How defensible is pro-social lying?

Dishonesty is one of the most complex, most interesting problems in practical ethics. Having an accurate model of reality is the basis for sane decisions, and deliberately causing other people to believe falsehoods makes their models of reality less accurate, impairing their ability to act rationally – often in unpredictable ways, because in a causal universe, falsehoods have to be connected to other falsehoods in order to make any sense. Lying is a breach of the listener’s autonomy: it implies that you don’t see them as entitled to the same information you have, which practically means that their ability to make rational judgments is not very relevant to you, at least in comparison to whatever it is you’re trying to achieve by lying. As a component of social relations, being able to rely on other people sincerely cooperating with you is immensely important, and notable violations of trust tend to damage relationships in ways that are very difficult to heal.
Still, most of our conversations tend to be peppered with all kinds of almost meaningless, apparently useful offhand lies. While most of them don’t seem to result in significant negative consequences, they do lessen the extent to which things people say can be expected to correspond to the things they think are true. (Up until a few years ago, I used to think of this as pretty much acceptable, and would frequently resort to butler lies out of laziness or exaggerate events to make my stories sound more interesting. In retrospect, going by the values I hold now (in the ripe age of 22!), almost all lies of this kind seem like a moral net negative, and I wish I would have understood the value of integrity as a virtue a lot earlier.)

In most situations, it’s pretty easy to choose to just not be a dick who lies about actually meaningful information, but so-called white lies, supposedly pro-social, are an aspect of communication I often find very stressful. In a culture where various degrees of dishonesty are pervasive enough to be considered part of basic politeness, not participating in lying of any kind is extremely difficult. People are expected to resort to white lies – such as telling a friend they’re busy when they simply don’t feel like hanging out right now – pretty much automatically, so that beginning to explain one’s actual reasons for anything is bound to sound vaguely suspicious. If a person says they’re unfortunately just too busy to meet today when they actually mean they just don’t feel like hanging out right now, what could the person who says they just don’t feel like hanging out right now actually mean? The possibilities are endless and exciting! (Secretly disliking you, trying to avoid all interaction and hoping you won’t ever ask again can certainly not be ruled out.)
Of course, saying you’re busy even if you aren’t isn’t usually even meant to distribute false information with certainty – it’s common knowledge that the phrase is sometimes just a polite way of saying you don’t feel like hanging out, and most people will in most situations effortlessly understand this possibility. This dynamic makes it hard to approach discussions with explicit honesty precisely because people are so used to the process of taking hints: when all messages are softened just a little bit around the edges to avoid the potential maximum amount of mildly hurt feelings, and you show up trying to communicate without embracing these almost universally accepted rules – with explicit honesty – the result is usually a specific kind of uneasiness best described as a weird superposition of maybe-hurt feelings everywhere forever.

Still, this aversion to distributing literally-false information is a very reasonable heuristic. The reason many more or less nerdy people deviate from the social rules described above – the rules which dictate you need to soften everything you say with courteous dishonesty, even if you end up complicating all honest interactions of everyone else in the community – is that there is an alternative set of rules to these interactions, which intuitively makes a lot more sense: saying approximately the things you actually mean. In theory, when things are expressed adhering to this second set of rules, everyone should be able to understand them almost in the same way, which to me sounds like a pretty great basis for information sharing.

I should stress that I’m not opposed to most forms of ambiguousness – I like some social guessing games a lot and I think they contribute a lot to the interestingness of human interactions, and I personally don’t have a lot of trouble reading nonverbal signals from people I know and interpreting them accurately – it’s just that I don’t want to say things if their more literal interpretation (which people also often need to use) is false, because it’s also very reasonable to interpret things literally, and I know that many people will probably do so and would often prefer to do so myself. Ambiguousness should be done in a way that leaves room for the use of the explicitly honest rule set too, and doesn’t disrupt it as described above, because there are times when honesty is truly needed and it isn’t always obvious whether this is in fact the case.
Even in a world where the rules of social information sharing have developed into the current complicated abomination which takes ages to learn in its entirety, and will backfire every once in a while even after you do, speaking the truth still seems like a valuable Schelling point: if I just explain things explicitly, and the listener is charitable, everything will probably go better than if I resort to saying something basically unrelated when I don’t have a clear idea of what would be the best thing to say to convey what I mean. But this method almost always requires a lengthy uninteresting explanation and more effort than just knowing what you’re supposed to say.

This is related to the reason we see this abundance of non-malicious lying. Effectively conveying an approximation of the truth and its relevant aspects is probably the most important function of white lies: most truths have to be understood as related to a body of other truths to make sense, and sometimes the situation is simply too complex to be explained in sufficient detail – the result of a short explanation in this case is actually only a half-truth, which could actually cause more damage than a considerate lie does.
For example: A likes to spend time with other humans a lot less than B, because A places a lot more value on the time they spend alone. B easily gets bored in isolation, is always up to social interaction unless it’s with a person they don’t like very much or is currently mad at, and very reasonably assumes on an intuitive level that everyone else is pretty much like them in this respect, because the people B usually interacts with really are as extroverted as B is. (B may still be failing at rationality in this case, but certainly not outrageously, so this is something everyone does all the time.)
It’s clearly not enough for A to tell B the truth I don’t feel like hanging out with you right now, even if it’s completely honest and doesn’t in itself mean A dislikes B: to tell the entire relevant truth and leave B with an accurate picture of their relationship, A would also have to figure out how their fairly fundamental, complicated social preferences differ from each other, and explain all of this to B. In a situation where the inferential distance between the participants is this large, A saying I’m busy (even if they aren’t, and it’s a lie in the speaker’s end) is actually closer to the truth in the relevant respects than I don’t feel like hanging out; as interpreted by B: B will accurately feel that A values spending time together, and ask again some other time, and all is basically well (until A is on their deathbed and has to explain why they never accomplished anything in their life other than 350k Reddit karma points, despite supposedly being very busy with important things all the time. But that’s only going to be awkward for a little while).

Anyway, this seems like the most important point where pro-social/white lies – the not-literal or only barely dishonest social niceties which emerged to soften interactions and make them cognitively less demanding – differ from malicious lying, where you purposely deceive the listener & have them believe more false things about the world than would be useful for them (e.g. to get something you want from them). One of the reasons I’m still pretty averse to white lies is that there is no easily detectable distinction in kind between them and these more serious deceptions, as all dishonesty will necessarily deprive other people of information they need to be able to interact rationally with the rest of reality, which again is a dick move. More importantly, accepting one kind of outright lying as an OK thing to do will make the rationalization of actually harmful, but temptingly self-serving lies a gazillion times easier. No one will approximately ever assess a lie they strongly feel like telling, deem it malicious or wrong, and proceed to just not tell it. It’s easy to convince yourself of that you’re only lying to be nice and doing the right thing, as avoiding the effort of explaining complex things or of facing a conflict rooted in something you did will always, mysteriously, feel a lot more appealing than being honest.

So, while I understand that some white lies seem on the surface level to make everyone’s life mostly easier, I’m reluctant to accept this is actually the case once you look at all the externalities. On the other hand, I’ve noticed that I often get exhausted around people who are very committed to this hey-let’s-simplify-interactions-by-just-telling-everyone-everything-explicitly thing. It could just be that choosing to adopt this mindset tends to correlate with not being intuitively good at reading others in social situations, which means that many of the people doing it are going to be bad at assessing which bits of information are interesting to the other person in the interaction, and which ones will just needlessly add to the cognitive load. I don’t actually have a solution to any of this, but would sort of like to find one.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s