I think using the phrase “level 6 lies” when referring to Scott’s taxonomy is itself at least a “level 6 lie”.
False. It is level 2: “Reasoning well and getting things wrong by bad luck”. I interpreted “the NYT routinely and deliberately misleading millions of people” to fit the definition of the word “lied”.
Unfortunately, a bunch of commenters thought that didn’t fit the definition; maybe that stricter definition is superior, but it isn’t common knowledge for most English speakers.
Your justification seems to me almost completely non-responsive to the point I was actually making, which is not about whether it’s reasonable to call what the NYT did in these cases “lying” but about whether it’s reasonable to call something at level 6 in Scott’s taxonomy a “level 6 lie”.
Scott classifies utterances into seven types in ascending order of dishonesty. The first four are uncontroversially not kinds of lying. Therefore, something on the sixth level of Scott’s taxonomy cannot reasonably be called a “level 6 lie”, because that phrase will lead any reader who hasn’t checked carefully to think that Scott has a taxonomy of levels of lying, where a “level 6 lie” is something worse than a level 5 lie, which is worse than a level 4 lie, … than a level 1 lie, with all these things actually being kinds of lies.
Whereas in fact, even if we ignore Scott’s own opinion that only “the most egregious cases of 6” (and also all of 7) deserve to be called lies at all, at the absolute worst a level-6 utterance is more-dishonest-than only one lower level of lie.
Further, you called these things “Scott Alexander’s criteria for media lies”, which is plainly not an accurate description because, again, more than half the levels in his taxonomy are completely uncontroversially not lying at all (and Scott’s own opinion is that only the top level and “the most egregious cases of” the one below should be called lying).
So even if you were 100% sincere and reasonable in regarding what the NYT did as (“routinely and brazenly”) lying, I do not see any way to understand your alleged application of Scott’s taxonomy as a sincere and reasonable use of it. I do not find it plausible that you are really unable to understand that most of its levels are plainly not types of lie. I do not find it plausible that you really thought that something that begins with “reasoning well, and getting things right” followed by “reasoning well, but getting things wrong because the world is complicated and you got unlucky” can rightly be described as “criteria for media lies”.
I could, of course, be wrong. Maybe you really are stupid enough not to understand that “according to X’s criteria for media lies, Y is a level 6 lie” implies that what X presented is a classification of lies into levels, in which Y comes at level 6. Or maybe the stupidity is mine and actually most people wouldn’t interpret it that way. (I would bet heavily against that but, again, I could be wrong.) Maybe you didn’t actually read Scott’s list, somehow. But you don’t generally seem stupid or unable to understand the meanings and implications of words, so I still find it much much more plausible that you knew perfectly well that Scott was presenting a taxonomy of mostly-not-lies, and chose to phrase things as you did because it made what you were accusing the NYT of sound worse. Which is, I repeat, on at least level 6 of Scott’s taxonomy.
And, again, none of this is about whether the NYT really did what you say, nor about whether it’s reasonable to describe what you said the NYT did was lying. It’s entirely about your abuse of Scott’s taxonomy, which (1) is not a list of “criteria for media lies” and (2) is not something that justifies calling an utterance at its Nth level a “level N lie”.
Look, instead of going to all this trouble to impute dark motives, maybe you could go look at the whole purpose of this post, the three NYT articles I cited, see what they’re writing about, and notice “wow, this well-trusted institution really was bending over backwards to deceive many thousands of people about the Ukraine war, it’s pretty cool that Trevor found this and put the work into a post pointing it out!”.
It looks like the disagreement stems entirely from this line:
On Scott Alexander’s criteria for media lies, these would be Level 6 lies; however, Level 7 lies are not practical for journalists, nor particularly necessary in a world where the lawyers-per-capita is as high as it is today.
That list was sent to me by a friend, and I figured citing it would be helpful. Upon going and looking through Scott’s other writings on the topic e.g. The Media Very Rarely Lies, it looks clear that Scott went to a lot of trouble to standardize a very specific definition for the word “Lie”. It is not surprising that lots people in this community got most of their understanding of news outlets deception from a few Scott Alexander posts (which are great posts), and subsequently expect short inferential distances from people who approached the topic from completely different backgrounds.
I haven’t spent the last 10 years on Lesswrong, and don’t really have experience with people finding galaxy-brained ways to write deceptive posts that are plausibly deniably disguised as mistakes. My understanding is that that sort of behavior is common among corporate executives. I instead spent the last 10 years in environments where people would just take mistakes at face value, and ask about the details of word definitions, instead of immediately jumping to accusations of deliberate dishonesty.
I did, in fact, read the post and the NYT articles, and I am not convinced that your description of what they do and what it means is correct. So, if my response to your article doesn’t consist mostly of the gushing praise your first paragraph indicates you’d prefer, that’s one reason why.
But, regardless of that: If you write something wrong, and someone points out that it’s wrong, I don’t think it’s reasonable to respond with “how dare you point that out rather than looking only at the other parts of what I wrote?”.
Scott is not using some weird eccentric definition of “lie”. E.g., the main definition in the OED is: “An act or instance of lying; a false statement made with intent to deceive; a criminal falsehood.” (Does that first clause soften it? Not really; it’s uninformative, because they define the verb “lie” in terms of the noun “lie”.) First definition in Wiktionary is ” To give false information intentionally with intent to deceive”. But, in any case, even with a very broad definition of “lie” the first four levels in his taxonomy are simply, uncontroversially, obviously not kinds of lying. Again, the first one is “reasoning well, and getting things right”.
If I say “There are seven classes of solid objects in the solar system: dust motes, pebbles, boulders, mountains, moons, small planets, and large planets” and you identify something as a small planet, you should not call it “a Level 6 Planet, according to gjm’s classification of planets”.
And, while I understand a preference for being charitable and not leaping to calling things dishonest that aren’t necessarily so … I don’t think you get to demand such treatment in the comments on an article that does the exact reverse to someone else.
False. It is level 2: “Reasoning well and getting things wrong by bad luck”. I interpreted “the NYT routinely and deliberately misleading millions of people” to fit the definition of the word “lied”.
Unfortunately, a bunch of commenters thought that didn’t fit the definition; maybe that stricter definition is superior, but it isn’t common knowledge for most English speakers.
Your justification seems to me almost completely non-responsive to the point I was actually making, which is not about whether it’s reasonable to call what the NYT did in these cases “lying” but about whether it’s reasonable to call something at level 6 in Scott’s taxonomy a “level 6 lie”.
Scott classifies utterances into seven types in ascending order of dishonesty. The first four are uncontroversially not kinds of lying. Therefore, something on the sixth level of Scott’s taxonomy cannot reasonably be called a “level 6 lie”, because that phrase will lead any reader who hasn’t checked carefully to think that Scott has a taxonomy of levels of lying, where a “level 6 lie” is something worse than a level 5 lie, which is worse than a level 4 lie, … than a level 1 lie, with all these things actually being kinds of lies.
Whereas in fact, even if we ignore Scott’s own opinion that only “the most egregious cases of 6” (and also all of 7) deserve to be called lies at all, at the absolute worst a level-6 utterance is more-dishonest-than only one lower level of lie.
Further, you called these things “Scott Alexander’s criteria for media lies”, which is plainly not an accurate description because, again, more than half the levels in his taxonomy are completely uncontroversially not lying at all (and Scott’s own opinion is that only the top level and “the most egregious cases of” the one below should be called lying).
So even if you were 100% sincere and reasonable in regarding what the NYT did as (“routinely and brazenly”) lying, I do not see any way to understand your alleged application of Scott’s taxonomy as a sincere and reasonable use of it. I do not find it plausible that you are really unable to understand that most of its levels are plainly not types of lie. I do not find it plausible that you really thought that something that begins with “reasoning well, and getting things right” followed by “reasoning well, but getting things wrong because the world is complicated and you got unlucky” can rightly be described as “criteria for media lies”.
I could, of course, be wrong. Maybe you really are stupid enough not to understand that “according to X’s criteria for media lies, Y is a level 6 lie” implies that what X presented is a classification of lies into levels, in which Y comes at level 6. Or maybe the stupidity is mine and actually most people wouldn’t interpret it that way. (I would bet heavily against that but, again, I could be wrong.) Maybe you didn’t actually read Scott’s list, somehow. But you don’t generally seem stupid or unable to understand the meanings and implications of words, so I still find it much much more plausible that you knew perfectly well that Scott was presenting a taxonomy of mostly-not-lies, and chose to phrase things as you did because it made what you were accusing the NYT of sound worse. Which is, I repeat, on at least level 6 of Scott’s taxonomy.
And, again, none of this is about whether the NYT really did what you say, nor about whether it’s reasonable to describe what you said the NYT did was lying. It’s entirely about your abuse of Scott’s taxonomy, which (1) is not a list of “criteria for media lies” and (2) is not something that justifies calling an utterance at its Nth level a “level N lie”.
Look, instead of going to all this trouble to impute dark motives, maybe you could go look at the whole purpose of this post, the three NYT articles I cited, see what they’re writing about, and notice “wow, this well-trusted institution really was bending over backwards to deceive many thousands of people about the Ukraine war, it’s pretty cool that Trevor found this and put the work into a post pointing it out!”.
It looks like the disagreement stems entirely from this line:
That list was sent to me by a friend, and I figured citing it would be helpful. Upon going and looking through Scott’s other writings on the topic e.g. The Media Very Rarely Lies, it looks clear that Scott went to a lot of trouble to standardize a very specific definition for the word “Lie”. It is not surprising that lots people in this community got most of their understanding of news outlets deception from a few Scott Alexander posts (which are great posts), and subsequently expect short inferential distances from people who approached the topic from completely different backgrounds.
I haven’t spent the last 10 years on Lesswrong, and don’t really have experience with people finding galaxy-brained ways to write deceptive posts that are plausibly deniably disguised as mistakes. My understanding is that that sort of behavior is common among corporate executives. I instead spent the last 10 years in environments where people would just take mistakes at face value, and ask about the details of word definitions, instead of immediately jumping to accusations of deliberate dishonesty.
I did, in fact, read the post and the NYT articles, and I am not convinced that your description of what they do and what it means is correct. So, if my response to your article doesn’t consist mostly of the gushing praise your first paragraph indicates you’d prefer, that’s one reason why.
But, regardless of that: If you write something wrong, and someone points out that it’s wrong, I don’t think it’s reasonable to respond with “how dare you point that out rather than looking only at the other parts of what I wrote?”.
Scott is not using some weird eccentric definition of “lie”. E.g., the main definition in the OED is: “An act or instance of lying; a false statement made with intent to deceive; a criminal falsehood.” (Does that first clause soften it? Not really; it’s uninformative, because they define the verb “lie” in terms of the noun “lie”.) First definition in Wiktionary is ” To give false information intentionally with intent to deceive”. But, in any case, even with a very broad definition of “lie” the first four levels in his taxonomy are simply, uncontroversially, obviously not kinds of lying. Again, the first one is “reasoning well, and getting things right”.
If I say “There are seven classes of solid objects in the solar system: dust motes, pebbles, boulders, mountains, moons, small planets, and large planets” and you identify something as a small planet, you should not call it “a Level 6 Planet, according to gjm’s classification of planets”.
And, while I understand a preference for being charitable and not leaping to calling things dishonest that aren’t necessarily so … I don’t think you get to demand such treatment in the comments on an article that does the exact reverse to someone else.