Re: Books on Intelligence

1

seems fatal to the possibility of belief that IQ measures something immutable or outside culture.

I really, really wish this were true. IME the IQ discussion is always, always about a belief first and a rationalization afterwards. I'm just as guilty.


Posted by: Witt | Link to this comment | 06-21-09 4:29 PM
horizontal rule
2

Is the goal to trap Shearer in this thread and lock him away here forever?


Posted by: essear | Link to this comment | 06-21-09 5:00 PM
horizontal rule
3

This is all well and good, but shouldn't the post have been titled "Mental Whateverness"?


Posted by: Di Kotimy | Link to this comment | 06-21-09 5:03 PM
horizontal rule
4

Now having read Cosma's review (feels odd to call the man by his first name, but whatever), two sections stand out. First:

Since IQ test questions are selected to be positively correlated, the appearance of g in factor analyses just means that none of the calculations was botched.

Stunningly obvious, and yet I never thought of it this way. Of course. Duh. A fabulous cocktail-party dismissal.

Similarly, what Raven's matrices test is not how well you can "educe relations", but how well you can find the patterns Raven liked -- personally, I can solve such puzzles only by guessing what was going through the test-maker's mind. In either case, to even begin to respond appropriately requires certain culturally-transmitted cognitive tools, and the motivation to use them on command.

A useful explanation for those of us who find analogies of this sort to be devlishly frustrating.


Posted by: Witt | Link to this comment | 06-21-09 5:10 PM
horizontal rule
5

Devilishly, even.


Posted by: Witt | Link to this comment | 06-21-09 5:11 PM
horizontal rule
6

"Peasants think like this, productive participants in modern bureaucratic society think like that."


Posted by: JP Stormcrow | Link to this comment | 06-21-09 5:47 PM
horizontal rule
7

It should certainly put paid to the idea that g measures something with a specific global or nearly global neurophysiological correlate. Which isn't to say that correlate doesn't exist, just that it isn't a correlate to g.

I keep thinking all of these questions were settled years ago, but silly me.

Tertiarily, none of this necessarily means that IQ tests aren't likely to correlate positively with outcomes; it would in fact be surprising if it didn't.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:00 PM
horizontal rule
8

none of this necessarily means that IQ tests aren't likely to correlate positively with outcomes; in the kind of society that produced the test.


Posted by: rob helpy-chalk | Link to this comment | 06-21-09 6:01 PM
horizontal rule
9

6: "Peasants are from Pluto, Bureaucrats are from Uranus"


Posted by: Di Kotimy | Link to this comment | 06-21-09 6:03 PM
horizontal rule
10

personally, I can solve such puzzles only by guessing what was going through the test-maker's mind.

This is true of every test ever devised. I tell my students this every semester: "You will do well on the test if you learn to think like me. While in college, if take classes from people who think a variety of different ways, you will learn to think a variety of different ways. If you are uncertain about what you will be doing in the future, you should cultivate a wide variety of thinking styles, because you don't know what will come in handy."


Posted by: rob helpy-chalk | Link to this comment | 06-21-09 6:05 PM
horizontal rule
11

I wish we could trap Cosma in this thread and I could bug him forever. Bayesians have been filling my head with all sorts of nonsense and it's just terribly seductive.

On The Other Hand my poor friend just finished a degree heavily featuring analytic philosophy and was excited to talk to me about the Doomsday Argument, which seems like as wrong an application of Bayes Theorem as could be constructed.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:06 PM
horizontal rule
12

8: I assumed that was obvious, yeah.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:06 PM
horizontal rule
13

"when the last bureaucrat is strangled with the guts of the last abstract reasoner".


Posted by: JP Stormcrow | Link to this comment | 06-21-09 6:09 PM
horizontal rule
14

On The Other Hand my poor friend just finished a degree heavily featuring analytic philosophy and was excited to talk to me about the Doomsday Argument, which seems like as wrong an application of Bayes Theorem as could be constructed.

I once sat through a whole day of talks about Boltzmann brains and related issues. Barely escaped with my sanity intact (and without blowing a gasket and screaming at people that they're all idiots).


Posted by: essear | Link to this comment | 06-21-09 6:10 PM
horizontal rule
15

... and then one inexorably has people leading themselves into terribly serious and earnest discussions of how it can be possible that they are not Chinese.


Posted by: essear | Link to this comment | 06-21-09 6:13 PM
horizontal rule
16

15: assume a uniform distribution over the porbability space containing having been hit with this chair.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:15 PM
horizontal rule
17

"porbability". Stupid phone.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:16 PM
horizontal rule
18

I'm here for you, Sifu. I'm ready to fight the Bayesians unto my last breath.


Posted by: Walt Someguy | Link to this comment | 06-21-09 6:22 PM
horizontal rule
19

15: That must get terribly confusing when some of the participants in the conversation are.

11: What is a Bayesian, in lay speak? I know what Bayes' Theorem is, and at the level I understand it at, it's fairly simple and useful for understanding things like test results with a given rate of false positives and so forth. But I do see discussions of 'Bayesianism' implying that it leads to systematic goofiness, and I don't really know what that's about.


Posted by: LizardBreath | Link to this comment | 06-21-09 6:22 PM
horizontal rule
20

There's a spectrum of goofiness, but in its purest form, Bayesianism is the claim that the only correct way to reason in the face of a lack of knowledge is by taking whatever it is that you don't know, and choosing a probability distribution for it.


Posted by: Walt Someguy | Link to this comment | 06-21-09 6:27 PM
horizontal rule
21
The step that converts N into an extinction time depends upon a finite human lifespan. If immortality becomes common, and the birth rate drops to zero, N will never be reached.

Sentences like this are why I love Wikipedia.


Posted by: mrh | Link to this comment | 06-21-09 6:28 PM
horizontal rule
22

19.2: oh, buh. I'll try, but will probably get it wrong. A Bayesian believes that it is possible to gain more information from a given set of data by assuming a prior distribution that is likely to be similar to the actual distribution of the data. A Bayesian further thinks this is true even if the prior distribution turns out to be pretty far off from the eventual final (posterior) distribution. A frequentist (as I understand it) thinks that this is likely to fuck you up, and that instead of working with the most likely posterior probability distribution out of a continuous range of distributions is fooling yourself into thinking you know something you don't, and your statistical inference should end at saying "given the frequency of this event in the past, it is this unlikely that it will occur on the next (or any future) trial."

I'm sure I fucked that up somewhere. But then, I do like me some Bayesian inference.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:30 PM
horizontal rule
23

I wish I had written what Walt did.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:31 PM
horizontal rule
24

instead of working with the most likely posterior probability distribution out of a continuous range of distributions is fooling yourself

Wha?


Posted by: nosflow | Link to this comment | 06-21-09 6:32 PM
horizontal rule
25

Bayesian thinking is what makes the authorities in '50s and '60s horror/sci-fi movies and The Twilight Zone look like idiots so much of the time.


Posted by: JP Stormcrow | Link to this comment | 06-21-09 6:36 PM
horizontal rule
26

24: change the first "of" to "that", if you would.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:37 PM
horizontal rule
27

I wish I understood Cosma's objections better, because Bayesian inference seems like a swimmingly great practical technique most of the time.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:39 PM
horizontal rule
28

2

Sailer's review of Flynn's book.


Posted by: James B. Shearer | Link to this comment | 06-21-09 6:42 PM
horizontal rule
29

22: A Bayesian further thinks this is true even if the prior distribution turns out to be pretty far off from the eventual final (posterior) distribution.

Ah. That does make sense, and seems as if it would lead to systematic goofiness.


Posted by: LizardBreath | Link to this comment | 06-21-09 6:42 PM
horizontal rule
30

So for example, if you have a coin, and you want to figure out the odds it will come up heads, you can't just start flipping it and seeing. You start by coming up with a probability distribution between 0 and 1 that represents your prior belief of the chances of it coming up heads. Your prior belief would be something like: you think there's a 5% chance that the coin will come up heads 33% of the time, a 90% chance that the coin will come up heads 50% of the time, and a 5% chance the coin will come up heads 66% of the time. (That's not a prior distribution anyone would actually choose, it's just an example.) Then, each time you flip the coin, you update that distribution by Bayes' theorem.


Posted by: Walt Someguy | Link to this comment | 06-21-09 6:43 PM
horizontal rule
31

29: surprisingly seldom, if the prior is chosen intelligently and the data is basically well-formed. Determining whether those two criteria hold is sort the art of it.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:47 PM
horizontal rule
32

The coin-flipping example that DeLong works through at the end of this post ("Cosma Shalizi Takes Me to Probability School. Or Is It Philosophy School?") is a good illustration.


Posted by: JP Stormcrow | Link to this comment | 06-21-09 6:49 PM
horizontal rule
33

Determining whether those two criteria hold is sort the art of it.

Well, yeah. The systematic goofiness possible would be where you didn't know enough about the data going into the problem to choose a prior intelligently, or at least to know that you had.


Posted by: LizardBreath | Link to this comment | 06-21-09 6:50 PM
horizontal rule
34

33: as seen in the Doomsday Argument, yes. What's interesting about Bayesian inference, though (and this is what people get carried away with, I think), is that even if the prior is totally, totally wrong, the posterior will converge (given enough data, and given that certain conditions which may or may not be common in the real world hold) on the true distribution.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 6:54 PM
horizontal rule
35

A frequentist thinks that when you say that a coin has a 50% of coming up heads that she means that if she flips the coin enough times, it will come up heads half the time. A Bayesian thinks this statement is nonsensical.

Sifu, the arguments you're giving for Bayesian inference are the frequentist arguments for Bayesian inference. The argument is not that Bayesianism is not a good method for inference, but whether every single other method of inference is illegitimate. In statistics, if a method eventually converges on the right answer given enough data, then a frequentist considers this a good argument for using that method. A Bayesian thinks that this is irrelevant, and only Bayesian methods are legitimate. Bayesianism is a totalizing philosophy that aims to eliminate all competitors.


Posted by: Walt Someguy | Link to this comment | 06-21-09 6:56 PM
horizontal rule
36

34 is the essential frequentist argument in favor of any statistical method. "Converge giving enough data" is a frequentist notion.


Posted by: Walt Someguy | Link to this comment | 06-21-09 6:57 PM
horizontal rule
37

35.2: because not using Bayesian methods is leaving perfectly good inference on the table, right? That sorta makes sense to me. Do I have to get all the way to universal use of uninformative priors to be a Bayesian. Like, if you can build a multi-level model, you're already embedding assumptions, and that doesn't seem scary.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 7:00 PM
horizontal rule
38

Argh this is a terrible thread to be participating in from my phone.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 7:01 PM
horizontal rule
39

All I know about Bayesian models is that it's pretty likely philosophers mess them up.


Posted by: Cala | Link to this comment | 06-21-09 7:01 PM
horizontal rule
40

39: Statement based on priors or data?


Posted by: JP Stormcrow | Link to this comment | 06-21-09 7:05 PM
horizontal rule
41

Pfffffft.


Posted by: Cala | Link to this comment | 06-21-09 7:11 PM
horizontal rule
42

It's also not true that given the prior you will always converge on the true posterior. Your prior has to basically put a positive density on the true model. Given that the world is a pretty complicated place, it seems pretty likely that we put a zero density on the true model all of the time. In any practical application, we need to iteratively improve our models until we find a model that embeds the true model. Bayesianism rules that kind of process out. The only valid method of inference is to start with a prior, and then move forward using Bayes' rule. If you start tinkering with your model after the fact, then you are no longer being a Bayesian.


Posted by: Walt Someguy | Link to this comment | 06-21-09 7:15 PM
horizontal rule
43

I didn't understand what you're asking in 37. You can use multilevel models in either Bayesian setting or a frequentist setting. In the frequentist setting, you would estimate the outermost set of hyperparameters from the data, rather than imposing a prior. In this particular case, this is equivalent to imposing an uninformative prior for the outermost set of hyperparameters, but that's not a universal truth about all frequentist procedures.


Posted by: Walt Someguy | Link to this comment | 06-21-09 7:18 PM
horizontal rule
44

42.last: even if you're tinkering with the model algorithmically?

43.last: but why not impose a prior if you think it might make the model mire efficient? I guess, in asking "why not", I'm not being truly Bayesian?


Posted by: Sifu Tweety | Link to this comment | 06-21-09 7:22 PM
horizontal rule
45

44.1. Yes. Only Bayes rule is allowed.

44.2 Basically.

Do you have in mind applications like spam filtering, say? Nobody objects to using Bayesian inference there.

The practical objection to Bayesianism is that by using a prior you're sticking your thumb on the scales. It's true that (assuming you put a positive weight on the true model) you will eventually converge on the right estimate, but you can choose the prior so that that process takes as long as you want. If you a priori think it's unlikely that the temperature hits 90 degrees in the summer, then you can put a low enough weight on that so that it will takes thousands of years of data to convince you that the weather really hits 90 degrees. At that point, a frequentist would say, "well, what does the data say." A Bayesian thinks that an argument of that form is the stupidest one imaginable.


Posted by: Walt Someguy | Link to this comment | 06-21-09 7:35 PM
horizontal rule
46

I'm largely thinking of practical machine learning approaches, yes. Maybe that's the disconnect.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 7:37 PM
horizontal rule
47

45.last: but that would be stupid.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 7:38 PM
horizontal rule
48

47: Which is stupid? The fate of your eternal soul hangs in the balance.


Posted by: Walt Someguy | Link to this comment | 06-21-09 7:40 PM
horizontal rule
49

48: choosing a prior so that it would take such a vast amount of data to inform it.

HELLBOUND!


Posted by: Sifu Tweety | Link to this comment | 06-21-09 7:46 PM
horizontal rule
50

Flynn, by the way, is a regular parliamentary candidate for the left-wing Alliance Party in New Zealand (they were in parliament two terms ago, but he's never been a serious MP prospect), and seems generally a good guy.


Posted by: Keir | Link to this comment | 06-21-09 7:49 PM
horizontal rule
51

No, you're headed towards a serene heaven where angels perform an experiment infinitely many times and then report the answer back to you.


Posted by: Walt Someguy | Link to this comment | 06-21-09 7:57 PM
horizontal rule
52

42: "It's also not true that given the prior you will always converge on the true posterior."

Which is why you should have sex with the lights on.


Posted by: Moby Hick | Link to this comment | 06-21-09 8:07 PM
horizontal rule
53

I sort of wish Emerson were here to hear how the onetime analytic philosophers may well not understand any of this at all, given that it seems to be about convergence on hard knowledge, which is not something philosophy necessarily deals with. Probability? But maybe the new-skool analytic philosophers do that.


Posted by: parsimon | Link to this comment | 06-21-09 8:15 PM
horizontal rule
54

Probability? But maybe the new-skool analytic philosophers do that.

I think the point is that they may try to do that, but do so poorly.


Posted by: Blume | Link to this comment | 06-21-09 8:21 PM
horizontal rule
55

54: Oh. I don't know analytic philosophy as it's now practiced. If it's concerned with narrowing down what we might reliably say we know, or at least have evidence for as calculated in a probabilistic manner, it doesn't resemble philosophy as I know it.


Posted by: parsimon | Link to this comment | 06-21-09 8:37 PM
horizontal rule
56

54: The point is that I suspect some of us do math badly.


Posted by: Cala | Link to this comment | 06-21-09 8:40 PM
horizontal rule
57

I've also now read the deLong post linked in 32, and get this a bit more.


Posted by: parsimon | Link to this comment | 06-21-09 8:48 PM
horizontal rule
58

54: The point is that I suspect some of us do math badly.

Actually according to a friend of mine the problem with philosophical uses of e.g. Sleeping Beauty isn't the mathematical prowess of the people involved but that they're terrifically misconstruing its point.


Posted by: nosflow | Link to this comment | 06-21-09 8:54 PM
horizontal rule
59

I don't see what the Doomsday argument has to do with Bayesian inference. I thought the troubling part of it was the use of the anthropic principle--the thought that we can get useful data out of the fact that we were born when we were, i.e. the idea that we're more likely to be born in a time with more people, with average characteristics, than not.

Plus I wish Eliezer Yudkowsky would come and skool all you Bayesian haterz.


Posted by: paranoid android | Link to this comment | 06-21-09 8:55 PM
horizontal rule
60

I proofread a dissertation that was all about Sleeping Beauty, the Doomsday argument, etc. The author was a thirder, I think, w/r/t the first-mentioned. I was convinced.


Posted by: nosflow | Link to this comment | 06-21-09 8:56 PM
horizontal rule
61

59: the wikipedia article is informative on the Bayesian aspects.

58: that strikes me as thoughtful. Although I dunno wtf Sleeping Beauty is in this context.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 8:58 PM
horizontal rule
62

42

... Given that the world is a pretty complicated place, it seems pretty likely that we put a zero density on the true model all of the time. ...

The world being complicated is a reason not to assign zero density to things.

Of course it is human nature to evaluate evidence within an unquestioned framework (or worldview) which leads to problems when the framework is wrong. But this just means humans can be bad at choosing priors.

... Bayesianism rules that kind of process out. ...

I don't really see this. Assign part of the probability to a general "not acceptably approximated by any of the other buckets" bucket.


Posted by: James B. Shearer | Link to this comment | 06-21-09 9:03 PM
horizontal rule
63

Yeah, what 62 says.


Posted by: paranoid android | Link to this comment | 06-21-09 9:07 PM
horizontal rule
64

From an article by Adam Elga in Analysis:

The Sleeping Beauty problem: Some researchers are going to put you to sleep. During the two days that your sleep will last, they will briefly wake you up either once or twice, depending on the toss of a fair coin (Heads: once; Tails: twice). After each waking, they will put you to back to sleep with a drug that makes you forget that waking. When you are first awakened, to what degree ought you believe that the outcome of the coin toss is Heads?

One half, or one third?


Posted by: nosflow | Link to this comment | 06-21-09 9:08 PM
horizontal rule
65

62, 63: but both of you are asserting the much weaker form of enthusiasm for Bayesian inference, from what I can tell, rather than the much stronger prescriptivist form of which Walt speaks.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 9:09 PM
horizontal rule
66

I've never seen that form personally.


Posted by: paranoid android | Link to this comment | 06-21-09 9:10 PM
horizontal rule
67

64: One half.

The tense of "to what degree ought you believe that the outcome of the coin toss is Heads?" is odd: to what degree ought you believe that the outcome was Heads.

In any case, the choice is between: is this the first time I've woken up, or the second? (i.e. Heads, or Tails?) There is no way to determine which it is, and absent further information weighting either way, 50-50.

This is interesting. Thanks. I probably really don't get probabilities.


Posted by: parsimon | Link to this comment | 06-21-09 9:28 PM
horizontal rule
68

They wake you up twice if it's tails. So if it's tails, this could be either the first or the second time you wake up—you don't know.


Posted by: nosflow | Link to this comment | 06-21-09 9:30 PM
horizontal rule
69

Not quite. Or rather, quite, but it's missing the point that makes it interesting. If the coin is tails, you'll be woken twice; if it's tails you're woken once. So there are three possible wakings, and you have no knowledge of which is which - but only one of them involves the coin toss coming up heads. So the probability is one third, the argument goes.

Arguments about it generally are missing the point, however.


Posted by: wispa | Link to this comment | 06-21-09 9:35 PM
horizontal rule
70

5/12


Posted by: text | Link to this comment | 06-21-09 9:37 PM
horizontal rule
71

If the issue is normative, then what's the point of illustrating it with a story as contrived and as boring as that one? I mean, with trolley problems and the like, there are at least interesting (hypothetical) results for making one choice or another.


Posted by: Michael Vanderwheel, B.A. | Link to this comment | 06-21-09 9:37 PM
horizontal rule
72

64

One half, or one third?

If you knew it was the first time you were woken up it would be one half but since you don't it is one third. Assuming of course you have been told the protocol and have good reason to trust what you have been told.


Posted by: James B. Shearer | Link to this comment | 06-21-09 9:39 PM
horizontal rule
73

Actually, there are standard constructions ("Diaconis-Freedman priors") under which Bayesian updating provably converges to false answers. In fact, establishing conditions under which Bayesian updating converges at all — never mind converges on the truth — is rather hard. So far as I know, all the sufficient conditions for Bayesian consistency also entail the existence of consistent non-Bayesian procedures for the same problems. As for 62's suggestion to "Assign part of the probability to a general 'not acceptably approximated by any of the other buckets' bucket", this doesn't work. All computable priors turn out to be pretty dogmatic (a point discussed in detail in Earman's Bayes or Bust?). Even if we wish that away, you don't just need a prior distribution over hypotheses, you also need a likelihood function. What is the likelihood of the data under the hypothesis "none of the above"? Depending on how you answer, the catch-all term either makes basically no difference, or it dominates the posterior distribution.

A strong conviction that Bayesianism is the one true normative theory of reasoning under uncertainty is extremely common in statistics, philosophy and economics departments; to say nothing of crazy people.

(I haven't read Nisbett's new book, but I admire his work, and am waiting for our library to get its copy.)


Posted by: Cosma | Link to this comment | 06-21-09 9:40 PM
horizontal rule
74

71: seriously, at least in the Monty Hall problem you win a goat.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 9:41 PM
horizontal rule
75

68: Yeah, I got that. Since you don't know if it's the first or second time you've woken up, what you ought to believe about the outcome of the coin toss is the same as before the coin was tossed: 50/50.

There's a lot of weight placed here on what you ought to believe, which is a little weird to me. Is it what you're justified in believing? Because in this Sleeping Beauty scenario, you're not justified in believing anything beyond a one-half probability of Heads or Tails.

Am I just being stubborn here? I'm getting that once the coin has been tossed, the deed is done and in fact one single thing happened; now we're just guessing at how it may have gone down, given certain limited possibilities for that. I really am not used to thinking in probabilistic terms.


Posted by: parsimon | Link to this comment | 06-21-09 9:41 PM
horizontal rule
76

There are all kinds of neutral or uninformative priors Bayesians can use -- uniform priors and so on. You can also check for robustness to different priors. Bayesianism never seemed kooky to me, frequentism always seemed kookier in a way, because so many processes are simply not repeatable.

I think philosophers do a lot of stuff with this now.


Posted by: PGD | Link to this comment | 06-21-09 9:41 PM
horizontal rule
77

71

If the issue is normative, then what's the point of illustrating it with a story as contrived and as boring as that one? ...

Simple cases are easier to analyze.


Posted by: James B. Shearer | Link to this comment | 06-21-09 9:43 PM
horizontal rule
78

28: Sailer takes the proper cultural imperialist view of the Flynn effect -- bureaucrats saturated in abstractions *really are* smarter than those untutored savages and peasants!


Posted by: PGD | Link to this comment | 06-21-09 9:45 PM
horizontal rule
79

73: can't some of these problems be addressed with more model complexity? And aren't you building assumptions into the data anyhow even in a frequentist model? How does this all apply to common-sense (as opposed to scientific) reasoning? Don't Bayesian techniques buy you efficiency?


Posted by: Sifu Tweety | Link to this comment | 06-21-09 9:46 PM
horizontal rule
80

75: Yeah, that's generally what the argument is about, and no, you're not being stubborn. They're both right. That's the problem.

Your answer is perfectly valid, and just rests on different assumptions than the other. Unfortunately they're philosophical assumptions, rather than something we can measure or establish. Hence the misuse.


Posted by: wispa | Link to this comment | 06-21-09 9:49 PM
horizontal rule
81

Simple cases are easier to analyze.

It's so simple that it does away with the whole normative dimension of the problem---with what one ought to believe.

Nobody cares how many times they woke up the test subject. The test subject doesn't care how many times they woke up the test subject. Which defeats the whole purpose of illustrating the problem with any such story.


Posted by: Michael Vanderwheel, B.A. | Link to this comment | 06-21-09 9:53 PM
horizontal rule
82

75

: Yeah, I got that. Since you don't know if it's the first or second time you've woken up, what you ought to believe about the outcome of the coin toss is the same as before the coin was tossed: 50/50.

Not really. Suppose whenever the toss is heads you are not woken and whenever the toss is tails you are. The if you are woken up you know the toss was tails. Next suppose you are woken up half the time when the toss is heads but all the time when it is tails. Now if you are woken up you don't know the toss was tails but the odds are (2/3) it was. Similarly if you are woken up twice when it is tails but once when it is heads the odds when you are woken up favor tails.


Posted by: James B. Shearer | Link to this comment | 06-21-09 9:55 PM
horizontal rule
83

78: I see that - but doesn't this create a problem for Sailer? As you correctly note, his discussion of this is culturally imperialistic - not genetically deterministic. Why couldn't a black bureaucrat be as good as a white one, on average and given the same cultural inputs?


Posted by: politicalfootball | Link to this comment | 06-21-09 9:57 PM
horizontal rule
84

73: an't some of these problems be addressed with more model complexity?

We're already talking about infinite-dimensional non-parametric models.

And aren't you building assumptions into the data anyhow even in a frequentist model?

Sure. In frequentist non-parametrics, you generally have some sort of constraints or regularization which you gradually relax to encompass more and more possible models.

Don't Bayesian techniques buy you efficiency?

Only if the prior biases you towards the truth; if the bias goes the wrong way, it can screw you arbitrarily badly. (There are really no such things as unbiased, uninformative priors.)


Posted by: Cosma | Link to this comment | 06-21-09 9:58 PM
horizontal rule
85

79: Define "efficiency".


Posted by: Walt Someguy | Link to this comment | 06-21-09 9:59 PM
horizontal rule
86

When you are first awakened, the probability of heads is one-half. When you are awakened (possibly for the second time if the flip was tails), the probability of heads is one-third. It's perfectly sensible to give different answers to these two different questions.


Posted by: Commenter-in-exile | Link to this comment | 06-21-09 10:00 PM
horizontal rule
87

81: For my part, I'd rather be woken up fewer times. Got a busy day at the trolly station ahead of me.


Posted by: text | Link to this comment | 06-21-09 10:02 PM
horizontal rule
88

83: I'm sure he would say that only certain races have the genetic gifts necessary to absorb the cultural capital it takes to become a good abstract bureaucrat. So IQ is getting at the capacity to learn when exposed.

The thing that hopelessly conflates genetics and culture is that it take many generations, centuries, to absorb a culture. People always think of cultural learning as easy and quick, but it's not.


Posted by: PGD | Link to this comment | 06-21-09 10:03 PM
horizontal rule
89

I have a really hard time seeing the case for an answer other than one-half. But, I can be pretty dense sometimes.


Posted by: essear | Link to this comment | 06-21-09 10:05 PM
horizontal rule
90

84.last: so if you're talking about strong priors with a good basis (from previous experiments or whatever) then this stuff is less of a problem? Because to be honest I've found the uninformative prior Kool-Aid pretty freaky.

Although I guess then I would ask about building strong, informative priors iteratively -- sure you still have model bias, but that's going to fuck you on first principles even as a frequentist, right? I mean, you still have to pick meaningful hypotheses.


Posted by: Sifu Tweety | Link to this comment | 06-21-09 10:05 PM
horizontal rule
91

Mostly this is the kind of question that makes me want to respond: why should I believe there's an answer?


Posted by: essear | Link to this comment | 06-21-09 10:07 PM
horizontal rule
92

89: when you are woken up on Monday, you should say the probability is 1/2. But if you are woken up on Monday and don't know if it's Monday or Tuesday, you should say 1/3.


Posted by: Barbar | Link to this comment | 06-21-09 10:08 PM
horizontal rule
93

As far as I can tell from looking at actual experimental results, whether the experimentalist is a frequentist or a Bayesian is irrelevant for my understanding of the science they produce. It plays roughly the same role as their religious beliefs: I don't care very much if they have one, but if they're passionate about it, I start to worry about whether to trust them as rational thinkers.


Posted by: essear | Link to this comment | 06-21-09 10:09 PM
horizontal rule
94

86 is right. It's really the key point of difference, and in part it's a problem in how the scenario has to be framed. Which is why both answers can be right, depending on exactly what you're assuming is "justified", etc.

In short, it's a completely useless waste of time of a gedankenexperiment that can never have an answer.


Posted by: wispa | Link to this comment | 06-21-09 10:11 PM
horizontal rule
95

73

... As for 62's suggestion to "Assign part of the probability to a general 'not acceptably approximated by any of the other buckets' bucket", this doesn't work. All computable priors turn out to be pretty dogmatic (a point discussed in detail in Earman's Bayes or Bust?). Even if we wish that away, you don't just need a prior distribution over hypotheses, you also need a likelihood function. What is the likelihood of the data under the hypothesis "none of the above"? Depending on how you answer, the catch-all term either makes basically no difference, or it dominates the posterior distribution.

This is confusing. If what you are saying is, it is not enough to just have a catch-all term you need a way to update the probability assigned to it as you get more data then I think you have a point.


Posted by: James B. Shearer | Link to this comment | 06-21-09 10:11 PM
horizontal rule
96

86: When you are first awakened, the probability of heads is one-half. When you are awakened (possibly for the second time if the flip was tails), the probability of heads is one-third. It's perfectly sensible to give different answers to these two different questions.

It depends on the subjective stance you're inhabiting. As wakee, it's always the first time you're awakened. Therefore one-half probability. If, as wakee, you want to always consider the possibility that it's the second time you're waking, then it's one-third.

Sure, I see that. It's more interesting to try to parse the way in which these are two different questions. I'm a slow starter on this.


Posted by: parsimon | Link to this comment | 06-21-09 10:14 PM
horizontal rule
97

95: If what you are saying is, it is not enough to just have a catch-all term you need a way to update the probability assigned to it as you get more data

Yes, exactly. I apologize for my lack of clarity.


Posted by: Cosma | Link to this comment | 06-21-09 10:14 PM
horizontal rule
98

In short, it's a completely useless waste of time of a gedankenexperiment that can never have an answer.

It doesn't need to have an answer to be a useful thought experiment.


Posted by: Criminally Bulgur | Link to this comment | 06-21-09 10:17 PM
horizontal rule
99

That's true, but this isn't useful either.


Posted by: wispa | Link to this comment | 06-21-09 10:18 PM
horizontal rule
100

(Useful outside the limited context for which it was formulated).


Posted by: wispa | Link to this comment | 06-21-09 10:19 PM
horizontal rule
101

You find the building supervisor and offer to give him the barometer to tell him an interesting gedankenexperiment if he'll tell you the height of the building.


Posted by: JP Stormcrow | Link to this comment | 06-21-09 10:28 PM
horizontal rule
102

so if you're talking about strong priors with a good basis (from previous experiments or whatever) then this stuff is less of a problem?

Yes. There are plenty of applications where Bayesian approaches make a lot of practical sense, and you can safely ignore religious arguments and get useful work done....


Posted by: soup biscuit | Link to this comment | 06-21-09 10:30 PM
horizontal rule
103

This thought experiment just doesn't do anything for me. I've been trying.


Posted by: essear | Link to this comment | 06-21-09 10:30 PM
horizontal rule
104

100: I was wondering about that. But apparently there's much more to be said about Sleeping Beauty, and, uh, the Doomsday argument and so on.

I'm vaguely reminded of beginner philosophy courses in which an instructor asks whether the stopped watch that's nevertheless correct on the time twice a day gives the truth with respect to the time at those times; or rather, whether someone who reports the correct time based on such a stopped watch knows the time.

Huh! That's a puzzle. Let's talk about propositional attitudes!


Posted by: parsimon | Link to this comment | 06-21-09 10:31 PM
horizontal rule
105

In general, I worry when people try to insist that it makes sense to talk about a probability in a setting where it isn't entirely clear what probability means. Like, there's this very strong conviction that a lot of people have that if we live in a patch of a big multiverse where, essentially, the laws of physics are different in different places, we must be able to define a probability and see that it is in some sense likely that we live in the universe we live in. And there are extensive and highly technical arguments about which "measure" on the multiverse is the right one. I get lost at step zero: why should there be such a measure? Why should this probability make sense? We see that we live in a universe with some properties, but I have no idea why we should conclude that these properties are preferred by some well-defined canonical choice of measure. And people who write these highly technical papers and have these elaborate arguments about choices of measure have never given me a satisfactory answer to why I should think the whole endeavor makes any fucking sense at all.

The "Sleeping Beauty" thought experiment seems like a baby version with similar issues. I just do not get it.


Posted by: essear | Link to this comment | 06-21-09 10:36 PM
horizontal rule
106

Huh, I still have this dissertation.

I think I will not read it again.


Posted by: nosflow | Link to this comment | 06-21-09 10:42 PM
horizontal rule
107

Oh, and I was wrong; he's a halfer, not a thirder.


Posted by: nosflow | Link to this comment | 06-21-09 10:43 PM
horizontal rule
108

I won't read the whole thing again.


Posted by: nosflow | Link to this comment | 06-21-09 10:44 PM
horizontal rule
109

Wouldn't the Sleeping Beauty problem be clearer if people were explicit about their filtration (i.e., what events can be distinguished at different points in time in the thought experiment)?

I too have trouble seeing the 1/3 answer.


Posted by: BA | Link to this comment | 06-21-09 10:54 PM
horizontal rule
110

The Halfers and the Thirders were two short-lived branches of Shi'a Islam. Historians believe both groups ran into trouble largely because they refused to recognize the validity of at least one Imam.


Posted by: Otto von Bisquick | Link to this comment | 06-21-09 11:00 PM
horizontal rule
111

The reasoning behind the one-third answer runs like this: there are three possible awakenings. Two of them are when the coin was tails, and one is when it was heads. You cannot distinguish those situations, and don't know whether it is the first or second awakening.

Since you have no information on that, you have to assume each is equally likely. So the probability of a heads is 1/3, the same as the probability of the single awakening in which that is true.

For the one-half answer, the line of reasoning is "it's a coin-flip, dummy".


Posted by: wispa | Link to this comment | 06-21-09 11:02 PM
horizontal rule
112

I'm with the people who can't understand the 1/3 answer....eg, why this talk about "three possible wakings"? The conditions you've been told of in advance let you know there will always be at least one waking, and you have no way of knowing if you've been woken before, so the experience of being awakened gives you absolutely no more information than before the game started. Which means you're back to basic probability: a coin toss is a fifty-fifty proposition. This particular formulation problem just doesn't work for me--it seems obvious that, as the scenario is constructed, being awakened is not a datum, since you will ALWAYS be awakened.


Posted by: lurker | Link to this comment | 06-21-09 11:02 PM
horizontal rule
113

112: The idea is that an unknown awakening is twice as likely to occur in the tails universe.


Posted by: wispa | Link to this comment | 06-21-09 11:04 PM
horizontal rule
114

This is one instance in the category of "weird-ass probability arguments in physics", but it's one where I think I agree with what they say. They call it a Bayesian approach, but it doesn't seem deeply committed to any very strong form of Bayesianism. Basically it's calling bullshit on various versions of the "we are all Chinese" fallacy.

I thought I could make an argument that the 1/3-er approach to the Sleeping Beauty problem is falling prey to the same fallacy, but I don't think it's quite that.


Posted by: essear | Link to this comment | 06-21-09 11:06 PM
horizontal rule
115

109 and others who aren't seeing the one-third answer, wispa's 69 clarified it for me:

If the coin is tails, you'll be woken twice; if it's tails you're woken once. So there are three possible wakings, and you have no knowledge of which is which - but only one of them involves the coin toss coming up heads. So the probability is one third, the argument goes.

The idea that there are three possible wakings (not two) is key: there are three possible worlds that could obtain here upon your waking. One in which you've awakened once and only once (coin toss was Heads), one in which you've awakened for the first time and will be awakened again, and one in which you've awakened for the second time, though you don't know that (coin toss Tails in the latter two cases).

This is weirdo possible worlds stuff, in which you're supposed to be calculating the probability that you're in one or another possible world that might be the case, but all you have for data going in is past information -- in this case the knowledge about the ramifications of the coin toss -- no current or new information to add to the development of hypotheses about the the truth of the matter. So you're flying blind.


Posted by: parsimon | Link to this comment | 06-21-09 11:08 PM
horizontal rule
116

112: The idea is that an unknown awakening is twice as likely to occur in the tails universe.

But there has to be some statement of what the assumed information is, and how that information updates the probability. And I just don't see a good way to frame that. Suppose I flip a coin, and post one incoherent Unfogged comment if it's heads and two if it's tails. Now suppose someone tells you that I posted at least one incoherent Unfogged comment. What's the probability the coin was heads? Is there any sense in which this thought experiment is not isomorphic to the Sleeping Beauty one? (To be more obvious, suppose I post one comment if heads and ten billion if tails, and repeat.)


Posted by: essear | Link to this comment | 06-21-09 11:10 PM
horizontal rule
117

115 was partially pwned, or not.


Posted by: parsimon | Link to this comment | 06-21-09 11:11 PM
horizontal rule
118

112--But being awakened, unknowing, provides no info about what "universe" you're in--being awakened unknowing will always happen--and since you already know the coin flip rule..........I feel, with essear, that there's a fallacy here, but I know my own limitations too well to expect to expose it......I want to say: "there are three possible worlds, and the first is twice as likely as the second and third to be the case"


Posted by: lurker | Link to this comment | 06-21-09 11:16 PM
horizontal rule
119

Like I said, it's not a useful experiment. The answer you come up with doesn't tell you anything productive, even about yourself. The reasoning process doesn't help much either.

And anybody who ever brings it up as part of making a point about statistics or probability should be shot.


Posted by: wispa | Link to this comment | 06-21-09 11:17 PM
horizontal rule
120

Based on 116, I'm hoping the coin tossed before parsi posted 117 landed heads.


Posted by: M/tch M/lls | Link to this comment | 06-21-09 11:18 PM
horizontal rule
121

The answer to all probability questions is that Jackmormon should not bet her own money on understanding any of them.


Posted by: Jackmormon | Link to this comment | 06-21-09 11:22 PM
horizontal rule
122

Whose money should she bet?


Posted by: nosflow | Link to this comment | 06-21-09 11:26 PM
horizontal rule
123

I want to say: "there are three possible worlds, and the first is twice as likely as the second and third to be the case"

I think this is more or less right.


Posted by: essear | Link to this comment | 06-21-09 11:26 PM
horizontal rule
124

118: It doesn't provide any information, no, and it doesn't have to.

It's important that one of the worlds will turn into another where the toss is the same. The one-third argument says that matters; the one-half argument says it's just a coin no matter what.

It may help to understand the reasoning by considering the possibility of being awoken once versus a much larger number of times. If you're not seeing it, you're really not missing out on anything.


Posted by: wispa | Link to this comment | 06-21-09 11:26 PM
horizontal rule
125

Maybe, given the Flynn effect, we should just wait a hundred years until this shit is all totally obvious and the discussion is unnecessary.


Posted by: essear | Link to this comment | 06-21-09 11:32 PM
horizontal rule
126

So philosophers concern themselves with this sort of thing these days, huh? What are they saying about it, for god's sake? I assume they're worrying about what kinds of things lead us to draw conclusions we feel are justified, in the absence of full information? In other words, how we reach conclusions we feel comfortable acting on (good-enough knowledge). I begin to see how this treats the human mind like a computing machine, yet again.


Posted by: parsimon | Link to this comment | 06-21-09 11:33 PM
horizontal rule
127

122: Your mother's?


Posted by: M/tch M/lls | Link to this comment | 06-21-09 11:33 PM
horizontal rule
128

114--Hmmm; I read yr #69 again, and just don't see how the move from "three possible wakings" to "probability of 1/3" is anything more than a mistake, rather than being a necessary outcome of possible-world thinking--it may be that a different word problem would make me see it somehow, but--


Posted by: lurker | Link to this comment | 06-21-09 11:36 PM
horizontal rule
129

114 in last s/b 124....


Posted by: lurker | Link to this comment | 06-21-09 11:36 PM
horizontal rule
130

It is a mistake, within the 1/2 analysis. Within that, you get two universes at the point of the coin flip, and there are two relevant points in time within one of them and one in the other. The 1/3 analysis maintains that since you don't know which point you're at, they're all separate.

Suppose it's one time versus one hundred times. When you're awoken, it's vastly more likely (crucially: as far as you know) that you're in the tails universe, during one of your hundred awakenings. So from your perspective, across all possibilities it's far more likely that the coin was tails than heads, and it's a better bet for you to say the coin is tails than heads.

Focusing too much on "the first time" or that you're always awoken once is a mistake either way, but both answers are valid given their presuppositions. It's pretty easy to reason yourself into either one, at least for me.


Posted by: wispa | Link to this comment | 06-21-09 11:46 PM
horizontal rule
131

128: There's a more elaborate, but slippery, argument in the PDF neb linked in 64. I'm convinced it's either (a) a fallacy in some way I can't quite pin down or (b) defining probability in an unusual way, but maybe I should revisit it at some time that is not 2 AM when I have a higher probability of making sense of it.


Posted by: essear | Link to this comment | 06-21-09 11:48 PM
horizontal rule
132

So philosophers concern themselves with this sort of thing these days, huh?

Some do, yes. There is, I suspect, no one thing that philosophers these days concern themselves with.

They concerned themselves with this sort of thing in your day as well, some of them.


Posted by: nosflow | Link to this comment | 06-21-09 11:50 PM
horizontal rule
133

128

114--Hmmm; I read yr #69 again, and just don't see how the move from "three possible wakings" to "probability of 1/3" is anything more than a mistake, rather than being a necessary outcome of possible-world thinking--it may be that a different word problem would make me see it somehow, but--

Suppose you ask what is the probability this is the first time you were awakened?


Posted by: James B. Shearer | Link to this comment | 06-21-09 11:50 PM
horizontal rule
134

There's clearly a related question that is getting at the 1/3 answer. Suppose that, on waking, the person in question is always asked if they think the coin is heads or tails, and is given $10 if they are correct. Then they earn twice as much money by always guessing "tails" as when always guessing "heads". This is the sort of question that is sensitive to counting the number of waking events. But I can't make sense of the "probability" question as one that depends on the number of waking events.


Posted by: essear | Link to this comment | 06-21-09 11:56 PM
horizontal rule
135

Maybe I should stop by this Elga dude's office and harass him.


Posted by: essear | Link to this comment | 06-21-09 11:57 PM
horizontal rule
136

132: Okay, yes. It's not my cup of tea, seeming to suppose, as it does, that we are all, in actuality, engaged in sorting through the possible worlds we might conceivably inhabit, that we might winnow them down to the true one, the actual world.

It's late, and I'm being mean to the possible-worlds philosophers for no good reason. Sorry.


Posted by: parsimon | Link to this comment | 06-22-09 12:00 AM
horizontal rule
137

If you do, essear, report back.


Posted by: nosflow | Link to this comment | 06-22-09 12:01 AM
horizontal rule
138

111: Since you have no information on that, you have to assume each is equally likely. So the probability of a heads is 1/3

This seems incorrect. If the three possibilities and associated probabilities are
1) coin flip heads, first and only awakening, probability p
2) coin flip tails, first awakening, probability q
3) coin flip tails, second awakening, probability r

One of the three must hold so p+q+r=1.
However, you also know that p=1/2 and that q+r=1/2, because of the coin flip. Ignoring that seems incorrect.
How you split the probability between q and r is up to you (but then it doesn't really matter).


Posted by: BA | Link to this comment | 06-22-09 12:04 AM
horizontal rule
139

138: See the PDF neb linked in 64, which presents a simple argument that p=q and that q=r. Maybe I'm just being really dense, but I don't see an obvious objection, except that I think it's sneaking in an assumption that all "awakening" events count equally. There's something sketchy about ignoring the fact that scenarios 2) and 3) always happen in conjunction, which I think makes the question being answered by "1/3" a different question than the one about the probability of the coin flip, which seems to obviously have the answer "1/2".


Posted by: essear | Link to this comment | 06-22-09 12:07 AM
horizontal rule
140

133--75%
134--aha! (thinks.......)--ok, now I really need to go to bed.


Posted by: lurker | Link to this comment | 06-22-09 12:10 AM
horizontal rule
141

I think there are other versions of the scenario in which the question is something like "what probability should you assign to the proposition that this is your first awakening?".


Posted by: nosflow | Link to this comment | 06-22-09 12:11 AM
horizontal rule
142

It's not the probability of the coin being one way or the other; it's what you ought to think of the proposition that it's heads. They are different questions.

Like I said, both answers are right, they're just not useful, and you should be able to argue yourself into either one without too much trouble. So long as it stays in the philosophy department and isn't being used for some probabilistic line of argument I don't much care what the answer is claimed to be.


Posted by: wispa | Link to this comment | 06-22-09 12:12 AM
horizontal rule
143

140

133--75%

In that case you should believe the probability of heads is 3/8. Since if this is the first time you were awakened the probability is .5 but if this is the second time you were awakened the probability is zero.


Posted by: James B. Shearer | Link to this comment | 06-22-09 12:27 AM
horizontal rule
144

wispa, philosophy has already crept into your area of interest, steathily!


Posted by: parsimon | Link to this comment | 06-22-09 12:29 AM
horizontal rule
145

stealthily, that is.


Posted by: parsimon | Link to this comment | 06-22-09 12:30 AM
horizontal rule
146

Suppose I flip a coin, and post one incoherent Unfogged comment if it's heads and two if it's tails.

Let me see if I'm following along. Is this the theory of the three possible wankings?


Posted by: Wrongshore | Link to this comment | 06-22-09 12:31 AM
horizontal rule
147

146: There's only one wanking, but there are three possible cryings.


Posted by: M/tch M/lls | Link to this comment | 06-22-09 12:34 AM
horizontal rule
148

But is it the best of all possible wankings?


Posted by: Otto von Bisquick | Link to this comment | 06-22-09 12:42 AM
horizontal rule
149

The best of all possible wankings can only be accomplished by God at the start of the Universe, before anything wankable has had a chance to die.


Posted by: Lambent Cactus | Link to this comment | 06-22-09 12:51 AM
horizontal rule
150

We live in a fallen world


Posted by: Lambent Cactus | Link to this comment | 06-22-09 12:51 AM
horizontal rule
151

I know. It's unfortunate. Not that there's something wrong with philosophy in itself, it should just stay out of the mathematics department (with a small exception for the relevant parts of logic).

The one philosophy class I ever took was fun, but I still have no idea what it was about.


Posted by: wispa | Link to this comment | 06-22-09 1:05 AM
horizontal rule
152

I think I understand the issue. While the 1/3 argument is plausible, there's nothing in the laws of probability that forces it on you. Forget about the coin flip. Suppose you do an experiment where you wake somebody up on Monday, give them a drug that makes them forget, and then wake them up again on Tuesday. If you are the experimental subject, what are the odds after waking that it's Monday? It's certainly plausible that it's 50%, but this is intuition does not derive from probability.


Posted by: Walt Someguy | Link to this comment | 06-22-09 2:04 AM
horizontal rule
153

re: 152

Well, it's a problem that derives from the intersection of probability with epistemology. And, complicated by the fact that one some accounts of probability, the two don't cleanly separate.

I suppose the Sleeping Beauty problem throws that into some sort of relief.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 3:12 AM
horizontal rule
154

So Prince Charming is actually a group of researchers tossing a coin around in the lab?


Posted by: eb | Link to this comment | 06-22-09 3:29 AM
horizontal rule
155

In the article Elga says:

H1 HEADS and it is Monday.
T1 TAILS and it is Monday.
T2 TAILS and it is Tuesday.

and then comes to the conclusion:

Combining results, we have that P(H1) = P(T1) = P(T2). Since these credences sum to 1, P(H1) = 1/3.

I haven't checked the reasoning he leads up to this with, but is the problem with this last step that (the descriptions at) T1 and T2 aren't actually (describing) outcomes? Or at least, that if they are (describing) outcomes they're (describing) the same one, namely:

Tails, and I wake on Monday.
aka Tails, and I wake on Tuesday..

Isn't the problem then that T1 and T2 aren't mutually exclusive? So you can have
P(T1) = P(T2) = P(H1), with the probabilities of the outcomes in the appropriate sense summing to 1, and yet still have
P(T1) = P(T2) = P(H1) = 1/2 .

I dunno; this is making my head hurt.



Posted by: KingOfVegetables | Link to this comment | 06-22-09 3:33 AM
horizontal rule
156

Is the subject allowed to ask the researcher to open a door without a car behind it?


Posted by: OFE | Link to this comment | 06-22-09 3:55 AM
horizontal rule
157

Hybrid vigor shows up in the Flynn review.


Posted by: eb | Link to this comment | 06-22-09 3:58 AM
horizontal rule
158

If you play the Sleeping Beauty scenario over and over, the coin would have turned up heads 1/3 of the time Sleeping Beauty wakes up.

Heads (1 awakening)
Tails (2 awakenings)
Heads (1 awakening)
Tails (2 awakenings)
...

To see why this is a "paradox," it helps to remember what the question is. The question is "when Sleeping Beauty wakes up for the first time, what probability should she assign to the coin having come up heads?" This is ambiguous, because it's not clear if you should consider the fact that when she wakes up for the first time, she doesn't actually know if it's the first time or the second time.

If she knows it's the first time, the answer is obviously 1/2.
If she doesn't know, the answer is obviously 1/3.

Elga thinks that this contradicts something known as the "Reflection Principle," but he's obviously wrong.


Posted by: Commenter-in-exile | Link to this comment | 06-22-09 6:59 AM
horizontal rule
159

105 gets it exactly right.


Posted by: Sifu Tweety | Link to this comment | 06-22-09 7:25 AM
horizontal rule
160

I'm with the Halfers.


Posted by: heebie-geebie | Link to this comment | 06-22-09 7:26 AM
horizontal rule
161

134: But I can't make sense of the "probability" question as one that depends on the number of waking events.

The "probability" game you can set up here that is "sensible" would be something like the following. Let's say there are a hundred simultaneous sleeping experiments being performed (starting at random times). Waking events are reported and people make book on whether it is a "tails" or a "heads" waking. 1/3 is the easily understood probability here.

I think the whole thing is predicated on silly semantic muddling with regard to the meaning of the question. More in enxt comment.


Posted by: JP Stormcrow | Link to this comment | 06-22-09 7:38 AM
horizontal rule
162

So Prince Charming is actually a group of researchers tossing a coin around in the lab?

He was always a tosser to me.


Posted by: Criminally Bulgur | Link to this comment | 06-22-09 7:44 AM
horizontal rule
163

For a scenario that I think is both more "motivated" and illustrates the muddledness I propose the following nasty little scenario.

Sadistic, but "fair" regime X plays the following nasty game with prisoners. Prisoner is brought into a room and told that someone in the next room is flipping a coin. Tails, you get shot in the next minute and heads you don't. After a minute you are asked what are the odds that tails came up. To encourage the prisoner to think "correctly", if you get the right answer you're done, if not you play again. I would claim this is logically the same setup, other than that you can ignore the initial probability in figuring out the right answer, does not matter if it is 1/10 or 99/100.

All that the multiple awakenings allows is the insertion of the dodgy explanation that the answerer "Upon being awakened, you receive no new information (you knew all along that you would be awakened)." (from Elga's paper). This is BS, you have the information that you have been awakened. The "one" question is obscuring the two different questions being asked.


Posted by: JP Stormcrow | Link to this comment | 06-22-09 8:00 AM
horizontal rule
164

I used to feel vaguely second-rate because I never used (and barely understood) Bayesian inference. After this thread, I only feel second rate because I went to state schools. Thank you unfogged.


Posted by: Moby Hick | Link to this comment | 06-22-09 8:01 AM
horizontal rule
165

You lost me on the Monday Tuesday T1 and T2

In T1, Schwarzenegger is the bad cyborg. In T2, he's the good one.


Posted by: Criminally Bulgur | Link to this comment | 06-22-09 8:13 AM
horizontal rule
166

164: you should use Bayesian inference! I learned it at state schools, and look at me!

(messes self)


Posted by: Sifu Tweety | Link to this comment | 06-22-09 8:20 AM
horizontal rule
167

167: Just to be clear, getting rid of my Bayesian inferiority complex didn't create the state school one, just revealed that it was there all along. I don't want to get rid of the state-school inferiority complex because I'm afraid of what will be underneath that. Probably concerns about 'performing masculinity'.


Posted by: Moby Hick | Link to this comment | 06-22-09 8:23 AM
horizontal rule
168

For what it's worth, I find the 'performing $gender' locution both irritating and wrong-headed.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 8:25 AM
horizontal rule
169

163: Elga sneaks in the "no new information" argument by asking about when Sleeping Beauty is first awakened. She gets woken up on Monday no matter how the coin turns out.

At the start of the experiment, you had credence 1/2 in H. But you were also certain that upon being awakened on Monday you would have credence 1/3 in H -- even though you were certain that you would receive no new information and suffer no cognitive mishaps during the intervening time. Thus the Sleeping Beauty example provides a new variety of counterexample to Bas Van Frassen's 'Reflection Principle'...

When Monday comes around Sleeping Beauty doesn't know that it's Monday anymore, which is exactly why she revises the probability of heads from 1/2 to 1/3. Her information has definitely changed!

Elga could get rid of the coin flips entirely. On every day of the week Sleeping Beauty is put to sleep and made to forget what day it is. Then she is woken up and asked what the probability that it is Monday. When she is woken up on Monday, what should she say? Before the experiment, the answer to this question is clearly 1. But when Monday actually comes around she'll give the answer 1/7. Whoa! Clearly philosophers have something profound to say about probability!


Posted by: Commenter-in-exile | Link to this comment | 06-22-09 8:28 AM
horizontal rule
170

169: I just learned the phrase yesterday. I'm going to try to work it into more conversations.


Posted by: Moby Hick | Link to this comment | 06-22-09 8:28 AM
horizontal rule
171

168: Huh. Doesn't it describe the dominance games you were talking about in the other thread fairly well?


Posted by: LizardBreath | Link to this comment | 06-22-09 8:42 AM
horizontal rule
172

If Sleeping Beauty chooses 1/3, she will be right 2/3 of the time. If she chooses 1/2, she will be right 1/2 the time. Which is why the correct answer, obviously, is 3/5.


Posted by: Spike | Link to this comment | 06-22-09 8:46 AM
horizontal rule
173

159
If she knows it's the first time, the answer is obviously 1/2.
If she doesn't know, the answer is obviously 1/3.

Elga thinks that this contradicts something known as the "Reflection Principle," but he's obviously wrong.

What does "obviously" mean?

Posted by: Commenter-in-exile

And yet we have yet another logic conundrum.

168: "Irritating" is fair enough, but why "wrong-headed"? It seems an apt description of a real situation. The description may be annoying, but it's still apt. It's a subset of peer pressure, just in more adult and professional situations than we normally think of peer pressure.


Posted by: Cyrus | Link to this comment | 06-22-09 8:47 AM
horizontal rule
174

And so we see the likelihood of any unfogged thread becoming a gender thread comes to dominate the prior probability that a thread for a post that mentions Cosma will continue endlessly about Bayesianism.


Posted by: Sifu Tweety | Link to this comment | 06-22-09 8:48 AM
horizontal rule
175

174: That is what a guy would say.


Posted by: JP Stormcrow | Link to this comment | 06-22-09 8:52 AM
horizontal rule
176

re: 173 and 171

Because the use of the word 'performing' gets the relationship between gender and identity totally wrong.

We aren't genderless higher-beings who adopt gender roles at will. We are men or women. That's partly why sexism is so pernicious -- it gets embedded in people's identities. Our gender identities are partly constitutive of who we are. It makes no sense to try to abstract gender away from the identities we have.

Of course, sometimes we perform roles that aren't those we'd naturally slip into, or we exaggerate, but to persistently talk of gender roles as 'performance' is badly mistaken (and a tad wanky).

I'm not arguing against the existence of sexism, or that gender identities aren't socially contingent. I'm completely in agreement with much of the substantive criticism that gets made about the ways in which gender identities are formed and the ways in which they are often problematic. It's the specific use of the word 'perform' that's a problem.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 9:00 AM
horizontal rule
177

I've made the point in 176 before, on a previous thread.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 9:00 AM
horizontal rule
178

Huh. And yet you have people talking about, e.g., self-consciously at the age of eight or nine realizing a need to fake an interest in professional sports in order to be perceived as socially appropriate. "Performing gender" seems like a straightforwardly accurate way to describe that sort of thing. It doesn't necessarily need to imply that in the absence of performance, gender roles would disappear, but there's some performance going on.


Posted by: LizardBreath | Link to this comment | 06-22-09 9:05 AM
horizontal rule
179

re: 178

No, but on this site people persistently talk about performing gender and in contexts other than this one, and I think that just gets the relationship wrong, or, at best, only accurately characterizes a very small part of the ways in which our gender identities function. It continually recasts gender as somehow inauthentic - as if it's something we do rather than something we are.

I get that there's a complicated process involved. We act according to social expectations, sometimes those social expectations are at odds with the way we'd ordinarily choose to act, doing so feels like an act of performance, plus, we are or become how we act, etc etc ad infinitum. But I just personally don't like the continual use of 'perform' to characterize what is a much more complicated relationship between action and personal identity.



Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 9:14 AM
horizontal rule
180
Posted by: | Link to this comment | 06-22-09 9:19 AM
horizontal rule
181

In 173, the paragraph beginning "Elga thinks" is a quote, but the question "What does 'obviously' mean?" is mine. Stupid italics.


Posted by: Cyrus | Link to this comment | 06-22-09 9:19 AM
horizontal rule
182
Posted by: | Link to this comment | 06-22-09 9:21 AM
horizontal rule
183

I endorse essear's comments upthread, and particularly the mildly contemptuous tone.

179: "perform" as I understand it in context specifically refers to those parts of gender presentation that require effort, as opposed to the parts that come naturally. Manly guys who really do like sports aren't performing their gender, but when I do the exact same things they do with regards to sports I am performing, since I'm faking it.


Posted by: togolosh | Link to this comment | 06-22-09 9:21 AM
horizontal rule
184

I'm sympathetic to ttam's pushing back on this usage, but when you add an adverb to "perform gender" -- e.g. "well" or "poorly" -- it becomes instantly very clear what's being talked about.

Perhaps it shouldn't be the only way gender gets talked about. But it's a very real part of it -- gender as a constitutive ideology that creates real pain for those who don't get it right.

Also, there's the tradition gender/sex distinction, though that may leave too little in between biology and (performed, or what you will) identity for you.


Posted by: Wrongshore | Link to this comment | 06-22-09 9:21 AM
horizontal rule
185

re: 183

Yeah, I get that usage, and sometimes in context it makes sense. But it just seems to elide a lot of what is going on in some of the contexts in which it is used.

We rarely, if ever, talk about people performing other roles which are partly constitutive of their identity.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 9:24 AM
horizontal rule
186

173: In the first two lines, "obviously" is used to indicate my opinion that the controversy only exists because of confusion about what the question is assuming; if the question is clarified, then all right-thinking people will agree on the answer.

In the third line, "obviously" is used to describe the argument I put forth in 169. Dunno, it just seems obvious to me.


Posted by: Commenter-in-exile | Link to this comment | 06-22-09 9:26 AM
horizontal rule
187

Perhaps we ought to.


Posted by: Wrongshore | Link to this comment | 06-22-09 9:27 AM
horizontal rule
188

187 to 185. And, why not, 175.


Posted by: Wrongshore | Link to this comment | 06-22-09 9:28 AM
horizontal rule
189

Completely off-topic:

From Ross Douthat's review of Mark Helprin's book, Digital Barbarism

In particular, one should never, ever write a book that includes, in its footnotes, "Posting No. 12" from thelede.blogs.nytimes.com, or "Posting 3:41" from missnemesis.blogspot.com -- or comments by "Peep," "Constantine" and "Anon," from Matthew Yglesias's blog.

Holy shit! Did Helprin actually quote me in his book? Does that mean I will be compelled to read this apparently outlandishly stupid book to find out?



Posted by: peep | Link to this comment | 06-22-09 9:28 AM
horizontal rule
190

performing other roles which are partly constitutive of their identity.

I'm trying to think of what those might be. I'm coming up with localism -- those threads where I'm doing New Yorker schtick, and Californians are doing the same. Some of that is performance, but it's almost purely playful; while I feel a real pressure to act in an appropriately feminine manner, I don't actually feel any social pressure to play the role of a loyal New Yorker -- that I do because it amuses me. So, still performance, but not performance that's terribly interesting.

What other kinds of roles are you thinking of?


Posted by: LizardBreath | Link to this comment | 06-22-09 9:29 AM
horizontal rule
191

I don't perform gender, I'm a real man. No acting here!

(Actually, I only watch the Super Bowl for the ads, and if I hear any bellowing from my housemates or family to indicate some history-making play I'll pop my head in and catch it on instand replay. And I can't stand golf, although maybe "can't stand" is putting it too strongly since I haven't tried since I was 10 or 12. Fortunately, I have chosen a career path at which gendered shmoozing activities are not required, and my friends share my geeky non-sports interests or at least have geeky interests of their own so they can be sympathetic.)


Posted by: Cyrus | Link to this comment | 06-22-09 9:30 AM
horizontal rule
192

ttaM and I are performing Scottishness. Soon we will attack each other with knives.


Posted by: ajay | Link to this comment | 06-22-09 9:33 AM
horizontal rule
193

I like how 'performing masculinity' sounds like it could have been used in an Enzyte commerial.


Posted by: Moby Hick | Link to this comment | 06-22-09 9:34 AM
horizontal rule
194

189: Yes, read it, but steal it. Better yet, pirate it or something so as not to increase the sales figures and money in his pocket.


Posted by: Cyrus | Link to this comment | 06-22-09 9:35 AM
horizontal rule
195

re: 190

Class, for a start. Age, for another. Race, nationality, etc. I mean it's so obvious it's almost vacuous to point out that our identities are multiply constituted in this kind of way, and that we dial up and dial down different bits in different contexts.

There's definitely an interesting discussion to be had about the nature of our identities. I just find -- without wanting to make a huge deal out of it -- that the word 'perform' is, often but not always, a dumb choice of word to use in having that kind of discussion.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 9:35 AM
horizontal rule
196

I dreamed I performed masculinity in my Maidenform bra.


Posted by: JP Stormcrow | Link to this comment | 06-22-09 9:35 AM
horizontal rule
197

on my knees


Posted by: JP Stormcrow | Link to this comment | 06-22-09 9:38 AM
horizontal rule
198

From Ross Douthat's review of Mark Helprin's book

A frightening phrase.


Posted by: nosflow | Link to this comment | 06-22-09 9:39 AM
horizontal rule
199

We rarely, if ever, talk about people performing other roles which are partly constitutive of their identity.

Which is too bad, since I find myself performing all the damn time. I perform UMC, I perform swipple, I perform tolerant non-elitist non-arrogant non-prick, and so on. I've recently started to put real effort into refining the elements of swipple performance, and it's a pain in the ass, but it seems like it's the least bad option.


Posted by: togolosh | Link to this comment | 06-22-09 9:41 AM
horizontal rule
200

198: Yes, and the proof of how bad Helprin's book must be is that even Douthat could hardly manage to say anything good about it.


Posted by: peep | Link to this comment | 06-22-09 9:43 AM
horizontal rule
201

199: "I perform tolerant non-elitist non-arrogant non-prick"

Not very well.


Posted by: Moby Hick | Link to this comment | 06-22-09 9:44 AM
horizontal rule
202

and pwned at 195 by the very person I'm responding to. Though since I can claim Scottish blood, perhaps I should perform as Ajay suggests and stab you, ttaM. Admittedly I wasn't aware of the stabbing thing, but I rather like it, as there are a number of people I 'd like to stab.


Posted by: togolosh | Link to this comment | 06-22-09 9:44 AM
horizontal rule
203

I only watch the Super Bowl for the ads

This is a classic class performance, no?


Posted by: Lambent Cactus | Link to this comment | 06-22-09 9:47 AM
horizontal rule
204

203: Same people that only read Playboy for the articles, probably.


Posted by: apostropher | Link to this comment | 06-22-09 9:48 AM
horizontal rule
205

202: But as ttaM will probably point out shortly, stabbing people isn't merely a performance for Scottish people -- it's who they are.


Posted by: peep | Link to this comment | 06-22-09 9:48 AM
horizontal rule
206

Scottishness is, regrettably, all about the stabbing, togolosh. And it doesn't have to be other Scots - there is a long tradition of stabbing foreigners too.


Posted by: ajay | Link to this comment | 06-22-09 9:49 AM
horizontal rule
207

190

I'm trying to think of what those might be. ...

Motherhood or fatherhood for example. Or Democrat or Republican. Or doctor or lawyer. Lots of examples.


Posted by: James B. Shearer | Link to this comment | 06-22-09 9:50 AM
horizontal rule
208

Yes, stabbing is constitutive of our identities qua Scot.

I stab, therefore I am.

Of course, our identities qua Scot are multiply constituted. So you could also headbutt, or hit with broken bottles.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 9:53 AM
horizontal rule
209

190: Think of it like code-switching. I can see pushing back on describing something as "performing gender" to the extent that describing it that way can give the impression that there's one authentic self, and then the self that must act interested in sports (or whatever), when it's really very common to move between different ways of self-expression.

This is a classic class performance, no?

Yup. I am not common. I am entertained by clever marketing (which is never as good this year as it was last year.)


Posted by: Cala | Link to this comment | 06-22-09 9:55 AM
horizontal rule
210

208: Ah, yes, no one could deny the multi-faceted nature of the Scot.


Posted by: peep | Link to this comment | 06-22-09 9:56 AM
horizontal rule
211

re: 209

I can see pushing back on describing something as "performing gender" to the extent that describing it that way can give the impression that there's one authentic self, and then the self that must act interested in sports (or whatever), when it's really very common to move between different ways of self-expression.

Yes, that is what I am getting at. It's the combination of the dubious appeal to authenticity and the over-simplification that makes me prefer other words.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 10:00 AM
horizontal rule
212

I watch the Super Bowl for the snacks.


Posted by: Lambent Cactus | Link to this comment | 06-22-09 10:01 AM
horizontal rule
213

Which is too bad, since I find myself performing all the damn time.

Me too.

I'm not sure I'm getting ttam's gripe at all, since teasing apart those aspects of behavior that are performance and those that are "constitutive of their identity" is a pretty interesting topic.

And the gender stuff is especially interesting in this regard, since there's so much legitimate disagreement about which parts are "identity" and which parts are "performance." That's why it gets talked about so much.


Posted by: politicalfootball | Link to this comment | 06-22-09 10:02 AM
horizontal rule
214

I can see pushing back on describing something as "performing gender" to the extent that describing it that way can give the impression that there's one authentic self, and then the self that must act interested in sports (or whatever), when it's really very common to move between different ways of self-expression.

I can see that. My initial response is to think that 'perform' is useful more broadly than just for gender, rather than to stop using it for gender. The conversation about social pressure not to excel academically for poor and minority kids, for example, seems like a natural place for this kind of usage of 'performance'.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:02 AM
horizontal rule
215

I mightn't have made myself clear earlier, so here's another way to put the same point:

Sure, if the experiment were repeated many times and you had a policy of saying "tails" whenever woken, the number of correct utterances you would make would be 2/3 the total number of utterances you made. But why should this correspond to any genuine sense in which the probability of your being correct is 2/3? You were still correct in 1/2 the trials; it's just that in the trials where you where you were correct you said the correct thing twice - once on Monday and once on Tuesday.

(This is basically essear's point at 134, I think).


Posted by: KingOfVegetables | Link to this comment | 06-22-09 10:02 AM
horizontal rule
216

I wish I could peform 'guy who finished the draft last week when it was due.'


Posted by: Moby Hick | Link to this comment | 06-22-09 10:02 AM
horizontal rule
217

203: Could be, I guess, but what class? SWPL itself? I get the impression that watching the Super Bowl is a cross-class activity, and the class markers are whether you watch it on a big screen or regular, crappy couch or good couch or sports bar, etc.


Posted by: Cyrus | Link to this comment | 06-22-09 10:06 AM
horizontal rule
218

215: I think that's the fundamental equivocation -- is a 'trial' an utterance, or a coin flip? If the trial is an utterance, then the probability of tails being correct is 1/3. If the trial is the coin flip, the probability of tails being correct is 1/2.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:07 AM
horizontal rule
219

217: There's definitely a social milieu in which it's ordinary to have or to attend a Super Bowl party despite lack of interest in football, just to pay attention to the commercials. That always seemed even duller than paying attention to football to me, but that's probably the type of Super Bowl party I'd end up at if I went to one at all.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:10 AM
horizontal rule
220

Yes, stabbing is constitutive of our identities qua Scot.

This absolutely thrills me. I'm going to exercise my right as an American to claim full membership in an ethnicity I hold only by the thinnest of threads. Matheson represent! Stabstabstab! headbutt! Drink, drink, stab, drink!


Posted by: togolosh | Link to this comment | 06-22-09 10:10 AM
horizontal rule
221

220: You've forgotten sustaining yourself on foods that are unexpectedly deep-fried.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:11 AM
horizontal rule
222

re: 214

But we already have a whole vocabulary for this -- it's not like the idea that the relationship between people's identities; the social roles in which they fit; the expectations placed upon them by the multiple communities in which they are embedded; the relationship between identity, class, gender; the multiple ways in which individual identity gets constructued; how we represent ourselves to ourselves, and so on, it's not like any of that is new. It's important and interesting stuff and social scientists, philosophers, anthropologists, cultural theorists, and many others have developed lots of ways of discussing it. I just find the reliance on the verb 'to perform' overly glib.

I suspect I am fighting a losing battle on this one, though.

re: 221

That's true. Deep fried foods, definitely. Generally being nasty, brutish and short.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 10:12 AM
horizontal rule
223

STAB STAB INNOVATE STAB


Posted by: OPINIONATED SCOT | Link to this comment | 06-22-09 10:15 AM
horizontal rule
224

220: You've forgotten sustaining yourself on foods that are unexpectedly deep-fried.

My experience eating a deep-fried Mars bar was anything but "sustaining".


Posted by: essear | Link to this comment | 06-22-09 10:17 AM
horizontal rule
225

219

There's definitely a social milieu in which it's ordinary to have or to attend a Super Bowl party despite lack of interest in football, just to pay attention to the commercials. ...

Absent a professional interest in advertising this strikes me as bizarre.


Posted by: James B. Shearer | Link to this comment | 06-22-09 10:19 AM
horizontal rule
226

re: 224

Your are forgetting the sophistication of the curry/haggis interface.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 10:19 AM
horizontal rule
227

My experience eating a deep-fried snickers bar was that it was amazing.


Posted by: nosflow | Link to this comment | 06-22-09 10:21 AM
horizontal rule
228

I get the impression that watching the Super Bowl is a cross-class activity, and the class markers are whether you watch it on a big screen or regular, crappy couch or good couch or sports bar, etc.

One could watch the Super Bowl because they care deeply about who wins.


Posted by: water moccasin | Link to this comment | 06-22-09 10:22 AM
horizontal rule
229

I think it's pretty clear that people highlight the "performance" of gender in order to make a polemical point about the ways gender roles can be constraining. But they can also come quite naturally, for most people some degree of gender is part of who they are.

It seems to me the whole "natural/social" role divide is pretty questionable when it comes to human beings -- we are naturally designed to play social roles. Maybe we only feel like we are under pressure to "perform" roles when the role has lost enough social support that conflicts around it are allowed to surface in consciousness.

What other kinds of roles are you thinking of?

LB, what about your other roles of a lawyer, a mother, a rational thoughtful adult-type person....


Posted by: PGD | Link to this comment | 06-22-09 10:23 AM
horizontal rule
230

222: This probably comes down to the fact that I don't do this sort of stuff professionally. I'm sure there is a better vocabulary for these issues out there, but I don't have it at my fingertips.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:23 AM
horizontal rule
231

228: "One could watch the Super Bowl because they care deeply about who wins."

We have a lot of that sort around Pittsburgh.


Posted by: Moby Hick | Link to this comment | 06-22-09 10:24 AM
horizontal rule
232

I like watching the SuperBowl.


Posted by: heebie-geebie | Link to this comment | 06-22-09 10:28 AM
horizontal rule
233

LB, what about your other roles of a lawyer, a mother, a rational thoughtful adult-type person....

I suppose my reaction to that is that there is no question (bracketing out 'mother', which is a little more socially fraught), that those roles are a matter of socialized choice -- thinking of those as 'performance' is so obvious that the word isn't interesting. Saying, for example, before going off and behaving in a reasonable manner about something annoying, rather than whining and being unhelpful about it: "I suppose I should go act like a reasonable person now," would be a very normal utterance. And the same with saying "Time to go be a lawyer."

I think the point of using 'perform' for gender is to claim that it's similar to other roles that are unquestionably 'performances' -- that being feminine is on some level the same sort of task as lawyering.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:29 AM
horizontal rule
234

231: In certain years.


Posted by: apostropher | Link to this comment | 06-22-09 10:29 AM
horizontal rule
235

I think the point of using 'perform' for gender is to claim that it's similar to other roles that are unquestionably 'performances' -- that being feminine is on some level the same sort of task as lawyering.

Which, I think, is clearly false.


Posted by: nattarGcM ttaM | Link to this comment | 06-22-09 10:30 AM
horizontal rule
236

215: Sure, if the experiment were repeated many times and you had a policy of saying "tails" whenever woken, the number of correct utterances you would make would be 2/3 the total number of utterances you made. But why should this correspond to any genuine sense in which the probability of your being correct is 2/3? You were still correct in 1/2 the trials; it's just that in the trials where you where you were correct you said the correct thing twice - once on Monday and once on Tuesday.

If you were never woken up when heads was flipped, and woken up one time when tails was flipped, then the probability of the coin being heads given that you were woken up and asked is 1, not 1/2. You know for sure that the coin was tails. I don't see why this logic can't be applied to the original example to give an answer of 1/3.

The opposing intuition is that you gain no information by waking up, since you wake up at least once whether the coin is heads or tails. But since the game is set up to wake you up more often when tails is flipped, the very fact that you are awake means that you're more likely to be in the tails branch.

In his paper Elga focuses on the first time Sleeping Beauty is woken up (Monday). Since Sleeping Beauty wakes up on Monday no matter what, he thinks it is very strange for her to revise her probability of heads from 1/2 to 1/3 on Monday. But of course Elga glosses over the fact that on Monday she no longer knows what day it is. With this loss of information, it's quite rational for her to revise her estimate.


Posted by: Commenter-in-exile | Link to this comment | 06-22-09 10:31 AM
horizontal rule
237

I love the Super Bowl because public places are so uncrowded while it's on.

Also, I don't watch television.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 10:31 AM
horizontal rule
238

that being feminine is on some level the same sort of task as lawyering.

To spin that out -- some things about lawyering are easy and natural to me, and I'd act that way even if I hadn't been to law school. And some are learned and effortful. I'm not here saying that gender is entirely an externally imposed role, just that it's like other roles that we recognize as in part learned and externally imposed.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:33 AM
horizontal rule
239

233: yeah, well put. I think of the "gender" role as a lot closer to "mother" than it is to a job -- something that comes naturally a lot of the time, feels biologically rooted, but also becomes a hassle at particular defined moments when social expectations for it conflict with what feels comfortable for you personally.

There are also similar overheated and personalized arguments about what it is to be a "real" parent.


Posted by: PGD | Link to this comment | 06-22-09 10:35 AM
horizontal rule
240

235: Depends on your experience. I certainly have days where being appropriately lawyerly comes more easily than being appropriately feminine.


Posted by: LizardBreath | Link to this comment | 06-22-09 10:35 AM
horizontal rule
241

I'd stay and chat, but the NYT says that we're not scared enough of North Korea, so I have to go panic now.

And for the 942nd time, dammit, NYT, "Hawaiian" refers to ethnicity, not residence.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 10:35 AM
horizontal rule
242

Sometimes the Super Bowl is on my birthday.


Posted by: heebie-geebie | Link to this comment | 06-22-09 10:36 AM
horizontal rule
243

241: Is there a one-word term for 'resident of Hawai'i regardless of ethnicity'?


Posted by: LizardBreath | Link to this comment | 06-22-09 10:37 AM
horizontal rule
244

that gender is entirely an externally imposed role, just that it's like other roles that we recognize as in part learned and externally imposed

all roles are partly learned and imposed. There's no such thing as an entirely "natural" way to be human.

As I said, talk about performing gender is a polemical attempt to highlight the "artificial" aspects of gender, and I don't see how you could deny that there are such aspects. But I can also see how harping on it to the exclusion of other perspectives could be irritating.


Posted by: PGD | Link to this comment | 06-22-09 10:38 AM
horizontal rule
245

218: What I'm having trouble seeing is how "Tails and it is Monday" could count as a different (i.e. mutually exclusive) outcome to "Tails and it is Tuesday". If the one occurs the other is bound to occur as well.


Posted by: KingOfVegetables | Link to this comment | 06-22-09 10:44 AM
horizontal rule
246

245: forget the coin flips, wake up Sleeping Beauty once a day and ask her what day it is. "Monday" and "Tuesday" are mutually exclusive outcomes although they are both bound to occur at some point.


Posted by: Commenter-in-exile | Link to this comment | 06-22-09 10:52 AM
horizontal rule
247

WOULD ALL YOU NON-PRINCE CHARMING MOTHERFUCKERS QUIT WAKING ME UP??? I'M GETTING TIRED OF THIS SHIT AND I DON'T CARE WHAT FUCKING DAY IT IS!!!1!!


Posted by: OPINIONATED SLEEPING BEAUTY | Link to this comment | 06-22-09 10:57 AM
horizontal rule
248

I just read the Self-Made Man: One Woman's Year Disguised as a Man book. It was pretty interesting on gender performance. Alot of what she did was pretty stunt-y though.


Posted by: lemmy caution | Link to this comment | 06-22-09 11:12 AM
horizontal rule
249

At some point, gender performance becomes practiced enough to be done unconsciously, but I don't think that makes it less of a performance. Stereotypical American boys aren't born liking to watch the Super Bowl; they watch and cheer in emulation of older males around them -- unquestionably "performance" -- until it becomes habitual.


Posted by: Hamilton-Lovecraft | Link to this comment | 06-22-09 11:14 AM
horizontal rule
250

I'm not entirely comfortable with the "performing gender" locution either, to be honest. It's frankly just shorthand for a particular stance toward the matter, a stance which may or may not hold at any given time.

I can't easily think of alternative formulations that don't take an entire paragraph to explain, getting into societal expectations, possible alienation from and resistance to same, reluctant compliance, blah blah blah.


Posted by: parsimon | Link to this comment | 06-22-09 11:20 AM
horizontal rule
251

236: On the Halfer interpretation I'm imagining, you get probability of 1/2 for Tails whenever you're woken, as long as you're woken at least once whatever the outcome of the flip, because the proportion of Tails among coin-flip trials where you are woken at least once is 1/2. But if you're not woken at all on Heads then the "Halfer" interpretation I'm imagining gives probability 1 for Tails, because the proportion of Tails in coin-flip trials where you are woken is 1. So your intuition that in the latter scenario the probability of Tails given your evidence is 1, is consistent with taking the number of trials to be the number of coin-flips. as opposed to the number of wakenings.

246: My picture is that "Monday" and "Tuesday" are not mutually exclusive outcomes, because they're not alternative possible states that the world can take.


Posted by: KingOfVegetables | Link to this comment | 06-22-09 11:21 AM
horizontal rule
252

241: All the Hawaiians so called in the article appear to be ethnic Hawaiians, not just resident Hawaiians, at least judging by their names.

IIRC this is a real issue when talking about the 'Stans; you need to differentiate between, say, Uzbekistani Tajiks and Tajikistani Uzbeks.


Posted by: ajay | Link to this comment | 06-22-09 11:23 AM
horizontal rule
253

Michael Bay should make a Bayesian movie: "In a world where there is a 1/3 chance that it is Monday..."


Posted by: Moby Hick | Link to this comment | 06-22-09 11:26 AM
horizontal rule
254

253: It's not impossible that I would go see that, but given his priors...


Posted by: Eggplant | Link to this comment | 06-22-09 11:30 AM
horizontal rule
255

251: Well, sure, if you want to define probability in a certain way, you can do that. My point is that there's absolutely nothing wrong with defining things so that the answer to the question is 1/3: it's a perfectly consistent way of looking at the world and doesn't lead to any paradoxes or contradictions. If Sleeping Beauty doesn't know what day it is, nothing bad happens if you allow the concept of "probability that today is Monday."


Posted by: Commenter-in-exile | Link to this comment | 06-22-09 11:34 AM
horizontal rule
256

Is there a one-word term for 'resident of Hawai'i regardless of ethnicity'?

Not really. Ethnicity and whether you were born and raised here are important information. "Local" is probably about as close as you get, but means not only lives here but also non-white and born and raised here (mostly raised might do, especially if parents are local; my son, who's mixed race and has lived here since before he was two, would have a marginal claim). "Local haole" means born and raised here but white. "Kamaaina" means born and raised here and of a family that's been here for a long time, generally white but maybe with a little bit of mixed race in there somewhere, except that it's also marketing-speak for anyone who lives here and can therefore be given a discount from tourist prices.

Some Hawaiians dislike "local" because they see it as claiming a connection with the place that belongs only to them. This stuff gets pretty complicated. For a lot of purposes, you just stick with "lives here."

All the Hawaiians so called in the article appear to be ethnic Hawaiians, not just resident Hawaiians, at least judging by their names.

One of eight, by my count ("Aikau" is Hawaiian; "Obama," "Lingle," "Lee," "Souki," "Tauiliili," "Brown," and "Connor" aren't (although "Mele" is)). And you really can't assume ethnicity from name around here.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 11:50 AM
horizontal rule
257

Clearly the solution is to change the name of the state. That way Hawai'ian is an ethnicity and Volcanistani or whatever is a regional identifier.


Posted by: togolosh | Link to this comment | 06-22-09 11:54 AM
horizontal rule
258

South Alaska?


Posted by: Moby Hick | Link to this comment | 06-22-09 11:55 AM
horizontal rule
259

256: Not really. Ethnicity and whether you were born and raised here are important information.

At that point I think you have to cut the NYT some slack for using "Hawaiian" as meaning "resident of Hawaii". They've got a legitimate need in the story to talk about "residents of Hawaii", and every other state uses a formation from the state name to mean 'resident of that state'. If there's not a preferable synonym, what else were they supposed to do?


Posted by: LizardBreath | Link to this comment | 06-22-09 11:56 AM
horizontal rule
260

"Sandwich Islands" is the obvious choice, but could raise certain political and other concerns.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 11:57 AM
horizontal rule
261

259: Say "residents" or "residents of Hawaii." I'm being somewhat peevish here, because "Hawaiian" as a geographic identifier is common outside the state, but it grates on local ears.

(Huh, "local" as an adjective in the preceding sentence turns out to work more broadly than "local" as a noun as defined above.)

And now I'm curious. Are residents of New Mexico "New Mexicans"? That's the closest analogy I can think of in the other 49 states. And would anyone refer to someone like me who moved to Puerto Rico as "Puerto Rican"?


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:01 PM
horizontal rule
262

You're Floridian or Texan or Nebraskan or Alaskan. But not Kansassian or Michigander unless you're being cutesy. Probably depends on the name of the state.


Posted by: heebie-geebie | Link to this comment | 06-22-09 12:04 PM
horizontal rule
263

And would anyone refer to someone like me who moved to Puerto Rico as "Puerto Rican"?

I don't think the oddness, such as it is, would be because of ethnicity issues. I live in Texas, but in many contexts it would seem odd to call me "Texan" (grew up in Florida, spent much time in the Northeast).


Posted by: Criminally Bulgur | Link to this comment | 06-22-09 12:08 PM
horizontal rule
264

NPH: are you in PR currently?


Posted by: soup biscuit | Link to this comment | 06-22-09 12:09 PM
horizontal rule
265

263: But isn't that because "Texan" is an ersatz ethnicity for white folks of a certain sort?


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:10 PM
horizontal rule
266

264: No. What I'm asking is if a white person raised elsewhere and transplanted to Puerto Rico becomes "Puerto Rican."


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:11 PM
horizontal rule
267

261
And would anyone refer to someone like me who moved to Puerto Rico as "Puerto Rican"?

262: In the northeast there are Vermonters, New Yorkers, Rhode Islanders, New Hampshire residents (maybe "New Hampshire-ites", but I don't think so), Mainiacs and Massholes. I don't know what people from Connecticut are called. Connecticutters?


Posted by: Cyrus | Link to this comment | 06-22-09 12:12 PM
horizontal rule
268

I wouldn't think it amiss in a news story.


Posted by: Cala | Link to this comment | 06-22-09 12:12 PM
horizontal rule
269

267: WASPs.

Sorry. I think Connecticutian (like Lilliputian) would be ideal, personally.


Posted by: Cala | Link to this comment | 06-22-09 12:13 PM
horizontal rule
270

Are residents of New Mexico "New Mexicans"?

I'd say yes, but I'm from Chicago. Or Berkeley. I moved here in 1975.


Posted by: Michael H Schneider | Link to this comment | 06-22-09 12:14 PM
horizontal rule
271

263, 265: I think in most states you'd comfortably refer to everyone who lives there as "Vermonters" or "people from Massachusetts" (come to think of it you're right, lots of states don't have collective nouns for residents) if you were talking about them generally. Once you started looking at individuals, you'd start getting picky about emotional affiliation -- someone who lives in NY but is planning to move back to Nebraska when they get married and have kids might not call themselves a New Yorker.


Posted by: LizardBreath | Link to this comment | 06-22-09 12:15 PM
horizontal rule
272

South Alaska?

West Carolina


Posted by: apostropher | Link to this comment | 06-22-09 12:15 PM
horizontal rule
273

Connecticutters?

Nutmeggers?


Posted by: LizardBreath | Link to this comment | 06-22-09 12:16 PM
horizontal rule
274

Sorry. I think Connecticutian (like Lilliputian) would be ideal, personally.

"Connecticutian" is reserved for speech acts that effect financial transactions.


Posted by: nosflow | Link to this comment | 06-22-09 12:16 PM
horizontal rule
275

271: I still call myself Nebraskan and I haven't lived there since 1993 and have no plans to live there again. None of the other states have measured up.


Posted by: Moby Hick | Link to this comment | 06-22-09 12:17 PM
horizontal rule
276

Connecticutters?

Connecticoots?


Posted by: apostropher | Link to this comment | 06-22-09 12:20 PM
horizontal rule
277

I've tried introducing myself as Moby Hick "of the Nebraska Hicks" but that gets me nothing but funny looks.


Posted by: Moby Hick | Link to this comment | 06-22-09 12:22 PM
horizontal rule
278

261

Say "residents" or "residents of Hawaii." I'm being somewhat peevish here, because "Hawaiian" as a geographic identifier is common outside the state, but it grates on local ears.

So what? It's the New York Times and Hawaiian can mean an inhabitant of Hawaii.


Posted by: James B. Shearer | Link to this comment | 06-22-09 12:23 PM
horizontal rule
279

Oops, meant to comment to 261 to say basically that the situation in Hawaii is one I've never thought of before. For a while now I've been aware of two types of places in America: places where you have to be there for three generations or more to be considered a native (like Vermont), and places where if you live there for two years you're a native for all purposes (like DC, or parts of some western states I believe). It's not a racial thing, or at least it's not just a racial thing, all the native Vermonters are white but so are all the newcomers too. So to answer the question in 261 I'd try to fit Hawaii and Puerto Rico into that template.


Posted by: Cyrus | Link to this comment | 06-22-09 12:24 PM
horizontal rule
280

278: And "Jimbo" can mean "James," but that doesn't mean it would be good form to address you by that name.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:25 PM
horizontal rule
281

Are residents of New Mexico "New Mexicans"?

Yes; or at least that was my experience in Santa Fe and environs. Interestingly, there "Anglo" was a residual category, embracing everyone who was not Native-American, "Spanish" (= descended from the Spanish speakers there at the time of the Mexican-American War; but there was some controversy about the right label for this group), or "Mexican" (= a more recent immigrant from Mexico or their descendant). English-speaking whites, blacks, Asians, etc. were all Anglos.


Posted by: Cosma | Link to this comment | 06-22-09 12:26 PM
horizontal rule
282

280: Although at least possibly entertaining.


Posted by: LizardBreath | Link to this comment | 06-22-09 12:28 PM
horizontal rule
283

266: Oh, I was just curious for a completely different reason...


Posted by: soup biscuit | Link to this comment | 06-22-09 12:32 PM
horizontal rule
284

You're Floridian or Texan or Nebraskan or Alaskan.

Describing people who live in those states in those terms is decidedly odd; it carries a sense that something beyond residency is meant. You'd usually say "native Floridian" or some such. More natural would be to say that someone is "from" whichever state, and even that is increasingly strained now that people are as prone as they are to moving around. You'd likely only say someone is a whateverstate-er if they were born and raised there.

Actually, this stuff is more regional, isn't it? You have your New Englanders, your Pacific Northwesters, your Southwesters, your Southerners. I mean, I'm a New Englander despite currently living in the Mid-Atlantic; but I was raised in New Hampshire, Massachusetts, Panama, Pennsylvania and upstate New York. And Massachusetts again. i.e. New England.


Posted by: parsimon | Link to this comment | 06-22-09 12:32 PM
horizontal rule
285

Further to 284: I guess that should be Southwesterners. I'm thinking Four Corners states. Texas and California are so freaking big that they get a category to themselves.


Posted by: parsimon | Link to this comment | 06-22-09 12:35 PM
horizontal rule
286

283: Thought maybe that was the case, but not until after I'd responded.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:36 PM
horizontal rule
287

Texas and California are so freaking big that they get a category to themselves.

I dunno, they're smaller than some of the canadian provinces, and that still goes regionally the same way


Posted by: soup biscuit | Link to this comment | 06-22-09 12:37 PM
horizontal rule
288

We unfortunates from borders between regions have no home, in this scheme.


Posted by: essear | Link to this comment | 06-22-09 12:39 PM
horizontal rule
289

271

... lots of states don't have collective nouns for residents ...

Not according to this .


Posted by: James B. Shearer | Link to this comment | 06-22-09 12:41 PM
horizontal rule
290

287: True, and I was debating about that. I'd still probably describe someone from Texas as a Southwesterner, and someone from California as a West Coaster (as distinguished from Pacific Northwesters, which latter also include people from the coast of British Columbia, actually).

Is it climate-related?


Posted by: parsimon | Link to this comment | 06-22-09 12:41 PM
horizontal rule
291

287: But Texas and California also have people in them.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:41 PM
horizontal rule
292

I dunno, they're smaller than some of the canadian provinces

Canada has 33 million people total, California just under 37 million, Texas has 24 million. The Four Corners states together have about 16 million; around 18 if you include Nevada in the Southwest, which I guess I would.


Posted by: Cosma | Link to this comment | 06-22-09 12:42 PM
horizontal rule
293

This will all become much easier when land line telephones go the way of the dodo and we all use nothing but cell phones. I'm a Vermonter because my phone number's area code is 802; it doesn't matter how long I live down here, I'm a Vermonter as long as I keep the phone number I had when I moved.


Posted by: Cyrus | Link to this comment | 06-22-09 12:43 PM
horizontal rule
294

280

And "Jimbo" can mean "James," but that doesn't mean it would be good form to address you by that name.

Poor analogy. It would be like I got upset at the use of Jamesian .


Posted by: James B. Shearer | Link to this comment | 06-22-09 12:46 PM
horizontal rule
295

292: Wow, Canada is, like, tiny!


Posted by: parsimon | Link to this comment | 06-22-09 12:48 PM
horizontal rule
296

But Texas and California also have people in them.

This is true.


Posted by: soup biscuit | Link to this comment | 06-22-09 12:51 PM
horizontal rule
297

294: Poor attempt at refutation. The point is that it's polite to refer to people by the names they use for themselves.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:51 PM
horizontal rule
298

Wow, Canada is, like, tiny!

Ssshhh. They like it like that. Don't let too many people know, they might move up there and ruin it.


Posted by: soup biscuit | Link to this comment | 06-22-09 12:52 PM
horizontal rule
299

Anyway, my point had nothing to do with relative population density --- regional identities cross state and provincial borders pretty commonly.


Posted by: soup biscuit | Link to this comment | 06-22-09 12:53 PM
horizontal rule
300

The point is that it's polite to refer to people by the names they use for themselves.

Would it be to much to ask for you guys to start referring to me as Your Sexiness from now on?


Posted by: Otto von Bisquick | Link to this comment | 06-22-09 12:57 PM
horizontal rule
301

299: And regional identity can be different from state identity. Rural BC and Alberta folks may identify with each other more than Vancouverites; Washington and Oregon divide regionally along the Cascades rather than the Columbia; etc.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 12:58 PM
horizontal rule
302

300: Yes.


Posted by: Moby Hick | Link to this comment | 06-22-09 12:59 PM
horizontal rule
303

And regional identity can be different from state identity.

exactly.


Posted by: soup biscuit | Link to this comment | 06-22-09 1:00 PM
horizontal rule
304

302: Selfish bastard. Where are your manners?


Posted by: Otto von Bisquick | Link to this comment | 06-22-09 1:02 PM
horizontal rule
305

I had a beer with His Sexiness recently. It was disconcerting.


Posted by: nosflow | Link to this comment | 06-22-09 1:03 PM
horizontal rule
306

288: We unfortunates from borders between regions have no home, in this scheme.

Sorry -- Plains States? Midwest? I'm not sure where, say, Idaho, Montana, North Dakota, etc. fit in: I think of them as the badlands, red dirt*, big sky country.

* This is coming from the fact that I was told by a friend from Idaho that schoolchildren invariably got the test question "Dirt is _____" wrong. The answer was supposed to be "brown," but the dirt/earth there is not brown, it is red.


Posted by: parsimon | Link to this comment | 06-22-09 1:04 PM
horizontal rule
307

300: No sweat, Your Sexiness. I prefer to be addressed as Master, btw.


Posted by: togolosh | Link to this comment | 06-22-09 1:05 PM
horizontal rule
308

Correction to 306: It wasn't Idaho, it was Wyoming, actually, where the dirt was red. Sorry about that.


Posted by: parsimon | Link to this comment | 06-22-09 1:07 PM
horizontal rule
309

Wow, that's a stupid question.


Posted by: nosflow | Link to this comment | 06-22-09 1:07 PM
horizontal rule
310

306: The actual badlands are a relatively small portion of southern South Dakota, portions of which look like a road runner cartoons and portions of which look bad, but not in an interesting way. Guess which portion went to the Dakota?


Posted by: Moby Hick | Link to this comment | 06-22-09 1:07 PM
horizontal rule
311

Dirt is red in lots of places.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 1:08 PM
horizontal rule
312

310: Yeah, I should have said both Dakotas. The actual badlands are not in any way bad at all, I didn't think.

I have no idea where the stupid dirt question was coming from, but apparently it stuck in people's craw that their kids were judged stupid because of it and related regionalisms in testing. What color is grass, after all? Yellow, a lot of the time!


Posted by: parsimon | Link to this comment | 06-22-09 1:16 PM
horizontal rule
313

Dirt is red in lots of places.
If iron's found in more than traces,
or blood is spilled with swords and maces,
dirt is red in lots of places.


Posted by: nosflow | Link to this comment | 06-22-09 1:16 PM
horizontal rule
314

What color is grass, after all?

Brown and sere.


Posted by: nosflow | Link to this comment | 06-22-09 1:18 PM
horizontal rule
315

297

Poor attempt at refutation. The point is that it's polite to refer to people by the names they use for themselves.

What name do Hawaiians use for themselves?

And is this rule applied to other nationalities?


Posted by: James B. Shearer | Link to this comment | 06-22-09 1:19 PM
horizontal rule
316

It is pleasing that hematite is so named.

||

The hipster cafe is playing Yes' "Roundabout".

|>


Posted by: nosflow | Link to this comment | 06-22-09 1:19 PM
horizontal rule
317

315: See 256 and 261.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 1:21 PM
horizontal rule
318

306-312: But we're sure they've got IQ tests right by now, right?


Posted by: Cyrus | Link to this comment | 06-22-09 1:29 PM
horizontal rule
319

Whoa, and now it's playing something from I believe the most recent The The album (NakedSelf). I would not have expected this.

I wonder if Gun Sluts will ever be released.


Posted by: nosflow | Link to this comment | 06-22-09 1:29 PM
horizontal rule
320

But we're sure they've haven't got IQ tests right by now, right?


Posted by: soup biscuit | Link to this comment | 06-22-09 1:31 PM
horizontal rule
321

I'm past 40 and therefore not smart enough to answer that question.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 1:33 PM
horizontal rule
322

A number of grad students here participate in a science education outreach program, wherein participants go to local elementary schools to teach lessons on science. The point is to show them what young working scientists are like, role models, etc. When a friend was preparing to teach a lesson on the seasons to grade schoolers, someone made the reasonable point to her that as these were SF kids, it would be reasonable of them to answer "foggy" when you ask them what summer is like, and "rainy" when you ask them what winter is like. (She had prepared on construction paper some more mainstream representations of the seasons.)


Posted by: Otto von Bisquick | Link to this comment | 06-22-09 1:36 PM
horizontal rule
323

318, 320: One neat thing that came up in the Nesbitt book, and according to Cosma's review is also in the Flynn book, is that cultural effects aren't limited to that sort of obviously culture bound knowledge (dirt is brown, not red; summer is hot, not foggy; tigers live in Asia, not the Bronx (that one was my father at age five, who'd seen tigers in the zoo)). IQ test items that are completely abstract and supposed to be culture free (Raven matrices, which are those weird little sequences of geometric forms where you get a bunch of shapes, and then have to pick which comes next) show big differences over time and big changes in response to education.


Posted by: LizardBreath | Link to this comment | 06-22-09 1:52 PM
horizontal rule
324

322: Goddamn the cultural imperialism of the four seasons. As a New Mexican child, I was constantly reminded by Eastern-U.S.-and-England-normative books that my native climate was not up to par.


Posted by: Bave Dee | Link to this comment | 06-22-09 1:58 PM
horizontal rule
325

323: Someone upthread made a comment about getting into the test-writer's head, which is absolutely right IME (I was GOOD at that in high school quiz bowl) and pretty much has to be cultural.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 1:59 PM
horizontal rule
326

324: How else were you supposed to learn about your innate inferiority?


Posted by: soup biscuit | Link to this comment | 06-22-09 2:05 PM
horizontal rule
327

Fair to say that a test is basically an argument between test-giver and test-taker and therefore favors people who like to argue?


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 2:07 PM
horizontal rule
328

Fair to say that a test is basically an exercise of authority on the part of a test-maker over a test-taker and therefore favors people who are conciliatory?


Posted by: nosflow | Link to this comment | 06-22-09 2:13 PM
horizontal rule
329

328: Also people who can remember to bring #2 pencil. I used a #4 pencil and blew my shot at Yale.


Posted by: Moby Hick | Link to this comment | 06-22-09 2:15 PM
horizontal rule
330

328: No. Well, sometimes, but only when the test-maker is bad at it.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 2:18 PM
horizontal rule
331

327, 328 -- Both are probably mostly wrong, but 328 makes more sense. To do well, the test-taker has to submit to the logic of the test-taker.


Posted by: peep | Link to this comment | 06-22-09 2:18 PM
horizontal rule
332

331: To do well, the test-taker has to submit to the logic of the test-maker.

(that is what I meant to say)


Posted by: peep | Link to this comment | 06-22-09 2:19 PM
horizontal rule
333

Back in the day - pre-Internet - there was an SAT test prep company that invented a fictional character called (if memory serves) Joe Blog - "Blog" in this case being just a random name, like "Joe Blow".

The key, they argued, was that SAT tests were designed to give guys like Joe an average score, and if you understood Joe, you could understand how to beat the test.

Early in the test, students were taught to answer like Joe, and later in the test, they were taught to avoid answering like Joe, because the tests were designed to get progressively more difficult.

They made a convincing case that you could identify Joe's answers in many cases without comprehending the content of the test at all.


Posted by: politicalfootball | Link to this comment | 06-22-09 2:21 PM
horizontal rule
334

Now the hipster cafe is either playing late Talk Talk or something from Mark Hollis' solo album.


Posted by: nosflow | Link to this comment | 06-22-09 2:23 PM
horizontal rule
335

320: Yes, that was my point.


Posted by: Cyrus | Link to this comment | 06-22-09 2:27 PM
horizontal rule
336

333: I read that book, I think. I already thought standardized tests were games (not fun games, but games), and that book was like discovering the rule book.

I had to hide that book under my bed, because my dad thought that one shouldn't prep for the SAT (he worried about inculcating test anxiety in his kids, and also seemed to believe that the SAT was a perfect test of what one knew, so all one would have to do is walk in and take it. In retrospect it's sort of surprising I got into college.) My mother checked it out from the library.


Posted by: Cala | Link to this comment | 06-22-09 2:31 PM
horizontal rule
337

334: That's some thrilling liveblogging you're doing there, neb. Of course the choice of music is basically an exercise of authority on the part of the cafe employees. To correctly identify what's being played you need to submit to the logic of the hipster barrista.


Posted by: M/tch M/lls | Link to this comment | 06-22-09 2:31 PM
horizontal rule
338

In my state there are four regional identities, none of which containe the name of the state: "near New York City," "Near Philadelphia," "Near Princeton," and "the shore."

And here in near Philadelphia, kids believe that eagles are green and the following contains no speling errors: "Phun phood phestival phriday."


Posted by: unimaginative | Link to this comment | 06-22-09 2:32 PM
horizontal rule
339

test-taker has to submit to the logic of the test-maker

I suppose, but my subjective experience was more like "nice try, you sneaky bastard" than "now I will try to emulate the way you think."


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 2:39 PM
horizontal rule
340

You are all horrible people for not having said:

UTAHN!!!!

I was absolutely incredulous when I saw that in a headline when I visited.


Posted by: JRoth | Link to this comment | 06-22-09 2:51 PM
horizontal rule
341

340: "Resident of the Strategic Whiteness Reserve"


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 2:54 PM
horizontal rule
342

I prefer to be addressed as Master, btw.

And I'm walking past and he yells out to me, "Hey Red". I was a cocky kid. I put down my violin. I go up to him, I said, "My name is not Red. If you want me, call me by my regular name. It's Master Haywood Allen." (Long pause...) I spent that winter in a wheelchair. A team of doctors labored to remove a violin.

Posted by: NickS | Link to this comment | 06-22-09 2:56 PM
horizontal rule
343

336: I shit my pants (metaphorically) at the prospect of taking SATs to get into college, figuring that kids in the US had all manner of prep that was denied me. I busted ass doing practice tests and the like only to find that because I'd been preparing for O and A levels my whole high school career I clobbered my US educated peers who tended to have absolutely shit test taking skills. Just not freaking out at the prospect of being tested is easily good for 100 points on your combined score.


Posted by: togolosh | Link to this comment | 06-22-09 3:07 PM
horizontal rule
344

338: It's "Downashore," please.


Posted by: Witt | Link to this comment | 06-22-09 3:08 PM
horizontal rule
345

"Performing gender" is an excellent description for what happens as you move further from the routine - if, for instance, you're gay or bi, or transgendered. Just because some people only need one or two layers of cultural programming to fit reasonably comfortably into a niche doesn't mean everyone does, and I find the language of performance helpful in building a connection there - it's still a performance even when it's one that comes easily. (It isn't necessary to struggle to be a good actor, if you've got a talent for it or the role is one that plays to your strengths, but you're still acting.)


Posted by: Ceri B. | Link to this comment | 06-22-09 3:16 PM
horizontal rule
346

I busted ass doing practice tests

IM(TU*)O the best strategy for people who already have an affinity for test-taking.

* Totally uninformed.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 3:18 PM
horizontal rule
347

328 is much more accurate than 327.


Posted by: PGD | Link to this comment | 06-22-09 3:27 PM
horizontal rule
348

Conciliatory or cunning, I should have said.


Posted by: nosflow | Link to this comment | 06-22-09 3:31 PM
horizontal rule
349

Just not freaking out at the prospect of being tested

On the other hand, I am always incredibly amused when I get a student coming in after a test, saying how they have terrible, terrible math anxiety and test anxiety and all the rest of it. Then you start chatting about the content of the test and 7 out of 10 times, it becomes clear that the kid has no freaking clue how to do any of the math and got a grade which accurately reflected that.


Posted by: heebie-geebie | Link to this comment | 06-22-09 3:33 PM
horizontal rule
350

Conciliatory runts rule the standarized test scene.


Posted by: JP Stormcrow | Link to this comment | 06-22-09 3:33 PM
horizontal rule
351

338 is interesting. I've lived in said state for about a year, but hadn't picked up on that. If I'm being vague I tend to say Central NJ.


Posted by: essear | Link to this comment | 06-22-09 3:38 PM
horizontal rule
352

Yes we do.


Posted by: Megan | Link to this comment | 06-22-09 3:41 PM
horizontal rule
353

Megan, a conciliatory runt?


Posted by: nosflow | Link to this comment | 06-22-09 3:41 PM
horizontal rule
354

Then you start chatting about the content of the test and 7 out of 10 times, it becomes clear that the kid has no freaking clue how to do any of the math and got a grade which accurately reflected that.

No wonder they have anxiety. Heebie, what do you do in these cases? Do you say "You should come in for help before the test, " or "You have to sit down and do more problems?"


Posted by: bill | Link to this comment | 06-22-09 3:45 PM
horizontal rule
355

Fond of standardized tests, at least.


Posted by: Megan | Link to this comment | 06-22-09 3:46 PM
horizontal rule
356

Conciliatory runts rule the standarized test scene.

They're good at performing cleverness.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 3:53 PM
horizontal rule
357

354: Either, depending. Usually I spend a little time picking the student's brain to find out how things are going for them in class and on the homework. If they claim everything is going great on the homework, I'll recommend that they come in a couple days before the test and work sample problems on the board with me. If they're struggling on the homework, I'll recommend that they come in as much as possible. They are often pretty honest if they're just not spending enough time on the homework, so then I can just grin back at them and let them connect the dots.


Posted by: heebie-geebie | Link to this comment | 06-22-09 3:53 PM
horizontal rule
358

Do you say "You should come in for help before the test, " or "You have to sit down and do more problems?"

"If I were as clueless as you are, I'd be anxious too."


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 3:55 PM
horizontal rule
359

"Connecting the dots" is a culturally fraught activity.


Posted by: essear | Link to this comment | 06-22-09 3:57 PM
horizontal rule
360

"Anxiety is very common when confronted with a test full of questions that you can't solve."


Posted by: heebie-geebie | Link to this comment | 06-22-09 3:57 PM
horizontal rule
361

It seems like I was pwned by 358, but I was trying to riff off of it.


Posted by: heebie-geebie | Link to this comment | 06-22-09 3:58 PM
horizontal rule
362

359: No, it's the first section on the test.


Posted by: heebie-geebie | Link to this comment | 06-22-09 3:58 PM
horizontal rule
363

363: Which reveals that the test is culturally biased against surrealists.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 4:00 PM
horizontal rule
364

363: No, the dots are connected with fish. And lawnmowers.


Posted by: M/tch M/lls | Link to this comment | 06-22-09 4:01 PM
horizontal rule
365

A dot in isolation is a beautiful, symmetrical thing, and you want it to be marred with these hideous anisotropic "lines"? Pshaw.


Posted by: essear | Link to this comment | 06-22-09 4:05 PM
horizontal rule
366

Connecting the dots is actually a deeply communitarian act of bold protest against enlightenment rationality. No wonder the dots-connectors do so poorly on math tests.


Posted by: nosflow | Link to this comment | 06-22-09 4:06 PM
horizontal rule
367

Essear illustrates the point precisely. "A dot in isolation." Next he'll start talking about rationality and preference satisfaction.


Posted by: nosflow | Link to this comment | 06-22-09 4:06 PM
horizontal rule
368

For being such a small state, New Jersey is rich in physical and culturally semi-distinct geographic regions. I would count it as:
- NY Suburbs (has its own rich sub-taxonomy of course).
- Valley and Ridge/Highlands, you'd think you were actually in NE Pennsylvania only somewhat more idyllic in a Norman Rockwell kinda way.
- Central NJ other than Trenton itself.
- Trenton. (which is "Central" despite being on the border, due to shape issues, the centroid of New Jersey nearly falls outside of the state borders)
- Northern Shore (down to about Tom's River, basically most of Monmouth County + a little bit).
- Pine Barrens (physiographically distinct, but very sparsely populated).
- Greater Philly.
- Lower Shore/ Cape May.
- Greater Delaware, Eastern Exclave.


Posted by: JP Stormcrow | Link to this comment | 06-22-09 4:15 PM
horizontal rule
369

357: Isn't it fun when a student comes to you for the first time about a week or so before the final exam, and within minutes you realize they basically have no idea what went on on the course.

The scary thing is they are doing a little better than the ones who tend to pop up on radar the first time a week after the final exam saying they can't understand how they didn't get at least a c+, after all they came to every class and handed in their homework.


Posted by: soup biscuit | Link to this comment | 06-22-09 4:40 PM
horizontal rule
370

I would insert the obligatory link to "Unskilled and Unaware of It", were I not commenting from a phone.


Posted by: essear | Link to this comment | 06-22-09 4:43 PM
horizontal rule
371

Connecting the dots is actually a deeply communitarian act of bold protest against enlightenment rationality.

Heh.


Posted by: parsimon | Link to this comment | 06-22-09 4:44 PM
horizontal rule
372

369.1: Do math teachers not have TAs in your math courses, who would ordinarily be first in line to notice that a student utterly and completely doesn't get what's going on?

This kind of thing obviously goes on in philosophy as well, and indeed we hear people here on this blog saying that they had no idea what was going on in the phil. course(s) they took.

Anyway, I think that no, in a larger state school, you probably don't have TAs.


Posted by: parsimon | Link to this comment | 06-22-09 4:51 PM
horizontal rule
373

372: In theory, absolutely. In practice you have people who just don't go to the lab sessions once they've run into trouble, and are apparently in deep denial for half the term.

It doesn't happen all that often, but every once in a while I get someone in my office who might have been able to turn things around, a month and a half ago.

The last intro course I taught, I had 4 TA's, one of whom turned out to be hopeless. That doesn't help either.


Posted by: soup biscuit | Link to this comment | 06-22-09 4:55 PM
horizontal rule
374

Anyway, I think that no, in a larger state school, you probably don't have TAs.

No, big state schools do have TAs. But my little school doesn't have a grad program, and we pride ourselves on having small classes, etc, so we do not have TAs. Or classes over 30ish.


Posted by: heebie-geebie | Link to this comment | 06-22-09 5:27 PM
horizontal rule
375

374, 375: Okay. It's been a while, and I really hadn't thought about factors like students who just don't come to TA sections (lab sessions), or bad TAs. But, of course.


Posted by: parsimon | Link to this comment | 06-22-09 5:36 PM
horizontal rule
376

338 is amusing to me inasmuch as I am probably the only native Pennsylvanian who says he grew up "near Trenton."


Posted by: Wrongshore | Link to this comment | 06-22-09 5:45 PM
horizontal rule
377

372: I was one of those people, but I should point out that I did get an A in the course notwithstanding. I just don't know what it was about. "Individuals, Societies, and Cultures" was the title, but I couldn't get more precise than that. I suspect good writing skills make up for a lot.

For the now-discarded Sleeping Beauty problem above, Commenter-in-exile explained the reasoning pretty well. The answer is very dependent on your interpretation of the problem, which is why it's a good philosophy question for pondering epistemiology and a bad probability question.


Posted by: wispa | Link to this comment | 06-22-09 5:54 PM
horizontal rule
378

I blame neb for ruining this thread.


Posted by: paranoid android | Link to this comment | 06-22-09 5:57 PM
horizontal rule
379

I blame neb for having a beer with Otto recently and not inviting me.


Posted by: Megan | Link to this comment | 06-22-09 5:59 PM
horizontal rule
380

It was a beer for the heteronomous.


Posted by: nosflow | Link to this comment | 06-22-09 6:01 PM
horizontal rule
381

I don't understand.


Posted by: Megan | Link to this comment | 06-22-09 6:03 PM
horizontal rule
382

Wait, having a beer with Otto is really, actually, disconcerting? On second thought: I'm sure that many people here as well as heteronymously* are disconcerting in real life, which is actually kind of cool.

* That word is misspelled over in the heteronomous place. I never did get that.


Posted by: parsimon | Link to this comment | 06-22-09 6:11 PM
horizontal rule
383

MUST I EXPLAIN EVERYTHING???

Having a beer with otto is not disconcerting. Having a beer with his sexiness might well be.


Posted by: nosflow | Link to this comment | 06-22-09 6:12 PM
horizontal rule
384

Did Otto's sexiness put the moves on you? Just a little too fast for your tastes?


Posted by: Megan | Link to this comment | 06-22-09 6:21 PM
horizontal rule
385

It's disconcerting for me, too, as my sexiness is not fully under my control. Sometimes it is there, sometimes not; sometimes it is well-mannered and a good conversationalist, sometimes it is a boor. And yet it is my sexiness, so though I cannot control it, I nonetheless feel like I am responsible for it.

And re 382.*, it's not clear that heteronomous is not in fact what is intended. But I won't pretend to understand to what those wily humanists might be slyly referring with their choice of that word.


Posted by: Otto von Bisquick | Link to this comment | 06-22-09 6:45 PM
horizontal rule
386

Heteronomy was precisely what was intended. "Heteronymy", a word I'venever before encountered, would be something else: differently named?

Is it possible that phormer philosopher parsley has never encountered the autonomy/heteronomy duo?


Posted by: nosflow | Link to this comment | 06-22-09 6:48 PM
horizontal rule
387

I thought that heteronomy was a variant spelling of heteronymy, and so did not pursue the issue further, but upon inspection, it does now seem to me that heteronymous would be apter.


Posted by: jms | Link to this comment | 06-22-09 6:51 PM
horizontal rule
388

"Heteronymy", a word I've never before encountered, would be something else: differently named?

I guess that's what I thought.

Like jms, I never looked into it, but I will say that when I type the word from scratch, as it were, it's with the "y" rather than the "o", and I have to correct myself. I'm not sure where that's coming from, then.


Posted by: parsimon | Link to this comment | 06-22-09 7:03 PM
horizontal rule
389

No, I don't think I've encountered heteronomous paired with autonomous. Knock me down with a feather.


Posted by: parsimon | Link to this comment | 06-22-09 7:07 PM
horizontal rule
390

"Performing gender" is a useful way of thinking about it for me. When I was six years old, my favorite TV show was Wonder Woman with Linda Carter. One day at recess I was alone and decided to see what it would be like to be Wonder Woman, specifically what it would feel like to run around with her hands hanging down, more or less limp-wristed, as Linda Carter played it for some reason -- I suppose, now, so she'd seem properly feminine. A friend of mine from the neighborhood who was a few years older than me noticed and took me aside: "Boys don't run like that. People might make fun of you." There are all sorts of little things like that that you've got to learn to come across as masculine or feminine, and I've often felt that I'm performing them one way or the other.


Posted by: Bave Dee | Link to this comment | 06-22-09 7:34 PM
horizontal rule
391

A friend of mine from the neighborhood who was a few years older than me noticed and took me aside: "Boys don't run like that. People might make fun of you."

And yet you still turned out gay. Just goes to show that you can never been too careful about teaching boys manliness from the very beginning.


Posted by: Not Prince Hamlet | Link to this comment | 06-22-09 7:49 PM
horizontal rule
392

I read that as heteronormous, i.e. the opposite of homonuscule.


Posted by: JP Stormcrow | Link to this comment | 06-22-09 8:22 PM
horizontal rule
393

in contemporary vernacular, "performing" is to Judith Butler as "deconstructing" is to Jacques Derrida....


Posted by: lurker | Link to this comment | 06-22-09 10:42 PM
horizontal rule
394

Some Hawaiians dislike "local" because they see it as claiming a connection with the place that belongs only to them.

Well, then, that's just tough. I think the NYT is absolutely right to refer to all long-term residents of Hawaii, whatever their ethnicity, as Hawaiian. If, god forbid, the British National Party's views were very popular in the UK, including the belief that all non-white people in Britain weren't really British, I should hope that the NYT would still refer to them as British, rather than pandering to the bigotry of the local population and calling them "resident coloureds" or whatever awful term the white population used.
If there's a need to distinguish haole from native Hawaiian, then they can call themselves "islanders" or something.


Posted by: ajay | Link to this comment | 06-23-09 5:15 AM
horizontal rule
395

Oy. I was suggesting forgiveness for the NYT in a context where there wasn't any point in drawing distinctions made by ethnicity. Saying that it's bigoted for Hawaiians to use their own language to differentiate between themselves and the people who showed up and annexed their country without asking a century and a half go, on the other hand, is going way too far for me.


Posted by: LizardBreath | Link to this comment | 06-23-09 5:35 AM
horizontal rule
396

Not for me. They're basically saying that someone could have lived on the islands all their life - and their parents and grandparents could have done the same - but they still aren't in some way of the islands because they're the wrong ethnicity. That's a pretty nasty belief and not one worth supporting. I wouldn't think much of an American, come to that, who suggested that a third-gen Hispanic US citizen didn't deserve to be called an American because his grandparents entered the country illegally.


Posted by: ajay | Link to this comment | 06-23-09 7:23 AM
horizontal rule
397

Returning to my off-topic:

I wonder if this is the comment quoted in Helprin's book:
Did anybody else get the impression that Helprin is espousing this policy because he is certain his works will become beloved classics, and if copyright is extended forever his ancestors will be wealthy aristocrats?

Not bad -- I think it holds up.


Posted by: peep | Link to this comment | 06-23-09 7:56 AM
horizontal rule
398

American doesn't, and never has, name an ethnicity. Would you call a Chinese person a bigot if they identified a third generation descendant of English missionaries in China as not Chinese?


Posted by: LizardBreath | Link to this comment | 06-23-09 7:56 AM
horizontal rule
399

397: Oh, no! I just realized I had meant to say his descendants would be wealthy aristocrats not his ancestors!!!

Helprin is right! I am a digital barbarian!


Posted by: peep | Link to this comment | 06-23-09 8:00 AM
horizontal rule
400

re: 398

The Chinese, being quite spectacularly racist, aren't necessarily the best example. But I get your point -- general usage on the whole nationalist/ethnicity dual-terminology thing is quite inconsistent.

FWIW, Scottish could be quite happily applied to, say, second generation Pakistani immigrants.


Posted by: nattarGcM ttaM | Link to this comment | 06-23-09 8:05 AM
horizontal rule
401

Sure, my hypothetical Chinese person might be racist for other reasons, but wanting to maintain a verbal distinction between ethnic Chinese and residents of China wouldn't be a reason itself.

In the Hawaiian context, the argument that it's bigoted for ethnic Hawaiians to refer to themselves exclusively as Hawaiian seems really similar to the sort of color/gender 'blindness' that Becks was griping about on the other thread. White Americans annexed their country without asking and now we're going to call them bigots for wanting to keep their name?


Posted by: LizardBreath | Link to this comment | 06-23-09 8:11 AM
horizontal rule
402

American doesn't, and never has, name an ethnicity

Yes, it does, and yes, it has. Certainly did in the sixteenth century. Certainly has done in the view of various more or less unpleasant politicians ever since the US was set up. And try asking someone in South America "what does a American look like?" They'll give you a very clear description.


Posted by: ajay | Link to this comment | 06-23-09 8:13 AM
horizontal rule
403

Does "Hawaiian Punch" refer to the state or the people? Inquiring babies are concerned.


Posted by: heebie-geebie | Link to this comment | 06-23-09 8:14 AM
horizontal rule
404

402: Somebody let the Irish in and screwed it up.


Posted by: Moby Hick | Link to this comment | 06-23-09 8:19 AM
horizontal rule
405

Certainly did in the sixteenth century.

I'm not following this -- in the sixteenth century, I don't have a strong sense that the word 'American' would have been used to describe anyone. The colonies from England hadn't been founded yet (unless I'm really confused, which is always a possibility), so there wasn't an English speaking population to be using the term. If you meant seventeenth century, I'd be really surprised if there were broad usage of 'American' to obviously exclude, say, farmers of Dutch origin in the NY area, or free blacks in New England, or so on.

Certainly has done in the view of various more or less unpleasant politicians ever since the US was set up.

I fucked up and said 'never'. Which means that I'm wrong, just like anyone who ever uses the word 'never' or 'always'. Still, I'd argue that 'unpleasant' in your sentence is doing real work there -- using 'American' to mean WASP-only has generally been a self-consciously tendentious usage, rather than a neutral distinction between American citizens of one ethnicity and another.


Posted by: LizardBreath | Link to this comment | 06-23-09 8:24 AM
horizontal rule
406

I'd disagree that it's a self-conscious usage, at least if you mean self-aware. Plenty of people screaming about immigration, or worried about American culture, etc., have a certain image in mind, and it's not a very inclusive one. White, non-Hispanic, Protestant. I know that foreign grad students often have a perception of what Americans look like (similar to how many Americans think England has no immigrants.)

Still, this doesn't solve the usage problem. Probably no rule is going to be good for every case. I'm inclined to believe that when a U.S. paper refers to issues, like North Korea getting nukes, that pertain to the whole population of a U.S. state, it's fine to use the adjectival form of the state to refer to residents of the state. When discussing ethnic diversity, probably best to use the word more restrictively.


Posted by: Cala | Link to this comment | 06-23-09 8:39 AM
horizontal rule
407

I'd disagree that it's a self-conscious usage, at least if you mean self-aware.

I'd say there's an unselfconscious default image of 'an American' as a white guy for lots of people, certainly. But, say, if you asked anyone if a black person with an American accent was an American, there's no one out there who'd unselfconsciously say "No, he's not American, he's black." And the same for people born in the US of Italian, or German, or Dutch origin -- there might be people who would say "No, they're not Americans", but very few people who'd do it, regardless of the century, without meaning to make a knowingly contested political point by it.


Posted by: LizardBreath | Link to this comment | 06-23-09 8:47 AM
horizontal rule
408

In the sixteenth century, I would reckon, "an American" would have been a native American. (More likely to be called an Indian, true.)

And it's not even the case that the people who think "American" means a certain ethnic type have to be terribly unpleasant: how many people, even in the US, would agree with, for example, the statement "Lucy Liu looks like a typical American woman"? More or less, would you say, than would agree with the same statement about Drew Barrymore?

Cala's talking sense, I think. "Hawaiian" is the adjective from "Hawaii", and it makes sense to use it for residents, just as you would for, say, plant life or scenery. In the rarer cases when you actually want to make a point about RACE, then find a different way of putting it. I would suggest "Kanaka", which is the Hawaiian word for "Hawaiian person".


Posted by: ajay | Link to this comment | 06-23-09 8:54 AM
horizontal rule
409

My take is that you get to use the ethnicity signifying place name if you have internalized at least some of the cultural elements of the relevant ethnic group. I'm African despite being genetically European because I have internalized a lot of cultural elements that belong to African cultures but not European ones. I'm not a Virginian, despite having lived there for a number of years, because I don't identify with the place, and I don't share the cultural attitudes that distinguish a Virginian from a New Yorker.

In the above I'm not talking about cultural things like food preferences or clothing styles: That stuff is epiphenomenal to what culture really is, which is attitudes towards other people and in particular the nature of the bonds between people and the correct way of interacting with others depending on their status within the cultural matrix. IOW the second generation Pakistani immigrant of ttaM's example is Scottish if he or she shares Scottish assumptions about who you drink with and who you stab. Just liking traditional Scottish food like curried haggis is insufficient.


Posted by: togolosh | Link to this comment | 06-23-09 8:57 AM
horizontal rule
410

I respectfully disagree. The experience of my friends who are visible minorities (people being surprised that they don't have an accent, etc) and of shiv (people are quite willing to say that Canadians don't really count as immigrants, as a blonde-haired blue-eyed guy no one thinks 'foreigner') indicates to me that not only is there a very strong default of what an American looks like, but that many, many people don't think that's particularly contested or a political point (and to be clear, I think it would be wrong for a paper to reify this usage by referring to an area as real America.) I think this is probably regional and influenced by one's proximity to a city.


Posted by: Cala | Link to this comment | 06-23-09 9:00 AM
horizontal rule
411

Cala: I'm inclined to believe that when a U.S. paper refers to issues, like North Korea getting nukes, that pertain to the whole population of a U.S. state, it's fine to use the adjectival form of the state to refer to residents of the state. When discussing ethnic diversity, probably best to use the word more restrictively.

ajay: Cala's talking sense, I think.

I'd agree that she is.

I would suggest "Kanaka", which is the Hawaiian word for "Hawaiian person".

No. That's the Hawaiian word for human being of any ethnicity, which was used by white people as identifying Hawaiians specifically (slur? Eh, I'm not sure of the level of offensiveness, either now or in the past. But using 'kanaka' to distinguish between Hawaiians and haoles is a white-person's usage, not a distinction that a Hawaiian speaking Hawaiian would make.)


Posted by: LizardBreath | Link to this comment | 06-23-09 9:04 AM
horizontal rule
412

In the past terms for ethnicity and terms for nationality (or for sub-national polities) tended to go together because there was an assumption that nations were ethnicities.

We tend to see that coming apart of these as being a consequence of migration and immigration, but, importantly, we also tend to see this coming apart as a good thing. The history of trying to enforce the identification of nation with ethnicity being, to say the least, pretty f'cking negative, and, it's usually a good rule of thumb that those who want to continue to identify nationality with ethnicity are, usually, people with pretty vile politics.

That gets more complicated in colonial and post-colonial situations, of course, but still, it's probably a good general rule to be suspicious of attempts to conflate nationality with ethnicity.

IOW the second generation Pakistani immigrant of ttaM's example is Scottish if he or she shares Scottish assumptions about who you drink with and who you stab.

Except, he or she probably doesn't. Nations are pluralities. Someone doesn't stop being Scottish just because they are Muslim and don't drink.


Posted by: nattarGcM ttaM | Link to this comment | 06-23-09 9:10 AM
horizontal rule
413

Huh. 410 seems reasonable to me, but doesn't seem to me to conflict with my 407. I'd absolutely agree that white-guy is the picture lots of people have of 'Americans', and that lots of people unthinkingly expect that anyone non-white and non-black is going to be an immigrant rather than born here.

But once someone (like Lucy Liu), who's non-white but also obviously not an immigrant is identified, I really think that anyone now taking the position that such a person isn't 'American' would be very uncommon, and someone thinking of that position as non-controversial would be incredibly uncommon. If you go back in time, that's going to differ depending on the ethnic groups you're talking about (Asians were socially and legally excluded from the concept of "American" for a long time) but I don't think there was a time when "American" identified one single ethnicity, rather than a group of ethnicities -- British Isles, German, Dutch, African-American, and so forth.


Posted by: LizardBreath | Link to this comment | 06-23-09 9:13 AM
horizontal rule
414

Yeah, I don't think "keeping it real" should be a requirement to claim a national designation. My parents are Cuban and don't particularly like to dance, talk over other people in conversations, or pray to the virgin.


Posted by: Criminally Bulgur | Link to this comment | 06-23-09 9:16 AM
horizontal rule
415

414 to 409, mostly.


Posted by: Criminally Bulgur | Link to this comment | 06-23-09 9:17 AM
horizontal rule
416

411: OK, good point; my mistake. But it still works, by analogy with "Inuit" which isn't the Inuit word for "Inuit" but the Inuit word for "person". If it's a slur, and I can't offhand see why it would be, it's no worse than "haole". Or just use "kanaka awy'i".

IOW the second generation Pakistani immigrant of ttaM's example is Scottish if he or she shares Scottish assumptions about who you drink with and who you stab.

This wasn't meant to be as offensive as it is, I assume, because you are unfamiliar with the horrible Norman Tebbit and his concept of the cricket test. To most Scots, this hypothetical person is Scottish if he was born there and identifies himself as such. End of story.


Posted by: ajay | Link to this comment | 06-23-09 9:21 AM
horizontal rule
417

awy'i

What's this? Google doesn't come up with anything, and it doesn't look like a Hawaiian word.

I'm getting shirty about this because names are important to people. Telling Hawaiians that they're bigoted for describing their ethnic group with the word in their language (or the English form of a word in their language) that describes their ethnicity, because white Americans took over their country and so we get to take their name as well, seems really unpleasant to me.

My instinct would be to do what Cala suggested above -- use Hawaiian generally as meaning 'resident of Hawaii' where that's all that's under discussion, but make the ethnic distinction where relevant. But if there's a problem with that, how about we let the Hawaiians keep their own name, and come up with a different word for residents of Hawaii, rather than calling them bigots for not renaming themselves to something we like better?


Posted by: LizardBreath | Link to this comment | 06-23-09 9:29 AM
horizontal rule
418

Oddly pertinent to 408-413 -- a hapa Japanese-American friend of mine wound up stopping at LL's house on a road trip with some other friends. They were all wearing kitschy Chinatown-themed sweatshirts. The person who knew her advised them to stow the shirts, as LL does not like that sort of thing.


Posted by: Wrongshore | Link to this comment | 06-23-09 9:29 AM
horizontal rule
419

417

... But if there's a problem with that, how about we let the Hawaiians keep their own name, and come up with a different word for residents of Hawaii, rather than calling them bigots for not renaming themselves to something we like better?

They don't have to rename themselves, they just have to tolerate a little ambiguity.


Posted by: James B. Shearer | Link to this comment | 06-23-09 9:33 AM
horizontal rule
420

Clearly, the Hawaiians have stumbled into a touchy area of Scottish politics, and vice versa. Mixing haggis and pineapple is never a good idea.

It looks to me like 100 percent of this dispute is caused by the fact that Scotland and Hawaii have very different histories and politics. Hawaiians are a colonized people who are now a small minority in their native land. It doesn't seem particularly offensive to me if they want to keep that name for themselves and, more importantly, this is in fact the usage people in the state of Hawaii, white, Asian, or native do use, at least according to NPH.

Being a "Scot" was more of an ethnic description than a political one from the Act of Union until very recently, but it's perfectly understandable that decent Scots would want to preserve the rights of immigrants to Scotland who've shown up in the past 50 years to call themselves Scottish and oppose racists who don't want that, especially as Scotland becomes more important as a political entity. It's dumb, though, to think of the two situations as directly comparable or as leading easily to a general rule about what people should be called.

These political/ethnic descriptions are contested because politics and ethnicity are contested issues (the same, btw, is true for "American" which has always been a highly contested term).


Posted by: Robert Halford | Link to this comment | 06-23-09 9:46 AM
horizontal rule
421

Rootless cosmopolitan as the ethnic identity of the future? My handful of Hawaiian friends left me with the impression that it's kind of a strange place for young people since property values are astronomical, making getting out on your own much harder than on the mainland. In some ways the ethnic attitudes they described reminded me of those held by people who only reluctantly left their neighborhoods.


Posted by: lw | Link to this comment | 06-23-09 9:48 AM
horizontal rule
422

My point in 409 is about ethnicity, not nationality. The Hawai'ian question that lead to this discussion is all about the ethnicity vs. geographical signifier. To use a signifier that combines the two as if the ethnicity it signifies is irrelevant seems to me to be straight out of the colonialist playbook, and native Hawai'ians are justified in objecting to it.

The counter argument as far as I can tell seems to be that culture is irrelevant, and the only thing that matters is who issues your passport. I can't wrap my head around this idea except as a noble lie to frustrate the creeping fascism that places ethnicity above all else*. I argue that the reality is complicated and depends on details to such a degree that there is no way to create an accurate generalized big picture view. IOW it's perfectly possible to live in a place and partially or completely adopt the ethnicity of the locals, and it's quite possible to segregate yourself and completely ignore the local culture. Which you choose seems to me important in deciding your ethnic identity, and the issuance of government papers has nothing to do with it.

* I've done this in other forums in arguments with crypto-fascists, but it's a huge oversimplification and a rhetorical ploy, not a statement of fact.


Posted by: togolosh | Link to this comment | 06-23-09 9:49 AM
horizontal rule
423

re: 420

It's not really a Scottish thing. You only have to look a the history of Europe through the 20th century, large swathes of Africa, and chunks of the Middle East to see that the identification of a particular nation or polity with ethnicity is problematic.

You're right that colonized nations are probably different, but I'd be pretty keen to argue for the general rule -- with possible exceptions -- rather than to ditch it.


Posted by: nattarGcM ttaM | Link to this comment | 06-23-09 9:51 AM
horizontal rule
424

Also, Ajay@416.last - I apologize if my example was offensive. I'm thinking about ethnicity in African terms and using examples from Hawai'i and Scotland, which is stupid. Unfortunately using the better examples requires a lot of backstory, which tends to bog discussion down.


Posted by: togolosh | Link to this comment | 06-23-09 9:54 AM
horizontal rule
425

Macedonia for Macedonians!


Posted by: JP Stormcrow | Link to this comment | 06-23-09 9:54 AM
horizontal rule
426

it's perfectly possible to live in a place and partially or completely adopt the ethnicity of the locals, and it's quite possible to segregate yourself and completely ignore the local culture. Which you choose seems to me important in deciding your ethnic identity, and the issuance of government papers has nothing to do with it.

I just think this is conflating "ethnicity" and "local culture" in a way that is atypical. An ethnicity can have a local culture associated with it, but typically it is something that you are born into, whereas a local culture is something you can come to adopt or not. The issue of local culture just seems orthogonal to, for example, the Hawaii issue.


Posted by: Criminally Bulgur | Link to this comment | 06-23-09 10:14 AM
horizontal rule
427

423 -- Of course you're right that the nation/ethnicity connection is extraordinarily problematic in general (really, is there anything that's been a bigger problem in world history, ever?). And I think we probably share the same general cosmopolitan values. My point is just that picking a name for someone's national or ethnic group is a complicated procedure, and one that often can't be done without knowing a lot of the backstory. For example, is an ethnic Hungarian in what's now Romania Hungarian or Romanian? Or both? Can you call yourself Swiss (not a nationality based on an ethnicity) if you haven't lived there for a few generations (when I lived there, the answer seemed to be "no," which, oddly, was something of a source of pride for my immigrant friends). Are indigenous folks in Chiapas Mexicans? Are Navajos Indians or "native Americans"?

These are pretty context-specific issues, and so it's worth paying attention to someone, like NPH, who actually knows what's going on in a particular location rather than just generalizing from your own experience or the politics of the place where you live.


Posted by: Robert Halford | Link to this comment | 06-23-09 10:16 AM
horizontal rule
428

426: Depends a lot on the local rules. In Samoa, a white person who married into a Samoan family and lived according to the local rules was considered pretty much a local, and their kids were as well. (Kids would be identified as mixed -- afakasi -- but seemed to be fairly thoroughly culturally integrated.)


Posted by: LizardBreath | Link to this comment | 06-23-09 10:19 AM
horizontal rule
429

In Samoa, a white person who married into a Samoan family and lived according to the local rules was considered pretty much a local, and their kids were as well.

But the marriage is establishing a kind of kin relation (i.e., entering a Samoan family) that is the crucial determiner, right? Or would simply learning the language, eating the food, etc., have the same effect?


Posted by: Criminally Bulgur | Link to this comment | 06-23-09 10:26 AM
horizontal rule
430

Yeah, you'd have to be family somehow, either marriage or 'adoption' (that is, a young single person moving in with a family). And you'd still be a palagi, but you'd also be part of the Samoan cultural/family system. So, not exactly what tologosh was talking about -- mostly I was thinking that for anyone who actually wanted to be adopted into the Samoan system, it'd be possible. Samoan culture can be hostile to outsiders, but it's pretty welcoming to people from other ethnicities who want in, and are willing to be wholehearted about it.


Posted by: LizardBreath | Link to this comment | 06-23-09 10:31 AM
horizontal rule
431

Samoans do the humpty dance.


Posted by: heebie-geebie | Link to this comment | 06-23-09 10:40 AM
horizontal rule
432

In Tswana culture you can become a Batswana simply by adopting the appropriate customs and attitudes, but that does require entering the cultural matrix in a way that is only possible with the active assistance of other Batswana. One of my sister's good friends from High School is white but absolutely Tswana. In fact, she's a sangoma (traditional healer, witch doctor, pick your preference), trained in the traditional manner. My sister, OTOH, is a semi-tswana, recognized as such by everyone. Despite being a christian she had her house placed under a protective spell by her sangoma friend, f'rex. I am culturally African in a sort of generic sense that is only really intelligible if you are familiar with the type (there's a surprising number of us out there), but not semi-tswana in the way my sister is.

429:Or would simply learning the language, eating the food, etc., have the same effect?

Ethnicity isn't food, music, language, or art. It's attitudes towards other people, values, assumptions about the way the world works. Art, music, language, table manners, all the rest, only exist to support, reinforce and express attitudes and values.


Posted by: togolosh | Link to this comment | 06-23-09 10:55 AM
horizontal rule
433

432: Hey, what do you think of those Alexander McCall Smith Botswana mysteries? I'm always kind of hesitant about 'white guy writes as insider in a culture not his own' books -- are they accurate and inoffensive, or are they something that Tswana are generally annoyed by?


Posted by: LizardBreath | Link to this comment | 06-23-09 11:00 AM
horizontal rule
434

And it's not even the case that the people who think "American" means a certain ethnic type have to be terribly unpleasant: how many people, even in the US, would agree with, for example, the statement "Lucy Liu looks like a typical American woman"?

Doesn't that just prove that terribly unpleasant people are everywhere?


Posted by: Jesurgislac | Link to this comment | 06-23-09 11:00 AM
horizontal rule
435

But once someone (like Lucy Liu), who's non-white but also obviously not an immigrant is identified, I really think that anyone now taking the position that such a person isn't 'American' would be very uncommon

LB's contention here surprises me, as I've had people come at me from this position all my life. More often than not, these people don't mean to be contentious -- it generally takes just a bit of gentle correction to set folks straight -- but the position that native-born citizens of Asian (and probably Middle Eastern and non-white Latino as well) descent are not "American" is not at all uncommon.


Posted by: jms | Link to this comment | 06-23-09 11:12 AM
horizontal rule
436

435: This is probably urban naivete, but I am honestly surprised; seriously people saying 'you're not American' despite knowing you were born here rather than people ignorantly assuming you were an immigrant?

The bit that I was really arguing, though, was that 'American' has always been a multi-ethnic identity, not that it's always included all ethnicities. I did know, for example, that mid 20th C, Asian-Americans would have been commonly regarded by white Americans as not actually American; I'd just thought this was more resolved than it is.


Posted by: LizardBreath | Link to this comment | 06-23-09 11:21 AM
horizontal rule
437

It's not "you're not american" but "you don't look like the typical american woman".


Posted by: nosflow | Link to this comment | 06-23-09 11:22 AM
horizontal rule
438

Well, that's going to be true of someone if the speaker has a picture in their head of the typical American woman at all. I think it's possible for someone who has a picture of a 'typical American woman' as a blue-eyed blond with a toothy grin, to nonetheless think of American as a multiethnic identity -- not consistent, maybe, but at a very ordinary level of inconsistency.


Posted by: LizardBreath | Link to this comment | 06-23-09 11:29 AM
horizontal rule
439

"you don't look like the typical american woman"

A Google image search for typical American woman (without quotes and moderate safe search on) reflects disagreement as to whether the typical American woman is thin and barely dressed or overweight. However, the second page of results includes this.


Posted by: apostropher | Link to this comment | 06-23-09 11:32 AM
horizontal rule
440

436: Honestly, I don't see that there's much of a difference, at bottom. People don't say to me, "you're not American," and defy me to contradict them. I'm not talking about people who go out of their way to be assholes. But even perfectly nice people not-infrequently do, and have always, assumed that I'm not an American citizen, or talked about "Americans" in a way that clearly excluded my family, and then when corrected, apologized that "well, of course you're an American citizen, but you know what I mean," etcetera.

Relatedly, I once took one of those online unconscious bias tests, I think from the Yale psychology department? This one asked you to differentiate between "American" and "foreign" faces. It presented you with well-known Asian-American faces (LL, Kristi Yamaguchi, Lieutenant Sulu, etc.), and well-known foreign white faces (Hugh Grant, Catherine Deneuve, etc.) and asked you to match them up as American or non-American. I totally failed it.


Posted by: jms | Link to this comment | 06-23-09 11:42 AM
horizontal rule
441

For the record, back in the day (the 80's), I knew a (ethnically) Hawaiian woman online (and she went by Hawaiian Punch) who explained lots of Hawaiian stuff to me... and sadly disclosed the fact that haole meant something between 'honky' and 'nigger'. However, as LB pointed out, the Hawaiian people were sort of conquered, driven back and otherwise killed by white men in the process of semi-colonizing Hawaii. So there is a lot of bigoty and racism against whites in Hawaii, but note why that enmity exists in the first place before you go high-horsing about. (There's a lot of anti-white racism amoungst blacks, hispanics and indians in the US as well, often attributable to the exact same causes.)

As someone who has too many relatives on the wrong side of the color line (both genetically and culturally) for southern comfort, while at the same time being a gray-eyed, blond white boy, I've frequently caught it from both directions, so I know it exists. I also know that I find the groups with the ancestors that have recently been enslaved, butchered and otherwise treated badly to be much more forgivable.

I'd also say that the usage 'American' has been generally understood to always refer to white people up until quite recently. Only post-1960 (or maybe post-1948) has it come to mean something other than 'white'. (Black men serving in Europe during WWII tended to be surprised to be called Americans by europeans.) Old people who came of age before 1960 probably still understand it that way, along with gobs and gobs of Southerners.

max
['I'm not sure how this fits in this here discussion, but I'll throw it out anyways.']


Posted by: max | Link to this comment | 06-23-09 11:46 AM
horizontal rule
442

People don't say to me, "you're not American," and defy me to contradict them. I'm not talking about people who go out of their way to be assholes.

But that's all I'm saying -- that someone who was explicitly taking the position that a non-white/non-black person born in the US wasn't American would be clearly being an asshole, and would be knowingly taking a contentious position about what 'American' means.

But even perfectly nice people not-infrequently do, and have always, assumed that I'm not an American citizen, or talked about "Americans" in a way that clearly excluded my family, and then when corrected, apologized that "well, of course you're an American citizen, but you know what I mean," etcetera.

This is people being thoughtlessly racist, and it's people whose mental image (although probably not their intellectual definition) of 'American' excludes some ethnic groups. But the fact that the intellectual definition of 'American' most people have is pan-ethnic (and for almost everyone it's at least multi-ethnic -- the same white people questioning your Americanness wouldn't think of questioning Michelle Obama's Americanness) means something itself, regardless of whether their unconscious biases line up with their intellectual beliefs.


Posted by: LizardBreath | Link to this comment | 06-23-09 11:52 AM
horizontal rule
443

Huh. Crossed with, and flatly contradicted by, max.

Maybe I'm living in a fantasy world, but I really wouldn't have thought there was a period where 'American' was explicitly understood by any large part of the population to exclude African Americans. (Excluding the pre-Civil War South, where I'm not sure.)


Posted by: LizardBreath | Link to this comment | 06-23-09 11:55 AM
horizontal rule
444

What then is the American, this new man? He is either an European, or the descendant of an European, hence that strange mixture of blood, which you will find in no other country. I could point out to you a family whose grandfather was an Englishman, whose wife was Dutch, whose son married a French woman, and whose present four sons have now four wives of different nations. He is an American, who leaving behind him all his ancient prejudices and manners, receives new ones from the new mode of life he has embraced, the new government he obeys, and the new rank he holds by being received in the broad lap of our great Alma Mater. Here individuals of all nations are melted into a new race of men, whose labours and posterity will one day cause great changes in the world.

Sorry, blacks and Asians!


Posted by: J. Hector St. John De Crevecoeur | Link to this comment | 06-23-09 11:58 AM
horizontal rule
445

Maybe I'm living in a fantasy world Manhattan.


Posted by: rob helpy-chalk | Link to this comment | 06-23-09 11:59 AM
horizontal rule
446

Heh. I'm going to take 44 as confirming my initial statement that "American" doesn't name an ethnicity.


Posted by: LizardBreath | Link to this comment | 06-23-09 12:04 PM
horizontal rule
447

I don't see the distinction in 445.


Posted by: soup biscuit | Link to this comment | 06-23-09 12:06 PM
horizontal rule
448

Rob was just being more specific.


Posted by: nosflow | Link to this comment | 06-23-09 12:07 PM
horizontal rule
449

447: I suppose you understand that this means war.


Posted by: LizardBreath | Link to this comment | 06-23-09 12:07 PM
horizontal rule
450

446: Sifu Tweety asking about being mire Bayesian?


Posted by: Jesurgislac | Link to this comment | 06-23-09 12:08 PM
horizontal rule
451

Argh. 444.


Posted by: LizardBreath | Link to this comment | 06-23-09 12:08 PM
horizontal rule
452

433: The McCall Smith books are popular in Botswana, and he's considered by Batswana to be "one of us" in the same sense that my sister is (My mom knows him and some of my sister's quilt work was bought by the production company for the movie, though I don't know if it's actually in the movie). They are accurate in both physical and cultural detail AFAICT, but I haven't read them myself, nor will I. My relationship to the place is complicated and emotionally something: Whatever the word or phrase is that means there is enormous emotional energy with poorly defined orientation, intensity cranked up to eleven but which emotion it is shifts like quicksand. Reading things that tap into that is bad for my mental health. I love that I'm from there, but you're only really *from* a place when you're away.


Posted by: togolosh | Link to this comment | 06-23-09 12:09 PM
horizontal rule
453

I know nothing -- except that my party is the American Party and that I really, really hate the Irish.


Posted by: Opinionated 19th Century American Party Member | Link to this comment | 06-23-09 12:14 PM
horizontal rule
454

Whatever the word or phrase is that means there is enormous emotional energy with poorly defined orientation, intensity cranked up to eleven but which emotion it is shifts like quicksand.

I diagnose you with labile cathexis disorder.


Posted by: nosflow | Link to this comment | 06-23-09 12:15 PM
horizontal rule
455

Huh. Crossed with, and flatly contradicted by, max.

I'm sorry!

Maybe I'm living in a fantasy world, but I really wouldn't have thought there was a period where 'American' was explicitly understood by any large part of the population to exclude African Americans. (Excluding the pre-Civil War South, where I'm not sure.)

If you want some high-quality entertainment (of a sort), may I strong recommend The Road to Disunion: Secessionists at Bay. Good stuff.

But yeah, more so than I would have thought, that's what American has meant. First it meant the people of the various states of British descent, and then after awhile it meant non-obviously immigrant Protestant whites, and then it seemed to mean non-obviously immigrant whites, ex Germans and Italians, and then it meant non-immigrants of most colors and so on.

Manhatten has always been Dutch at the core, so NYC tends to be (unusually) not so much racist as classist. Or so I would gather from my Grandmother (RIP, late of NYC, Weehawken and Cape Cod). After all, she married a black man, and liked to explain to me how Hitler would have loved me for being blond and so on. (She also liked to explain how the Hindenberg flew over her house and she heard it blow up.) Except, of course, she deeply disliked Asians because of Pearl Harbor, the Korean War and Vietnam.

Dallas is a sort of southern NYC in a way; color matters (depending), but much less so than just exactly how much money you have, which is very very important indeed.

So, at any rate, it (still) surprised the fuck out of me to see Arkansas, Tennesse and WV go out of their way to diss Obama, but, well, I did hear those Appalachian persons, and they will not have to tell me again. Since the election, I've been paying very sharp attention to all this code word nonsense, and whaddya know.

max
['They ain't kiddin'.']


Posted by: max | Link to this comment | 06-23-09 12:19 PM
horizontal rule
456

451: Wimp. You could have come up with some tortured rationale that we could all have waterboarded, pretending it was an evil mouse.


Posted by: Jesurgislac | Link to this comment | 06-23-09 12:24 PM
horizontal rule
457

Has anyone bothered to determine whether the sign in 439 says "hungry" or "angry"?


Posted by: Wrongshore | Link to this comment | 06-23-09 12:37 PM
horizontal rule
458

"Angry". Neither a "u" nor an "h" would create the little angled bit of red we see near the immigrant vagina–bearer's head.


Posted by: nosflow | Link to this comment | 06-23-09 12:41 PM
horizontal rule
459

457: Could be aggry, begry, conyngry, gry, higry-pigry, iggry, meagry, menagry, nangry, podagry, puggry, or skugry.


Posted by: Jesurgislac | Link to this comment | 06-23-09 12:44 PM
horizontal rule
460

Why, I was just at that same page, Jes. Meagry makes the most sense in context.


Posted by: apostropher | Link to this comment | 06-23-09 12:45 PM
horizontal rule
461

Except, from that list, it could only be conyngry or nangry.


Posted by: apostropher | Link to this comment | 06-23-09 12:46 PM
horizontal rule
462

more importantly, this is in fact the usage people in the state of Hawaii, white, Asian, or native do use, at least according to NPH

For the record, this is all I really started out to say. Hawaiian activism as a political movement is problematic (as opposed to the cultural renaissance, which is pretty much unequivocally a good thing), as is most of the mainland-funded counter-activism. But just as a matter of usage, it doesn't make a lot of sense to refer to a group of people by a name they wouldn't recognize as applying to themselves.

So there is a lot of bigoty and racism against whites in Hawaii

Really not my experience at all. There's some places you'd want to be careful if you don't belong, there's some stuff in some schools that isn't good, and you run into a few people who would pretty apparently be happier to see you if you weren't white, but it's mostly pretty minor stuff. I think most of the reputation to the contrary comes from people who spend time here without ever making a connection to local culture, particularly military people and tourists of a certain sort.


Posted by: Not Prince Hamlet | Link to this comment | 06-23-09 12:57 PM
horizontal rule
463

452
My relationship to the place is complicated and emotionally something: Whatever the word or phrase is that means there is enormous emotional energy with poorly defined orientation, intensity cranked up to eleven but which emotion it is shifts like quicksand.

Ambivalence?


Posted by: Cyrus | Link to this comment | 06-23-09 1:17 PM
horizontal rule
464

462: There's some places you'd want to be careful if you don't belong, there's some stuff in some schools that isn't good, and you run into a few people who would pretty apparently be happier to see you if you weren't white, but it's mostly pretty minor stuff.

Fair enough, since you would know better. Punchy explained it at some length, but that was 22 years ago, she was careful to say she didn't think that way (it was the relatives, etc.) and honestly, she didn't speak English that well. (On the phone, she tended to lapse into Hawaiian quite frequently when groping for the english equivalent.)

max
['{shrug}']


Posted by: max | Link to this comment | 06-23-09 1:21 PM
horizontal rule
465

464: I think there probably was more of some sorts of bad behavior directed at white people 20 years ago (also more white immunity from any sort of mistreatment if you were the right sort of white person).

On language, was it Hawaiian or pidgin that she lapsed into? AFAIK native speakers of Hawaiian are thin on the ground these days and would have been more so 20 years ago. But strong pidgin (which is more of working/lower class thing than a Hawaiian thing) can be pretty hard to understand.


Posted by: Not Prince Hamlet | Link to this comment | 06-23-09 1:38 PM
horizontal rule
466

457: after all the effort the photographer went to, to get that nice ambiguous shot, 458 sure is a letdown.


Posted by: Hamilton-Lovecraft | Link to this comment | 06-23-09 1:39 PM
horizontal rule
467

463: Perhaps passionate ambivalence captures it. Deconstructed, ambivalence is perfect, but as used in practice the word tends to mean "sort of meh," which is exactly wrong.

In 454 nosflow is, as usual, pretentiously correct and almost completely useless. That fits my needs a lot of the time though, so labile cathexis is my working descriptor.


Posted by: togolosh | Link to this comment | 06-23-09 1:39 PM
horizontal rule
468

My first thought on looking at the pic in 439 was that the person holding the sign doesn't look like an immigrant - she looks like she could well be AmerIndian. Then I realized I had no idea where the picture was taken. Probably Borneo.


Posted by: togolosh | Link to this comment | 06-23-09 1:48 PM
horizontal rule
469

461: Conyngry.


Posted by: Jesurgislac | Link to this comment | 06-23-09 1:52 PM
horizontal rule
470

469 -- I think you're right. The protest babe in 439 was clearly Mary Tofts.


Posted by: Robert Halford | Link to this comment | 06-23-09 2:01 PM
horizontal rule
471

Tofts was forced to admit on 7 December 1726 that she had manually inserted dead rabbits into her vagina and then allowed them to be removed as if she were giving birth.

Wow.


Posted by: nosflow | Link to this comment | 06-23-09 2:05 PM
horizontal rule
472

465: On language, was it Hawaiian or pidgin that she lapsed into? AFAIK native speakers of Hawaiian are thin on the ground these days and would have been more so 20 years ago. But strong pidgin (which is more of working/lower class thing than a Hawaiian thing) can be pretty hard to understand.

When I was writing 464 I started to say that her English sounded like pidgin, so it's entirely possible her Hawaiian was pidgin. (Very strong accent.) She also told me (true or not, do not know) that her folks came from way backwoods or maybe the reservation.

Family had moved to the big island and she married some guy there, and then wound up in Oahu. So she had computer access due to his miltary aceess. Thus, bored housewife chonicles, or pissed off housewife chronicles. I vaguely remember her referring to him as something like 'that fucking bastard haole' and then immediately apologizing to me.

max
['She was so mad.']


Posted by: max | Link to this comment | 06-23-09 2:06 PM
horizontal rule
473

she had manually inserted dead rabbits into her vagina

As opposed to having them mechanically inserted, one presumes.


Posted by: apostropher | Link to this comment | 06-23-09 2:08 PM
horizontal rule
474

There are no reservations in Hawaii. Closest thing would be the island of Niihau, which is privately owned, closed to outsiders, and populated entirely by Hawaiians. If her family was from there, that would make sense.

IME "haole" is generally used neutrally. It's when there are adjectives attached that you have to watch out.


Posted by: Not Prince Hamlet | Link to this comment | 06-23-09 2:09 PM
horizontal rule
475

What's the big deal? I heard from ogged that 40% of women have manually inserted rabbits into their vaginas at one time or another.


Posted by: Hamilton-Lovecraft | Link to this comment | 06-23-09 2:13 PM
horizontal rule
476

And, to be clear, there are lots of subcultures here. I've lived here for the better part of 20 years and my wife and many of my friends grew up here, but someone else's experience of the place could be totally different.


Posted by: Not Prince Hamlet | Link to this comment | 06-23-09 2:14 PM
horizontal rule
477

475: Well, duh, you need bait to get the dogs to lick properly.


Posted by: Not Prince Hamlet | Link to this comment | 06-23-09 2:15 PM
horizontal rule
478

manually inserted rabbits

Prestidigitation has gone too far!


Posted by: LizardBreath | Link to this comment | 06-23-09 2:16 PM
horizontal rule
479

Prestidigitation has gone too far!

"Manually" means with the hands, not with the digits, LB.


Posted by: nosflow | Link to this comment | 06-23-09 2:18 PM
horizontal rule
480

The best fake rabbit births are hand crafted.


Posted by: rob helpy-chalk | Link to this comment | 06-23-09 2:20 PM
horizontal rule
481

Hey Rocky, watch me pull a rabbit out of my vagina!


Posted by: rob helpy-chalk | Link to this comment | 06-23-09 2:23 PM
horizontal rule
482

479: So, you're supposed to palm the rabbit like a basketball? This sounds difficult.


Posted by: LizardBreath | Link to this comment | 06-23-09 2:23 PM
horizontal rule
483

It is difficult. It's not for everyone. But the few who've mastered the trick can attest that it brings uncommon satisfaction.


Posted by: nosflow | Link to this comment | 06-23-09 2:26 PM
horizontal rule
484

palm the rabbit like a basketball

SLAM DUNK


Posted by: apostropher | Link to this comment | 06-23-09 2:27 PM
horizontal rule
485

Here's the Hogarth print mentioned in the Wikipedia entry on Tofts.

William Hogarth published the print "The Cunicularii or the Wise Men of Godalming in Consultation", showing St André and Manningham witnessing Mary "giving birth" while John Howard skulks at the door turning down an offered rabbit with the words "it's too big".


Posted by: JP Stormcrow | Link to this comment | 06-23-09 2:43 PM
horizontal rule
486

NPH: There are no reservations in Hawaii. Closest thing would be the island of Niihau, which is privately owned, closed to outsiders, and populated entirely by Hawaiians. If her family was from there, that would make sense.

That's what I think she meant, but not entirely sure. (Keep in mind, she typed like a modern teenager txts (only the accent and the extra Hawaiianish vocab was a surprise), which is why she got nicknamed Punchy in the first place.) She would explain to the haole the whole fake Hawaii vs. real Hawaii thing and go from there to explaining that her folks were pure Hawaiian and so on. Hard to say if it was true, but she certainly acted as though it were.

And, to be clear, there are lots of subcultures here. I've lived here for the better part of 20 years and my wife and many of my friends grew up here, but someone else's experience of the place could be totally different.

's fine. It was one of those weird online things where racial AND ethic AND class AND geographic boundaries all get crossed at once, and so communications routinely got crossed up. As it comes back to me, I'm pretty most of what she said agreed with what you've said, excepting that she did think there was a lot of rcism in Hawaii. (I.e. she emphasized that she wasn't one of those Hawaiians that hated white people, which surprised white people who didn't know there were Hawaiians that hated white people to begin with. 'Why would they?') That kind of cultural crossup doesn't happen much anymore because there are enough people online such that they can all cluster together with their own kind. So if she's online these days I'd imagine she talks mainly to other ethnic Hawaiians.

max
['Or working class Hawaiians. Or something like that.']


Posted by: max | Link to this comment | 06-23-09 3:14 PM
horizontal rule
487

From 470:

Tofts was forced to admit on 7 December 1726 that she had manually inserted dead rabbits into her vagina and then allowed them to be removed as if she were giving birth.

Aaaaaah!


Posted by: heebie-geebie | Link to this comment | 06-23-09 3:44 PM
horizontal rule
488

Oh wow. Never have I been so pwned.


Posted by: heebie-geebie | Link to this comment | 06-23-09 3:45 PM
horizontal rule
489

488: That's the problem when you take time to do research.


Posted by: Jesurgislac | Link to this comment | 06-23-09 4:03 PM
horizontal rule
490

448: It was more meaningful coming form your than nose-flow.


Posted by: rob helpy-chalk | Link to this comment | 06-23-09 4:11 PM
horizontal rule
491

490 was so well-intentioned that I'll overlook the startingly high density of typos.


Posted by: heebie-geebie | Link to this comment | 06-23-09 4:14 PM
horizontal rule
492

||

OT: Is there any kind of book review thing. I'd be very interested in discussing Hal Luft's new book Total Cure: The Antidote to the Health Care Crisis. The bullet point version is available on his website.

I'd be very interested to know pain perdu's thoughts and soup's, in particular.

|>


Posted by: Bostoniangirl | Link to this comment | 06-23-09 4:17 PM
horizontal rule
493

Aaaaaah!

I'm hoping that's a "that's horrifying"-type aaaaah and not a "that's so satisfying"-type aaaaah.


Posted by: M/tch M/lls | Link to this comment | 06-23-09 4:56 PM
horizontal rule
494

492: it's not incremental. It's close to single-payer. If you're allowed to wipe the slate clean and start over then it's easy to improve on what we have. But that's not going to happen.

It might be something we gradually evolve towards -- the end of health underwriting (at least, aboveboard health underwriting) starts the movement toward a single risk pool. Payment and delivery system reform could happen through a public plan that uses its size and market power. Political independence is a possibility for a public plan. Political independence for Medicare could be achieved by giving MedPac more power. People are talking about that but it looks like it isn't going to happen any time soon. Giving someone independent the power to cut physicians reimbursements (a 20 percent cut if you followed the "sustainable growth rate" guidelines) is giving them the power to get you kicked out of office.


Posted by: PGD | Link to this comment | 06-23-09 5:10 PM
horizontal rule
495

493: Have you never stuck dead rabbits in your vagina, M/tch? If you had, you would know how satisfying it is to have them removed again.


Posted by: Otto von Bisquick | Link to this comment | 06-23-09 5:20 PM
horizontal rule
496

Thanks PGD. A huge portion of it is not incremental. I liked the idea of separating primary care docs from specialist payments so that the powerful specialists would not have as much of a say. I also liked that it gave PCPs the ability to charge what they want and have that reflected in their fees/premiums. It would allow doctors who want to spend time with their patients instead of ordering tests do that. It seemed more interesting than Medicare for All.

I don't think that it will happen now, but I wish that it was getting more air time. That stupid 2% solution book was everywhere.


Posted by: Bostoniangirl | Link to this comment | 06-23-09 7:06 PM
horizontal rule
497

494: Further, it looked to me like this got away from the SGR model by making it possible for providers to earn more from their expertise while doing fewer tests.


Posted by: Bostoniangirl | Link to this comment | 06-23-09 7:08 PM
horizontal rule
498

I am panting up from the street-corner behind you all because this

"' being feminine is on some level the same sort of task as lawyering.'

is clearly true for me -- being feminine is more work for me, less natural, than whatever it is I do at work. (Switch browsers to get through the faffing-off limits I set up for myself, yeah, but whatever); much more work.

To vaguely revisit a thread from some time ago that I was expecting to go in a completely different direction than it did, one of the most interesting (yet mutually annoying!) conversations I've ever had about sexuality, or gender, was with a transitioning woman. We both have strong gender identities, and the gender identity in question is called female in both cases, but I think it demonstrates that I shouldn't have to act feminine, and she thought it demonstrated that she needed to learn how. Definitely performance for both of us. We spent some time trying to work out what it was that was a female identity, if it wasn't femininity, but we didn't get much to agree on.


Posted by: clew | Link to this comment | 06-23-09 7:38 PM
horizontal rule