No Wonder So Many People are Depressed
Moderator: bbmods
-
Dave The Man
- Posts: 45001
- Joined: Fri Apr 01, 2005 2:04 pm
- Location: Someville, Victoria, Australia
- Has liked: 2 times
- Been liked: 21 times
- Contact:
-
Dave The Man
- Posts: 45001
- Joined: Fri Apr 01, 2005 2:04 pm
- Location: Someville, Victoria, Australia
- Has liked: 2 times
- Been liked: 21 times
- Contact:
-
stui magpie
- Posts: 54843
- Joined: Tue May 03, 2005 10:10 am
- Location: In flagrante delicto
- Has liked: 132 times
- Been liked: 168 times
To me, the flaws raised in reviews of Pinker's books (quoted above and in posts to follow) really undermine his credibility. One might have hoped that he'd have learnt from the mistakes of the earlier book, but instead the impression is that he simply does not care and has wilfully repeated and amplified them in his latest book, perhaps seduced by the undeserved influence popular writing has given him.
Kolbert above questions his "fishy math". Elsewhere, Ada Palmer notes (without any concern) that "charts are easy to niggle at: a chart of declining war deaths per annum beginning in 1945 might look very different had it started in 1600". I think both are valid concerns. Yes, if his charts really mean something, they should not be dependent on some magic starting point like 1945 (and it is the author's responsibility to show the reader that it indeed is not dependent on that). From what Kolbert writes, Pinker normalized the death toll by population but not by timespan, when surely the relevant metric is the rate of deaths per capita per year. And those are just the non-technical problems...
Kolbert above questions his "fishy math". Elsewhere, Ada Palmer notes (without any concern) that "charts are easy to niggle at: a chart of declining war deaths per annum beginning in 1945 might look very different had it started in 1600". I think both are valid concerns. Yes, if his charts really mean something, they should not be dependent on some magic starting point like 1945 (and it is the author's responsibility to show the reader that it indeed is not dependent on that). From what Kolbert writes, Pinker normalized the death toll by population but not by timespan, when surely the relevant metric is the rate of deaths per capita per year. And those are just the non-technical problems...
David, I'm not critiquing the literary style or structure. I'm critiquing the scientific reasoning and use of data (statistical methodology). In this regard, what is claimed in the books is not in dispute. Whether those are correct claims to make from a scientific perspective is in question. If people were claiming direct quotes and specific graphs were in the books, and he were claiming he was misquoted or no such graphs were in the books, then the terms of debate would be totally different and we'd need copies of the books just to begin the argument.
For example, it is a fact that he has chosen to look at war deaths (as a proportion of population) per event and not per time. That is not in dispute. (I know that Kolbert has not simply misunderstood Pinker or made a false claim, and that I have not simply misunderstood her, because Pinker's response to her shows it. I was leaving discussion of his response for a future post. You will no doubt be pleased to learn that I have read his response, though you may be less pleased to learn that I think it's rubbish.) The question is whether that is the logically correct or practically reliable thing to do. At the point of that question's raising, the book is neither necessary nor sufficient to address it.
Is that clear?
It's also worth questioning whether the actual claims requiring domain-specific expertise that he makes are correct. This raises an interesting question, namely, what the best way to do this is. If you think that merely reading the book is a good way to do this, then you must assume the reader is an expert in all of those domains (or at least one, in which case he can at least evaluate the relevant subset of claims). If one is not an expert, it's actually more useful to heed the words of reviewers who are experts in those areas. I am not an historian specializing in the Enlightenment, for example. If there is a professor who is such an expert, his review is more useful for evaluating Pinker's historical claims concerning the Enlightenment than my naive reading of Pinker's claims. If they disagree, who would you be more inclined to believe? The expert or the dilettante? The answer may partly depend on how much you respect that field of study and how much you respect the dilettante.
[David, do you actually own a copy of the earlier book? In some sense I hope you do not, because Pinker does not deserve the book sales. But it would be useful if you do, because Pinker in one of his counterattacks does refer to some pages in it.
And also, David, with or without your own copy, you have read the earlier book, though not the latter, so if you think Kolbert et al. have misrepresented him more generally (surely she has not directly misquoted him), you should comment on it.]
For example, it is a fact that he has chosen to look at war deaths (as a proportion of population) per event and not per time. That is not in dispute. (I know that Kolbert has not simply misunderstood Pinker or made a false claim, and that I have not simply misunderstood her, because Pinker's response to her shows it. I was leaving discussion of his response for a future post. You will no doubt be pleased to learn that I have read his response, though you may be less pleased to learn that I think it's rubbish.) The question is whether that is the logically correct or practically reliable thing to do. At the point of that question's raising, the book is neither necessary nor sufficient to address it.
Is that clear?
It's also worth questioning whether the actual claims requiring domain-specific expertise that he makes are correct. This raises an interesting question, namely, what the best way to do this is. If you think that merely reading the book is a good way to do this, then you must assume the reader is an expert in all of those domains (or at least one, in which case he can at least evaluate the relevant subset of claims). If one is not an expert, it's actually more useful to heed the words of reviewers who are experts in those areas. I am not an historian specializing in the Enlightenment, for example. If there is a professor who is such an expert, his review is more useful for evaluating Pinker's historical claims concerning the Enlightenment than my naive reading of Pinker's claims. If they disagree, who would you be more inclined to believe? The expert or the dilettante? The answer may partly depend on how much you respect that field of study and how much you respect the dilettante.
[David, do you actually own a copy of the earlier book? In some sense I hope you do not, because Pinker does not deserve the book sales. But it would be useful if you do, because Pinker in one of his counterattacks does refer to some pages in it.
And also, David, with or without your own copy, you have read the earlier book, though not the latter, so if you think Kolbert et al. have misrepresented him more generally (surely she has not directly misquoted him), you should comment on it.]