« Promises and Lies | Main | Ten Lessons of ObamaCare »

Friday, November 08, 2013


Feed You can follow this conversation by subscribing to the comment feed for this post.

Would another option be to say that what he is doing is not saying anything but merely expressing his contempt or defiance? It is as if Larry is giving the audience the middle finger with words.

To give credit where it's due, that counterexample is, if I recall correctly, a variation of an example Jeff Brower gave in an unpublished paper on "Aquinas on Lying" that I read years ago (Aquinas, if my recollection continues to hold, averred the pretty radical view (a) that a lie is an intentionally untruthful declaration to another [full-stop] and (b) that all lying is morally defective).

To Daniel: I can imagine situations where that sort of expressivist view would accurately describe the situation, but I can also imagine a case where it does not, namely, one in which Larry wishes to express his contempt ,by means of asserting something he knows to be false and knows everyone else knows to be false (without also intending to deceive anyone into believing it's true).

To The Maverick: I think (C) is promising.
What you say sounds similar (and perhaps is extensionally equivalent) to Aristotle's pros hen and Boethius's ,ab uno and ad unum relations between one or more terms with another term that has the central or focal meaning (though I've never really thought about how this relates to Wittgenstein and family resemblances.) A well worn example: the sense in which exercise is "healthy" is different from the sense in which a diet is "healthy," but both are properly predicated of their subjects insofar as they are related in the right way to the central sense of "health," namely the health of an organism.


Please do not use HTML codes unless you know how to turn them off using a forward slash. If you are unsure how to italicize and bold use asterisks.


I thought of that too, but Larry is also lying.


Talk of analogy seems fruitful. The peripheral cases are analogous to the central or paradigm cases. Would you agree that a paradigm case of lying must involve an intention to deceive?

Why are we so attached to the Vallicellian definition of a lie? Ordinary language? How many counterexamples or non-paradigm examples would be sufficient for abandoning it?

Dr. Vallicella, how do you handle the half-truth and equivocal truth counterexamples discussed in the "More on Lying" comment thread?

"Would you agree that a paradigm case of lying must involve an intention to deceive?"

Yes, that seems to me to be the paradigm case.

I think it is interesting, though, that in instances like the Larry-court-case it seems pretty obvious that THIS TOO is a lie. But why is it obvious if there is no intention to deceive? I think it might be because in the Larry-court-cased (a) the liar believes that what he is saying is false, (b) the motivation is deviant, and (c) the paradigm cases of lying exemplify (a) and (b) (as well as an intention to deceive).

I partially agree with Daniel K. The Larry case is not a case of lying; rather it seems to me that it is a case of stating a known falsehood as an act of defiance and contempt. The partial agreement is because I think Larry did *say* something: namely "I did not do it", thereby expressing contempt.

I, therefore, prefer option A. It has the advantage of preserving our central intuitions that lying requires an intention to deceive and at the same time we can still explain Larry's case as a case of stating a falsehood with the intention to express contempt.

I reiterate my opposition to Bill's view that lying requires stating a falsehood; such a view is vulnerable to obvious counterexamples. It is better to require that a lie is intentionally misrepresenting what one believes to be the truth with the intention to lie (along Davidson's proposal). I have yet to see a proposed counterexample to this definition.

The last phrase in my previous post should obviously be "with the intention to deceive."


Do you have a Davidson reference for me? The problem with the Davidson proposal, assuming it is really his, is one that I mentioned before: it implies that there are true lies. What you call counterexamples to my definition are merely examples of a different concept of lying bearing a strong family resemblance to my concept that captures the paradigm cases. We have to give up the idea that there is some one essence of lying that can be captured in a definition. What we are learning from this discussion is that there isn't.

As for Borland's Larry example, surely what Larry is doing can correctly be called lying in English. Same with the Clinton case. He lied even though he fooled no one. It is just that this use of 'lie' diverges from the paradigmatic uses which do involve intent to deceive.


As I said in an earlier thread, a separate post is need to discuss partial truth. That gets us into the question as to what the vehicles of lying are. A single noncompound statement is either true or false.


I do not recall where Davidson made the said suggestion. I will have to look it up.

I am not ready as yet to give up on the idea that lying has a viable core definition for the following reasons.

First, I consider the intention to deceive a necessary condition. This intention, however, is possible only if the agent believes that a deception is possible. Since in the Larry case such a belief is absent, it is not a counterexample. Moreover, since we can legitimately construe the Larry case as defiance and contempt, rather than lying, it is preferable to do so. Thus, there are reasons against viewing the Larry case as lying and there are reasons to view it as a case of defiance and contempt that fit the facts. I cannot see why not do so. The Clinton's case is a red herring. Clinton most likely believed that he could in fact deceive the public and thus minimize the damage. There are countless such cases when it comes to politicians. Their first instinct is always to deny the transgression hoping that it will go away.

Second, Davidson's proposal does in fact entail that there can be true lies. But this should not be surprising. After all (as I have illustrated with an example in a previous post) just like we may wrongly believe that something is true, we may also believe that something is false, when in fact it is true. I doubt you wish to deny this. So the possibility of cases of true lies follows from a more general fact that we are all fallible when it comes to the truth. And so it is not surprising at all that this general fallibility affects many aspects of human affairs, including when it comes to the nature of lying. Now add to this general fact the intention to deceive and you get cases of a lie that is true, although the one who lies fails to know that.

And, finally, as a general methodological strategy, before I am willing to abandon a definitional route (i.e., essences) in a given case in favor of entering the muddy waters of Wittgenstein's "family resemblance" notion, I expect powerful considerations that demonstrate that no other alternative is available. In my judgment, this has not been shown in the case of lying.

I would like to offer a fourth potential response to Tully Borland's recent challenge to the "intention to deceive" clause.

In my proposed response, the "intention to deceive" clause is given up as a necessary condition of a statement's being a lie. However, I replace it with conditions that I think are closely related to the "intention to deceive" clause and do its work.

Here is my proposed definition:

A lie is a false statement made by agent x when agent x has
a) knowledge that the statement is false, and
b) the intention of stating the false statement, despite knowing its falsity.

The purpose of (a) is to ensure that false statements that are thought to be true by agent x and uttered under this impression do not get included in the class of statements that are lies. For example, suppose a news anchorman is reading a teleprompter that just so happens to have a factual error. The newscaster trusts his team of teleprompter writers, who get their facts right most of the time. When he reads the factual error on the nightly news, he does not lie. Why? Because he did not utter the statement with knowledge that the statement was false.

The purpose of (b) is to capture the intentional element in the original "intention to deceive" clause. I also intend it to block counterexamples in which agent x has knowledge that a given statement is false, but accidentally utters it (perhaps because of a mental hiccup). So, using my newscaster example again, imagine that the newscaster is reading a news story that contains the following fact: "There was a bank robbery on 5th Street." This time, the newscaster knows this fact to be true. In fact, he was there to witness it. However, when he reads the nightly news, he miss-speaks and says, "There was a bank robbery on 4th Street." Has he lied? This time he has knowledge that the sentence "There was a bank robbery on 4th Street" is false. Yet, he clearly did not lie, since he did not have the intention of uttering the false statement. He intended to read the news correctly, but failed.

Now we can apply this definition to Tully's case and say that corrupt Larry lied. When he coldly looks at the jurors and states "I did not do it," he makes a false statement with knowledge that it is false, and he does so intentionally. Liar!

The obvious weakness of my definition is that it gives up on the paradigm definition, which includes the "intention to deceive". A second weakness is that it seem susceptible to counterexamples which involve certain kinds of jokes and teases. For instance, my father, who is a school teacher, regularly makes statements like, "George Washington gave the Gettysburg Adress," in order to check whether his students are paying attention. I don't think such cases are lies, but perhaps intuitions divide on such counterexamples. It seems like we could build in other conditions to handle them, though.

Mr. Shields,

Your account of lying is basically Aquinas's, which I quickly glossed in a comment above as "an intentionally untruthful declaration to another [full-stop]."

Knowledge in your clause (a), however, seems too strong. One can tell a lie in certain circumstances if one merely believes that what one is stating is false.

But because of the cases like the one you mention (e.g. joke telling), I'm skeptical that there is a set of necessary and sufficient conditions that will fit the bill for an adequate definition.

Here's what I'm getting at. One the one hand, it's EASY to give a set of necesary and sufficient conditions for lying: one lies iff one lies, or x is a lie iff x is a lie and there is no Santa Clause in world Alpha. But that doesn't give us the nature of a lie (the genus & differentia) and it's not very helpful. On the other hand (presuming that lying has a nature), there will be a set of necessary and sufficient conditions for a lie, but explicating that in terms of a definition could be quite complicated (an extreme example would be analyzing a lie as one might analyze one of Leibniz's complete concepts--L is a lie iff L is an assertion believed to be false, L is not blue, L is not God, L exists, either L is not me or 2+2=5....) But we don't have time for that kind of thing!

I have seen enough examples of borderline cases of lying that I'm skeptical there is a definition of lying (in terms of necessary and sufficient conditions) that we can grasp (other than the painfully obvious ones). That makes me inclined to adopt the paradigm/focal meaning view. But perhaps there is an ingenius grad student who will write a paper showing that I should believe otherwise.

The comments to this entry are closed.

My Photo
Blog powered by Typepad
Member since 10/2008



June 2024

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
Blog powered by Typepad