If this is in fact all I am doing you should be able to show it, not tell it. You have not yet been able to do so.
Please. You keep calling it a "polite proposition." That's such insanely jaded and unsupported language that it poisons anything that comes after. It's also a concept for which you have no data.
What percent of "polite propositions" lead to harrassment significant enough to merit a request that they not occur?
I think perhaps you should just stay away from arguments using analogies, because your use of them seems to be confusing even yourself.
[...]
Nobody has argued that fear is unjustified in every possible scenario, for example, so bringing up the scenario of someone accosting you with a hand in their jacket to illustrate the point that fear in some scenarios is justified is irrelevant and unproductive.
Haha. That's a pretty weak effort at trying to slink out of such a silly mistake.
You keep making the mistake. That analogy wasn't about fear, it was about dealing with the proper statistical reality given the available facts. It's really not complicated.
You are wrong. Not as a matter of opinion, but as a matter of mathematical fact. In any universe where probability works as it does in ours, P(A|B)=P(B|A)P(A)/P(B). P(A|B) does not equal some entirely new value totally unrelated to P(A).
If you don't even grasp this most fundamental tenet of applied epistemology, I don't think you're ready for the more advanced scenarios where you do get to exclude some facts which might have previously informed P(A) in light of new information.
Sure, I looked at what I wrote. That was incorrect. Is it not "irrelevant." That's why I deleted that section.
The issue with respect to Bayes is the lack of quality in our baseline estimate.
Is it the perecent of PEOPLE who are sexually assaulted when they ride the elevator? Is it the percentage of WOMEN? Is it the percentage of WOMEN when they're alone with men?
We can include or exclude an enormous number of situations that radically alter that baseline statistic. Assuming we could ever establish the possibility of being assaulted after a cold propositioning of a stranger, our percent of even jsut "women assaulted on hotel elevators" as a baseline can change drastically.
If we include all elevator rides, that percent for assaults will be tiny. If we just exluded all isntances of women riding the elevator alone, our initial estimate shoots up pretty significantly (in proportion to the original number).
Given that we have exactly zero numbers on any of this, and the claims minimizing the incidents of sexual assault on hotel elevators are deeply flawed, I don't see any grounds for declaring that apprehension irrational.
So stop presenting irrelevant or trivial analogies as being relevant to the point we are actually trying to discuss, if every time we pin you down it turns out that your analogy was irrelevant or trivial.
The problem is not that we are being childish, it's that we are insisting on relevance and you seem unable to present analogies, arguments or evidence that are actually relevant.
No, the problem is that you're trying to an analogy directed at one issue to all issues on a subject. It's just a ridiculous thing to do. I don't understand why you keep doing it.
There's not even an argument, here. Yes, I agree, applying the analogies I offered in the way you're suggesting yeilds inapplicable results.
The tutu straw man you are attempting to engage with is absurd because the probability of being robbed or assaulted or whatever is unaffected whether or not you are wearing a tutu. P(A|B) where A is "I will be robbed" and B is "I am wearing a tutu" equals P(A). You can assemble any number of such irrelevant factors - I am wearing my underpants inside out, I am humming a song by Queen, my nose is itchy - and none of them will significantly modify P(A).
I think you have my argument backwards. This is a discussion that was imported from another thread, so I'm not sure that you've read what lead to it.
It's a spin-off of Dawkins claim that an elevator was a safe place. He expressed befuddlement that someone could be concerned about assault on elevator--just get off on the next floor.
The introduction of all of the rapes and assaults that occur on elevators are meant to reject the notion that it's a uniquely safe place. Because it's an isolated environment, it's a more likely place for an assault to occur than a bar, or even the hallway outside the elevator, empty or otherwise. It's probably a less likely place than a secluded alleyway or a wide range of other places.
I would say that the elevator is likely a factor that doesn't make an assault more or less LIKELY, it makes it more unavoidable should it occur. It's a factor that heightens the apprehension, not the source of it.
Whether or not a person is more vulnerable in a lift is not irrelevant. In fact this thread is all about that topic, and you have been arguing that indeed people are more vulnerable in a lift. P(A|B) where A is "I will be sexually assaulted" and B is "I have been politely sexually propositioned in a lift at a conference for skeptics" might or might not equal P(A), and it might or might not be so significantly higher than P(A) that a rational person would be concerned. We are trying to sort that issue out. You are not helping by arguing out of one side of your mouth that it is irrelevant and out of the other that it is cause for justified concern.
More vulnerable in a lift that a given number of alternative scenarios, like a bar. Are you arguing against this point? You think you are safer from assault alone in an elevator than in a public place, all things being equal?
Notice that you're once again setting up "B" in such a way that you're likely to get the result you want, even though you actually have no data for that situation.
"B" is propositioning someone you've never talked to before.
Aha! A tiny bit of light dawns. P(rape|in a lift at a skeptic's conference) is indeed probably higher than P(rape|in a crowded bar). You have taken a step in the right direction.
Now the question is, what is the actual value of P(rape|in a lift at a skeptic's conference)? Is it of sufficient magnitude to merit rational concern?
First we would have to know what rate would generate a rational concern. Then we would have to know the intermediate steps sufficient to generate concern and their likelihood.
Again, assault is just the unavoidable worst-case scenario.
Where do I start? I guess anywhere is as good as anywhere else with a paragraph this terrible. To begin with, you have already made this claim ("everyone is from academia"), albeit in weaker form, and already been challenged to provide evidence. Making it again more emphatically but with absolutely no evidence does you no credit.
If the female attendance at these conference is going to increase, where are those women going to come from? What sorts of women will be interested in the skeptical movement?
What is your evidence that everyone at a skeptic's conference is a career philosopher or the equivalent? Unless a very substantial number of attendees are career philosophers your argument falls over immediately. (You did not respond to the point that just because Ian Plimer is a geologist and a prominent atheist that we cannot conclude that anything which is true of the geologist community must automatically be true of the organised atheist community).
Again, that point was brought up in my original post on the subject. Obviously the relationship is not directly equivalent.
Moving on to another equally serious error, it simply does not follow that because there is a high rate of Observation C in Community D, and a non-zero rate of Observation C in Community E, that therefore the rate of observation in both communities is identical.
Again, "identical" is just your hyperbolic reaction to the issue.
The issue is the relationship between the communities. Community E is almost entirely composed of a subset of people from Community D. The women in both communities (many of whom are members of both) describe exactly similar treatment.
Community E is new, it's small, it doesn't have much of a history. What we know of that small history is that the women complain of the same treatment that they receive in Community D. It's instructive to observe the situation in the feeder community to avoid making the same mistakes.
Compare: "There are lots of murders in Detroit. I have heard anecdotes of murders in XXXXXXXX. Therefore the murder rate in Detroit and XXXXXXXX is the same, until someone else shoulders the burden of proof to show otherwise". Sorry, but you don't get to jump from "X is non-zero" to "X is whatever value I say it is unless you prove otherwise".
Right, but XXXXXX is composed of transplanted citizens from Detroit. I suppose it's possible only the non-murderers moved, but given that there is a demonstrable murder rate in the new, smaller XXXXXX, it's safe to conclude that some of that behavior travelled with the population.
Straw man. I'm not advancing the claim they are totally separate, and in fact since I already gave you my estimate that the overlap in population was in the 1:10 to 1:1000 range accusing me of pretending that they are totally separate is evidence that you aren't tracking this conversation very well.
You're the one using the word "identical." From the beginning it's been an obvious issue of the overlap. Almost every speaker at these "skeptical" events has a high profile existence in Academia. The attendees are something of a black box. We don't have any good data. I would bet, however, that >90% have college degrees, and a significantly higher percent have advanced degrees than the population as a whole. "Skepticism" is essentially a multi-discipline effort of the academic system. There are some folks, like Randi, who enter via a different community, but there's a really close relationship.
We would need survey data to know for certain, but I would put my money on it being much closer to 1:10 than 1:1000.
What I am doing is challenging you to provide your own estimate and reasons to back it up. Your claim, your burden of proof, and handwaving or calling things "retarded" is not proof. What's your evidence of an overlap in academic philosophy conference attendance and organised skeptical conference attendance sufficiently strong to entail that we can generalise from one to the other?
Again, the only unique thing about "Philosophy" is the available data. The issue is one that is general to academia. The relationship between skepticism and higher education is based on the fact that almost every major figure in skepticism also works in academia (there are obviously some exceptions).
The women who exist in both worlds, like the feminist skeptics with higher profile blogs, describe exactly the same sexist treatment in both.
The inference is strong, but not perfect.
Evidence? Last time I checked we didn't know the man in the lift was present for anything in particular, so it sounds to me like you are making things up. I welcome correction if I am factually wrong on this point.
I suppose you can argue that Watson is lying about everything because all we really have to go by is her description of events. She says he was at the panel, it seems like a good bet since he was attending the conference and that was the keynote. I suppose it's possible that he wasn't, but Watson discussed the talk and the threats at the bar where he was present, but silent.
Are you a Continental philosopher perchance? They also have a habit of squirting out a cloud of forced epistemological doubt when threatened, and they also have the habit of not noticing that this forced doubt cuts both ways.
If you don't know the figures you are asking for then you have no basis for your own opinion. I put it to you that since you manifestly have an opinion, you should answer your own questions, since either you have some idea of what those figures are and you should share it or you have no idea and your opinion is baseless.
That's the entire point. We have a virtual blackout on relevant data. It's impossible to say with any degree of certainty what sort of threat Watson faced in that elevator. You can fiddle with the statistics based on the degree to which you narrow or broaden the baseline estimate. "Sexual assault on an elevator" includes countless times people were alone, countless times they were with groups, family, people they knew...it's just a messy, massive blob of data--and it's not even that. It's just a bunch of guesses based on google searches.
The question that immediately arises is whether you have any better data which you can show us. I'm guessing you don't, which is why all you can do is kick up dust, raise doubts, demand other people provide hard numbers and then try to position your unsupported opinion as correct until proven otherwise.
Where's the data you base your opinion on?
There isn't any solid data available. Just because I can't provide any better information, the default isn't that ****** analysis becomes good.
Look, in 15 minutes I doubled that guy's list of sexual assaults on elevators. It's an idiotic list. Google searches are not a definitive way to guage the likelihood of an assault.
It's flawed on so many levels that it's basically useless:
1) Doesn't define "acquaintance"
2) Collects data poorly from a poor source (google search)
3) Seems to think that a conviction is the only way to confirm that an assault occurred
4) Does not deal with under-reporting, which is a persistent problem in sexual assault statistics
5) Thinks that the ability to avoid a full, completed rape is somehow evidence that an assault is unlikely to occur
6) Eliminates relevant facts from the scenario
7) Assumes that sexual assault is the only thing that could justify a woman's apprehension on the elevator
It's not a good effort.
I look forward to you doing it and getting back to us, but I'm not holding my breath waiting for you.
Why would I be the one to do it?