No I don’t. If you paid any attention to the denotations of what I have written on the topic so far, rather than imposing connotations from your own preconceptions, you would see that I have said that
a) I’m interested in “sex before dawn” because I expect it has good, data-rich scholarly arguments.
b) I doubt its well reasoned because Pinker is almost always better reasoned than his debate opponents
c) that doesn’t mean he’s right. It’s very likely that some of his views are exaggerated, partly because he’s a popular writer
d) for that reason, I expect his scholarly opponents like Jablonka and Lamb to be able to add valuable nuance.
I was trying to point out that, regardless of whether Pinker has taken exaggerated positions, he has “deliberately misrepresented key evidence in support of a position.” Is doing that well-reasoned?
Demonstrating in painstaking detail that he has done so is the entire point of the first third of the post. Did you not find that convincing?
Convincing of what? I really feel like you think arguments are soldiers and that that is ALL that arguments are. I can’t do much to help that. I’m sorry, but I can’t. You need to observe more arguments.
That Steven Pinker is unreliable when it comes to evolutionary psychology because he has misrepresented his evidence on at least one occasion. That doesn’t make his conclusions wrong, but it should make them more suspect.
I really feel like you think arguments are soldiers and that that is ALL that arguments are. I can’t do much to help that. I’m sorry, but I can’t. You need to observe more arguments.
I really don’t think that. I think it’s a good idea to be wary of Steven Pinker because I don’t expect that he will always point towards the truth.
Would it be better if I phrased my question as “Have you updated your opinion of Pinker’s credibility based on the knowledge that he misrepresented evidence on one occasion? If not, was it because you believe that the evidence demonstrating that he did so was inaccurate?”
That Steven Pinker is unreliable when it comes to evolutionary psychology because he has misrepresented his evidence on at least one occasion. That doesn’t make his conclusions wrong, but it should make them more suspect.
This is a great example of an error that, now that I think about it, is extremely common. I should write a longer article about this, since I haven’t seen any complete explanations of the phenomenon. I hope you don’t mind my using your comment as an example; you are definitely not the only person who makes this mistake, your comment just the one that happened to be the one that crystallized some concepts in my mind.
Your statement is literally true, but it’s also enough inferential steps away from the statements we’re likely to care about, that it’s effectively meaningless. It is four inferential steps away. (1) Even supposing that Pinker was maximally wrong (his opinions were generated by throwing dice), any given conclusion is still at least as likely to be true as it would be if Pinker hadn’t said anything at all. (2) Pinker is not maximally wrong; even if he did have a pattern of misrepresenting evidence, there would still only be a small chance of him having done it on any particular instance. (3) It’s not established that there is a pattern; there’s just one example, which could be atypical. And (4) it’s not even necessarily established that there is such an example; I might disagree with your interpretation.
As discussions proceed, they get further and further away from the original conclusion. And on the topic of human sexuality, they start at a longer than normal distance to begin with, because there are so many nasty traps for studies to fall into (especially people lying on surveys and to researchers). I think that on an unconscious level, people track the average inferential distance, and respond negatively if it’s out of range.
(1) Even supposing that Pinker was maximally wrong (his opinions were generated by throwing dice), any given conclusion is still at least as likely to be true as it would be if Pinker hadn’t said anything at all.
It’s not that he does no better than chance, exactly. It’s that he promotes hypotheses in an unjustified way. And so for any given hypothesis that you hear, you should ideally discount its probability by the difference between the number of bits of evidence the hypothesis should need and the number of bits of evidence you expect him to actually have. If that’s a big difference, then he’s not worth paying attention to.
Your other points are solid, though. I definitely have a habit of underestimating inferential distance.
This sounds like a better argument than I have seen from you earlier, indicating to me that you do have the ability to make such arguments when you are inclined, at least some of the time.
With respect to its content, yes, I think this is a reasonable conclusion regarding how one should respond to Pinker’s conclusions. However, as I noted, my observation is that Pinker argues well. When you have a good argument in front of you its primarily the argument and not the evidential value of the opinion which you have to confront.
Yep. When evaluating a dispute between someone and Pinker I started by saying that I could take it for granted that his arguments are sound, but that it seemed to me that their arguments might be worth-while and scholarly. (I could also have said fact-filled and valuable, especially if not taken literally, as a source of added validity).
In practice, arguments that aren’t sound are practically never fully valid, but they are frequently a valuable complement to sound arguments as part of how validity of belief is achieved.
Yep. When evaluating a dispute between someone and Pinker I started by saying that I could take it for granted that his arguments are sound
Did you mean valid here? Its the soundness of Pinker’s arguments that I am claiming to have undermined, so I’m kind of confused.
I think that my internal argument-evaluation-algorithm focuses most of its effort on premises and treats the subsequent reasoning as a mostly mechanical and straightforward process. Getting enough and good enough data together instinctively seems like the greater obstacle to me, possibly because I do fairly well with formal logic.
Ah. The problem with treating reasoning as mechanical is that almost no-one actually does reasoning reliably enough for that to work. If they did, the quality of public debate would be completely different.
No I don’t. If you paid any attention to the denotations of what I have written on the topic so far, rather than imposing connotations from your own preconceptions, you would see that I have said that a) I’m interested in “sex before dawn” because I expect it has good, data-rich scholarly arguments. b) I doubt its well reasoned because Pinker is almost always better reasoned than his debate opponents c) that doesn’t mean he’s right. It’s very likely that some of his views are exaggerated, partly because he’s a popular writer d) for that reason, I expect his scholarly opponents like Jablonka and Lamb to be able to add valuable nuance.
That was an incredulous question. Like asking someone if they deny the theory of evolution.
I was trying to point out that, regardless of whether Pinker has taken exaggerated positions, he has “deliberately misrepresented key evidence in support of a position.” Is doing that well-reasoned?
Demonstrating in painstaking detail that he has done so is the entire point of the first third of the post. Did you not find that convincing?
Convincing of what? I really feel like you think arguments are soldiers and that that is ALL that arguments are. I can’t do much to help that. I’m sorry, but I can’t. You need to observe more arguments.
That Steven Pinker is unreliable when it comes to evolutionary psychology because he has misrepresented his evidence on at least one occasion. That doesn’t make his conclusions wrong, but it should make them more suspect.
I really don’t think that. I think it’s a good idea to be wary of Steven Pinker because I don’t expect that he will always point towards the truth.
Would it be better if I phrased my question as “Have you updated your opinion of Pinker’s credibility based on the knowledge that he misrepresented evidence on one occasion? If not, was it because you believe that the evidence demonstrating that he did so was inaccurate?”
This is a great example of an error that, now that I think about it, is extremely common. I should write a longer article about this, since I haven’t seen any complete explanations of the phenomenon. I hope you don’t mind my using your comment as an example; you are definitely not the only person who makes this mistake, your comment just the one that happened to be the one that crystallized some concepts in my mind.
Your statement is literally true, but it’s also enough inferential steps away from the statements we’re likely to care about, that it’s effectively meaningless. It is four inferential steps away. (1) Even supposing that Pinker was maximally wrong (his opinions were generated by throwing dice), any given conclusion is still at least as likely to be true as it would be if Pinker hadn’t said anything at all. (2) Pinker is not maximally wrong; even if he did have a pattern of misrepresenting evidence, there would still only be a small chance of him having done it on any particular instance. (3) It’s not established that there is a pattern; there’s just one example, which could be atypical. And (4) it’s not even necessarily established that there is such an example; I might disagree with your interpretation.
As discussions proceed, they get further and further away from the original conclusion. And on the topic of human sexuality, they start at a longer than normal distance to begin with, because there are so many nasty traps for studies to fall into (especially people lying on surveys and to researchers). I think that on an unconscious level, people track the average inferential distance, and respond negatively if it’s out of range.
It’s not that he does no better than chance, exactly. It’s that he promotes hypotheses in an unjustified way. And so for any given hypothesis that you hear, you should ideally discount its probability by the difference between the number of bits of evidence the hypothesis should need and the number of bits of evidence you expect him to actually have. If that’s a big difference, then he’s not worth paying attention to.
Your other points are solid, though. I definitely have a habit of underestimating inferential distance.
This sounds like a better argument than I have seen from you earlier, indicating to me that you do have the ability to make such arguments when you are inclined, at least some of the time.
With respect to its content, yes, I think this is a reasonable conclusion regarding how one should respond to Pinker’s conclusions. However, as I noted, my observation is that Pinker argues well. When you have a good argument in front of you its primarily the argument and not the evidential value of the opinion which you have to confront.
I think that I’m placing more emphasis on whether his arguments are sound, and you’re more concerned with their validity.
Yep. When evaluating a dispute between someone and Pinker I started by saying that I could take it for granted that his arguments are sound, but that it seemed to me that their arguments might be worth-while and scholarly. (I could also have said fact-filled and valuable, especially if not taken literally, as a source of added validity).
In practice, arguments that aren’t sound are practically never fully valid, but they are frequently a valuable complement to sound arguments as part of how validity of belief is achieved.
Did you mean valid here? Its the soundness of Pinker’s arguments that I am claiming to have undermined, so I’m kind of confused.
I think that my internal argument-evaluation-algorithm focuses most of its effort on premises and treats the subsequent reasoning as a mostly mechanical and straightforward process. Getting enough and good enough data together instinctively seems like the greater obstacle to me, possibly because I do fairly well with formal logic.
Yep. My mistake.
Ah. The problem with treating reasoning as mechanical is that almost no-one actually does reasoning reliably enough for that to work. If they did, the quality of public debate would be completely different.
Absolutely agreed. I know I don’t reason as reliably as I seem to generally expect. This is my bug, not the world’s.