The background probabilities cannot be ignored. If the most honest, reliable, and trustworthy man in the world tells you that he just met THE Abraham Lincoln on the street, there’s a good chance he’s mistaken.
If the Pope tests positive for AIDS, there’s a better chance that the test is wrong than if a homeless person tests positive for AIDS using the same test.
People just have a hard time accepting this if race and racial statistics or generalizations are involved.
Because the result is that in some situations, a claim made by a person of one race is less credible than a similar claim made by a person of another race.
If a 20 year old black guy claims that a NYC cab driver refused to pick him up, it’s a lot more believable than if a 40 year old white woman makes the same claim. Of course this doesn’t mean that 20 year old black guys are more credible than 40 year old white women.
Not because people who meet Abraham Lincoln are rarer than people who meet Bill Clinton, it’s because Lincoln’s dead. Sorry to be the one to bring you the bad news.
Of course, because that has nothing to do with the credibility of a person, but with the likelihood of an event occurring.
I have absolutely no problem accepting that one race lies more than another about any given thing, just as soon as someone actually shows some evidence that that is the case. Hasn’t happened yet. Don’t expect it to. Will still gladly accept it if it does.
Which situations would those be?
I have no problem believing that a 20 year old black guy has a harder time getting a cab in NYC than a 40 year old white woman. That has nothing to do with the credibility of either. If both black males and white females both lie about cab service in NYC at the same rate, they are equally credible.
If you think that’s not the case, then you can answer the two questions posed above in bold. No one else wants to, even though they want to claim that there is a statistical difference in the credibility.
Once again, it’s not a matter of how rare a given plausible occurrence is that takes away from credibility. It’s how likely a given claimant is going to lie about it that affects it. Prove that black women are more likely to falsely accuse white men of rape than white women do, and I’ll be on your side.
Without information about who they are, there is no scientific way I can do this. The veracity of their claim is a function of their veracity as a claimant. The word “veracity” derives from Latin for “truth”. This is no mere coincidence.
Again, I’ll have to make a random guess. Without knowing anything about these people, I have nothing to go on at all.
See, this is spectacularly illogical. For all I know, both twins are native to Andorra, so why would it be implausible for them to have been there? When you talk about probability in the sense that you’re using it, the assumption is that you’re dealing with random events that have yet to occur. In other words, you’re still stuck in prediction mode.
But even if we were talking about random events, your thought process is still screwy. If I roll a die that has five blue sides and 1 red one, there is a 17% chance that it’ll come up red. After rolling, if I closed my eyes and guessed which side had come up, I’d guess blue. But this is still very much a prediction because I have yet to witness the outcome of the event.
But say you kept your eyes open. Sweetly, I ask you to tell me the outcome of the roll. I trust you to tell me the truth because I’ve never known you to lie about anything. Plus, you have no incentive to lie, because we’re just playing a stupid game. There is no sex, food, or money on the line, in other words. And you are not color-blind, so I know you know what blue and red look like when you see them.
You claim that the die came up red. Now at this point I have three options:
to believe that you are telling me the truth
to believe that you are lying
to abstain from making a judgment until I see the evidence myself.
Which should I choose? Well, since I know you (hypothetically) and can attest to your good credibility, I choose option 1. If I didn’t know you and were unable to attest to your credibility, I’d choose option 3, but only if I wanted to be completely agnostic about it. If I wasn’t in the mood for agnosticism, I’d choose to believe you if only because there is nothing implausible at all about a red coming up and I have no reason to believe you’d lie.
The event has already transpired and I can not go back in time and apply a prior predicition to the present reality. There may have been a 83% chance that a blue came up, but the more important question now is this: What is the probability that magellan will tell me truth about the outcome of the roll? Because now the question is not what is more likely. The question is how credible is this claim.
A claim does not exist if there is no claimant. You can not separate one from another. To examine the credibility of the former you have to examine the credibility of the latter. Especially if you are dealing with an event that is well within the realm of possibility.
You do. I don’t. Because I understand what probability means. I also understand that after a nonrandom event happens, I can’t go back in time and say how probable it is, simply based on useless numbers about what other people have done.
Wrong. In exactly the same way as the rape scenario, we would need some indication of how frequently 20 year old black guys and 40 year old white women make such claims. Otherwise, we can make no inference. None. You’re sticking an intermediate, dependent variable (whether someone makes a claim) right into your probability tree, then saying that without any historical data on such claims, one can start making predictions about their veracity. This is simply not the case.
You are drawing a probability tree that looks like this:
The cases where no claim is made are omitted to save space, because they are irrelevant.
You are saying that without any data whatsoever regarding the [makesclaim] nodes, we can make some assessment of how likely, relatively speaking, true and false claims are. We can not. If, for the sake of argument, we assume that each race is equally likely to lie, then we would notice that the probability of a given claim being true is the number of true claims for a given race divided by the total claims made by a given race. This ratio is independent of the likelihood of a person being ignored, as the above tree shows.
There is. This is true of *any *claim. Those are the avenues by which you would be able to determine credibility.
Of course, it does. If two people you don’t know call you on the phone and one says “I’ve been to Paris” and the other says “I’ve been to the moon”, in a vacuum, one of those claims is more credible that the other. If I apply the condition that only one of them is telling the truth and you have no other information to go on and you have to choose one, you’d be foolish not to conclude that the claim of having been to Paris is more credible. You just have to play the odds. Remember, the conditions imposed are 1) one is lying 2) at this point, as far as you know the people themselves are equally credible 3) you have to choose which one you think might be lying.
I think you mean this.
In the individual cases, you have to assume the credibility of both people to be equal. The way you constructed the question (with such a large disparity in percentages) is helpful, but with the limited information given you can’t put an actual number on it. But it is safe to say that the odds of the white woman being the one that was raped are higher.
Let’s add a third person to your scenario: a white male who claims to have been raped. Wouldn’t you say that his claim is less credible—less likely to be true—than either of the two women? How about if he claims to have been raped by a woman?
The way I see it, while we have to approach things assumiing they are all equally credible as people, the credibility of their claims in this instance are rankable:
Whaite woman
Black woman
Male raped by male
Male raped by female
Why? Simply because this is how the particular acts rank in incidence in the real world.
She has less credibility only in regards to this particular claim. Not as a person in the general sense. If we change the question, she could easily be the one with the more credible claim. For instance, if the claim is “I have been the victim of racial bigotry.”
But all that would allow you to do is say “those who have proven themselves to be less credible are less credible”. That’s a given. The scenario in question is exploring a different situation, one in which the the credibility of each person is unknowable. And that does not render you unable to judge at all. You can judge based solely on the credibility of the claim itself.
But claims themselves can be judged for veracity, as in their likelihood of being true. If one person claims to have invented a way to make his bike go 10% faster and the other claims to have invented away to make his bike go 500% faster, do your really think those claims, devoid of other information, are equally credible? If so, we’ll just have to agree to disagree and just move on.
Sure you do. You have the fact that Andorra contains a tiny fraction of the number of people that Sitzerland does. And the fact that an even tinier fraction of non-natives that have visited Switzerland have visited Andorra. Your ignoring these facts is an strenuous attempt at remaining willfully ignorant (of the situation) and probably forfeiting a million dollars. This is ridiculous. (No snarkiness intended.)
It’s not implausible, it’s just less likely. The twins can be from Switzerland, as well. And the chances of that happening are (assuming equal incidence of twinery) MUCH greater. Why do you refuse to take into account real world statistics. (The Population of Andorra is just over 70,000 people. Swizterland has almost 7,500,000 people. I would guess that visitation to the two countries is even more skewed.)
Yes, and we’re forcing a prediction. If I tell you to predict the outcome of a nickle I’m about to flip and offer you that miiilon again, I doubt you’d say that it will land on it’s edge, although that is a possibility.
Okay, but you’re usiing a scenario in which all six outcomes have an equal probanility of happening. Go back to the nickle. Three outcomes, two equally very likely, one opossible but improbable. If I told you it landed on edge, are you telling me that you would have the same degree of confidence in my claim than id I told you it landed heads, or tails?
Again, for the 201th time, in this situation you are making a PREDICTION about an unobserved event. You didn’t witness the event take place and no one has presented a claim, so this is akin to closing your eyes, rolling a die, and making an educated guess.
So it makes sense to guess that a human and not a monkey has done this, because it is inherently implausible that an illiterate monkey could perform such a feat.
Now, if the keeper claimed that he saw a monkey typing the note, you’d be stuck with the same options that I presented in my previous post. You could choose to believe that he’s telling the truth, you could choose to believe he’s lying, or you could abstain from judgement until you get more evidence. Since the situation in this case is so implausible based on what we know about monkeys, I’d probably think the guy was pulling my leg.
But I didn’t arrive at this conclusion using probability estimates. There is no known probability that a monkey will type Hamlet’s soliloquy. AFAIK, that’s never happened before.
Once hypotheticals that are based on implausible outcomes that have nothing to do with the laws of probability start being trotted out, I get impatient and bored.
I’ve learned this from the trainwreck threads of recent, magellan: anyone who has to create an outlandish scenario in order to illustrate whatever point they are trying to make is clearly on the losing side of the debate. Ninja liquor store robbers, monkey typists, and nickels that land on their edges are not comparable to rape accusations or anything like it.
It was unfortunate but inevitable that this discussion would turn and go this route, when you have two parties arguing two entirely different scenarios, more often than not you won’t be acheiving any sort of closure to a debate.
One party has clearly laid down the stipulations this discussion was meant to take place under, and one party has clearly chosen to ignore those stipulations. I understand this was a difficult topic to discuss, and it was good that we did so relatively civil especially considering the forum we’re in.
If your willing to blatantly disregard the most vital parts of the topic at hand, i.e. the fact that the honesty of the two women is held equal, all you know are the races of the people involved, and that you have irrefutable numbers showing you which was the more likely individual to be raped than there really is nothing more to say. I suppose you drive in your car without a seatbelt because even though statistics show that your more like to survive with it on, you don’t know if the people who have told they lived through a car crash because of it are actually honest as individuals. I’m sure you wouldn’t believe graduating high school would actually increase your potential for financial growth in life, because even though statistics show that is the tendency throughout the population, those individuals reporting it could be lying. Sure you could live through a car crash without a seatbelt, and it is possible to become a billionaire after dropping out from highschool, but look at the numbers. Bollocks to playing the odds based on likelihood ever in life, eh?
Let’s use your seatbelt example, and let’s agree that wearing a seatbelt saves lives and have the statistics to verify it.
You have two people in the identical situation, save one was wearing a seltbelt and the other wasn’t.
Person A says: I was in a car crash and walked away and I wasn’t wearing a seatbelt.
Person B say: I was in a car crash and walked away and I was wearing a seatbelt.
Why would you not believed person a was in a car crash, compared to person B?
That’s all that YWTF is asking.
Now sure if Person A says, it was being driven by a super intelligent typewriting monkey, who was running away ninjas, who just robbed a liquor store and riding flying green dragons, who were being controlled by the Zombie Ghost of Abe Lincoln, you might have to ask for a little more detail…just a little mind you.
The last sentence is true. The preceding is less so. It can easily be shown mathematically that the number of possible character strings that the monkey could type of the same length as the fragment of Hamlet quoted is of the order of 10[sup]100[/sup].
It’s not merely “inherently implausible” that the monkey should happen to hit upon Hamlet’s opening sentence, it’s a mathematical impossibility that such a thing could happen at random within the lifetime of the universe. It’s not so much a case of closing your eyes, rolling a die and making a guess. It’s more like closing your eyes, rolling 100 ten-sided dice, and confidently declaring that they have not all come up showing 1.
So if you don’t reach this conclusion using probability estimates, you damn well ought to. I would not probably think the guy was pulling my leg; I would know for a fact that he was lying, crazy, or himself the subject of a deception.
Well, we have to start somewhere. The point about a reductio ad absurdum is you need to cite an absurdity to make it. But you’ve painted yourself into this logical corner where you refuse to admit that even the most patently implausible claim should not be treated with the same respect as the most patently plausible one.
When you receive an email promising to make you a millionaire if you will help to smuggle a sum of money out of an African country, do you carefully examine the track record of the person making the offer, in order to determine if he has a known reputation for truth or falsehood? Or do you consider the claim itself wildly unlikely, given that the received wisdom is that no-one ever does make money this way except for the people who lure in suckers with promises of unearned wealth?
The main reason why we are in disagreement is because you think population-level statistics provide meaningful information about individuals. You do not have “irrefutable numbers showing you which was the more likely individual to be raped”. That’s the whole point. All you have are numbers telling you about a group of nameless, faceless people. You know nothing specific about Janice, Marie, or ywtf.
For all you know, Janice is a 200 lbs weight-lifter who’d be able to kick any guys ass who attempted to attack her. Marie might weigh 100 lbs soaking wet and be easily overtaken by a rapist. ywtf might have a hobby of walking the streets half-naked in the middle of the night, drawing unwanted and possibly dangerous attention to herself. You just don’t know who is more at risk. Not if all you have to go by is race.
If it was possible to use stats like you’re using them, guess what? They’d probably be admissible in court. But they are not. Ask yourself why.
Lincoln’s dead, and therefore the odds of running into him are very low.
I found the following question:
You need a few other pieces of information. I will assume that there are 10,000 rapes per year, all of which are reported, and an additional 1000 false reports per year, distributed randomly among all women. I will assume a population that is half black and half white.
Thus, if there are 10,000 rapes per year, 100 of those rapes are white on black, and 9,900 are black on white.
In addition, 500 black women and 500 white women file false reports.
Thus, the total number of rape reports by black women is 600, 100 of which are true reports and 500 of which are false. So without knowing anything, the chances that a black woman’s rape accusation is false are about 85%.
On the other hand, there are 9,900 true reports by white women and 500 false reports. The chances that a white woman’s report is false is about 5%.
One thing bears emphasizing: In this example, black and white women are equally likely to make up a rape accusation. But once you factor background statistics in, the chances that a given accusation is false are different between the two groups.
But you’d have to assume that a monkey would behave in a way that would permit this experient to take place in accordance with this probability. That’s what is implausible.
Monkeys, just like humans, do not work randomly. In real life, a monkey would be much more likely to bang blindly on the keys with its fists, fling turds at the keyboard, try to fit the thing in its mouth, etc…than sit and patiently hit random keys with robotic-like precision. It’s an implausible story because we’d have to assume that a monkey would behave in a decidedly unmonkey type of fashion.
But even the absurdity needs to needs to rest upon plausible premises, or else your argument reads like a caricature of itself.
Ignoring the last part of your statement as it does no good (except I hope you feel a bit better after getting that out), and if we were to use the same parameters that should have been used for the initial statement, Person B’s story is more likely to be true. Why? For the same reasons it is more likely the white woman’s story was true.
If you know one and only one of the two stories are true, if you know nothing about their character or their tendency to tell lies, and you know what the statistics show there is no other logical person to chose.
Sure, Person B could have been in an accident without a seatbelt and walked away, however the statistics show that is far less likely to happen compared to someone wearing a seatbelt in a car crash and walking away.
You are making a huge leap in deduction to automatically assume that is somehow a statement that if an individual were to tell me that he was not wearing a seatbelt and walked away that I would not believe him.
you with the face do you see that portion of my quote that you bolded? Now read the first part of that sentence. You DO NOT KNOW if Janice is a 200 lb ass kicking machine, you DO NOT KNOW if one of the women had superpowers, that is the key. You somehow think that odds are less likely for a woman to be raped by a man if they are 200 lb body builders, but odds cannot be applied to a white man raping a black woman for whatever reason?
Once more, there is a complete and utter failure to carry this debate on in the context that it was initially set up. Once more, there is a failure to understand this is not a discussion on applying any conclusion to the judicial system. Once more, there is a failure to address anything about the actual subject at hand.
The concept of “odds” has no place in a discussion about…
NONRANDOM, UNPREDICTABLE, INDIVIDUAL HUMAN BEHAVIOR!!!
And this is where both you and Bricker keep getting tripped up. This is the fatal flaw in your thinking. I guarantee you that if a white man wanted to rape a black woman, the laws of probability will not stop him from doing do. If a white man wanted to NOT rape a white woman, he could exercise free will and not do so.
A flipped coin has no choice but to land on its head or tail according to statistical odds. A rolled dice is the same way. That’s why we can make safe predictions about events involving coins and dice.
You can not make the same kind of predictions about…
NONRANDOM, UNPREDICTABLE, INDIVIDUAL HUMAN BEHAVIOR!!!
I believe a 200 lbs weight-lifter will be less at-risk than a 100 lbs waif not because of “odds”, but because of what I know about rape. Rape is often a crime of violence. Strong people are able to ward off attack better than weak people. Ergo, a 200 lbs weight-lifter probably is not going to be as vunerable as a waif. I didn’t have to consult any statistics to reach this conclusion. I’m relying on common sense.