I'm sorry I took so long to post this. My computer broke a little while ago. I promise this will be relevant later.

A surgeon has to perform emergency surgery on a patient. No painkillers of any kind are available. The surgeon takes an inert saline IV and hooks it up to the patient, hoping that the illusion of extra treatment will make the patient more comfortable. The patient asks, "What's in that?" The doctor has a few options:

  1. "It's a saline IV. It shouldn't do anything itself, but if you believe it's a painkiller, it'll make this less painful.
  2. "Morphine."
  3. "The strongest painkiller I have."

-The first explanation is not only true, but maximizes the patient's understanding of the world.
-The second is obviously a lie, though, in this case, it is a lie with a clear intended positive effect: if the patient thinks he's getting morphine, then, due to the placebo effect, there is a very real chance he will experience less subjective pain.
-The third is, in a sense, both true and a lie. It is technically true. However, it's somewhat arbitrary; the doctor could have easily have said "It's the weakest painkiller I have," or "It's the strongest sedative I have," or any other number of technically true but misleading statements. This statement is clearly intended to mislead the hearer into thinking it is a potent painkiller; it promotes false beliefs while not quite being a false statement. It's Not Technically Lying. It seems that it deserves most, if not almost all, the disapproval that actually lying does; the truth does not save it. Because language does not specify single, clear meanings we can often use language where the obvious meaning is false and the non-obvious true, intentionally promoting false beliefs without false statements.

Another, perhaps more practical example: the opening two sentences of this post. I have been meaning to write this for a couple weeks, and have failed mostly due to akrasia. My computer broke a few months ago. Both statements are technically true,1 but the implied "because" is not just false, but completely opposite the truth - it's complex, but if my computer had not broken, I would probably never have written this post. I've created the impression of a quasi-legitimate excuse without actually saying anything false, because our conventional use of language filled in the gaps that would have been lies.

The distinction between telling someone a falsehood with the intention of promoting false beliefs and telling them a truth with the intention of promoting false beliefs seems razor-thin. In general, you're probably not justified in deceiving someone, but if you are justified, I hardly see how one form of deception is totally OK and the other is totally wrong. If, and I stress if, your purpose is justified, it seems you should choose whichever will fulfill it more effectively. I'd imagine the balance generally favors NTL, because there are often negative consequences associated with lies, but I doubt that the balance strictly favors NTL; the above doctor hypothetical is an example where the lie seems better than the truth (absent malpractice concerns).

For what common sentiment is worth, people often see little distinction between lies and NTLs. If I used my computer excuse with a boss or professor, and she later found out my computer actually broke before the paper was even assigned, my saying, "Well, I didn't claim there was a causal connection; you made that leap yourself! I was telling the truth (technically)!" is unlikely to fix the damage to her opinion of me. From the perspective of the listener, the two are about equally wrong. Indeed, at least in my experience, some listeners view NTL as worse because you don't even think you're lying to them.

Lying does admittedly have its own special problems, though I think the big one, deception of others, is clearly shared. There is the risk of lies begetting further lies, as the truth is entangled.  This may be true, but it is unclear how Not Technically Lying resolves this; if you are entirely honest, the moment your claim is questioned seriously, you either admit you were misleading someone, or you have to continue misleading them in a very clever manner. If you were actually justified in misleading them, failing to do so does not appear to be an efficient outcome. If you're able to mislead them further, then you've further separated their mind from reality, even if, had they really understood what you said, you wouldn't have. And, of course, there's the risk that you will come to believe your own lies, which is serious.

Not Technically Lying poses a few problems that lying does not. For one, if I fill in the bottom line and then fill in my premises with NTL's, omitting or rephrasing difficult facts, I can potentially create an excellent argument, an investigation of which will show all my premises are true. If I lied, this could be spotted by fact-checking and my argument largely dismissed as a result. Depending on the context (for example, if I know there are fact-checkers) either one may be more efficient at confounding the truth.

While it may be a risk that one believes their own lies, if you are generally honest, you will at least be aware when you are lying, and it will likely be highly infrequent. NTL, by contrast, may be too cheap. If I lie about something, I realize that I'm lying and I feel bad that I have to. I may change my behaviour in the future to avoid that. I may realize that it reflects poorly on me as a person. But if I don't technically lie, well, hey! I'm still an honest, upright person and I can thus justify visciously misleading people because at least I'm not technically dishonest. I can easily overvalue the technical truth if I don't worry about promoting true beliefs. Of course, this will vary by individual; if you think lying is generally pretty much OK, you're probably doomed. You'd have to have a pretty serious attachment to the truth. But if you have that attachment, NTL seems that much more dangerous.

I'm not trying to spell out a moral argument for why we should all lie; if anything, I'm spelling out an argument for why we shouldn't all Not Technically Lie. Where one is immoral, in most if not all cases, so is the other, though where one is justified, the other is likely justified as well, though perhaps not more justified. If lying is never justified because of its effect on the listener, then neither is NTL. If lying is never justified because of its effect on the speaker, well, NTL may or may not be justified; its effects on the speaker don't seem so good, either.

To tie this into AI (definitely not my field, so I'll be quite brief), it seems a true superintelligence would be unbelievably good at promoting false beliefs with true statements if it really understood the beings it was speaking to. Imagine how well a person could mislead you if they knew beforehand exactly how you would interpret any statement they made. If our concern is the effect on the listener, rather than the effect on the speaker, this is a problem to be concerned with. A Technically Honest AI could probably get away with more deception than we can imagine.

1-Admittedly this depends on your value of a "little while," but this is sufficiently subjective that I find it reasonable to call both statements true.

As a footnote, I realize that this topic has been done a lot, but I do not recall seeing this angle (or, actually, this distinction) discussed; it's always been truth vs. falsity, so hopefully this is an interesting take on a thoroughly worn subject.


New to LessWrong?

New Comment
85 comments, sorted by Click to highlight new comments since: Today at 9:36 AM
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

Maybe I'm just deluding myself, or as Michael Vassar would put it, expressing codes of morality that I learned by being socialized by fantasy novels as a kid...

...but for me, one of the primary motivations against lying is "Once a man gets a reputation for lying, he might as well be whistling in the wind." At least if you get a reputation for Not Technically Lying, your words still mean things, they just have to be carefully double-checked. Depending on how much trouble you go to in order to Not Technically Lie, when anyone else would just lie and be done with it, it expresses an odd sort of respect for the truth. A lot of that would depend on context and motivation, I expect. And maybe if you aren't socialized by fantasy novels you just don't care about the difference at all.

I think there are grades of Not Technically Lying, too. For example, there's Not Literally Lying According To Sentence Syntax But Lying If You Added Words That Appear By Gricean Implication, like the missing "because" in the opening paragraph. In contrast, Subtly Changing the Subject or Not Answering the Original Question, actually seem to me substantially less similar to outright lies - the original sentence becomes a lie if an implied word "because" is said out loud; Subtly Changing the Subject is more... semantic.

It seems likely to me that a reputation for Not Technically Lying, being highly unusual, would be highly salient, and so might lead to less trust than a reputation for lying at an average frequency – i.e., the null reputation, not salient at all.

I'm pretty sure that this is correct.

More precisely, I'm pretty sure that one simply doesn't want to have ANY reputation regarding trustworthyness, truth, or whatever. Make issues of truth salient and you loose. Even a reputation for always communicating honestly (no efforts at deception) costs you status because it makes you a less valuable ally, less capable of desirable forms of partiality, and above all, weird. Being seen as a trusted neutral third party is at best a weak consolation prize, and one that is only possible if you are also seen as either a) not having your own agenda, or b) not having an agenda that anyone is allowed to question.

By contrast, politicians who are caught in lies repeatedly pick themselves up and go back to being high status politicians after wiping the dirt off their faces.

You can have in someone's eyes a reputation of lying to everyone else but being truthful to this particular person because they're special. I've seen such cases.
3Eliezer Yudkowsky15y
Pope's weird. Wouldn't have much status if he were normal, he'd just be Chair (not CEO) of a large international corporation.
In addition to the obvious position-of-authority thing, it might be relevant that the Pope's weirdness is a factor of (or at least can easily be attributed to) his situation, not his disposition (as honesty would be).
The Pope is a good neutral third party. He has taken the consolation prize of being the World's Most Moral Man because he can't be Vladimir Putin or Barack Obama, both of whom have more friends and more power.
Is it actually possible to get a reputation for NTL? I think the closest you can get is a reputation for whining in weird ways when called on clear-cut lying (and probably less specific than that). Is it even possible to get such a reputation among nerds? ie, people who have a poorly calibrated sense of what's a weird reaction

Is it actually possible to get a reputation for NTL?

Yes. I have such a reputation amongst some of my friends. In fact, one time I was hanging out with two friends, one of which knew of this reputation (Amy) and one of which didn't (Beth), and so the following conversation occured:

  • Amy: Oh, but you probably shouldn't go on the roller coaster.
  • Me: Why not?
  • Amy: Don't you get motion sickness on airplanes and stuff? You could throw up on the rollercoaster.
  • Me: I've never thrown up while on a roller coaster.
  • Amy: But have you ever been on a roller coaster?
  • Beth: Amy! (Tone is as if Amy had been particularly rude to me or something.)
  • Amy: What? He never actually said he's been on one.
  • Beth: Yeah, but clearly he means that...
  • Me: (interrupting, 'cause I did want to get on the roller coaster and wanted to eliminate this conversation which was delaying my goal) I'll explicitly state that yes, I have been on roller coasters before, and I have never puked while on one.
Thanks for the story! I say it doesn't count. There's probably a bit of "no true Scotsman," but hear me out. My current position is that people don't get a reputation for being honest or dishonest (cf Psychohistorian); reputation probably consists of a list of allies and measures of loyalty to a generic ally. For some meanings of "reputation" you didn't have one: Beth didn't know (though I'd need more details). Moreover, your reputation among your friends wasn't about your honesty, it was about a game you played. It wasn't that you used NTL to manipulate people, or how to extract secrets from you. I am a little impressed that you got the reputation at all. Are you and your friends nerds? (how about Beth?) So my new questions: 1. Can one have a reputation for being dishonest, as opposed to a reputation for not having allies? 2. Can one have a reputation for being honest? can that be positive? maybe a reputation for being honest about who are your allies?
Yes, we're nerds. I don't about Beth because she was Amy's friend rather than my friend, and I never really spoke to her again after that. Also, I think in my circle of friends, NTL is considered "honest", so if your definition of "reputation" allows that I had a reputation of being an NTL-er (an NTL-ar?), then I'd also have one for being honest.

In fact, people will probably remember you saying what you misled them to believe instead of the technically true thing you actually said.

There's an anecdote in Surely you're joking, Mr. Feynman in which as a frat brother in college Feynman stole a door off someone's room and hid it in the basement. After two or three days, the head of the frat house called everyone to a meeting and asked them all one after another, "Did you take the door?" When Feynman was asked, he replied in a sarcastic tone of voice, "Yes, I took the door." People heeded his tone instead of the literal meaning of his words, and when he finally relented a couple of days later, no one recalled the phrasing of his technically true reply.

I believe Feynman did not even use a sarcastic tone of voice. Feynman stole the door, and when everyone was arguing about it, Feynman said he knew how to solve it: Have everyone swear on their honor not to lie, and then ask them point blank whether or not they took it. So the head of the house went through each member, asking them to swear, and then asking if they took the door. When they reached Feynman, he swore, and then said that yes, he did take the door in a deadpan serious voice. They told Feynman to stop joking around, and then just went on to the next member in line, asking them to swear and whether or not they took the door, etc. And like you said, when they later found out that Feynman had indeed taken the door, nobody remembered that he had actually said "yes" when asked.

Feynman didn't want to set himself up for the NTaL; his idea wasn't the honor pledge but a plea for the anonymous thief to return the door as secretly as it was taken, with a heaping of praise for the thief's ingenuity. (Feynman was honest, but nobody ever accused him of humility.)

But when someone else came up with the honor-pledge solution, he did the only thing that comported with his ethic of literal honesty.

"Jack, did you take the door?"

"No, sir, I did not take the door."

"Tim: Did you take the door?"

"No, sir! I did not take the door!"

"Maurice. Did you take the door?"

"No, I did not take the door, sir."

"Feynman, did you take the door?"

"Yeah, I took the door."

"Cut it out, Feynman, this is serious! Sam! Did you take the door..."— it went all the way around. Everyone was shocked. There must be some real rat in the fraternity who didn't respect the fraternity word of honor!

. . .

Sometime later I finally admitted to taking the other door, and I was accused by everybody of lying. They couldn't remember what I had said. All they could remember was their conclusion after the president of the fraternity had gone around the table and asked everybody, that nobody admitted taking the door. The idea they remembered, but not the words.

It may be clearer to say that Not Technically Lying will actually get you a reputation for being crafty and untrustworthy, because people will have to put so much extra effort into figuring out how they should react to what you say, rather than just being able to trust it.
Agreed. Most people know that most people lie; sometimes out of greed, sometimes for noble reasons. But it is generally accepted that people tell bullshit to each other at some frequency. It is also generally accepted - especially within politically charged places such as the media, political debate and office politics - that people tell NTLs.
Indeed we have a word for it "spin" and a political profession associated with it as well. Although spin can cover outright lies the majority of the time it is just selective truths being told.
And in recent times, "Spin" or "Spin Doctors" has become one of the worst derogatory terms you can throw at someone. Does US politics have the concept of "spin" yet? Somehow I want to say that it is better to tell an honest lie than commit spin
We distrust someone using their SKILL and intelligence to deceive us--perhaps because it further obscures the truth, because we feel that if they can outwit us like that, the world suddenly becomes smoke and mirrors and we don't know what side we should be fighting for. If someone tells a lie, that keeps the game simple--no word play, no clever tricks that might have to be reasoned past, producing an existential angst that there may be nothing beyond the spin. With a lie, well, it can be easily falsified, and it can even be culturally accepted--because, hey, we all say we didn't eat two cookies when everyone had one or wanted one but didn't get it in time. But we don't all try to confuse people and make them THINK--especially in settings where it's socially agreed we don't/shouldn't have to.
I'm pretty sure that this is correct. More precisely, I'm pretty sure that one simply doesn't want to have ANY reputation regarding trustworthyness, truth, or whatever. Make issues of truth salient and you loose. Even a reputation for always communicating honestly (no efforts at deception) costs you status because it makes you a less valuable ally, less capable of desirable forms of partiality, and above all, weird. Being seen as a trusted neutral third party is at best a weak consolation prize, and one that is only possible if you are also seen as either a) not having your own agenda, or b) not having an agenda that anyone is allowed to question. By contrast, politicians who are caught in lies repeatedly pick themselves up and go back to being high status politicians after wiping the dirt off their faces.
It seems the main reason to put serious intellectual resources behind NTL is because you get to feel justified in the fact that you're telling the truth. At least liars will (generally, in my experience) actually feel bad when you catch them. If you have a reputation for NTL, unless I think I'm much, much smarter than you, I can't trust you, and even then I probably can't trust you unless I cross examine you after any vaguely ambiguous statement. Liars may have less respect for the truth, but if they at least feel guilty and generally try to avoid lying, they have more respect for me than someone who has no problem saying anything technically true to get what he wants. I also don't think the whistling in the wind thing is, well, true. Or, at least, I think it's really, really hard to get a reputation for being a liar, and, even if someone gets one, that person's words are still of some value, unless they're actually pathological. I don't think I've really heard of anyone who has a real reputation for lying. And a few lies here and there won't do it, since if they did, pretty much everyone ever would have a reputation for being a liar. I'd think it's gotta be very big, very wrong, and relatively visible among the relevant sphere of people, however large or small that sphere might be.
If you can be accused of having intended to deceive it counts as a lie in most social functions. If you can't be, as in changes of subject or not answering the question, that counts as good manners. It can be infuriating to the questioner if you are less subtle than you think you are, but only really upsets fairly rare people. With those rare people though, it may be a better idea to just admit that you aren't willing to answer or to place a condition on answering. We have had conflicts over this in the past where you would have done better by doing so in any event.
Grice's maxims are a very interesting factor to introduce here, and I think encapsulate very well what Psychohistorian is saying. You might argue that breaking any of them (Quality, Quantity, Relevance, Manner), with the possible exception of Manner, is a form of lying. Breaking Quality is direct lying. Breaking Quantity is excluding true information that is necessary for the listener to really understand what's going on. And what PH did here is a good example of breaking Relevance: introducing the computer information where it was actually irrelevant, thereby creating a not-technically-a-lie out of the fact that we assume he is complying with Grice's maxims. Manner is a less obvious case. You might argue that you were not-technically-lying by breaking Manner if you spewed mountains of true information and buried the part that you want to not-technically-lie about among all the other true stuff so that no one took any notice of it. But that could also be expressed as breaking both Quantity and Relevance.
I'm smiling, shaking your hand. "Yes, this was a very productive conversation! I'm sure glad I met up with you and we had this discussion!" You think I'm going to help support you in your efforts to do whatever you're going to do with the ideas or information you've shared with me tonight. Instead, I'm going to report it to another group that will stop you from doing what you want, take your ideas, or exploit the information in a way counter to your goals. I've just lied by breaking the maxim of Manner. [Edit: It's arguable that breaking Quantity would also cover this--I didn't say WHY it was productive--but it seems clear to me that it's essentially a violation of Manner.]
"once a man gets a reputation as a liar, he might as well be struck dumb, for people do not listen to the wind." - Colonel Richard "Pop" Baslim, from Citizen of the Galaxy, by Robert Heinlein. One of my favorite books, growing up.
Distinguishing NTLs by their syntax or semantics won't yield ethically useful conclusions for a utility maximizer because semantics and syntax, categorically conceived, lack any direct relationship to the NTL's consequences. You need a pragmatic standard; the common-law standard for fraud is a good starting point. For an act to count as fraud, the deceiver must intend to deceive and the deceived must reasonably rely on the false information. If the doctor tells the patient the saline solution is his most powerful sedative, would a reasonable person be justified in assuming the solution is a sedative, or would a reasonable person necessarily ask what the substance is? The answer, of course, depends on making the "reasonable person" more precise, but, if that's the kind of inquiry that's ethically appropriate, it points to the importance of preserving the patient's autonomy, even when deceiving him, and it speaks for some NTLs as better than lies. Some NTLs serve to maximize the patient's autonomy by allowing him to be unreasonable if he chooses and cooperate with his own deceit, while not foreclosing further inquiry if the patient really wants the truth. From this standpoint, I think "Don't distract me" is a more deleterious deception. The doctor silences the patient and ignores the possibility that the patient may choose to be informed, rather than comfortable. An outright lie that isn't credible could serve the same purpose and obtain the same ethical evaluation, except for the difficulty in manufacturing a suitable lie.
I just had a funny thought. In the spirit of your theory of Newcomblike problems, you will lie (or NTL) if you can extrapolate that the other person would have been thankful for it. On the other hand, adopting this strategy makes stupidity and self-deception beneficial to you, so scratch that.

It probably comes from reading too many fantasy novels and from playing Ultima IV during my formative years, but I somehow acquired a strong aversion to telling literal lies, to the point where I basically can't do it at all. On the other hand, using the truth with an intent to deceive feels fine; it doesn't trigger the same "OMG this is bad and wrong" emotional reaction I get when I think about lying outright to someone. It's like the restrictions on Aes Sedai in Robert Jordan's Wheel of Time series.

I also tell people this.

The last line makes it, in my opinion. Would upvote twice.

Among lawyers, this can show high status. Lying to a court is completely unacceptable, and can make an advocate useless to future clients. However NTL is OK.

The pupil barrister: "To be honest..."

Head of Chambers: Lawyers are always honest. The lawyer will say, "To be frank..."

(taken from the BabyBarista blog)

Lots of people say "don't talk to the police--they'll take things out of context"; that is, they'll NTL about the conversation (and substantively lie about their interpretation). Some people say "don't talk to the police--they'll fabricate an incriminating remark." But such people never say that the police will fabricate a conversation. Is there some magic line that the police won't cross that tells us more about how most people operate than the concept of NTL in this post?

(an alternate explanation is that police don't fabricate incriminating remarks; they just take things out of context and the victim lies about having stumbled into the trap; the person recounting the story doesn't want to accuse the victim, an ally, of lying, so he accuses the police.)

I would be interested to see what people think of some other options:

4) "I'll tell you later." [Fails to answer the question; is neither a Technical Lie nor actively misleading; may lead the patient to believe it's a painkiller or similar anyway]

5) "I'm doing everything I can to help you; please don't distract me." [True statement and a request; may lead the patient to believe it's a painkiller or similar anyway]

6) "It's a 300-osmolarity isotonic solution of chloridated alkali metal." [Technobabble; the average patient will have no clue what this means. Unless I'm misusing chemical terms found on Wikipedia, it's completely accurate. Will probably discourage further questions, especially if stated rapid-fire in a curt tone.]


"It's a 300-osmolarity isotonic solution of chloridated alkali metal -- the strongest painkiller I have. I'll explain the details later, but now please don't distract me -- I'm doing everything I can to help you." (stated rapid-fire in a curt tone.)

I think 'not actively misleading' is just another form of Not Technically a Lie. If you're saying X in the hope that it will lead them to believe a false statement Y, your statement is either a lie or a NTaL, regardless of how cleverly you construct X.
These three are all basically honest, but their effects are unpredictable. 6 and especially 4 may actually make the patient significantly worse off, since the first is evasive and abrasive, and the second is meaningless and prone to misinterpretation ("chloridated alkali metal!? I had no idea it was that serious!"). 5 does no harm, but likely does little good and may be seen as evasive. Given that placebos can actually still work when people know they are recieving them, 1 may be better than all these options, as may 1+3, which seems essentially honest.
If I were the patient, I imagine I would shut up and be (relatively) content with 5, but would be royally pissed off with the doctor for saying either 4 or 6. Unless I were very seriously intimidated, I think I would be extremely likely to follow up 4 with "Why can't you tell me now?" and 6 with "But what is it for?"
If I were the patient, I imagine I would interpret both 4 and 6 to mean 5, and then cease distracting the doctor from helping me. But that could just be my imagination.
"Has it got dihydrogen monoxide in it?"
I'd read through the technobabble in 6 and I suspect a large number of other people would too. (You only need high school chemistry.) I'd back calculate the reason and it would end up a nocebo. Don't bank on people being dumb - plenty aren't. 5 sounds the best to me. 4 could be read as 5, as condescension, or as near-overt dissembling depending on tone and content.

Very few people retain a high school chemistry background. Fewer retain it when in intense pain.

And furthermore, lots of people are really, really dumb. I got an A in honors high school chem and that was only five years ago, and I had to visit Wikipedia to check every word more obscure than "solution" in that description of saline - the average patient probably did worse or didn't take the class at all, took it longer ago if they did take it, and remembers less. Heck, there's probably a scarily high percentage of people who wouldn't even understand you if you told them it was saline. Now there is an interesting question: can you be held morally responsible for lying if you just don't have the time/patience/ability to explain something moderately complex to someone really stupid? What if they don't even know the word "placebo" - then what do you do? Explain scientific controls and psychosomatic effects while you're trying to extract a bullet, or what have you, from your patient?
Yes, but we're retarded, too. Never forget that.
We're not retarded. We're advanced
ROFL... no way. I personally fooled a nurse into signing a petition against dihydrogen monoxide. people are Stupider Than You Realize
I would imagine that's not a case of stupidity, but of the brain working in a way that's (usually, more or less) efficient. Instead of analyzing the specific words you're using, the nurse, who has no reason not to trust you, analyzes the content of what you're saying, the urgency and manner with which you're presenting the evidence against this chemical that's just "blahblahblah" to the brain. This is a way of filtering out irrelevant content and only paying attention to what is (likely) to be relevant. I had a related problem when learning to drive--my brain doesn't instantly process "right" or "left" as belonging to the specified direction, but when the instructor or person giving the test bellowed a word at me, I knew to turn and turned whichever way made more sense to me in context--which wasn't always the right decision. I don't think everyone has this thinking style, as evidenced by my instructor's irritation with me, but it's certainly not overall a bad one--in general, it's probably better to pay attention to information from the environment when operating heavy machinery, to the emotional content of a social situation rather than to etymological clues, and so on.

I would say that making an unnecessary injection for the purposes of creating a false belief is already going way past lying. What do words count for compared to actions?

Words are presumably more intrusive on the mind, injecting crafty inferences that are far out of the search space offered by other actions.
Sometimes. Clearly NOT in the case given in the example. Also not in many other cases. There's a reason people think prediction markets might be a good idea. For that matter, there's a reason why it's illegal to impersonate a police officer and many other types of people distinguished by costume etc.

The example introduces a difficulty unrelated to the topic of the article. The statement "this is the strongest painkiller I have", or even better "This will help your pain" is such that its truth value depends on it having been pronounced. This complicates the issue of whether it's true, because "platonically", taken outside the context of being pronounced, it's "not technically a lie", while when it's actually pronounced, it becomes true to the situation, less misleading.

This is an intuitive difficulty akin to tha... (read more)

I think there's a significant difference in having a reputation for lying, and a reputation for truth but Not Technically a Lie. The difference is, it is meaningless to converse with a person with a reputation for lying on a subject they are likely to lie about. One can however converse with someone with a reputation for NTL, because you know they will answer truthfully -- but you have to avoid filling in vagueness like you would for an honest person. If need be, you can choose an exact wording and ask them to repeat it. Note: my presumption is that you ca... (read more)

If Sam expresses themselves in such a way that the most likely interpretation of their expression is A, in a situation where Sam knows that NOT A, Sam is doing something which is expected to result in their audience believing a falsehood. Presumably that's what a consequentialist cares about. No nervousness required, and intent is relevant only insofar as it influences actual behavior. Whether we call what Sam is doing a "lie" or not, and whether we decide Sam is a bad person or not, is an entirely different question. I agree with you about the differential reliability of an outright liar and a NTL-er, though.

It occurs to me that NTL may be a safer deception technique than outright lying for another reason: an honest person may fail to disclose relevant information purely by accident - they did not realize that their audience was ignorant of it - and thereby leave a false impression. Intent is therefore harder to demonstrate in NTL cases.

Perhaps I'm just arguing for the conclusion I want, and thus digging myself deeper, but in your example, 3 seems entirely truthful to me. You're administering saline, sure, but you're administering it in order to treat pain. If you had a better means of treating pain, you would be using that. Given those circumstances "The strongest painkiller I have." actually seems accurate.

It is truth, but you are explicitly saying those words so that the hearer (the patient) forms a false belief about the world. So it cannot be really truthful because most people in that situation would, after hearing example 3, believe that they are being given something that has more affect than a placebo.

I agree with this. It's closely related to the saying "Promoting less than maximally accurate beliefs is an act of sabotage. Don't do it to anyone unless you'd also slash their tires.", which I've always interpreted to be about much more than just lying.

My main objection to NTLs is that they seem to be generated when a person is trying to convince themselves that they're not the type of person who lies, even though they are (where misleading == lying). If you want to be the type of person who never misleads people, you can't NTL. If you think you... (read more)

Agree, and also note that if slashing someone's tires would significantly reduce the trauma of a painful experience then I may just do that for them.

Much of this is academic once you realize that in communication, especially effective communication (and, really, if we're rationalists concerned with "winning communication" what other sort are we concerned with?), the truth value of a statement from our internal interpretation is irrelevant, and, indeed our intention is also irrelevant unless it is indicated over some other channel during our communication, because none of these are accessible to the people we are communicating with.

Your statement may be accurate and come from honest intent w... (read more)

bears repeating.

Feynman's Who Stole the Door? story is a great example of NTL and demonstrates that "can be accused of having intended to deceive" is a very low standard.

i think people who (think they) are frequently misinterpreted are more likely to be sympathetic to not technically lying. I have had to go out of my way quite a lot to stop myself accidentally not technically lying to people, some of whom would, when I explained what I meant, not believe me. This at first caused me to take the attitude: "LISTEN TO THE WORDS IM SAYING, YOUR HEURISTICS OBVIOUSLY AREN'T WORKING." Now that I know, to some extent, how people will misinterpret me, I don't feel at all guilty about talking as I naturally do, in a manner ... (read more)

I dunno, I hear this sort of thing sometimes, and if others are using heuristics that presumably work when communicating with other people, maybe you're the one that's missing something!
this comment really confused me but I think I have figured out why. Did you take me to, by "heuristic," be referring to more than information interpretation shortcuts? I'm not talking about "they said X, so say Y" or anything that is a heuristic for action, or responses that people have learnt which "work." I'm specifically complaining about when heuristics for determining what X said repeatedly elicit "no that's not what I meant" or similiar as a response, and people don't even try turning off the heuristic, looking at the words, and applying dictionary definitions or even asking what people mean by the words, if they are used idiosyncratically. I realise there is a small crossover where someone's heuristic for action may be to deliberately misunderstand me, or pretend to but this is kind of out there so I assume that isn't what you mean I'm missing. Is my diagnosis correct? I assumed it would be clear from context I was using heuristic this way but socially I don't explicity use "action heuristics" often (or maybe just saliently) enough for that interpretation to occur to me so that's some lacking awareness on my part if I'm correct as to what happened here.
No, I understood that, what I meant was if you're having trouble communicating, and repeatedly saying X but transmitting the wrong meaning, you might also be using incorrect heuristics to determine what to say to convey a particular concept. Of course this is just a possibility, and depends on what sort of person you're talking to! It sounds like you're talking about relatively unreflective people, in which case most of the blame for miscommunication can safely be assigned to them. It's just that this is a problem nerds sometimes run in to.

From my personal observations, it seems very hard for the human brain to live with lies which are not technically lies, and at the same time remain un-damaged in other ways. I have this concept marked as one of the major leading causes of "pretzel-brain".

The most interesting example I have found so far is of an individual who thinks of himself as always honest - and he maintains this image by never lying at the meta-level. He introduces himself as a liar to all new people he meets, and, if asked directly if a statement he's made is a lie, he will answer truthfully, to the best of his ability. It's not technically a lie if people know you're honest, and can catch you in the act if they try.

I've actually noticed the opposite with at least a few people: they don't like lie-lies, but are fine with NTLs because they're still "telling the truth". But those people may have pretzel brains. For what it's worth, I see NTLs as a type of lie: the distinction doesn't seem important.
Does this individual count conversational conventions along the lines of "Really?" to be direct inquires as to the truthfulness of statements?
I don't really know. He's a generally unusual fellow - if I had to guess, I'd say yes, he takes them to be direct inquiries. At the same time, though, he openly despises conventions, so he may behave inconsistently if any inquiry comes as a result of a convention rather than an honest attempt to get the truth. I think I do alright at figuring out when a brain's been pretzel-ed. Trying to figure out the pretzels that form in my brain and the brains of those around me, though being a lot of fun, is not something I am very good at.

The one thing that stands out for me in this is that it seems to go from the same "figures don't lie but liars sure can figure" assumption that NTL is much easier fool people with than making stuff up.

But, in my experience, that's not true. There are indicators when someone is NTL, versus actually being honest, just as I've noticed over the years that there are indications when a statistic is being taken out of context.

Most forms of deceit are either very short term, or fall quite rapidly to logic of the form "If this is true as it stands, w... (read more)

4Eliezer Yudkowsky15y
Not every post gets a Wiki page. The first time you want to refer to this concept in a blog post or comment, it gets a Wiki page.
Good concepts get wiki pages, not necessarily posts. Test by finding an occasion to use a concept isn't always necessary and isn't always sufficient, the judgment of concept's notability may be incorrect either way. In this case, the concept seems salient enough to me.
4Eliezer Yudkowsky15y
But a lot of people may think XYZ is a "good concept" and crowd the Wiki. Keeping it down to concepts that people find themselves using in discussion outside the original post, is, indeed, our version of Wikipedia's notability rule.
Can you please explain how you think the Wiki can get "crowded"? Are we going to run out of hard drive space, or is someone going to print this out, or what?
The Wiki itself can't become crowded, but categories, lists, "see also" sections, and the like can. Not sure if this is Eliezer's point.
The Wiki itself can't become crowded, but categories, lists, "See Also" sections, and the like can.
The dangers of crowding the wiki are not very intuitive, which makes some kinds of "crowding" fine, and others not so much. One problem is redundancy: each given topic should be focused, so that a small cluster of pages is sufficient to cover it, while not making the rest of the pages repetitive. This, for example, suggests that creating pages for blog posts is a bad idea, because the info would be shared by the blog post pages and concept pages that discuss the concepts introduced in those blog posts. This problem is typically healed by merging pages, rewriting the content, throwing out redundancy, and factoring out the less relevant points into separate pages. The wiki has a lot of growing to come to a stage where this becomes sufficiently relevant. Another problem is lack of structure. Bits and pieces of each topic are all over the place, so it's hard to systematically see them together, especially if you visit the wiki for the first time. This is solved by interlinking, categorization (currently, the former gets better, the latter is in disarray), and by summary pages. Yet another problem is expected quality of articles. If there are too many articles, most of them will be bad, and so reading the wiki would be more like treasure hunting. Keeping high quality requires keeping the speed of creating new pages below the speed of sufficiently improving the existing articles. This is somewhat in conflict with requirement for better structure, and in alliance with fighting redundancy. At the same time, adding stand-alone high-quality articles doesn't hurt, as long as they have a prospect of being integrated in the structure of the wiki eventually.

I wonder if being literally honest but willing to tell Not Technically Lies serves culturally in the same way that being willing to tell "little white lies" but not break an official oath/vow does. People know you're willing to be deceptive, so you're not going to be cornered into spilling any secrets they tell you, you're not going to make a tactless ass of yourself, etc. But they also know that they can trust you if they're willing to spend the time and effort to corner you on anything that's exceptionally important to them.

The NTL restrictio... (read more)

I guess this is where I come down on this:

  1. Whether you say 2 or 3, you are deliberately encouraging false beliefs in your patient.

  2. Assuming that your patient is not an abnormal human decision maker, he would probably prefer less pain over some trivial but accurate information about what substance is in a bag.

  3. If you are a utility-maximizer or a volition-extrapolator you would tell the guy whatever lie seems most likely to mitigate his suffering.

Very good, what the patient wants is more important than my conscience.

You should try to get a better conscience... one that wants what the patient wants.

Is this what you read when you came up with this idea?

From an ethical perspective, there is little difference between L and NTL. If we define a lie as 'an attempt to deceive' (as Michael Vassar puts it), they are just the same.

In practice though, for people it does seem to make a significant difference; it seems that many people avoid plain lying, and instead try some NTL or some related class of non-truth. E.g., not answering 'No, I did not do that', but instead saying 'Where could I have found the time to do that?'. For some reason, that's easier.

But again, from an ethical viewpoint it seems not very useful... (read more)

A lie is a knowing statement of untruth, almost always made in the hope that it will be mistaken for a sincere statement of truth. Deception is far larger than lies. As for intent - it's difficult to show, and depends partly on the qualities of the listener. Especially stupid and small-minded people often accuse others of trying to deceive them when the real problem was that they leapt to an invalid conclusion. My experience is that people without a great deal of self-candor will often accuse others of deception rather than considering the possibility that they themselves were dumb.
I agree with this. NTL is, by definition, not lying. In response to the original post I would ask for clarification on these points: * Is the problem with lying the lie itself? * Should deception be considered wrong? You can deceive people through silence and you can quickly find theoretical scenarios where non-action becomes clever deception. The basic objection I have to the original post is that Lying is being compared to Not-Lying when it seems like the intent was to compare Lying to Deception or possibly "other forms of deception". If the problem with Lying is Deception than the examples at the beginning are misleading. If NTL is bad, it is bad on grounds of deception, not because it looks so much like lying. To draw a comparison: Lying is to Murder as Deception is to Killing. Finding a scenario for Not-Technically-Murder is certainly possible but if the action should be considered analogous to murder for moral purposes than make it Technically-Murder. If both Murder and Not-Technically-Murder are bad because they are unjustified Killing than the subject has little to do with Murder and everything to do with unjustified Killing.

I know that this is simply an exercise but I think it points out my epistemic belief that there are no such things as gray areas in deceit.

The doctor could have easily have said "It's the weakest painkiller I have," or "It's the strongest sedative I have," or any other number of technically true but misleading statements

In reality the Saline IV is neither a painkiller nor a sedative whatsoever therefore:

The strongest painkiller I have

is actually as much of a lie as the morphine statement.

Now of course you go on to say this:


... (read more)