Monday, February 15, 2010

Asserting and endorsing

So, I've been puzzling for a long time over the existence of lying-like phenomena. For instance, I set up an automated website that gives false information about the weather, presenting it as true. When the website falsely says: "The current temperature in Waco is -2oC", most people will, I think, say that I haven't, strictly asserted this. After all, I might even be dead by this point. However, what I did is relevantly like lying.

Consider also a continuum of cases. I sign a letter to a third party. I know the letter contains false central claims. But there is a continum here based on how much I've contributed to the letter and how aware I am of its contents. As long as I know that the letter contains false central claims, it seems to matter little morally whether I am the sole author of the letter or have signed it without knowing any particulars about the contents—in both cases, I am dishonest in a way equivalent to lying. For purposes of the morality of honesty in communication it should not be necessary to draw the line between fully asserting and signing.

A friend suggested to me that one talk of a "constructive assertion", where "constructive" is a legal alienans adjective: a "constructive A" is not an A, but has the same normative consequences of Aing. But I think there is a simpler solution. We simply talk of the endorsement of propositions. It is wrong to endorse when one knows one is endorsing a falsehood. (That's the easy case. It may be tricky to come up with a rule that handles cases where you do not have full knowledge.)

"Endorsing propositions" generalizes "asserting propositions". For to assert a proposition is simply to present a proposition and endorse it. Actually, it may be that every full act of endorsement includes a presentation as well. After all, one's act of endorsement has to somehow indicate what proposition one is endorsing, and one has to do that by means of some linguistic or quasi-linguistic convention, such as setting down one's after a written expression of the proposition or nodding after someone else has expressed the proposition orally. If one takes this wide view of presenting a proposition, then one does in fact assert what the website says and what the letter says, even when one does not know the content of either. If one does not like this conclusion, one can give a narrow account of presenting.

A consideration in favor of the wider view is this. Suppose you assert that p, and I say: "Yes." It seems fairly reasonable to say that I've asserted that p by means of a prosentence. But there will be cases when I say it without having paid any attention to what you said, and I agree out of cowardice or trust. If one thinks that I don't assert in the website case, one should say that I don't assert in this case. But I think in this case I do assert. For take the case where I say: "Yes, p." In that case, it sure seems that I have made an assertion, even if I repeat "p" completely mechanically and absentmindedly. If so, then likewise a "Yes" without the "p" part should be an assertion. And the only reasonable thing for it to be an assertion of is that p. (One might say that "Yes" is an assertion of "I agree with what you've just said" or "What you've just said is true." But that's probably not correct, because then the "Yes" has additional entailments that the proposition that p did not have: it entails that you spoke and that I exist.)

In any case, even if one not convinced by my suggestion that all endorsement involves assertion (or re-assertion), I think it is correct to say that the general phenomenon for the morality of lying to discuss is endorsement.


Heath White said...

I think this is helpful, and one place to look for particulars is an account of the speech-act of assertion. For example, Brandom says that to assert is to undertake a responsibility to justify a particular content. (viz. the content you just asserted.) Consequently, if you assert a content which you know you cannot justify, or have no intention of justifying, or can justify only in underhanded ways, etc., is to undertake a task which you cannot well perform. It's analogous to making a false promise.

I don't think you have to take Brandom's account of assertion as final, but the point is that the speech act of assertion ought to be defined in terms of its normative significance, and then that significance can give you hints about the morality of lying and close relatives.

Alexander R Pruss said...

Right: assertion = expression (content) + endorsement (illocutionary force). Then one can give an account of the endorsement as an undertaking of some sort. If the Brandom account is right, then when I endorse a co-signed letter, I undertake a responsibility to justify its content (e.g., by justifying the authoritativeness of the primary author).

My own preference, however, would be to start with the moral side, where I have clearer intuitions, and then go from there to the normative significance of endorsement.

Here's an off-hand worry about the Brandom suggestion. Doesn't justifying a content involve making further assertions ("p because q")? If so, then while Brandom's claim may be true, it doesn't seem to be a full account of the normative content of an assertion. (Unless one Ramseyfies?)

The second worry is this. Suppose I want to take on a theology teaching job at a religious school whose doctrines I am agnostic about. I promise the school to justify the doctrines to all and sundry, and especially the students. Thus, I am committed, whenever it's relevant, to give the best arguments for p. And I can do it as well as anybody who actually believes p. My promise is, then, the undertaking of the responsibility--but it's not an assertion or endorsement.

I think this is a general problem with normative accounts of things. Such accounts can be illuminating, but they are apt to fail to get at the heart of the matter, because isomorphs of the same normative structure of responsibilities and permissions can be produced with a different kind of normativity. Imagine a variant on chess, legal-chess, where the normative structure is exactly the same, except that the rules are enshrined in criminal law. Thus, to move a pawn three spaces is a crime. Merely to specify chess by giving the rules is to leave it indeterminate whether we're dealing with chess or legal-chess. The same applies in the case of linguistic things. Imagine a world where God explicitly commanded those who have expressed p in a certain tone of voice to justify p. Expressing p in that tone of voice then does undertake a responsibility to justify the content, but it does not do so according to assertive (or endorsive) normativity, but according to divine-command normativity. (If you think assertive normativity is divine-command normativity, just vary the example--maybe it's the criminal law that commands this.)

(The point remains even if all normativity falls under one general kind--I myself think all normativity is moral--becuase there will still be subdivisions within that kind.)

Heath White said...


On your first worry: Brandom’s view of the content of p is that it is the network of other contents which are reasons for it and which it is a reason for. So “assertion” doesn’t show up in the analysis, but “content” does. And yes he would just embrace the circularity involved in defining content.

On the second: I have wondered about this myself, though not exactly in the ways you mention, which are helpful. I’ve thought about contexts in which people are expected to justify assertions without necessarily believing them. E.g.: Presidential press secretaries at the podium; spokesmen for military or corporate operations; your agnostic religion teacher; think tank intellectuals who are paid to come up with arguments for donors positions; etc. (Most of these are morally problematic; they have connections with Frankfurt’s BSers.) I’ve thought about two different kinds of solutions.

The first is to appeal to motive: you undertake to justify a claim, with the motive of getting at the truth. As opposed to: with the motive of defending your superior’s position, or whatever. The second is to appeal to context: you undertake to justify a claim in any context whatsoever. As opposed to: when at the podium, when in the classroom, in public, etc.

Your idea, that the kind of norm involved makes a difference, is one I will have to give further thought to. (And it may be that these are all variations on a single theme.)

Alexander R Pruss said...

The truth-directedness suggestion is helpful. The idea here might be that norms have a point in addition to the rule.

Maybe a problem with my religion teacher is this. She either thinks that the doctrines are justifiable or not. If she thinks they're justifiable, then she should believe them. if she thinks they're unjustifiable, then she shouldn't play the game of justifying them.

(Is it a responsibility to actually justify, or to do the best one can? Surely the former. For if the latter, then when I agree to teach on Leibniz and Spinoza, I assert their doctrines, as I undertake to do the best that can be done to justify their doctrines.)

Heath White said...

Another thing, which I don't exactly know how to work in: all the quasi-assertions are parasitic on real assertions. That is, there would be no justifications for your religion teacher to offer, if someone did not believe them and come up with those justifications. Or at the very least, the *possibility* of genuine assertions is presupposed. I'm not sure how much ice that cuts.

Alexander R Pruss said...

You could come up with justifications for a proposition nobody's ever held, though, right?

Heath White said...

Yes, I guess you could. (This might be the basis of a certain kind of law school training. "Defend the proposition that Heath White is a kangaroo. Go!")

enigMan said...

Surely I could, without doing anything like lying, assert a proposition which I do not endorse. I could be teaching, for example, or passing on a message. What I would endorse would be, not so much the proposition as its assertion. Even so, I would be presenting it assertively. Now, I might be doing something like lying if I thought that the lesson's facts were false, or if the message had been designed to be deceptive. But I might be quite indifferent about such details, and such might not be the case. And I might trust, not that such was not the case, but that if it was it was as well that it was.

enigMan said...

(upon reflection, I'm not so sure:)