened in the Clinton administration, the government simply stops functioning, an outcome worse than either proposal. Footnote Endless such situations, both real and imagined, could be listed here.


D.  Distress patterns and painful emotion

Finally, there are simply situations in which people are emotionally unable to agree to any resolution other than their own, no matter how unreasonable their preferred outcome might be. For example, I have known several people, both men and women, who fought off any kind of physical contact – a hug, a pat on the back, even shaking hands – because they had been abused as children, physically and/or sexually. Any physical contact immediately brought up old, painful memories, so much so that they were unable to see the gesture in its present-time context. Determined, intentional therapeutic discourse over months or years changed this, but when a decision needs to be made more quickly than that, their position is unalterable. Footnote


IV    Summary

Universal, uncoerced agreement is rare even under the best of circumstances. This means that while we can aspire to agreement, we rarely realize that happy state. Lacking morality’s gold standard of agreement, we find ourselves in a moral never-never land, where we are guided only by the rather uncertain compass of our intuitions. However, intuition lacks the discipline that actual discourse provides, allowing all sorts of extra-moral forces to push our intuitions this way and that. We cannot be certain – indeed, we have good reason to doubt – that a group’s collection of intuitions will even approximate acceptable moral norms.

            If intuition were all that was left us, we could reconcile ourselves to this uncertainty; we might even console ourselves that intuition can still produce good judgments, even if we can’t defend them. I believe, however, that we have not yet reached the true frontier of publicly justifiable moral considerations. With the understanding of morality’s ground presented in Chapter 2, I try in the rest of this chapter to describe moral considerations that lie beyond agreement.