Kavanaugh: A Probability Poll

There’s some literature suggesting that people are more careful when they think in probabilities. If you ask them for a definite answer, they might give it and sound very confident, but if you encourage them to think probabilistically they might admit there’s more uncertainty.

I wanted to look into this in the context of the recent Supreme Court confirmation hearings, so I asked readers to estimate their probability that Judge Kavanaugh was guilty of sexually assaulting Dr. Ford. I got 2,350 responses (thank you, you are great). Here was the overall distribution of probabilities. Horizontal axis is percent chance he did it; vertical axis is number of people who responded with that percent:

This looks weird because people were most likely to give numbers rounded off the the nearest ten.

I separated responses into bins from 0 – 9%, 10 – 19%, and so on to 90 – 100%. Keep in mind that the last bin is slightly larger than the others, so it might make it unfairly look like more people gave extreme high answers than extreme low answers. I also switched the vertical axis to percent of responses in each bin. Smoothed out, it looks like this:

This looks pretty balanced, and it is: the average probability is 52.64%. This is probably a fake balance based on all the different demographic skews involved cancelling out: 2.5x as many Democrats as Republicans answered the survey, but 9x as many men as women did.

Here are the results broken down by party (blue is Democrats, red is Republicans):

And here are the results broken down by gender (blue is men, pink is women):

There should be an interaction between party and gender, because men are more likely to be Republicans, and women Democrats. I didn’t have enough data to investigate this too carefully, so I can’t say whether gender controlled for party remains significant or not.

I asked two questions to assess participants’ level of background knowledge: where did Kavanaugh go to law school? (correct answer: Yale) and what is Kavanaugh’s wife’s name? (correct answer: Ashley, but a shout-out to everyone who wrote “Mrs. Kavanaugh” and to the one person who wrote “beer”). Here are the probabilities of people who got both questions right (gold) vs. both questions wrong (green):

People with high background knowledge were more extreme in their answers, and slightly more likely to think Kavanaugh is innocent. I worry that I made a mistake in the questions I chose, since people who are more sympathetic to Kavanaugh might be more likely to know about his family. In retrospect, I should have asked at least one question about Dr. Ford.

What about neutral people? Do such people exist? I looked at people who were neither registered Democrats nor Republicans, and who rated their liberal-vs-conservative ideology, on a one to ten scale, as 4, 5, or 6. These people looked like this:

This chart makes it look like they’re slightly leaning towards guilt, but I think that might be a function of the binning; their mean probability of guilt was 49.85%, about as close to “totally uncertain” as you can get. Neutral people with more background knowledge were, again, more likely to lean innocent, with a mean of 41%.

I asked people whether they felt the evidence that Kavanaugh may have committed sexual assault was sufficient to reject his nomination to the Supreme Court, regardless of any other reasons to vote for or against him (like his legal opinions). 55% of respondents thought that yes, his nomination should be voted down; 45% still supported him. Here is a list of support for confirmation by probability of guilt:

Of people who thought there was only a 0 – 9% chance Kavanaugh was guilty, 98% thought he shouldn’t be rejected from the Supreme Court on this basis alone. Of people who thought there was a 90-100% chance of guilt, 96% thought that was sufficient to reject the nomination. Of people who thought there was 50-50 chance of guilt, about 50% still supported him and 50% opposed him.

This question suggests there is no real consensus about how plausible an accusation has to be before it means someone should be denied nomination to the Supreme Court. People generally agreed that if there was below a 25% chance the accusations were true, he should definitely be confirmed, and if there was above an 80% chance, he definitely shouldn’t be. But between 25% and 80%, people were pretty split on whether the Senate should err on the side of not confirming a potential assaulter, or wait until it was beyond a reasonable doubt. If we were trying to make these answers into a guideline for how a Senator should vote, it looks like they would be satisfying the most people if they voted to confirm if they thought the accusations had a less than 50% chance of being true, and to reject if they thought they had a more than 50% chance. I wonder how many people would endorse this rule as written.

I also asked people whether they would reject Kavanaugh in the hypothetical universe where he had immediately admitted to the accusations, then apologized for his actions and said he had changed as a person since then. About 55% of people said they would accept him in this scenario, meaning he gains about 10% support in the SSC demographic compared to the real-world situation. But the question was poorly worded and I’m not sure how many people answered yes they would reject him, accidentally meaning to say yes they would accept him.

Here is a graph of how people answered this hypothetical compared to how guilty they thought he was:

There’s no correlation. This makes sense: how guilty you think he is in this universe shouldn’t affect your opinions about a hypothetical universe where you know he’s guilty – but for some reason I’m still surprised. I guess I expected people’s partisan biases to sneak in, even if they didn’t make sense. Maybe the question was so confusing that answers to it are basically random.

Overall, when asked to use probabilities people were able to admit to a little bit of uncertainty in their answer. They could give probabilities that were well-formed and self-consistent. But none of this came close to removing the partisan bias and the strong difference in opinions. There is no consensus in the general SSC demographic, and even unbiased people as a group are unable to send a coherent signal. This is not a good way to get beyond confusion and disagreement on an issue like this one.

You can download the raw data (slightly cleaned up) here.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

533 Responses to Kavanaugh: A Probability Poll

  1. Michael Watts says:

    Poorly directed counterpoint: I went to take the survey, but left without filling it out after seeing that the questions were “how likely do you think it is that Kavanaugh did what he’s accused of” and “if he did, what would be the appropriate response”.

    I don’t see that those questions are… relevant… to the Kavanaugh issue as it played out in reality. I also don’t see where people are getting answers from. You have an accusation with the following properties:

    1. There is no evidence supporting it.

    2. There is no evidence contradicting it.

    3. There cannot be any evidence in the future either supporting or contradicting it, because it is set more than 30 years in the past.

    This doesn’t seem like enough to go on if you’re trying to produce a probability estimate. You can try to go for a base rate (already hard; data on this is inherently low-quality)… and once you have the base rate, there’s nothing else you can do. I would guess that this is why there’s such a strong signal of party affiliation in estimates — people provide their party’s preferred answer because there are no other data points pulling in any direction.

    (The title of this post contains a typo — Kavanuagh instead of Kavanaugh.)

    • CorruptedHardware says:

      Besides legal evidence, there are other (weak) forms of evidence that could move your estimate away from the base rate, for example character assessments of the people involved, and the specifics of their testimony.

      • Ketil says:

        Besides legal evidence, there are other (weak) forms of evidence that could move your estimate away from the base rate

        In theory, but I think it is pretty clear that nobody does this. Everybody starts with a prior of 99% depending on where they stand politically, and the evidence is too scant to move that in any meaningful way – even if they somehow manage to avoid confirmation bias and other cognitive biases.

        • Alsadius says:

          That does not match the observed data. There would not be so many people in the middle if we were all just partisan yes-men. Never mind all of the partisans going against interest off on the tails – yes, the tails are thin, but they’re still a meaningful number of people(100-200 from eyeballing it?) in total.

        • Joseftstadter says:

          The majority of people who knew Kavanaugh at Yale, including Republicans, tend to think he’s guilty. The vast majority of people, including Republicans, who had interactions with DKE fraternity brothers at Yale between 1984 and 1988, certainly believe he was quite capable of the offense Ford describes, and is certainly guilty of perjury. I assume that’s why the FBI refused to talk to most of those people.

          • The Nybbler says:

            Even if that were true (which seems very unlikely, or rather it seems very unlikely any such survey of all of Kavanaugh’s Yale associates or all people who had interactions with DKE members in those years has actually been performed), what difference would the opinion of people who were in no position to have knowledge of Kavanaugh’s guilt or innocence make?

          • “The majority of …”

            How do you know these things?

          • Mark Atwood says:

            The majority of people who knew Kavanaugh at Yale, including Republicans, tend to think he’s guilty.

            {{Citation needed}}

      • baconbits9 says:

        These are only forms of evidence if you know how to interpret them.

    • scmccarthy says:

      I think you’re conflating evidence in the sense of “information I can use to estimate a probability” with evidence in the sense of “information relevant and to a criminal trial and sufficient to determine (or not) guilt beyond a reasonable doubt”.

      1. There is no evidence supporting it.

      2. There is no evidence contradicting it.

      3. There cannot be any evidence in the future either supporting or contradicting it, because it is set more than 30 years in the past.

      That’s silly. Yes, it would be hard to take this to trial. No, it’s not impossible to gain information about the past.

      • Michael Watts says:

        Sure, it’s possible to gain information about the past. But it’s not possible to gain any arbitrary information about the past. We know that ancient Egypt’s economy involved making gold art objects and trading them for foreign copper and wood. We know what those objects looked like — they survive today.

        We know, based on documentary evidence, that ancient Mesopotamia’s export economy was based more around manufacturing cloth. We don’t know what that cloth was like, and we never will; unlike gold, cloth rots.

        The claim that something happened thirty years ago in a private room without leaving any physical evidence of any kind is one of those “no evidence could theoretically emerge in the future” kind of claims.

        • Murphy says:

          Well… evidence could emerge in future.

          If someone filmed brett kavanaugh in a drunken rant shouting “She shouldn’t-a talked! I shoulda done more than just tried to stick my dick in her, I shoulda killed her as well!”

          Or a text message between Christine Blasey Ford and a friend turned up with something like “I know it’s not true but we can’t let GOP scum onto the supreme court, I’m gonna take the bullet on this one and claim he assaulted me”

          Or if it turned out that whoever owned the house it happened in back then was a super-creepy pervert with hidden cameras in every room that caught the whole thing and his heirs discovered them.

          It’s entirely *possible.*

          It’s just *unlikely*.

        • sympathizer says:

          The claim that something happened thirty years ago in a private room without leaving any physical evidence of any kind is one of those “no evidence could theoretically emerge in the future” kind of claims.

          That’s not true, and that’s not even what Dr. Ford’s claim is.

          There are so many crime mysteries that start out with a seemingly unsolvable mystery that it’s a genre unto itself: Locked-room mystery.

          But Dr. Ford’s accusation is not about a locked-room mystery. There is at least one claimed eye-witness (Mark Judge) and any of the people Kavanaugh and Ford interacted with in the days before and after the party this was supposed to have occurred on will have potentially relevant recollections.

          • Murphy says:

            “Locked-room mystery”

            There’s a reason why 99.something% of crime novels end with the perpetrator standing up and boldly confessing all: the evil voice doesn’t stand up in court and most of the time when they “solve” the mystery they don’t so much solve it as lay out a vaguely plausible way that the evidence could have ended up as found.

            https://www.youtube.com/watch?v=5cFzABv0xMU

            Take almost any crime drama with a dramatic confession once the protagonist has figured it out and instead replace the confession with the accused going “nope. I didn’t do it.” and they have basically nothing and the novel or TV show ends on the unsatisfying conclusion that they might just be dragging an innocent person through the courts.

        • hopaulius says:

          Let’s suppose we could put one of Justice K’s or Prof. B-F’s high school friends in a time-travel device that would also restore them to their appearance at the time. It would be nearly impossible for them to find the party. There is no date. There is no address. There is no reference to the name of an owner or resident of the home. All we have is something alleged to be stored in the memory of one person, or up to three people. The one making the charge, when she was asked about her most vivid memory, said approximately, “The hippocampus stores the laughter.” This is a different statement from “I remember laughter.” It is a general statement about something that might be stored in the hippocampus of someone who had undergone a crisis. It is the sort of sentence one might construct if one were trying to defeat a polygraph, or avoid the appearance of lying on TV. If our time traveler could go back to 198? and notice the future professor exiting a house, she might ask, “Hey, did you see Brett in there?” I’m not sure what the follow-up would be: “Did he try to rape you?”

    • Ketil says:

      Absolutely. It is impossible to determine the truth 35 years after the fact, so people just stick to their political biases, and rationalize it after the fact. Conservatives are certain he is innocent, liberals are certain he is guilty, and people with no political opinion make it a toss-up. And hey, she lied about her fear of flying, didn’t she? And he tugged his sleeve every time he talked about alcohol, that’s a tell, right? And so on. But we really have no way of telling, and the only thing the statistics show is that we, the SSC commenters, suck at our ambition of rationality and epistemic honesty.

      But what is bothering me, is that I started in the 50/50 camp. I guess I found the story of Ford being dragged from a party into a bedroom for the purpose of rape a bit incredible – most people would react to something like that. But a sexual situation that got out of hand or unpleasant, maybe. Likewise, it wasn’t inconceivable that she has developed false memories, through therapy or otherwise, or that she was making a deliberate, false accusation to bring down a hated political enemy.

      But the more I read about it, the more I find that I’m becoming partisan. I’m starting to lean towards one side, and the more I read, the more confirmation bias and a tendency to ignore the counterarguments does its work. This is disconcerting. Anybody else have the same experience?

      • Status_Signal says:

        My biggest concern is that my information diet is becoming low quality and partisan. I’m not getting exposed to decent right-wing news sources. I’m seeing very anti-Kavanaugh stuff on Reddit that I occasionally look at and find is very low quality, but presumably it’s still shifting my viewpoint.

        • Murphy says:

          if all I did was graze reddit I’d just think there was a half dozen women confirming the claim and that it was thus pretty damned solid.

          It was over on quora that I saw some posts where someone broke down the list of what was testified to by some of the other people which boiled down to

          “I saw him hanging out at a party”… where that party apparently included someone getting raped and the person confirming being a 22 year old hanging out with a bunch of 16 year olds and apparently not taking any issue with the rape-party.

          + a load of people basically saying “Oh ya, I knew him, I could *totally* imagine him acting like that”

          • Conrad Honcho says:

            The Swetnick claims are particularly difficult to dissect, because they appear to be written in an intentionally confusing way. When I first read them, I came to believe they meant Swetnick was accusing BK/MJ of 1) drugging/spiking punch for the purpose of raping girls and 2) standing in line for the gang rapes.

            When she had her interview with NBC, however, she said she never actually saw them putting anything in the punch. This seems like a contradiction, but no, when you look back at her original statement, she did not specifically say she saw them spiking the punch. She says she “became aware of efforts” by them to spike the punch. Now I don’t know how one makes efforts to spike punch beyond actually spiking the punch (maybe procuring the substances?) but I think the vast majority of people reading that sentence would comprehend it to mean she claims she saw them spiking the punch. But technically, no, “became aware of efforts to” is not the same thing as “saw them do it.”

            Also, she said in the interview that she was not aware of the gang rapes at the time. It only became clear to her that that’s what was going on after she was raped herself.

            But yes, she’s 3 years older than Kavanaugh, so I would have liked an explanation for “why were you, 3 years out of high school, going back to high school parties? How did you find these parties? Who invited you?”

        • JulieK says:

          This Jonah Goldberg column is a good summary of the right-wing view.

          • Conrad Honcho says:

            That was an excellent summary. And I am a right-winger who does not usually agree with Jonah Goldberg. One effect of the Kavanaugh hearings has been to unify the Republican party. I don’t think there’s any wing of the Republican party that would disagree with that column.

          • DavidS says:

            I’m in the UK and only getting bits of this (so can’t really judge how much of what he says there is fair) but this was helpful in understanding a thoughtful conservative view (I’ve mostly seen very tribal people on both sides)

      • Slocum says:

        I’m an libertarian-leaning independent. I would say that I started out at 50-50 and moved toward greater doubt of Ford based on details that emerged later: her bogus fear of flying, the dubious claim that she was unaware of the desire to interview her in California, her bogus story about the two front doors, the lack of corroboration by her friend supposedly at the party (along with Ford’s suggestion that the friend didn’t remember because she had mental problems). They ex-boyfriend’s letter detailing how Ford had helped prep friend Monica McLean for a polygraph when she was applying for an FBI job. The refusal to release the original therapist notes to the committee (which would have been the only even minor corroborating evidence). The almost complete scrubbing of her internet history (which prevented any assessment of the strength of her partisan biases).

        As a small-l libertarian, I wish the nominee was somebody other than Kavanaugh due to his record on civil liberty cases, and wouldn’t have minded if solid evidence had emerged to support Ford’s claims which forced Kavanaugh to withdraw. But, to my eyes, the opposite happened — this looked more and more like a calculated, coordinated last-minute attempted hit job, with Ford playing at least some role in the coordination (falsely claiming a fear of flying to justify further delay).

        • Careless says:

          You stole my post!

        • qwints says:

          For context, I’m left wing and wish Kavanaugh was not on the Supreme Court. I agree that Keyser (Ford’s friend) gave a statement that makes it less likely that Ford’s story is true, especially the part about not knowing Kavanaugh. I’d also put the ex-boyfriend’s letter, not providing the therapists notes and internet scrubbing on that side, though at little weight.

          I think the rest are really motivated reasoning. A fear of flying isn’t rare nor is choosing to fly despite having that fear. What the California interview would entail wasn’t clear. Ford’s house does have two doors. And there’s no motive to lie about those things. These sorts of things are the equivalent of the people harping on Kavanaugh lying about the meaning of “devil’s triangle.”

          • J Mann says:

            Completely agree – Kavanaugh and Ford both have the kinds of minor inconsistencies that I would expect from live testimony, and probably should be weighted similarly.

            (For the record, the thing about the two doors is whether she got a second door because she needed a second exit from her master bedroom as a result of trauma or whether she got it because she wanted to sublet a portion of her house to a therapist who listed it as her place of business after the remodel, to Google employees, etc. Allegedly, Palo Alto real estate is so valuable that the second thing is a possibility, but I don’t think the final answer is really known.)

          • Edward Scizorhands says:

            My trust in both Ford and Kavanaugh went down, but like J Mann said, this was probably inevitable, since no one has a memory like a VCR.

            We just got to drag two people with no real change.

          • As I read the situation, the way the accusation played out was clearly political strategy. I wouldn’t be astonished if the accusation was true, wouldn’t be astonished if it was false.

          • Slocum says:

            No, fear of flying isn’t uncommon. But she flew frequently for many other reasons, including vacation. And including for the hearing when she finally agreed to appear. Regardless of what the California interview might have entailed, the point is that she claimed she didn’t know that the committee wanted to interview her at all – not through her legal team and not through the news media. Anybody paying attention knew about the committee wanted to arrange an interview as quickly as possible, and yet somehow Ford didn’t? I guess reasonable people can assign different probabilities to that, but it looks to me (as with the fear of flying) as very likely another calculated falsehood intended to delay the process. And yes, her house has two front doors, but apparently the second one was put years before the 2012 therapy sessions and with the purpose of being able to rent out part of it. The reason the ex-boyfriend’s letter has weight is that it is detailed and specific. He claimed to have witnessed Ford help Monica McLean prepare for a polygraph for a prospective FBI job. Nobody, AFAIK, has denied that Ford lived with McLean or that McLean did go to work for the FBI at that time. But Ford denied, under oath, that she’d ever helped anybody prepare for a polygraph. Why did she lie about that? Because she had presented herself as a waif who previously knew nothing about how polygraphs worked. And why put weight on the therapists notes? There were suggestions of inconsistencies between the account given to the therapist in 2012 and the current accusation (number of people involved, Ford’s age at the time of the attack). Keeping them hidden appeared to me intended to hide any inconsistencies. I believe that if the notes they strengthened her case, they would have been released, and since they weren’t, I have to downgrade the probability there, too.

            I really can’t think of anything that came out after Ford’s accusation were leaked that made me think the probability of truth should be upgraded.

          • qwints says:

            @slocum, like I said, I agree that the ex-boyfriend’s letter makes Ford less credible. I personally don’t think that letter is very credible, but I can see why someone else might. Same direction for the therapist’s notes, but I think it’s also likely someone would not want those released regardless of whether they helped them.

            she claimed she didn’t know that the committee wanted to interview her at all

            What are you referring to here? She did not make that claim in her testimony.

            Here’s the exchange

            MITCHELL: OK.

            Was it communicated to you by your counsel or someone else, that the committee had asked to interview you and that — that they offered to come out to California to do so?


            [objection to attorney-client material]

            FORD: I just appreciate that you did offer that. I wasn’t clear on what the offer was. If you were going to come out to see me, I would have happily hosted you and had you — had been happy to speak with you out there. I just did not — it wasn’t clear to me that that was the case.

            The offer was to fly female staff investigators to meet Dr. Ford and [her counsel] in California, or anywhere else, to obtain Dr. Ford’s testimony.” I can’t find anything published that suggest Grassley offered to have the committee itself travel to California, although Grassley, Tillis and Mitchell (the prosecutor) all refer to the committee being willing to go to California.

            But Ford denied, under oath, that she’d ever helped anybody prepare for a polygraph

          • J Mann says:

            @qwints

            That’s fair, although I wonder how Ford could not have known what the offer was, given that Grassley was making it both to her lawyers and publicly.

            If I understand correctly,

            – Ford might be somewhat uncomfortable flying, but she flies as many as several times a year, including some trans-Pacific flights.

            – When Ford’s lawyers indicated that one of the reasons they couldn’t meet a particular deadline was that Ford can’t fly and had to make arrangement, they were not being honest. (Assuming that exchange with Grassley’s staff has been reported accurately).

            – Ford’s position is that she let her lawyers represent that she had difficulties flying in the hope that Senate staff would conduct a public hearing in California. (Possibly in her house or school, since she says she “would have hosted” Senator Grassley).

            – Ford was not willing to conduct a pre-hearing interview with Senate investigators, or at least was willing to let her lawyers prevent one.

            – Ford’s lawyers proposed a number of conditions for her testimony, but IIRC, they never proposed that the testimony occur in California.

            Again, it’s not really informative, but it doesn’t strike me as adding up.

          • Slocum says:

            “She did not make that claim in her testimony…”

            Fair enough — I remembered that wrong. She claimed she didn’t know what the details of the offer were rather than not knowing about it at all. But still, given that she supposedly was afraid to fly but eager to be heard, it would seem the natural response would be to ask for clarification and try to make things work. Which she and her legal team obviously did not do.

          • qwints says:

            Okay, so it’s more the idea that she intentionally delayed the testimony. I get that. I agree that the probability of the negotiation and testimony playing out the way it did is higher if she’s intentionally lying than if she’s telling the truth, though I’m not sure on the magnitude. I just don’t get this claim that people who fly (even frequently) can’t have a fear of flying. I know several people who fly regularly who hate flying. A search shows numerous articles that have nothing to do with the Ford case. And when I see people make the argument, I find it hard to consider their rest of their argument seriously.

          • J Mann says:

            Yeah, I honestly thought that Ford said she didn’t know that Grassley was trying to set up an interview with her, so you’ve updated me. It’s like the idea that Kavanaugh said that “Ralph Club President” was because of “spicy food” – a combination of live testimony + telephone.

            (Although frankly, the idea that she didn’t know the terms of the interview offer is silly since Grassley’s staff were both emailing them to her lawyers and posting them on twitter. Not a big deal, but I assume she was wiggling a little to explain the delay strategy.)

            The thing on flying is back when Grassley was trying to put this hearing together, his staff claim that Ford’s lawyers said one of the reasons they couldn’t meet the original deadlines was that Ford was afraid to fly and needed to make arrangements to drive across country. I don’t think that’s in any of the letters, so it may be a miscommunication.

          • Yosarian2 says:

            I interpreted that stuff as less “intentional attempt to delay” and more as “deep hesitation about going public/ testifying in public” about the allegations, which remember she really didn’t want to go public about in the first place. Totally not surprising to me that she was hesitant about testifying in front of Congress. It’s also totally not surprising that any existing phobias she can normally deal with would become much more extreme in those circumstances.

            Overall I don’t think the “delay tactics” theory fits her behavior any better that the IMHO more likely “someone who honestly just never wanted to go through this ugly process at all” theory.

          • Protagoras says:

            How does Keyser’s statement show anything? I know Kavanaugh said repeatedly that Keyser’s statement refuted Ford, but nothing memorable is supposed to have happened to Keyser, and all she’s said is that she doesn’t remember (as is hardly surprising for 36 year ago events of no significance to her). She even says she believes Ford, which admittedly doesn’t really count for much in Ford’s favor when she doesn’t remember, but surely at least doesn’t count against Ford.

          • The Nybbler says:

            As with the dog that didn’t bark in the night, sometimes absence of evidence is evidence of absence. Keyser, Ford’s friend, was supposedly at this party with Kavanaugh, Judge, Smyth, and Ford. Ford at some point goes running out of this party without so much as a fare-thee-well…. and Keyser doesn’t remember this at all, or having ever been at a party with Kavanaugh? Sure, it’s not conclusive, but it’s evidence.

          • “A fear of flying isn’t rare nor is choosing to fly despite having that fear. ”

            “Being sick isn’t rare, nor is choosing to play golf despite being sick. I just didn’t want to, I mean, I was too sick to go in to work this mourning. Why would anyone think me dishonest?”

    • slightlylesshairyape says:

      The universe doesn’t particularly owe you enough data to make a probability estimate over any particular set of facts.

      The political process doesn’t allow the Senate to duck the binary decision of whether or not to confirm. By extension, it becomes something of a binary political decision.

      The sum total of this is that while you can have whatever epistemic uncertainty, you can’t really duck the question. As a (I assume) non-Senator, you can perhaps claim that you don’t know how you would vote.

      • Michael Watts says:

        No, the universe doesn’t owe me enough data to make a probability estimate over a set of facts. What would that mean? If I could just take data from the universe because I wanted it, I would be certain about everything; probability would be outmoded.

        But Scott can take whether it’s possible to make a probability estimate into account when asking survey takers to make probability estimates.

        This reminds me pretty strongly of a point which I believe I also encountered here. I’ll put the idea in blockquotes, but hopefully it will be obvious that I’m making this up based on a vague memory:

        When polled “What’s better: A, B, C, or no way to decide?”, 30% of respondents chose “no way to decide”. But when polled “What’s better: A, B, or C?”, it was not the case that 30% of (other) respondents refused to answer. The D group from the first poll was lying!

        On second thought, it’s true that the respondents disagree, but it’s not obvious that the first group are lying. Why not consider that 30% of the second group are lying, and provided a meaningless answer because “no way to decide” wasn’t there?

        I submit that nearly 100% of responses to this survey follow that model — they provided a number because Scott asked for one, but the number was not justified, and does not reflect the judgements of the respondents.

        • slightlylesshairyape says:

          OK, but IRL there are cases where you have to decide between A,B,C even in the absence of sufficient data.

          On the other hand, there are indeed cases where you can say ‘no way to decide’.

          The K drama is a case of the former type. Regardless of the sufficiency of data, either he is confirmed or not. It has to be decided. And if we think the truth of the allegation is relevant to that decision, then it has to be estimated.

          • Michael Watts says:

            Why would we think the truth of the allegation is relevant to the decision to confirm? By hypothesis, we cannot investigate the truth of the allegation. How can it be relevant to anything?

            What is the difference between the Kavanaugh allegation and the claim “God hates fags”? If God really did hate fags, what would the material consequences be?

            Is there a purpose to estimating the probability that God hates fags? Forget purpose — is there a meaning?

          • slightlylesshairyape says:

            Why would we think the truth of the allegation is relevant to the decision to confirm?

            Because we’ve decided a priori that we don’t want to confirm a rapist to the court. And we have to make a binary decision about this nominee. Voting ‘yes’ is a positive affirmation that he’s not a rapist.

            I really don’t understand why you think that because we don’t have much data somehow it’s not relevant.

            What is the difference between the Kavanaugh allegation and the claim “God hates fags”?

            Well, if my son or daughter was gay, the claim that “god (exists within some particular theologically defined bounds and) hates fags” would be pretty relevant to my binary decision whether to attend their wedding or acknowledge their sexuality. I have to decide whether homosexuality is an affront to the all-benevolent ruler of the universe, I don’t have a choice but to resolve it.

            [ Well, in this case maybe there are a few choices than 2. Unlike the WBC, I might be permitted by my theology to accept that GHF but not to condemn them. ]

        • Deiseach says:

          That second question is slightly different, though. You’re presenting them with three options and asking them to choose one. Someone might think “Ugh, they’re all terrible, but if I have to choose, then C is the least worst/lesser of three evils”. Whereas if you give someone the option not to choose, they can legitimately go “Great! I think they’re all terrible, I pick none!”

    • Scott Alexander says:

      I think if you believe that, you go with base rates for a randomly selected accused person of being guilty. If like many other people I’ve talked to you think there is some evidence, then you would say that.

      • Jon Gunnarsson says:

        But how do you get a base rate? We have no idea how many accusations of sexual assault are true because there are so many cases where (like this one) there is insufficient evidence to decide.

      • Conrad Honcho says:

        Except Kavanaugh is not randomly selected. He’s an extremely high profile person in the most polarized political atmosphere any of us remember.

        • vV_Vv says:

          Except Kavanaugh is not randomly selected. He’s an extremely high profile person in the most polarized political atmosphere any of us remember.

          Which is strong evidence that the accusation is politically motivated.

      • vV_Vv says:

        I’ve answered 1% to the probability of guilt question, my reasoning is the following: the baseline rate for a random man committing sexual assault is around 1-2%. On a purely Bayesian basis, accusation is evidence of guilt, but the apparent political motivation and timing of the accusation cancel this down to about baseline level. The many inconsistent details and memory gaps in Ford’s account are mild evidence of the accusation being false. On character level, Kavanaugh’s history of binge drinking is weak evidence of sexual misconduct, but his long career as a successful man free from sexual misconduct allegations is weak evidence of innocence. Overall, I’m not much moved from the baseline rate, hence my 1% probability estimate.

        The Supreme Court confirmation process should hold standards similar to a court of law, IMHO, applying presumption of innocence for the game-theoretic reason not to incentivize false accusations, hence for this purpose the accusation is not evidence of guilt. Similarly, it should not consider the apparent political motivation of the accusation as evidence of its falsehood, in order not to incentivize political figures to do bad things because they could think they could get away with them by claiming political persecution. Therefore we are again left with the baseline rate and weak evidence both towards guilt and towards innocence. Hence I answered that the Court should confirm.

        • vV_Vv says:

          Hence I answered that the Court should confirm.

          Sorry, I meant the Senate Committee on the Judiciary should confirm Kavanaugh.

        • oncomer says:

          Is the baseline rate for a random man committing sexual assault shown somewhere to be around 1-2%, or is that a personal estimate? I mean the question honestly, and am willing to believe if it is the former.

          Also, and separately, how do you compare different pieces of weak evidence in different directions? Kavanaugh’s history of binge drinking and Ford’s imperfect testimony are, as you say, both weak evidence increasing and decreasing probability. I would have argued that being a member of a fraternity in the eighties who binge drinks is a stronger predictor for committing sexual assault (and is in fact the stereotypical demographic for committing sexual assault) than the kinds of inconsistencies and memory gaps Dr. Ford displayed are a predictor for lying about sexual assault.

          I appreciate that you showed your reasoning, even though it invited criticism. It helped me to better understand how someone with your position might think about the issue.

          • Doctor Mist says:

            I was surprised by that as well, and did a little googling to see if I could find a source.

            I didn’t, so I did a little Fermi estimate. According to wikipedia, the U.S. seems to have about 30 rapes per 100K people, and there are about 3.2M people, so that’s about 960 rapes per year. There are about 1.6 million men in the U.S. and each each has maybe 50 years in which he might commit a rape. If 2% of them commit one rape in their lifetime, that’s 640 rapes per year.

            So the order of magnitude is not ridiculous, as long as you assume most rapists are not serial offenders (or get caught pretty soon). Whether those assumptions are plausible I could not say.

          • Glen Raphael says:

            @Doctor Mist

            There are about 1.6 million men in the U.S.

            Wait, what?

          • Doctor Mist says:

            Wait, what?

            Oh, crap.

            Let’s try that again. There are 320M people, so:
            about 96K rapes per year (since there are 30 per 100K)
            about 160M men (since we are about 50/50)
            If 2% of men commit rape once in their lifetimes, and those men commit it this year rather than in the other 49 years I am assuming them to be capable of it, then we would expect to see 64K rapes per year, and as before that’s the right order of magnitude.

            I must have just typed clumsily before; I’d think I would have noticed if I really thought I was saying there were only a million and a half men in the U.S.

            I’d still love it if somebody knows a citation for the original claim. My Fermi analysis is kind of sensitive to how many times a rapist rapes before getting caught or reforming. It also neglects the fact that a 20-year-old is a lot more likely than a 70-year-old (right?) but I don’t think that matters — instead of assuming that any given rapist might decide to commit his rape at age N on a uniform distribution, I could have assumed that all rapists were 20 years old, and just divided the number of men by 50, which would have given me the same answer.

          • Viliam says:

            My Fermi analysis is kind of sensitive to how many times a rapist rapes before getting caught or reforming.

            I would assume an uneven distribution, because I suppose that:

            – psychopaths are more likely to commit crime than ordinary men, and are more likely to commit it repeatedly even after being caught;

            – subcultures with less rule-of-law (if you live in a ghetto where the police does not bother to enforce the law, or if you have a rule that “one does not talk to cops, ever”) have more interpersonal crime;

            – difference in opportunity, e.g. some men have more access to vulnerable potential victims;

            – difference in personality, e.g. asexual or gay men are less likely to rape women.

          • vV_Vv says:

            I remember that I’ve read that estimate on a source that also said that most rapes are committed by serial offenders, who tend to have a general pattern of anti-social behavior.

            I can’t find that source anymore, but Wikipedia cites a study that estimate male perpetrators at 2.4% of the male population and RAINN says that “Perpetrators of rape are often serial criminals”.

        • Luke G says:

          On a purely Bayesian basis, accusation is evidence of guilt, but the apparent political motivation and timing of the accusation cancel this down to about baseline level.

          You’re implying that about 50% of superficially-credible accusations of sexual misbehavior against political figures are false? That sounds far too high to me.

          The many inconsistent details and memory gaps in Ford’s account are mild evidence of the accusation being false.

          Memories fade after 35 years, except for the extremely emotional highlights. Liars tend to fill in the details to appear more credible! Having some fuzzy memories shouldn’t count against Ford, and may actually count in her favor if you agree the fuzziness is consistent with expected memory decay (perhaps mitigated by the fact that Ford is a psychologist and might know how to fake memory decay accurately).

          On character level, Kavanaugh’s history of binge drinking is weak evidence of sexual misconduct, but his long career as a successful man free from sexual misconduct allegations is weak evidence of innocence.

          Besides being a drinker, Kavanaugh also was a multi-sport athlete. This means he was more than likely surrounded by a culture where women are treated poorly, and significantly increases the baseline rate of sexual assault. His yearbook entry suggests Kavanaugh either didn’t think much of women during high school, or had peer pressure not to think much of women.

          Plenty of men behave badly in HS and college and go on to successful, scandal-free adult lives. I don’t believe his adult success cancels the drinking and HS jock-culture very much.

          Overall, I’m not much moved from the baseline rate, hence my 1% probability estimate.

          I feel like you’ve missed some of the most important pieces of evidence however:

          * Ford repeatedly asked for an FBI investigation, Kavanaugh repeatedly dodged the question of FBI involvement when posed to him. Raises the likelihood Kavanaugh has relevant skeletons he doesn’t want unearthed.

          * Ford raised the accusations before Kavanaugh was nominated, when he was only on the shortlist. This dramatically reduces the probability Ford was acting on political motivations: you wouldn’t be very motivated until after the nomination.

          * Ford apparently communicated parts of her story earlier, to her husband, therapist, and some friend. There might be some doubt over each of these claims, but overall should count in her favor.

          * Ford knew who Kavanaugh’s drinking buddies were: PJ, Mark Judge. So she almost surely actually was at a party with Kavanaugh. Moreover, for that memory to stick with her for 35 years, something memorable must’ve happened there.

          * Kavanaugh’s own calendar shows a beer party with PJ and Mark Judge with timing that aligns with Ford’s testimony: summer of ’82 a month before Judge worked at a grocery store. (Ironically Kavanaugh used this calendar to support his innocence.)

          The Supreme Court confirmation process should hold standards similar to a court of law, IMHO, applying presumption of innocence for the game-theoretic reason not to incentivize false accusations, hence for this purpose the accusation is not evidence of guilt.

          There’s already a huge incentive not to make false accusations, such as perjury, and destroying your reputation, life, and career if it’s found out you’re lying. In fact there’s already a huge incentive not to make true accusations, which you’re surely aware of with the #MeToo movement. The game theory is already on the side of the accused.

          In the case of the Supreme Court (and similarly for CEOs or other high-profile positions), you generally have a plethora of qualified candidates. To preserve the legitimacy of the institution, you don’t want reputational problems with the individuals casting a cloud over what they do. It’s completely logical to fall back to other candidates if one has even 10% chance of being guilty. It sucks if you’re the victim of a false accusation, but there’s a lot of things that suck about being a public figure that you must accept if you want to be a public figure. The US judicial code admits as much, saying “A judge must expect to be the subject of constant public scrutiny and accept freely and willingly restrictions that might be viewed as burdensome by the ordinary citizen.”

          • PeterDonis says:

            Ford raised the accusations before Kavanaugh was nominated, when he was only on the shortlist.

            This is not my understanding of the timing. Ford sent her letter to Sen. Feinstein on July 30th; Kavanaugh was announced as the nominee on July 9th.

          • idontknow131647093 says:

            Most of what you said is not provable or utterly against the evidence, but this sticks out as the worst idea propogated by your post. Indeed, it is insanely bad.

            Memories fade after 35 years, except for the extremely emotional highlights. Liars tend to fill in the details to appear more credible! Having some fuzzy memories shouldn’t count against Ford, and may actually count in her favor if you agree the fuzziness is consistent with expected memory decay (perhaps mitigated by the fact that Ford is a psychologist and might know how to fake memory decay accurately).

            No no. Memories fade INCLUDING AND ESPECIALLY HIGHLIGHTS. Or rather, highlights are always filled in. Ask anyone where they were on 9/11, they will always have an answer. That answer will almost certainly be wrong unless it is generic ala “school” or specific “121 state”. Indeed based on records taken on 9/11 (journals and emails) most people from my school fabricate what happened on that day. Basically, every important event from the year/fall is rolled into 9/11 for people who went to my middle school. There is an evacuation that stands out to me, most people remember it on 9/11, it was a different day according to my AOL.

            This is obviously an

          • Edward Scizorhands says:

            * Kavanaugh’s own calendar shows a beer party with PJ and Mark Judge with timing that aligns with Ford’s testimony: summer of ’82 a month before Judge worked at a grocery store. (Ironically Kavanaugh used this calendar to support his innocence.)

            If you are talking about July 1, Ford’s own team says that date doesn’t work.

            Memories fade after 35 years, except for the extremely emotional highlights

            Emotional highlights get distorted, too. If you are suspicious that the Republicans will forget this in a few months, you are not alone.

            (FWIW, I think a 1% chance of Ford being true and accurate is too low.)

          • J Mann says:

            Ask anyone where they were on 9/11, they will always have an answer.

            Just for the record, I was in my living room, playing Civilization on my laptop and listening to NPR in the background. My single clearest memory is the way the NPR host’s (Karl Kassel?) voice broke when he announced the second plane hit. Later on I was walking someplace and a guy told me he was Muslim and we hugged.

            Of course, if you told me you were sure one or both of those things didn’t happen, I wouldn’t be 100% confident, even though they’re some of my clearest memories today. They’re clear because they’re the ones I’ve recalled from that day, but the act of recall itself edits memories.

          • “Kavanaugh also was a multi-sport athlete. This means he was more than likely surrounded by a culture where women are treated poorly, and significantly increases the baseline rate of sexual assault”

            Did it hurt when you pulled that logic out of your ass?

            “Ford apparently communicated parts of her story earlier, to her husband, therapist, and some friend. There might be some doubt over each of these claims, but overall should count in her favor.”

            I like the “some doubt” and “parts of the story.” “I say I told you X, you don’t remember this, but that’s evidence in my favor, right? I’ve got a witness, like, sort of.”

            “Ford knew who Kavanaugh’s drinking buddies were: PJ, Mark Judge”

            They weren’t his only drinking buddies. Knowing two out of a dozen of someone’s friends does not require inside knowledge.

          • vV_Vv says:

            You’re implying that about 50% of superficially-credible accusations of sexual misbehavior against political figures are false?

            In this climate of extreme political polarization and #MeToo social panic? Yes.

            Memories fade after 35 years, except for the extremely emotional highlights. Liars tend to fill in the details to appear more credible! Having some fuzzy memories shouldn’t count against Ford, and may actually count in her favor if you agree the fuzziness is consistent with expected memory decay

            Oh come on! So if Ford gets a detail right it’s evidence that she’s telling the truth, if she gets a detail wrong it’s also evidence that she’s telling the truth!

            Kavanaugh also was a multi-sport athlete. This means he was more than likely surrounded by a culture where women are treated poorly, and significantly increases the baseline rate of sexual assault.

            The jerk jock villain is a movie stereotype based on envy.

            Is there any evidence that successful athletes tend to treat women more poorly than the average? Probably not, and my prior is that successful athletes, like successful people in general, tend to be more pro-social than the average.

            Ford repeatedly asked for an FBI investigation, Kavanaugh repeatedly dodged the question of FBI involvement when posed to him. Raises the likelihood Kavanaugh has relevant skeletons he doesn’t want unearthed.

            That was Feinstein sitting on the accusation, not Kavanaugh’s fault.

            Ford raised the accusations before Kavanaugh was nominated, when he was only on the shortlist. This dramatically reduces the probability Ford was acting on political motivations: you wouldn’t be very motivated until after the nomination.

            Seems to be false.

            Ford apparently communicated parts of her story earlier, to her husband, therapist, and some friend. There might be some doubt over each of these claims, but overall should count in her favor.

            There is no record of her naming Kavanaugh prior to his nomination. The therapist notes don’t report his name.

            Ford knew who Kavanaugh’s drinking buddies were: PJ, Mark Judge. So she almost surely actually was at a party with Kavanaugh. Moreover, for that memory to stick with her for 35 years, something memorable must’ve happened there.

            Or she just had passing familiarity with Kavanaugh’s social circle.

            Kavanaugh’s own calendar shows a beer party with PJ and Mark Judge with timing that aligns with Ford’s testimony: summer of ’82 a month before Judge worked at a grocery store.

            Duh. How is this evidence of guilt?

            There’s already a huge incentive not to make false accusations, such as perjury, and destroying your reputation, life, and career if it’s found out you’re lying.

            It’s extremely difficult to prove that she is deliberately lying, because the events allegedly happened 35 years ago, the account is quite vague and any detail that doesn’t match can be hand-waved as misremembering, as you just did in your comment.

            Therefore Ford had little to lose and lots to gain from making the accusation, like the $ 627,000 she got on her GoFundMe, in addition to praise of half of the country.

            In fact there’s already a huge incentive not to make true accusations, which you’re surely aware of with the #MeToo movement.

            I’m aware that the legal case against Harvey Weinstein is falling apart and that Asia Argento, one of the early accuser and leader of the movement, has admitted to paying off a former child actor to keep quiet about an underage sex allegation (which she denies, but there is a selfie of them lying in a bed possibly undressed). So tell me about the #MeToo movement.

            To preserve the legitimacy of the institution, you don’t want reputational problems with the individuals casting a cloud over what they do.

            It must also not open itself up to manipulation by false accusations.

      • Glen Raphael says:

        I think if you believe that, you go with base rates for a randomly selected accused person of being guilty.

        Isn’t a more relevant reference class the base rate for a randomly selected very famous and high status person? In that case, the base rate is probably quite high.

        On the other hand, perhaps we should use the base rate for a non-randomly selected accusation. By which I mean: Toxoplasma selected that we should hear about this story. If the story were easily resolvable either way we wouldn’t have heard about it so the simple fact that we’ve heard about it endlessly in the media suggests that it can’t be decided and has a moderate likelihood of falsity.

        So three different ways to construct a reference class give us either a low, high, or moderate prior likelihood of falsity. Science!

        • Isn’t a more relevant reference class the base rate for a randomly selected very famous and high status person? In that case, the base rate is probably quite high.

          Why?

          I would think that the probability of a famous and high status person committing attempted rape after he became famous and high status would be low, both because he has more to lose from an accusation than a random person and because a famous and high status person would have easier access to sex than a random person, hence less incentive to attempt rape. I can imagine more complicated arguments in both directions, but those seem the obvious ones.

          Off hand, I don’t see why someone who is going to be famous and high status in the future would be more likely to attempt rape now than a random person would.

          Could you explain your reason for the opposite conclusion?

          • NoRandomWalk says:

            I think the underlying premise isn’t that famous people commit assault at different rates, but that, as a function of being famous, they are more likely to be falsely accused of assault by individuals with a political or attention-seeking motive, or heck just because famous people tend to have interactions with more people increasing the chance of being the final object of mistaken identity.

            There are a lot more phantom sighting of Elvis than my neighbor Bob.

          • Conrad Honcho says:

            Off hand, I don’t see why someone who is going to be famous and high status in the future would be more likely to attempt rape now than a random person would.

            “Rape” no. Unwanted touching attempting to initiate consensual sexual intercourse that we now describe as “sexual assault” when it fails, yes.

            Consider James Franco, who was accused of (among worse things), pushing a girl’s head into his lap in his car. He was not trying to rape this girl. That is, if she objected he was not going to force her. In James Franco’s world, girls get into his car for the purpose of putting their heads in his lap. The girl who objected to having her head pushed into his lap is probably the 10th girl that week whose head James Franco pushed into his lap. The other 9 were delighted to have their heads in James Franco’s lap, and in fact having their head in James Franco’s lap was the reason they got into his car.

            While Franco has a 90% success rate with that technique, men who are not as famous, handsome, and successful as Franco do not push girls’ heads into their laps because they will almost certainly have a >90% failure rate.

          • vV_Vv says:

            Unwanted touching attempting to initiate consensual sexual intercourse that we now describe as “sexual assault” when it fails, yes.

            I’d like to see some evidence that very famous and high status people do this at a higher frequency than the general population. My prior is that sexual assault, like most crimes, is mostly done by low status men, but I might be wrong.

            But in this case this is irrelevant since Kavanaugh was not famous or particularly high status (at least compared to his peers) when he was 17.

          • Conrad Honcho says:

            I don’t know how one would measure such a thing.

            If we’re talking about forcible rape, then yes, that’s low status men. If we’re talking about “unwanted touching/groping,” then that’s in many circumstances indistinguishable from an attempt to initiate sexual intercourse that fails. A high-status man who’s used to women responding positively to his advances with little effort (i.e., movie star, pro athlete, rock star) will have more attempts at initiation, but without perfect success, a greater absolute number of failures.

            A story that starts “he asked me to come up to his room to see his collection of erotic Japanese lithographs and then he put his hand on my leg” can either end with “and a good time was had by all” or “therefore sexual assaulting me.” The difference being whether or not he was correct that she would appreciate that sort of thing.

          • vV_Vv says:

            A story that starts “he asked me to come up to his room to see his collection of erotic Japanese lithographs and then he put his hand on my leg” can either end with “and a good time was had by all” or “therefore sexual assaulting me.” The difference being whether or not he was correct that she would appreciate that sort of thing.

            But in this case we are expanding the definition of “sexual assault” so much to make it essentially meaningless.

            I mean, taboo “sexual assault”, let’s say that a man who once touched the leg of a woman who followed him to his room in the mistaken, though plausible, belief that she was interested in having sex with him, should we regard him as a social pariah unfit for any public office?

          • Doctor Mist says:

            more likely to be falsely accused of assault

            No, I think that was Glen’s second scenario.

            If I understood the first scenario, it’s a reference to the #MeToo revelation that men in power who, as DavidFriedman points out, you might suspect to be less likely either to need or to risk performing a sexual assault, in fact seem to be quite willing to do so with abandon. Absolute power corrupts absolutely, and all that. (Of course, there’s an availability bias there — we have heard about lots of powerful men who have credibly been accused, but it’s still only a minute fraction of all powerful men — the powerful men who lead wholesome lives do not make the headlines for it.)

          • Glen Raphael says:

            @DavidFriedman:

            Could you explain your reason for the opposite conclusion?

            My mistake – we’re actually in agreement. When I said “the base rate is probably quite high” I intended to mean the base rate of false accusations was high. My claim is that famous well-connected people tend to attract false accusations, because famous people’s names and faces and descriptions are extra-salient.

            Memory mistakes are common and the chance of mistaken identity does not track with the certainty level of the rememberer. If somebody remembers having a bad experience once and they happen to accidentally conflate their vague unremembered attacker with a certainty that it was some specific person, the person they accidentally falsely accuse is much more likely to be somebody famous than a nobody. It might be somebody they recently saw on tv or read about in the news or were arguing about online rather than, say, NoRandomWalk’s next-door neighbor Bob. So being famous makes one unusually likely to attract false retrospective accusations.

            Complicating this further is the fact that the sort of person who is likely to become a Supreme Court Judge is probably far more and better socially connected than the average person. When we look back on our own experience we probably didn’t go to as many parties or meet as many people of get remembered by as many people as did a prospective Supreme, so our personal intuition as to the size of the relevant attack surface is an underestimate.

          • Conrad Honcho says:

            But in this case we are expanding the definition of “sexual assault” so much to make it essentially meaningless.

            Most states do not have a crime called “rape.” They have “sexual assault” in various degrees, and I think largely because they wanted to taboo the term “rape” so they could get people to understand unwanted touching, or sex acquired through coercion is still very bad even if it doesn’t fit into the standard concept of “rape” as “stranger in the dark alley with a knife.”

            I mean, taboo “sexual assault”, let’s say that a man who once touched the leg of a woman who followed him to his room in the mistaken, though plausible, belief that she was interested in having sex with him, should we regard him as a social pariah unfit for any public office?

            Look at the accusations against Trump during the election. I tuned most of them out after the first one or two met that model, but I recall one being about a woman who said he asked her to come up to sit next to him on an airplane. When she did he was “all over her like an octopus.” She said no, and left. He did not stop her from leaving.

            Assuming it’s true, that’s basically the same scenario. Attractive lady, why did the think you man who has a notorious reputation as a billionaire playboy invited you to sit next to him? This is, I believe, the sort of thing people talk about when they talk about Trump “sexually assaulting women.”* How many other women completely understood why the billionaire playboy wanted them to sit next to him, and were quite willing to be pawed like an octopus by a rich and famous man?

            * Put the recanted accusation by Ivana in their divorce proceedings in a different bucket.

          • The original Mr. X says:

            Memory mistakes are common and the chance of mistaken identity does not track with the certainty level of the rememberer. If somebody remembers having a bad experience once and they happen to accidentally conflate their vague unremembered attacker with a certainty that it was some specific person, the person they accidentally falsely accuse is much more likely to be somebody famous than a nobody. It might be somebody they recently saw on tv or read about in the news or were arguing about online rather than, say, NoRandomWalk’s next-door neighbor Bob. So being famous makes one unusually likely to attract false retrospective accusations.

            Back when I was studying psychology for A-level, I remember coming across where a woman was raped with the television on in the background, and ended up “remembering” that the man hosting the TV show was the one who raped her. (Fortunately for him, of course, he had a cast-iron alibi, as he was on live TV when the crime occurred.)

          • mdet says:

            The girl who objected to having her head pushed into his lap is probably the 10th girl that week whose head James Franco pushed into his lap. The other 9 were delighted to have their heads in James Franco’s lap

            I gotta disagree that this isn’t sexual assault. Now it clearly shouldn’t be as much of a strike against him as if he was someone who didn’t take No for an answer, and we should make that distinction clear. But I fully support establishing the norm “Even if you’re the kind of celebrity who 9/10 women want to be touched by, the harm done by being groped by a stranger without your consent is great enough, and clarification is easy enough, that you still need to get some kind of quick confirmation before groping”.

    • Yosarian2 says:

      I feel like we have significant evidence.

      My probability started at maybe about 65%.

      It went up a lot when Ramirez came out and accused him.

      It went up more when Julie Swetnick came out and accused him.

      It went up a little more when we found out that at least in general terms Dr. Ford’s description of his youthful drinking and partying behavior was correct.

      And then in the testimony, well, this one is clearly more subjective, but Ford seemed extremly credible while Kavanaugh seemed mostly be trying to lash out at people and win partisan points, and he also seemed to have a pretty casual relationship with the truth in a lot of ways.

      I come out something like a 80%-90% chance that he assaulted Ford, and a higher chance that he assaulted at least someone. I am honestly confused how people ended up with a low probability, considering everything.

      • onyomi says:

        I think I correctly predicted that the additional allegations would help Kavanaugh’s chances, rather than hurt them. In addition to the possibility of something that should up people’s priors lowering them, there is also the issue discussed in this thread, namely that it’s a misapplication of Bayesianism to assume that the probability of at least one allegation being true can only go up as more allegations come out.

        This fails to take into account the fact that the appearance of new, incredible accusations should up your prior on peoples’ willingness to come out with fake allegations in general and in this situation in particular.

        So, for me, the Ramirez accusation and especially the Swetnick accusation lowered my appraisal of the truth of the Ford allegation once they were debunked (they very briefly raised them because, when you the “new allegations!” headlines the initial reaction is to think “uh oh, it’s not just a one-off but a pattern!”; however, looking into the details of those new accusations had the opposite effect and also increased my suspicion of the motives of those involved more generally: an attempt to pull an accusation “Gish gallop” or “moral-political DoS attack” (see tweet in first link).

        My appraisal of Ford’s credibility further dropped when we learned more about e.g. her story about installing a new door in her house due to trauma (in addition to the made-up fear of flying) and the FBI investigation came up with no corroboration after re-questioning those claimed to have been present.

        I answered the survey as “40%” with that being a probability on Kavanaugh having had, at some point, unwanted sexual contact with Ford, regardless of whether he remembered it or realized it was unwanted at the time. I’d say I’m now more like 35%. I will also say I can’t understand how anyone can seriously give an answer close to 0 or 100 given how few facts we’re working with, and how old the relevant memories are.

        • Murphy says:

          There’s an old expression “damn with faint praise”

          Damn with faint praise, assent with civil leer,
          And without sneering, teach the rest to sneer;

          From an old book review:

          “Funnier than Psycho… more chilling than Jeeves
          Takes Charge… shorter than War and Peace…”

          I think there may be a corollary for the inverse.

          When Jimmy Savile sex-abuse stuff hit the news, dozens of credible cases turned up. Rolf Harris, ditto. Cosby again ditto. They’re damned by lots of strong evidence.

          On the other hand, when half the people of power in the country start combing someones history for anything, anything at all with their name plastered across the media…. and the results…. are they find some people willing to say “I could totally imagine him doing that” and a date that could maybe kinda sorta fit and the revelation that he drank beer and was an asshole as a teenager…..

          I think that has the inverse effect, ie “you searched everything you could find about his life…. and that is the most damning thing you came up with???”

          it’s a little bit like the police proudly showing off results like these from “weapon sweeps”

          https://imgur.com/a/PDPRoN0

          • I would agree if the question is “did he routinely attempt rape.” But I don’t see any conflict between “he tried to force a girl once when he was seventeen and drunk” and “he never did anything like that again.”

        • Yosarian2 says:

          Can you explain how you think Ramirez was “debunked”?
          If anything, we at least have a photo of her with Kavanaugh before the incident, which seems like strong evidence that he was not telling the truth that he has never met her. The fact that he was apparently trying to refute her allegations before they became public seems even more damming.

          But if there’s evidence that her allegations were “refuted” I’d like to hear it.

          • Murphy says:

            “we at least have a photo of her with Kavanaugh before the incident”

            If you look at my facebook feed there’s various photos from over the years at various events. Sometimes smiling for the camera with a classmate and her extended gaggle of friends or some guy the guy next to me was chatting to the moment before or a friend of a friend of the person who’s 21’s we’re at….

            Do I know their names? Nope. If one of them accused me of being part of a plot 15 years ago and I said “who the devil are you? I don’t even know you” that would be entirely honest. My actual social circle is fairly modest at only a few dozen people I really knew or hung out with. But I’m certain I’m in photos with many times that, if you count larger group photos then add a few more orders of magnitude.

            doing a vague fermi estimate and with how many photos people take nowadays you could probably place me in at least one photo with something like 10% or more of the population of the city I grew up in within about 1 year of age of myself pretty easily.

            I was not a very social person. Someone who regularly attended parties and went out a lot could outstrip my own estimate for myself pretty easily.

          • Yosarian2 says:

            To be clear, I agree that the photo is not very strong evidence in her favor, there are alternate explanations as you say, but it is evidence. I just found the claim that she was somehow “debunked” to be shocking since as far as I am aware all of the admittedly limited evidence we’ve gotten since her accusation has so far supported her story and weakened his.

          • Conrad Honcho says:

            I don’t believe Kavanaugh ever denied knowing Ramirez. He denies ever exposing himself to her. No one has been able to corroborate her story, and it’s my understanding she’s uncertain it was Kavanaugh who did this.

            I don’t know if I would call that “debunked,” though. Just weak or improbable.

          • J Mann says:

            @Yossarian2

            If anything, we at least have a photo of her with Kavanaugh before the incident, which seems like strong evidence that he was not telling the truth that he has never met her.

            Kavanaugh did not deny meeting Ramirez. In his September 25 interview with Dem and GOP investigative staff, Kavanaugh said:

            Do you know Deborah Ramirez?
            Judge Kavanaugh. I do.
            When did you meet her?
            Judge Kavanaugh. I knew her in college.

            (Kavanaugh did deny meeting Swetnick, and stated that while he doesn’t remember meeting Ford, he might have. Could you be confusing Ramirez with Swetnick or Ford?)

            You also write:

            The fact that he was apparently trying to refute her allegations before they became public seems even more damming.

            It’s pretty frustrating that the press won’t release the timeline in their possession. The NYT has emails from Ramirez looking for support for her account (and conceding that at that point, she wasn’t sure if it was accurate). Somebody has texts showing that “Brett’s guy” was trying to gather information to refute Ramirez’s story. The New Yorker knows when they contacted Kavanaugh for his side of the story.

            If Kavanaugh’s team started investigating after they learned that Ramirez was contacting people or after the New Yorker contacted them, that wouldn’t IMHO be suspicious. If they were contacting people before Ramirez and the New Yorker’s activity, that would be pretty much 100% damning.

            My guess is that if the latter were true, the news coverage regarding the “Brett’s guy” texts would have included the dates.

          • The original Mr. X says:

            “Debunked” might be a bit strong, but the fact that she apparently had to spend six days asking around and consulting her solicitor before concluding that it was Kavanaugh should raise a big red flag. I mean, that’s pretty much a textbook example of the sort of situation that would lead to unreliable recollection.

        • The original Mr. X says:

          My appraisal of Ford’s credibility further dropped when we learned more about e.g. her story about installing a new door in her house due to trauma (in addition to the made-up fear of flying) and the FBI investigation came up with no corroboration after re-questioning those claimed to have been present.

          Don’t forget her refusing to hand over her therapist’s notes to the Senate Committee, despite being repeatedly asked to and despite the fact that one of the main arguments in her favour had always been “But she said she was abused to her therapist, so clearly she wasn’t making this up.” When somebody refuses to disclose a key piece of allegedly corroborating evidence, my confidence that it is actually corroborating goes down significantly.

      • Chevalier Mal Fet says:

        I started about the same, but went in the opposite direction. I started at about 50%. Then, as each piece of evidence piled up, and all uniformly pointing in one direction, I slid my probability down. Down a little bit when Judge failed to corroborate, down a bit more when all the other witness failed to corroborate, down more when he showed the calendar. The fact that he was angry at being accused of gang rape during the hearing did not alter my judgment in any way – maybe made me a little more favorable to him.

        The Ramirez and Swetnick accusations never even passed the laugh test for me. Both were extremely dubious to start with and both basically fell completely apart on further examination (Swetnick’s interview where she walked the whole thing back, the follow-up New Yorker story on Ramirez). Similarly, the weak perjury accusations collapsed one after another (like when everyone was insisting that he lied about the “Devil’s Triangle” and it later emerged that, yup, it was in fact a drinking game), which lowered my estimation of any of the reporting on the issue.

        I wound up at about 10% that he assaulted Ford, with a higher probability that something happened to Ford that summer but she misidentified Kavanaugh. I’m honestly confused how you could get such a high probability, given everything

        • Yosarian2 says:

          My impression is that Judge’s story seemed to contradict itself and things he earlier had said, and the weakness of his story is why the Republicans didn’t want him to testify.

          • I don’t think Judge’s denial is much evidence either way, since if Ford’s story is true it is quite likely that Judge would lie about it.

          • Joyously says:

            Republicans didn’t want Judge to testify because it would be a long series of Democrats asking him about his embarrassing drunken escapades as a way to cast an air of drunken debauchery over Kavanaugh by association.

            The FBI interviewed him for hours and didn’t seem to turn up anything damning.

            But either way he’d be fully expected to lie about the incident at hand, and the odds of breaking him and getting him to burst out with “You’re goddamn right we did it!” are very low.

      • Aapje says:

        @Yosarian2

        Are you aware that Swetnick’s only actual observation of Kavanaugh doing anything was him standing in a group at a party? The rest of her story linking him to ‘rape trains’ is conjecture.

        I don’t see how this should move the needle towards him being guilty. If anything, it should lower the probability, since it shows the level of motivated reasoning that this nomination can produce.

      • eqdw says:

        THE FOLLOWING IS NOT MEANT AS A CRITICISM OR DISAGREEMENT WITH YOU, BUT SIMPLY USES YOUR COMMENT AS A JUMPING OFF POINT TO DISCUSS SOMETHING DIFFERENT.

        And then in the testimony, well, this one is clearly more subjective, but Ford seemed extremly credible while Kavanaugh seemed mostly be trying to lash out at people and win partisan points, and he also seemed to have a pretty casual relationship with the truth in a lot of ways.

        I watched both of these testimonies and came to the exact opposite conclusion.

        I find it extremely curious that this seems to be a pretty common occurrence. Tons of people are watching the exact same testimonies and coming to the exact opposite conclusions. How does this happen? Why does this happen?

        How is it possible that two different groups of people, who are acting honestly and in good faith to the best of their abilities, can see the same things and come to opposite conclusions? Is it peer pressure? Are we really all that much slaved to our political factions? Are we not actually watching the same things (eg different edits of the testimonies)? Is there a secret government mind-control ray that’s indoctrinating half the population (of course, it would be the other half. I decided on my beliefs through pure, untainted reason /s)

        This is a seriously weird effect, I think it needs explaining, and I think it’s indicative of a deeper problem.

        • Yosarian2 says:

          That’s interesting, and I’m not sure what the answer is.

          Based on some of the commentary I think it might have a lot to do with empathy, and which person between the two the listener felt more empathy for.

          I don’t think it’s just political faction either, at least not in the narrow sense; I feel confident that if this was a liberal judge being accused most of the same people would still believe her over him, and I think if this happened in an equally credible way a Democratic President would have withdrawn him. Maybe the “believe the victim” idea is just more deeply ingrained in people on the left?

          But yeah; not only did basically everyone I know find her credible, they all thought it was so objectively obvious she was credible that they were expecting the Republicans to drop the nomination and appoint someone else after hearing her testify, and were shocked and horrified at how the Republican Senators responded later.

          • Edward Scizorhands says:

            Even if we had no cultural or political biases, we would still be basing or assumptions on what we told ourselves ahead of time the way “people ought to act” when under pressure. And this is based on a bunch of experience we bring to the table and aren’t even aware of.

            I remember reading Hamlet in high school and I thought Hamlet’s plan — to have a play of a guy being murdered by his brother, and see if his uncle “acts guilty” — was shit, because even if innocent his uncle would still feel bad about it. The teacher blew me off. But there is no playbook for how people are “supposed” to act when lying or telling the truth, and even if there were for something as high profile as this both people would have been coached from the same playbook.

          • J Mann says:

            @Edward Scizorhands

            You were correct, IMHO, about Hamlet. The thing is, if Claudius were innocent, he might very well see the play as a direct threat against his life. (Hamlet has arranged the play and it is about a nephew killing his uncle.)

            Until we see Claudius at prayer, the audience doesn’t actually have enough evidence to know if the specter’s accusation is true.

          • tscharf says:

            Two Words: Bill Clinton.

            I suggest revising your expectations, ha ha. My take is this is more a “Believe the victim” vs “Believe the (lack of) evidence” battle.

            I put less weight on subjective testimony. Both of these people were coached by people very good at their jobs. I have known some really likable and gifted liars in my time. Suffice it to say that Meryl Streep could have knocked it out of the park with a false accusation. Put her up against somebody on the autism spectrum and she will win every time.

            When you have nothing else to go on but a credibility assessment in a he said / she said then there just isn’t enough information to make a material decision on.

            Who do you believe most is an interesting debate, but the real question is what is the level of evidence required to prevent Kavanaugh from being confirmed? It’s too late to ask that question now, as the post-hoc rationalizations are already set in stone.

          • rlms says:

            What about Bill Clinton? Probably most normie Democrats would be dubious about accusations against him, but I expect people here who think he was a good president would nevertheless tend to believe he probably did sexually assault various women. If not I’d be interested to hear why.

          • Protagoras says:

            @Edward Scizorhands, Not sure about that teacher of yours; as you note, Hamlet’s plan is crap, and there are certainly interpretations of Hamlet on which one would expect him to be employing a crap plan.

        • baconbits9 says:

          Its because people are pattern matching.

          He was angry and defiant in testimony and she was composed.

          One person sees this and thinks “He is acting guilty and lashing out to protect himself while she is acting truthful and confident” and another sees it and goes “I’d be angry if I was falsely accused and she seems rehearsed and to calm for someone claiming traumatic attack.” (I had to reword this a few times just to try to make it sound like balanced positions to me).

          The issue is that no one has any idea how to weigh the ‘evidence’, but anyone can come up with a narrative to justify why the evidence supports their position as they aren’t bound by physical laws but interpretive frameworks (they don’t even have to be consistent, I complained early on that people supporting Ford were simultaneously forgiving her memory lapses about the event while being sure that what she remembered must be true). So the majority of opinions just get pushed one way or another by individual biases.

          I would also state that 50/50 in this case isn’t neutral, in a parallel universe where you knew the likelihood of such an accusation being true you would be able to start with a “neutral” position and make updates from there, but those odds are unlikely to be 50/50 and so we are starting off our neutral position with an expected error that we know neither the direction or magnitude off. People make this mistake all of the time, to highlight imagine that such an accusation was 95% likely to be true with a bare bones outline of the case. This actually would alter how you should view the evidence, it should now take very strong evidence to give him a pass and any weak additional evidence on the accusers side should actually be viewed as quite strong. Having an inaccurate frame of reference actually distorts how you should interpret evidence and the only thing that people (who are looking for accuracy) ought to be acknowledging is that we do not have good reason to believe we have an accurate frame of reference.

          • baconbits9 says:

            What actually bothers me right now is that The Black Swan was recently reviewed and discussed and one of Taleb’s points was that “I don’t know and can’t put a probability on it” is an important position to be able to take.

          • eqdw says:

            @baconbits9

            “I don’t know and can’t put a probability on it” is the core of my position on the Kavenaugh thing. I then refine that with some judgement calls based on metadata, but overall I think that there is close to zero hard evidence for _any_ position on this issue.

          • dick says:

            What actually bothers me right now is that The Black Swan was recently reviewed and discussed and one of Taleb’s points was that “I don’t know and can’t put a probability on it” is an important position to be able to take.

            Agreed, and I think all the SSC Kavanaugh threads could be lightly edited and published as an essay titled, “Why the Ludic Fallacy is Bad: a Case Study”.

        • Dan L says:

          This is a seriously weird effect, I think it needs explaining, and I think it’s indicative of a deeper problem.

          There is a substantial amount of data indicating that humans are categorically awful judges of sincerity, to the point where even professional interviewers and interrogators scarcely beat chance when denied access to outside evidence. It’s not surprising that judgements will then break along partisan lines.

          This is a disturbing finding, but it is also inconvenient and unflattering so it tends to be ignored. Kinda like “it’s actually pretty easy to produce a false confession”.

          • Aimless Aardvark says:

            I am a lawyer by trade. I spend a lot of time asking people questions who have strong incentives for a certain story to be true. I have, on occasion, been able to confirm people are lying under oath. This has almost always been as a result of outside evidence directly contradicting the person; in my experience there are no consistent or obvious tells in how a witness tells a story which accurately indicate truth or falsehood.

            My job has taught me to have zero faith in my own ability to judge the truth based on how a witness presents under questioning. What people are usually evaluating when they think they are determining a witness’s truthfulness is in fact the witness’s likability. There is for all intents and purposes ZERO correlation between how a witness presents in court and whether they are telling the truth. Some people tell stories well, some don’t, some people are sympathetic, some aren’t, some people have charisma, some don’t.

            Now, from my perspective, how a witness presents is important as a practical matter, as it may impact me winning or losing a case for my client. But it doesn’t move the needle at all on whether they are telling the truth.

            I did not watch all of the Kavanaugh hearing, but I watched excerpts from both Ford and Kavanaugh testifying. Nothing in how either of them gave their evidence impacted my opinion on their credibility, and it dismays but does not surprise me that so many people drew such strong opinions from absolutely useless data.

            All that to say, I agree with you.

        • eqdw says:

          Comment update:

          After leaving this comment I realized that I had not actually watched Ford’s testimony in full, I’d only seen clips of it. In the interest of full transparency and honesty, tonight when I get home I’m going to watch her testimony in full online and see if I still hold the position I do.

          However, I watched all (or most; I tuned in early enough to get like 2, 3 hours worth? I think) of Kavenaugh’s testimony and I still really don’t understand what people mean by saying he was angry. He was remarkably composed given all that he’s going through. I have to wonder if the people saying it was obvious how angry he was have never actually seen someone be actually angry. Like, throwing things into walls level angry.

          Or maybe I don’t know what I’m talking about. I don’t know. It’s still absolutely bizarre to me that people are coming to such opposite positions. But then again, for that matter it’s bizarre to me that him being angry is taken as evidence that he _did_ it. If I was falsely accused of a crime that I knew I didn’t commit, and if I was being harassed and my family was being harassed and my future was being threatened over a falsehood, that would make me very angry. But if I had at some point in the past committed a horrible assault against somebody, and now the world had found out, I don’t imagine myself getting _angry_ about that in the first place.

          Of course, I can’t imagine myself committing an assault like that in the first place, so maybe I should hedge against typical mind fallacy and accept that aggressive, violent people have different emotional responses than I do.

          In any case, my core point remains: this is a curiously large difference of opinion

          • eqdw says:

            Another update.

            I’ve now listened to the first half of Dr. Ford’s testimony at this youtube link here.

            1) My first impression is that her testimony seems superficially plausible and is not obviously wrong to me. This is roughly the same way that I feel about Kavenaugh’s testimony.

            2) A handful of things that I’ve seen so far have made me question her reliability. Specifically, the random neurobiology lesson that has been all memed about on right wing social media really does strike me as bad. This makes her lose a minor amount of credibility to me. However, in her defense: I am listening to this testimony specifically to develop an opinion about it. I listened to the Kavenaugh testimony while drinking and playing ZeroK. It is possible that he committed some similar type of gaffe. So, while this definitely makes her less reliable in my mind, it doesn’t necessarily make her proportionally less reliable than him.

            3) For various reasons that are 100% idiosyncratic to myself and not something I would expect other people to agree with, I am biased to think she is less trustworthy.

            Overall, after having listened to her testimony, my position is not largely changed (although I realize in my original post I may have overstated my perspective on her testimony): Her testimony rubs me the wrong way but is superficially plausible. His testimony still feels fairly solid to me. Neither are obviously wrong.

            I remain perplexed as to why people think Kavenaugh’s testimony is obviously untrustworthy.

          • HeelBearCub says:

            @eqdw:
            Why does Kavenaugh’s testimony seem untrustworthy?

            Because he lied about his drinking. He lied about whether he would party on a weeknight by saying he was too busy with schoolwork and football paractice and working out. He lied about whether his calendar showed a party on a weeknight. He lied about whether people had refuted Ford’s testimony. He lied about other things.

            He also engaged in the kind of “how dare you question me” high dudgeon, both during his prepared statement and under questioning that smacks of not wishing to address the actual allegations. Ex: When asked whether he ever drank to excess his response was that he went Yale. This is a non-sequitur, at best. When asked whether he had ever blacked out drinking his response was “Have you?”

            He all but claimed that Ford was a plant by the Democratic machine, while also claiming that he simply thought Ford was mistaking him for her actual assailant.

          • The Nybbler says:

            Because he lied about his drinking

            Has not been shown. He has said he on occasion drank to excess, and he said he had never blacked out from drinking.

            He lied about whether he would party on a weeknight by saying he was too busy with schoolwork and football paractice and working out. He lied about whether his calendar showed a party on a weeknight.

            He did not:
            “The calendars show a few weekday gatherings at friends’ houses after a workout or just to meet up and have some beers. But none of those gatherings included the group of people that Dr. Ford has identified. And as my calendars show, I was very precise about listing who was there; very precise.”

          • Doctor Mist says:

            He all but claimed that Ford was a plant by the Democratic machine, while also claiming that he simply thought Ford was mistaking him for her actual assailant.

            Both could be true.

          • eqdw says:

            @heelbearclub

            You are misunderstanding what I mean when I say their testimony seems reliable.

            When I am watching their testimony and judging for reliability I am explicitly discounting the facts available outside of the testimony.

            Why? Because if I say “Kavenaugh seems untrustworthy because he said this thing, when outside facts say this other thing”, this is not a statement about his testimony, this is a statement about the outside facts.

            Specifically, the question I’m trying to investigate here is “how can two people take in the exact same data (the testimony) and come to exact opposite conclusions?”. If you’re factoring in data from outside the testimony, then the answer becomes trivially obvious: They’re not going after the same data. In that case, there’s no mystery, and this topic stops being interesting to me

            Additionally, to soapbox for a second: Based on my media bubble, I’m aware of dozens and dozens of datapoints that suggest that Ford is a knowing liar who _actually is_ a democrat plant, and I am aware of virtually no credible evidence that Kavenaugh is the same.

            However, I’m also-also aware of the fact that someone like you will say the exact same thing, only politically bit-flipped.

            This suggests to me that, in actual fact, every source of “information” on this controversy is untrustworthy, made up of lies and/or partisans who are aggressively trying to push for their own position, regardless of truth.

            This, in turn, sends me right back to my original position of “I have no data, therefore null hypothesis, large error bars, and I don’t care”

        • cryptoshill says:

          @tscharf – the answer, as brought up by the words “Bill Clinton” is – “If the accused is in my outgroup, any single shred of evidence is enough to guarantee that this person should never ever hold political office ever. If the accused is in my ingroup I demand a time machine”

          Personally, I have a lot of biases against “any accusation in which the people who are critical of it are being called to the carpet for not believing women enough”, but I still have to admit that drunken partiers aren’t exactly the least likely to be doing these things – so I would estimate the probability of Kavanaugh committing such an act at 15-30%, I am willing to raise that estimate if we consider “consensual sexual shenanigans that became nonconsensual at some point, whether or not it rose to the level of a crime (including here that there’s an implicit argument here about what constitutes the revocation of sexual consent).”

          NOTE: I did not answer the survey.

          • tscharf says:

            The interpretation of who you believe when there is a large amount of uncertainty miraculously comes down on the side of your prior biases. When Bill Clinton was accused (Paula Jones et. al.) the left believed him at the time. After he finished serving his leftist purposes, people now reevaluate those claims more negatively. There are two moving parts here though, one is whose ox is getting gored, and the other is the believe women madness which is rather recent.

            In any event, the left will be tested eventually. They have eaten plenty of their own like Franken, but he is just replaced with another left Senator. The real test comes when a politically meaningful accusation is made.

            Imagine someone accuses one of the liberal SC justices tomorrow of something similar 30 years ago without any corroborating evidence. Are we to expect calls for his impeachment from the usual suspects in the media? I’ll go out on a limb and suggest that won’t happen and everyone on both sides will scream hypocrites at each other.

      • invective says:

        Accusations are not evidence.

        To my knowledge there is no evidence at all.

        Every claimed witness has refused to corroborate her statement.

        What exactly would have to be presented for you to believe his innocence?

        • Yosarian2 says:

          If an accuser is willing to testify about what happened under oath, then that is clearly evidence.

          • But it’s very weak evidence in a situation such as this where there were hundreds, probably thousands, of potential accusers in an equally good position to much such an accusation and with a strong motive to do so.

          • baconbits9 says:

            @ David Friedman

            Only if you made a bunch of assumptions about the frequency of such claims.

          • tscharf says:

            The frequency of claims against right leaning judges for the SC seems to be rather higher than the background rate. N may be small, but it implies something else is going on.

          • rlms says:

            Why would a scheme to falsely accuse right-wing supreme court candidates of sexual assault not extend to right-wing politicians?

          • tscharf says:

            Because they are elected, and that takes care of itself.

          • invective says:

            This is not evidence.

            Military pilots and astronaughts have claimed to see aliens, along with thousands of regular people.

            Is this evidence?

            If I claim to have personally met jesus, would you take this as evidence?

            Please tell me exactly what a man of impeccable moral credentials for the last 30 years would have to do to convince you otherwise?

            I find this casting to the wind of pressumed innocence terrifying.

            Forget the frequnecy claims others have mentioned. As someone else has stated, it’s extremely unlikely that you win the lottery, yet SOMEONE does.

            I’m from Scotland, and perhaps evidence and proof mean something else here, but this whole fiasco has come across as repulsive to me.

            Never, ever take a claim at face value without proof. Ever.

          • invective says:

            Also, by your own logic, his testomony is evidence that it did not happen.

          • eqdw says:

            They both testified under oath, and they both gave conflicting testimony. Based on this, we know definitely that at least one of them said something that was false, under oath.

            Consequently, for me this _heavily_ discounts “gave testimony under oath” as evidence for that testimony’s trustworthiness.

      • tscharf says:

        My confidence went up that Kavanaugh was innocent with the additional allegations. Even though they should be treated independently, I found that the media framing and uncritical parroting of these allegations showed that there was a search for guilt instead of search for truth going on. I will note that the NYT refused to print these stories until they were outed by others. It was a just throw these things out there and see what happens mentality.

        The combination of allegations looked exactly like an orchestrated hit job that the media was more than happy to take part of, and it ended us being a major disservice to Ford. It easily allowed the right to conflate the accusations to their advantage.

        • oncomer says:

          Why does the media treatment of the accusations change your confidence in their truth one way or the other?

          • Edward Scizorhands says:

            All the many many times the media said that Kavanaugh has committed perjury, seriously you guys, we got him now … only to totally forget about it a day or two later and want to talk about something else, pushes me away from believing that Kavanaugh was lying.

            In isolation the media’s lies shouldn’t do that, but I have to do something to correct for the fact that I’m swimming in sea of lies.

          • tscharf says:

            The short version is that all information is being channeled through the media, the media has some baseline credibility, and the further stories showed a lack of journalistic integrity. This lowers my confidence that the first story was treated with journalistic integrity.

            The media did a lot of pounding on the table, as in a constant stream of appeals to emotion. Other unrelated survivor stories and so on. This struck me as an avoidance of examining the facts of the case at hand. Nothing tips one off quicker to media bias than the anecdotal stories they choose to tell. It doesn’t make Ford’s story true or false, but it indicates what the media prefers it to be. I don’t recall reading any anecdotal stories about men falsely accused.

          • oncomer says:

            The algorithm that writes the bottom line of your thought process shouldn’t be “Believe whatever the opposite of the media wants me to believe.” “The media” is not a reliable authority, but it is also not a reliable anti-authority. News coverage having a short attention span, or uncritically parroting allegations, is not unique to Kavanaugh coverage– its presence does not distinguish between a world in which Kavanaugh is innocent or guilty, because it is present in either case.

            It would be one thing if we suspected that media coverage had suppressed important information, but that is pretty much impossible when almost everyone, including organizations with opposed slants, are drawing from near-identical sources– mostly the publicly broadcast hearings. Instead the “Sea of Lies” is mostly use of unbiased tone, emotion, etc. In this case, no group can stop their ideological rivals from reporting any facts inconvenient to their interpretation. This kind of reporting unduly affects public opinion, but shouldn’t actually prove persuasive to someone like you who is trying to critically assess the facts of the case. It doesn’t effectively muddy the water if you are paying attention.

            So if in this case biased media coverage can’t prevent you from learning pertinent facts about the case, and you are capable of disregarding appeals to emotion, and the media coverage would be the same regardless of the truth of the matter, it might be a mistake to take a cue from it. The world’s cleverest arguer may point out that the sun is shining, and yet it is still probably daytime.

          • tscharf says:

            I don’t suspect they suppressed information, I know they did. The WP knew about Ford’s friend and that she couldn’t corroborate the event when they first were told about it. For reasons I cannot fathom, they did not report this information. It came out about a week later. The more I looked into it, the worse it got. The moving target of dates 1980’s, early 80’s, mid 1980’s, summer 1982? There is more, these are all summed up by the prosecutors report to the Senators.

            If you determine the standards of reporting have changed, then you should examine their evidence more skeptically. It doesn’t change how I evaluate the evidence, but it changed how much I could trust their view and whether I needed to look harder if I expected to find the truth.

            They rarely highlighted how little evidence there was against Kavanaugh. The mass psychosis of “Believe Women” was affecting their coverage.

    • The Big Red Scary says:

      “I went to take the survey, but left without filling it out”

      The same here. I didn’t follow the story carefully, found most sources of information on it dubious, and wasn’t sure what was the point of assigning a 50% probability.

      Some important under-addressed issues seemed to me to be 1) setting up perverse incentives (as others have pointed out) and 2) being so caught up in speculation about the accuracy of the accusations that there was little time to discuss other less speculative reasons one might support or oppose the appointment of Kavanaugh.

      It is surely worthy of note that a SSC poll exhibits such partisanship. I wonder if the results would look significantly different if readers who abstained from voting on it and had participated.

    • arlie says:

      I also abandoned the survey unfinished, with a feeling that it was asking the wrong questions, and no repsonse I gave would be meaningful. The truth is, I see that whole shitstorm as farther evidence of two things:
      – American “beliefs” depend on party affiliation
      – American “justice” is about selecting people based on party loyalty, particularly including predictability of future votes, and absolutely nothing else matters.

      That was actually more evident when the Great Leader was nominating people for the supreme Court position who appeared to have about as much knowledge of the law as I do – i.e. not enough for the job, unless of course the job is really just to vote the right way based on political considerations.

      It’s just that attempted rape etc. etc. is more of a hot button issue than incompetence … and of course in this case, there was always the hope (on one side of the aisle) that it might work. Whether or not the guy was guilty of more than being a very nasty piece of work as a youth doesn’t matter all that much if your concern is whether he’ll do what you want now, perfectly legally 🙁

      I also didn’t much care whether this specific party loyalist would get confirmed – some hyper-politicized, loyal to the cause Republican was pretty much guaranteed. If anything, there was a politcal advantage to confirming one who was more likely than average to misbehave in the future badly enough to get himself impeached 🙁 (All this presuming that there was no chance of getting someone who believes in unpopular things like rule of law :-()

      Anyway, that’s why I didn’t follow the case at all closely, and had no clue about either “are you well informed” question.

      • Careless says:

        “some hyper-politicized, loyal to the cause Republican was pretty much guaranteed.”

        Have you seen the results the Republicans have gotten from their nominations? The Democrats? yes, 100% loyal to the party. Republicans? Maybe 50%

  2. Robert L says:

    If you were a Senator, should you vote probabalistically? If you think the chances are 66% that Kavanaugh was guilty, what is wrong with rolling a die and voting against on 1-4 but for on 5 or 6?

    • Michael Watts says:

      What’s wrong with it is that it’s stupid. It’s stupid in the sense that people who learn that you’re making decisions that way will lower their opinion of you, but it’s also stupid in the more objective sense that following this strategy will make you wrong more of the time. With that methodology, you’ll make the “right” choice 5/9 of the time, whereas rolling the die and voting against on a 1-6 would make you right 6/9 of the time.

    • Philosophisticat says:

      If confirming a guilty K is at least half as bad as failing to confirm an innocent K, then voting probabilistically has a lower expected value than voting against, and if confirming a guilty K is less than half as bad as failing to confirm an innocent K, then voting probabilistically has a lower expected value than voting for. There is no argument I can see for voting probabilistically – it is always irrational.

      • slightlylesshairyape says:

        I think this is it. People are focused very strongly on the question of guilt, but no one actually has explicitly stated what they think the utility of the off-diagonal outcomes are.

        Over on the sub-reddit, I made a comment that implicated the payoffs. I realize now I should have modeled it first.

  3. cactus head says:

    >But between 25% and 80%, people were pretty split on whether the Senate should err on the side of not confirming a potential assaulter, or wait until it was beyond a reasonable doubt.
    Is this ‘reasonable doubt’ supposed to be taken as the legalese meaning of the term? Are there any lawyers here who can clarify what the term means in theory and in practice, i.e. what probability is it supposed to correspond to (if it is supposed to do so in the first place)? What proportion of people found guilty beyond reasonable doubt are actually innocent?

    • theodidactus says:

      Hello
      2L law student here.
      The courts are famously mum on the exact probabilities, and in fact I believe there is at least one case where the court holds that one is forbidden to quantify reasonable doubt. The usual metaphor is you would order your most important affairs around it, IE risk your life savings or life on the amswer. personally I think its useful to say higher than 95% but lower than 99%.

      There are other standards. Preponderance of the evidence at 51%, probable cause at below that ( my personal helpful qauntification of probable cause is 33% or lower depending on the crime)

      • Gazeboist says:

        Never forget “clear and convincing”! (Usually described as ~75%)

      • massivefocusedinaction says:

        Wow, I guess I’m risk averse, I wouldn’t even stop to consider a bet where my live savings was at risk with a 95% chance of success, I’d even reject 99%, and start listening to the odds with something like 99.9% or so.

        • sandoratthezoo says:

          Surely it would depend on the reward as well? I’d be willing to risk my life savings on a 95% probability of 10x-ing them. Not sure I’d be willing to risk them on 2x-ing them.

  4. Deiseach says:

    I asked people whether they felt the evidence that Kavanaugh may have committed sexual assault was sufficient to reject his nomination to the Supreme Court, regardless of any other reasons to vote for or against him (like his legal opinions).

    Personally, for me “may have” was doing a lot of heavy lifting there. If it were proven, or at least established very well, that he did commit sexual assault – yes. He might have done? No.

    And the waters were further muddied by “define sexual assault in this instance”. We had three accusations: from least to most serious they were – drunken indecent exposure, drunken groping/attempted rape (giving that the most weighty charge), sober participating in drug rape gang parties.

    “When he was young and stupid and at college, he got drunk at a party and thought it would be a great joke to waggle his bits in a woman’s face” – stupid idiot behaviour, deserved a kick in the pants at the time, but not disbarring him from, nearly forty years on, sitting on the Supreme Court (unless he is still getting drunk and waggling his bits in strange women’s faces, which nobody seems to have claimed is happening).

    “When he was still in high school, he and a cabal of other guys were spiking drinks with alcohol and/or drugs to get women unable to resist or refuse, then lining up to gangbang those women” – well, that’s the most serious charge and if there’s any proof at all, then not alone should he not be a Supreme Court judge, he should probably be looking at some amount of jail time. Then again, even the accuser walked it back to “I saw him and his pal in the drug gang rape line, but I can’t say they actually did participate in drug gang rape”.

    That leaves Ford’s accusation, the one both “not batshit crazy” and “fairly serious” of the three, and again we don’t have anything like independent corroboration from anyone else, and I’m not going to re-argue the whole case here all over again.

    So on balance – no, the possibility he might have dunnit should not debar him from the seat.

    EDIT: I’m probably biased because of the experience, as I’ve said, of witnessing at first hand a lynch mob mentality over a rape accusation where it turned out the accused parties were innocent and the accuser and her witness were lying/self-deceived, but before that came out everyone from the Director of the Dublin Rape Crisis Centre on down (and up) was going with “plainly guilty, no presumption of innocence, lock up the monsters, monitor them for life so they can’t do this again”. It was even more egregious because when the account was firmly examined it immediately fell apart, but it seems like nobody initially wanted to do the hard questioning of the ‘victim’ because come on, that would be too horrible and mean, even when she changed her story once the physical impossibility of what she claimed had happened was demonstrated (they went with a second charge based on another instance she claimed happened instead).

    • slightlylesshairyape says:

      I don’t know. If we want to assure with 95% probability that there are no rapists on the Court (a reasonable goal, IMHO), then we need to be 99.4% sure that each of the 9 justices is individually not a rapist.

      • Deiseach says:

        If we’re going to go into what people may have done in the past, take Justice Bader Ginsburg. She was born in 1933. Think of social attitudes in the 40s and 50s and what it was acceptable then to say about minorities (both ethnic and sex/gender/orientation). Can anyone be absolutely sure she never ever said anything that would nowadays be considered beyond the pale?

        If deadnaming is violence, did she ever do that? Ever use a term for gay people, POC, etc. that is now considered evidence of racism and homophobia? Never ever said anything back in 1953 that today in 2018 would not be acceptably woke?

        Glass houses and stones.

        • Dan L says:

          Do you actually believe that being old is about the same evidentiary level as sworn testimony, or are you just parroting something you don’t believe in order to make a politically expedient false equivalence?

          • Rick Hull says:

            No, and no. You seem to have a charitability deficiency. There is no evidence of a potentially unwoke utterance by Ginsberg 70 years ago. Deiseach is asking about a hypothetical if such evidence were uncovered.

          • Dan L says:

            Even if it were a hypothetical, it’s egregious enough that I’m not sure it’d be uncharitable to dismiss it as JAQing off. But no, hypotheticals don’t end in accusations of hypocrisy.

        • slightlylesshairyape says:

          Do you actually believe that saying something beyond the pale is the moral equivalent of attempted rape?

      • tscharf says:

        Being a rapist doesn’t preclude you from being the best arbiter of constitutional law on the court, ha ha. I get that we want morally upright people which implies they might be better at that job. Secondly, there seems to be a lot of dot connecting from teenage groping to sexual assault to attempted rape to now rape.

        If any partisan who lived in the same general area as a nominee decades ago can make a disqualifying accusation with no supporting evidence then I suggest we will have a lot of open seats on the court in the next few decades. Call me cynical but I very much doubt only really victims will be stepping forward with those incentives in place.

    • Aapje says:

      @Deiseach

      Swetnick actually revised her claims more than that. She now actually claims to merely have seen Kavanaugh standing in a group at a party, not a line. Nearly all of her claims seem to be conjecture and/or things she heard from others, not things she observed herself.

  5. Radu Floricica says:

    Since it looks a lot like motivated cognition, I’d be curious to see all the graphs separated by tribe.

  6. TooDamnSexy says:

    In a perfect world I would have liked to see a survey where we could have expressed a probability range rather than point estimates, or failing that express opinion on hypotheses of varying strength. I think there might be more interrogation to do about the shape of the disagreement. Consider the following suite:

    A) What is your probability estimate that Dr Ford’s testimony is accurate in all or most material respects?

    B) What is your estimate that Dr Ford is sincerely attempting to relate the facts as she recalls them, regardless of accuracy?

    C) What is your estimate that Judge Kavanaugh truly has no memory of committing a sexual assault?

    D) What is your estimate that Judge Kavanaugh was telling the whole truth about his general behavior as a young man?

    E) What is your estimate that one or both parties have materially interfered with the process of gathering and publicizing information about this case?

    All of the above can be answered in line with a partisan stance, but at least in my case (generally left of center) the answers would be A) 30-50% B) 80-90% C) 60-70% D) 30-50% E) 50%. This reflects where my uncertainty lives beyond a simple “what credence should we lend to sexual misconduct allegations absent other evidence”.

    If we saw more questioning along these lines I wonder if we couldn’t get a better sense of the lines that divide the two major party factions in your data.

    • J Mann says:

      Good questions, but IMHO “telling the whole truth about his general behavior as a young man” is insufficiently specific.

      At a certain point, the map becomes the territory, so to literally tell the whole truth, he would have needed hours.

      More directly, I think most people would gravitate towards either

      1) “He answered questions accurately, but didn’t volunteer extra information except where it would help his case, which is pretty much what every lawyer everywhere tells their witness to do and completely reasonably, and was in line with Dr. Ford’s similar approach to testimony” and

      2) He made a bunch of statements which while largely individually accurate, are misleading when viewed in total, and I (the speaker, not me) personally believe that some are false, for example whether he specifically knew that Manuel Miranda was accessing Democratic judiciary materials in the ‘aughts.

      I’d personally say #1 is true, which would mean that I think that (a) Kavanaugh didn’t share the whole truth but (b) no one ever does outside, possibly, of a really honest confessional memoir, and (c) his behavior was both reasonable and typical.

    • grendelkhan says:

      Have you seen this Bayes’ rule calculator on the topic? Differing Red and Blue priors would explain a lot of the variation. I wonder if it would be possible to resolve some of the differences empirically, or if these questions are so tribal that facts don’t really matter.

      (For example, ‘Prior probability of assault’ for men is pretty empirical, and it’s definitely well above 0.01.)

      • Garrett says:

        The problem is that we’re running headlong into the ecological fallacy.
        s/men/blacks/
        and see how that reads.
        And it doesn’t much help us answer the question of: what did this specific person do?

        • grendelkhan says:

          Please look at the actual calculator; this isn’t just about taking some priors and reading them as conclusions; it’s about trying to make the best inferences from what we know, including our priors.

          s/men/blacks/ and see how that reads.

          The number’s roughly half as large; as I recall, self-reported propensity to rape doesn’t differ markedly by race, though if you include women, you’d be cutting the numbers in half. It would be kinda weird to split it out that way, I guess. But I’m more interested in getting the most accurate answer possible.

          And it doesn’t much help us answer the question of: what did this specific person do?

          Sure it does! For example, it’s incredibly unlikely that Brett Kavanaugh is a serial killer, or a terrorist, because the prior probability of a man being either of those things is really, really low. And it’s still pretty darned unlikely, knowing anything else, that any unspecified man is a rapist.

          The outcome would be different if rapists were incredibly rare, or if they were the majority of men. Base rates are an important input into the reasoning here. You don’t do better by pretending you don’t have priors; you try and make them more precise, and reduce your uncertainty as you integrate more evidence. If you don’t, you’re throwing valuable bits on the floor.

      • ajfirecracker says:

        Set it to “Blue Assumptions” and put the prior that he’s a rapist (pre-allegations) at something sane, like 0.002 (1 in 500, still 8x higher than FBI stats 1 in 4,000) or 0.01 (1 in 100, which I think is an extreme upper bound based on victimization surveys).

        It produces a result of Kavanaugh almost certainly being innocent. In short, the Blue/Democrat assumption was based on rape hysteria, not sound reasoning.

        • grendelkhan says:

          The base probability that a man is a rapist is (very) roughly one in ten. See Lisak and Miller (2002) (which found 6.4%) or McWhorter (2009) (which found 13%). There are some significant error bars, but it’s pretty clear that the number isn’t “1 in 500”.

          If you’re wondering why the FBI (I think you mean UCR) stats don’t match up; this is what you get when you ask men if they’ve forced women to have sex with them even when they didn’t want to, which is consilient with what you get if you ask women if men have forced them to have sex even when they didn’t want to. The UCR measures crimes reported to the police; instruments like the NCVS report responses to ‘has someone raped you’, which is different from responses to ‘has someone [definition of rape]d you’. The results have been repeated around five times from the ask-the-victim point of view and twice from the ask-the-perpetrator perspective. We’re pretty certain about it.

          • J Mann says:

            I’m not a good stats guy, but I’ll note that of those percentages, the majority seem to involve violation while the victim is incapacitated as a result of alcohol or other drugs, not violence.

          • baconbits9 says:

            Lisak and Miller found that the base rate for men who use physical violence to rape is 1/3rd to 1/5th of that rate. These are the only rapes and attempted rapes that clearly satisfy the definition of rape through this methodology. The other claimed rapes require the victim’s appraisal of what happened, which isn’t included.

          • grendelkhan says:

            Yes, the primary method rapists use is to attack people who are incapacitated, rather than to rely entirely on physical force. If you want to describe that as not being violent, that’s your prerogative, but I disagree there.

            (The questions Lisak and Miller used included: “Have you ever had sexual intercourse with someone, even though they did no want to, because they were too intoxicated (on alcohol or drugs) to resist your sexual advances (e.g., removing their clothes)?”.)

            baconbits9, note that the perpetrators said that they had sex with people who didn’t want to have sex with them, but who were too impaired to stop them. That’s still rape–the thing where you have sex with someone even when they don’t want you to, whether you overcome their resistance because you’re stronger or because they’re somehow impaired.

            (I seem to remember that the psychological effects are also comparable whether the method is impairment or physical force, but I don’t have the cite handy. I’m reasonably sure that the legal consequences are similar, just much harder to prove, which is why rapists tend to rely on the method that’s safer (for them).)

            Even if you believe that it’s not really rape unless pure physical force is involved, the numbers are still well above one in a hundred. And more to the point, this is the method that Kavanaugh was accused of using, or attempting to use (alcohol was definitely involved), so about one in ten is the correct base rate to pick, so far as I can tell.

          • Ketil says:

            I’m surprised that the numbers are this high, after all this is self-reported, so the real number is unlikely to be lower. Here’s a quote from Lisk and Miller:

            Of the 1,882 men in the total sample, 120 (6.4%) met criteria for rape or attempted rape. A majority of these men, 80.8%, reported committing rapes of women who were incapacitated because of drugs or alcohol; 17.5% reported using threats or overt force in attempted rapes; 9.2% reported using threats or overt force to coerce sexual intercourse; and 10% reported using threats or overt force to coerce oral sex.

            Most of the rapists (63%) are repeat offenders, and also have violent or abusive behaviors in general (58%). Slightly surprising, perhaps:

            [T]he mean number of total violent acts committed by “overt-force” rapists (14.5, SD = 22.4) was greater than that committed by “intoxication-rapists” (8.4, SD = 22.4), but the difference
            was not statistically significant (t(118) = -1.35).

          • J Mann says:

            @grendelkhan – sorry, I absolutely didn’t mean to suggest that the behavior isn’t outrageous or that the stat isn’t important, only that we need to decide whether our background rate for the Bayesian analysis is “percentage of men who attempt to violate women through physical force” or “percentage of men who attempt to violate women through any means.”

            As I said, I’m not good at Bayesian calculations, so I honestly don’t know which is the more appropriate starting number.

          • baconbits9 says:

            @ grendlekhan

            The only direct quote in the paper given from the question about drug and alcohol induced sex is

            “too intoxicated (on alcohol or drugs) to resist your sexual advances (e.g.,
            removing their clothes).”

            This is a broad statement that could cover a ton of non rape situations.

            1. My partner and I drop X together and then have sex later.
            2. My partner and I go out and get smashed on new years eve, come home and have sex.
            3. A person who wants to get over their inhibitions about sex goes out and gets drunk with the intention of hooking up with someone randomly.

            These situations are not rape unless, at the very least, the woman states that they are rape. They cannot be ascertained by asking a single question about if you have ever had sex with a woman on drugs or alcohol.

          • grendelkhan says:

            baconbits9: This is a broad statement that could cover a ton of non rape situations.

            That’s a common critique, but I don’t think it holds up. I quoted one of Lisak and Miller’s drug-and-alcohol related questions above: “Have you ever had sexual intercourse with someone, even though they did not want to, because they were too intoxicated (on alcohol or drugs) to resist your sexual advances (e.g., removing their clothes)?” (p.77, emphasis mine). This is the mirror image of the question asked of women by, for example, Schwarz and Leggett (1999): “Have you engaged in sexual intercourse when you didn’t want to but were so intoxicated or under the influence of alcohol or drugs that you could not stop it or object?” (p. 269)

            The answer in turn, was similar to what you get when you ask: “Have you had sexual intercourse when you didn’t want to because a man gave you alcohol or drugs?”, as Koss et al. (1987) did (p. 167). (It was broadly supposed that this question was producing a lot of false positives, but that seems not to be the case.)

            You wouldn’t say, in any of those situations you listed, that you had sex with someone who didn’t want to have sex with you, but they were too intoxicated to resist. I’m sure there are people out there who would find fault with those situations, but that’s not what I’m talking about here, and that’s not what the numbers reflect.

          • And more to the point, this is the method that Kavanaugh was accused of using, or attempting to use (alcohol was definitely involved)

            I thought Ford claimed she had only had one drink, in which case she would not have been incapacitated. It was the attackers who were said to be drunk.

        • mdet says:

          grendelkhan is doing a great job citing the data, but from personal experience I have no idea how you could call 1 in 500 a “sane” estimate. I remember in highschool I *knew* a few guys who would brag to each other about pinning a girl down and not letting her go. I *knew* guys who would say things along the lines of “if she says no just get more drinks in her”. For 1 in 500 to be reasonable, I would have to know thousands of guys. (I did not, and the guys I did know well were disproportionately the nerds, not the partiers.) On the other hand, I didn’t witness these assaults, so some of them could easily be teenage boys lying about how much they get laid. Still, I wanna know what your personal experience is like that you think 1 in 500 is a plausible estimate.

  7. Deiseach says:

    I feel I should thank Scott for making a good faith effort like this. There’s been an awful lot of partisan commentary on both sides about this whole affair, and I’m rather pissed off right now because of a piece of commentary that was linked over on the sub-reddit (I don’t want to go into more detail than that).

    By contrast, Scott is trying to be fair and turn the heat down. This makes me appreciate SSC more than ever.

    • What I thought most interesting about the results was that there was a very strong division by political party, only a weak division by gender. But that may reflect characteristics of this population that would not hold for a more random selection of people.

      • HeelBearCub says:

        At this point, people still willing to call themselves Republican (and those that no longer will) is probably doing that sorting for you. I’d want to know how skewed this sample was on gender before I drew any conclusion on raw percentages.

      • NoRandomWalk says:

        @DavidFriedman
        it depends on who you identify with.
        A woman would not necessarily identify with Ford – she might identify with Ashley Kavanaugh.
        What public polling there has been seems to show that the gender gap exists for young single people, but not for married individuals.

      • Nicholas Weininger says:

        In fact AIUI this is the pattern that holds for the general population as well (e.g. R women are no less likely than R men to believe Kavanaugh, D men no less likely than D women to believe Ford), so it slightly reduces my prior estimate of how different SSC readership is from the general population.

  8. Jon Gunnarsson says:

    I have no idea whether Kavanaugh is guilty. If you put a gun to my head and forced me to give a probability estimate, I’d say 40%, but that’s really just a made up number. As long as we don’t have any remotely strong evidence one way or another, it really doesn’t matter to me whether the number is 20% or 80%. Either way, Kavanaugh should be confirmed because doing anything else would create perverse incentives.

    If Kavanaugh is rejected based on an unproven and unprovable accusation with no evidence behind it, then it’ll be open season on any high profile political appointee. All you need to do is find one partisan willing to perjure themselves in exchange for fame and notoriety to make an accusation that cannot be disproved. This is why the presumption of innocence is so important and we abandon it at our peril.

    • zzzzort says:

      I understand the argument that this might give bad incentives, but it seems somewhat presumptuous to say that it would actually happen. If kavanaugh had been pulled, and the next person had a previously unknown assault allegation pop up, then that seems like better evidence that a high standard for nominees is unworkable. In this scenario there’s still plenty of time for the GOP to say, ‘see, this is a witch hunt, we must confirm hardiman or no republican is safe.’

      • NoRandomWalk says:

        @zzzzort my reading of the situation is that it already did happen.

        Senator Hirono saying “I put his denial in the context of everything that I know about him in terms of how he approaches his cases”,

        The media-partisan apparatus acting seemingly in consort, showing a general willingness to treat the 2nd and 3rd allegations with slightly less but still way more credibility than they deserve,

        Knowingly ignoring the laws of conservation of evidence by saying evidence in favor of Kavanaugh is really evidence against, when in the alternate world where the facts went the other way they totally would not treat it as evidence in his favor.

        The extent to which the center-right was radicalized over this issue cannot be understated. They certainly don’t think they need additional evidence to conclude that if they tried to nominate someone else a similar thing would occur.

        • zzzzort says:

          I don’t understand all of what you’re saying, I disagree with some of it, but all of it seems to only apply to kavanaugh. The argument about incentives only applies to future nominees. Whether or not kavanaugh had been confirmed, I don’t see this as a strategy (to the extent that it was a deliberate strategy) that could have worked more than once, for better or for worse. If the next nominee has a he said-she said accusation, it will be viewed more suspiciously. If the accusers are cynical partisans or paid operatives, then after N iterations the likelihood of discovery goes to unity pretty quickly. All of which is to say, I think the sweeping claims of moral hazard are a bit overblown.

          • J Mann says:

            It’s probably a little bit of exaggeration for effect, but if removing Kavanaugh (for example) triples the number of unfalsifiable assertions against public figures and triples the number of firings as a result, that’s still an effect.

          • NoRandomWalk says:

            @zzzzort if you specify what parts you disagree with/I failed to state in a way you understood my meaning, I am more likely to respond with something that makes sense to you. Intellectual turing tests are usually failed by both parties.

            My understanding of the moderate republican view (informed mostly by National Review articles and never-trump twitter feeds), is that the way the three most-prominent allegations were treated by the democrats and the media is evidence that:
            1. How credible a claim is on the merits has very little relation to how strongly it is used to demand someone “believes all women” and “do what I happen to want politically right now”.
            2. If the republicans withdrew Kavanaugh, progressively more tenuous allegations will create more political circuses down the line.

            Confirming Kavanaugh is basically viewed as ‘standing up to the bully’.

            I think your position is “at least one allegation was credible, so if in the future none of the allegations are credible the Republicans aren’t signaling weakness by withdrawing Kavanaugh”

            But this is only logical if Republicans, Democrats, both think the other agrees publicly and privately in good faith about which allegations are credible and which are not.
            No such common knowledge exists. Certainly not in public or body politic, maybe not in private. Coordination is really hard.

            It seems the consensus on the right is that the response by the Republicans Now determines whether or not the tactics of bad-faith character assassination are viewed by their opponents as an effective method in the Future.

          • zzzzort says:

            It seems the consensus on the right is that the response by the Republicans Now determines whether or not the tactics of bad-faith character assassination are viewed by their opponents as an effective method in the Future.

            I guess that is what I’m disagreeing with. There are lots of cases of perverse incentives that for a variety of reasons just don’t matter that much in the real world, e.g. life insurance.

            Let’s make a non-partisan analogy. Say I meet a friend at the pub, and he says he’s forgotten his wallet, and asks if I can pay. Now, if I pay for him that creates an incentive for him to leave his wallet at home in the future (i’m on record as never accepting repayment for drinks). But in practice I don’t expect this to be an issue. Certainly the first time it happened I would take his word that it was a mistake, knowing that if he tried it again I could point to a pattern as justification.

            Personally I see that consensus among the right on how slippery this slope is as the result of a few factors:
            -Desire to shift the terms of the debate from uncomfortable specifics to nice abstract principles.
            -Desire to accuse the accusation of being in bad faith without accusing blasey ford herself of acting in bad faith.
            -Universal tendencies of tribes to view opposing tribes as being ruthless and efficient as compared to one’s own tribe.

          • SamChevre says:

            From a right-wing perspective, a very similar thing already happened, and was successful, with Roy Moore.

            In both cases, the unproved allegation was supported by the target’s participation in behavior that is legal, but more widely accepted by the right than the left. (Age-disparate marriage in Moore’s case, fraternity membership in Kavanaugh’s.)

          • Jaskologist says:

            Proposed: if Kavanaugh and Moore had happened in the opposite order, Moore would have won his election. Republicans are now going to be extremely suspicious of sexual allegations against their candidates, especially if they come out at a politically convenient time. The presumption will be that it’s just a smear job.

            (Also, when did the Right become the group in favor of frat parties and the Left opposed? I’m not saying that didn’t happen, but I think it’s very very recent, as in, only in the past few months.)

          • zzzzort says:

            I can see the moore example affecting the psychology of the current GOP, and really see the argument that if the order was reversed moore might have won. I don’t buy that moore counts as a victim of unfounded allegations, though. In support, see mitch mcconnell: “It’s outrageous to compare Brett Kavanaugh to Roy Moore”

          • I think this is at least the third round. The first was Hill/Thomas. It failed that time. On the one hand, the accusation was of less serious behavior than attempted rape, but on the other hand it was made by someone much closer to the person accused and involved much more recent behavior.

            And the general pattern of alleging improper sexual behavior against a high profile candidate goes back somewhat farther than that.

          • tscharf says:

            The 2nd and 3rd accusations against Kavanaugh can arguably be seen as these perverse incentives bearing their fruit. The media shouldn’t have touched those with a ten foot pole.

          • rlms says:

            Wow, I agree with Mitch McConnell for about the only time ever! No, the issue with Moore was not that “age-disparate marriage” is disapproved of on the left, it’s that locking a 16 year old girl in your car, groping her, and forcing her head towards your crotch is generally viewed in a negative light. I like to think that that view cuts across the political aisle, but maybe not.

            Is there any reason Bill Clinton hasn’t been brought up?

          • SamChevre says:

            rlms

            It seems either I was unclear, or I was mistaken on the facts.

            My understanding of the Moore accusations was that the “groping a 16-year-old” was the accusation, but was denied by Moore and not confirmed otherwise. The claim everyone agreed to, including Moore, was that he’d dated 16-year-olds. Different groups differed on whether the second claim substantiated the first.

            On “when did the right become the pro-frat side”, I think it happened when academia and feminism became two major key driver on the left; both groups despise fraternities (and sororities). The right is a coalition (as is the left), and some parts have long been fairly libertine (the “economic issues” Republicans and some of the libertarian-leaning ones as well).

          • If I correctly remember the Moore case, the two strongest claims were that he had tried to seduce someone younger than sixteen (I think fourteen) and that he had offered to give a young woman, I think sixteen, a ride home and instead aggressively groped her and, when he gave up on going further, told her that because he was high status nobody would believe her if she told them what he had done.

            I’m not sure what the evidence was on either of those.

          • rlms says:

            @SamChevre
            The accusations of groping against Moore were not confirmed in the sense of legally proven, but I think they would widely be considered to be both a lot more probable than those against Kavanaugh and a lot more damning if true. As we have seen from the poll results, a reasonable person could live with a guilty Kavanaugh in office; I don’t think the same applies to Moore.

          • SamChevre says:

            rlms

            I think we agree on the facts of the Moore case, and on its relative strength relative to the Kavanaugh case.

        • NoRandomWalk says:

          @zzzzort
          I agree with a lot of what you said, especially “republicans claim to believe her to be mistaken not lying” being explained by that being the only politically viable defense. I think there’s a huge number who think “if true is still doesn’t matter” who don’t say that’s their real defense. Your other beliefs are also completely true. I think majority of both sides were arguing in bad faith.

          My points are to explain the sliver that is ‘people who consider themselves center-right, or consider themselves republicans but vehemently dislike everything trump represents’. i.e. reader base of national review, never trumpers, etc.

          To your non-partisan example, I think the crucial distinction is capacity for coordination. ‘Your friend’ is a stable entity. Who the democrats are keeps changing over time (elections vote in new people, the electoral base dies off or changes its opinion), and crucially there’s a bunch of people holding different positions in the same tribe and they don’t all agree so coordination is hard.

          The only way to respond to such a messy situation is introducing a schelling fence.
          A distinct, clearly definable threshold you can rally the troops around without too much internal arguing.

          The center-right has identified that point as ‘allegations must have non-zero contemporaneous corroborative evidence to be acted upon’.

          • zzzzort says:

            The concept of a schelling point only makes sense if there’s a slippery slope, if each time a strategy works makes it more likely to work in the future. In cases where there is a negative feedback (each time my friend ‘forgets’ his wallet makes me less likely to spot him money in the future’), there’s no need for a schelling point.

            If we want to model this with some abstract game theory, i would suggest a noisy iterated prisoner’s dilemma. The GOP wants to punish defection, but there’s no way for the GOP to be sure if the democrats are acting in good faith. Optimal strategy (people think) is a generous tit for tat. The GOP seems to be going for an ungenerous tit for tat, where even if this instance isn’t bad faith, they’ll still punish it in case the next one is. Though it would be interesting to see what people’s views are about probability of the ‘good faithness’ of these accusations.

          • I agree with a lot of what you said, especially “republicans claim to believe her to be mistaken not lying” being explained by that being the only politically viable defense. I think there’s a huge number who think “if true is still doesn’t matter” who don’t say that’s their real defense.

            I suspect there are also many who believe she is deliberately lying but that saying so is less politic than saying she is mistaken.

            Of the three alternatives—deliberate lie, mistake, truth—I find mistake the least likely.

          • NoRandomWalk says:

            @zzzzort
            “The concept of a schelling point only makes sense if there’s a slippery slope”

            It can also make sense as a coordination mechanism threshold.

            Because there are tons of different actors on both sides, there’s no way to send a signal to some actors ‘we treat allegations of this strength seriously, but credibly resolve to not put up with you weaponizing less serious allegations’ without sending a signal to other actors ‘we are capable of being bullied by less serious allegations’.

            Therefore, you need a schelling point: something that is a demarcation of ‘seriousness of allegation’ that allows you to send the same signal to everyone on ‘the other side’.

            In this context, I see moderate republicans as coordinating around ‘allegations need contemporary corroboration’ as a defensible schelling point under the belief that this will prevent partisan democrats from trying to weaponize allegations without contemporary corroboration (which are easy to make up and tend to be not credible) in the future.

            Key Note: I am not saying all the non-contemporaneously-corroborated allegations in this case were easy to make up. The crucial point is that everyone can agree on what contemporaneous corroboration is, whereas ‘easy to make up’ and ‘legitimate to vilify for considering if made up’ are not agreed upon.

          • tscharf says:

            There is a further argument left untouched that it did happen, but Ford is overstating what really happened and conflating 1980’s awkward groping with 2018 sexual assault / attempted rape. If there was proof Kavanaugh was in the room with her then it is likely the argument would be over what really happened and intent.

            One thing I don’t like is retroactively applying the 2018 moral lens to 1980 behavior. I don’t have a problem with society redefining its rules of sexual conduct, just don’t do so retroactively. This is not a justification for legitimate rape. What was called a sexual assault in 1980 and what is called sexual assault in 2018 are different.

          • zzzzort says:

            @NoRandomWalk

            It can also make sense as a coordination mechanism threshold.

            That’s a good point, which I hadn’t really considered. I still don’t think it’s the point that most people on the right are making. There is a lot of thought going into setting up a system that is robust against political operatives cynically making false accusations, without much evidence that that is likely to be a problem.

            The question of where to draw the line of sufficient evidence should just reduce to the question of probability of guilt. Contemporary corroboration seems pretty arbitrary, we should all just agree that we require evidence sufficient to establish x% probability of guilt. A natural point might be 50%, which the SSC community seems to get behind. I would probably argue we should be more stringent that ‘less likely than not to be a perjuring sex criminal,’ but at least i’m comfortable identifying that as a value difference.

          • NoRandomWalk says:

            @zzzzort
            “Contemporary corroboration seems pretty arbitrary, we should all just agree that we require evidence sufficient to establish x% probability of guilt.”

            This. This is the crux of our disagreement.

            It is extremely difficult to agree with your partisan opponent on a politically charged case whether the likelihood of guilt is above/below X%.
            The entire narrative on the right is that the mainstream media is biased in subtle and not-so-subtle ways. Whether or not you accept this premise informs your priors about base rates of evidence so as to explain a huge percentage of the variance in ‘is he more likely to be guilty than not’, before we even get to all the incentives to not report your true assessment of guilt!

            Contemporary corroboration is arbitrary, sure. But it has two major advantages:
            1. It’s objective. We can agree on what counts as contemporary evidence even if we disagree on how strong it is.
            2. It’s difficult to fake after the fact, solving for the disagreements about ‘base rates’

            When discussing slippery slopes and moral hazzards and what social coordination mechanisms can do about them, it is objectivity that is going to be your most-powerful hammer.

            To your point about ‘what is the actual position of people on the right’ I’m not confident in my view. Its plausibility is informed in large part by what I observe as a genuine collective feeling of being continually gaslit, resulting in a demand for an objective schelling point that a bad-faith politically powerful actor has fewer degrees of freedom to abuse.

          • zzzzort says:

            @NoRandomWalk
            Glad that we got somewhere, though it still seems somewhat far from where we started. Your argument, as I understand it, is that political institutions suck at determining facts with political implications, so we should take away that responsibility by having bright line rules.
            -I haven’t seen many people are making this argument, and i’m pretty sure the original poster wasn’t. If this is going to be the right’s position, they’re going to need to communicate it a lot more clearly, or the whole point of having a bright line is lost.
            -I’m not sure I like this outcome on the merits. Say everyone is 80-90% sure that someone is guilty, but there’s no contemporary corroboration. You could imagine multiple accusers, a bunch of circumstantial evidence, maybe the accused confessed to a friend at some point before recanting. I don’t find it a good outcome to elevate that person just for the sake of the norm.
            -Formulating a rule after an important issue seems normal, formulating a rule for an important issue seems fishy, formulating a rule and saying that adherence to that rule is more important than the merits of the important issue is mighty suspicious.

          • zzzzort says:

            And just to argue the alternative position, I’d be a lot happier if partisans were up front when they have a different view of reality, rather than couching that different view of reality in some sort of process question. If 51 republican senators stood up and said that they were 95-100% sure kavanaugh didn’t do it, i would disagree with the assessment, but not with their decision to vote for him. Having a universal standard to circumvent those judgements seems to make actual communication harder, even if we agree on superficial things.

          • NoRandomWalk says:

            @zzzzort
            I’m glad we’re getting somewhere. I don’t agree we’ve moved beyond the original post.

            To re-quote:
            “Kavanaugh should be confirmed because doing anything else would create perverse incentives…[or] it’ll be open season on any high profile political appointee. All you need to do is find one partisan willing to perjure themselves.”

            I’m not saying politics implies the mess we’re in necessarily. There were time in american history where stakes were high but there was bipartisan mutual good faith.

            I’m saying that, right now, there is no such good faith.

            What solves this? Any criterion of evidence that ‘verifiably cannot be falsified after the fact without getting at least two related persons to agree to lie’.

            “Formulating a rule and saying that adherence to that rule is more important than the merits of the important issue is mighty suspicious”.

            In context, what is ‘suspicious’ from the perspective of republicans is that the democrats/mainstream media engages in what they perceive as a partisan witch hunt.

            They respond to this by demanding objective standards because they no longer believe that their opponents will agree to any ‘objective factfinding process in good faith.’

          • NoRandomWalk says:

            Also, there is a remedy to such a crude solution. Namely if the democrats convinced the republicans they were acting in good faith.

            Just a few examples of what might’ve done this:
            1. If Feinstein investigated the allegation confidentially without sitting on it for 2 months while helping Ford get partisan lawyers
            2. If the allegations against Keith Ellison were taken seriously
            3. If Sen Hirono didn’t say Kavanaugh’s judicial philosophy informs her belief that Ford is telling the truth
            4. If Sen Booker, Sen Harris didn’t quote him out of context or call him evil.
            5. If the democrats actively disavowed the allegations of Julie Swetnick when it became obvious she was lying
            6. If the mainstream media spent at least 5% of the effort they exerted trying to prove he was lying to investigate whether she would have reason to. Almost every single piece of verified evidence in his favor was discovered by partisan right-leaning source despite most of the qualified investigative resources being at places like CNN, New York Times, etc.

          • zzzzort says:

            I think that’s conflating a lot of things under the heading of bad faith. You mention a lot of things that democrats and the media did, but do you think blasey ford is acting in bad faith? There’s a big difference between senators grandstanding and taking quotes out of context, and someone perjuring themselves to accuse someone of attempted rape. Do you think not confirming kavanaugh would have led to a significant number of people perjuring themselves to make such accusations?

            If using the existence of a credible allegation for partisan advantage is considered a defection, then I disagree with enforcing this norm through rules on the quality of the allegation. Rules better suited to prevent this in the future might be a duty to report for senators on the committee, an end to senators directly asking questions at confirmation hearings, or a move to more autonomous non-partisan investigations. But to a large extent making political hay out of things is the reason politicians exist, and expecting them to do otherwise is foolish.

            If the creation of the allegation is the defection, then the right needs to argue that blasey ford made up the allegation, which it is conspicuously not doing.

            Instead, the right seems to be saying that blasey ford was earnest (if mistaken) but that if kavanaugh withdrew the next time the accuser would have been a soros funded acorn operative. This is the step I find hard to swallow.

          • HeelBearCub says:

            @zzzzort:

            which it is conspicuously not doing.

            I’m not sure it’s all that conspicuous. There is a great of waffling back and forth and not being very clear about what they mean precisely because they want to imply she is lying, while also claiming to say they believe her to have been assaulted.

          • tscharf says:

            There isn’t much danger of perjury when you are the sole witness of your alleged story. People aren’t accusing Ford of lying for political reasons, she absolutely, positively, without a doubt could be lying. Proving she is lying is about as difficult as proving she is telling the truth in this case. It is likely we will never know.

          • Joyously says:

            I agree with a lot of what you said, especially “republicans claim to believe her to be mistaken not lying” being explained by that being the only politically viable defense. I think there’s a huge number who think “if true is still doesn’t matter” who don’t say that’s their real defense. Your other beliefs are also completely true. I think majority of both sides were arguing in bad faith.

            Some people are treating the “mistaken” theory as transparently ridiculous when I really don’t think it is. There are two basic forms of “mistaken” theory:

            1 – Something happened but it wasn’t him
            2 – Something happened with him but it wasn’t as bad as this.

            People do misidentify their attackers, memories do alter over time, people have false and mistaken memories of things which they’re really certain about and which you’d really think they wouldn’t, and genuinely traumatic memories have been found to grow more traumatic over time (e.g. soldiers with PTSD start to remember they were under sniper fire when that wasn’t part of their memory before, and they even incorporate other attacks they’ve seen on the news or in movies into their own memories.)

      • Murphy says:

        I understand they were somewhat on the clock. Had kavanaugh been dropped then that would have a strong chance of becoming a de-facto victory for the dems because putting a new candidate in and confirming would take time if they could delay things until after the midterms they’d have good odds of being able to block any future nominee at which point they can just wait, pointing to the Reps doing the same during obama’s term.

        • zzzzort says:

          -I don’t really understand how the clock was that close. They might have needed to confirm someone in the lame duck, which might have upset dems and been somewhat unpopular, but pushing through kavanaugh also upset dems and was somewhat unpopular.

          -That doesn’t speak to the issue of incentives, which only pertains to future nominees. You can argue that the GOP had to confirm kavanaugh to get the supreme court seat (though I disagree with the assessment), but that doesn’t mean they had to confirm kavanaugh to be able to confirm anyone to anything ever again (as some of the more hyperbolic rhetoric posited).

          • Murphy says:

            Given that he went through 51-49 and 35 of the 100 seats are up for grabs right now…. they really were very close to the time limit. Supreme court seats are extremely politically valuable.

            I don’t understand you “confirm anyone to anything ever again” comment.

          • zzzzort says:

            -New congress doesn’t get seated until January, which is plenty of time for a motivated senate.

            -original poster said

            it’ll be open season on any high profile political appointee

            The theory being that if one nomination is sunk by an unsubstantiated allegation, then in the future all nominees will face unsubstantiated allegations from politically motivated people, and that because the first person was denied confirmation, all future nominees will also be denied. This is usually framed as something that only the GOP is vulnerable to. I think this fear is overblown.

          • Confirmation is going to be somewhat harder with a lame duck senate because Democrats running in deep red states will either be reelected or out, so have no incentive to vote for confirmation—as one in fact did.

          • zzzzort says:

            There’s good reason to believe that manchin would have voted differently if his vote had mattered, and better reason to believe that murkowski could have been swayed on a nominee with less baggage.

          • PeterDonis says:

            There’s good reason to believe that manchin would have voted differently if his vote had mattered

            Huh? The vote was 50 to 48. If manchin had voted no, it would have been 49 to 49 and Kavanaugh would not have been confirmed.

          • Evan Þ says:

            If the vote was 49-49, Pence would’ve broken the tie by voting to confirm.

          • PeterDonis says:

            If the vote was 49-49, Pence would’ve broken the tie by voting to confirm.

            Ah, that’s right.

          • Thomas Jørgensen says:

            It really, really was not. That is actually why I think all the arguments about bad faith and “liberal plots” are ridiculous, and honestly, make the people making them look very, very bad.

            The federalists had a list of 25. Picking someone out of that hat with cleaner personal finances, less entanglement with the Bush era and, oh, yes, no sexual assault allegations would have been easy.

            That is obvious to the democrats from the word go. Heck, it is obvious to Ford. Stopping Kavanaugh does not get you a liberal justice, it just gets you a different conservative one. That is not much of an incentive to lie.
            (And no, accuse the next guy too is not a viable plan! That would predictably blow up in your face. For one thing, that list has women on it. Please grant your opponent the dignity of ascribing at least minimal intelligence to them)
            It is a strong incentive to come forward if you have been personally victimized by the present candidate and want to be able to look yourself in the mirror, but the partisan motivation just does not exist.

            If you want to feel paranoid, here is a more plausible democratic plot.

            The democrats have better dirt on Kavanaugh than this. They brought forward all the stuff they felt confident that the republicans would ignore out of sheer bloody partisanship. Come election season, miraculously, they are going to find out where the money for baseball tickets went, and then the republicans are the party that put a criminal on the supreme court after being informed in exquisite detail that he was a dirtbag.

          • Edward Scizorhands says:

            Picking someone out of that hat with cleaner personal finances, less entanglement with the Bush era and, oh, yes, no sexual assault allegations would have been easy.

            They didn’t know about the sexual assault allegation, because those with the knowledge of it didn’t say anything about it.

            Stopping Kavanaugh does not get you a liberal justice, it just gets you a different conservative one. That is not much of an incentive to lie.

            Feinstein sat on the accusation until the 11th hour. Clearly, the motive was to try to stall Kavanaugh at least past the midterms. Delay is a very viable and easy-to-understand political tactic here, not least of all because SCOTUS is now hearing cases, and keeping Kavanaugh off the court keeps it at 8 people instead of 9.

            Also, there was about a 50/50 chance that the GOP would have done something really stupid, like call Ford a fakey-fakey liar bitch right out of the gate. I was pleasantly surprised, a little bit, and I think the Democrats in Washington were, too, that the Republicans on the SJC said “okay, sure, let’s get Ford down here to hear what she has to say.”

            (I’ll say, yet again, even though I shouldn’t have to, but bad actors mean I have to, that I give a low probability to Ford simply making all this up.)

          • Thomas Jørgensen says:

            The character reference letter that got produced? That came out way too fast. The republicans knew it was coming. Presumably because every office in Washington leaks like a sieve, though there are more damming reads on it.

            But more importantly, ask yourself. If Kavanaugh had dropped dead at the hearing from an outraged induced heart attack, do you *really* believe it would have taken the Republicans enough time to replace him that the lame duck would have expired?

            No?

            Then you have to admit that the Democrats probably also did not have much in the way of hopes of stopping the appointment of a conservative justice. Just this specific one. Thus, not much of an incentive.

          • Edward Scizorhands says:

            The character reference letter that got produced? That came out way too fast. The republicans knew it was coming.

            You could ask the people who put it together. If you read news sources even slightly out of your own bubble, you would know this.

            https://www.weeklystandard.com/virginia-hume/about-that-letter-from-women-in-support-of-brett-kavanaugh

            We have social media these days. You can get a bunch of people who knew each other in high school to all come together, especially if they are upper-class and working in politics or journalism.

            One of the signers involved was the Renate girl, and she asked to be removed once she learned what BK wrote, and even then she didn’t change her story to say that this letter had been planned ahead of time.

            But you be you.

            *EDIT*

            do you *really* believe it would have taken the Republicans enough time to replace him that the lame duck would have expired?

            1. The political calculus changes after the midterms and during the midterm, so maybe not.

            2. I still think there’s a better than 50% chance they could still get Barrett through, but who knows what new surprise bomb would be hidden until the 11th hour this time. And even at a 60% chance of them getting her through, that’s a 40% chance for the Democrats to hold out until the new Senate.

            3. The Supreme Court is hearing cases RIGHT NOW. Here is the schedule, you can see what they are hearing today, Tuesday the 8th. https://www.supremecourt.gov/oral_arguments/calendars/MonthlyArgumentCalOctober2018.html Keeping the new justice off of the court, even for one more day, is a political victory.

          • Thomas Jørgensen says:

            As I said, throwing dirt at the second candidate is not a viable plan. Because that makes the Democrats look like obvious bad faith actors to the median voter, not just to hardline republican partisans, and the R have the votes to confirm on strict party lines, and would.

            Especially since the obvious fall-back candidate is a woman that the sitting senate already approved of. That is not “60% chance of getting through”, that is as near to a sure thing as politics ever gets. The seat was never in play. Talking like it is, means you are not reading the situation even remotely right.

            I mean, I am for example pretty sure Feinstein was hoping the republicans would alienate women voters super hard over this, and that is why she pushed it as late as she could, to maximize the impact on the midterms, but that is just basic strategy once Ford dropped into her lap, and certainly not a reason to think it was a plot – Risk/reward all wrong, because to the extent that the republicans did alienate women voters here, it was an unforced error on their part. Nothing the democrats could actually do to stop the other side from just going “Eh, we do not owe K that many favors, Barrett it is”.

          • Edward Scizorhands says:

            That is the kind of political analysis you get from a character inside an Orson Scott Card novel, where the characters can predict exactly how things will play out and act completely rationally.

            In the real world, the Democrats are already set to pay a serious price at the polls because of the past few weeks, with the chance of them taking the Senate dropping to 20%. Why would anyone listen to predictions from them on how they wouldn’t do something that hurts them in the polls? Just because something turned out to be a bad idea doesn’t mean it wasn’t deliberate. People fuck up all the time. (Which is why I gave such a serious chance of the Republicans calling Ford a liar.)

            Believing your own press releases is the most common mistake in politics.

          • Careless says:

            It is a strong incentive to come forward if you have been personally victimized by the present candidate and want to be able to look yourself in the mirror, but the partisan motivation just does not exist.

            Given that there were other accusations after, I don’t believe you are correct. Of course the partisan motivation exists.

        • tscharf says:

          A proposed theoretical deal was to drop Kavanaugh in exchange for a promise from the left to confirm person X before the election. This would have isolated whether this was about Kavanaugh or the seat itself. It’s about both of course, but I think most people, especially on the right, believe this was just a tactic to prevent the seat from being filled, not about Kavanaugh.

          • Evan Þ says:

            I don’t think the Right would’ve trusted such a promise. (Nor, IMO, should they have.)

          • tscharf says:

            Probably not, which is something that needs fixed in the long term. Politics may be war, but it doesn’t have to be scorched earth 24/7.

          • cryptoshill says:

            I think this may have been Ford’s intention with giving the letter to Feinstein in the first place. Ford is not a known political actor, and may have expected a more reasonable outcome than actually occurred.

            These events had substantial personal consequences for Ford and Kavanaugh, and no real impact on the truth-finding of the case at hand or anyone’s opinion on Kavanaugh.

          • Thomas Jørgensen says:

            Which makes no sense. There is no way Kavanaugh derailing does not just get you a lame duck session confirmation of some other conservative. Seat was never in play in that sense.

          • J Mann says:

            1) You never know, sometimes the horse learns to sing. The next nominee might turn out to have evidence that would turn some of the GOP senators (say a string of virulently racist emails from last year, or a sex tape showing a rape last year).

            2) The lame duck would have had its own problems. With no pressure from their voters, the GOP would be unlikely to pick up any of the nominally moderate Red State dems like Manchin, Tester, Heitkamp, etc. Therefore, if even 2 GOP senators flipped, Trump loses the chance to name someone before the next Senate.

            3) Forcing Heitkamp, Tester, etc. to vote no may help the GOP pick up some of the Red state seats.

          • Thomas Jørgensen says:

            But that kind of (thin) hope is not something you mount a major political effort on.
            Best read I have on it is that the democrats went to war here because

            a: They had to. Ignoring a highly respectable accuser in the me-too era just was not on. Does not matter if this was ever a winnable fight for them, they had to be seen to be fighting, lest their base loose enthusiasm.

            b: Tactically, there was a really good chance the bunch of very old white dudes on the judiciary committee would screw the republicans with female voters for the next decade or two by being horrible people.

            Anita Hill didnt stop the appointment, but those hearings had one hell of a toll at the polls come next elections. Hoping for “Year of the woman, round two”? That is a far more solid hope.

            And they did, in fact, step in it a few times, though, well, not as horribly as they could have. Presumably, the RNC had people following them around whispering in their ear “REMEMBER: You are on camera, and the record never dies!”

          • J Mann says:

            Yeah, I agree with that. My rough guess of what happened is:

            1) As long as the accusation was confidential and Ford wasn’t requesting to come forward, Feinstein actually didn’t think this was important to go forward with.

            2) Eshoo’s staff, DiFi’s staff, or Ford’s legal team leaked to the other Dem senators that the letter existed, and some of their staff leaked its existence to the Intercept.

            3) Once a plausible but unverifiable accusation was out in the open, both sides’ base demanded that they fight.

          • One point possibly relevant to the whole story is that Feinstein is under pressure from her left in the election. Due to California’s primary system, both senatorial candidates are Democrats. So some of her acts may have been motivated by her political interest, not her party’s.

      • SamChevre says:

        moved

      • Jon Gunnarsson says:

        As others have pointed out in this thread, Kavanaugh is hardly the first high profile man who has been subjected to dubious accusations of sexual misconduct. Some of these accusations ruined the careers and lives of the people accused, while others merely tarnished their reputations. The more often such accusations are ruinous, the bigger the incentive to make false allegations.

        What you’re basically saying is let’s negotiate with the terrorists just this once. But then the next time terrorists kidnap a bunch of people, the same logic will tell you to make another exception and so on. So what makes the Kavanaugh situation special such that we’re supposed to disregard presumption of innocence and just believe accusers just this once?

    • beleester says:

      The evidence might be thin, but I want to push back on the idea it would be possible to engineer a similar accusation against literally anybody.

      Ford told her therapist about the attack in 2012, before Kavanaugh was on anyone’s radar. The idea that she was laying the groundwork for a false accusation just in case they needed to smear him six years later strains credibility.

      This isn’t ironclad evidence against Kavanaugh – there are plenty of ways Ford could be mistaken about her attacker’s identity or the details of the attack – but it’s strong evidence against the idea that the Democrats made up the story whole-cloth in 2018, or that it would be equally easy to make the same accusation against any nominee.

      • The Arcadian says:

        Kavanaugh was on the radar for the Supreme Court in 2012, according to The New Yorker, which explicitly predicted him as the most likely Supreme Court nominee if a Republican defeated Obama.

        https://www.newyorker.com/magazine/2012/03/26/holding-court

        • Edward Scizorhands says:

          That article came out in March 2012, and her visiting her therapist was in May 2012, so it’s exactly what you would see from her if your prior expectation was that she was seeing his name and deciding to start laying the ground work to stop him.

          • JPNunez says:

            But that works too if she was freaked out by the possibility that her attacker could end up in the supreme court and so she told her therapist.

            I mean, surely you are not expecting rape victims to be relaxed about the possibility their attackers ending up in the Supreme Court and thus not commenting it to their therapists?

          • Edward Scizorhands says:

            But that works too if she was freaked out by the possibility that her attacker could end up in the supreme court and so she told her therapist.

            Yeah. This is, like, the fourth time I’ve pointed out this fact about timelines here. And every other time I’ve pointed out that “obviously there is another explanation that makes sense, that she just remembered her legitimate attack from seeing his name in the news.” I wondered how necessarily that disclaimer was to say every single time. Now I know.

            I’m just blowing up the “this accusations pre-dated Kavanaugh’s name being in the press.” That March 2012 New Yorker article was written by a guy who currently does news analysis on CNN for the Kavanaugh hearings, and do you think he’s mentioned it?

      • Murphy says:

        I agree with you…. but I also like hypotheticals so I’m gonna run this one out.

        Lets try to assign a cash value to major political positions.
        How much is a supreme court justice seat worth?

        Both GOP and the Dems have yearly budgets close to the billion mark just for regular campaigning across the country. Presidential campaigns also close to the billion mark each.

        On average justices last for about 17 years though for partisan issues the deciding vote can be what matters, so it might be easier to think of the supreme court as 1 position you can control by tipping the balance by 1 seat one way or the other.

        Or put another way, if you could somehow have an actual auction, how much money could you raise from every church in america if the thing being sold was a supreme court position the week before roe-vs-wade?

        As such.. it doesn’t seem unreasonable to say that the price/worth of a seat is in the orders of hundreds of millions of dollars. easily.

        … but that much money is hard to play with without notice so lets assume an order of magnitude less money actually readily available for this.

        If you were tasked with belt-and-bracers-ing for your party re: the next supreme court nominee and it was expected in, say, a decades time. what could you do with a budget of a few million to lay the groundwork for a scandal? If you wanted to lay the groundwork for some credible accusations and had some willing partisans (or the cash to search out some willing partisans) how much would it cost?

        well, most supreme court nominees come from the circuit courts so that’s a good place to look for future candidates. About half nominated by each party. You’d probably want to concentrate on the ones most strongly opposed to your party and the super-elderly ones are less of a threat since they won’t live long if nominated. You could probably cull the list down to 50 or so candidates that your party found really objectionable.

        What could you do if you had ,say, $100K per person and the goal of manufacturing something for the future? You don’t need to make a fuss today. Everything is open to you. You can delve into their history looking for any angry ex’s who might be willing to say terrible things about them for the right price. You may even be able to do some things under the guise of journalists looking for scandal. You can try to get them to actually do something the public would disapprove of, like figuring out their dream partner and dangling a few models on a thread for an affair with pictures and other proof in case you ever need to pull it out of the bag.

        If they really are incorruptible and squeaky clean you could just get one of your fresh-faced partisans placed in the same building with them as a clerk or admin of some kind with the task of getting alone with them off camera before handing in their notice and make sure that your partisan turns up in a hospital somewhere that evening crying and refusing to file charges against anyone but saying lots of things that can be used later. a few years later you can point to medical records from the hospital and say “this was all before the person was even on the radar”.

        Honestly…. my main thought is that if I was trying to frame someone for something like this with the value of the supreme court seats and wanted to lay some political landmines with the kind of resources I’d reasonably expect such a conspiracy to have…. I think they’d be able to create something vastly better with more corroboration and much more damning.

        I’m sure you can think of more you could do with a few hundred thousand dollar budget per person.

        Which actually moves me slightly towards believing more that this Ford one is very slightly more likely to be real, if it was manufactured then something much more powerful could have been manufactured.

        • Jiro says:

          Which actually moves me slightly towards believing more that this Ford one is very slightly more likely to be real, if it was manufactured then something much more powerful could have been manufactured.

          By conservation of expected evidence, “the evidence isn’t powerful, so it is more likely to be real” implies “if the evidence was powerful, it would be more likely to be fake”. Is that a proposition you would defend?

          You can’t have a catch-22 where if the evidence is weak, that proves it’s real, and if the evidence is powerful, that also proves it’s real.

          • J Mann says:

            I’m mostly pro-Kavanaugh, but I agree with Murphy on this point. It’s not just that Ford’s story had weaknesses, but it had weaknesses that a smart liar could eliminate. For example:

            – If she and her husband are intentionally lying, he could have said she told her earlier, or she could potentially have found someone else to claim she told her.

            – She could agree with her husband that she named Kavanaugh in therapy, instead of saying she doesn’t recall that, but does recall saying one of her attackers could have been on the Supreme Court.

            It’s still possible that she’s lying, of course, but IMHO, I don’t think it’s likely that a crafty liar would introduce those details, either through incompetence or as some kind of double bluff.

          • Jiro says:

            I’m mostly pro-Kavanaugh, but I agree with Murphy on this point. It’s not just that Ford’s story had weaknesses, but it had weaknesses that a smart liar could eliminate.

            That’s still subject to conservation of expected evidence, though.

            So if you claim that “no sabotage” is evidence for the existence of a Japanese-American Fifth Column, you must conversely hold that seeing sabotage would argue against a Fifth Column. If you claim that “a good and proper life” is evidence that a woman is a witch, then an evil and improper life must be evidence that she is not a witch.

            If you think that the evidence is more likely to be real because it contains these weaknesses, you would also have to think that if the evidence did not contain these weaknesses, it would be less likely to be real. And I’m sure you don’t actually think that.

          • J Mann says:

            @Jiro,

            Using your framework, I guess I disagree about whether those details make the evidence weaker or stronger in this particular case.

            There are other cases where a particular collection of evidence makes something more plausible or less. Sometimes I do a job interview where every single answer from the interviewee is basically perfect. They have exactly the interests that would make them a fit for the job, their attitude is perfect, they have just the right flaws, etc. At some point, I start to raise my subjective probability that they have prepared for the interview and are being insincere. (And yes, an even smarter liar would introduce just the right amount of normal inconsistencies to fool me, but what are you going to do.)

          • Murphy says:

            You’re mixing together things.

            P(evidence is real/fake)

            vs

            (evidence is weak/strong)

            If you were going to plant fake evidence that your neighbor Bob was a Japanese spy would you

            1: plant a radio transmitter and fake code books on his property.

            2: edit some of his highschool SA’s to include subtle dog whistles implying that he admires Japanese culture.

            The former is awesome evidence that bob is a spy.

            The latter is terrible evidence that bob is a spy but the latter is not the kind of thing you’re going to place if you want everyone to believe bob is a spy. So while the latter isn’t going to convince me Bob is a spy it might move me 1/1000th of the way towards me believing he may be and I’m not going to bother considering that it’s a plant by a conspiracy.

            The former, 1, meanwhile is all-or-nothing. Either I believe it and bob is a spy or I believe in the conspiracy and Bob isn’t.

          • carvenvisage says:

            @Jiro the idea is presumably that these are trivial imperfections that someone with a purposefully deceptive mindset might be less likely to allow/tempted to avoid, as they hunt for what is superficially convincing and neat, than messy random reality would throw up.

            i.e. not weaknesses persay but purposeless imperfections which in the context of appearing guileless might actually be strengths.

          • Jiro says:

            @Jiro the idea is presumably that these are trivial imperfections that someone with a purposefully deceptive mindset might be less likely to allow/tempted to avoid

            Which still implies that people without those imperfections are more likely to be deceptive.

        • slightlylesshairyape says:

          You missed the key problem with conspiracies — the more people are in on it, the more opportunities for it to come out into the open. If you are running around planting evidence on dozens of would-be-nominees decades in advance, your exposure is massive.

        • zzzzort says:

          I feel like the lesson of the watergate scandal is that partisan conspiracies tend to be pretty poorly organized, and have a good chance of being exposed at some point. Given the risk of exposure, I would think you’d get more bang for you buck from just spending more on oppo research for supreme court nominees. It does seem like this is an area of under-investment.

          • Murphy says:

            Given the of the value of the things in the balance it seems bizarre that there’s no competent conspiracies. People manage to put together competent conspiracies when their goal is to add unofficial chips to card readers to steal credit card info, people manage to put together competent conspiracies in military and espionage matters, why does all the competence fall away?

          • Garrett says:

            I suspect that for most people, a good chunk of the payoff isn’t in the results themselves, but in being the person who got the results.

            If you are into credit card fraud, the goal is money. And everybody knows the process is illegal. So everyone has motivation to cooperate.

            If you are trying to buy/cheat/steal SCOTUS seats you only get the payoff if people know you were responsible. And, likely, because brains are weird, if you are the first name associated with the scheme. So everyone has significant motivation to defect.

          • Mark Atwood says:

            My moderately weak hypothesis about the rarity of competent partisan conspiracies is that it’s the result of the rarity of competent professional partisans.

            By which I mean, not the elected politicians from a party, and not the citizen voters who support a party, but instead the people who actually are the party machinery, including the constellation of closely allied thinktanks, contractors, cutout ngo’s etc.

            You would need a set of people with planning horizon greater than 10 years, strong group cohesion, basic office skills, excellent research skills, some people who can do what the tech industry calls “technical program management”, and the ability for everyone keep their mouth shut, even under the temptation to put it on your resume or brag about it to that cute guy/girl you’re trying to impress. Basic tradecraft, decent infosec, and even widespread basic “adulting skills” would be a plus.

            And you will need a small army of professional investigators, also none of whom will leak or fuck it up.

            And then you need a set of directors and paymasters to conceive and fund it, while themselves also with the necessary planning horizon, organization skills, access to resources, the ability to allocate those resources without public disclosure, and the ability to do it all forever without using having done all this as coin for self promotion.

            And also you do this through constantly shifting landscape of vicious individual and group status infighting, and wave after wave after wave of “fire, reorg, rehire”.

            My observation is that the D orgs and the R orgs are both terrible at all this, just differently worse broken in somewhat different ways.

            When I read “Shattered” by Amie Parnes and Jonathan Allen, I did so not with an eye towards the politics, or even the tactics, but just on their own internal organization and people management. My takeaway was that particular campaign, and the larger party apparatus, treat their people terribly, and elevate people who are terrible at people management and terrible at organization and coordination. And I don’t think that particular point in history was uniquely bad.

          • Chad Gonczy says:

            You have no idea what the denominator is. Thats the problem.

          • Murphy says:

            @mark

            I’m not sure that holds. Private firms that collect intelligence have turned out to exist in the past. People can be united by their love of money and then hired by whomever has cash and needs their services.

            That’s basically capitalism and lots of companies manage to enforce NDA’s for long periods, often apparently even for extremely illegal activity. In the cases we do find out about it sometimes turns out to have been decades… implying many groups can probably do so fairly stably.

            Though that being said I’m also fairly constantly baffled by the apparent almost complete lack of particularly competent terrorists and the profiles of terrorists and political partisans are very similar. So I guess your hypothesis is pretty reasonable.

        • One problem with your scenario is that a coordinated conspiracy on that scale risks leaks–someone involved either changes his ideology or sees the opportunity for a profitable betrayal.

          My understanding of the claim is that Ford says she told her psychiatrist she had been assaulted but not by whom. The unwillingness of Ford (or the psychiatrist) to provide the Senate committee with the psychiatrist’s notes suggests that the details don’t make a very good fit with the current story, although there are obviously alternative explanations.

          So the question becomes one of how many women there were who could plausibly have made some such charge—anyone the right age and living in the right area, or who went to college with him, or who clerked for him, or …—who had comparable evidence that something bad and sexual had happened to them in the past. That last requirement might lower it from thousands to hundreds, but I think it’s still a sizable number.

      • J Mann says:

        Ford told her therapist about the attack in 2012, before Kavanaugh was on anyone’s radar.

        I see your point, but in addition to the point that Kavanaugh was publicly being named as a potential Supreme Court pick by the Romney campaign, what we know about the therapist’s notes are that according to the Washington Post, the notes say that Ford was attacked by four students “from an elitist boys’ school” who went on to become “highly respected and high-ranking members of society in Washington.”

        Ford’s husband says that she named Kavanaugh in the therapy session, and Ford says that she told the therapist that one of her attackers might someday be on the Supreme Court, but we haven’t seen the notes or heard from the therapist because Ford’s attorneys made the strategic decision to withhold evidence in an effort to prolong the investigation. (IMHO).

        So the notes are definitely evidence, but if you grant the possibility that Ford and her husband are lying, then it’s possible that she was attacked by four boys from a DC private school and leveraged that to the current accusation.

        • what we know about the therapist’s notes are that according to the Washington Post, the notes say that Ford was attacked by four students “from an elitist boys’ school” who went on to become “highly respected and high-ranking members of society in Washington.”

          If that’s right, it doesn’t fit Ford’s later account, according to which only two of the students attacked her. And I don’t think the other one, Mark Judge, ended up as a “respected and high-ranking” member of Washington society. He seems to be an author and journalist known largely for writing about his own alcoholism.

          • J Mann says:

            Yes, that’s one of the inconsistencies. Ford and her husband told the Post that was an error by the therapist and that Ford had actually said that “there were four boys at the party but only two in the room.” I agree that Mark Judge doesn’t seem to be a respected and high ranking member of DC society, but presumably Ford would chalk that up to more error or lack of clarity by the therapist.

            During her testimony, Ford clarified that “[t]here were four boys I remember specifically being there: Brett Kavanaugh, Mark Judge, a boy named P.J., and one other boy whose name I cannot recall. I also remember my friend Leland attending.”

      • eqdw says:

        Devil’s advocate. I do not believe this but I am proposing it as a possibility:

        You don’t need to have Dr Ford go to her therapist and make up a false story half a decade prior just on the off chance that they need to politically assassinate someone in the future.

        The US is a gigantic country with tons of people in it. If you look hard enough, you can find singular examples of just about anything you want. Or, if it’s not an exact example, a vague and ambiguous one that you can hammer into shape.

        If you want to politically assassinate someone, you don’t have to have already laid the groundwork for this way in the past. What you have to do is find someone who has made some sort of claim like this to a therapist in the past, and who also had in the past some passing interaction with the target. You can massage the facts the rest of the way.

        I’m not saying this happened in this case. I intentionally didn’t pay attention to the Kavenaugh controvery while it was happening and am not familiar with the data. Obviously if it’s documented on record that in 2012 she told a therapist “brett raped me” then everything I said above is invalid. but, well, follow the thought experiment. How many women did Kavenaugh have social exposure to in his youth (eg shared a college class, went to same party, whatever)? Extrapolating from my own life, 1k-5k seems plausible. How many of those women have told their therapists that they’ve been assaulted in the past? Given campus activism stats, the number is closer to 1/5 than it is to 1/10,000. Let’s use a conservative estimate of 5%. 2500 * 0.05 => 125 people who a sufficiently motivated political assassin could use to mount such an attack against a high-profile politician.

        That is, envelope math tells us that there are 125 people whose pre-existing life stories could potentially be weaponized against someone. A good fraction of them might not work. eg. they met the guy in one year but reported the assault as having happened in a different year. Many of them might not go along with any such weaponization. There’s a lot of things that could invalidate any one of them for this strategy.

        All I’m saying is that it’s not necessary to have a confederate making up lies a decade before you spring your trap, in order to spring a trap like this

      • vV_Vv says:

        Ford told her therapist about the attack in 2012, before Kavanaugh was on anyone’s radar.

        If I understand correctly, she didn’t name Kavanaugh, or anybody else, to her therapist. The details of the story (e.g. the number of people involved) also changed.

        • J Mann says:

          If I understand correctly, she didn’t name Kavanaugh, or anybody else, to her therapist. The details of the story (e.g. the number of people involved) also changed

          The evidence is flexible enough that you can argue one way or the other.

          1) According to the Washington Post, the therapist’s notes don’t include Kavanaugh’s name and state that she was attacked by 4 students “from an elitist boys’ school” who went on to become “highly respected and high-ranking members of society in Washington.” Ford and her husband say that was an error, and that while Ford said there were four other people at the party, she only said that two of them attacked her.

          2) Ford’s husband says that she said Kavenaugh’s name in the session, and Ford says that she doesn’t recall whether she said his name, but she recalls that she said that one of her attackers might be on the Supreme Court someday. (Which presumably narrows down the possible candidates to at most a few people, even in the DC prep school scene.)

          I’m basically pro-Kavenaugh, but I don’t think those minor inconsistencies are much to work with – they’re consistent with Ford’s account, or with a hypothesis that Ford and her husband are leveraging that session.

          I’d be much more interested in how Ford’s account changed (or didn’t) over the course of therapy, but that’s highly private information.

    • slightlylesshairyape says:

      As long as we don’t have any remotely strong evidence one way or another, it really doesn’t matter to me whether the number is 20% or 80%. Either way, Kavanaugh should be confirmed

      This is crazy to me. Not being a rapist is a really important qualification for a justice.

      It’s like saying, well, we don’t know that this guy is a Russian spy, but if the odds are less than 20% we should let him work at the FBI.

      [ What’s more, if you say you are wiling to confirm at 40%, then you are accepting a 99% chance that one of the nine justices will be guilty of rape. ]

      • J Mann says:

        – I would say that at 40%, you have a 40% chance of naming a perjerous sexual assaulter to the Court if you name him, and a 60% chance of destroying a man’s life after what has been by all accounts decades of public service and kindness to those around him. At 80%, those odds change.

        – That doesn’t tell you how you should balance those odds of course.

        – The OP’s perspective is rules based – I infer that he thinks the consequences of acting on accusations that are both unfalsifiable and unverifiable are worse in the long run than acting on them.

        • grendelkhan says:

          and a 60% chance of destroying a man’s life

          I don’t think this was credibly on the table.

          Brett Kavanaugh wasn’t on trial; jail wasn’t a threat. The reputational effects were the same either way, whether or not he was confirmed. If his life was ruined, he’d still be… a circuit-court judge in the DC circuit, which is a pretty prestigious job. He wasn’t even up for a recall like Aaron Persky was. (For what it’s worth, I opposed Persky’s recall.)

          The downside simply wasn’t as bad as people made it out to be.

          • J Mann says:

            I won’t argue if you want to use a different term, since Kavanaugh wouldn’t literally be dead, but IMHO the consequences were quite serious.

            – He loved teaching at Harvard. Done.
            – It’s hard to imagine him ever speaking at another college or Bar event, particularly if he hadn’t been confirmed. He would basically either have to become a right wing advocacy figure or retire from public life.
            – He loved coaching his daughters’ basketball teams. Done.
            – His wife has a public job (I think she’s city manager of Chevy Chase or something like that). She’s being deluged with hate mail, death threats, wishes that their daughters would be raped, etc.
            – His daughters are going to go to college. For the rest of their lives, the thing most people are going to know is that half the country thinks their dad is a lying rapist.
            – It’s questionable whether he could stay in his current job. Supposedly, what made Kozinsky finally retire when he got #MeTooed was that he could no longer get quality students to sign on as law clerks.

            If you assume for the purposes of argument the possibility that Kavanaugh is literally innocent, then what happened to him is quite serious, and not something I would wish on any innocent person.

          • keranih says:

            I don’t think this was credibly on the table.

            I have a deep, immediate, visceral negative reaction to people who say, “It is no big deal to be framed for gang rape.”

            I think that either rape is no big deal, in which case being thought the sort of person who might do such a thing is also no big deal, or else yes, it is horrific to assault other humans sexually, and so no one would want that thought of them.

            It is not rational to want it both ways. (Might be human to want it, but not rational.)

          • grendelkhan says:

            J Mann: IMHO the consequences were quite serious.

            I agree. Like I said, the reputational effects are the same either way; whether or not he was confirmed to the Supreme Court, all of that is true.

            I should have been clearer–this really has been no picnic for him, and I’m not trying to imply this has been perfectly convenient. But the thing at stake with the nomination wasn’t whether or not he’d be put in jail, or whether or not he’d ever teach at Harvard again, or whether or not he can get clerks to sign on to work with him (though given this polling, I wonder if that would really be an issue).

            The only thing directly at stake here was whether or not he’d be elevated to a lifetime seat on the Supreme Court. The rest, good and bad, doesn’t hinge on that.

          • J Mann says:

            @Grendelkhan. Agreed. I’d further refine the point to:

            – To some extent, Kavanaugh’s name would be more tarnished if he withdrew or was voted down than if he was confirmed, in that (a) I think undecideds would see either outcome as some evidence in favor of guilt or innocence respectively, and (b) he now has the opportunity to put in a few decades of court service free from future credible accusations, as Thomas has done.

            – I agree that his name is still substantially tarnished even after confirmation, but it’s less so after confirmation than it would be after withdrawal or rejection. (For example, Ryan Lizza and Chris Hardwick survived accusations, and Keith Ellison seems like he might. Intuitively, that causes me to assume it’s more likely that they’re innocent than say, Matt Lauer who resigned in disgrace, even though I’m not particularly familiar with any of their accusations).

            – This also goes to the incentives. If the next accusation is handled through the confidential investigation process before the completion of the FBI check, that makes it more likely that the next nominee’s reputation will survive confirmation, albiet at an increased risk of a cover-up. I’m not sure how that plays out in the long run.

          • Edward Scizorhands says:

            If his life was ruined, he’d still be… a circuit-court judge in the DC circuit, which is a pretty prestigious job

            If Kavanaugh got shot down, it would be proof that Kavanaugh was guilty, and the Democrats wouldn’t let him keep that job. They have the power to impeach him.

            At a certain point it would have been best for Kavanaugh to withdraw, but doing so would be seen as an admission of guilt, and so he had only two paths forward: get confirmed to SCOTUS or lose most of the trapping of the upper-class lifestyle he’s spent the last 20 years building.

            If we just wanted to silently deny him the seat, Diane Feinstein had the perfect opportunity to do that. She could have told fellow Republicans on the SJC that she had a letter from an accuser while Kavanaugh was on the short list but before he was named, and the SJC could have told Trump “hey, nothing against your boy, but something’s come up and please pick someone else.” There was quite a bit of resistance to Kavanaugh within conservative circles, and walking him back before he gets nominated is the easy path to victory. Feinstein, I believe purposefully, closed off that path to create a political bombshell she hoped to use against the Republicans.

            This was Ford’s stated goal for giving Feinstein the letter privately. It would have dragged neither Ford nor Kavanaugh through the mud.

          • vV_Vv says:

            The downside simply wasn’t as bad as people made it out to be.

            The downside was pretty bad for him personally, but more importantly it was bad for the American society in general: if Kavanaugh wasn’t confirmed it would have signalled that unverifiable sexual assault accusations were a viable method of bringing down political adversaries. This would have emboldened future false accusers.

          • Doctor Mist says:

            Feinstein, I believe purposefully, closed off that path to create a political bombshell she hoped to use against the Republicans.

            Amen to that. I used to be moderately favorably inclined toward Feinstein, but her behavior in this fiasco has been absolutely disgraceful.

          • PeterDonis says:

            She could have told fellow Republicans on the SJC that she had a letter from an accuser while Kavanaugh was on the short list but before he was named

            As I understand the timing, this wasn’t possible; by the time Feinstein got the letter (July 30), Kavanaugh had already been nominated (July 9).

            However, she certainly could have privately shared information with the committee (and the FBI, if that were deemed necessary) right after she got the letter, instead of waiting until the middle of September when the fact of its existence was made public. And the fact that she didn’t do that strongly suggests that she was more interested in, as you say, use the accusation as a political bombshell, than in actually investigating it.

          • Edward Scizorhands says:

            Thanks for the timing update. I’ll have to update my priors a bit.

          • HeelBearCub says:

            But Ford sent a letter to her congressperson first, before sending a letter to Feinstein, I believe.

          • J Mann says:

            My understanding is that Ford contacted the Washington Post and a staffer at her Congressperson’s (Anna Eshoo?)’s office on July 6.

            Ford told the Judiciary Committee that she wanted to get her information in front of the President and the Senate as soon as possible, but she didn’t specifically know how to contact the Senate, and the people with her at Rehobeth recommended she call the Post and Eshoo.

          • PeterDonis says:

            My understanding is that Ford contacted the Washington Post and a staffer at her Congressperson’s (Anna Eshoo?)’s office on July 6.

            Do you have a source for this? I haven’t been able to find a reference to such a contact that early on in any of the timelines I’ve read.

          • J Mann says:

            @PeterDonis – her team has said the July 6 date a number of times. For example, in her testimony, Ford said:

            On July 6th, I had a sense of urgency to relay the information to the Senate and the president as soon as possible, before a nominee was selected. I did not know how, specifically, to do this.

            I called my congressional representative and let her receptionist know that someone on the president’s shortlist had attacked me. I also sent a message to the encrypted Washington Post confidential tip line.

            I don’t think the actual messages are public, but during the testimony, Rachel Mitchell and Ford both had screenshots of a July 6 WhatsApp conversation between Ford and the Post, presumably provided by Ford’s team.

          • Doctor Mist says:

            she didn’t specifically know how to contact the Senate

            Wait, what? A grown person with a doctorate can’t figure out how to contact her own Senator, who is ranking member of the Judiciary Committee?

            I was angry enough about this fiasco that I sent a letter to Feinstein myself. It took my literally two minutes online to find out how.

          • Edward Scizorhands says:

            A letter that keeps itself private? I can send a letter if I don’t care who reads it and gets read by an aide.

            Your representative is supposed to, well, represent the people, and be accessible.

            I’m not going to hold this point much against Ford. (The fact she was simultaneously reaching out to the Washington Post is a strike against “I wanted to stay anonymous.”)

        • slightlylesshairyape says:

          decades of public service and kindness to those around him

          A quick digression on this point is in order.

          I never disagreed with anyone that pointed out that BK did all these good things. I disagree with the idea that this implies that it’s implausible to believe he attempted to rape Ford. It just doesn’t follow. Good people occasionally have done bad things.

          It’s some combination of the fundamental attribution fallacy and the idea that rapes are committed by ‘bad hombres’. Some rapes are, but some are committed by perfectly ordinary men that go on to love their wives and kids and dog and coach softball and all that. The line between good and evil runs through every man’s heart.

          • J Mann says:

            I was thinking about that point too.

            IMHO, a little bit of this was a feeling about how many untestable assertions it takes to MeToo someone.

            My impression previously was that somewhere between three and six similar accusations was about what it takes to drive a public figure like Kosinsky or Lauer out of public life. I get that the Supreme Court is a more important job than whatever it was that Matt Lauer did, but I think some of the pro-“due process” people saw this as a shift towards one accusation (or 1.5 or 2, depending on how you see it) being enough.

            To that, I do think that his prior life is relevant for two reasons. (1) It does make assaulting someone and lying about it somewhat less likely, although you can argue about how much and (2) it increases the amount he has to lose if we assume he’s guilty.

          • The original Mr. X says:

            Good people occasionally have done bad things.

            They occasionally have, but not very often. (If they did do bad things often, they wouldn’t be considered good people in the first place.) So whilst having an apparently good character doesn’t disprove an accusation, we should have a strong presumption (the strength being proportional to how good the person is) that they’re innocent.

          • The original Mr. X says:

            To add to my above post: where people have been credibly accused of sexual assault, such as with Harvey Weinstein or Theodore McCarrick, there have generally been rumours and complaints about them for years, but nobody was willing to go public about them. With Kavanaugh, on the other hand, everybody who’s known him over the past 30 years seems genuinely shocked at the thought of him assaulting someone.

      • Jon Gunnarsson says:

        So let’s say you’re in charge of hiring at the FBI. You want to fill an important and prestigeous position and your top candidate is Bob, who has an exemplary record and all the necessary qualifications. Then along comes Alice, who you know dislikes Bob for political reasons and doesn’t want him to be hired. She tells you that 30 years ago she was at a party with teenage Bob, who told her that he was a Russian Spy. She offers no evidence whatever for the story apart from her word.

        What do you do? It seems to me that the most sensible course of action is investigate Bob again. If you find evidence of him being a spy, you don’t hire him. But what if you don’t find anything? Are you supposed to just pass on what you thought was the best candidate based on an accusation that has zero evidence behind it?

        • Edward Scizorhands says:

          It seems to me that the most sensible course of action is investigate Bob again

          This was the sane path forward, and it was in Diane Feinstein’s hands to make it happen. But she sat on Alice’s testimony.

        • slightlylesshairyape says:

          I mean, if there’s credible evidence that Alice specifically has a personal grudge against Bob, you can surely factor that in. But you also have to factor in that Bob was part of a group that traveled to Russia (maybe innocently) and wrote about how bad the American Imperialist project in Vietnam was (probably correctly).

          At the end of the analysis, you also have to agree that the disutility of hiring the second-most-qualified guy instead of the first is greatly outweighed by the disutility of hiring a guy that turns out to be a mole. So your decision making has to take into account (at least for the FBI) that a false negative is far more harmful than a false positive.

          Contrast that with the traditional criminal assumption where the disutility of imprisoning an innocent man is assumed to be many times that of acquitting the guilty.

      • Not being a rapist is a really important qualification for a justice.

        How about “not having unsuccessfully attempted rape once when he was a drunk teenager?”

        For what it’s worth, there is moderately good evidence that a president of the United States was a rapist–evidence known before he was elected. He admitted to having accepted paternity of a child born to an unmarried mother, strongly implied that he had slept with her but didn’t know if the child was his, and she strongly implied that he had raped her.

        • Doctor Mist says:

          Confirming Grover Cleveland in the face of credible allegations of sexual assault, and his smug refusal to deny the charges undermines the legitimacy of the Presidency.

          (cf. this)

      • rlms says:

        The word “should” is overloaded here. I agree that in some senses if the number is as high as 20% he should not be confirmed, and if it is 60% there are some further sense, but I feel like there is some sense where even if it is 80-90% there some sense in which he should be, or rather shouldn’t not be. That sense is a combination of innocent-till-proven-guilty/beyond-reasonable (even though that doesn’t necessarily apply exactly here) plus the fact that I don’t actually think whether he is or isn’t a rapist would much affect how he’d act on the court. Compare/contrast with Bill Clinton.

        • Compare/contrast with Bill Clinton.

          I had a low opinion of Clinton as a human being, not mainly based on sexual allegations, but I think he was a pretty good president, at least compared to other recent ones.

  9. Aging Loser says:

    What is the probability that at least 50% of powerful men throughout history have behaved badly while drunk? What is the probability that at least 40% of powerful men throughout history have behaved badly while drunk? And so forth.

    Yes, that sort of man will be that sort of man. Eliminate that sort of man? What an interesting experiment! What is the probability, if we eliminate that sort of man, that 75% of humankind will starve to death? That 50% will starve to death? And so forth.

    • beleester says:

      What are your probabilities for those outcomes? Do you actually think mass starvation is a likely outcome from powerful men not being allowed to commit drunken rapes, or are you just leaping for the most extreme consequence possible and using it as an everything-proof shield?

      Also, “powerful” is not an innate quality of a person, it’s something you can gain or lose. If “that sort of man” doesn’t get into a position of power, some other sort of man will. So unless you want to argue that getting blackout drunk is evidence that you’ll be a better leader than a teetotaler, your argument doesn’t hold any water.

      • Aging Loser says:

        beleester —

        Human societies collapse without leaders, and leaders have certain “innate qualities” that make them leaders (powerful).

        You can either (1) set up a Brave New System designed to prevent people of that sort from behaving badly while drunk at parties attended by unsupervised girls and women who aren’t prostitutes, or (2) return to the millennia-old practice of not permitting girls and women to attend parties except under the supervisory protection of male relatives.

        • beleester says:

          There are a lot of negative qualities that powerful historical leaders have had, but I see no reason to believe that they’re an essential part of the Generic Leader Personality rather than something that can come and go depending on social forces.

          A thousand years ago, the main feature that allowed a leader to rise to power was “ability to kill your enemies.” By your logic, we need to allow people to commit murder, because otherwise we’re punishing the sort of people who will become our leaders! Are you really going to risk billions dying of starvation just to keep a few murderers out of the halls of power?

          Why is “willingness to commit sexual assault” an essential leader trait that we can’t do anything about, while “willingness to commit murder” is not?

    • zzzzort says:

      OK, let’s see how the project of keeping SSC from becoming too right-wing goes:
      Estimates I’ve seen is that the overall rate of men who perpetrate sexual assault is a few percent (still disturbingly large, but not enough to cause the collapse of civilization). While that number may well be higher in more powerful positions, the selection effects would need to be really strong. The rate of sociopathy is estimated at being ~5x higher in CEOs. (i personally question the added social value of having those people in positions of power, but your mileage may vary.)

      Also, few people are suggesting that kavanaugh be ‘eliminated’. They are suggesting he not be given more power.

      Finally, the ultimate goal is not ‘men commit assault and then are exiled’, it is ‘men do not commit assault in the first place’. Human power dynamics are complicated and entrenched, but presumably respond to incentives.

      • Aging Loser says:

        zzzort —

        “sexual assault” means whatever the Progressives want it to mean, when they think that they can use the label to destroy anyone who represents a vestige of Western Civilization.

        “Human power dynamics … presumably respond to incentives” — yes, so Lenin presumed.

      • Garrett says:

        overall rate of men who perpetrate sexual assault is a few percent

        I think this is highly dependent upon what set of acts you include in “sexual assault”. This could comprise anything from a stranger yanking a victim into dark alley and engaging in violent, forceful, sexual intercourse, to copping a feel in a bar, to a married couple having sex after splitting a bottle of wine for a romantic dinner.
        And to my limited knowledge, the specific kinds of things that Kavanaugh was accused of are insufficiently studied.

      • rlms says:

        Don’t feed the trolls please.

    • grendelkhan says:

      zzzort covered some of this, but in a broader sense, this reminds me of the “Dr. House” problem–the idea that hypercompetence is a tradeoff with antisocial or otherwise vile behavior, that we must accept people who hurt others because only they can do great things. And while the idea is poetic, so far as I can tell, it’s false; it’s an excuse used by caustic or toxic people for failing to control themselves. It’s a bad meme.

      • slightlylesshairyape says:

        Why are you sure this idea is false? There’s a few reasons to believe it’s true:

        1) Competent people are more likely to succeed at bending or breaking the rules without consequence. This means they have more to gain from a defection strategy (from either or both of an odds/payoff POV).

        2) Competent people are recognized as valuable to their respective niches and so there’s a subconscious action by those around them to minimize their faults. This is optimal at some margin: a group /should/ try to acquire and retain competence.

        To put (2) another way, in a world where people can freely move between them, a group that punishes/expels wrongdoers irrespective of their competence will be outperformed by one that lets the more competent slide at marginally higher rates. This happens both directly and indirectly.

        3) Competent people are aware that they are valuable and so implicitly account for (2) when deciding when and how to follow the rules.

        My personal experience has seen some anecdotal examples of all 3, but I’ll admit it’s hardly data.

        This line of reasoning does not somehow rely on an a-priori correlation between being anti-social and competent. Rather, it’s a posterior correlation that arises out of behavioral changes that, on the margin, push the competent towards being more anti-social.

        • grendelkhan says:

          I guess I’m running partly on anecdotes as well. I work with a lot of high-performing people in a technical field. A single person can do pretty impressive things, but at some point you have to work with others. If you make your environment toxic, you’ll be lucky if your net effect is positive at all, let alone hugely positive.

          Specifically, I disagree that hypercompetent jerks are a positive asset, and furthermore, that they wouldn’t respond to incentives narrowing their options. Consider Enron, for example, which hired cutthroat analysts unencumbered by ethics, and was in turn devoured by them.

          I agree that if hypercompent people were competent enough to more than cover the losses they inflict on others, then there would be an asset to keeping them around; I believe that’s it’s very rare for anyone to be that competent. There are a lot more wannabe petty tyrants than there are Elon Musks or Linus Torvaldses. (And they, reportedly, aren’t all that bad as bosses go!)

          • For what it’s worth, I’ve known five Nobel prize winners in my field and five judges at the level just below the Supreme Court. Of the ten, one, who I didn’t know very well, appears to have bent the rules some, although not, so far as I know, at the level of attempted rape. The four I’ve known best I would all describe as clearly good people.

          • slightlylesshairyape says:

            I think you are missing the ‘on the margin’ part 🙂

            It’s in no one’s interest to create a toxic environment, not the rock-star and not the rest of the band or the supporting groupies. But on the margin, the rock-star will probably obey the rules a little less readily, violate protocol when it suits them more often and just generally be a bit more a jerk than they would otherwise be.

            And on the margin, the band and groupies will (on average) accept it and give him a bit more slack than if someone of moderate talent attempted the same.

            Of course, some people will overplay their hands — the amount of marginal leeway a person given isn’t printed out and posted on the wall with the rest of the rules, it’s discovered through social negotiation. A few times the rock star may overplay and be smacked down (in just enough proportion to remind them to stay marginally closer).

            In other words, it’s all about small incentives.

    • arlie says:

      You seem to be making the assumption that powerful men generally do good and important things for other people, collectively, and that there’s a shortage of replacements who could and would do at least as much good while not “behaving badly while drunk”.

      Let’s split “behaving badly while drunk” into 2 categories – making an ass of oneself, and harming or attempting to harm others, including rape. I don’t much care about people making asses of themselves; in this context, what I’m concerned about is violence.

      IMNSHO, if we could arrange with miraculously, 100% certainty, that commiting rape would result in an immediate, fatal heart attack, the supply of ‘powerful men’ would remain essentially unchanged. Probably ditto for any other specific violent and commonly illegal behaviour. Some might avoid misbehaving, who otherwise would have indulged themselves; others would simply be replaced by better people.

      I say that’s a net win.

      • slightlylesshairyape says:

        I think this is backwards. Power corrupts.

        Or, since the Simpsons did it: “you ever try going mad without power? It’s boring, no one listens to you.”

  10. Dan L says:

    I asked people whether they felt the evidence that Kavanaugh may have committed sexual assault was sufficient to reject his nomination to the Supreme Court, regardless of any other reasons to vote for or against him (like his legal opinions). 55% of respondents thought that yes, his nomination should be voted down; 45% still supported him. Here is a list of support for confirmation by probability of guilt:

    Of people who thought there was only a 0 – 9% chance Kavanaugh was guilty, 98% thought he shouldn’t be rejected from the Supreme Court on this basis alone. Of people who thought there was a 90-100% chance of guilt, 96% thought that was sufficient to reject the nomination. Of people who thought there was 50-50 chance of guilt, about 50% still supported him and 50% opposed him.

    This question suggests there is no real consensus about how plausible an accusation has to be before it means someone should be denied nomination to the Supreme Court. People generally agreed that if there was below a 25% chance the accusations were true, he should definitely be confirmed, and if there was above an 80% chance, he definitely shouldn’t be. But between 25% and 80%, people were pretty split on whether the Senate should err on the side of not confirming a potential assaulter, or wait until it was beyond a reasonable doubt. If we were trying to make these answers into a guideline for how a Senator should vote, it looks like they would be satisfying the most people if they voted to confirm if they thought the accusations had a less than 50% chance of being true, and to reject if they thought they had a more than 50% chance. I wonder how many people would endorse this rule as written.

    I’m a little disappointed that this part didn’t include any controlling for party affiliation – I strongly suspect that it is a better predictor of confirmation choice than perceived chance of guilt. (And that the seeming lack of consensus is just a shifting ratio of Reds/Blues at each probability.) This has been a perfect case for seeing people pick which side they’re on, then adjust their evaluation of evidence for strategic benefit. I’d like to see that taken a step into the meta level, to see if people are also adjusting how much evidence they think is meaningful.

  11. NoRandomWalk says:

    Does anyone think that a strong reason not to confirm is that a large portion of the country would view resulting 5-4 decisions as illegitimate?

    I bit the bullet and came down on the side of ‘even if 95%+ likely he did it, if he doesn’t admit it and there’s exactly zero contemporaneous evidence of any kind, he should still be confirmed because partisans have shown so little interest in investigating veracity of accusations that this precedent would result in much less credible accusations in the future being treated with the same weight because of how effective a political strategy it is’.

    But I am interested in seeing a discussion of which is the greater long-term harm:
    Weaponization of low-credibility allegations, or
    Perceived illegitimacy of the highest court in the land

    • Conrad Honcho says:

      I don’t think the assault accusations move the needle much on “illegitimacy of the courts.” I do not think Democratic partisans would accept overturning Roe regardless. And the talk about court packing started before the accusations.

      • qwints says:

        I agree. From a Democratic perspective, the Garland-Gorsuch saga was far more damaging to trust in the court. This maps closely to Thomas – a liberal to conservative switch, the presumed 5th vote to overturn Roe (remember that Souter was widely perceived to be equivalent to Scalia during the Thomas confirmation process) and accusations of misconduct. The Dems had congress and the presidency after the 1992 election, but took no action to attack the court.

    • Some Troll's Serious Discussion Alt says:

      Was any Trump nomination going pass that legitimacy test? The people who hate him seem to really hate him. Gorsuch was kinda tolerable because he didnt move the balance of power much, but Kavanaugh does.

      • HeelBearCub says:

        Gorsuch himself didn’t particularly move the needle for me any more than the simple failure to even hold a hearing for Garland.

        Kavanaugh was a gross display of altering the process for partisan benefit from the early on, right up until the end. Kavanaugh talking about the country reaping the whirlwind and revenge on behalf of the Clintons is icing on the cake.

        If Democrats ever manage to put a majority of liberals on the court, I expect the Republicans to pack the court when they are next in power. Maybe the Deomcrats will actually do it first, but I doubt it. Between Thomas’s current willingness to be explicitly political in appearance and now Kavanaugh’s confirmation process, I think we are pretty much done with the fiction of an apolitical judiciary.

        • Conrad Honcho says:

          Do you have any thoughts about Ruth Bader Ginsberg’s public airing of her opinions on Trump? Is that different in some way, or a manifestation of the same rot?

        • Edward Scizorhands says:

          Kavanaugh was a gross display of altering the process for partisan benefit from the early on,

          What did you mean by this?

          • HeelBearCub says:

            At the moment, I’m not going to go digging for a summary article pointing out what has been different about this process.

            I’ll start by pointing out that the normal process for reviewing the written records of any nominated justice so that any privileged or confidential material can be held back is to ask the national archives to do the review. Because Kavanaugh has such a voluminous record, mostly owing (I believe) to his time as a Republican staffer in the W. Bush WH, the archives would have needed until late October to complete the review.

            Rather than allowing them to do this review, the Republicans simply had Bush WH lawyers review the documents for release, and withheld 100,000 pages out of 550,000 pages. They claimed that they withheld them because of assertion of executive priviledge, but in fact Trump’s WH did not actually invoke executive privilege for those documents.

        • On the subject of court packing–expanding the number of seats in order to alter results–it’s already been done. Once.

          • tscharf says:

            It is entertaining that the same people who have been writing about the “death of democracy” for two years turn around and discuss court packing and justice impeachment on technicalities in the next breath. It’s a bit difficult to take it seriously.

          • HeelBearCub says:

            In 1868, I presume you mean. Specifically as to the issuance of paper currency?

            I’m not sure your framing is quite true. The court was reduced in size to prevent Johnson from getting any appointments. The increase when Grant took the presidency was likely happening regardless, I would think.

          • In 1863 Lincoln appointed Stephen Field, a pro-war Democrat, as the 10th justice on the Supreme Court.

          • Evan Þ says:

            On the other hand, I’ve always read that it was as much to bring the California circuit into harmony with the rest of the country by giving it a Supreme Court seat as any other reason.

            (Tangentially, Field also gave us an excellent dissent to the aptly-named Slaughterhouse Cases.)

    • tscharf says:

      Illegitimate is certainly today’s media buzzword. The impossible to prove fact but still likely to be true is that the left would consider any conservative majority illegitimate and a mirrored liberal majority to be perfectly legitimate. One piece of evidence for this is that people have convinced themselves it is illegitimate without a single decision having been made.

      I can’t speak for other people, but a conservative majority on the SC doesn’t mean all kinds of chicanery to promote conservative causes to me, but it is a holding action against the excesses of identity politics and mob rule. Freedom of speech, equal protection.

      • Jaskologist says:

        But it is a gift. Why not use this Ring?

        • tscharf says:

          The temptation may be overwhelming. Anytime there is a pile of money somewhere, there will be a fight for it. The power to veto nominees without evidence will not go unused. The power to direct culture through the SC may also not go unused. It’s an almost certainty that there will be conservatives out there trying to give the SC an opportunity to abuse their power. It’s up to the SC to not take that opportunity. It’s reasonable to wait a few years and see what actually happens. From the conservative side SC abuse is more defined as reinterpreting the Constitution as a living document. I’d prefer to have justices that are Constitution Bible thumpers, the literal word.

    • Plumber says:

      “…I am interested in seeing a discussion of which is the greater long-term harm:
      Weaponization of low-credibility allegations….”

      @NoRandomWalk

      I’d be surprised if that tactic doesn’t have diminishing returns

      “….or

      Perceived illegitimacy of the highest court in the land”

      Whatever reverence the Court still holds is long overdue to expire, which may be a good thing.

  12. Michael Pershan says:

    People with high background knowledge were more extreme in their answers, and slightly more likely to think Kavanaugh is innocent.

    This result reminds me of Daniel Kahan’s research on public perception of risks — things like the risk of climate change or of vaccines. Lots of people think that knowledge of the issues necessarily reduces controversy, but…

    Seeming public apathy over climate change is often attributed to a deficit in comprehension. The public knows too little science, it is claimed, to understand the evidence or avoid being misled. Widespread limits on technical reasoning aggravate the problem by forcing citizens to use unreliable cognitive heuristics to assess risk. We conducted a study to test this account and found no support for it. Members of the public with the highest degrees of science literacy and technical reasoning capacity were not the most concerned about climate change. Rather, they were the ones among whom cultural polarization was greatest.

    Kahan, Dan M., et al. “The polarizing impact of science literacy and numeracy on perceived climate change risks.” Nature climate change 2.10 (2012): 732.

    • That’s a somewhat special case, because it’s an issue where the media presentation considerably overstates the implications of the science. I would guess that uninformed people, asked how much global temperature or sea level has risen, would substantially overestimate.

      What’s more interesting about Dan Kahan’s work is that, where a position has become linked to group identification, the more intellectually able someone is, the more likely he is to support his group’s position, whether that means believing in evolution or not believing in evolution.

      Dan’s explanation is that this is rational behavior. Whether you believe in evolution has very little effect on the world in general but a large effect on how you interact with the people around you, making it in your interest to hold whatever position identifies you as one of them. The brighter you are, the better you will be at talking yourself into the beliefs that it is in your interest to hold.

      • tscharf says:

        You must have missed today’s news that we only have a single decade to avoid climate catastrophe. It seems that >1.5C is now officially catastrophic.

  13. VolumeWarrior says:

    Knowing nothing else except that people are 50-50 on this makes me lean strongly towards the “didn’t do it” side.

    The people who say he is unfit sound a lot like they’re virtue signaling. Saying that Kavanaugh definitely did it might be insane, but no one will doubt you are a Virtuous Person Against Rape. It’s like we have this norm where you can intentionally lobotomize yourself and as long as you yell for the right things, your tribe will overlook your ability to do basic math.

    Whereas the other side, which also thinks rape is bad (everyone thinks rape is bad…..), is saying reasonable things like we can’t convict someone based on hearsay alone, a “believe all victims 100%” norm creates moral hazard, etc. So at worst, they’re trying to signal intelligence and maturity by leaning on common sense. And if you’re going to use every current event as a prop to beat the other tribe with, this is not a bad place to be.

    • zzzzort says:

      -Personally, I dislike the term virtue signalling, as it always seems like a bravery argument in disguise.

      -The distribution seems pretty darn symmetric to me, and well explained by partisan affiliation. In particular, how are 100% people less sane than the 0% people?

      -You might be over-estimating the consensus on ‘sexual assault is bad.’ cf. Aging Loser above, or the current US president.

      • VolumeWarrior says:

        In particular, how are 100% people less sane than the 0% people?

        If you vote for “guilty”, you virtue signal that you’re a good person who cares about punishing very bad icky people. If you vote for “not-guilty”, then you signal rationality (low value) and possibly get labeled a rape apologist (extremely negative value).

        The fact that a large number of people are willing to bear this cost indicates that the actual facts and common sense are more compelling in that direction.

    • grendelkhan says:

      VolumeWarrior: Whereas the other side, which also thinks rape is bad (everyone thinks rape is bad…..),

      This is inaccurate. See this polling, specifically, page 17 here. In terms of yes/no/unsure, when asked if Kavanaugh should be put on the Court even if the allegations were true, Republicans came down 54/32/14. No other group rose above 50%, though White Evangelical Christians (48/36/16) and rural people (44/42/14) came close.

      For some qualitative salt on that polling, here you can see a group of women saying that even if it was true, it would be no big deal.

      • VolumeWarrior says:

        Yeah, I think those polls are strange. It’s definitely conflated with peoples’ desire to come down hard on either side of the issue. In my experience, people also really suck at answering “If X, would you Y”, particularly when trading in their group’s sacred symbols.

        But if you asked them the direct question: “is rape bad”, you’d get basically 100% agreement. This is what I meant. My prior is strongly that Republicans would not legalize rape if given free reign, and may even have harsher penalties for it given their affinity for the criminal justice system.

        • grendelkhan says:

          VolumeWarrior: But if you asked them the direct question: “is rape bad”, you’d get basically 100% agreement. This is what I meant. My prior is strongly that Republicans would not legalize rape if given free reign, and may even have harsher penalties for it given their affinity for the criminal justice system.

          If you look at the video, nobody said that rape was okay. They just say that [whatever our guy did] wasn’t real rape. And that’s how it’s legitimized; you massage the truth such that when your guy does it, it’s okay. (See: the principle that a seventeen year old’s actions should or should not affect the rest of their life, which is hardly a principle at all.)

          There’s plenty of rationalization there (note how one of the women spins an story from thin air about how Ford had a crush on Kavanaugh, and he didn’t call her back in the morning, and she felt bad about that, for example). The word has plenty of power. Taboo the word, and see what shows up.

          • VolumeWarrior says:

            So, caveat that discussions shouldn’t devolve into what the media can get insane people to say on camera.

            They just say that [whatever our guy did] wasn’t real rape

            I don’t really know the facts, details, etc, of what he’s being accused of. But assuming it wasn’t “real”, it must have been something else. Assault? I am sure basically 100% of people think assault is bad and should be criminally punished.

            Although, when you start asking people about the Kavanaugh case, they are bound to do mental acrobatics. I’m saying if you ask the GENERAL question, there is almost no disagreement.

          • The question here isn’t whether rape, or attempted rape (what was actually alleged), or assault, is bad. It’s whether having done such a thing as a drunk teenager is bad enough to disqualify someone for the Supreme Court decades later.

            I think attempted rape is bad. But if the only thing I knew against Kavanaugh was Ford’s story, assumed to be true, and he was in all other ways an exemplary candidate, I would be in favor of confirming him.

          • Doctor Mist says:

            But if the only thing I knew against Kavanaugh was Ford’s story, assumed to be true, and he was in all other ways an exemplary candidate, I would be in favor of confirming him.

            I think there are probably a lot of pro-Kavanaugh people who would agree with this in their heart of hearts, but in these times they dare not risk actually saying so aloud. Unfortunately, the only course that leaves is to pretend to actively disbelieve Ford’s story.

            I’d like to believe that someone who has been an upright citizen for three decades after doing something bad as a teenager (as the Democrats alleged about Kavanaugh) would be able to strike a blow for sanity by admitting the transgression and pointing to all the effort he has made since then to atone. But the recent history of #metoo shows that, once you are in its sights, a mea culpa does you no good at all — you will be destroyed regardless, unless you are sufficiently useful like Keith Ellison.

    • Placid Platypus says:

      we can’t convict someone based on hearsay alone

      There’s a difference between “convicting” someone and just deciding we don’t trust them enough to give them a lifetime appointment to one of the most powerful positions in our government.

      • VolumeWarrior says:

        Sorry. “Convict” probably reads as in a legal sense, but I meant in an informal social sense.

        I understand the argument that our highest offices need to be “beyond reproach”. But I hope it’s obvious that there has to be some kind of filter on what constitutes a red flag. Maybe Ford’s testimony should pass that filter, maybe not. But the current debate is so bad that it’s “no gatekeeping” vs “at least some gatekeeping”. Which is actually a pretty good summary of many culture war topics…

  14. gin-and-whiskey says:

    I think you made a mistake by going for absolutes instead of relatives. I would have done:

    a) What is the % probability you assign to an initial accusation of sexual assault? In other words: If you hear “Sally accused Bill of sexual assault,” without knowing more about Sally or Bill, what is your prior regarding Sally’s accuracy/Bill’s guilt?

    b) What is the % probability of criminal activity, above which you think a justice should not be confirmed?

    then…

    c) Having heard some evidence in this case, what is your % about Kavanaugh in particular?

    I think a lot of women will say that (a) exceeds (b). Most men will say that (b) exceeds (a). (c) will split on party lines.

    • Placid Platypus says:

      Overall in this conversation I’m surprised how high people are willing to put B (the % probability of criminal activity, above which you think a justice should not be confirmed). Personally I think even a 30% chance of having committed a crime as serious as attempted rape is more than I’m willing to accept in a nominee to such a powerful and unaccountable position.

      • Marklouis says:

        Why are you surprised? We routinely elect (and re-elect) people to important positions with very questionable past behavior. If anything, i’m inclined to go the other way: why would we be so hard on a Justice when the voters clearly do not heavily weigh allegations of questionable behavior? (As one real-time example: Bob Menendez is likely to be re-elected in New Jersey despite pretty damning evidence of corruption).

        The revealed preference of the voter base is that we don’t care about misdeeds unless it’s the other tribe. That’s fine…let’s just stop pretending otherwise.

    • Edward Scizorhands says:

      I had a hard enough time coming up with the exact percentage for Scott’s “how likely do you think it is that Ford is telling the truth?” question. Given a completely open-ended question like deciding between two randos, I would have closed the questionaire immediately.

  15. Alfred MacDonald says:

    my reply to this sort of thing is, and will probably always be, “I can’t know anything about the probabilities on this issue because what you’re asking me to estimate probability from is just gossip/hearsay/chisme, and asking me to infer probability from that is asinine.”

    • slightlylesshairyape says:

      Senators still need to vote. If we believe that the possibility of a nominee committing sexual assault is relevant, we have to come to some conclusion . . .

      • Michael Watts says:

        The possibility exists for every nominee. The accusations against Kavanaugh didn’t change that — like the claim “God hates fags” I asked you about upthread, they lack Popperian falsifiability. As such, there is no reason for those accusations to alter whatever conclusion you would come to in their absence. (And, wonder to behold, that’s exactly what we observe in reality!)

        When people make this same “you have to reach a conclusion” argument using religion-coded keywords, we laugh and call it “Pascal’s mugging”. The ability to state a claim doesn’t mean other people need to, or should, give that claim more than zero consideration. Claims need to have some internal strength.

        • slightlylesshairyape says:

          I responded upthread. There is no not-deciding. The same is true in a lot of contexts — a person is in the hospital and their son says the man wouldn’t want life support and the daughter says the man would have. The Doctor can’t sit around in some hypothetical land and decide that because the statement about the patient’s desires is not falsifiable he cannot come to a conclusion. No matter what, the Dr either leaves the life support on or takes it off.

          Also, the `God Hates Fags’ is only an example in the case where a decider has a linked binary decision such as “attend gay child’s wedding or don’t”. In that case, the decider must make a theological decision (maybe not whether ‘god hates fags’ is literally true, but some related ones regarding sin and so forth) no matter what.

          • Michael Watts says:

            I’ve responded to you also. My upper comment (here) is written in large part with reference to your response (there) as well as your comment here. You appear to be working off of some alien model of “deciding”, “not-deciding”, sin, and grace.

            I’ve been pretty clear about my view of the accusation and how it would be appropriately handled: because the accusation is literally meaningless (see: “there is no evidence”; “there cannot be any evidence that emerges in the future”; “the accusation lacks Popperian falsifiability”), it does not alter the probability that Kavanaugh committed sexual assault, and deserves absolutely zero consideration. As such, any decisions that were made or provisional before the accusation was aired remain exactly as valid afterwards as they were before. Saying “but you have to make a decision” is a non sequitur as far as I can see. It is true, but it is unrelated to any current issue.

            You appear to be worried that the Supreme Court might be metaphysically tainted by the fact of an accusation, despite that accusation’s complete separation from anything a positivist would recognize as reality. I am saying that the accusation is meaningless. Don’t transfer your anxiety about “having a rapist on the court” onto “having someone on the court who was once accused without evidence of being a rapist”. They aren’t similar things.

            Note that, because this accusation does not make Kavanaugh more or less likely to have committed sexual assault or whatever other offense, replacing him with another nominee is not even an improvement from the perspective of “but there might be a rapist on the court”, unless the replacement is a woman.

          • slightlylesshairyape says:

            it does not alter the probability that Kavanaugh committed sexual assault, and deserves absolutely zero consideration

            This is ridiculous. Because we can’t figure out with precision the truth of what happened, individuals cannot decide that it’s relevant to their own decisions?

            I will agree to one thing, our decision making processes are alien. You seem to be very devoted (almost religiously, I might say) to a particular positivist model. I’ve adopted a different (Bayesian) one.

          • Michael Watts says:

            “Bayesian” doesn’t contrast with “positivist”. Positivism is just the idea that if there are no consequences to the existence of something, it cannot meaningfully be said to exist.

            What, in your model of the world, is the difference between (1) the claim that Kavanaugh committed sexual assault at an unspecified time and place before no witnesses who would admit to seeing it and leaving no evidence behind; and (2) the company’s claim in this comic strip that employees who work themselves to death will be generously rewarded in the afterlife?

            Why would one merit more consideration than the other?

    • grothor says:

      what you’re asking me to estimate probability from is just gossip/hearsay/chisme, and asking me to infer probability from that is asinine.

      I don’t think this is true. Suppose it came out that there was conclusive evidence about his guilt or innocence, but you did not know what the answer was; you only knew as much as you know now. If, before you found out, someone offered you a bet, your $1 to their $1,000, for *either* position, would you take either of those bets? If you’re willing to take both of them, then you think the probability is between .1% and 99.9%. I expect you’d be willing to take less extreme odds than that. So you do know something about the probability, even if you have a lot of uncertainty.

      • Michael Watts says:

        If, before you found out, someone offered you a bet, your $1 to their $1,000, for *either* position, would you take either of those bets? If you’re willing to take both of them, then you think the probability is between .1% and 99.9%.

        This is quite obviously false; it implies that people play the lottery only because they don’t realize what the odds are.

        As a matter of gambling, I’d be happy to take either of those bets regardless of my assessment of what was likely to have happened, because the impact on my life of gaining $1,000 is minor, but the impact of losing $1 is a flat zero. I’d be even happier to take both bets, which would be a guaranteed gain of $999.

        • grothor says:

          I should have said “either of them” not “both of them”. Offering both to someone is an obviously bad idea.

          This is quite obviously false; it implies that people play the lottery only because they don’t realize what the odds are.

          No, because there are plenty of bets people would *not* take with those odds. If the outcome in question was “Will Donald Trump sprout a unicorn horn in the next six hours?”, I suspect many fewer people would be willing to bet in favor of that, even at 1:1000 odds.

          As a matter of gambling, I’d be happy to take either of those bets regardless of my assessment of what was likely to have happened, because the impact on my life of gaining $1,000 is minor, but the impact of losing $1 is a flat zero.

          Great! In that case, you should be willing to take the following bet: If the sun rises tomorrow, you give me $1. If it does not, I give you $1000. I’m happy to use a third-party arbiter for this, if you prefer. I’m also happy to take this same bet with you daily.

          • Michael Watts says:

            there are plenty of bets people would *not* take with those odds. If the outcome in question was “Will Donald Trump sprout a unicorn horn in the next six hours?”, I suspect many fewer people would be willing to bet in favor of that, even at 1:1000 odds.

            I agree with this, but it’s not support for the original proposition that being willing to bet on something at 1:1000 odds implies a belief that the something’s “true probability” is bounded at least 0.001 away from zero / one. That proposition still implies that lottery tickets with negative expected value wouldn’t sell if people understood the odds.

            It is support for the idea that people can intuitively model probabilities that are so much less than 0.001 that an ultra-low-stakes bet at 1:1000 odds still looks like a waste of money.

            I’m also happy to take this same bet with you daily.

            This isn’t right; repeating the interaction changes the analysis.

        • J Mann says:

          IMHO, Michael is correct that most people have non-neutral risk preferences over some ranges – they make negative expected value bets because they prefer having an out of the money long shot to having the cost of the bet, and they buy negative expected value insurance because they would rather not have the risk of the downside in that case.

          That doesn’t take away from grothor’s underlying point that we often need to (and do) make decisions based on uncertain information, or that putting money down often clarifies the mind.

          We could probably construct a largely risk neutral hypothetical (enough money that you don’t want to lose it, not so much that winning is exciting ($200 even money?), the bet will be paid by an impartial computer God and no one will ever know how you bet so that there is minimal signalling value). In that case, it would be interesting how people bet.

      • baconbits9 says:

        This is world shifting. One of them is telling the truth about the encounter occurring and one is not, so any probability that you put on either side that is not either 0% or 100% is, by definition, an incorrect assessment. Adding the information that the situation can be resolved shoves it into this world of 0 or 100%. The point with assigning probabilities to unknowns doesn’t work in a world of knowns.

        • grothor says:

          I agree there is something subtle and important about asserting that the truth is known and will be revealed. But there is some chance that it is known or will be known, and that it will be revealed. I do not think I would be willing to bet with 99.9% confidence that Kavanaugh will not be convicted of sexual assault or something similar for something that happened in the 80s, for example (though I do not think someone who takes that bet is crazy).

  16. RavenclawPrefect says:

    Well, that’s rather disappointing. My initial stance was “I know so little about this situation that I can’t reasonably make a good judgment”; after reading the unit of caring arguing in favor of guilty, I updated to maybe 85% in that direction. Seeing the level of partisanship even within SSC, I’m probably back down to ~60% on the outside view, and annoyed that we apparently can’t do better.

    What’s really annoying is I can feel myself being biased, taking in evidence from one side with grudging acceptance and the other with ease, and I don’t know how to turn this process down. I think I’ve managed it in a few cases – my System 1 is now neutral enough on gun control and the minimum wage that a compelling meta-analysis in either direction could shift my views easily, and it hasn’t always been that way – but in general I’m not sure how to account for this beyond forcing greater uncertainty into my beliefs.

    Clearly there are some people who are swimming against the partisan tide, though; I’d be interested in hearing from anyone with a view contrary to what they’d like to believe (in the sense that they can feel themselves biasing towards one end but are forced to the other by the state of the evidence). Good responses in this vein on topics other than Kavanaugh would be appreciated too; I’m a lot more interested in the general mental phenomenon than the actual details at hand here.

    • Eponymous says:

      I lean somewhat right overall, and strongly right on judicial picks.

      I weakly lean towards Kav being guilty, maybe 2:1 odds. My reasoning is (1) given his drinking/partying history and the culture he was part of, the accusation is plausible (prior reasonably high), and (2) Ford was a good witness and her story hangs together well. I gave small weight to other pieces of evidence, but those were the main ones.

      I haven’t done an explicit probability estimate, mind you. Just intuitive judgement after considering the components.

      It might be relevant that I strongly debiased myself, and may have overdone it. Like, I noticed that this topic was breaking down along partisan lines, and decided that I wouldn’t let myself fall into that trap. So in the end I may have been leaning towards finding him guilty to prove to myself that I’m an “independent thinker”.

      From a strategic perspective I think the GOP took on an unnecessary risk in pushing Kav through (i.e. more evidence surfaces later, Dems launch an investigation/impeachment proceedings, this energizes Dem voters, etc.)

      • grendelkhan says:

        If it helps, I’m generally center-left, and I made an effort to debias myself on this, mainly after being very surprised about the Duke Lacrosse accusations being false, and doing some significant soul-searching after that. I came up with about two-to-one odds that Ford’s story was true.

        • Mark Atwood says:

          Can you expand on this story about your soul searching? I’m genuinely curious.
          I promise, I won’t snark.

          I admit, I was surprised too about the Duke Lacrosse accusations being false, but realized in myself it was mostly remaining antipathy to college athletes.

          • grendelkhan says:

            Sure. It’s not really public anywhere; I didn’t write about it. I just remember being very certain that this story, which fit a narrative I believed–privileged, misogynist jocks perpetrating horrific violence–and when it turned out to be false, I felt like I, personally, had played a part in that injustice, even though I don’t think I said anything about it to anyone, or even left a comment. But I felt like I’d made a tremendous mistake.

            I cast a leery eye at MeToo because of this, and wondered at how to distinguish signal from noise here. I read what I could about how predators actually behave, and tried to use my reason as best I could on this case, though it didn’t endear me to people when I did it.

            I try to carry around that reminder of how wrong I’d been, as if I myself had been part of a screaming mob. As if I myself had made that mistake, instead of just following along when others did.

    • I’d be interested in hearing from anyone with a view contrary to what they’d like to believe (in the sense that they can feel themselves biasing towards one end but are forced to the other by the state of the evidence). Good responses in this vein on topics other than Kavanaugh would be appreciated too

      I’ve been in a discussion in a couple of FB threads on how common rape is, along with a related discussion about lesser forms of harassment, and that discussion plus looking at the victimization survey data, including data specifically on college women, plus conversations with two women I know well, significantly altered my estimates. In particular that, plus some conversation, persuaded me that it was quite likely that the rate among college women was substantially higher than among women in general.

      That was partly based on data, partly on rethinking my reasons for not expecting that to be true. I still think feminists considerably overestimate the frequency, but not by as much as I thought they did a month ago.

      • tscharf says:

        Most of the women I dated had a bad experience of some kind or other. I would like to see a little more granularity between the binary rape / not rape. Violent rape by someone you don’t know versus lack of verbal consent when drunk are pretty far apart. What seems strange is that while I’ve known women with bad experiences, I’m not aware of any man I know having been a perpetrator and would swear the bulk of them to be incapable of doing it. Somethings is not adding up, or else a small minority of men are doing most of the evil work. Confusion on consent seems to be a big problem.

        • SamaelTheDelicateOne says:

          Apologies in advance for poor formatting, poor form etc. This is my first SSC comment and my first use of this commenting system in general.

          …Somethings is not adding up, or else a small minority of men are doing most of the evil work. Confusion on consent seems to be a big problem.

          I have not dug to see if there is further (later) research along this precise line of thought, but that said, there is at least one academic paper addressing this.

          The abstract lays it out pretty neatly.

          The repeat rapists averaged 5.8 rapes each. The 120 rapists were responsible for 1,225 separate acts of interpersonal violence, including rape, battery, and child physical and sexual abuse

          As far as “confusion on consent” goes, I imagine that men who are bad at reading nonverbal cues are overrepresented in the population of abusers. More to the point, failing to recognize fear in the faces of others could be the critical difference between interpreting a first would-be-victim’s total verbal silence as “assumed consent” vs “terrified acquiescence.” Once such a mistake is made, the behavior would be prone to self-reinforcement due to the psychological consequences of radically recontextualizing one’s past behavior.

          This stands out for me as one possible explanation for why people cannot imagine anyone they know as being capable of such abuse on account of a related set of experiences I had with recognizing fear in others.

          When I was younger, I terrified people with verbal assaults in a handful of incidents. Some were to cover my escape from a dangerous situation, others were simply because I was angry, but most importantly I didn’t always intend to scare people or realize that I had.

          It was not until years later that it would occur to me to correlate the faces of those I intentionally bullied with those people whose behavior, to me, had seemed at the time purely inexplicable. IE I had no idea that I was scaring people because I did not realize “Oh, that is what being scared shitless looks like.” I’ve grown to accept this, but upon first realization I was mortified to learn that I was a bully.

          When I tell friends or family these stories, they usually think it is awesome and/or hilarious. If you had asked anyone who knew me and had not personally witnessed such an incident, then you would have had little idea I was capable of it. When I was a child, my father was similarly verbally abusive and when my parents briefly separated there was nobody we knew that was able to wrap their heads around the man they knew and the behaviors ascribed to him.

          FWIW I am aware of what subset of the population is known for missing facial fear cues. I actively do my best to not live up to the stereotype(s).

          • Careless says:

            Lisak’s paper is bunk.

            To quote Reason on it,

            The basis of Lisak’s 13-year old paper was not his own research but data collected as part of one student’s master’s thesis and three dissertations, none of which were about campus sexual assault.
            The most widely quoted figures—that 90 percent of campus rapes are committed by serial offenders and that they average six rapes each—were calculated on a total of 76 non-traditional students who were not living on a college campus, and whose offenses may or may not have happened on or near a college campus, may or may not have been perpetrated on other students, and may have happened at any time in the survey respondents’ adult lives.

            The students were as old as 71.

            When I asked how he was able to speak with men participating in an anonymous survey for research he was not conducting, he ended the phone call.

          • SamaelTheDelicateOne says:

            @Careless

            Wow! Thanks for the heads up/link on this. Shame on me for not doing a perfunctory ” + invalid” google search 🙁

        • adder says:

          or else a small minority of men are doing most of the evil work.

          Damn, I’ve heard this asserted as being backed up by some major study. ‘This” being the metacontrarian view that yes, actually, most sexual assasult is by a small number of repeat offenders. I wish I had a source… does this ring any bells for anyone?

          • Thomas Jørgensen says:

            It is a very solid finding, as social science goes, I remember it showing up in the surveys that taboo the word rape to get people to self-report, in surveys of who convicted rapists are, and how many victims they each had, and in analytics run on terrifyingly huge rape-kit backlogs. One-time offenders exist, but they make up a vanishingly small number of all instances of rape, and rape, as a crime, is more or less entirely the work of repeat offenders that average 10 cases or more each. Cosby was an outlier, but.. not that extreme an outlier.

      • Joyously says:

        Not comparable to rape at all, but I do think this story illustrates that “lived experience” really can skew people’s estimates of how common things are:

        I was out late in the city with my brother. We were headed back to our Air BnB at about 11 PM when we realized we’d left our bag at the last stop. I wasn’t wearing good shoes so I waited while he ran back for our bag. A guy approached me, starting talking, invited me to come hang out with him and his friend. I told him no, I was out with my brother. My brother comes back, talks briefly to the stranger, and the two of us left.

        Brother: “What did that guy want?”
        Me: “He wanted me to come play games with him and his friend.”
        Brother: “… Why would anyone do that?”

        I just thought it was funny how dumbfounded he was, because from my point of view it’s completely normal that every so often strange men ask you to come with them or get in their car.

      • orangecat says:

        In particular that, plus some conversation, persuaded me that it was quite likely that the rate among college women was substantially higher than among women in general.

        I’d expect that to be true just as a function of age. Women who don’t go to college are even worse off: https://www.newsweek.com/women-without-college-degree-risk-sexual-assault-720948

    • The Nybbler says:

      The Unit of Caring article was not well thought out. I’ll repost what I said on the subreddit:

      I think it’s vanishingly unlikely that she lied about Kavanaugh six years ago to her therapist and husband.

      She did not release the therapy notes. The portion she may or may not have shown to the Washington Post did not include Kavanaugh’s name (a former therapist I asked said she would not have included the name in the notes even if it was given). So this boils down to “her husband supports her claims”.

      which is true to reports of assault and not consistent with false allegations when they happen

      We do not have a ground truth for factually-false allegations, only demonstrated-false allegations; most rape allegations are not proven one way or another.

      her carefulness about details and about correcting the record over even minor confusions when she testified before Congress

      This appears to be making a virtue out of the vice of changing her story in minor ways.

      She described Judge hopping onto Kavanaugh’s shoulders, knocking off his balance and allowing her to escape – and corroboration was found that Judge was a high-energy screwball person who liked pulling that exact trick on Brett Kavanaugh in particular.

      It was? I certainly can’t find it. Of course, if this was publicly known before the accusations were made, it weighs the other way.

      Judge’s memoir portrays him and Kavanaugh as heavy drinkers who were frequently out of control, which is corroboration.

      Corroboration of what?

      We know they moved in the same social circles, as Kavanaugh’s calendar includes entries meeting with a boy Ford was dating.

      Ford actually refused to use the word “dating”, and as far as I know Squi hasn’t said he ever went out with Ford (as would be required for “corroboration”). That Kavanaugh and Ford were both suburban Maryland prep school students is not disputed in any case.

      We know Kavanaugh went to his friends’ houses for small, informal underage drinking parties that summer, because those are on his calendar too.

      Yes, and? Just showing that pieces of background in Ford’s story are true doesn’t corroborate her central allegation.

      And we know that Kavanaugh, age 17, at least sometimes enjoyed bullying women to impress his friends, because his yearbook entry includes a demeaning joke about the school slut.

      Well, no, we don’t know that. Even if “Renate Alumnus” was supposed to imply Renate was a slut, a few words in a yearbook are a long way from “sexual assault”, so this is the non-central fallacy applied to “bullying”. (Also Renate obviously went to a different school)

      That’s…. a lot of corroboration, actually.

      Corroboration of background information that no one disputes, not corroboration of the central allegation. This is a pretty common fallacy — take a bunch of mundane statements and one controversial one, and claim proof of the mundane statements is proof of the whole.

      I’ll add that the article fails to address the lack of corroboration from Judge (perhaps to be expected, as an alleged co-assailant), PJ Smyth, and Leland Ingham Keyser.

  17. sclmlw says:

    What I saw was Republicans basically said, “this should be treated like a criminal trial, where the burden of proof rests on the accuser,” while Democrats said, “this should be treated like a job interview where the burden of proof rests on the accused.”

    This wasn’t either of those things, though. This was a partisan struggle for control of the judiciary for years to come. That means that if you’re a Republican the burden of proof rests on the accuser, and if you’re a Democrat the burden of proof rests on the accused, and of you’re in the center you’re confused as there’s no standard given the uncertainties involved. This largely matches the observations here, and my own anecdotal experiences.

    • thedufer says:

      This was a partisan struggle for control of the judiciary for years to come.

      Regardless of which side you’re on, that’s clearly false. Gorsuch got through the nomination process with a reasonable margin, and there’s plenty of time before midterms to try again (which historically shouldn’t be relevant, but given the Garland mess probably is now). Kavanaugh, even before the accusations, seemed like a particularly unfit nominee (compared to the population of SC nominees, of course – this is not a claim that the average person is more fit or anything dumb like that). It’s hard to imagine the Republicans have literally no one more qualified than him.

      • Ghillie Dhu says:

        Gorsuch was essentially a hot-swap for Scalia; no change in overall strength on the Court so the stakes were lower. His nomination still faced a filibuster* from Democrat Senators.

        Given all the noise about Kavanaugh’s confirmation being rushed (it was longer even before the accusations than most, although he had a much larger paper trail to comb through), it’s not unreasonable for the GOP to infer that there wasn’t “plenty of time before midterms to try again”.

        *I suspect that, after the 2013 elimination of the filibuster for lower court nominees by the Democrats, Gorsuch was a carefully-laid trap to give the GOP an excuse to finish it off for SCOTUS nominees as well (“If they’re filibustering such an uncontroversial nominee as him, they’re clearly acting in bad faith so we are justified in removing their privilege.”); if so, then he’s the GOP’s conception of the least-opposable acceptable nominee (i.e., everybody else is going to get more pushback).

        • sclmlw says:

          I wondered during Garland’s nomination why the Democrats didn’t eliminate SCOTUS filibusters. As you noted, clearly they didn’t have scruples for lower court nominees. And Republicans were being at least as intransigent with Garland as Democrats were with Gorsuch.

          But going back and looking at the rhetoric of the time, they appear to have believed that Clinton would absolutely win in November 2016, after which they could nominate a less centrist judge for the Court. They could then credibly claim that any Republican objections were just the same old obstructionism, and end up with a much better outcome once the new nominee got sworn in.

          The fact that Democrats didn’t respond to the November loss by changing the rules and pushing through Garland between November and January was likely due to a number of factors, including that they couldn’t figure out how to ‘justify’ it in the same way Republicans ‘justified’ it with Gorsuch, and that there was likely insufficient time to do so.

          My personal biases: Getting rid of the filibuster for lower court nominees was bad; getting rid of the SCOTUS filibuster was bad; complete intransigence in working with the other side (by both sides under Obama, and again under Trump) has led to really bad outcomes where each side feels justified trashing the system for short-term gains. Thus, the Kavanaugh fight was never going to be conducted in good faith. The problem isn’t this one fight, it’s how we got here.

          • Jaskologist says:

            Democrats couldn’t do those things. Republicans had gained a majority of the Senate in 2014.

          • sclmlw says:

            Right! Thanks for the clarification.

          • Edward Scizorhands says:

            Most observers think playing hardball on Gorsuch, which destroyed the filibuster, was a bad move. If Democrats had gone along on Gorsuch with 60 votes, the filibuster would have been preserved for Kavanaugh, and even if you think the Republicans were extremely unprincipled and just doing power-grabs, they wouldn’t have had the votes to end the filibuster now.

          • eccdogg says:

            I am not sure I understand you on Garland. He was not filibustered, the Republicans were in control of the Senate. The Democrats did not have the power to eliminate the SCOTUS filibuster at the time. Did I misunderstand you?

            ETA: Sorry I missed Jaskologist comment.

          • sclmlw says:

            No, I forgot that Republicans held the Senate in 2014, so my whole comment was moot, as Jaskologist pointed out.

      • sclmlw says:

        You’re saying that the difference between Gorsuch and Kavanaugh is not the time factor? I find that hard to believe. Consider from the Right’s perspective: to get a new nominee confirmed prior to the next session of congress (January 2019), they want to assure sufficient time for the nomination process. It took Kavanaugh from July to October, or about 3 months, so that leaves almost exactly equivalent time for the next nominee. That also creates a more pressing ticking political clock, and both sides know it. Most on the Right I’ve talked to believe Kavanaugh was falsely accused and railroaded (no evidence of that, but ideology is not about evidence). So in their mind the narrative goes something like this, “Well, they’re just going to do this same thing again to the next nominee; if we establish a precedent with this nominee that we consider allegations of misconduct alone to disqualify a candidate, the other side will do the same thing with the next nominee.” (Of course, if there’s time for a full investigation this calculus might change, but any theoretical second-shot nominee would face at least the time constraints as Kavanaugh, so the calculus doesn’t change in dealing with uncertainty about sexual assault allegations for which you don’t have time to determine substantiation.) From the perspective of the Right, the game-theoretic result of rejecting Kavanaugh is to ensure the next nominee certainly doesn’t have sufficient time to be nominated.

        (Now, the flip side of that is when I point out to my Republican friends that the reason you can’t get a nominee through when your side doesn’t hold the Senate is because they set a precedent with Garland that you have to hold the Presidency and the Senate to get a nominee through. Whatever, it’s all short-term political gamesmanship that sets long-term, seemingly unintended precedents.)

        Meanwhile, I think for Republican strategists, they’re more concerned about the nomination fight continuing into the midterm elections. November isn’t the right time to talk about nominees accused of rape; if the next nominee gets immediately accused of misconduct, they’ll have to defend that nominee while running re-election campaigns. By confirming Kavanaugh, they’re hoping the public has a short enough attention span that they’ll be sufficiently tired of the topic by November and the confirmation will hurt them less than it would if they were fighting some new battle about a new nominee. And I think Democrats see the possibility of actively preventing an ideological shift of SCOTUS as a big driver of voter turnout, so they really wanted the Republicans to withdraw Kavanaugh for the same reasons Republicans really wanted to get the confirmation in the bag. Three months for a confirmation – during which time you have to deal with Senators fighting for reelection – is too high-risk for Republicans to accept when the ideological leaning of the Court is their ultimate concern.

        • By confirming Kavanaugh, they’re hoping the public has a short enough attention span that they’ll be sufficiently tired of the topic by November and the confirmation will hurt them less than it would if they were fighting some new battle about a new nominee.

          I’ve seen it plausibly argued that the confirmation fight is going to help the Republicans hold the senate while making it harder for them to hold the house.

          The argument is that the Kavanaugh fight energized both bases, that there are a bunch of Democratic senators running for reelection in deep red states, and that energizing the Republican base is going to make them more likely to lose. Presumably the one such senator who voted for confirmation agreed.

          • sclmlw says:

            If that were true, wouldn’t the Republicans have been better off waiting for midterms before scheduling the vote? They had the option of waiting one extra month “to decide” the matter, during which time they could have had their cake (the possible electoral boost you refer to) and eaten it too (confirmation of their nominee during lame-duck period).

            I’d guess certain votes they needed (the one Democrat who voted yes and perhpas more moderate Republicans who voted yes or present) could not guarantee their vote for confirmation after the midterms. One more example of how statistics don’t necessarily carry forward to specifics.

          • eccdogg says:

            I watched the election betting odds during the whole confirmation process.

            As the odds of Kavanaugh getting approved went up the Senate odds for Republicans went up but the House odds for the Democrats also went down slightly.

            By that metric confirming Kavanaugh was a win all around for Republicans

          • Careless says:

            If that were true, wouldn’t the Republicans have been better off waiting for midterms before scheduling the vote?

            Only if they won. If they lost, under two months would be a very short window to get someone else through the process

  18. Jiro says:

    How much of what this is measuring is the way people make estimates, rather than how polarized they are about Kavanaugh specifically?

  19. eqdw says:

    My perspective on all this Kavenaugh situation was informed by exactly two data points:

    1) The more aggressively public institutions and authorities are trying to convince me of something, the less likely it is that that position is true (on the grounds that, if they had the truth on their side, they would not have to be so aggressive in their persuasion)

    2) The more people I encounter who support a given position for bad faith or otherwise non-rational reasons, the less likely it is that their position is true (on the grounds that truth is not established by, eg., how happy or sad it makes somebody feel).

    In this case, on (1), I witnessed the entirety of my mainstream society come down, aggressively, on “He’s a rapist” while I only witnessed a handful of kooks on the pro-Kavenaugh side engaging in bad faith. And on (2), well, I’ve yet to meet a person offline who actually supported this guy, but I sure met a lot of people who opposed him for reasons I considered invalid.

    I’m satisfied that this decision process led me to the correct position (“There is insufficient data to demonstrate either guilt or innocence, and so fall back to innocent-until-proven-guilty. Meanwhile, the dramatic degradation of norms and etiquette around this needs to be very strongly punished/disincentivized, and the best/least-violent/least-destabilizing way to do that is to demonstrate that it doesn’t work, which was in fact what happened). However, I’m concerned that this decision process is highly susceptible to sampling bias.

    Imagine, for instance, that 80% of all people are bad-faith or otherwise non-rational, and 80% of all social authorities are really just using their authority to push their preferred position. In this case, my decision process will lead me to knee-jerk oppose whichever faction I happen to be more exposed to.

    As I live in a relatively left-leaning area, have relatively left-leaning peers and colleagues, and observe that most social authorities (media, news, etc) are relatively left-leaning, if you assume that everybody is just equally terrible, my decision process will reliably send me to the right leaning conclusion, regardless of truth or falsehood.

    • Placid Platypus says:

      There is insufficient data to demonstrate either guilt or innocence, and so fall back to innocent-until-proven-guilty.

      I’m not sure this is the most logical fallback given that this isn’t a criminal accusation that can lead to a conviction and sentence. The question is whether he’s an appropriate person to hold one of the most powerful and unaccountable positions in our government. In that context it seems reasonable to hold him to a higher standard than “Well, he might not have done it.”

      • eqdw says:

        I believe, very strongly, that if some body or institution with social authority (eg the FBI, the Senate, whatever) publicly concluded that he had in fact done what he was accused of, then his life would be over. He would likely lose the positions he already held, and would likely have a hard time finding employment for quite some time. I believe that he would be harassed in public whenever recognized, same as the politicians who supported him were. I believe his family would have to put up with same. His kids would be harassed at school by friends and peers _and teachers_ as “that rapist’s kids”. His wife probably leaves him. I believe there’s a small chance (say 5%) in this situation that he blows his brains out, and a large chance (say, 30%) that he ends up with his life effectively destroyed.

        Given my assumptions, I think “innocent-until-proven-guilty” is a reasonable strategy to take. If other people have different assumptions, maybe that strategy is not reasonable for them to take.

        In addition to the above, there is a secondary strategy that synergizes with that one, which is that ‘the burden of proof always lies with the person making the claim that contradicts the status quo. In this case, the status quo is “some guy I’ve never heard of is living his life” and the claims are (1) “this guy raped me” and (2) “you should be very angry about this”.

        In general, it’s not my responsibility to change my mind on a given issue just because somebody wants me to. It’s actually their responsibility to convince me that I am currently wrong and that I should change my mind to be correct. And, in general, it’s not my responsibility to care about things. We all have our lives to live, we all have responsibilities, obligations, things to think about. Other people are not entitled to our deliberative consideration by default. If somebody wants to get me to care deeply and act upon something that they care about, they have the burden of proof to convince me to do so. Otherwise, I would be vulnerable to mental denial-of-service attacks as people spam me up with causes and demands that I care about them, and I would be stuck having to retroactively do the work to figure out which ones are real or not.

        Dismissing the above as an instance of “well he might not have done it” is reductive and uncharitable. As a trivial demonstration of this: if every superficially plausible accusation against someone causes a positive moral obligation on everyone to actively disprove it, then I have several accusations I’d like to make. Some are true. Most are false. But, well, once I make the accusation, I don’t think “The accused _might_ not have done it” is sufficient to clear their names. Y’all better get to work disproving these things.

        Obviously if I did that, this would be absurd, you would tell me to frig off, and you would be absolutely correct to do so. I’m applying this principle to this kavenaugh stuff.

        Final point worth mentioning: I’m saying what I believe. I’m saying what I use to establish truth or falsehood and how I judge confidence in beliefs. I am _not_ saying how others should do this. I am _not_ saying how, eg, the senate should do this. They are in a different position. Different rules apply.

      • sidereal says:

        > In that context it seems reasonable to hold him to a higher standard than “Well, he might not have done it.”

        Well, sure. It’s the same sort of logic that applies to job applicants: they don’t have to *prove* you did anything wrong. They can turn you down simply because they got a wierd vibe off you. And it’s totally reasonable too, assuming they are more job applicants. And to be sure, isn’t any chance that a supreme court justice is a rapist too high?

        However, in this particular case, while it’s not a criminal proceeding, it’s not a regular job application either. There is a tremendous amount at stake, which should raise your prior that someone would make a false, pointed accusation for strategic considerations to a non-negligible level. Further, this would set a precedent for future character assassinations (whether or not this is one) which feels like a moral hazard to me. I dunno, it’s really tough.

    • arlie says:

      I sympathize with both of your heuristics, and have used them for years. I still use them in some contexts, especially when I’m not concerned enough to engage my full rational evaluation process (i.e. I’m using my “fast” more than my “slow” system.)

      But I’ve stopped using them as part of my “slow” system, after collecting too much evidence of people supporting good causes in horrifically bad ways. Between never-rational feeling-oriented types who appear incapable of using their under-developed “slow” systems, and rational actors communicating based on research about what will be most convincing to more people (hint: not rational argument), I now discount the presence of bad-faith and irrational fellow travellers, and simply try to look for rational people making rational, data-based arguments.

      That’s hard – it’s easy to create plausible-sounding, seemingly data-based arguments that selectively omit relevant information so as to support any desired conclusion, not to mention simply making up evidence. We do not currently appear to have a sufficiently robust public review process to rely on this behaviour being routinely noticed and publicized – hence there’s incentive to escalate it farther, putting fact and logic checkers even farther behind. My confidence in my own estimates of truth have dropped significantly as I realized this, particularly in areas most prone to persuasion attempts, and lacking obvious “follow the money” incentives.

      • eqdw says:

        I feel that people supporting good causes in horrifically bad ways is weak evidence against those causes being good.

        If I’m currently undecided on an issue, that means that I do not know, a priori, that it is good. If I go and look for reasons why it is good, and the only thing I get is evidence that I would consider to support the hypothesis that it is bad, I will update in the direction of believing the issue to be bad.

        I believe that this is important for both an object- and meta-level reason. Object level: if someone makes a bad argument, I take that data at face value. If, for example, someone argues in favour of position X by saying something that I know is a lie, I update in the direction of “position X is favoured by liars. I am not a liar. Therefore, weak evidence in favour of not-position-X’. I suppose in a sense this is guilt by association.

        More importantly is a meta level. I am aware that I am very, very bad at figuring out what is true or not. I am also aware that there are lots of unscrupulous people people out there who would try to manipulate me into supporting a position they prefer, regardless of truth value. And finally, I am aware that there is a lot of noise out there, and that good clean strong data is hard to come by.

        I want to strongly disincentivize bad data, as a way to raise the signal/noise ratio of available data and, hopefully, slowly converge to truer beliefs. In order to do this, there has to be some form of penalty to disincentivize the spreading of bad data. In this case, it’s my small little activist stand against the world: If you want me to adopt your position, you have to help me clean up the data landscape by preventing the spread of bad data. If you won’t do this, if you will tolerate yourself or your peers spreading bad data in the course of advocating for your position, then I will add a little penalty cost to that position that needs to be overcome before I adopt it.

        I’m aware that this is probably absurd, and I’m aware that (in the short term, at least) this probably marginally prevents me from converging on true beliefs. However, given the massive error bars that I hold on most of my beliefs, + the relatively small size of each individual malus I apply, I’m not too concerned about this.

        The only concern I have is what I laid out in the original post, which is that this might generate incorrect conclusions if I get overwhelmed with _too many_ bad arguments, especially if the underlying set of people I’m sampling from has a bias in some direction.

    • Nicholas Weininger says:

      You are right to be concerned. Certainly it is extremely typical of humans of all political persuasions to rationalize away misconduct by people on their own side in the face of a threat from the outgroup; the cases of Bill Clinton and Teddy Kennedy demonstrate that, for example, self-avowed liberal feminists are not exempt from this, and I can see few reasons to believe Kavanaugh supporters would be any better at avoiding this than others, and some reasons to believe they may well be worse in this case.

      • eqdw says:

        I would politely request that you reread what I posted. At no point did I make any mention of excusing the people who support kavenaugh from being held accountable for this behaviour.

        What I said was that I have literally never actually met someone who supports kavenaugh, and so I have no data on those people at all.

  20. Plumber says:

    I’ll just repost what I already have:

    “….I want to push back a bit on the concept that a judge must have exemplary character….”

    @Mark V Anderson,

    Speaking only for myself, my opposition to Kavaugh had nothing to do with his character. 

    I mostly opposed him because of what to me seem multiple anti-worker pro employer decisions (the OSHA vs. Seaworld one is most prominent in my mind, but there’s list), and I’d be happier if he wasn’t a Yale law school graduate (I’d be happier still if he wasn’t a law school graduate at all and passed the Bar after serving an apprenticeship, but we haven’t had a Justice like that since the 1940’s).

    But what was revealed about his background made me like him even less, no not the drinking and alleged harassment and assault, but insteas his prep school/country club privileged roots while claiming “he had no connections” really made me want him not on the court and opposed even more. 

    My wife on the other hand, was convinced of and livid about his alleged groping, et cetera and followed the case much more closely than me.

    My boss also followed the case and was convinced that Kavaugh “was being railroaded” and was angry about the “sexual politics” (and says he and his wife fought over it, as she had a different opinion).

    So my wife (on one side) being angry this and last week made home life less pleasant, and my boss (on the other side) has been grumpy as well, making work harder. 

    For myself, I hope that I don’t hear about it much more, but I fear that it’s now more likely to be more decisions like Janus, which will make me angry.

    And to continue: What Kavaugh did as a youth is of infinitesimal interest to me compared to what decisions he’s made as a judge.
    Roe v. Wade is often talked about, so I’ll state my position on it:
    I think it’s a bad precedent that is anti-democracy.
    It’s said that Republicans are more likely to overturn Roe v. Wade (I’ll believe it when I see it), but that still doesn’t sway me to vote Republican as I care far more about protecting the Voting Rights Act and protecting collective bargaining.

    Kavaugh is guilty, guilty of anti-union decisions (I haven’t bothered to look up any other of his judicial decisions besides labor issues because those are enough for me to oppose him) as to his youth?

    One side portrayed him as a drunken prep school lout, the other as a devoted father and husband, fine, those “facts” I’ll call even, and I really don’t care about that, it’s what he may do as one of our nine Kings that concerns me, not this “character” sideshow.

  21. AM says:

    Oh, that was the point? I thought carefully about the hypothetical, but for the probability question I just made up a number because I don’t like/find it difficult to weigh evidence.

  22. Luke the CIA Stooge says:

    Alright this has beem driving me nuts for the entire length of the hearing, and yet no one is saying it.
    NONE OF IT MATTERS

    A life time appointment to the Dictatorship of Nine is infinitely more important than a thousand rape victims.
    If either side found away to replace an ideological opponent on the court with a recently reanimated Adolf Hitler I’d expect them to do it and thank their lucky stars they could change a permanent opposition vote to a swing vote.

    And if you have any ideological commitments at all, this is the correct way to think about it.
    If you are actually prolife and think abortion is murder then millions of babies a year are being murdered in a crime worst than the holocaust, at least hitler wanted to stop after 12 million jews were wrapped up, pro choice advocats want to kill millions of babies a year FOREVER.

    likewise if your prochoice the only thing standing between millions of women and forced vaginal occupation is the dictatorship of nine.
    If you believe women have a right to control their body then forcing a woman to carry to term is the equivalent of rape(actually worse than rape, rape last average 20min, it hurts, then its over, being forced to carry to term is 9months of having your sexual organs forcibly occupied then 8-50hours of the worse pain humans can experience and possiblily mangling your sexual organs for life)
    At least hitler planned for the rape of eastern europe to stop at somepoint, prolife advocats want to worse than rape millions of women every year FOREVER.

    Now multiply that across every potical issue that will be decided by the supreme court (ie every political issue)

    Now compare losing that fight to one rapist getting away with it, or one respectable family having their lives ruined over false alegations. It doesn’t matter.

    If Hitler would vote your way you’d want him on the supreme court.

    If it came out tomorrow that Kavanaugh did 9/11, then anyone with any ideological commitments would be most concerned that the idiot swing voters might let this affect their resolve, because a few million dead in the middleast and 2 decades of war is small potatoes compared to control of the dictatorship of 9.

    And i just want someone somewhere to ackowledge it!

    If we assume any of us actually mean what we say about politics, the kavanuagh hearings don’t matter, and neither do elections, investigations, character or debate.
    What matters is grabbing enough positions of stregth before the inevitable civil war.

    • Some Troll's Serious Discussion Alt says:

      If we assume any of us actually mean what we say about politics,

      Im hopeful that we mostly don’t.

      • Luke the CIA Stooge says:

        Thats the scary thing political rehtoric has a way of becoming the goal whether anyone wants it too or not.
        In 1925 in Mein Kamph Hitler had genocidal rhetoric about killing the Jews but then in 1942 core party members were still caught off guard at the Wannsee conference when Heydrich anounced that was the plan. Eichmann especially since now he’d learnt yiddish and studied zionism for nothing.

        Likewise i doubt most russian communists actually thought they were going to execute the kulaks, landowners and royal family.
        But once you’ve picked a side and repeated the rhetoric its hard to stop even if your losing let alone when your in power.

        This is clearest on abortion in the US. If abortion is murder and the US has preformed more abortions than jews died in the holocaust then isn’t the US commiting a holocaust?
        Aren’t good people supposed to fight against those commiting the holocaust?
        Wouldnt killing a lead executioner in a holocaust be a heroic thing?
        Then wouldn’t killing abortion doctors be heroic?

        Most prolifers would be adamantly opposed to that logic but you’ll never corner them on where its wrong because then theyd have to admit they dont consider an abortion the moral equivalent of shooting an innocent jewish child. And therefore they don’t think a fetus is human baby life.

        We can tell the same story with the left. captain America punched Nazis to stop to the spread of fascism. Oh look fascism’s spreading time to punch Nazis. Once you’ve publicly avowed that your opponents are nazis you kinda have to do it, otherwise your a sypathizer or hypocrite.

        The problem is what happens if the nazis keep winning. Captain america was cartoon how have good on-the-right-side-of-History upstanding americans actually dealt with nazis in the past?

        Americans on both sides have already publicly commited to having a civil war if the otherside doesn’t unconditionally surrender and they’re recommiting every day. Its just so far they’ve thought its rhetorical.

    • arlie says:

      *roflmao*

      I was about to chide you for not realizing that you can’t run a country, long term, without compromise, and then got to your last paragraph.

      What matters is grabbing enough positions of stregth before the inevitable civil war.

      It can be argued that the main thing wrong with the US today, is the diminishingly tiny fraction of the population who have any idea of the scale of harm likely to result from e.g. civil war, and who thus never think about the ultimate effects of extreme “us vs them” politics.

      And that’s true whether or not your post is intended as satirical.

      • Garrett says:

        Thought:

        If civil war is so bad, give your opponents what they want and they won’t bother.

        No thanks.

        • arlie says:

          Once those are the only choices, it’s already too late.

          [Edit – but having people *claiming* those are the only choices is at best very weak evidence for this beign true.]

    • Evan Þ says:

      If you seriously think we’re on the verge of a literal civil war, the rest of your comment doesn’t make sense. The South controlled the Supreme Court before the Civil War, and that didn’t help them – in fact, it probably hurt them because it radicalized the North. Similarly, having Reanimated Hitler on the court would give your opponents excellent ideological ammunition which would almost certainly hurt you in the future civil war.

      Yes, having abortion (allowed / banned) (take your pick) for twenty years or so until the impending civil war would be a horrific thing. But losing that civil war and having it (allowed / banned) indefinitely into the future would be even worse.

    • carvenvisage says:

      If you believe women have a right to control their body then forcing a woman to carry to term is the equivalent of rape(actually worse than rape, rape last average 20min, it hurts, then its over, being forced to carry to term is 9months of having your sexual organs forcibly occupied then 8-50hours of the worse pain humans can experience and possiblily mangling your sexual organs for life)

      I’m not sure if you’re wilfully stupid or actively trolling, but trying to reduce rape to physical friction is not a mistake anyone can honestly make, and probably not one one anyone can make without purposeful malice.

      Some reasons which I hope are superflous beyond pointing out the obvious:

      1. Being physically held in place, unable to defend oneself or escape, overpowered, helpless, violently subjected to another’s mercy, is (extremely) unpleasant in itself. -You would not enjoy 20 minutes wrestling with alexander karelin, sprung on you at a moment’s notice, even if you knew for a fact that this was just a prank bro.

      2. Nor you would not enjoy being held over a cliff edge for 20 minutes by someone who is not sure themselves of what they’ll do. Being subject to someone’s willingness to do crazy violent things, (and in an immediate position to be killed) is again something most people do not hope to experience very often and preferably never.

      3. Do you think Swans and wolves mate for life because of shared witticisms over the dinner table? — As contemptible as you might find it, humans and especially women have a tendency to be soft cuddly creatures who easily form attachments, and in particular have “evolved” over millions of years to secure a protector for their offspring, and not to reject said disgusting little creatures who put them to so much trouble. (for a crude analogy, imagine you had been programmed with a button designed to make you worship josef fritzl, -would you enjoy being tied up and having it pressed for 20 minutes? -whether or not you can override/ignore the programming, or it has any hold?)

      4. Beyond which, society/people actively believe in the control/sanctity of their “private” (hint) parts. Sensitivity about such matters, beyond being natural is something we have affirmed and elevated, one of the things distinguishing us from creatures who we have the right to kill, eat, and wear the skin of, — i.e. not just beyond reproach but so far beyond reproach as to make you a blasphemer, a preacher of abominations.

      5. Nobody wants to kill a baby, and most people are at least pretty uneasy about killing a foetus. there is no corresponding duty, meaning, or advantage to getting raped.

      6. usually when you have a baby you have some control over the matter. Abortion limits just shorter the timeframe, there’s always going to be a point past which the average women would not consider killing a baby. Even a total abortion ban (whatever its merits) would not leave people without agency or avenues of control.

      • Luke the CIA Stooge says:

        I think you’ve purposefully ignored the comparison. Which is being raped to being forced to go through childbirth.
        Which most people would acknowledge as horrifingly worse than rape if it involved someone impregnating you against your will in a way that didn’t include rape (say an OBGYN sneaking it in their while your under).

        The issue is prochoice advocats think denying access to abortions is the equivalent of forced childbirth (ie worse than rape) while prolife advocats generally don’t. (atleast according to their respective rhetorics)

        • carvenvisage says:

          I think you’ve purposelfully ignored the comparison.

          You can think whatever you like (“purposefully” lol, not enough to say I ignore what I hyperfocus on like a squirrel PCP), but it’s already clear that you make a habit of adopting “thoughts” more to anger those you oppose than based on what is useful or true, so you don’t need to announce it a second time.

          Which is being raped to being forced to go through childbirth.

          The comparison was rape as specifically reduced to “it hurts for 20 minutes and then it’s over”, to being denied abortion, which I spent the 100.0%-entirety of my post addressing. (Do I need to do a detailed breakdown or is reminding you the of contradiction between “ignored the comparison” and 1/3rd of my numbered points being devoted exclusively and not just implicitly enough?)

          Which most people would acknowledge as horrifingly worse than rape if it involved someone impregnating you against your will in a way that didn’t include rape (say an OBGYN sneaking it in their while your under).

          The comparison was to being “forced to carry to term” by not offering abortions (a mainstream-ish worry of the left), not to secret artificial insemination (bizarre idea you’re just inventing that has no relevance to mainstream leftist concerns) or straight up corner-case raping someone while they sleep. (whichever newly introduced idea it was you meant by “sneaking it in”)

          And no it isn’t “horrifyingly worse” because this is not an engineering problem about raw physical stress applied to the body and still less about discomfort or inconvenience, as if people would rather get raped than do a 9 month enlistment on an oil rig.

          -Have you ever wondered why rape is treated in the same breath as murder despite the disparity in permanent physical damage done to the victim? I’m guessing the answer is no, as you don’t seem like you wonder about anything, just apply whatever schema can most badly mangle the facts while still providing a semblance of functionality, -but think of it as a problem to solve:

          Why do you think that could be? (hint: why is shoving a gun in someone’s face a serious crime when that directly does 0 permanent physical damage?) (answer: because of the implicit threat and the high potential psychological damage which can last a lot lot longer than 9 months)

          _

          The issue is prochoice advocates think denying access to abortions is the equivalent of forced childbirth (ie worse than rape) while prolife advocats generally don’t. (atleast according to their respective rhetorics)

          Sure, and insofar as some people do act as though it is on the same level, I think it’s an interesting point. -Except for the totally needless sideswipe about how rape ‘just hurts for a bit then it’s over’ like it’s a calf being branded or something.

  23. Eponymous says:

    Also, I’ll throw out a seemingly unpopular opinion that I’m not 100% convinced of, but at least think is plausible:

    At the end of the day, the truth of these allegations doesn’t matter much.

    Set aside all the rhetoric and framing, the “cultural moment” we’re in, etc. What will matter for the state of the world 50 years form now will be Kav’s rulings as a supreme court justice. Whether or not he did something bad when he was a drunk teen doesn’t really affect my judgement of how good or bad these rulings will be. Nor do I have a strong belief they will differ in any particular way from the replacement level justice Trump would have nominated instead of Kav.

    Admittedly, these allegations do lower my assessment of his character when he was 17. They don’t move the needle too much for me on his current character, particularly when weighed against his long judicial record.

    I also think that incentives created by how this process played out will have a negligible impact.

    • NoRandomWalk says:

      I’d like to push back on this (ideally without discouraging you from stating more unpopular opinions).

      Let’s table the societal benefits of appearing to increase the incentive to not sexually harass folks when you’re young and immature.

      Doesn’t it matter if he’s lying about it now?
      The Supreme Court is an anti-democratic body, with a historically valuable appearance of being non-partisan. Its credibility is extremely important so that its decisions are viewed as legitimate.

      I grant that his long judicial record may be more evidence (I have no idea) than his lying is, but layperson voters don’t have the bandwidth to evaluate his judicial record.

    • “unpopular opinion”

      Unpopular in expression but almost unanimous in belief. Virtually everyone on both sides cares more about what he is going to do in the future, than about what did or did not happen, which latter is a mere factual question about the past. But precisely for this reason, people believe he did or did not do it voluntarily, in order to bring about their desired ends. It is the same way people hold religious beliefs because they are needed in order to continue living the lives they desire, not because they care whether the things are true or false.

      • Eponymous says:

        I can readily believe that most political operatives don’t really care one way or the other, and are engaging in performative outrage for political purposes. But I think a lot of regular people are really upset about the allegations (one way or the other), probably because it’s a much more “regular human level” debate than his judicial philosophy.

        Then again I don’t really understand regular humans.

    • herbert herberson says:

      ultimately, the most important question answered by the truth of the matter isn’t whether Kavanaugh did sexual assaults in high school and college (particularly in light of the fact that it seems very likely that he didn’t do any since then, as there would have been more allegations and that they were relatively minor unless Swetnick is true); it’s whether or not he lied under oath about it to the entire country

      • Eponymous says:

        I think it would take a rare person to admit to the allegations, particularly if he thought denying them was the only way to preserve his SCOTUS appointment. So I don’t think we can conclude anything from his choosing to take this course, except that he isn’t among the very rare breed of people who would fess up to it (conditional on guilt of course).

        I agree that being caught in perjury would sink his nomination, and rightfully so since that violates The Rules. But I don’t think his lying greatly changes my estimation of how good or bad a justice he will be.

    • I also think that incentives created by how this process played out will have a negligible impact.

      That part I disagree with. If the result of how this played out is that potential nominees in the future expect to go through what Kavanaugh went through, a significant number of them, arguably of the better ones, will decline. If I consider the two judges I know who I would be happiest to see on the court—as it happens both are too old to be likely candidates—I can easily imagine either of them doing so.

  24. keranih says:

    People generally agreed that if there was below a 25% chance the accusations were true, he should definitely be confirmed, and if there was above an 80% chance, he definitely shouldn’t be.

    That this agreement still exists is one of the very few hopeful things I’ve seen come out of this whole mess.

    • Eponymous says:

      Don’t read my comment just about yours then 😉

    • Tarpitz says:

      As a vaguely libertarian-leaning non-American with no great love for either the sexual politics of the activist left or Kavanaugh’s seeming pro-executive power inclinations, I like to think I don’t have much of a political dog in this fight. I find it very surprising that most people would not view a 25% probability of guilt (of a serious sexual assault) as disqualifying; my bar would be set much lower than that, and I didn’t think I would be atypical in that respect.

      • tscharf says:

        If 4 people were known for a fact to be at a party and it is known that one of those people did the deed to a blind folded victim, that is 25% probability +- 0% based on evidence. The current debate is something where the probability is closer to a random guess without evidence. One cannot qualitatively or intellectually discriminate an estimate of 5% and 95% in this case. Confidence of the estimate is missing.

      • Edward Scizorhands says:

        If the hiring committee had reliable information, somehow, that there was a 25% chance he was a rapist, they would not have selected him.

        A lot of the problem comes because the accusations were made publicly. That leaves a 75% chance that his life gets destroyed[1] unfairly.

        If I were on the selection committee in the White House back before Kavanaugh was named, knowing what I know now, I would tell them not to select him, because my confidence he did what Ford said is somewhere between 10% and 90%, and why bother when there are still many others? But the path to “just” have him “fail his job interview”[2] was lost somewhere between Ford’s and Swetnick’s accusations (inclusive).

        [1] He would not end up homeless, but the life he’s built up over the past 20-30 years gets undone.

        [2] If we buy this narrative, can we just say that Merrill Garland just didn’t get his job interview either?

    • Dan L says:

      I have a comment on this upthread, but after looking over it a bit more I’ll strengthen it: I’m afraid Scott (and you) are badly misinterpreting the data. “People” in this case isn’t a single category that is making a consensus judgement about what level of doubt they find acceptable. Instead, the line you’ve quoted is talking about two distinct groups at the edges of the distribution: one disbelieves the accusations and thinks Kavanaugh should definitely be confirmed (and is heavily Republican), and the other believes the accusations and thinks Kavanaugh should not be confirmed (and is heavily Democratic). There’s a strong correlation between the trio in each group, but I find the supposed causal direction dubious.

      In terms of counterfactual belief, it isn’t hard to find someone opposed to Kavanaugh’s confirmation on pure policy grounds regardless of the accusations. Conversely, it isn’t hard to find someone who says they would still support Kavanaugh even if they were certain Ford’s accusations were true. Pulling from a few polling nuggets, I estimate that if you add these groups together you’d get ~65-80% of American voters. I don’t find that particularly hopeful.

  25. tscharf says:

    A better questions is:

    Q: What is the probability that anybody else knows whether Kavanaugh is guilty? (besides the people involved)
    A: 0%

    It’s like asking which way a coin is going to flip from two highly loyal “team heads” and “team tails” tribes. I find the question asked to just be a proxy for how loyal one is to certain tribes. Given the high level of uncertainty involved everyone falls back to their prior biases (and why not?).

  26. NeedleFactory says:

    You “separated responses into bins from 0 – 9%, 10 – 19%, and so on to 90 – 100%”, introducing (as you remarked) a slight bias. I belatedly suggest: round every percentage to the nearest multiple of 10 unless it is exactly an odd multiple of 5, in which case round “in both directions” (i.e., award a half-vote to each of the two nearest multiples of 10).

    This reduces all estimates to the eleven numbers 0,10,20,…,100. Because people tend to give round numbers, these are the only unbiased values your data provides; start there.

  27. cmurdock says:

    And here are the results broken down by gender (blue is men, pink is women):

    This graph is kind of hard to read. Partly because of the colors-overlapping-effect, and partly because the “pink” and “blue” you picked both just look like slightly-different shades of mauve.

  28. JPNunez says:

    I think any allegation of Dem conspiracies to sabotage Kavanaugh’s nomination with invented rape accussations, has to answer why didn’t this happen to Gorsuch.

    Unless your theory is that Ford lived her life in wait for Kavanaugh to be nominated, which does not exclude that she was in wait because Brett actually sexually assaulted her.

    Now, yes, fake rape accussations do happen, but my prior for them is low. The part that’s insane here is that we have two men declaring to congress that they may have been the guys that assaulted Ford, in a weird perverse Spartacus ripoff. That…just does not happen normally? The worst part is that there are two. I mean, if I am a roman general, each additional I-am-Spartacus dude only lowers my estimates that I actually caught Spartacus.

    • Jaskologist says:

      My understanding is that false confessions actually do happen normally in high-profile cases. Remember, the question isn’t “would a normal person do this?”, it’s “would any person in the country do this?” and some non-zero bit of the country is literally certifiably insane.

    • Ghillie Dhu says:

      why didn’t this happen to Gorsuch

      Timing constraints & the then-extant filibuster for SCOTUS nominees.

      Gorsuch was nominated years before the next election, so the marginal benefit of dragging the process out a little longer was much lower; also, he was filibustered (the minority could still do that at the time) so there wasn’t an obvious need for a Plan B.

    • Edward Scizorhands says:

      has to answer why didn’t this happen to Gorsuch

      1. There were accusations against Gorsuch that he was sexist and such things. The idea that he got through without controversy is re-invented history. Democrats were angry about Merrill Garland (and in a way I completely understand).

      2. There were people showing up in red cowls to Kavanaugh’s hearing even before Ford came forward, because this was the end of Roe or something. While I give low chances to Ford knowingly lying, when your own side screams its head off that the sky is falling, you don’t get to act surprised later that people accuse you of freaking out. Own the circus, own the monkeys.

      • JPNunez says:

        Wait.

        Are you pretending that repealling Roe v Wade, or at least limiting abortion as much as possible, is not an objective of the republicans?

        • Edward Scizorhands says:

          Are you pretending

          Cool story, bro.

          Repealing Roe has just as much to do with The Handmaid’s Tale as whatever the Republicans plan is.

          Do your crazy cosplay all you want, it’s a free country. But don’t come back a few weeks later and insist that you[1] couldn’t have done something irrational because it wouldn’t make sense and would be crazy and oh geeze why would we ever be crazy?

          Anyway, it was pretty obvious that the reason people were so much more freaked out over Kavanaugh replacing Kennedy instead of Gorsuch replacing Scalia was that Kennedy was the swing vote instead of keeping the levels the same. Which perfectly answers why they were freaked out so much more about Kavanaugh, even before Kavanaugh’s name was announced. Which perfectly answers anyone feigning innocence “well shucks, it don’t make no sense to only freak out now and not before over Gorsuch.”

          [1] the generic you

          • Michael Watts says:

            There is another element — nominating a replacement for Scalia was 100% expected. Scalia/Garland was major news in every media channel, what happened with Garland was a big source of outrage, and getting the SCOTUS pick was a primary element in many people’s 2016 votes. That Trump nominated someone to the slot after winning did not come as a surprise.

            Nominating a replacement for Kennedy was a surprise, and people feel much worse about those.

    • Conrad Honcho says:

      I think any allegation of Dem conspiracies to sabotage Kavanaugh’s nomination with invented rape accussations, has to answer why didn’t this happen to Gorsuch.

      In addition to what others have said, #MeToo didn’t exist during Gorsuch. What social capital #MeToo had on the right has been completely burned in the Kavanaugh fiasco, however.

      • JPNunez says:

        Eh.

        It could have easily been the case that started the MeToo movement. It’s not like the allegations against Clarence Thomas went unnoticed in the feminist movement.

        • Conrad Honcho says:

          I don’t think so. #MeToo was started by exposing literal serial rapist Harvey Weinstein and the conspiracy of silence around him with scores of witnesses, evidence, and criminal charges. It would never have been launched by 36 year old accusations of groping/attempted assault at a high school party with no corroboration. The only reason anyone took that seriously was because of the social capital accumulated by #MeToo.

          • JPNunez says:

            Even if this counterfactual was true, does not mean that Ford would not have come forward without the Weinstein case.

            Anita Hill came forward literally decades before #MeToo. Ford talked to her psychiatrist years before #MeToo. These two datapoints cannot be explained away by saying that only #MeToo gave Ford and the Democrats a way to attack Kavanaugh.

          • Conrad Honcho says:

            I’m not saying she wouldn’t have come forward. Maybe she would have. I’m saying it wouldn’t have ground the government and media to a halt without #MeToo. I’m not saying “no one would have cared” but definitely far fewer people would have cared.

    • The original Mr. X says:

      The part that’s insane here is that we have two men declaring to congress that they may have been the guys that assaulted Ford, in a weird perverse Spartacus ripoff. That…just does not happen normally? The worst part is that there are two. I mean, if I am a roman general, each additional I-am-Spartacus dude only lowers my estimates that I actually caught Spartacus.

      To be fair, Ford did claim that there were two people involved (Kavanaugh + Judge), and it’s not clear from the reports I’ve seen whether both men are claiming to be the Kavanaugh here or whether one’s claiming to be the Kavanaugh and the other to be the Judge.

      • JPNunez says:

        I guess it is a possibility.

        • Conrad Honcho says:

          Ford’s yearbooks describe the girls at the school as rather promiscuous. She admits to attending drinking parties with boys at the age of 15. It’s entirely possible she had sexual encounters with other boys, which those men now pattern match to the events she described.

    • The Nybbler says:

      If you’re conspiracy minded, you can postulate that Ford (et al) intended to make this claim when President Romney nominated Kavanaugh. President Romney didn’t happen, but when Kavanaugh was later nominated, the setup from then was easy to build on. There wasn’t such a history with Gorsuch.

      • JPNunez says:

        You have to be v conspiracy minded for that, cause the prior should be that a woman wouldn’t want to expose herself to a lifetime of harrassment by crazy rightwingers for coming forward.

        Again, if a woman comes forward like that, chances are she is not lying -altho that does not say that she is right. I assign some small probability to her being wrong on Brett’s identity-.

        • The Nybbler says:

          Except it also gets her the adulation of those who would hail her as a hero for keeping a Roe-killing conservative off the court.

          Given that Ramirez and Swetnick (both of who seem rather likely to be lying, especially Swetnick) came forward, I would not assign such a small probability to a liar coming forward.

        • Again, if a woman comes forward like that, chances are she is not lying

          True if there is only one possible woman who can do it. If there are a couple of hundred, each of whom has a .01 chance of coming forward with a lie, the odds that one will do so are high.

          • xq says:

            I think “a large majority of accusations of sexual misconduct are true, even in politics” fits the evidence pretty well, once you exclude crazy people and similar.

            Of the four last presidents, two have multiple plausible accusations against them and two have none. Simply being widely known and hated by millions of people isn’t enough.

            Aside from the presidents, high-profile accusations of sexual misconduct seem pretty rare. For example, there are about seven competitive senate races in 2018. Roy Moore proves that allegations of sexual assault can be decisive, yet as far as I am aware zero of these races involve such allegations. And control of the senate is much more valuable than forcing the Republicans to replace Kavanaugh with an ideologically identical judge. There are also probably 70 competitive congressional races; do any of them involve such allegations? None of this seems consistent with a .01 chance for each women who knows a Republican candidate in an important and undecided race to lie about sexual misconduct.

            The number of unproven allegations in politics also doesn’t seem unexpected relative to the number of unambiguous cases; nor relative to areas that draw from a similar population like media, law, or entertainment.

          • JPNunez says:

            While each additional woman increases the probability one of them is lying, we gotta remember that they also increase the probability at least one of them is telling the truth. On top of that, serial harrassers/rapists will normally leave a bunch of victims behind -hence the serial part-, which is why sometimes we get Bill Cosby getting accussed by dozens of women at once.

          • Jaskologist says:

            While we’re discussing probability, are there numbers on how many rapists/harassers aren’t serial? My gut feeling is that the kind of person who does this to one person does it to a lot of people.

            Which would also imply that having only one victim come forward is evidence against the allegation if no other accusers pop up.

  29. J Mann says:

    Just as a data point, is it now common belief that Bill Clinton is very likely to be a rapist? (As a review, Juanita Broadrick’s evidence is that Clinton violently abused and raped her over her protests, she can identify where and when she was at the time of the alleged attack, the Clinton team has always refused to release Bill Clinton’s schedule for that day, and several people say that Broadrick informed them of the rape shortly after it occurred.)

    If so, it’s kind of extraordinary. Bill is on the Board of the Clinton Foundation, Hillary said during the campaign that she doesn’t believe Juanita without consequence, Bill spoke at the 2016 Democratic Convention, he spoke at Aretha Franklin’s funeral, he made bank giving the Keynote speech at the Ripple conference, he and Hillary are currently headed out for a national speaking tour, etc.

    I get that being a public intellectual isn’t the same thing as being a Supreme Court Justice, but if I were >50% convinced that Kavanaugh drunkenly attacked a woman at 17, I would be at least as convinced that Bill Clinton violently and soberly raped a woman as an adult and covered it up for years, and he doesn’t seem to be suffering any consequence.

    • Marklouis says:

      Good observation but it goes beyond “public intellectual.” The American Left was basically fine with the idea of having Bill Clinton as First-Man where he undoubtedly would have had influence on policy. Truth is, we only care about past misdeeds when they were committed by the other tribe. Let’s just admit it and move on instead of pretending we are defending some ill-defined notion of values.

    • Teddy Kennedy is another example of the pattern, although that was longer ago. There isn’t any evidence that he raped anyone, but it is undisputed that he drove a car containing himself and a woman off a bridge, succeeded in getting out himself, did not get her out, then went home and went to bed without telling anyone that there was a dead or dying woman in his car under water.

      He not only went on to have a successful political career, he was viewed as a leading figure in the progressive movement of the day.

      • J Mann says:

        It’s definitely true that the rules were different when Kennedy was operating. In addition to Chappaquiddick, I think the Kennedy-Dodd “waitress sandwich” story is generally accepted to be true.

        I guess my question on Clinton is that people are not acting like they believe today that it’s very likely he’s a violent rapist, nor did they act that way during Hillary’s campaign. Do they actually believe it?

        • Edward Scizorhands says:

          With Clinton or Trump or Roy Moore, you have a lot more evidence than you have with Kavanaugh. You have multiple accusers who made reports of the incidents to other people near the time it happened, which helps when dealing with something as fickle as memory.

        • Marklouis says:

          Ted Kennedy was most recently elected to the Senate in 2006. That doesn’t seem like a different era with “different rules.” What we are debating is how much allegations of long-ago criminal actions matter. In the case of Ted Kennedy, the answer is very little.

          The Kavanaugh situation is about politics, not “values.”

        • Doctor Mist says:

          It’s definitely true that the rules were different when Kennedy was operating.

          Sexual rules, certainly.

          If some random Joe — or a Republican — had done what he did in the Chappaquiddick incident, he would not have fared nearly so well as Teddy did. The only real thing Kennedy lost was the Presidency, which he arguably might have taken away from Carter otherwise.

    • Evan Þ says:

      Bill Clinton engaged in overt sexual behavior, while President, with an intern working for him. That is undisputed, and by prevailing modern standards that constitutes sexual harassment at the very least.

    • JPNunez says:

      Bill Clinton is not a nominee for the supreme court, which is what we are discussing. Sure, some people would have preferred he had a lower profile during the 2016 election, but certainly he could have been higher profile. He was not really a political candidate.

      The resignation of Al Franken seems like a much more closer parallel. Franken quit after pressure from both sides.

      Nice try.

      • Conrad Honcho says:

        But with Franken there were photographs.

      • J Mann says:

        Is “nice try” a reply to me? I promise you I asked the question because I’m seriously interested in the answer, because I’m curious how people think about this issue. I am a little sad that I didn’t get people to engage the question more directly, and will think about if there’s a better way to phrase it.

        Let me put it to you – do you believe that Bill Clinton held a woman down as she begged him to stop, violently raped her, said “you better put some ice on that” and left, that he groped Kathleen Willey against her will, and that he had state troopers bring a state employee to his hotel room where he exposed his junk to her and asked her to kiss it? If you do think that’s likely, should he suffer some consequence, or is it OK that he’s a highly paid public speaker and respected public figure? Should Hillary suffer some consequence for believing that he’s innocent?

        • JPNunez says:

          Oh Bill Clinton is def a rapist.

          But I don’t know why this is relevant given that he is giving talks, writing a book, attending funerals, etc. You are free to boycott those.

          If he was a candidate to congress, or a nominee to the SC, it’d be a thing. I don’t doubt that the democratic party can still extract value from Bill’s charisma, but they are careful on not positioning him where his past can be exploited by the opposition. If you are a republican, you can get a lot more mileage from thrashing Hillary for whatever thing -including defending Bill- than attacking Bill himself.

          Which is why I brought up Franken. He was accussed, and the establishment promptly kicked him out.

          Yes, yes, there was more proof against Franken. Maybe if the FBI had not carried out a joke, very limited investigation of the Kavanaugh issue, they’d find something -probably more witnesses-. Or perhaps not. But we won’t know now, and thus I default at believing him guilty because there was a concerted effort to make the investigation limited.

        • brmic says:

          Yes, I believe Bill Clinton is a rapist.
          He should suffer some consequence and I’d boycott him. That said, while I disagree with the moral calculus of those who invite him, I don’t fault them personally.
          As to Hillary: No. I base this on my reaction to Cosby’s wife who ‘did not see’ much worse crimes and many more of them. And yet, even there I have a hard time condeming her, though she can be seen as aiding Cosby’s crimes. But I don’t know what went on inside their marriage and I think the extent to which people can lie to themselves if they’re in love is so massive that I wouldn’t want to blame the spouse without evidence of co-conspiracy.

        • J Mann says:

          Thanks @JPNunez and @brmic – it seems to me as if people aren’t acting like he’s a rapist, so I was curious about what’s going on there.

          (@JPNunez – I hadn’t noticed it before, but I like your avatar)

  30. Eliezer Yudkowsky says:

    Always give the median probability in polls and surveys, not the mean probability; and tell people in advance that you’ll do so.

    Median probability means that people don’t have an incentive to give more extreme responses than their true probability in order to skew the mean. Every response to a side of the median shifts that median the same amount. It’s also more politically informative because of the Median Voter Theorem. And it more closely resembles a prediction-market process, where people bet to a side of the market price rather than stating their true prices. (Albeit in a market, people can bet more to shift the price more when they think the price is further off true. But each dollar is still shifting the “median price”.)

    (I don’t know if research has been done on the accuracy of median vs. mean probabilities when dealing with nonsuperforecasters, but I testably predict more accurate median probabilities than mean probabilities when aggregating groups of nonsuperforecasters.)

    • J Mann says:

      You have a lot higher opinion of poll respondents’ numeracy than I do. 🙂

    • Scott Alexander says:

      Median was 50%.

      • alphago says:

        Median was 50%.

        I just downloaded the data, but I calculated 60% for the median probability that he’s guilty?

        Perhaps I’m misunderstanding something about how the data is labeled, but I just took the median of column H (“Probability”); I excluded the null columns from the calculation. On the other hand, taking the mean did match your 52.64% figure.

      • quanta413 says:

        Sure, but isn’t Eliezer’s point that if you don’t report anything but the median, then people have less incentive to skew the results by picking high or low values?

        Eliezer, please correct me if I’m misunderstanding.

      • NoRandomWalk says:

        Does anyone actually think people respond to polling trying to skew the results?

        Even when I feel strongly and know I’m outside of the mainstream on something, ex consensus is 50 and I’m at 20, I don’t say I’m at 0 to ‘get the average closer to my position’.

  31. Sebastian_H says:

    This is hard for me because the question covers too large of a range. Also I don’t believe I can adequately distinguish at a rate much better than 5%, though I can distinguish between 5% and vanishingly small.

    Do I think that NOTHING happened to Ford and she is all just making it up? No. I’d put that in the <5% chance.

    Do I think that SOMETHING, which Ford interpreted as traumatic, happened to Ford? That's the flip side of the above question. I'd put it at least 95%.

    So then, is she wrong that Kavanaugh was involved in this event? She apparently knew him, and mistaken identity among people you know is a lot lower than people just random eyewitness problems. I'd put that at 5%. And really only because I think she might have been drunk. She says she wasn't, but that is the kind of self-protective lie you tell yourself that may be totally convincing 35 years later.

    Given all the testimony, my best educated guess about what happened is the following (I'd put it in the 50-55% range):

    Brett and Mark were drunk, and drunker than they admit to. At least one of them thought she was willing at first. When it became very clear to them through their drunkenness that she wasn't, they let her go.

    If that is what I believed happened, how do I answer the question? That isn't attempted rape (the most common charge talked about). Thinking that isn't the right interpretation is supported by the fact that in the escape part, they didn't try to keep her from leaving the room, and didn't try to get into the bathroom. They were well positioned for it to be more than just an attempt if that is what they wanted.

    In my main case interpretation, a lot of how Brett reacts to the attempted rape charge makes sense. He certainly doesn't code it as any such attempt. Also I think he is DRAMATICALLY underplaying his drinking (probability 70-75%) which makes there a possibility of blackout total non-remeberance even if he committed a rape attempt.

    So is my main interpretation of the events enough to stop the appointment? I'd tend to think no, BUT I think it makes the chances of it being an actual rape attempt HIGHER than whatever the base rate is. I don't really know what the base rate is, so I have trouble putting a number to that. But I'd say it puts it quite a bit higher than the base rate whatever that is (say 3X higher).

    I think Kavanaugh exhibits being in a prep school drinking culture which also might influence the base rate.

    I think Kavanaugh almost certainly was misleading about some of the interpretations of the side evidence stuff (the devil's triangle, the other yearbook stuff, etc.).

    All in all I'd tend to say that my main case alone shouldn't be enough to stop the confirmation, but that the surrounding stuff edges up the main case to worse cases, more than the mitigation takes it down. And the way he went into attack mode against Democrats in general toward the end may be within the range of understandable reactions, it makes it hard to believe he will be able to transcend that later, which is what we want from Supreme Court justices. So on balance he shouldn't have been confirmed even though I'm not sure he made a rape attempt (pretty sure) or other sexual assault (sort of sure).

    Given that, how should I have answered?

    Edi I’m convinced he’s lying about the impact of his drinking. Its an understandable lie, but it makes me code other stuff as more suspicious.

    • The Nybbler says:

      So then, is she wrong that Kavanaugh was involved in this event? She apparently knew him

      Who, besides herself, says she knew him? Ingham says she wasn’t at any party with Ford and Kavanaugh. So does PJ Smyth.

      If you can’t put them at a party together, all this speculation about what happened is building castles on air.

      I think Kavanaugh almost certainly was misleading about some of the interpretations of the side evidence stuff (the devil’s triangle, the other yearbook stuff, etc.).

      Devil’s Triangle as a drinking game has been confirmed. Same with “boofing” meaning farting.

      • Eugene Dawn says:

        Devil’s Triangle as a drinking game has been confirmed. Same with “boofing” meaning farting.

        You’ve said this a few times, but I haven’t seen anything about it; do you have a link?

        • Evan Þ says:

          Regarding “Devil’s Triangle,” see Fox or National Review, both reporting on “four high school classmates” of Kavanaugh’s. Current Affairs points out that “the evidence on this now conflicts” with Kavanaugh’s roommate Roche still having claimed it references sex.

          • The Nybbler says:

            What would a Yale guy know about Georgetown Prep slang?

          • J Mann says:

            He says he learned it from Georgetown Prep guys in his class. (FWIW, I learned Euchre at Boys’ State, so if anyone ever says it’s a sex act, let the record show it’s also a card game).

            IMHO, that’s not particularly important – the bigger deal, IMHO, is that one of the guys who says its a drinking game is listing in the same yearbook as Kavanaugh with the entry “Devil’s Triangle (Founder of the Term)” and a third guy is listed as “3, 4: Lost in Devil’s Triangle.” If the term was a well known term for a sex act, “Founder of the Term” wouldn’t make much sense, and “Lost in Devil’s Triangle” makes much more sense if it was understood to be a drinking game than a sex act. (That also sheds some light on how it got past the Yearbook Advisor, not that he seems to have been a strict dude).

          • The Nybbler says:

            I learned Euchre at Boys’ State, so if anyone ever says it’s a sex act, let the record show it’s also a card game

            On the other hand, based on what I learned as a youth, don’t believe anyone who tells you that 69 is just the product of 23 and 3. It is also Dolly Parton’s bust size (at the time).

        • The Nybbler says:

          Devil’s Triangle

          https://www.nationalreview.com/corner/kavanaugh-hearings-devils-triangle-drinking-game/

          “Boof”, from the 2004 book “Art of the Fart”

          https://archive.is/GTdYl

        • Glen Raphael says:

          @Eugene Dawn:
          Here’s CNN: Georgetown Prep alumni back up Kavanaugh’s testimony on ‘devil’s triangle’

          The last several paragraphs of the CNN article names six people who confirm the use:

          “‘Devil’s Triangle’ was a drinking game we came up with in high school,” they wrote. “It was a variation on the game ‘Quarters.'”

          The four, DeLancey Davis, Bernard McCarthy, Jr., Paul Murray and Matthew Quinn, said while they did not remember where the name came from, none of them used it in the yearbook “to refer to any kind of sexual activity.”

          “To us, it was just a game with glasses in the shape of a triangle,” they wrote. “If the phrase ‘Devil’s Triangle’ had any sexual meaning in the early 1980s, we did not know it.”

          Greg Aceto and Bill Van Pelt, IV, wrote to Grassley and Feinstein as well. They said in their first year at Boston College they lived with Quinn.

          “Matthew taught us a drinking game called ‘Devil’s Triangle’ that he had played with his friends in high school,” they wrote. “We did not understand ‘Devil’s Triangle’ to have any sexual meaning. It was simply a game that used cups or glasses of beer placed in the shape of a triangle.”

    • scmccarthy says:

      Very well said. I have a similar impression of the situation.

  32. Rusty says:

    It would be interesting to get your take on the reliability of memories. Does it make a difference when they are recovered under therapy, does in make a difference when they are of a really significant event, does it make a difference when they are from a long time ago. I felt persuaded Dr Ford was telling the truth in the sense of reporting what she remembered but really unsure about what I could infer from that.

    • At a tangent, what should we infer from the refusal to make the psychiatrist’s notes available to the senate investigators? I see four alternatives:

      1. The account in the notes is sufficiently inconsistent with Ford’s present account to cast doubt on the latter.

      2. The notes show other things about Ford that would make her look like a less reliable witness—for instance that she had told the psychiatrist about other traumatic events in her past which she and/or the psychiatrist later concluded were false.

      3. The notes show things about Ford irrelevant to her testimony but very embarrassing to her, for instance that she had had an extramarital affair.

      4. The psychiatrist sees it as an issue of principle never to betray the privacy of her notes on her patients.

      • Doctor Mist says:

        The psychiatrist sees it as an issue of principle never to betray the privacy of her notes on her patients.

        I wonder if Scott can comment on the likelihood of this, contingent on Ford requesting the release of these notes.

      • The original Mr. X says:

        Re: 4, presumably Ford herself could authorise the psychiatrist to release the notes.

        • Joyously says:

          “Portions” of the notes were shown to the Washington Post. She also said she’d give the notes to the FBI *if* the FBI agreed to interview her. Putting a condition on handing over evidence you believe to be coorborating is… odd.

      • Ketil says:

        5. It was couples therapy, so perhaps her husband objects?

        BTW, she doesn’t/didn’t seek help for anxiety, phobia, and PTSD? It looks a bit strange that she allegedly has struggled with this – as a psychologist herself, even – for thirty years, and then incidentally brings it up in couples therapy.

        With the ex claiming she never had any fear of flying in the nineties, and the doubts about installing the extra door, my maximum likelihood* estimate is that she acquired mental problems more recently, and through therapy or otherwise, have convinced herself that a sexual assault must be the reason. I.e. a classic false memory situation.

        The claims about the door, the fear of flying, and preparing for the polygraph test should be verifiable. Likewise, there should be hundreds or thousands of people who can corroborate of her having mental problems.

        *) Maybe something like:
        False memories: 60%
        Lying outright: 20%
        Kavanaugh did it: 20%

        (The latter can be broken into:
        consensual sexual encounter or fooling around that turned awkward/sour: 15%
        actual intent to sexually assault or rape: < 5%)

  33. quitelikelyblog says:

    As one of the Democrats who thought it was quite likely Kavanaugh assaulted Ford I have a bit of a tough time imagining what’s going through the heads of people who think it’s very unlikely he did so. I can understand thinking “well there’s not really any conclusive evidence” but how do you get from “Blasey Ford says she is 100% certain that she remember being attacked by Kavanaugh” to “he probably didn’t do it”?

    • Edward Scizorhands says:

      I’m not endorsing this POV, but you could get to “probably didn’t” if you push the chance of her knowing she’s lying up to around 25%, and then add another 40 or 50 percentage points to her being mistaken in her recollection. For her being mistaken, if I were steelmanning that I would point out that every eyewitness she says was there, including eyewitnesses hostile, neutral, and friendly towards her, don’t remember it. “Everyone else is misremembering but me” is easy to call nuts with just a little bit of uncharity.

      (If you think Republicans will forget all about this presumption of innocence ideal the next time any kind of criminal justice reform rolls around, I’m right there with you.)

    • J Mann says:

      I’m not super confident, and I don’t think any one of these things are dispositive, but here’s what gives me pause.

      1) I don’t think 30 year old memories are very reliable. People think they are, but my understanding is they change over time and with recollection, so if Ford had said she was 100% sure Kavanaugh threw the winning basket at such and such game and Kavanaugh said absolutely not, I’d put the probability at around 50/50.

      2) My understanding of so called “flashbulb” memories formed under trauma is that they seem much more clear and certain, but they’re equally susceptible to change, if not more so. So the trauma doesn’t make Ford’s memory more susceptible to error than any other memory a 50 year old might have of their sophomore year in high school, but it also doesn’t make it any less susceptible to error – it just makes it feel error proof.

      3) There are a lot of weird details that don’t disqualify Ford’s story individually, but overall make me uncomfortable. In no particular order:

      a) Ford has very specific memories of some non-traumatic details (that she definitely only had one beer, that she wasn’t on medication, that Smythe and Keyser were there but Garrett definitely wasn’t, etc), but no memory of other details, like whose house it was, where the house was, or how she got home. How she got home is a particularly weird detail – the party was as much as a two hour walk from her house, she was too young to drive, she didn’t have a cell phone, and she’s sure she left by herself without any of the other participants. I would normally think that would be memorable. No one has come forward to say they remember giving her a ride. (Although I guess it’s possible that she just walked back to the Country Club or a pay phone and called a cab, depending on where the house was?)

      b) Keyser says she has no recollection of the party or of ever meeting Kavanaugh. Keyser’s FBI interview is confidential, but it apparently doesn’t have anything that Dems could lever or that swayed Flake or Collins, nor did Murkowski refer to the investigation in justifying her vote (as far as I know). Despite being lobbied by Ford’s friend, Keyser has never come out and said something like “I went to a number of parties with Christine, and this easily could have happened.” Normally you would think that these events – going to a party with her friend and then having Ford flee without explanation or getting a ride from anyone, ditching Keyser at a party where she apparently didn’t know anyone* – would be more memorable than an average party.

      * If Keyser knew Judge or Smythe, you would normally think she’d say so, but maybe not.

      c) Smythe says that he has no recollection of the party. (I don’t put much probative value on this one – I’d love to know what he told FBI investigators about what he did remember – but it is one more thing).

      d) Judge, Smythe and Keyser all spoke to the FBI under oath, and apparently didn’t shift the narrative.

      e) None of Ford’s brothers, mother or father have said they believe her story. It’s completely possible that they’re all assholes, or that the family is estranged for other reasons, or it’s possible they have reason to doubt it.

      f) Kavanaugh was correct that he did document his schedule fairly comprehensively in his calendars. Ford’s team has apparently ruled out the July 1 date as a possibility, so we’re left with the idea that the gathering occurred on one of the days when Kavenaugh’s calendar shows him to be in town and not otherwise occupied. Given that Ford states that he and Judge had been pre-drinking and they had a full schedule of summer football practices, the gathering is possible, but the calendar and football schedule presents some challenges.

      g) Kavenaugh has a lot of character references, including numerous women who knew him in high school and, IIRC, every actual girlfriend. He doesn’t appear to have gotten in any trouble of any kind after law school. I’m accustomed to seeing multiple accusers in these cases, and with the arguable exception of Ramirez, who is alleging different conduct and who wasn’t sure earlier this summer, we don’t have any. Again, not iron clad by any means, but when you read the character references, you wonder why people are willing to lie for him or how he was able to hide this side of his personality.

      h) None of the parent’s houses of any of the kids are plausible. I understand the current theory is that the party occurred at Judge’s grandmother’s house, but haven’t looked into it.

      i) My assumption is that the Washington Post, Ronan Farrow, Democratic opposition researchers, and Ford’s lawyers have had months to try to prove this up, and have not come up with anything beyond what we’ve heard, which is really thin. (Basically Ford, Ramirez, and people who saw Kavanaugh drinking).

      So to sum it up, I’m struggling. Without Kavanaugh’s denials, I’d assume that Ford was probably remembering things accurately, and vice versa, so I guess that I’m inclined to believe both of them in isolation. I really do believe that the current science is that (i) 30 year old memories are not as accurate as people think and (ii) traumatic memories seem clearer and more accurate than non-traumatic, but in fact are as error-prone as normal memories, so one way to square things is that at least one of them is remembering incorrectly.

      • DeservingPorcupine says:

        This is a good rundown, I think.

      • slightlylesshairyape says:

        I don’t get g). What in the world stops a person that would do such a thing in high school from not repeating that?

        That seems like a cartoon-villain straw man. It’s totally consistent that he did what Ford alleged and that at some point around college or law school he sobered up and stopped. In fact, I’d bet significant money that this is a highly typical outcome for sexual predators at the high-school/college level, especially in very affluent/connected places like Yale.

        • Edward Scizorhands says:

          If you believe that most sexual assaulters are serial assaulters, the fact that more of these reports haven’t come through is evidence against him being an assaulter.

          (It’s possible he was doing the thing kinda like Ford described, and realized afterwards that it was a shitty thing to do and mended his ways. If so, it sounds like something to remember and he should say he remembers it.)

          • slightlylesshairyape says:

            The claim isn’t that most assaulters are serial assaulters, but that most assaults are committed by serial assaulters. Both can easily be true at once.

            Anyway, it’s possible that he stopped for moral reasons. Or because he was afraid of getting caught. Or that he needed to stop drinking to focus on ever-harder studies (in this respect, the laughable difficulty level of freshman level classes is a major disservice) and the rest followed. Or any combination of the above.

        • J Mann says:

          I was impressed by this article, as well as by the lack of misconduct after freshman year at Yale. As I said below, it’s definitely possible that an offender straightened out, or that this was a one time slip up that he since regretted, but the lack of apparent subsequent misconduct does reduce my internal probability on this accusation.

          Similarly, the fact that lots of women who knew him then are willing to stick their neck into this crapstorm to testify to his good character as a high schooler speaks for him, and IMHO lowers the probability, at least somewhat.

        • Joyously says:

          I do not think that deliberate attempted rape is something that happens exactly once. That has always been one of the major reasons I think it is less than <50% likely this happened exactly as she described.

          There's twenty thousand people angrily declaring that they saw him sloppy drunk, that he passed out at parties, that he got loud and obnoxious. If it was anything like normal behavior for him to grope or force himself on girls when he was drunk we'd know by now.

          And like J Mann below I found this significant: “He was always there taking care of us,” she says. “I was a year younger, and he was like my big brother. He wouldn’t let any guys mess around with you. If anybody was drinking, he would be the one taking care of you. Not everyone in his friend group was like that, but he always was.”

      • Edward Scizorhands says:

        but no memory of other details, like whose house it was, where the house was, or how she got home. How she got home is a particularly weird detail – the party was as much as a two hour walk from her house, she was too young to drive, she didn’t have a cell phone, and she’s sure she left by herself without any of the other participants.

        In the past week or two, I’ve read an argument from a fair-minded woman (Ioffe? McArdle?) that had gone through some kind of assault, and they also describe remembering the specific incident a lot but completely losing the memory of the next several hours. It’s like some kind of memory-writing hormone in your brain gets used up all at once[1] when a traumatic event happens and takes a while to return.

        I’m very skeptical that you can remember 30 year old events with accuracy, but her not remembering this detail is almost expected.

        [1] I ain’t no psychiatrist.

        • J Mann says:

          Each of those details in my list is definitely possible, especially individually. It’s when you put them together that I get uncomfortable. In the case of the memories, it’s the combination that she’s certain of specific non-traumatic memories that help her case (definitely only one beer, definitely no medication), while not certain of something that seems like it would have been very memorable, and no one else recalls giving her a ride when she left the party. Definitely possible (especially if she had what we used to call “mad money” and was able to hail a cab), but a little eyebrow raising.

        • Joyously says:

          It’s like some kind of memory-writing hormone in your brain gets used up all at once[1] when a traumatic event happens and takes a while to return.

          I believe you’re thinking of Megan McCardle, btw–I read her recently about how her boyfriend hit her, and now she can remember the moment very clearly but not what they were arguing about or a lot of what happened after until she told her roommate the next morning. But I don’t think it’s that the traumatic event made it *more* likely to forget other details, just that you forget those details like you forget everything else.

          I got hit by a car about eight years back and I can remember some things but not others. Wouldn’t be able to say what day of the week it happened, for example, or what kind of car it was. And there are other details which I’m pretty sure I remember (like what street corner it happened on and the sweater I was wearing) which for all I know I could be wrong about.

          The lesson I think isn’t that forgetting stuff like that makes it likely that you’re lying, just that people forget things, even traumatic things, and memories change, and the longer it’s been the less reliable memories are and it’s not *fair* that this makes it harder to know for sure what happened back then but it’s true.

          (I also have at least one traumatic memory which I’m almost certain never happened at all, by the way.)

      • Joyously says:

        This is where I am, and why I’ve come down at <50% likelihood of guilt, especially b and g and i, and to some extent e and f.

      • Ketil says:

        I’ll add:

        j) the testimony of the ex-boyfriend, who claims she didn’t suffer PTSD or anxiety in the 90s, that she helped her friend, who was applying for a job with the FBI, practice for a polygraph, and that she never mentioned any assault.

        (I see that the friend now denies the part about the polygraph, so perhaps the ex isn’t much value as a witness on the other things either:
        https://www.washingtonexaminer.com/news/longtime-friend-denies-ford-helped-her-prepare-for-polygraph-exam)

        k) allegations that the extra door was installed (in 2008) not because of her anxiety (which was first revealed in 2012), but in order to rent out part of the house.

      • tscharf says:

        Another point is that I can believe one prep school boy might be inherently aggressive and capable of rape. What we supposedly have here is two prep school boys who are theoretically brought up responsibility, have bright futures ahead, and decide as a team to forcibly rape a 15 year old girl they don’t know at a public party with other people they know there. I very much doubt this, although it is not impossible. Every case where this “preppy white men are evil gang rapers” is alleged it falls apart (Duke, UVA). I don’t know, but my guess is the instances of forcible rape by virgins is pretty rare.

        Much more likely if it really happened was drunk boys fooling around and confusion on intent and the details.

    • sandoratthezoo says:

      I am very agnostic about whether Kavanaugh did this, so I’m not precisely who you’re talking to. But I think that the answer for how you get to “very likely that he did not do it” is:

      1. You say, “Nobody can be 100% certain of any memory for 35 years ago without corroborating evidence.”
      2. You say, “There is immense desire for people beyond a certain level of left-politics to find something to sink Kavanaugh’s nomination, such that some people will either lie to the public or create memories for themselves to try to sink it.”
      3. You say, “If it were true, there would be stronger corroborating evidence. The lack of found corroborating evidence is evidence that it is not true.”

      And that’s basically it.

      Something I’ve found personally helpful in this hyper-partisan world we live in is to remember that what I think about difficult-to-be-certain-about issues has literally 0 meaning. The world would not be any different if I were convinced of Kavanaugh’s guilt or innocence.

    • The Nybbler says:

      Here’s the Republican lawyer’s case against the accusations:

      https://www.documentcloud.org/documents/4952147-180930-Rachel-Mitchell-Memo-Kavanaugh.html

      Basically, there’s no corroborating evidence of the central claims, there’s no corroborating evidence of Ford and Kavanaugh even being at the same party, and there _should have been_ corroborating evidence of the latter (the testimony of Keyser (Ingham) and Smyth, neither of whom remember any such party). Add that to the inconsistencies in her story, and Kavanaugh’s notebooks (which neither mention such a party, and no the June 1 1982 one is not it, nor have suspicious holes), and I think that not only is there not a preponderance of the evidence making Ford’s case, but there’s clear and convincing evidence _against_ it.

    • Ketil says:

      I think her story doesn’t add up much (i.e. it doesn’t fit my priors). The accusation is assault with intent to rape. Yeah, so the plan is that K. holds her down on the bed, while J. ….jumps repeatedly onto the bed until they all tumble down onto the floor and she escapes? While they laugh? And two 17 year olds, intent on raping a 15 year old girl, are then unable to hold her back?

      If the story had been that she and K went to the bedroom to make out, and J comes barging in and she gets cold feet, tries to wriggle free, and K resists her struggles, I’d find it much more credible.

      The 100% comment is one thing that actually weakens her story. Nobody can be 100% certain of memories from last week, much less thirty-odd years ago. And she’s a psychologist, surely she must be aware of this. Perhaps I’m wrong (but I’ve seen several papers supporting my view), perhaps she’s just ignorant in spite of her work and education – but there is some probability she is being deliberately deceitful.

    • Conrad Honcho says:

      I started with something approximating an open mind, that perhaps both are telling the truth as they remember it. He denies it, and this sort of behavior does not at all mesh with the behavior all of his character witnesses attest to. Drunk, yes, handsy, no. So perhaps she’s innocently mistaken, or he was blackout drunk.

      I eventually settled on “she’s deliberately lying.” My reasoning is as follows:

      1) I find it very hard to believe she does not remember how she got home. She states she ran out of the house without speaking to anyone, so no one from the party took her home. She did not have a cell phone. There are not pay phones in residential areas. If the party was with within a mile of the place she says it was near (the country club) then it would be 6-8 miles from her home. I understand trauma can make you forgot lots of details surrounding the trauma, but the experience getting home would almost certainly also be traumatic. Walking home for 1.5-2 hours in the dark, terrified the rapists are coming after you. Running to the neighbor’s house and begging to use the phone, and waiting there crying for mom and dad to arrive. One would remember these events. Yes, people forget what happens after the danger/scary event is over, but the danger and scary event did not end when she left the house.

      2) While watching her testimony, I thought her body language looked like that of a liar. Slight curling of the lips, throwing the head back, laughing at points while describing very bad experiences. It’s called “the duper’s delight.” The joy liars get when they believe they’re the smartest person in the room, getting one over on all these rubes.

      3) She told several other verifiable lies, and as Da Nang Dick Blumenthal reminded us, falsus in uno, falsus in omnibus.

      3a) She said they needed to delay the hearing because her fear of flying required she drive across the country. Whether her fear of flying exists or not, she’s clearly able to get over it on a regular basis (she travels often for work/vacation), so it is not true that she could not fly to the hearing. Also, the letter her ex-boyfriend sent to the SJC says they dated for 8 years and she never mentioned any fear of flying, and they even rode in small planes together. He could be lying, though.

      3b) She claimed the reason she went to the therapist in 2012 with her husband was over conflict around getting the second front door installed. But there’s nothing about the story she told that has anything to do with lack of a front door exit. She easily fled out the front door. Does her house not have a back door? And the permit for the 2nd entrance was pulled in 2008, not 2012, and there are photos from 2011 showing the door already installed. And the door was used as a second entrance for renters. I wonder if whatever space the second door opens into has privacy locks to prevent casual entrance from the main living area. Might make it difficult to flee from rapists through that space. And if you’re scared of rapists, why invite strange young men to live in your house?

      3c) Her ex-boyfriend swears he witnessed Ford coaching her “beach friend” former FBI official Monica McLean on taking polygraph exams, but Ford testified she never advised anyone about polygraphs. It’s possible he’s lying or that she forgot, though. McLean swears she didn’t receive such coaching, but McLean also has a credibility problem (see 4).

      3d) Ford said she contacted her representative because she had no idea how to contact the Senate or the White House to warn them about Kavanaugh. But her lifelong friend McLean (who she was with when later composing the letter to Feinstein, and who also attended her SJC hearing) was a career FBI official, a Public Information Officer with the NY FBI under Preet Bharara, who worked with Chuck Schumer. Ford is surely aware that her lifelong friend who she vacations with is just one or two personal phone calls away from the Senate Judiciary Committee.

      4) The FBI has texts that show McLean attempted to tamper with the testimony of Leland Keyser, pressuring her to tell a story more favorable to Ford. I assume they got these texts from Keyser, during the interview. This does not seem like truth-seeking behavior to me. Also, McLean released another statement through her lawyer denying this. This seems like a pretty obvious thing to be able to prove. And McLean’s statement went not to the FBI or the Senate (under penalty of felony) but to the media. If McLean’s lying about this…

      5) Speaking of McLean, her attorney is David Laufman, a former DOJ National Security official who was present when Peter Strozk was interviewing Hillary Clinton. One of Ford’s attorneys is Michael Bromwich, who represented Andrew McCabe. Ford herself may not be a political operative, but everyone around her is running in the same circles as the lawfare division of The Resistance.

      6) So what motivation does she have to lie? Well, if you believe the media or the left (but I repeat myself) Trump is Hitler, and he’s appointing a Supreme Court Justice who is also Hitler, and they’re going to Double Hitler everyone. What’s a little lie to stop Double Hitler? And she gets the fame and the adulation of all the sorts of people she cares about across the nation and in Palo Alto (cover of Time Magazine!). And her million dollar GoFundMe payout was predictable. Bromwich would certainly know that given the money McCabe received.

      7) The Resistance is a broad group of people who oppose anything Trump does through any means available. Some of them harangue their friends on FaceBook. Some run conservatives out of restaurants. Some steal papers off the President’s desk and slow-walk his policy initiatives. Some riot in the streets and make death threats. Some open fire with a rifle at Republican baseball practices. “Lying about rape” is certainly not beyond the pale here.

      So, with all of that, I would very much like to see an FBI investigation into McLean and Ford and the process by which these accusations came to be. I’m sure they will welcome an FBI investigation to clear their good names.

      • Jaskologist says:

        One of the reasons not knowing how she got home is especially suspicious:

        In the early version of the story Ford told her therapist, she was in her late teens when the attack took place (fits with what she originally told the Washington Post about this taking place in the mid 80s). In that timeline, there’s little mystery about how she could have gotten home: she could have driven herself, which would not be a memorable thing.

        The trouble is that in this timeline, Kavanaugh is away at college by that point. So to implicate him, the date needs to be moved earlier, which also places it before she can drive. And that’s what opens up the hole about transportation, because she either has a harrowing multi-hour walk home alone in the dark, or tearfully begging a stranger to borrow their phone to call someone for a ride, which means at least 2 potential people who would be able to corroborate elements of the story. People she never provided.

        • Conrad Honcho says:

          Yes, that was my thinking as well. That is, if she was describing an event that did occur, but substituting Kavanaugh for her actual assailant.

          It is convenient that her story is maximally unfalsifiable. But I didn’t want to get into that as to reasoning I believe it’s a constructed story, because then you go down the “so I can clearly not choose the wine in front of me!” rabbit hole.

      • J Mann says:

        I think I’m broadly pro-Kavanaugh, and see your point, but am going to challenge you a little on some of the particulars.

        1) Second door: I understood Ford to be saying not that she entered counseling as a result of the second door, but that the second door was something that came up during counseling:

        I had never told the details to anyone — the specific details — until May 2012, during a couples counseling session.

        The reason this came up in counseling is that my husband and I had completed a very extensive, very long remodel of our home and I insisted on a second front door, an idea that he and others disagreed with and could not understand.

        In explaining why I wanted a second front door, I began to describe the assault in detail.

        I think that’s consistent with “we applied for a general remodel in 08 and completed it by 10, and the remodel was one of many things we talked about in counseling.” (On the other hand, if they actually got the door so they could sublet a portion of the house and if it didn’t actually provide escape opportunities, then her testimony was misleading, but I don’t think we know that for sure.)

        2) Re: the trip home. I’d also expect that to be memorable, but it’s not quite as ridiculous as people make it sound. My recollection of suburban life during the 80s is that everybody was supposed to have $20 hidden somewhere in case their driver was drunk, they got ditched, a date got aggressive and they needed to walk, etc. (Sometimes called “mad money.”) So she could have walked to a gas station, used a pay phone and called a cab. Alternately, if we assume the party was at Mark Judge’s grandmother’s house, that’s a 15 minute walk back to the Country Club.

        3) Some of the other wiggling during testimony – the idea that she wanted to contact the President’s staff and Senate, but instead called Eshoo and the Post because she did not “specifically know” how to contact the Senate, or the idea that her lawyers suggested she didn’t like to fly because she was hoping the Senate would conduct a hearing in Palo Alto which she would have hosted – strikes me as implausible but sort of normal witness testimony.

    • I have a bit of a tough time imagining what’s going through the heads of people who think it’s very unlikely he did so.

      I actually think there is a fair chance he is guilty and a fair chance he isn’t and she is deliberately lying. The strongest argument for the latter is the unintuitive one.

      Suppose we start out by assuming that for a woman like Ford to invent such a story is very unlikely, that even given the political incentive to do it there is only a five percent chance she would. Further suppose there are two thousand women who could have been victims of the sort of thing she says she was a victim–right age range, lived somewhere making it plausible that they attended a party with Kavanaugh when he was in high school or at Yale. Assume half of them have political views that would make them, like Ford, very much want to stop Kavanaugh. Assume that a tenth of those can produce as much soft evidence as Ford (in her case, that she told her psychiatrist that she had once been the victim of an attempted rape–with details unavailable because the psychiatrist won’t share her notes—but one can imagine lots of other things of comparable strength).

      That gets us to a hundred women, each of whom has a .05 chance of making such a false accusation. The probability that none will is .95^100=.006. So the chance that at least one such false accusation will be made is 99.4%.

      I’ve made up the numbers, of course, but I don’t think they are absurd, so I can see someone believing that that’s the relevant calculation.

  34. benquo says:

    It seems like the poll unfortunately compresses things into a Bayesian framework where this is the sort of problem that can better be solved with a decision theory conditioning directly on the evidence. One of the major constraints on the evidence available is that multiple parties acted to limit that evidence. Ford asked for confidentiality, and then someone (most plausibly one of her friends) leaked the story at nearly the last minute. Then, Kavanaugh decided to blatantly stonewall and probably lie under oath about the reliability of his memory and related factors. Then, the Republicans refused to allow an in-depth investigation. People on both sides might be justified in saying that they are uncertain, but that some sources of uncertainty need to be screened off by deciding against whichever “side” created the uncertainty, lest it create an incentive to conceal information in the future.

    The obvious solution would have been for the Senate to eliminate the time pressure by passing a special-purpose law allowing the current decisionmakers to retain their control over the Kavanaugh confirmation regardless of the outcome of the midterm elections, and then fully investigate.

    • Evan Þ says:

      The obvious solution would have been for the Senate to eliminate the time pressure by passing a special-purpose law allowing the current decisionmakers to retain their control over the Kavanaugh confirmation regardless of the outcome of the midterm elections, and then fully investigate.

      Such a law would be unconstitutional. Plus, even if possible, removing the incentive of the upcoming midterms would change some Senators’ (or former Senators’) behavior.

  35. Tarpitz says:

    Tangential, but I’ve seen a few people referring to Sixteen Candles as indicative of contemporary culture, implying that we are supposed to think Ted raped Caroline. Am I the only one who thinks that’s a straight-up misinterpretation of the film? I have always read that morning-after scene as implying quite clearly that no sex – consensual, non-consensual or in-a-grey-area-due-to-booze-and/or-shifting-cultural-standards – took place. Am I way off base here?

  36. janrandom says:

    > If we were trying to make these answers into a guideline for how a Senator should vote, it looks like they would be satisfying the most people if they voted to confirm if they thought the accusations had a less than 50% chance of being true, and to reject if they thought they had a more than 50% chance. I wonder how many people would endorse this rule as written.

    That’s basically exactly the explanation and suggested rule given by Susan Collins in her explanation for her vote (actually a speech quite worth to hear).

  37. benwave says:

    It would be interesting to see if there was a significantly different result for people living inside USA and those outside it. I am a well left-of-centre respondent who picked a probability of 80%, then also answered that this evidence, considered in isolation of other factors, was not sufficient to reject him. I see from the results that this was an extremely uncommon answer, and I wonder whether it’s mostly my own idiosyncrasies or if the fact that I’m insulated from this result somewhat by not living in the USA influenced my decision

  38. Basil Elton says:

    Sorry if am asking something everybody knows already, but I couldn’t find the answer on Google or in the comments here. So: is there some explanation to counter a priori improbability of two people coming up with accusations about events that happened 30 years old, exactly at the same time when Kavanaugh was nominated to the supreme court?

    That is, I’m not saying that if those assaults really happened, the a priori probability of reporting in a given moment of time is distributed uniformly over 30 years. But it is distributed over 30 years, it looks to me that it should spike immediately after the events, then drop to almost zero and remain there in the following decade, and then rise after the beginning of the third-wave feminism and remain at some roughly constant level afterwards, with some spike around the nomination. That is, probability density really is higher near the nomination, but overall most of probability, or at least a huge chunk of it, is distributed over the preceding years. While the alternative theory of false accusations obviously assigns approximately all probability to the time period around the nomination. Therefore, before we look at any evidence, it starts off with considerably higher chances of being true. And even higher, if instead of “reporting” we consider “telling about it anyone at all who would remember”.

    Is there anything to balance this off, like (at least) some explanation on Ford’s and Ramirez’s side why they didn’t come up with this before? Or am I making some mistake in the calculations?

    • oncomer says:

      Ford reportedly told her husband and therapist years ago. I don’t know why she would have told Senator Diane Feinstein until Kavanaugh’s nomination. It is very costly to your reputation to accuse powerful people of crimes you know you cannot prove. I would expect most people not to come forward, and for them to be more likely to come forward when half of congress would be more likely to listen to and protect them. I would expect them to be more likely to come forward when the name of their abuser is in the national newspaper daily. I would expect them to be more likely to come forward when their abuser is on the cusp of receiving power and honors. Remember that it is the point of senate hearings to investigate the candidates, so it should not be surprising when things are dug up.

      • Basil Elton says:

        Oh, I didn’t know that there were other people, husband and therapist, testifying that she told them about the case beforehand. This answers my question, basically by “something everybody knows already” I meant exactly that kind of things.

        • The Nybbler says:

          Her husband gave a statement saying she mentioned the case and Kavanaugh’s name in therapy. The therapist did not testify.

  39. Nancy Lebovitz says:

    In regards to background knowledge, I don’t think knowing Kavanaugh’s wife’s name is all that important. Ford’s profession is about as central as where Kavanagh went to college.

  40. Eponymous says:

    Question: did anyone revise their probability estimates after seeing the results of the survey? After reading the comment thread?

    I didn’t, but I’m wondering if maybe I should. I guess I would revise towards the consensus of 50%, but I’m not sure how much would be reasonable, particularly given that the distribution of the results don’t inspire much confidence that they were derived from a very rational process.

    I’m wondering if instead I should adjust my estimate of the collective rationality of SSC commenters down.

    • J Mann says:

      I’ve revised my assessment that Kavanaugh is guilty upward somewhat as the result of the discussion, but my confidence in my ability to estimate (quite low) has not changed. The survey was interesting but didn’t alter my analysis.

    • Edward Scizorhands says:

      I think I inched up a little on Kav being guilty, but inched down on my ability to be sure of anything.

  41. Doctor Mist says:

    Did anybody notice the playful sparring between Senator Hirono and Judge Kavanaugh about whether Yale or Georgetown was the better law school? Would you have initiated that exchange with someone you seriously believed to be a rapist?

    • J Mann says:

      Yeah, the ability to act civilly with people in between periods of calling them scum is pretty common with lawyers, even the inferior ones who went to Georgetown.

      Also, from Hirono’s perspective, a lot of the men she’s dealing with are probably rapists. (As grendelkhan cited upthread, there’s data supporting that assumption).

      • Doctor Mist says:

        Sure, but I think you’re talking about situations where calling them scum is just rhetorical posturing.

        Regarding your other point, hmmm. I guess the alleged statistic that 1-2% of men are rapists means that there’s at least a couple of senators and a handful of representatives.

        • Tarpitz says:

          Only if you assume that Senators and Reps are representative of the population as a whole in this respect. I find it easy to imagine just-so stories whereby the proportions are much higher (erstwhile entitled drunken frat boys over-represented among Congressmen) or much lower (rape statistics for the nation as a whole driven by very high levels of offending by a criminal underclass). I don’t think anyone has a good way of knowing what the appropriate base rate to use is.

          • Conrad Honcho says:

            I just find it interesting that nobody’s really questioning the prior that attendees at elite institutions are more likely to be rapists. I always thought I was supposed to think if someone attended Georgetown Prep, Yale College and Yale law, that meant they were the cream of society, and not likely to be a rapist. Why are our top institutions turning out rapists instead of model citizens? Where can we find model citizens who are unlikely to be rapists if not our premiere institutions?

            What’s the prior on “assumed rapist” for decidedly non-elite high schools in the ghetto?

          • Thomas Jørgensen says:

            … Exactly the same, except the ghetto rapists may rack up fewer crimes before ending up in prison, due to not being as good at covering their tracks.. Or maybe not, it depends on how poorly your ghetto is policed.

          • Conrad Honcho says:

            So, if someone says to you “I went to Yale Law,” you think “better watch out for this one, as Yale Law attendees are more likely to be rapists?”

          • Thomas Jørgensen says:

            Not more. Equally likely as anyone else. You cannot tell by looking, or by social status, it is just an (unlikely. 2-3 % or so) possibility for every man you meet, and an extremely unlikely one for the women, too.

            But as for whether I watch out? Yes, I suppose, technically I do.
            Years ago I picked up the habit of never leaving a drink unattended. (This is mostly to lend social cover. If everyone abides by this rule, then any of my female friends that do this are not being paranoid) Otherwise, not particularly, because I routinely fail to worry about much more likely problems. I am just not a worrier by nature.

          • Conrad Honcho says:

            Then you’re not talking about probabilities but social norms you like. “Equally likely” would be a rather shocking finding to me. This sounds more like political correctness than actual correctness.

            At least then, you would push back against the left-wing insinuations that Kavanaugh’s pedigree makes him more likely to be a rapist?

            Personally I’d go beyond that and say a Yale grad is statistically less likely to be a rapist than a graduate of South Central Ghetto High.

          • The Nybbler says:

            @Conrad Honcho

            I think the assumption is that attendees at elite institutions are more likely to get away with sexual misconduct, and they know it, and therefore they’re more likely to engage in it. There are certainly examples of this (like the Stanford dumpster guy and his light sentence), but I don’t know if it actually holds overall.

            In general the decadency and degeneracy of elites is a pretty common meme.

          • The original Mr. X says:

            Sed contra, attendees at elite universities are more likely to have good impulse control, low time preference, high conscientiousness, etc., all of which make it less likely that they’d commit crimes (including sexual assault).

          • Conrad Honcho says:

            What X said.

            In general the decadency and degeneracy of elites is a pretty common meme.

            Yes, it’s a meme, but I don’t think it’s true. “Rich people are evil” is more like the thing poor people tell themselves to feel less bad about being poor.

            Also, if we’re talking about sex crimes, we’re talking about sex crimes perpetrated by the elite males against the elite females. Maybe you could tell me rich guys like abusing low-class women, but I find it hard to believe they’re getting away with abusing the daughters of other elites.

  42. MB says:

    I see some striking similarities to the equally polarizing Dreyfus affair.
    Obviously, some people want Kavanaugh to be guilty, while others want him to be innocent. I’m in the second camp. As of right now, none of us has enough information to decide one way or another. Maybe we’ll know in a couple of months or years, but for now almost everyone has a strongly felt opinion about the matter.
    My take: the careful scrubbing of the main accuser’s social media accounts, the refusal to turn over notes and other corroborating evidence, the lack of specificity, and especially the delay in issuing the accusation (in two months a lot of progress could have been made with the investigation) are all clear evidence that the accusers did not want the decision to be based on evidence.
    In the Dreyfus affair, Dreyfus was eventually exonerated based on material evidence, but here such material evidence will be hard to come by, due to the vagueness of the accusation and to its lack of specificity (so far neither the time nor the place has been established).
    Not providing any concrete evidence or the time that could be used to exonerate the accused seem deliberate choices. Maybe one of the lawyers knew enough history to frame the accusation in this manner.
    They wanted the decision to be based on a visceral emotional reaction, on one vividly described scene, on nothing but the personal credibility of the accuser and of the accused. Either one believes the accuser or one believes the accused, with no supporting evidence either way. A polarizing psychological drama, in other words, not an orderly investigation.
    And then they portrayed the accused in a stereotypical way in the media: frat boys drink; white men from privileged backgrounds treat women as toys. By contrast, the accuser was made into a representative for all abused women, with whom every victim can identify. If she’s afraid of flying, who isn’t? She’s an everywoman with no particular characteristics; no personal information was allowed to leak.
    To me, again, this looks like a deliberate decision to move the investigation away from the actual evidence, in order to get a conviction based on carefully cultivated stereotypes. For a good while now, movie and literary villains have looked like Kavanaugh. It’s a constantly reinforced message, which pays on such occasions.
    But, of course, we’ll still be having this dispute for many years to come, just as neither Dreyfus’s conviction nor his exoneration actually convinced anyone. Kavanaugh will probably still be heckled, if not worse, 20 years from now. The accuser will always be feted and cherished in certain circles.
    It’s impossible to put the genie back in the bottle. Reason will not persuade anyone, especially since people were urged to trust their instinctual response and their personal assessment of the trustworthiness of the parties (with a thorough groundwork having been laid to persuade people that the idea of universal truth is oppressive, their personal, deeply felt, embodied conviction is the one that counts, that some ways of thinking are inherently oppressive, and alternative, more revolutionary ways of thinking are needed).
    In the end, the anti-dreyfusard camp and their successors were discredited by their WWII collaborationism (and “discredited” means that many were put on trial and sentenced to death or jail after the war). Likewise, nothing short of civil war or foreign occupation will put an end to the Kavanaugh controversy, and I don’t wish for either.

    • Edward Scizorhands says:

      Burning social media to the ground prior to being outed for something on a national level is a very good idea for anyone. It doesn’t show guilt. Same thing with Kavanaugh hiring a lawyer when he was accused of assault: it doesn’t show guilt.

      The reason Kavanaugh didn’t burn all of his social media is because he, like everyone else already part of the national conversation, has already gone through the process.

      • MB says:

        Couldn’t agree more. Ideally, not having any online presence whatsoever is an even better idea.
        But to me it’s striking how the two months’ preparation time before publicly rolling out the accusation was spent erasing evidence instead of finding more of it.
        My belief, also supported by other circumstantial evidence, was that this was part of a deliberate media strategy of appealing to emotion and stereotypes, instead of concrete evidence, in the accuser’s camp.
        As for Kavanaugh, my impression is that high-ranking public officials, with few exceptions, are too busy (or like to give this impression) to maintain a consistent online presence. This matters even more for a judge, who could justly be accused of prejudice when publicly espousing political opinions.