Thursday, March 4, 2010

What's a lie? (redux)

There seems to be a constellation of things that are lie-like and people may or may not view them as lies. There seems to be three elements of a traditional lie: deception, saying something the sayer believes false, and an element that is viewed as worthy of condemnation. If a speech act has at minimum any two of these elements, there is a chance it may be construed a lie.

Consider this scenario where someone says something he doesn't believe for the sake of hurting someone else but he does so without trying to deceive the person (or anyone else). If Bob gets angry at his wife Jean and tells her "You're a fat!" out of spite, even though neither he nor Jean thinks of Jean as fat, then has he lied? He said something he doesn't believe true and his act may be generally considered condemnable but it's not deceptive. We can think of other examples with any two of the elements without the other. It seems that there may be many things that are lies but have only two, but any two combination, of the elements to be a lie while traditional lies have all three elements. Where to draw a line between those that have two elements that are lies and those that don't is contingent. I doubt there is anything that has just one of the elements that people will consider a lie but I could be wrong.

What's a lie?

It's not clear to many philosophers what a lie is. There are some definitions but all are problematic for one reason or another. The traditional philosophical definition defines a lie as something said which the sayer believes false with the intention of deceiving someone. There are problems where there are speech acts that don't seem to fall definitively in the lie category under any definition I've heard or thought up. Take "lies" involving the use of equivocations or implicatures as the most obvious examples. I don't know if these issues involving these speech acts will ever be resolved; that is I don't know if there is a fact of the matter about these speech acts that makes them lies or not. They may just lie (no pun intended) in some gray area.

However, philosophers have also debated whether or not bald-faced lies are really lies. Bald-face "lies" do not involve deception. Consider this scenario. A student who has been caught cheating is summoned to the dean's office at her college. It is the policy of this college that no matter how strong the evidence is for cheating, a student must explicitly admit to the act of cheating to be subject to disciplinary actions. The dean has conclusive proof that she cheated (perhaps he has security photos). Both she and the dean knows she has cheated. Moreover, she knows that he knows she cheated because he shows her the photos of her in the act. But she says that she didn't cheat so as to not be subject to those disciplinary actions. Has she lied to the dean?

Apparently, everyday people will say a definitive 'yes' as was determined in a study by an "experimental philosopher". This suggests that a true definition of lying that draw boundaries inclusive of bald-faced lies (e.i., without the deception criterion) will be the far more accurate definition. One of the authors (Don Fallis) favors his own definition which the study suggests is the one that best fits the responses from the study subjects among several rival definitions including the traditional philosophical definition (originally from Bernard Williams, I believe).

However, I am not so sure about that. I'm agnostic between the traditional definition and Fallis's. There is a possibility that people who classify bald-face lies do so not because they think of them in the same way as more straightforward lies but because they have some alternate notion of "lying" which they realize at some level are misnomers.

Take this example. If asked, most people will probably say that a killer whale (Orcinus orca) is a whale. But the term "killer whale" is a misnomer as these creatures (orcas) are not really whales at all (they are dolphins or members of the family Delphinidae).

Now if one were to show people a picture of different kinds of dolphins (something most people presumably don't think of as whales), next to a picture of different kinds of whales, say humpback whales (Megaptera novaeangliae), blue whales (Balaenoptera musculus)and right whales (Balaena australis), and show them a picture of a killer whale and let them "classify" where best the orca belongs, I think we'd get an interesting and more nuanced picture, one that is more accurate of people's concept of marine animals (cetaceans in this case). I think more people will cluster the killer whale with the dolphins than the (true) whales.

Just because people may think that killer whales are whales does not mean that they will have a non nuanced conception of whale-hood and may have even conflicting concepts of whales. They may use "whale" differently for different purposes fully but implicitly acknowledging salient differences. Likewise, I suggest may be the case with lies.

I say a better way to test people's implicit, intuitive understandings of the definition of lying and whether or not bald-faced lies are really lies in the same sense as straightforward lies is to set up a similar thing.

Have at one end, descriptions of scenarios where there clearly are lies (straightforward lies which involves deception) while at the other end, have scenarios describing intentional speech acts that the teller believes false but without the intent to deceive and are uncontroversially not lies (such as instances of sarcasm, telling of jokes, or instances of acting or mimicking). The later are clearly not lies and most people in the study above have attested to that fact. Then ask people where bald-faced lies best fit in that conceptual space. I think you'd get a much more conflicting picture with roughly equal numbers of people choosing one end and others picking the other end and many choosing somewhere inbetween. This may show that bald-faced lies are in some gray area much like equivocations and implicatures rather than clear instances of lies.