Wednesday, August 22, 2012

Civility in Argument

There is a very interesting but, I think, very incorrect article on civility in argument by Scott Aikin and Robert Talisse at "3quarksdaily". The essential line of argument is essentially:

(1) Civility in argument is not conflict aversion.
(2) Civility in argument is not politeness or calmness of tone.
(3) "Argumentation is the process of articulating our reasons for holding our beliefs."
(4) Civility in argument is that set of tendencies enabling exchange of reasons among disputants.
(5) Civility in argument has three dimensions: Representation (actually engaging with each other's reasons); Reception (giving a proper hearing to one's opponent's reasons); and Reciprocity (presenting reasons that one's opponents can see the relevance of).
(6) Therefore civility in argument does not have to do with "being nice, calm, or even polite" but with being a sincere arguer.

In fact, I think it's clear that every single point here is wrong, although in general this is due to dropping qualifications rather than from any egregious error. It is true that civility in argument is not only conflict aversion and that it is not only about tone; it is true that one of the ends of argumentation is articulation of reasons for holding one's beliefs; it is true that one of the reasons for civility in argument is to enable exchange of reasons; Representation, Reception, and Reception are some of the things that are sometimes relevant to civility in argument; and sincerity in argument is conducive under some conditions to civility in argument. But the qualifications are all extraordinarily important. It's also the case that the account given of Reciprocity is inconsistent with the account given of argumentation; Reciprocity as Aikin and Talisse describe it cannot be a general dimension of civility in argument given how they describe argumentation, and argumentation as Aikin and Talisse describe it can be severely interfered with by Reciprocity as they describe it. In fact, it looks very much like they've rigged their account of Reciprocity in order to get a particular conclusion about democracy that they think good.

It is a mistake, I think, to think there is some special thing going under the name 'civility in argument'; rather there's just civility, and it can be relevant to argument, and when we look at cases where it is relevant to argument we find common patterns. One of the interesting failures of the argument of Aikin and Talisse, for instance, is in its lack of any good explanation for why we would call civility in argument 'civility' at all. Civility is the practice of maintaining civic good relations, where civic good relations are those relations that make society 'work' in some sense. (The old Aristotelian name for them would be civic friendships.) Society is a way in which we live together, and civic good relations are those relations among the people in society which make living together possible and viable, either because they are pleasant, or because they are useful, or because they are good for us. In this sense you can have civic good relations with people you don't really like at all; in such cases you simply recognize that, whatever problems you have with them, you have to live with them to achieve certain good things, and you act accordingly.

Civility in argument, then, involve approaching argument in such a way as to maintain good civic relations -- that's the only reason to call it civility in the first place. Thus conflict aversion is, right then and there, part of what civility in argument is about. Aikin and Talisse are right to the extent that civility is not just about conflict aversion -- it requires positive practices in addition to avoiding conflict -- and and it is not about just any kind of conflict aversion -- it is about avoiding conflicts detrimental to civic good relations. But dropping conflict aversion entirely, as they do, is an extreme move. There are also problems with their argument for it. They say:

On some accounts, civility is equivalent to conflict aversion; one is civil insofar as one is conciliatory and irenic in dealing with one’s political opponents. Civility in this sense seeks to deal with disagreement by disposing of it. Civility of this kind is little more than a call for compromise at the expense of one’s own commitments. Hence this kind of civility might be inconsistent with actually believing anything.

This argument seems to consist entirely of leaps, however. Being conciliatory and irenic in dealing with one's opponents is not the same as dealing with disagreement by disposing of it; in fact, it may sometimes be quite the opposite -- it's consistent with letting a disagreement go on longer than it has to by not trying to force the matter here and now, or with taking the time to argue in part on your opponent's own principles, just for the sake of argument, not because you are disposing of the disagreement but because you are prioritizing sub-disagreements and thus only focusing on the most essential matters. Likewise, neither being conciliatory nor disposing of disagreements is the same as compromise at the expense of one's own commitments. Obviously it will depend on your commitments, since it may well be that the disagreement in question is genuinely not very important, or that a live-and-let-live approach will allow you to be both conciliatory and to dispose of the disagreement by rendering the whole point moot. And even if it did involve a call for compromise at the expense of one's own commitments, it wouldn't follow that it could be inconsistent with actually believing anything; it would be consistent with believing in things consistent with such compromises, for instance.

Likewise, we can see that civility in argument can't just be about tone because civility itself isn't just about tone. And it's true that tone might sometimes not be even relevant, precisely because tone is sometimes not the big issue for civility itself. But it does not follow that tone is irrelevant. Maintaining civic good relations requires approaching people in particular ways, and it is a simple fact of human nature that, where there is a choice of relevant tones available, some tones will be more conducive to this than others. As the saying goes, you catch more flies with honey than with vinegar. Further, a lot of practices that are themselves designed in part to maintain civility -- practices of professionalism, for instance -- themselves require that tones be kept within a certain range. A professor being nasty in an argument with a graduate student is being unprofessional and uncivil in argument, and it simply does not matter how sincere, or how well-reasoned the professor's argument is: the professor has a special professional involvement in the society between professors and graduate students. A large number of circumstantial factors, including the power difference between professors and graduate students, the effect of professor-student interaction not just on the two parties but on the whole department, and others, converge to make it absolutely essential to professors to avoid using hostile and antagonistic language toward graduate students to the extent humanly possible. Professors who speak nastily to graduate students are abusing their position, and harming civic good relations in their profession. Whether and to what degree tone is relevant will depend on circumstances; but while it sometimes will not be important, it will sometimes be essential.

As with their rejection of conflict aversion, there are problems with the argument given for this point. They say:

Argument is a form of confrontation, one with words instead of weapons, and any norm that prevents argument from displaying the critical edges of disagreements undercuts what inspires the argument to begin with. Furthermore, it is possible to fail at proper argumentation and yet maintain a calm and respectful tone of voice. In fact, under certain circumstances, one patronizes one’s interlocutor precisely by sustaining one’s composure.

This argument, however, involves a completely different account of argumentation from the one they will later go on to give; if argumentation is a process of articulating reasons for holding one's beliefs, then its most common and natural outcome will not be arguments in the sense of disagreements and confrontations, but arguments simply in the sense conclusions-and-reasons-for-them. And even if we restrict our considerations to cases where we are presenting our conclusions-and-reasons to someone, and to actual arguments, not all arguments are disagreements -- it is possible to play devil's advocate or just argue just in order to be arguing (for fun, perhaps), for instance. And even if we restrict our consideration to disagreements, not all disagreements are confrontations, nor are all disagreements (or, for that matter, confrontations) inspired by things that would be undercut by anything preventing argument "from displaying the critical edges of disagreements". Even more than this, there is nothing about restrictions of tone per se that eliminate confrontations or disagreements, nor do they necessarily prevent argument "from displaying the critical edges of disagreements". And the suggestion that you can patronize someone by sustaining one's composure borders on complete nonsense: there are lots of different ways of sustaining one's composure, and patronizing people will always at the very least require more than merely sustaining one's composure.

Argumentation is certainly about articulation of reasons, and can certainly be for articulating reasons for beliefs you hold, but argumentation as such has nothing to do with beliefs. In many ways, in fact, the most interesting kind of argumentation occurs when people are not articulating reasons for beliefs they hold but for positions that none of them hold but that they find intriguing; and part of the reason this kind of argumentation can be so interesting is that it can lead to new positions that were simply not on the table to be believed or not in the first place -- things people hadn't even had a chance to think of before. Likewise, Aikin and Talisse claim that the point of articulating reasons is "to put them on display so that they may be examined and evaluated"; but this is certianly not true. The point of articulating reasons is to have available reasons. It may be true that the point of having available reasons is so that they may be examined and evaluated, but this is not at all the only reason why one would articulate your reasons for your beliefs. You might do it because you think it's important to have reasons for beliefs. You might do it just from habit. You might do it just to let people know what they are; sometimes people argue not to evaluate reasons but simply to clarify what they are in the first place. This actually is a matter of some importance. Aikin and Talisse claim, "When we argue specifically in response to disagreement, we supply our reasons for the purpose of demonstrating to our interlocutor their strength, and the comparative weakness of the reasons that support opposing views." But this, while very often true, is by no means universal. When we argue specifically in response to disagreement, we may not be trying to demonstrate the strengths of our reasons and the comparative weakness of opposing reasons, we may be simply trying to show that we have relevant reasons at all, that in taking the position we are, we are not being arbitrary or stupid. Thus its simply false to say that argumentation has within it the idea that you should believe only what the strongest reasons support; this may well be true, but it has nothing to do with argumentation, which isn't necessarily about what one believes, and even when it is, it isn't necessarily about giving the strongest reasons. Likewise, they are simply wrong to say that if argumentation lacks the background commitment to possibly revising one's views, it is pointless. Such a background commitment may often be very important to broader considerations of rationality, but argumentation is quite capable of having many points. In fact, such a claim doesn't even follow from the account of argumentation given by Aikin and Talisse: in their account it is reasons, not beliefs, that are put forward to be evaluated, and thus argumentation will still have a point if one intends to keep one's beliefs but wants to have the strongest reasons for them.

Given this we can see that many tendencies enabling exchange of reasons among disputants will be relevant to civility much of the time, but that, depending on circumstances, some of those tendencies won't always be relevant to civility. An interesting question is whether all civility in argument requires tendencies enabling exchange of reasons among disputants. This will obviously usually be the case. And I think there's a strong argument that civic good relations require the general possibility of exchange of reasons among disputants. I'm not convinced, though, that civility in a particular argument will always require tendencies enabling the exchange of reasons among disputants in that particular argument. There is, I think, a great deal to be said for the idea that sometimes civility in argument requires giving the argument a rest for a while. I think even more than this, there are situations where, due to some asymmetry, exchanges of reasons may not even be the point of argumentation. In philosophy conferences, for instance, one may argue on a new topic with new arguments in your paper; but while you are communicating reasons in a philosophy paper, you are not exchanging reasons unless you make the paper part of an exchange of some sort; if you are the first person you know ever to have talked about a subject, you are just giving your arguments. Philosophy conferences typically have question-and-answer right after the paper, and in those cases there will sometimes be an exchange of reasons (although many questions in a Q&A are purely clarificatory). There are reasons why these Q&A sessions are added, and they all have to do with the fact that delivering a paper, considered in itself, is an asymmetric situation: you are simply telling people what your argument is. This doesn't mean there's no argumentation; it doesn't even necessarily mean there is no exchange of reasons (people usually craft philosophy papers in order to take part in exchanges of reasons); but it does mean that there might not be any exchange of reasons at all, since it might be completely one-sided argumentation, and for good reason.

It follows from this that none of the three dimensions of civility in argument mentioned by Aikin and Talisse are either necessary or sufficient. I do agree that they are commonly needed for civility in argument. But one can be civil in argument without them, depending on the circumstances, and civility in argument will often require other things.

Of the three dimensions Aikin and Talisse note, I don't have much to say about Representation and Reception. I probably wouldn't have much to say about Reciprocity, either, except that Aikin and Talisse, by their own argument, shouldn't be including it. They had originally said that argumentation was a matter of articulating reasons for one's beliefs, and that one does so in order to test and evaluate them. Reciprocity as they gloss it, however, requires that sometimes one not be allowed to articulate one's reasons for one's beliefs or put them forward to be examined and evaluated; Reciprocity is therefore not a tendency enabling exchange of reasons among disputants and cannot be a form of civility in argument on their account. (I suspect that Aikin and Talisse insist on Reciprocity for reasons of Rawlsian liberalism, or something similar; but Rawlsian liberalism is inherently inconsistent with the idea that public argumentation is about articulating reasons for one's beliefs or necessarily tied to testing the quality of one's reasons. It's consistent with it sometimes being so, of course; but the thesis of public reasons requires that public argumentation not be primarily about reasons for belief, but at least primarily about (1) reasons for action that (2) will be accepted by other people as being of at least the right kind.) Since argumentation in fact does not necessarily involve articulating one's beliefs, or putting forward such articulations to test one's reasons, this means that it is in fact true that Reciprocity is a common pattern in civil argumentation. But since Aikin and Talisse's problem is not being egregiously wrong but mostly treating things that are often true as if they were always true, it's clear that Reciprocity will sometimes not be required for civility in argument.

And this actually makes a lot of sense in general. One thing that Aikin and Talisse don't seem to grasp in their discussion of Reciprocity is that how one answers the question, "Is this reason relevant to that position?" can be heavily path-dependent. But in fact it is so: A may be relevant to B in a non-obvious way that can only be seen if one accepts C, for instance. Because of this people can disagree about whether reasons are actually relevant, due to larger background disagreements, and merely because other people don't think a reason relevant doesn't mean it isn't. And while there are cases where the reasons for disagreement are known by everyone, there will be many cases in which it will simply be up in the air whether the reasons in question maintain Reciprocity. This, so far from being a problem, is often not even remotely important; argument goes on without it. We've already seen that argumentation does not necessarily involve an exchange of reasons, but even when it involves an exchange of reasons, when you are giving reasons to each other, it's sometimes just not important whether the other person thinks they are good reasons.

Much of the argument given by Aikin and Talisse seems to be geared not to understanding civility in argument as such but to defending a particular conception of democracy. "Democratic politics," they say, "is all about argument." This is manifestly not true about actual democratic politics. I don't think it can even be defended in this unqualified way for most democratic politics in an ideal democracy. Democratic politics will sometimes be about avoiding arguments in the first place, for instance; and lots of politics of any kind has nothing whatsoever to do with argument. That arguments are important to democratic politics is certainly true, but I'm not, for instance, really sure that the success of democratic politics "depends upon our ability as a citizenry to reliably make the distinction between argument and sophistry"; I simply don't think it's that fragile. There are other desirable things in politics that do depend on the citizenry being able to make such a distinction reliably, but democratic politics itself doesn't depend on it. All the mechanisms of democratic politics depend for their success on reliable venues for rhetorical communication; good rhetorical communication will often require argument. But you can have a pretty successful democratic politics based less on argument than on human sympathy, and one finds in fact in democratic politics that resolution of serious disputes is very often achieved not by argument but by compromise or by mutual sympathy or by a live-and-let-live approach. We should certainly be thankful for that; otherwise our societies would have failed long ago.