A recent issue of the New Yorker has Malcolm Gladwell’s discussion of the recent book Why? by Columbia sociologist Charles Tilly. Since I’ve not read the book, I can only comment based on the discussion, but Tilly’s book might have implications for how we understand the actual process of exchanging reasons. Tilly argues that when we express reasons for action, either to recommend courses of action to others or to justify our actions to them, these reasons will fall into four general categories:
1. CONVENTIONS: An example might be "no one likes a sore loser."
2. NARRATIVES: Stories "circumscribe time and space, limit the number of actors and actions,
situate all causes ‘in the consciousness of the actors,’ and elevate
the personal over the institutional."
3. CODES: Similar to conventions, but that invoke procedural rules and categories: "If a loan officer turns you
down for a mortgage, the reason he gives has to do with your inability
to conform to a prescribed standard of creditworthiness."
4. TECHNICAL ACCOUNTS: These are explanations "informed by specialized knowledge and
authority" that aim to be exhaustively detailed.
(Gladwell — and maybe Tilly hmself — are not too careful with the justificatory/explanatory distinction among reasons, but it seems like the claims apply to the justificatory kind.)
Tilly suggests that none of these reason kinds is the best; instead, reasons are more or less suited to play different roles, depending on the relationship that the person describing the reason has to her audience and others involved. So here’s Gladwell’s example of how different reasons could be used to discuss one and the same phenomena:
"Here are four kinds of reasons, all relational in nature. If you like
Cheney and are eager to relieve him of responsibility, you want the
disengagement offered by a convention. For a beleaguered P.R. agent,
the first line of defense in any burgeoning scandal is, inevitably, There is no story here.
When, in Cheney’s case, this failed, the Vice-President had to convey
his concern and regret while not admitting that he had done anything
procedurally wrong. Only a story can accomplish that. Anything else—to
shrug and say that accidents happen, for instance—would have been
perceived as unpardonably callous. Cheney’s critics, for their part,
wanted the finality and precision of a code: he acted improperly. And
hunting experts wanted to display their authority and educate the
public about how to hunt safely, so they retold the story of Cheney’s
accident with the benefit of their specialized knowledge."
In sum, "effective reason-giving, then, involves matching the kind of reason we
give to the particular role that we happen to be playing at the time a
reason is necessary."
I think there’s a good deal for the moral philosopher here. Just a few issues that came to my mind:
One way to understand disagreement about normative moral theories is to see them as disagreements about the weight, relative or absolute, to be assigned to various reasons: The Kantian and the utilitarian subscribe to different "codes" in Tilly’s sense, whereas perhaps some feminists, virtue ethicists, and care ethicists may favor narrative as the principal kind of moral reason. There might be an intriguing research project to investigate which of these reasons people provide when they engage in moral reasoning or dialogue (and perhaps which, if any, they recognize as explicitly moral reasons.) This work might also shed light on what a rational moral agent is: Is it someone who is responsive to certain of these reason kinds, or is the rational moral agent the one who understands that no one of these kinds says all we need to say about moral reasons?
I wonder what basis Tilly gives for dividing reasons for action into these particular categories? My immediate reaction (surely colored by my philosophical background) is that it is a rather strange division in a number of ways. Where is a reason like “I just wanted to” supposed to fit in here? Is there a convention that tells us that wanting to do something is (or can be) a good reason to do it? Of course, I must admit that it’s not clear to me how conventions are supposed to work. The example offered – “no one likes a sore loser” – doesn’t say anything about reasons for action at all. I suppose we are assuming that people want to be liked – but then isn’t that the real source of the reason?
What unites true claims about reasons for action, I think, is that all of them link with some sort of good that can be accomplished or promoted by the action for which, it is claimed, there is a reason. But the notion of goodness doesn’t seem to appear at all in the four-category division Tilly proposes—not even in an indirect or submerged manner.
Moreover, the Cheney example suggests that what Tilly is really interested in is not so much (normative) reasons as explanations or, perhaps better still, exonerations. It might be true that, relative to a certain purpose (of leaving a certain audience with a certain impression), certain claims about one’s reason for action will be preferable to others. It does not in any way follow that the truth about one’s reasons for action will be relative in this way. Moreover, the idea that reasons for action are relative to one’s audience seems to leave us in the dark with respect to first-person practical deliberation. It might be that, when trying to impress an audience in some way, I want to choose between various reason statements with that end in mine; but surely when I am on my own deciding which action would be best, what I want to know is what my reasons really are, and which are strongest – and there is no relativity here.
It seems, perhaps, that what Tilly is really writing about is not reasons, but rather statements about reasons. And of course it’s true that different statements about one’s reasons (including, sometimes, false ones) will serve different purposes. Is that interesting? I wonder. But of course, like Michael I haven’t read the book; and unlike him, I haven’t even read the Gladwell piece! So my concerns might be quite misdirected, when all is said and done.
For what it is worth, the distinctions mentioned strike me as vague and arbitrary, but I bet they could be improved upon if you drew from Von Wright’s “Varieties of Goodness”. He clearly and consistently distinguishes forms of goodness, e.g. techinical, medical, utilitatian, substantive, hedonic, welfare, and his distinctions do seem to track different types of reasons. He also connects this discussion to debates about moral theory and what we would now think of as theories of rationality.
Even if the buckpassing account of good is wrongheaded, that does not get in the way of the comparison.
I think that its somewhat important to recognize Tilly’s credentials, and thus his theoretical perspective, is predominantly (for obvious reasons) sociological. The point is that an account of reason-types such as this with an eye towards their effectiveness doesn’t make any claims about the normative status of” kinds” of reasons (and this is why these types of reasons cannot be classified into any kind of hierarchy).
It is, I think, important to bear in mind the distinction between such types of reasons, with “moral reason(ing)” (however we may understand it) as such which grounds our assent to the reasons-offered (of whatever type they may be). Thus one would say, to borrow Michael’s elucidation, that certainly a Kantian or a utilitarian may express itself in terms of a “code,” while other groups may prefer other rhetorical tools, but each side would (presumably) be just as comfortable if their “principles” were expresed in any rhetorical form, but for purposes of efficacy, may prefer one mode of expression over others.
On the other hand, what I think may actually be quite interesting is whether, in analyzing the contextual efficacy of these various reason-types, we may discover the ways in which a single reason (abstractly considered from its rhetorical dress) may drive us to conflicting judgments when set in different reason-types. For instance, we may hold a particular reason to be perfectly valid as a moral norm when dressed as a “code,” but when we see the code “in action” in a narrative, we may come to a very different conclusion. And far from necessarily implying any kind of rhetorical hierarchy (which Tilly quite rightly seems to reject), this process can quite easily work in reverse, where the narrative motivates assent, while the “code” expressed therin seems objectionable.
Going at it in this way, it seems that the rhetorical analysis of reason-giving strategies might open up interesting perspectives for the normative evaluation of reasons in general. For instance, can we really say that the reason is normatively valid only if it meets with our satisfaction no matter how it is expressed (i.e. if it can be expressed in any of the four types – or in whatever other classification one may choose) in which case normative validity carries strongly universalistic implications, or do we have to accept a more “contextualized” notion of normativity itself, without necessarily laying claim to any kind of over-arching normative content to unify them all?
Corwin and Troy,
Yes, it doesn’t seem like Tilly is trying to give us an ontology of reasons in the philosopher’s sense, but attempting to analyze reason-giving rhetoric, and it strikes me that what he has to say about how his reason-types vary in their effectiveness and in how they are used in different relationships might tell us something about how people *actually* engage in moral reasoning. Philosophers tend (I think) to suppose that moral reasons are determined by some combination of facts about the world plus some normative element (one’s desires, a moral theory that tells us why those facts are morally relevant, etc.) But I wonder how effective reasons so depicted really are in shaping actual people’s moral beliefs. It seems to me that a good deal of actual moral reasoning looks more like narrative than the provision of a code or theory deductively applied to facts. Also, is it plausible to think that any one reason is expressible (without loss of of its persuasive force) in all four of these modes? In a larger sense, I’m wondering what the practice of reasong giving in actual discourse tells us about the topography of reasons generally.