Call For Adversarial Collaborations

An adversarial collaboration is an effort by two people with opposing opinions on a topic to collaborate on a summary of the evidence. Just as we hope that a trial with both prosecutor and defense will give the jury a balanced view of the evidence for and against a suspect, so we hope an adversarial collaboration will give readers a balanced view of evidence for and against some thesis. It’s typically done for scientific papers, but I’m excited about the possibility of people applying the concept to to less formal writeups as well.

For example, a pro-gun activist might collaborate with an anti-gun activist to write a joint article on the evidence for whether gun control saves lives. We trust each person to make sure the best evidence for their respective side is included. We also trust that they’ll fact-check each other and make sure there aren’t any errors or falsehoods in the final document. There might be a lot of debating, but it will happen on high-bandwidth informal channels behind the scenes and nobody will feel like they have tailor their debating to sounding good for an audience.

I don’t know to what degree true adversarial collaborations are really possible. It might be that people who disagree on high-level issues might not be able to cooperate on a survey of the field at all. But I’d like to find out.

So I’m offering a prize, plus a chance to get the results published on SSC, to any teams (probably of two people each) who want to do adversarial collaborations. If you want to participate, comment on this post with what subject you’d like to work on and what your opinion is on the subject. Or look through existing comments, find someone who has the opposite opinion to you on a subject you care about, and reply to them saying you want to be their foil. After that you can exchange emails and start working.

If at least five teams participate, there will be a prize of $1000 for whichever team I think does the best work. There might also be a prize of $250 for a second-place team if they do exceptionally good work, though I am not promising this. Thanks to everyone who donates to this blog’s Patreon for providing the money to make prizes like these possible. Here are some more rules:

1. You will write an essay summarizing your joint summary of the evidence regarding a controversial topic you disagree on. Strongly recommend that this be a single factual issue, like “Does gun control save lives on net?”, rather than a vaguer moral question like “Guns – good or bad?”, though it can still be a pretty broad topic – I would love to see people write about Caplan’s case against education, for example. Even though most of the examples here are political, this doesn’t have to be; it could involve controversial topics in medicine, history, religion, et cetera.

2. You will write the essay as a united front. Please don’t write “Alice says this study proves guns save lives, but Bob says it’s wrong and this other study proves guns are bad.” Instead you are going to have to come to an agreement on how to describe each study. For example “Here is a study purporting to show that guns save lives. It seems to accurately describe what is going on in rural areas, but it might be of limited applicability elsewhere.”

3. You will come to at least some sort of unified conclusion, even if that conclusion is “There’s not enough evidence in this field to be sure either way and we should default to our priors/biases”.

4. The essay should be similar in length, tone, and amount-of-research to one of my Much More Than You Wanted To Know essays, eg here and here.

5. By entering the contest, you are giving me permission to publish your essay on SSC (with full attribution to you, of course). You can also publish it wherever else you want. I will probably publish the winning essay, and I might or might not publish the others depending on how good they are.

6. Because of (5), please don’t research any topic that I would not be able to publish on SSC if you came to a taboo conclusion. If you want to do an adversarial collaboration on taboo topics, you can feel free to arrange it in the comments, but it won’t be considered an official entry, it won’t be eligible for prizes, and I probably won’t post it (I might link it if it’s posted somewhere else). If you’re wondering whether a specific topic is taboo, you can ask.

7. If you’re officially proposing a collaboration or responding to a proposal, please put those comments in bold so people can find them amidst the discussion in the comment section. I may edit the timestamps on comments to bring these to the top, or even to bring the most interesting ones furthest to the top. If you have many opinions you’d be willing to try an adversarial collaboration on, consider posting the one where you disagree most with the SSC consensus, so that you have the most chance of finding a collaborator. If you get many responses, please talk to the people involved, choose one, and mention your choice clearly on the comment so other people don’t keep asking.

8. I’ll update everyone on the next Open Thread about the state of the competition, whether it’s actually going ahead, whether there are at least five teams, et cetera. I’ll also post the closing date by which all entries must be completed and submitted to me by email. Assume for now this will be around July 1, though I’m happy to shift that a little bit if there’s strong demand.

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

605 Responses to Call For Adversarial Collaborations

  1. RC-cola-and-a-moon-pie says:

    I worry that if commenters are not experts in the subject matter, but rather merely informed laypersons with policy views in the area, that the write-ups may tend to skew toward whichever of the pair is a more thorough researcher and more persuasive advocate. True expert adversarial collaborations sound enormously valuable.

    • Scott Alexander says:

      I agree that’s a risk. One of my goals with this is to get a good understanding for in what situations adversarial collaborations do vs. don’t work. Another goal is to raise awareness enough that some actual experts get interested.

    • Joel Salomon says:

      … the write-ups may tend to skew toward whichever of the pair is a more thorough researcher and more persuasive advocate.

      Very much like a trial then, with the jury empanelled to decide which side has the better lawyer.

      • RC-cola-and-a-moon-pie says:

        True, but two points. First, lawyers receive years of training and at least theoretically have a degree of expertise in certain kinds of advocacy relevant to trial prssentations. And second, the legal system imposes certain justiciability requirements for lawsuits to be within the jurisdiction of the courts precisely to ensure that litigants have real skin in the game to incentivize their respective counsel to do a thorough job of investigating and presenting the facts. One could argue that these criteria are exactly what would be missing in lay adversarial collaborations. (I guess the third point is that the legal system arguably presents a poor model for getting to the truth of complicated scientific questions.)

        • albatross11 says:

          One thing I think you’d want from this kind of collaboration is two people who are willing to accept/acknowledge the weak points in their own argument and the strong points in the other side’s arguments.

          I’d also say that it’s hard to feel like you’ve gotten anywhere when you’re looking at basically a clash of theories/models with no way of empirically deciding anything. That’s a place where the better arguer will likely prevail. But if two clashing models of the world lead to very different predictions, you can look to see how the different predictions work out.

    • Ben Thompson says:

      It would be great if we could get two such unbalanced teams working on the same question, skewed in different directions. Then we’d get some insight into reason vs. authority.

      (…But I guess that presupposes two experts on opposite sides of an issue.)

    • Brad says:

      True expert adversarial collaborations sound enormously valuable.

      One of the best classes I ever took was a course on death penalty law team-taught by a pro-death penalty professor and the lead litigator of an anti-death penalty organization. But I’m not sure that either, but especially the litigator, would have been willing to make some of the concessions they did during the course about the strengths of the other’s position in writing.

    • J Mann says:

      @RC-cola-and-a-moon-pie, I would be willing to take the other side of that if you want to collaborate. 😉 My guess is that one side will leave more facts on the table than the other, even when both parties are operating collaboratively in good faith and trying to help each other out, but not by much.

      (I’m joking, of course, but also serious – let me know).

  2. losethedebate says:

    I might be interested in collaborating on an article about Mercier and Sperber’s Argumentative Theory of Reason (I lean towards believing it to be true). However:
    1. I would likely not be able to start work on this until after May 19th, when my semester ends.
    2. I am at best an interested amateur on this topic.

    Edit: link formatting
    Edit 2: forgot to include my own stance on the topic.

    • textor says:

      I could cooperate. I think this theory is beautiful and part true, but part wrong. This is really a separate problem for adversarials (adversatrials? OK I get it) – positions allowing effective opposition may not be the polar opposites, and not every disagreement is easily described by two general stances (such as pro-life and pro-choice).

      • losethedebate says:

        Sorry it’s taken me a couple days to reply, end of semester is a busy time. If you’re still interested in a collab on this, you can email me at v r papenhausen (without the spaces) at gmail. I also have a friend who could do a three-way collab with us, filling out the “stridently opposed” position (though I haven’t asked yet if he’d be interested), if you think your position isnt strongly opposed enough to make for a good collab with only a supporter of the theory.

  3. Joel Salomon says:

    It might also be useful for separate write-ups in cases where the joint paper becomes impossible—including both sides’ explanations of how the collaboration broke down.

    • keranih says:

      100% support this reporting of negative findings. Hopefully it would help people involved in future projects lessen failure rates.

  4. Jacob says:

    Scott, is an adversarial collaboration eligible if it’s already in progress?

    Spencer Greenberg recently ran a study on power posing, and I’m in the process of preregistering my own analysis plan of his data. The central research question is “Does holding a power pose improve mood?”

    Spencer came to this research expecting a positive answer, based on his own experience power posing. I come to this having written several articles about the entire field being p-hacked abomination of null effects. We are both going to look at each other’s analyses.

    I have enormous respect for Spencer, so I will certainly try to reconcile our analyses and arrive at a shared conclusion. I hope he feels the same.

    Of course, since he initiated the research and collected the data, he owns it and gets to decide where it may or may not eventually be published. But if you say we’re eligible, I’ll ask him and work to make a submission that fits your criteria.

    • Scott Alexander says:

      I didn’t know there were still defenders of power posing around, and I’m worried this would kind of come across as beating a dead horse. But I won’t explicitly ban it.

  5. salimfurth says:

    I’m interested! I’m a conservative/free-market research fellow and willing to write about economic policy, especially in urban economics. One possibility would be to pair me (a pro-growth, pro-property rights YIMBY) with a NIMBY or growth-skeptic. Another would be on the likely net benefits of economic regulation.

    • Anonymous Bosch says:

      Another would be on the likely net benefits of economic regulation.

      This sounds way too broad to have a hope of producing something useful.

        • onyomi says:

          Minimum wage seems to be one where the literature is very split.

          • pjiq says:

            for minimum wage, I did this fun econometric research paper that showed potential effects (some were positive, others negative) of higher minimum wages by making comparisons btwn different US states, if anyone does this debate I’d be happy to send them my work.

          • oli cairns says:

            I’m happy to do a the minimum wage is likely to decrease employment contribution

        • Gil says:

          I’m not volunteering but I’d love to see a well hashed out discussion of the limits of zoning/restrictive covenants. I’ve been around a lot of urban planners and planning academics, and this topic is one where I think there is a lot of genuine disagreement, not so much on the edges (ie. no zoning vs extremely restrictive zoning), but where on that spectrum is the optimum point that produces the best economic and social outcomes.

          Another related topic might be gentrification: net econ/social positive or negative.

          • cryptoshill says:

            At least in “urbanist” communities there are vocal and populous (although their overall representation is unkown – I am using populous to mean “my market-urbanist Applause Light statements are frequently upvoted) anti-zoning groups as well as groups who want all of the typical welfare-zoning (in this I include rent control, required minimums of “affordable” housing, etc) laws. I don’t think I see any support among that group of “neighborhood character” type rent-seeking-disguised-as-aesthetic legal action.

      • Eponymous says:

        @Anonymous Bosch:

        People have done research trying to connect overall measures of regulatory burden (e.g. total lines of regulatory code, aggregate measures like freedom index) to economic outcomes.

        I’m not saying this research is any good. Just that it exists.

    • sclmlw says:

      Here’s the collaborative adversarial discussion I would like to see:

      1. Which laws (at a local, state, or national level – pick one) that have actually passed achieved the goals the law was explicitly set out to achieve? (Wholly, in part, or not at all.)
      2. What are the defining elements of laws capable of achieving stated goals, versus laws incapable of doing so?

      I’m not a strong enough partisan on this subject to consider myself a good candidate as a collaborator, but if you have a strong interventionist you’d like to collaborate with I think this idea would be a fascinating way to narrow your broad interest down.

    • bulb5 says:

      I feel like I’d be significantly outgunned in an adversarial collaboration, but I’m interested in the intersection of behavioral economics and regulation – I believe many regulations that try to protect us from ourselves do more good than harm.

      I prefer to remain anonymous on this blog, but I’ll send you an email if you’re interested.

      • toBoot says:

        I’m interested in this too, and am happy to discuss. I took a class on Law, Econ, and Psychology (which was essentially a class on regulation and behavioral economics) in Law school and have kept up with some of the literature.

        I’m generally a big proponent of regulations that promote information disclosure, but not of ones that prohibit transactions.

        I don’t wanna throw my email up here unless necessary, so lemme know!

  6. userfriendlyyy says:

    Base claim: A Job Guarantee is a much better and more urgent social safety net than an UBI.

    Reach Claim: Capitalism will descend into neofeudalism or neo-fascism without it.

    I don’t think anyone would be able to do an adversarial collaboration on the reach claim because it’s hard to make much in the way of evidence based claims in that area.

    • Alsadius says:

      I probably don’t have time to do this properly, but if I did, I’d actually defend the status quo against both. We’ve spent a lot of time wringing welfare traps out of the current system, and while it’s imperfect it’s at least much better than what it used to be. Both of those proposals would bring them charging back. Given how astonishingly destructive welfare dependence is, that is a very big problem for me even before you consider the price tag.

    • Randy M says:

      What does “urgent” mean here?

    • Izaak says:

      I’m interested in at least talking about doing this with you, with the caveat that I’m just starting a very busy month in my life, so it might be hard for us to make the deadline.

      I prefer UBI to a Job Guarantee, but I agree that something is necessary to avoid capitalism descending into neofeudalism or neo-fascism, so we’d be arguing the base claim.

      You can email me at:

    • Ash says:

      I was going to post about UBI as the best-case welfare policy for the near term future, so I think that is close enough to this ballpark to be doable. I am not an economist per se, but I specialize in meta-analyses and statistical evidence (development econ + program evaluation) so that can help I imagine. Let me know if you are interested (or if anyone else is) with a reply and we can set up an email exchange.

      Also I see no reason it has to be 1v1, 2v2 etc are viable formats if perhaps having a few more kinks.

    • Erusian says:

      If you guys want to do a three-way or simply to argue with a more opposed position, I’d argue with you that UBI and Job Guarantee are both policies that are bad not only for general economic output but inferior to more practical policy prescriptions in helping the lower classes.

      • Edward Scizorhands says:

        Definition question:

        Does job guarantee mean, by definition, that the government is the employer for hard-luck cases? Normal employers (like our current mix of for-profits, non-profits, and governments) can employ nearly everyone as long as you make the wage subsidy big enough, which I guess is the distinction between wage subsidy and job guarantee.

        • Erusian says:

          I’m taking it to mean that the government is the employer. In particular, several specific policy prescriptions have come out of some more radical left-wing think tanks that suggest the government offering employment to all comers at $15/hour with benefits.

          I’d be glad to discuss wage subsidies, but depending on the specific policy prescriptions can’t guarantee uniform opposition.

      • Tracy W says:

        Can I support you on that, Eursian?

      • ADifferentAnonymous says:

        What do you mean by “more practical policy prescriptions”?

      • pjiq says:

        I’d research and make an argument in favor of UBI (but not a job guarantee)

        • Erusian says:

          I’m fine with debating just UBI, presuming that we pre-agree on what ‘UBI’ specifically means. Alternatively, we can agree to look at multiple proposals. I just want to debate specifics rather than generalities.

          • pjiq says:

            UBI often means a basic income that by definition can pay the bills- which does seem unsustainable, at least in certain situations but a basic income in the sense of “some kind of citizen’s dividend, similar to the Alaska Permanent Fund” I’d be interested in arguing for that.

            Basic Income therefore I define as “cash payouts from the government to all citizens,” (regardless of whether those payments provide a “liveable income.”

            Anyways let me know if that’s specific enough for you or if you were thinking of something completely different-

          • outis says:


            Basic Income therefore I define as “cash payouts from the government to all citizens,” (regardless of whether those payments provide a “liveable income.”

            I guess English may not be your first language, so allow me to clarify: in “Universal Basic Income”, “universal” is the part that means “to all citizens”, and “basic” is the part that refers to “providing for basic life needs”. You took out the wrong part.

          • Note that most UBI proposals involve significant tax increases

          • pjiq says:

            Regarding outis’ comment on my definition of UBI (and how it takes out the “basic” part) I’d argue that my defining “basic” as “forming an essential foundation or starting point” (for one’s financial well being, in this case) is not necessarily absurd or evidence that the English language is something I have utterly failed to grasp ;). This is the Google dictionary definition of the word “basic” after all. But I agree that UBI typically refers to something that at the bare minimum covers basic needs, which is why I clarified my more flexible position in the first place.

            Reading through the Wikipedia page on “basic income” in order to clarify further syntax concerns, I came across:

            An unconditional income that is sufficient to meet a person’s basic needs (at or above the poverty line) is called a full basic income, while if it is less than that amount, it is called partial.

            Therefore I will rephrase that I am hoping to argue for a “full or partial universal basic income” rather than exclusively for a basic income that is “full”, in hopes that this will be sufficiently clear.

            Erusian, if you still wish to discuss this subject more in depth, contact me at pjiq786 [at]

    • cassander says:

      It seems to be that a wage subsidy is an almost categorically better idea than a job guarantee. it costs the same amount of money and encourages finding actual productive work for the people in question, instead of just having half dig holes and the other half fill them.

    • Brian Bleakley says:

      I’m interested in this and sort of disagree with your claim. I’d like to weigh in on this, but as you mention it’s hard to bring much data to this discussion. brianbleakley at gmail

    • MostlyCredibleHulk says:

      I can plausibly see how you can do research on “better”, though “urgent” is a value judgement IMHO, so I am not sure how you can research if something is urgent or not. But I completely can’t see how the “reach claim” could be verified in research – especially given that there’s no universally accepted definition of “neofeudalism” or “neo-fascism” as far as I know.

    • userfriendlyyy says:

      Sorry, I had something come up in my personal life.

  7. Soeren E says:

    I am interested in adversarial collaboration on AI safety. I mostly agree with the main points expressed in the book “Superintelligence: Paths, Dangers, Strategies” by Nick Bostrom.

    We will need to reduce the scope quite a bit, as a cannot commit to an ambitious essay.

    • jdly says:

      I’d be interested in this topic! I’m an engineer who knows a bit about current ai technology and I’ve been very skeptical of ai risk for a long time, so this is a great opportunity for me to write something about it.

      A good thesis on my part might be there is a negligible chance of humans creating an artificial general intelligence within the next 1000 years.

      Edit: to clarify, I specifically believe the risk is so small it isn’t worth worrying about. I mean it in the sense that donating to places like MIRI is a waste of money.

      I’d prefer to work with someone who is knowledgeable about the subject, or at the very least has the time to do some serious research, because I think Scott is looking for a pretty lengthy essay.

      • Douglas Summers-Stay says:

        I’d be interested in coming from the other side of this (that AGI will probably happen sooner than that). I work as an AI researcher, and have some relevant publications.
        It seems like talking about *when* AGI will happen requires coming to a consensus about how we can know how long far future technologies (more than 20 years out) will take to arrive, and how we can judge whether or not AGI has arrived.
        Let me know if you’re interested. I could contribute together with Soeren, if you are both want to.

        • jdly says:

          Glad to hear! I’d still be down to defend the position I made in my original post if you’d prefer (or are more qualified) to talk about AGI rather than ASI, but after sleeping on it I think a better question to explore with regards to AI safety is as follows:

          What is the existential risk of AI technology (compared to other existential risks)? I think answering this question is good because we can come up with some sort of consensus based Drake equation-esqe fault tree where we try to nail down (with associated uncertainties) how likely and/or difficult various conditions for dangerous superintelligence are.

          My position would be:
          Even getting to AGI will be very hard and take a very long time.
          Even if we get to AGI, it is unlikely that it would be able to recursively self improve.
          Even if it can recursively self improve, it is unlikely that that self improvement would be exponential.
          Even if that self improvement is exponential, it is unlikely that it will be exponential for very long.
          Even if an AI is able to sustain exponential self improvement for a long time, it is unlikely that that humans won’t also have some sort of intelligence augmentation that levels the playing field and reduces the existential risk.

          Again, we can focus on AGI if you want, and I do think it would be interesting to do some sort of first principles write up where we nail down definitions and give the readers a layout of the current state of technology and what needs to happen for AGI.

          I’m kind of leery about putting my email on the machine readable internet, so I guess you (or anyone who is interested) can contact me at my throwaway sinenomine1337 at g mail and I’ll give you my real contact info from there.

      • Soeren E says:

        I would like to collaborate with you on writing an essay about this topic.
        To make my claim explicit: I assign a 75% risk of Artificial General Intelligence being developed within 1000 years.
        Most of the 25% is global catastrophes etc., so we should consider conditioning on this to make our disagreement more salient.
        I reserve the right to update as I write the essay :).

        Would you be willing to assign a percentage to your belief? I would like to narrow the scope to not consider if MIRI etc. are effective organizations. Also, unless the temporal discount rate is really low, it is not worthwhile to care at all about events in 1000 years, even if they are very likely.

        I would consider myself quite knowledgeable about AI Safety in general, but only moderately so about AGI development. Would you be interested in adversarial collaboration with both me and Douglas Summers-Stay?

        Feel free to email me

        • sclmlw says:

          I agree that a 1000 year time horizon is a little long, and beside the point. Perhaps a better question in regards to this issue is to balance out the perceived probability of developing AGI versus the perceived ability of humans to control said AGI (for example, by crafting effective morals testing). And putting this all in context of something that makes sense to consider technologically, I think, means you have to have a time horizon that is within the potential lived experience of someone reading this blog.

          Let’s say few readers of this blog are less than 18 years old (or few people capable of doing anything about AGI in their lifetimes are <18). Let's assume a lifespan of 90 years (life expectancy has been increasing for humans, but there is no evidence that lifespan is increasing so we should default to a generous assumption of lifespan based on the current non-trend of lifespan stagnation). That gives until 2090 to develop AGI in a time horizon that is meaningful in the sense that we ought to think about doing something soon.

          Contrast that with the question, "say we developed AGI; how long would it take us to develop the ability to perform effective morality testing on it prior to giving it any kind of power?"

          Say the answer to the question of how long AGI will take is around 72 years, but it will take us 12 to figure out the morals testing thing, then I think there's little cause for concern over the next half century. If, however, morals testing is hard, and takes 150 years, then we're in a bad way, and should consider panicking.

          Ultimately, the question we want to answer is, "should we be worried about this taking over and subordinating human control?" Or more importantly, "can we stop it from doing so?" And in that sense, it doesn't make sense to worry either about something that's not going to affect us anyway, or about something we can't fix.

    • johnsonmx says:

      I strongly believe that The Complexity of Value Thesis is Wrong (unfalsifiable / unnecessary / intrinsically confused) and is essentially functioning as a ‘cognitive stop sign’ for work that could actually generate progress on the Value Problem. I.e. I think CEV and Fun Theory have failed to mature into productive research directions, not for want of trying, but because value should be defined at the level of qualia, not the level of intensionality. Progress- or even greater clarity- here could substantially help AI safety efforts.

      The talk I gave on this two weeks ago;
      The argument in one infographic;
      Relevance to AI safety;
      Why we can’t be realists about suffering and non-realists about consciousness.

      (Mike Johnson, Qualia Research Institute)

      • skef says:

        A lot of your opponents are going to take this as falling short of an argument.

        Suppose person A lives a full, happy life that features frequent interaction with true friends. Person B lives a full life of similar feeling that many would be comfortable labeling “happy”, but B’s frequent interactions are with people who actually dislike B and are paid to pretend to be B’s friends. We can also consider C, who’s entire subjective experience is driven by a computer simulation. C has never interacted with anyone of his kind, his “children” are just simulations, etc.

        On hearing about these lives, many people will consider them to differ in value even though the subjective experiences are similar. An argument against value complexity would need to somehow address such concerns. Your argument is too low-level to do so.

        • johnsonmx says:

          I think you’re conflating ‘correct’ and ‘consistent with common intuitions’– I generally agree with Eric Schwitzgebel when he suggests that

          Common sense is incoherent in matters of metaphysics. There’s no way to develop an ambitious, broad-ranging, self- consistent metaphysical system without doing serious violence to common sense somewhere. It’s just impossible. Since common sense is an inconsistent system, you can’t respect it all. Every metaphysician will have to violate it somewhere.

          This seems to suggest that if we take “consistent in all ways with common sense intuition” as an axiom for how to build some sort of computational morality, then we may be doomed before we even begin, because common sense intuition is internally inconsistent.

          But, I think this is even a little orthogonal to the Complexity of Value Thesis(?).

          Ideally, we should talk about theories of value/consciousness on the object level, in terms of predictions. And I think our alternative to CVT offers some novel predictions.

          • skef says:

            I think you’re conflating ‘correct’ and ‘consistent with common intuitions’

            Except that I didn’t make any claim about what was correct. My claim is that your arguments don’t touch on the intuition. You can add: “And I’m also stipulating that intuitions of this sort are irrelevant in order to make some progress”, but I’m reasonably confident that Schwitzgebel would then consider you to be a philosopher exploring one area of a problem space, rather than as someone who is addressing the argument that you claim to be.

            But, I think this is even a little orthogonal to the Complexity of Value Thesis(?).

            We don’t necessarily even need to go up that far. Suppose that the intuitions carry some weight. The common thread seems to have something to do with the attitudes that lead to our good feelings being accurate. What is or isn’t accurate? (That is, what does the concept of accuracy validly apply to?) Some sort of content. So that weight alone can bring intensionality back into the picture.

            Moving up a level (and presuming the relevance of intensionality to the question, which I take from your summary that you accept), value would then depend in part on a relation between qualia, the thought patterns that resulted in those qualia, and some sort of accuracy condition on those thought patterns. That puts value partly “out in the world”, where things that can have positive value for one consciousness can have negative value for another. It is at least plausible that the sort of complexity posited by the thesis arises from such interactions.

          • johnsonmx says:

            My expectation is that if we grant that qualia are ‘real’, the principle of parsimony suggests we should try to explain moral value wholly in terms of qualia. I.e., I would say accuracy would be instrumentally useful for optimizing good qualia over time, but I don’t see it as an intrinsic good.

            That puts value partly “out in the world”, where things that can have positive value for one consciousness can have negative value for another. It is at least plausible that the sort of complexity posited by the thesis arises from such interactions.

            I like this thought experiment. Different beings will react differently to different stimuli (apparent complexity of value) but this doesn’t mean there’s not a common thread across different ‘flavors’ of positive qualia (unity of value thesis).

          • skef says:

            The thought experiment has some attractive features, but the picture it paints is based on a far less radical view than yours.

            Suppose for the sake of argument that in one year a fast-takeoff event results in a super-intelligent A.I. It arrives at the right theory of material/qualia relation and, not quite having figured everything out “uploads” everyone into a unified simulation where everything seems to work out better for the consciousnesses, and therefore according to the qualia theory of value, actually does work out better for them.

            Then there are two further stages.

            In the second stage, the A.I. breaks the simulations up to better optimize the qualia of each consciousness. Now no consciousness interacts with another consciousness, but all of them have a “better time”.

            In the third stage, the A.I. abandons its previous focus on continuity and switches to the following strategy: 1) It maintains a small “lab” of short-term conscious simulations in an attempt to develop better and better “peak experiences”. 2) It devotes the bulk of its consciousness equipment to “playing” as many copies of its current peak experience as possible. 3) It attempts to increase the amount of equipment available in order to play more of those copies.

            I take you to be “on board” through stage 2. What about stage 3? If qualia are the only good, what is the basis for the significance of individuals? It’s nice to think of each of us experiencing bliss in some future simulation, as the linked thought experiment seems to suggest. And the A.i. could of course make use of that niceness — perhaps the peak experience would include some feeling of togetherness and community. It won’t be true, of course, but the idea that it would need to be is presumably just another facet of the mistaken impression that there is value beyond qualia.

            So if we follow the implications of your view, is this third stage bad, and if so why?

          • johnsonmx says:

            I like the clear stages of your thought experiment; however, our intuitions actually diverge during part 1. In short, if we’re interested in understanding qualia I suspect we should pay attention to what atoms are doing, not what bits (e.g. simulations) are doing, because there is no objective fact of the matter about which computations/simulations a given physical system is instantiating. Tegmark has written about this in his Perceptronium paper, although I think the fuller argument comes from McCabe here.

            All that aside, I do agree with you that identity is a weird thing to consider in ethics, since it’s sort of an inherently fuzzy concept, and any way we carve it we’re going to get some counterintuitive results. For framing this issue I’d defer to Andres’s writeup here (feel free to skip the text and just look at the pictures). I consider myself halfway in between an open individualist and an empty individualist, in the terminology he uses. (I.e., not a closed individualist.)

          • skef says:

            I like the clear stages of your thought experiment; however, our intuitions actually diverge during part 1. In short, if we’re interested in understanding qualia I suspect we should pay attention to what atoms are doing, not what bits (e.g. simulations) are doing, because there is no objective fact of the matter about which computations/simulations a given physical system is instantiating.

            This is not a context where I can expect what I’m about to do to not come across horribly, let alone to work. But here I go anyway:

            Are you fundamentally trying to help, by way of coming to accurate beliefs about complex subjects (regardless of whether correctness ultimately matters in the end), or are you hoping to trade whitepapers for cash? If the latter — fine, just ignore this conversation.

            If you’re trying to do the former, and someone poses a thought experiment that may count against your view, make an effort to improve it as best as you can, rather than finding an easy way of dismissing it.

            In this case, the substrate/simulation issue isn’t inherent to the objection. The A.I. can just as easily start by envating brains into a shared simulation, and then move to individual simulations, and then concentrate on growing custom brain tissue with optimized subjective experience. That changes some features of the example — perhaps there are certain limits on the experience that can be induced in a brain based on what it has or has not already experienced — although if memory can be manipulated aside from consciousness (as the second linked author seems to suspect), such limits are probably modest.

            This particular instance is far less important than the meta point. If your general approach to objections is to note that things are complicated and link to someone else’s paper, you will likely wind up spending your life preaching to and getting grants from the converted. We could joke about that being entirely compatible with your purported view, but that’s only actually true on a solipsistic level — if you don’t wind up influencing people towards a better outcome, you’ll only have pleased yourself.

            With respect to the second link, yes identity is complicated, and seems to be wrapped up in concepts of value. And those concerns can’t just be waved away with a vague invocation of “counter-intuitive”.

          • johnsonmx says:

            I noted one set of objections to your thought experiment in my first paragraph, and in the second paragraph noted that, although these objections weren’t inherently fatal, identity is inherently fuzzy, which in turn make it difficult to draw clear conclusions from arguments which at their core revolve around our intuitions about identity (as your thought experiment seems to).

            I would add to this the point Eric Schwitzgebel made, that intuition is not always going to be a good guide for ethics, especially in novel contexts. When one adds these three things together, it seems to me the structure of the thought-experiment is unsalvageable. I did appreciate the thought that went into it however.

            It seems to me you’re perhaps halfway in between offering helpful comments and being frustrated in a counterproductive manner. Thanks for the feedback on argument style, it is appreciated.

          • The principle of parsimony only should use the simplest explanation that works.

            there is no objective fact of the matter about which computations/simulations a given physical system is instantiating.

            That isn’t a fact, and simulated qualia make a strong argument against it. If there is a fact of the matter about the qualia in a simulation, you use that to distinguish between simulated heaven and simulated hell.

      • Reasoner says:

        I kind of agree, but not because of anything having to do with the definition of value. By the time we are able to create AGI, I think our ML capabilities will be at the point where we’ll be able to train a model for human values that knows our values more deeply than we ourselves know them, in the same way Amazon can recommend products that users didn’t know they needed. Even if that’s too hard, I think we’ll be able to train a model for what it means for an AI to behave like a good servant. MIRI et al. play a motte and bailey game where capabilities will be super strong when it comes to implementing UFAI but super weak when it comes to implementing FAI.

        Too busy to do an adversarial collaboration about this right now but might like to in the future–I could write a lot more about this.

        • rlms says:

          MIRI et al. play a motte and bailey game where capabilities will be super strong when it comes to implementing UFAI but super weak when it comes to implementing FAI.

          What do you mean by this? I’m not much of a MIRI fan, but I’ve always got the impression that they think FAI will immanentize the eschaton.

      • What happened to the option for “qualia are real but not mathematically tractable” ?

        • johnsonmx says:

          I wouldn’t be inherently opposed to the idea, is there anything else in this reference class you could point to as an example?

          • In what sense, a.fully worked.out theory? I don’t think that is.very relevant. If you are trying to argue from an exhaustive list of possibilities, the existence.of a possibility.not on the list is a problem.

      • says:

        I am interested in collaborating with you about the complexity of value thesis and qualia. I strongly endorse the complexity of value thesis and I don’t think value should be defined at the level of qualia.

        I am a philosophy PhD student with research in consciousness and AI safety. I’ve also had the standard training in ethics and metaethics.

        • johnsonmx says:

          Hi Daniel,

          This sounds promising! What sort of paper would you like to collaborate on?

          A few thoughts that might spark something:
          – A possible focus could be on laying out the arguments for complexity of value (CVT) and unity of value (UVT). I’d also like to discuss a little about what each framework has generated. E.g., CVT has inspired people to think about indirect normativity approaches like CEV and fun theory, and I think Christiano’s work as well(?). Perhaps the current status of CEV/FT could be discussed. Likewise UVT has led to QRI’s body of work.

          – Finding the exact way to double crux here seems really important.

          – I’m hoping we could frame questions of metaphysics & ethics as empirical predictions. I.e., if we do X, and Y happens, then it’s evidence for CVT. I can say more about why I think this is possible.

          – It seems like there’s this interesting memetic polarization that happens between views here. The LW memeplex tends to speak in terms of utility functions, preference satisfaction, complexity of value, universal turing machines, etc. These are taken to be fundamental, whereas other things (like qualia) are taken to be leaky reifications. On the other hand, formalists (like Tegmark, Tononi, QRI) would suggest most of these are leaky abstractions, but qualia is a natural kind, emotional valence is a natural kind, and so on.

          – Would you self-identify more with the LessWrong memeplex, or the academic philosophy memeplex?

          I’ll also email this.


    • Jedediah says:

      I’d take this on. I’ve been a programmer for 30 years. I would argue the following points:

      * Nobody actually knows what AI is
      * Nobody knows if superintelligence is possible
      * AI is probably the safest kind of software
      * There’s no reason to worry that AI will do silly things (e.g. turning everything into paperclips)
      * There’s no way to control what people do with AI
      * Humans will probably never coexist with AI
      * We shouldn’t care about future humans more than AI

      PM me if interested –

  8. Ozy Frantz says:

    I’m interested in doing adversarial collaborations on parenting!

    About My Views On Parenting:
    –Attachment parent but not hardcore about it
    –Plan to homeschool, but pro-homeschool-regulation
    –Free-range kids
    –Gender-diverse/feminist parenting
    –In favor of both cosleeping and cry-it-out if it works for the family
    –Breastfeeding has relatively minor health benefits except for IQ
    –Pro-epidural, but we do too many C-sections
    –Positive discipline
    –Non-abusive spanking probably doesn’t have long-term negative consequences
    –Violent media probably does have short-term negative consequences
    –Children should have more autonomy
    –Teenagers should essentially be treated as adults
    –Sex-positive; in particular, we should be less hysterical about childhood and teenage sexuality
    –Parenting has an effect on children but not as much as the average parent likes to think
    –Ask me if you have another idea!

    Of course, we’d collaborate on some more specific, narrow issue.

    Note that I am currently taking care of an adorable new baby and so am likely to be slow in writing the collaboration.

    • a reader says:

      What do you mean by “Gender-diverse/feminist parenting” ? If it’s about transitioning in childhood, I may be interested – my position being that children shouldn’t take puberty blockers.

      • Aapje says:

        It’s probably more anti-gender norms.

      • Ozy Frantz says:

        I don’t want to argue about trans issues at present.

        Gender-diverse/feminist parenting is, as Aapje said, anti-gender-norms; gender-diverse is often used instead of gender-neutral to convey that if a child wants to behave in a gender-conforming way, that’s wonderful. Examples of how it applies in my own life: I will let my son wear dresses if he wants to; I make an effort to buy books with female protagonists and gender-non-conforming protagonists; I will buy my son dolls, a toy kitchen, and other conventionally female toys; I ask people not to assume that my son is going to be violent; I make an effort to gently correct children’s pronouncements that only girls can X or boys all Y.

        • len says:

          How do you even plan to study this? Is there even a dataset that can be used? Even assuming such a dataset exists, how do you even measure parenting effectiveness — are you just going to use life outcomes of the child, controlling for a bunch of factors?

      • mo says:

        I would be interested in taking the children should take puberty blockers in certain circumstances position with you if you are interested.

        • a reader says:

          @mo: ok. If you give me an email – rot13 if you don’t want to give it to spambots – I will send you a first draft about my point of view and my arguments, probably next week.

          Very briefly, I’m not totally against transition – if a child has strong, constant gender dysphoria from 3 to 18, maybe transition is the best course – but afaik, according to many studies, in the majority of cases, childhood gender dysphoria ceases after puberty and those children become usually cis gays/lesbians as adults – so no need for lifelong hormones and invasive surgery in those cases.

          • mo says:

            Sure, my email (rot13) is
            zrtnzbabpyr [at]

            We might not have a disagreement depending on what you mean by “children.” My position is that puberty blockers are sometimes advisable starting right after the first signs of puberty (around age 12).

            If this sounds like something you disagree with, then I look forward to working with you.

        • Deiseach says:

          How do puberty blockers work, exactly? Do they affect the other changes of puberty such as growth and so on? If you have an eleven year old starting on puberty blockers, will they still look like an eleven year old when they are fifteen in a class of their fifteen year old peers who have gone through puberty?

          I have no idea what is involved and this is why I can’t make a properly informed decision. If you’re going to be just as tall etc. as everyone else, only difference being your secondary sexual characteristics have not developed, that’s a different case to stunted growth and the like.

          • rlms says:

            A quick google turns up this, which claims that they block all aspects of growth, not just secondary sex characteristics (but this is temporary, you grow as normal after stopping taking them).

          • Back when I wrote Future Imperfect, I raised the possibility of puberty blockers, not knowing that they already existed.

            I was thinking not in terms of children who might want to be transsexual but of parents who might want to postpone the problems of dealing with adolescent children. Has anyone been discussing that side of the technology and related moral issues?

            Lots of possibilities for sf plots there.

    • Evan Þ says:

      I might be interested in a collaboration on “Teenagers should essentially be treated as adults,” though I’m not sure whether we could agree on what standards that claim should be judged by? Frankly, I’ll need to think about which standards I’d accept.

      Alternatively, could you elaborate on what kinds of homeschooling regulation you’re in favor of? Though, I’m pretty sure we do disagree on what standards to judge that by.

      • Jiro says:

        I wouldn’t even know what teenagers should essentially be treated as adults” means. My first question would be whether equal treatment with disparate impact counts as “treating them the same” (for instance, if you require that someone have a college degree and a job in order to be treated a certain way). You pretty much have to say that it counts (or else you’re not treating *anyone* the same, not even treating adults the same as other adults), but if you do, it can end up affecting almost no real world cases.

        I would also recommend that if you do this, you exclude age of consent. If you even think of discussing it, you’ve created a place that will be populated by a few rationalists, and a hell of a lot of witches (not to mention the danger of being mistaken for one of the witches.) (Edit: This falls under 6) anyway.)

        • Aron Wall says:

          Since Ozy was talking about parenting, I assume it meant something like: “Parents of teenagers should give them nearly as much autonomy as they would give their adult children”.

      • Deiseach says:

        What age teenagers are you talking about? There’s a difference between thirteen and seventeen.

        And “treated like adults” includes the way some American states will charge minors as adults even for crimes carrying the death penalty*. If you’re going to say a sixteen year old has all the rights of being legally twenty-one eighteen (darn it, eighteen is still too young!) then they have all the responsibilities and risks as well.

        You can’t say fifteen year old Joe should be considered an adult if he wants to take drugs and parents have no say in his decision, but Joe is only a poor little child who shouldn’t be treated on the same level as an adult when he gets arrested for those same drugs.

        *I do not approve of this! Just in case you took away from this that I think fourteen year olds should get the lethal injections!

    • sarth says:

      I might be able to mount a defense/collaboration for:

      Gender diverse/feminist parenting

      teenagers should essentially be treated as adults

      less hysterical about teenage sexuality

      Depending on whether we could find a well defined claim.

      • Shannon Alther says:

        I would be willing to argue that teenagers should not essentially be treated as adults, depending on our definitions of “teenager” and “treated as an adult”. If you’re interested, my email is (ROT13) funaabanygure [at] gmail dot com.

    • fion says:

      Sounds like a really interesting one. I’d love to read an adversarial collaboration on some of those topics.

    • janrandom says:

      I’m so well aligned with this (I can’t find any single point I relevantly disagree with) that I would probably offer to work for Ozy’s side of any argument too.

      One minor question: “Free-range kids” and “Children should have more autonomy” are both given and I’m not sure what the distinction might be. Or is free range more about outdoors?

    • ADifferentAnonymous says:

      I won’t have time to collaborate, but I’m most curious about
      > Violent media probably does have short-term negative consequences

      My understanding is that the evidence is only for effects immediately after the exposure. To me this would suggest the parenting approach “pay attention to how your kids act after violent media, and restrict it if you notice problems”, rather than “don’t let your kids consume violent media”.

      So I guess I technically agree with that bullet point, but disagree with the impression I get from its inclusion.

    • AG says:

      Aren’t all of these individual claims subsumed by the greater debate on if parenting details matters at all? I thought the twin studies literature were fairly confident that anything outside of abuse is basically irrelevant to where the kid will eventually end up.

      Or rather, you cannot generalize parenting practices. They have to be tailored to each child’s unique personality. Some kids will respond better to certain practices/environments, other kids will respond better to the antithesis of those same practices/environment. This is probably what renders everyone into the noise band at the study level.

    • Cliff says:

      I think it’s quite likely that the only benefit to breastfeeding is now immune-related, not IQ. Formula is much more advanced than it was 40 years ago.

      • Douglas Knight says:

        Do you believe that breastfeeding had an effect on IQ even 40 years ago?

        • albatross11 says:

          Have people looked closely at the IQ/breastfeeding correlation w.r.t. educational / income traits of the parents? You could imagine it being a real thing, or being an artifact of the kind of mothers who do and don’t breastfeed.

          • Cliff says:

            Typically formula was for the high income, and poorer mothers breastfed, so if anything you would think this would go in the opposite direction. I don’t have any strong opinion about 40 years ago but I would assign >50% chance of an IQ effect at that time.

    • Matt C says:

      > pro-homeschool-regulation

      For most people favoring homeschooling regulation I would assume:

      a) they have no good reason to believe regulation is needed (for example, actual evidence that homeschoolers are clearly deficient in X way compared to their SES-comparable counterparts)

      b) if they have a), they have no good reason to believe that this could be solved with regulation (for example, previous, similar social service interventions that actually moved the needle the right direction with respect to X).

      In some cases I might agree that b) should be disregarded, but in most cases, I do not think identifying a problem is the same thing as having a good reason to believe politicians/regulation can actually fix the problem.

      Now, you being Ozy, I would not be surprised if you had some evidence for a) and maybe b). If you did, I’d be interested to hear about it.

      (I don’t think this is big enough to make a project out of. Mostly curious where you’re coming from, and maybe you know something I don’t.)

    • Shannon Alther says:

      Depending on what we mean by “teenager” and “treated as an adult”, my position is that teenagers should not essentially be treated as adults. If you’re interested, my email is (ROT13) funaabanygure [at] gmail dot com.

    • You don’t mention the issue of schooling/unschooling.

      Please note that the parent of an adorable baby who is a utilitarian or anything close should feel morally obligated to bring said baby to the next SSC South Bay meetup for me, and others, to adore.

      Saturday May 12th, 2:00, 3806 Williams Rd, S.J. CA 95117.

  9. lamaybe says:

    Not vaccinating is a legitimate medical choice from a personal and public health point of view.

    • Alsadius says:

      Point of information: Are you referring to adults not vaccinating themselves, or parents not vaccinating their children?

      • lamaybe says:

        I’m referring to parents forgoing all of the routine childhood vaccines (or choosing selective vaccination) as opposed to following the CDC schedule.

    • Ozy Frantz says:

      I’m interested in what specifically you mean by “not vaccinating”: the case for not vaccinating for polio in the US is much stronger than the case for not vaccinating for measles.

      • lamaybe says:

        I’m referring to parents forgoing all of the routine childhood vaccines (or choosing selective vaccination) as opposed to following the CDC schedule.

        • Aapje says:

          Are you assuming that the number of non-vaxxers is so small that you still have herd immunity or don’t you believe that herd immunity gives substantial benefits?

        • sclmlw says:

          It sounds like the real assertion/claim should be stated as, Not following the CDC vaccination schedule is a legitimate medical choice from a personal and public health point of view.

          That’s probably a much easier position to defend, and one that I’d be on board with defending, as an immunologist. However, if your position is that “avoidance of all vaccinations is a legitimate medical choice” I would say it’s fine from a personal perspective, but from a public health point of view it’s a disastrophe. As Aapje notes above, the only way you could substantiate the “from a public health point of view” part of that statement would be by demonstrating that public persuasion efforts are sufficient to maintain herd immunity.

          As a motte-and-bailey argument you could point out that we’ve consistently achieved herd immunity without compulsory vaccination (that’s the motte). But the bailey most anti-vax proponents I know want to live in is to claim that strong pro-vaccination campaigns should be gotten rid of, and not vaccinating should be viewed as normal and okay. Again, I’m fine with a few individuals doing what they want, and I won’t defend compulsory vaccination, since it’s clearly not necessary to achieve public health objectives. But if you want to defend getting rid of public health campaigns that favor vaccination, or say that not getting vaccinated should be normalized, I’m willing to do an adversarial collaboration with you.


          Immunology is my area of expertise.

          Claims I am willing to support:
          – Compulsory vaccination is not necessary to achieve public health objectives
          – The CDC vaccination schedule is far from ideal, and responsible people can vaccinate along a different schedule

          Claims I am willing to vigorously oppose:
          Vaccination as public policy is unwarranted
          Normalization of vaccine-free parental choices should be encouraged

          If you’re interested, you can reach me at my screen name on Gmail.

    • Zeno of Citium says:

      Pleased that someone is willing to put this out there even though I disagree with you, I’d love to see an adversarial collaboration on this. I don’t have the expertise to do the correct side, sadly, although maybe if *no* actual scientists, or at least people that studied medicine or epidemiology volunteer I have a lot of the background knowledge.
      What do you mean by “legitimate?” Do you mean something along the lines that (per below) forgoing all childhood vaccinations will lead, on average, to as good or about as good medical outcomes as using the normal vaccination schedule?
      Also, the big question for you: do you believe that vaccination can cause autism, and if so, have you read any of the enormous body of evidence that proves they don’t and that the originator of that claim is a literal fraud?

      • I’d love to assist the pro-vaccine epidemiology side if assistance is needed, since I’m probably qualified, but I’m too busy to do the project myself. Maybe we could tag-team that side for some of it?

        If you’re interested, you can find me on twitter, facebook, or gmail using obvious username.

      • lamaybe says:

        Zeno, I mean that the available evidence isn’t compelling that forgoing all childhood vaccinations would lead to significantly better medical outcomes in countries with greater economic resources like the US, and that are realistic reasons to think that childhood vaccines contribute to the epidemic of immune hyperactivity occurring in said countries with greater economic resources. There is no evidence that vaccines cause autism.

        • Zeno of Citium says:

          Thanks for the clarification, and this is interested but definitely out of my area of expertise – why the developed world has so many autoimmune diseases etc. is an interesting question (other than “people with these diseases used to just die”). I look forward to your team’s results if you get a partner.

    • melolontha says:

      I think you need to break this down into something more empirical and less value-laden: I can see a debate over ‘legitimacy’ foundering on irreconcilable disagreements about moral foundations (assuming it doesn’t founder on semantics first).

      • lamaybe says:

        How about

        In countries with greater economic resources, typical childhood vaccination leads to significantly better medical outcomes than not vaccinating, and does not contribute to immune hyperactivity.


        In countries with greater economic resources, typical childhood vaccination does not lead to significantly better medical outcomes than not vaccinating, and may contribute to immune hyperactivity.

        • rahien.din says:

          Are you pre-conceding some point that, in countries with lesser economic resources, typical childhood vaccination would lead to significantly better medical outcomes?

          If vaccines do produce significantly better medical outcomes, wouldn’t this necessarily consider or overwhelm any contribution to immune hyperactivity?

          Why raise the question of immune hyperactivity at all? That doesn’t seem epidemiologically or pathophysiologically tractable.

          • lamaybe says:

            Yes, I pre-concede that in countries with fewer economic resources typical childhood vaccination improves outcomes by significantly preventing the diseases they are supposed to prevent, diseases which have bad outcomes in the context of malnourishment & other public health issues. If someone wants to get into the debate with me, we can start looking at the tractability of the evidence for immune hyperactivity.

        • albatross11 says:

          I think the question needs narrowing:

          a. Would if be better for my children not to get their shots (or some shot), assuming everything else stays the same?

          b. Would it be better if everyone stopped having their kids get their shots (or some shot)?

          Those are two very different questions, and it’s quite plausible to me (as a non-expert with a layman’s interest in the subject) that right now, the potential side effects and rare bad reactions of taking some of the standard shots is not worth the benefit, because (say) M, M, and R are all pretty rare in the US anyway, but that if this behavior became widespread, we’d get commonly circulating diseases again and the incentives would shift.

          If that’s true, it’s exactly the situation in which you can justify a law requiring everyone to give their kids the vaccine–you’re staring at a prisoners’ dilemma and need to enforce the don’t-defect option on everyone.

          • lamaybe says:

            If we consider vaccines to be low risk to an individual, high benefit to a population, it’s pretty easy to argue that a given individual choosing to forgo routine childhood vaccines in the context of a high-vaccinating population is maximizing personal benefit, albeit at risk to the high benefit to the population. I’m proposing the second, more controversial argument; if most people in the US stopped getting routine childhood vaccines, what would outcomes look like? I’m proposing that we’d only see quite modest harms (at risk levels that many parents would be willing to tolerate) and quite possibly some benefits.

          • albatross11 says:

            Do we have examples where countries have stopped the use of some vaccine that was pretty effective on a still-circulating disease for some reason, that we could use to get a base rate?

            ISTR that there have been widely-used vaccines that got pulled from the market for awhile because of contamination or complications. Maybe starting from those, we could get a reasonable projection for what happens when, say, the MMR becomes optional and most parents stop giving it to their kids.

            We can also look at the easiest cases:

            a. Smallpox vaccine is given to the military and to researchers working with pox viruses[1], but not to anyone else. Military recipients are (so far) living in a world with zero smallpox, and I think the smallpox virus is relatively risky (it’s a live cowpox virus infection poked into your arm with a hooked needle, IIRC), so that defines one end of the tradeoff.

            b. Tetanus and rabies vaccines for humans are given for diseases that don’t circulate among humans. (Tetanus is caused by a bacterial toxin from a wound; rabies is spread by animal bites but I don’t think human-human spread is a big issue anywhere.) So there’s no herd immunity issue to worry about at all there. (IMO, you’d be a fool not to get tetanus shots every 10 years or so, and absolutely nuts not to get the whole post-exposure treatment (I think vaccine + antibodies) if you’d been bitten by a rabid animal.)

            c. Influenza vaccines are not very good–I think the best evidence is that most years, they maybe make you a little less likely to get sick with the flu, and make your infection a little shorter if you do get it. Flu evolves around the vaccine/existing immunity all the time, so it’s not clear how much herd immunity you get, either. Maybe this is a good place to start, because it’s likely to be a push either way?

            [1] Confusingly, chickenpox isn’t a poxvirus, it’s a herpes virus.

          • Cliff says:

            This might be totally wrong, but I always get the flu vaccine because I figure the more exposure to flu viruses the better, and anecdotally my reaction to getting the flu even before I take the flu shot for that season has improved dramatically since I started doing this. I.e. I used to be out of commission for 4 days, now symptoms pass in 1 day

        • sclmlw says:

          I’ll take the pro-vaccine side of this claim. I am a clinical researcher with a PhD in immunology. I’m not an epidemiologist, but I’m relatively familiar with epidemiological arguments in favor of vaccination in first-world countries. As an immunologist who primarily studied auto-immune and auto-inflammatory disease pathogenesis in grad school I have solid qualifications for debating that part of your claim.

          Feel free to send me an email at my username on Gmail.

    • Drew says:

      If I can suggest a refinement of ‘legitimate’: “Parents routinely make decisions that have more public health impact than the decision to vaccinate their children.”

      Skipping vaccines probably has some risk. But so does everything.

      If you can show me that vaccines matter less than something mundane — like sending kids to summer camp — I’d start to believe that the focus on vaccines is more a matter of special pleading than a sober analysis of risks.

      I don’t have enough of an opinion to argue the topic myself, but that’s how I’d expand ‘legitimate choice’ into something empirical.

      • Jiro says:

        We generally think that the acceptable amount of public impact from someone’s actions is less if the action accomplishes no instrumental or terminal goals. So I don’t think your framing works–even if we tolerate more risk from summer camp, parents actually think summer camp is good for its own sake.

        • lamaybe says:

          Aha! Therein lies the rub. I actually think there may be benefit to allowing your immune system to encounter the vaccine-preventable illnesses on its own.

          • Deiseach says:

            I actually think there may be benefit to allowing your immune system to encounter the vaccine-preventable illnesses on its own

            I’d broadly agree with this, and the amount of vaccinations young children undergo in their first five years of life have certainly increased drastically since I was a child. And I got the measles naturally, and it never did me any harm! (wibble wibble)

            But on the other hand, seeing how easily the kids in the service come down with all manner of ailments, and having children there with compromised immune systems for various reasons – what’s a common childhood ailment that one child will recover from could have those kids rushed off to the emergency department and needing intensive care. Little Johnny doesn’t get vaccinated and brings an illness in to preschool and little Susie has multiple allergies or a particular syndrome, it is not going to end well.

            So herd immunity is the way to go and mass vaccination is the way to do it. If you’ve got a fairly healthy seven year old you might be sanguine about letting them catch chickenpox, but if your seven year old has health problems which mean they get constant respiratory infections and have been on a regime of a cocktail of antibiotics since they were a baby to fight those off, you’re going to take a different view of the risks and of your neighbour upping those risks by letting their kid roam around and catch natural ailments which they then pass on to the other kids they interact with.

          • lamaybe says:

            Deiseach, something about the settings doesn’t allow me to respond directly to your comment.

            I mostly see your comment as a philosophical one; the question of whether an individual can choose increased chance of personal benefit that might lead to increased risk of harm to others, or whether we should make our own medical decisions based on the health status of others. It’s an interesting question, but not at the heart of what I’m trying to debate.

            What I’d like to get adversarial about is whether or not we have strong enough evidence to decide if mandatory routine childhood vaccination is on the whole substantially beneficial to public health in countries with greater economic resources.

          • ADifferentAnonymous says:

            >Deiseach, something about the settings doesn’t allow me to respond directly to your comment.

            SSC has a maximum reply depth.

          • Jiro says:

            I actually think there may be benefit to allowing your immune system to encounter the vaccine-preventable illnesses on its own.

            Maybe, but my point is that you have to argue it. You can’t get around the need to figure out which side is right by pointing to how many risks we are willing to tolerate–how many risks we are willing to tolerate is going to be affected by who’s right.

          • sclmlw says:

            The problem is framing this as “how many risks we are willing to tolerate.” A better frame is to consider to what extent the decision not to vaccinate produces a negative externality. Or to be more fair (since vaccination is an action you take, and not vaccinating is an action you don’t) to what extent vaccination produces a positive externality. You could consider both of these ideas in light of two proposed forms of collective/government involvement:

            1. Does the evidence support compulsory vaccination in a cost/benefit analysis?
            2. Does the evidence support non-compulsory promotion of vaccination as a public policy initiative in a cost/benefit analysis?

            I think the evidence is against the first, but strongly favors the second.

          • albatross11 says:

            Doesn’t this turn on how much weight you put on individual freedom? That is, it might be that most Americans would come to agree that mandatory vaccinations were a bad policy because the society-wide benefit is very small relative to the intrusion in personal liberty, but maybe the Chinese government would see things differently.

          • sclmlw says:

            Maybe at some point, but this really is an empirical question. For example, say we had a disease that kills 5% of people who get it. Let’s call it Sinfluenza. Knowing transmission rates versus vaccine efficacy, we calculate that in order to achieve herd immunity we need to vaccinate 80% of the population. By which I mean that at a vaccination rate of 80% we will not see outbreaks, because there can’t be a chain of transmission. In that case, the debate is either:
            a.) Can we achieve 80% vaccination in a way that does not negatively impact public opinion (since a democratic system will get rid of the vaccination policy if it’s obnoxious)
            b.) Is the method in (a) sufficiently acceptable to justify the benefits of vaccination.

            Again, this is all empirical. Say Sinfluenza vaccine campaigns are sufficient to get up to 65% vaccination, but for some reason people don’t like to get the vaccine. We might see epidemics from time to time, but with only 1/20 people who get the disease dying there’s not enough of a public concern that people change their behavior about it.

            So we consider to mandated vaccination. This would likely cause a backlash of people not supporting all vaccination on the grounds of personal liberty, so we may see decreased vaccination for other pathogens as a result of this policy. Then we look at the absolute number of people dying in the Sinfluenza outbreaks, and notice only 40 people die each year from them and consider that maybe mandatory vaccination is a net negative as a public policy. If there’s a 5% probability that the public will vote against public funding of vaccine and other public health research, but after mandatory vaccination that probability jumps to 35% we should probably not do mandatory vaccination. The risks are far greater than the potential benefits.

            Compare that with a public health campaign. Say a public health campaign increases Sinfluenza vaccination from 65% to 82%. Deaths due to Sinfluenze drop from 40/year to 1-2/year. Say we spent $380 million on the campaign, or $10 million per life saved. Is there a better place in public health where these dollars could have been more effective? Perhaps we could do some nutrition campaign for the same cost that saves 1,000 lives, or $380,000 per life saved. Suddenly the public policy doesn’t look like such a good idea, either.

            You’re right that the value of individual freedom itself is hard to quantify, and definitely deserves a voice in this kind of debate. But there are a lot of other factors that can be quantified in a meaningful way.

          • sharper13 says:


            The empirical number which I’d say is missing from your discsussion of what to discuss around a “Sinfluenza” vaccine, is what are an individual’s odds of a negative reaction to the vaccine itself.

            To make up some round numbers, if the vaccine reduces your risk of dying from Sinfluenza in the case of an epidemic by 50%, the odds of an epidemic are only 10% during the effective period of the vaccine, but the vaccine itself kills 5% of the people who get it, then even though the vaccine is effective, I wouldn’t take it. If you have someone who has in the past shown a negative reaction to a similarly formulated vaccine, I presume that would also increase their odds of reacting poorly to the Sinfluenza vaccine.

            In other words, I think any analysis has to also factor in the odds of various negative outcomes of being given the vaccine as well. For many vaccines you can (or at least used to) be able to get those from the CDC. For example, it’s been a negative odds proposition for a while (No cases originating in the U.S. since 1979) to get vaccinated against Polio using the live virus. For a few years before they switched from the live virus version, the vaccination itself was the main way people contracted Polio in the U.S. Recently, because of the world-wide use of the oral vaccine, 3x as many people get Polio effects from being vaccinated as get Polio itself.

            An interesting sideline would also be to compare disease contraction rates among vaccinated children vs. adults for whom vaccinations are projected to have worn off. My prior would be that for many of the recommended childhood vaccinations which wear off, there isn’t much actual herd immunity in the U.S. because adults rarely get additional vaccinations for themselves once out of K-12 school.

          • albatross11 says:

            One thing I always think when I see vaccine safety discussions is that we’re not discussing science so much as law and regulation. It’s absolutely possible to make unsafe vaccines–there are examples where vaccines had dangerous contaminants or nasty side-effects.

            When you say that vaccines on the normal schedule are reasonable, you’re saying you believe the people maintaining that schedule (public health authorities, drug regulators, drug companies) are doing a good job making sure the vaccines offer a good risk/reward tradeoff. That comes down to thinking they’re capable of doing a good job and have the right incentives to do a good job.

    • azantium says:

      I could be part of a group on this subject. I am currently pro-vaccination, but I have put a fair bit of time into the subject in the past, and it is apparent to me that the standard consensus position is manufactured from poor data, and there has been insufficient insufficient testing of vaccine aluminum to be as confident of its safety as we claim to be.

    • sclmlw says:

      Announcement of collaboration, for those interested in participating:

      lamaybe and I have agreed to do an adversarial collaboration on this subject, where he is planning to defend the following two statements (and I will take the opposition to them):

      1. In countries with greater economic resources, mandatory vaccination as public policy does not convincingly lead to the best public health outcomes

      2. In countries with greater economic resources, normalization of vaccine-free parental choices should be encouraged

      We’re working out some mechanism for pre-registration of sub-claims, before defending each position and exchanging literature. Given a broader interest in this topic, we would like to invite adjunct contributors to this subject, with the caveat that we don’t want to overwhelm one side by sheer volume. If you’re interested in contributing to the discussion, please post a comment below noting whether you would be willing to contribute for or against the first and/or second assertions above.

      • albatross11 says:

        This topic is interesting to me, but I don’t have time to contribute much formally. But I do wonder what “vaccine free parental choices” means. Does that mean “don’t bother with the flu shot, it doesn’t help much anyway?” Or “refuse the post-exposure treatment after being bitten by a known-rabid dog?” Or something in-between?

  10. peacetreefrog says:

    I’d be interested in reading a report from people on opposite sides of the political spectrum talk about Venezuela, and how it’s either the inevitable progression of socialism or not really true socialism at all.

    I’d also be interested in reading an in depth, multi-perspective discussion on the US vs more liberal, larger safety net European countries in terms of social services, quality of life, economic well being etc.

    I think either of those could be good opportunities for people on opposite ends of the spectrum to collaborate on. They seem less “persuasive essay topic” and more “here’s a summary of the facts” type topics.

    • sharper13 says:

      Having recently done a summary on it elsewhere, I’d be willing to take your first topic and turn it into a more specific “Socialist policies, including economic nationalization, in Venezuela are the primary causes of the country’s current economic troubles.

      I suppose someone on the other side of that would either be someone arguing a strict negative, or else arguing an alternate cause, such as the people involved, etc…

      • peacetreefrog says:

        I’d be interested in reading this summary (regardless of whether someone takes up the challenge of summarizing the other side) if you wouldn’t mind posting it.

        • sharper13 says:

          The below was originally written in response to an article blaming the Venezuelan economic issues on oil price declines, so you’ll see references to that. I’m posting this as-is, rather than rewriting it. In the four months since I wrote it, things have only gotten worse, in terms of lack of food and in terms of oil production (despite recently rising oil prices, up $10/barrel in that time), to the point where we’re seeing news stories about treason charges for oil workers in a futile attempt to get production back up at government-run PDVSA. How bad is it according to Reuters?

          “About 25,000 PDVSA workers resigned between the start of January 2017 and the end of January 2018, out of a workforce last officially reported at 146,000, Reuters reported last week. The resignations – including high-level professionals that are now almost impossible to replace – have only accelerated since Quevedo arrived, two dozen industry sources told Reuters.”

          I’ve never been to Venezuela. This isn’t original research, it’s collating from publicly available sources. I did run the results past three people who live in Venezuela and they agreed it describes what they’ve seen/experienced.

          The article contains some facts, but it also includes opinions and as Gilberto pointed out, it leaves many facts out, mostly about the government as related to the economy.

          Here are some additional facts and opinions to consider:

          From 1998 to 2018, oil production in Venezuela is down from 3.5 million barrels per day in December of 1997 vs 2 million in October of 2017.

          So what happened in the last 20 years? From Wikipedia :
          “After Hugo Chávez officially took office in February 1999, several policy changes involving the country’s oil industry were made to explicitly tie it to the state under his Bolivarian Revolution. Since then, PDVSA has not demonstrated any capability to bring new oil fields on stream since nationalizing heavy oil projects in the Orinoco Petroleum Belt formerly operated by international oil companies ExxonMobil, ConocoPhillips, Chevron and Total. Chávez’s policies damaged Venezuela’s oil industry due to lack of investment, corruption and cash shortages.”

          Probably just a fluke, though, right? I mean, steel production in Venezuela increased from 3400 tons in 1998 to about 4600 tons in 2008. The steel industry was nationalized by the Venezuelan government in 2008 and production declined to under 1600 tons. Huh, definitely a pattern forming. Similar stories of lower production and losses in the other industries after they were taken over: aluminum, cement, gold, iron, farming, transportation, electricity, food production, banking, paper and the media.

          How well does the government run the nationalized oil company, PDVSA? Reuters:
          “The output fall could not come at a worse time, with the economy in crisis and the socialist government struggling to pay its foreign debt.” and “Compounding the situation, another eight managers and employees of state oil company PDVSA in eastern Venezuela were arrested in recent days for fiddling production figures, chief prosecutor Tarek Saab told reporters.

          In a major corruption sweep engulfing the oil sector, about two dozen high-level executives have already been arrested in recent weeks, ridding PDVSA of much of its top brass.”

          Without the government takeover, even if oil companies were only competent enough to continue production levels and not grow them (as they’ve done previously over time), Venezuela would have almost twice as much hard currency coming in from oil sales.

          Have oil prices just completely collapsed? Well, not exactly. They’re lower than they have been recently, but that just puts them back at the same levels as they were in 2004 (in real dollars), still much higher than they were for the two decades before that.

          Instead, they chose to steal the companies from their owners and turn them into cash machines for their political allies. Studebaker comes up with three options, two of which are investments and one of which is actually spending/consuming, not “investing”. Guess which one Venezuela’s government chose once they took over the oil industry (and other industries)?

          Studebaker compares their policies to Saudi Arabia, but says Venezuela is more dependent on oil than anyone else. In fact, they’re 8th, with 7-8% of their GDP from oil. The UAE and Kazakhstan are about 14%. Saudi Arabia is 21%. Oman is 25%. Iraq is 28%. Kuwait is 30%. Angola gets about 34% of their GDP from oil production. (Stats from the World Bank and The World Factbook)

          Huh, funny you don’t hear about people starving and rioting in the street from the results of “lower oil prices” in all those countries. Actually, none of those other countries which are even more dependent in their economy on oil exports are even in a recession, let alone an economic collapse.

          Maybe the difference has more to do with how the government has run the economy it’s taken over (including the oil industry) than it does with a drop in oil prices.

          Studebaker gives the government lots of credit for improving poverty levels, etc… but the entire region improved similarly. In fact, Peru, Brazil and Panama all did even better at reducing poverty during the same time and none of them nationalized industries in the same way and spent the previous profits, so it’s unlikely that’s why Venezuela had the gains Studebaker brags about. Instead, reduction in poverty is the normal state of an economy which begins poor and grows over time.

          The number of private companies in Venezuela was 14K in 1998. In 2011 it was 9K. (ABC News) The government there has taken over and runs at least 500 companies and loses money on at least 70% of them. Some news reports say they’ve ruined thousands of companies. They’re 3rd to last (167/170) in the corruption index.

          So no, their problems aren’t just about oil prices. Their problems, including a big chunk of the oil revenue losses, are a direct result of the socialist government of Venezuela under Chavez and Maduro taking over large portions of the economy.
          I tried to fix the links for the repost, but if one doesn’t work right, let me know.

      • Watchman says:

        From experience of debating with socialists, I suspect that most counter-arguments would indeed deny the socialism of whatever caused Venezuela’s problems. This might be something of a trope but might also be a genuine argument that can be made: it’s not been made well in my presence, but if you’re going to find a socialist who can argue this in a way that is understandable to non-socialists, rather than rely on assertions about what socialism is, then SSC is probably a good place to look.

        • sclmlw says:

          My problem with the No True Scotsman claim socialists make of all failed socialist states is that they then proceed to argue the same principles that were the underpinnings behind founding failed not-socialism in the first place.

          Such that even were I to grant the premise, “Socialism didn’t destroy Venezuela, because that’s not True Socialism.” I’m left to ponder how the next socialist movement is planning to avoid the fate of Venezuela.

          If the pattern is, “persuasive tyrant takes charge by hiding his tyrannical leanings while simultaneously promising a socialist utopia” and you’re arguing we should follow some movement that’s promising a socialist utopia, you’ve got to show me how that movement is fundamentally different than all past movements that devolved into tyranny and ruin. How do I know the next socialist movement is really going to be True Socialism, and not more tyranny and ruin? That’s what they need to be preaching. That’s the fundamental problem they need to solve. Because when I read accounts from the old USSR, they seem at least as ideologically pure as today’s socialist and communists.

    • fion says:


    • belvarine says:

      I don’t foresee this going very well. “Socialism” is a huge umbrella and this will rapidly degenerate into an argument over true Scotsmen. Seriously there are a lot of pamphlets out there.

      Also, I’m not sure it’s possible to argue over this in good faith.

      Opponents of socialism can credibly argue Venezuela is the “inevitable progression” of socialism because, well, it happened. They would enter the discussion having won that debate.

      Advocates of socialism can credibly claim that failures in Venezuela are a set of data points to be considered during subsequent socialist experiments, so the next “improved” iteration must be the actual “inevitable progression” of socialism. This side would also enter the debate having already won.

      How are you going to keep these parties from talking past each other?

      • peacetreefrog says:

        I think it might be helpful in this case to flesh out how true of a Scotsman Venezuela is. I’m libertarian and very pro capitalism, and read Bryan Caplan’s debate on socialism with Elizabeth Bruenig. One of Bryan’s strongest arguments (to him at least), seemed to be that Venezuela is socialist, and turned out horribly.

        But that argument wasn’t that persuasive to me, because I don’t know anything about Venezuela. I think it’d be a valuable use of time to have two people on opposing ends of the spectrum flesh this out, and — unlike some other topics in this thread — I think there are a lot of facts to unearth about what happened down there that both people could agree on. Then people can make up their own minds — was this poor outcome a unique situation because of the people involved/the (non-applicable to other socialist situations) particulars of Venezuela or the inevitable result of some well-meaning but doomed to fail socialist policy?

        I don’t nec think it’d turn out in the style of alternating persuasive essay like some of these other more philosophical topics might.

    • honoredb says:

      I’m not really qualified to do analyze this myself, but the way I like to frame this is something like All else being equal and not knowing anything about how it will be spent, should we expect a marginal increase/decrease in a country’s level of state spending as a percentage of its GDP to improve or harm the welfare of its citizens? In other words, given a simple quantitative definition of socialism level and a simple quantitative definition of “good”, like life expectancy or something, is a higher socialism level good or bad?

      • albatross11 says:

        Wouldn’t you need to account for how much of the country’s GDP is currently spent by the government in there somewhere? At the extreme end, the Laffer curve shows that in principle, there’s a point where raising taxes costs you revenue. (Though whether that ever really happens in practice is not so clear.) If you assume more spending requires more taxes, it seems like this must put some kind of upper bound on when it could be good to raise government spending, even if the ideal government fraction of the GDP were very high.

        • honoredb says:

          Sure, it might be easiest to assume the current number is somewhere randomly distributed in the (US..Sweden) range (about 37%..50%) and that a 1% change has the same effect anywhere in that range, but that might miss some important effects. A similar approach could be, if both collaborators agreed that in a modern developed liberal capitalist democracy the ideal number is somewhere between 10% and 90%, to look for evidence about where that number is.

      • outis says:

        All else being equal

        But that is precisely the problem with so many disastrous public policies: the assumption that an economic intervention will change only the thing it intends to change and everything else will stay the same. Which is pretty much always false.

  11. reasoned argumentation says:

    Because of (5), please don’t research any topic that I would not be able to publish on SSC if you came to a taboo conclusion. If you want to do an adversarial collaboration on taboo topics, you can feel free to arrange it in the comments, but it won’t be considered an official entry, it won’t be eligible for prizes, and I probably won’t post it (I might link it if it’s posted somewhere else). If you’re wondering whether a specific topic is taboo, you can ask.

    Here’s a fun game – let’s count how many adversarial collaborations will be won by one side because the best arguments against them are taboo.

    This game could also be called “why do replication crises keep popping up?”

    • Aapje says:

      I assume that they will be abandoned rather than that one side will win. Or are you thinking about situations where a side won’t actually use the best arguments for their own side?

      • reasoned argumentation says:

        Or are you thinking about situations where a side won’t actually use the best arguments for their own side?

        Yes, that’s exactly what I’m describing.

    • Nearly Takuan says:

      That itself could probably be a thesis by which you could collaborate against someone, which doesn’t seem like it would be considered taboo.

      I don’t want to weakman your side of things, so please do correct if I’ve got the wrong read on this. But I also find myself very interested in learning what position you’re advocating. Right now, my best guess is that your claim would be something like:

      Weighing potential (ethical?) risks against potential (scientific? humanitarian?) benefits, investigating politically-incorrect topics (e.g. correlations between IQ, ethnic background, political philosophy, income, “life outcome”, autism or other end-of-bell-curve genetic trait, etc.) has nonnegative results on net.

      According to survey results, I am below the average level of intelligence for an SSC reader/commenter. I also work and study in fields that have almost nothing to do with this topic. I am, therefore, doubly underqualified to collaborate on something like this. I’d definitely read any report(s) that came out of an adversarial collaboration on this or a similar subject, though.

      • realwelder says:

        Such a collaboration would either be unpublishable here due to taboo, suffer from the very restriction it seeks to describe, or fail.

        For in establishing the taboo arguments are strong enough to overtake the orthodox view it would be necessary to make them sufficiently strongly and to agree with them. To avoid them or to make them in a sufficiently weaker form that the taboo is not broken would be to fall victim to the claim under discussion. And to disprove the taboo arguments is to invalidate the claim.

        • albatross11 says:

          There’s also another built-in skew. If you believe these topics should seldom or never be aired in public, then you’re probably not interested in providing a high-quality set of arguments about why, since that would weaken the decision not to air them in public.

          I mean, if you get someone making an eloquent argument that if the public knew the truth about [forbiddden topic with great social relevance] and what it implies, they’d support [insert terrible policies here], and it’s widely believed, that kind-of undermines the whole point of not wanting to publicize the forbidden topic.

          • Nearly Takuan says:

            If there’s a special topic that deserves to be made un-taboo because the benefits of knowing the Truth outweigh any harm done by the investigation, then I agree that there’s no way to talk around the case coherently without at some point presenting the case itself and making a strong argument for/against it, but I also think it’s a separate argument from the one I proposed.

            If instead you were to dig up older research, from before certain things were taboo or from cases where somebody didn’t care and got results and published them anyway, then I suspect you could have more of a conversation about what effect (if any) that had on those who were exposed to it. Here there’s still a risk of signal-boosting an otherwise-fringe argument that turns out to be surprisingly coherent, persuasive, and/or correct, but also causes those who read it to behave badly for some reason.

            These risks can be further mitigated by choosing research that is already fairly public, but then doing some meta-research on that to find out what influence (if any) ${taboo_article} had on ${audience}’s opinions.

            Then there are proxies. Things that are pretty well-studied, that deal with taboo topics, that seem not yet to be themselves considered taboo, at least in the mainstream (not 100% sure whether that can be inferred to mean it’s not taboo here either). As a specific example, consider the Implicit Association Test: there exists a group that supports research using IAT as a good way to investigate things that are otherwise difficult for people to admit to themselves or others even in a scientific context; also, there exists a group that perceives the first group as looking for an excuse to shout RACIST! at everything and silence The Discourse. If two individuals came together who agreed that IAT is/isn’t detecting something, but disagreed over whether it is/isn’t useful on net (or whatever), then neither side would really need to draw on taboo evidence to make their case.

            Another proxy could be subjects that aren’t yet taboo in this space, but might be considered taboo (or be surrounded in rhetoric that suggests it’s taboo) elsewhere. For example, people seem to talk pretty freely about IQ here (as long as it’s not entangled with other id-pol stuff), but there exist other communities that either feel that the subject is already too “tainted” by its history or believe (for one of several possible reasons) that even in perfect isolation the concept of IQ is dangerous. I think it’s possible for someone to abstractly defend a position like this in principle, even if they don’t agree with the exact argument being made, but I’ll admit this is pretty inconsistent with my general opinion of the ACLU.

            Or you can go broader still. After all, what’s so special about taboo subjects, specifically? If one position is that approaching Truth by t units is a pursuit of sufficient intrinsic value to be worth it up to some (high) cost/consequence threshold M, while the other is something ∈ {it’s too difficult to determine in advance whether a cost will or won’t exceed M; Truth alone isn’t worth it beyond a lower cost threshold N; the ratio t:M is flexible and requires definition by other functions or factors; …}, then two parties could choose different stances here and then pick a smaller set of non-taboo examples to use as case studies.

    • MugaSofer says:

      If you want to collaborate anonymously, you can just ask. But we already have anonymous taboo arguments for everything under the sun and they don’t seem that good.

  12. jonmarcus says:

    Do you plan any way to verify the positions of the collaborators? For instance, I’m in favor of gun control. How would you keep me from collaborating with a supposed gun right’s supporter, who miraculously finds herself convinced that my position was right all along? (i.e. with someone else in favor of gun control who only pretends to be my opponent)

    • Alsadius says:

      An adversarial collaboration being posted on SSC is not a big enough prize to be worth playing games to win, and anyone who’d try is almost certainly too ham-fisted to pull it off. (And even if they did, it wouldn’t meet the contest criteria).

      That said, if this catches on in a really big way – e.g., before a law is passed, you’re expected to have an AC or two about it – then yes, this would be a real concern.

    • Edward Scizorhands says:

      (I moved this comment from elsewhere because you beat me to it.)

      Scott could require some attempt to point out why they think they are a good advocate for their side. Even old forum posts here defending the idea. If they have never written anything about their side, would they be able to stand up for their side?

    • melolontha says:

      I was worried about this too, but I wonder if it’s a real problem? I think it would matter if someone prominent did this, because people would (with some justification) defer to their authority. (If someone I know to be really smart and to believe X turns around tomorrow and says that she’s changed her mind, I will probably update in favour of not-X, even if I don’t have the time/ability/inclination to fully understand her arguments. So if she was only pretending to believe X in the first place, she’s successfully manipulated my opinion.) But if someone with no pre-existing reputation tries this, I’d be silly to fall for it, unless someone like Scott gives it his seal of approval. And to earn that, I hope it would have to be about as even-handed as the honest adversarial collaboration it’s pretending to be.

      • melolontha says:

        Perhaps the bigger problem is dishonest or relatively incompetent people crowding out more competent, honest people who would like to take the same position. They could waste the time of an expert on the other side, and perhaps deter other people from addressing the same question (because ‘it’s already being done’). So I think it’s important to demand some signals of competence and honesty before agreeing to a collaboration, and not to worry too much about topic overlap.

    • Walter says:

      It’s a shot at a thousand bucks. Not worth getting shady over.

      • sclmlw says:

        I think it would be too difficult for a partisan collaborating in bad faith to produce a sufficiently convincing (but ultimately defeated) argument, such that Scott would be willing to grant the essay first prize (or even a close-second-he-really-wanted-to-reward prize). So the money does incentivize true collaboration, without requiring external vetting. Otherwise, you risk all your hard collusion work just getting tossed out, with no real chance of the $1,000 reward.

    • Watchman says:

      Are there that many people who read SSC (which in itself implies a moral flexibility not apparent in the sort of missionary personality this scenario might appeal to), can present themselves as a reasonable advocate of an opposing position to their own, including mastering the often different language involved in adopting that position, and would feel there was value in doing this to advance an argument only within the community of readers of SSC? And considering this is SSC, how many of the tricksters supposed adversaries would actually accept the capitulation anyway? This looks to be unlikely to me unless I’m missing something about the community here. Remember that this is one of the few places on the Internet where readers come for open debate with openish minds.

  13. wilarseny says:

    I would be interested in an adversarial collaboration on some U.S. legal topic. I graduated from a CCN school a few years back and am in private practice; I wouldn’t characterize myself as an expert on anything except maybe the intersection of U.S. trademark law and internet domain names, but have more than passing familiarity with a number of legal topics. No particular expertise with empirical study of law, but some interest. Would expect a collaborator to also have graduated law school.

    Potential topics I can think of that might fit Scott’s criteria:

    Do Originalist methodologies lead to more consistent judicial decisionmaking?

    Do plaintiffs fare better, worse, or comparably in state/federal court vs. arbitration for non-class actions? In a particular type of claim (e.g. labor/employment claims)?

    Is there bias against men in family law proceedings (divorce, custody, etc)? [This seems like probably the best one that I can think of for the content of this blog]

    I am sure there is empirical research on each of the above, so I don’t think we would need to reinvent the wheel on any of these topics. I am pretty gray tribe overall but tend to lean more conservative when it comes to jurisprudence/rule of law/constitutional claims, etc, so I would be looking for a left-leaning collaborator (assuming that we end up disagreeing on the base claim, whichever we would pick). For timing, I would want to take a couple of months to look into any of these topics, not least because I have some rough work deadlines in the next few weeks. Let me know if anyone is interested!

    • Zeno of Citium says:

      I will signal boost this to some lawyers I know that have a surprising amount of time on their hands. Do you have positions already on the topics you’re interested in?

      • wilarseny says:

        Thanks, Zeno! I am also surprised to hear of lawyers with time on their hands.

        Sure – I’d stake my initial positions as follows:

        Do Originalist methodologies lead to more consistent judicial decisionmaking?
        Though the question is not well-defined, I would guess yes.

        Do plaintiffs fare better, worse, or comparably in state/federal court vs. arbitration for non-class actions? In a particular type of claim (e.g. labor/employment claims)?
        I have some familiarity with both the empirics and anecdata here, and I would guess that plaintiffs win less in arbitration, but win larger judgments when they succeed.

        Is there bias against men in family law proceedings (divorce, custody, etc)?
        Pretty unsure on this one, but my guess would be that men pay more in alimony than women and are granted custody less often. If this is true, though, may be hard to know whether this reflects bias in family law courts, or can’t be disentangled from underlying differences in pay, employment, criminal conviction rates, etc (which may themselves reflect bias in one direction or another, but that seems outside the scope of the question presented).

        • rlms says:

          It would be odd if men did not pay more in alimony without adjusting for income, since they earn more (on average, and the gap should be bigger in divorcing couples which are more likely to have lower income women). Likewise, it would be odd if women were not granted custody more often than men, since presumably they request it more (I expect they are also granted it more even after adjusting for that though).

          • wilarseny says:

            That sounds right to me. If we need to further refine the claim for the purposes of an adversarial collaboration, I would guess that even after adjusting for income, education, criminal convictions, and whatever other objective criteria one might consider, men are still paying x% more than women in alimony and being awarded custody y% less often. In other words, I would guess that there is some systemic bias that remains even after disentangling the possible cofounders (though admittedly I would not be surprised if the effect size were small, or turned out not to be statistically significant).

        • Aapje says:


          The data shows that men seek custody less often, but that they do get granted custody about equally often if they seek it. This can be because:
          1. men are interested less in getting custody
          2. men or their lawyers correctly predict a bias and don’t waste their time/money on hopeless cases
          3. men or their lawyers incorrectly predict a bias and the courts are actually fair

          The last one seems least logical, because then one would expect that men would be granted custody more often, because that prediction would surely not be random, but would take the chance of victory into account.

          The former seems to not be the case: “Policy makers could benefit from considering that similar to fathers in other studies (Arendell, 1995; Maccoby, Buchanan, Mnookin, & Dornbusch, 1993), these fathers expressed a clear desire to have shared physical custody.”

          So the second seems most likely, although a combination of effects may also be the cause. I’m not aware of sufficient evidence to quantify this, however.

          However, I would actually argue that the legal standards which are used by the courts cause severe disparate impact, because of the way in which gender roles work. So even if the courts were to be unbiased in applying the rules, there would still be huge bias, as the rules themselves have a gender bias (“In its majestic equality, the law forbids rich and poor alike to sleep under bridges, beg in the streets and steal loaves of bread.”).

          • a reader says:

            However, I would actually argue that the legal standards which are used by the courts cause severe disparate impact, because of the way in which gender roles work. So even if the courts were to be unbiased in applying the rules, there would still be huge bias, as the rules themselves have a gender bias (“In its majestic equality, the law forbids rich and poor alike to sleep under bridges, beg in the streets and steal loaves of bread.”).

            Gender roles have a basis in biology – women are (on average) more suited to take care of children (more people-oriented, more empathy), and more in charge of their care, so I think it’s quite normal that courts give the child to the mother more often than to the father, especially when the child is not a teen (in my country afaik a child older than 10 can choose the parent). The interest of the child is the most important here.

          • Aapje says:

            I think that men (on average) tend to be more skilled at some parenting tasks and less skilled at others, but that both are important.

            Less empathetic parenting can for example be helpful to make the child more resilient & independent. More systemizing parenting can help the child better understand the world, rather than just ‘feel’ it.

            I think that it is a mistake to assume that children benefit from just experiencing one type of parenting style.

            I would argue that shared parenting should be the default.

          • rlms says:

            I don’t think a study of 14 men who responded to an advertisement that stated they were particularly looking for “men who had maintained involvement with their children” is very strong evidence for anything (except the claim that more than zero divorced men want to see their children, which I don’t think anyone is disputing).

          • Aapje says:


            Maccoby, Buchanan, Mnookin, & Dornbusch, 1993 is a 3-yr longitudinal study of 1,124 divorcing families.

            But you are correct that the evidence is weak.

          • Cliff says:

            Obviously huge selection bias, but children of single men have WAYYY better outcomes than children of single mothers. Like children of single men actually have many outcomes better than children of two-parent households and children of single women are fifty times worse, that sort of thing.

          • rlms says:

            Sure, but I was talking about the study you linked. I can’t find Maccoby et al. anywhere so I don’t know what it’s claiming.

        • RC-cola-and-a-moon-pie says:

          I would also be careful of the originalist one from the other side. We really don’t care about “consistent” rulings per se. Here’s a decision rule: The party that comes first alphabetically wins. That decision rulewill produce rigorously predictable, consistent outcomes across all manner of hard cases. We want to be originalist because originalism is actually correct, not becaus it provides greater consistency than some postulated alternative.

          • Brad says:

            Yeah, I’m not sure how a collaboration on that one would be particularly productive. Besides your point about consistency:

            – You’d basically have to limit yourself to the Supreme Court. For every other court the obligation to follow binding precedents and different levels of respect for that obfuscate the effect of underlying judicial philosophies.

            – As to the Justices, while I’d argue that every one from those appointed by Reagan on has been influenced by originalism, there’s never been a working majority of dedicated orginalists. That means to find uncompromising originalist opinions you’d need to look at pluralities, concurrences, and dissents. We’d expect those to be different from majority opinions even without any differences in judicial philosophy.

        • Deiseach says:

          Is there bias against men in family law proceedings (divorce, custody, etc)?

          Completely unqualified opinion here but I would say:

          (1) Certainly regarding custody, yes.

          But in large part this was a correction (over-correction if you like) to what went before, where men were automatically granted custody of their children in a divorce or separation (largely because pre- no fault divorce, divorce could only be obtained for a handful of reasons which had strong connotations of moral turpitude attached). An abusive husband could hold the threat over his wife’s head that “you try to leave me/divorce me, you will never, ever, ever see your children again till the day you die”. A woman might be very reluctant to leave a bad marriage where she knew the children would be left to the father’s mercies and she had no recourse in law until the 1839 act was passed.

          I think there were also a lot of cases of men divorcing and then re-marrying to start a new family with a new spouse (who was probably already lined up and hence the divorce) and so not looking for custody in the first place, at least in previous decades. How it shakes out now, I couldn’t say.

          The pendulum probably has swung to the other extreme, where it takes extraordinary circumstances for fathers to be awarded sole custody.

          (2) That being said, and basing the following on limited experience with not a really stellar sub-set of society, the cases I’ve seen where fathers want custody are guys trying to game the system. Most examples were men who had no interest whatsoever in the child/children they had produced (often but not always by different women) until they realised/thought they could obtain something by it. The stand-out example but not the worst or only one was the guy who claimed he was an artist and said he needed social housing with an extra room for a studio (under the regulations as a single man he was only eligible for one-bedroom accommodation). Having been refused and having fought to a standstill, all of a sudden he comes in claiming he needs another bedroom because he’s seeking joint custody of his nine year old daughter and if granted access will have her to stay at weekends.

          That this would conveniently leave the room empty for five days of the week and he could then use it as the studio he claimed he needed – well, that was just a coincidence, right? Up to the moment he was turned down on grounds of “you’re a single guy”, there was no evidence he was involved in the child’s life whatsoever (and in social housing they do look for this to be able to make need-based decisions on allocations).

          Women do this too, to be fair. There’s a lot of shitty parenting out there where people use their biological/fostered/nearest family member taking care of them kids as game chips to get something more out of the system. But generally it was the guys suddenly discovering an interest in parenting when they needed/wanted something more than they were otherwise entitled to.

          I’m jaundiced, because I’ve seen cases where having the kids raised in the woods by wolves would turn out better than the shitty way they were being treated by family.

          So I suppose I incline towards “yes men can be subject to bias but in the main it works out”.

        • Zeno of Citium says:

          Alas, none of my lawyer friends were interested. Hoping this happens though.

    • Anonymous Bosch says:

      Do Originalist methodologies lead to more consistent judicial decisionmaking?

      This is difficult if only because you run into the no true Scotsman problem. Is the decision making inconsistent because of the originalist method itself, or because humans are imperfect and originalism was abandoned in this case?

      • wilarseny says:

        Definitely agree that this is the most ill-formed of the questions, and would require a fair amount of specification to get something workable. But I would imagine legal academics have already done a fair bit of work trying to figure this out and developing methods to measure it.

        • Anonymous Bosch says:

          But I would imagine legal academics have already done a fair bit of work trying to figure this out and developing methods to measure it.

          I don’t read law journals that closely but I’m not really aware of any method of quantifying originalism. I’m not even sure how you’d do such a thing.

          • wilarseny says:

            I worked as a research assistant in law school and spent some of that time coding judicial decisions and bios of judges for a paper about whether political donations reflected more/less in the decisionmaking of judges. The professor would have a few different people working on the same dataset independently, and would then combine their results. I… admit I’m a little skeptical of this method even (or especially!) having done it, but it’s at least one way to take a qualitative corpus of judicial opinions and convert that into a quantitative dataset. In my professor’s research, we were only looking for liberal/conservative, but you could probably design a more rigid coding set just for detecting originalism (e.g. only look at judges who explicitly declare themselves as originalists, or opinions matching a pre-defined set of search terms (“original meaning” + whatever, citations to Websters 1760 edition, etc]) and then manually code those. May not be the most reliable, but it’s at least a sheen of empiricism. Would not be surprised to learn that there are better methods.

    • cassandrus says:

      Regarding the question of arbitration, the emerging consensus among the sophisticated plaintiffs’ bar appears to be that arbitration is as good if not better than courts for larger business v. business disputes. Whatever you lose in terms of discovery are (more than) made up for by constrained appellate rights and greater speed. This view is pretty new—most plaintiff-side practitioners were of the view until the last few years that arbitration was the kiss of death. Conversely, entities that tend to be on the right side of the v. in these big business-business disputes are suddenly a lot less enthusiastic about sticking them in every contract.

      Can’t speak to other contexts such as labor. I’ve heard that FINRA arbitration can be a total disaster….

    • Patten says:

      I cannot tell if we disagree enough to have an adversarial collaboration, but I’d be very interested in doing one about On the net, employees are better off in arbitration than trial court with me taking the side that employees are on the net better off in arbitration. I’m a T14 law grad with employment law/HR experience, but only in the public sector, so I’m interested in seeing how the private sector differs. We’ll also need to decide if union arbitration should be included or not in this article, but I’m willing to argue this side either way.

    • RamblinDash says:

      Long time lurker. I registered to reply to this post! I am very interested in doing a collaboration on:

      Do Originalist methodologies lead to more consistent judicial decisionmaking?

      My strong prior is that Originalist methodologies do not cause consistent judicial decisionmaking.

      I would want to start our collaboration (if you want to do one) by more precisely defining our research question. I am a 2016 graduate of a T14 law school and participated in journal research while there, and can prove privately if necessary.

      • toBoot says:

        I’m kind of interested in this, but the topic seems kind of problematic. My prior is that Supreme Court Justices who self-identify as originalists tend to write opinions whose dicta and holdings are more philosophically consistent with their dicta and holdings in other decisions than justices who do not self-identify as originalists. Is this what we mean by judicial decisionmaking?
        Because if it is, then I think it’s almost tautological. Originalism and Textualism/strict constructionism are methodologies aimed at philosophical consistency. Justices like Breyer who consider themselves pragmatists/living constitutionalists are not aiming as much for philosophical consistency as they are trying to understand what the law was getting after in a wholistic way so that it can be applied to circumstances that were not contemplated when the law was passed.
        Also, I kinda feel like this has already been done by people who are way more expert than I am.
        I’m a 2010 graduate of a top law school with some time on my hands, and would be interested in something legal-y. Though one of the things that is kinda lovely about our legal system is that it’s got plenty of adversarial collaboration built into it, so it might be harder to find a new topic.

        • RamblinDash says:

          Also, I kinda feel like this has already been done by people who are way more expert than I am.

          I hear that.

          My prior is that Supreme Court Justices who self-identify as originalists tend to write opinions whose dicta and holdings are more philosophically consistent with their dicta and holdings in other decisions than justices who do not self-identify as originalists. Is this what we mean by judicial decisionmaking?

          This isn’t quite what I was getting at. When I said I don’t think originalism causes consistent judicial decisionmaking, I meant to convey that I don’t think that knowing a judge self-identifies as an originalist really gives you additional information about the likely outcomes of cases if you already know how liberal or conservative the judge is, appointing president’s party being a (noisy) proxy therefor.

          One reason for this is that there are many forms of originalism. Some forms of originalism really would constrain judges and thus knowing the judge adhered to that form would be valuable bayesian evidence even given the appointing president or some other measure of political ideology. However, such forms of originalism have been generally discredited as normatively unjustifiable and so nobody adheres to those. Think what’s (pejoratively) called original-expected-applications originalism–the idea that if the framers of the provision didn’t think/intend it to apply to a specific factual scenario, then it doesn’t.

          The forms of originalism most popular among scholars and judges today, most typically public meaning originalism (AKA semantic originalism), are insufficiently constraining to have a causal impact on the outcomes of cases. It may affect the way the opinions get written but in my view probably does not affect the way a judge votes.

          Additionally, the self-identified originalist judge sometimes just ignores originalist concerns when they don’t lead to preferred outcomes. This is most striking in affirmative action cases.

          Finally, there is a kind of originalist two-step, where a judge will select different (mutually inconsistent) forms of originalism as help the judge reach ideologically preferred outcomes. For example, you have lots of judges testifying at confirmation hearings that they are just applying constitution as written, or justifying a vote in some cases (think same-sex marriage) as basically original-expected-applications. However, in different lines of cases the judge will insist on applying original public meaning to reach an outcome which would not be reached on original-expected-applications.

          For all these reasons, I don’t think a precommitment to originalism has a causal effect on the way a judge votes (and thus the outcome of cases), although I don’t dispute that it may affect how opinions get written.

  14. salimfurth says:

    Here’s a separate proposal (I won’t do two): in which policy areas should the U.S. imitate Switzerland? There’s a factual component and a persuasive component, and the adversarial collaboration should be specifically helpful in this case, since it will help weed out Swiss policies that merely move us toward one or other pole of our own political spectrum.

    Why CH? I think Switzerland is interesting and underused as a comparison case (relative to, say, Sweden and the larger EU countries), and I’d appreciate the excuse to learn more about it.

    My partner would be someone left of center, and likely focus would be on economic policy, but not exclusively. Direct democracy would be an important component.

    • Chlopodo says:

      I misread that as “invade Switzerland” at first, and momentarily thought you were proposing a much more entertaining topic.

    • shakeddown says:

      I might be up for that.

    • zluria says:

      To make a concrete suggestion, I would be interested in having a collaboration about the Swiss educational system.
      At 8th grade, Swiss students have to take a standardized test, and only the top 1/3 or so go on to study at a high school that prepares them for college. The rest choose their future profession and go into an apprenticeship. By age 18, they are certified professionals.

      This system has many advantages, some of which I’m sure will occur to you immediately. Should the US adopt it?

  15. Freddie deBoer says:

    The Soviet Union was not an authentically Marxist or communist government.

    • Aapje says:

      Presumably, that would be an analysis to what extent Marxism-Leninism (mis)matches with Marx’s writings?

    • jonm says:

      Would you consider the view that sustained Marxist/Communist government is incompatible with human psychology/economics to be compatible with that statement?

      • Freddie deBoer says:

        I think that’s the kind of larger-bore idea that would get too big and shaggy. Trying to come up with a good Marxism one that won’t end up being “good or bad,” even though I’m firmly in the good camp.

        • Gilbert says:

          Throwing out some fairly specific anti-communist talking points I might be willing to argue for:
          – Marx and Engels basically saw the collectives as fundamental and individuals as exemplars of collectives. This is at tension with a standard liberal account where individuals are fundamental and collectives are collections of individuals.
          – Commiting atrocities to further future communism is therefore not fundamentally add odds with Marxism-as-meant-by-Marx. Of course Marxists could still instrumentally argue that the actual historical atrocities didn’t actually help with bringing communism about.
          – On any “originalist” reading of Marx and Engels, the labor theory of value and its particular application to work power (“Arbeitskraft”, the ability to do work as opposed to actual work, don’t know what the usual English translation might be) are critical to their economic analysis. By “originalist” I mean what they actually meant and were contemporarily understood to mean. (I am aware modern Marxists sometimes disagree.)
          – The labor theory of value is, depending on how you want to interpret it, either vacuous or empirically false. An actually somewhat useful theory (modern economics after the marginal revolution) has since arrived.

          These might be too specific though, I don’t think they leave sufficient material for an essay of Scott’s desired length.

    • jdly says:

      I think it would be hard to find someone to argue that the system in the Soviet Union met the definition of communism as laid out by Marx. Instead, I think the position is that the people in charge were genuinely trying to implement communism as laid out by Marx, but the incentive structure that resulted caused the whole thing to go wrong. Furthermore, any future attempts to implement Marxism will inevitably run into the same problems regardless of the intentions of the people doing the implementation. Thus, communism has been tried even though nobody actually created Marx’s system.

      I don’t know how useful this comment is but maybe you can find something or someone to argue against?

      • bbeck310 says:

        Agree with this critique; maybe the better debatable topic is: The Soviet Union’s problems were/were not caused by Marxism.

        Not sure I have the chops to debate it, but I’d be willing to take a shot from the anti-communist side if no one else wants to (i.e., problems were caused by Marxism).

      • sclmlw says:

        It sounds like the point you’re trying to argue is, The case of the Soviet Union demonstrates that communism, as outlined by Marx, cannot be created in the real world.

        This is the fundamental complaint I have about the No True Scotsman argument that dismisses all prior communist movements. “They didn’t do it right, we’ll do it right” kind of glosses over the details of what “they” did wrong and why. And those details are really important if you want to avoid more tyranny and ruin. The benefit of this kind of adversarial collaboration would be to outline where the Soviet Union went wrong, and to consider what elements of current communist theory need to be reevaluated or reformed in order to make it more than a purely hypothetical political philosophy.

    • Michael Arc says:

      How about the New Deal? I think that it was more Marxist than the USSR, but by the end of WWII, I think that they were both control systems.

    • Erusian says:

      I will argue that the Soviet Union was widely recognized as Marxist by the majority contemporary communists and that the ‘not truly communist/Marxist’ narrative was developed largely by anti-Soviet communists. Despite this, the position was not adopted by the majority of Marxists until after the fall of the Soviet Union. I will further argue that contemporary recognition by Communists better defines what is Communist than arguments from abstract theory.

      • sharper13 says:

        +1 for you to be the counter-party on this one. 🙂

      • Lyle_L says:

        I think that topic would lead to a primarily theoretical argument about historical definitions, whereas I’d be more interested a debate around what a contemporary Marxist revolution would look like.

        So I’m more with Jdly, that the counterpoint to Freddie’s statement would be more along the lines of A totalitarian regime similar to what occurred in the Soviet Union is the likely result of an authentically Marxist government

        • Gazeboist says:

          Probably better to stick to the “attempt to create an authentically Marxist government” formulation in that case.

        • Erusian says:

          You’ll notice there’s a meta-level question (what defines something as Communist) and an object level one (does the USSR qualify). I think without tackling both you’ll end up arguing over definitions. By forcing an explicit definition and using the words of people at the time, you can reach a more definitive answer. There is a positive, definitive answer to the question: Did Comintern ever call the Soviet Union Marxist.

          Also, there’s something ironic about accusing me of being historically focused when we’re discussing a regime that disappeared thirty years ago.

          But people are definitely free to pursue whichever they think is better.

      • Hoopdawg says:

        The truth value of your first proposition is largely determined by the vast amounts of people conforming to the ideology of their own country. My take: false since the Sino-Soviet split. (Also, in serious danger of turning into an ad populum fallacy.)

        The second one basically only requires deliberation about the nature of language and semantics. I’d venture it’s simultaneously true for layman usage and false for any kind of meaningful formal scientific/philosophical discussion (like, say, in the kind of adversarial collaborations we’re currently talking about), where strict theoretical definitions must be agreed upon by both parties at the very start.

        EDIT: a few general thoughts about this whole chain, in order not to litter the comments with several small posts:
        – A “communist government” is an oxymoron.
        – None of the “communist” states describe or described themselves as such in any capacity. (Though some of them had parties nominally dedicated to eventual establishment of communism. But the parties at the very least understood their own ideology and did not make claims that would obviously contradict it.)
        – Economic policies of governments can and should be discussed on their own merit. (There is no need to introduce the nominally claimed ideology of the people in charge into the equation.)
        – Regardless of not being communist in any meaningful way, and regardless of being totalitarian hellholes both Soviet Union and China are examples of economic success achieved despite (or perhaps because of) going against liberal economic orthodoxy.
        There is no intellectually consistent position that concedes that modern China is capitalist while denying the same categorization for Soviet Union at any point of its history.

        • Erusian says:

          The truth value of your first proposition is largely determined by the vast amounts of people conforming to the ideology of their own country. My take: false since the Sino-Soviet split.

          You’ll have to explain how it relies on the vast majority of people conforming to their country’s ideology. Because as far as I see it doesn’t. I said the vast majority of Communists, not the vast majority of Russians.

          (Also, in serious danger of turning into an ad populum fallacy.)

          Please explain how. At best it’s an appeal to the authority of expertise and self-description. At no point did I talk about the general public.

          The second one basically only requires deliberation about the nature of language and semantics. I’d venture it’s simultaneously true for layman usage and false for any kind of meaningful formal scientific/philosophical discussion (like, say, in the kind of adversarial collaborations we’re currently talking about), where strict theoretical definitions must be agreed upon by both parties at the very start.

          Again, I did not say identification by people or Russians or anything like that. I said by Communists. You are venturing to use theoretical definitions to declare meetings of thousands of expert, degreed, influential Marxist theoreticians mistaken. While that might be the case, their definitions must be considered. If you are declaring them wrong simply by saying your definitions are right and theirs were wrong, then you’re effectively divorcing yourself from the movements that ran all actual Communist governments and claiming their name. While you might be correct in some spheroid chicken of uniform density sense, it’s certainly not practical, falsifiable, or real world knowledge.

          – A “communist government” is an oxymoron.

          Any form of Communism, in a practical sense, necessitates a government.
          No non-governmental form of communism has ever existed on a national level.

          – None of the “communist” states describe or described themselves as such in any capacity. (Though some of them had parties nominally dedicated to eventual establishment of communism. But the parties at the very least understood their own ideology and did not make claims that would obviously contradict it.)

          Sort of true. You’re right on the ideological front. On the other hand, the All Union Communist Party has ‘communist’ in its name, doesn’t it? That counts as describing. Likewise, we call certain Christian sects Millenarian despite the fact none of them believe Christ has actually returned yet.

          – Economic policies of governments can and should be discussed on their own merit. (There is no need to introduce the nominally claimed ideology of the people in charge into the equation.)

          Agreed. I’d argue that Leftists are as guilty of doing this as Rightists, but perhaps you’d agree there.

          – Regardless of not being communist in any meaningful way, and regardless of being totalitarian hellholes both Soviet Union and China are examples of economic success achieved despite (or perhaps because of) going against liberal economic orthodoxy.

          I’ll take you up on the first bolded proposition. China and the Soviet Union are examples of how deviating from economic orthodoxy and free market capitalism lead to suboptimal economic outcomes, except where such outcomes are defined to include other metrics than solving the problem of scarcity, wealth creation, and efficiency.

          There is no intellectually consistent position that concedes that modern China is capitalist while denying the same categorization for Soviet Union at any point of its history.

          I mostly agree. China is not capitalist.

    • Tracy W says:

      I’ll take you on that, though I predict we will arrive at “they tried to be but reality got in the way.”

      • Reasoner says:

        Right. Instead of a collaboration over what counts as “authentically Marxist”, maybe it’d be more useful to collaborate over what the fall of the USSR should teach us about policy period? For example, if the fall of the USSR was due to random misfortune and the same plan could have worked under other circumstances, that seems worth knowing. If the fall of the USSR teaches that certain quasi-Marxist systems don’t work (but others still might work), that also seems worth knowing.

    • Levantine says:

      Personally, I’m always confused by the phrase ‘Marxist government.’ Marx is mainly known as a student of capitalism. He made predictions about capitalism’s replacement by socialism and communism. He was also a philosopher, a poet, a profuse writer who, unsurprisingly, contradicted himself during his writing career. He expressed himself only a little about communism and socialism. Two Marxist scholars can come to mutually contradicting conclusions, without that being outrageous.
      The Soviet Union’s system changed substantially between the 1920s and the 1950s, as well as between the 1950s and the 1980s.

      I’m just willing to follow a discussion on this, and eventually give input, for which I expect neither to be accepted, respected nor financially rewarded.

      • Freddie deBoer says:

        Marxism is the why; communism is the how. Communism is a series of cascading worker revolutions that destroy the state, kill god, and replace capitalism. What follows is a system of semi-autonomous workers’ collectives that operate under egalitarian democracy under the principles of each working according to their ability and receiving according to their need. It’s a stateless system of government but a system of government nonetheless.

        • albatross11 says:


          Is that a definition or a prediction?

          I mean, I might define a proper Catholic theocracy as one in which the Pope, bishops, and priests use the power of the Church to see to the well-being of everyone, Catholic or not, in a joyous future of peace, prosperity, and freedom. But it may still be that in practice, putting the Catholic church in charge of the government generally leads to inquisitors warming up the hot irons to use on heretics or popes calling for crusades against various heathens.

    • LadyJane says:

      I’m worried that such a discussion would end up devolving into a debate over semantics, regarding what does or does not count as authentically Marxist/communist. In order to prevent that, I would frame my counter-argument as such: The majority of the Soviet Union’s political and economic policies, include those that failed or resulted in violations of human rights, were a result of the government’s attempt to implement Marxist ideas. As a secondary claim, I would add: The Soviet Union implemented Marx’s ideas as well as any society possibly could, given the economic and technological conditions of the time, and economic/technological conditions have not yet changed enough that any modern attempt to implement those ideas would fare any different.

      I’m currently writing a thesis on the role that ideology plays in driving the actions of modern nation-states, so this is very well within my wheelhouse. If you’re interested in doing a collaboration, send me an e-mail at While I disagree with a lot of your views, I’m nonetheless a huge fan of your work and I think you have a lot of great insights, so I would be honored to collaborate with you on this.

      • Freddie deBoer says:

        You mean an impoverished feudal economy that was ravaged by two world wars to a degree that exceeded that of any other warring nation was the best possible attempt to achieve a system which explicitly states that a healthy and advanced capitalist economy needs to exist beforehand? But there I go arguing.

  16. christhenottopher says:

    OK here’s one that seems reasonably constrained I’m interested in.

    What is the effect of low skill immigration on the effectiveness of government institutions (defined by solvency, level of corruption, and user satisfaction)?

    Me: I am a very pro-immigration (even Open Borders supporting) person and libertarian-ish in overall politics.

    This is not a discussion of whether low skill immigration is an overall net good or bad, that’s way too broad, just on the question of their effect on government institutions.

    • NoRandomWalk says:

      I would be interested in taking the other side of this. Contact me at

      • christhenottopher says:

        I’ll send an email of what I’m thinking right now and we can see if we can make this work.

    • decodyng says:

      I have a not-yet-sufficiently-researched belief that overly high levels of immigration, particularly without effective integration strategies, can lead to less trust in and effectiveness of government. Based on this, I would be potentially interested in participating in a collaboration here.

      That said, I think we might want to parse out different kinds of effectiveness (i.e. are we strictly looking at efficiency, or are we also looking at political stability and public trust?). To the extent I have concerns about immigration, it’s more from the stability/public trust angle, rather than the corruption or solvency you initially mentioned, so if you are more interested in those, we may not be that adversarial after all.

      • christhenottopher says:

        Personally, I view political trust as at most an instrumental rather than terminal value, though stability is important depending on the definition. The problem I’d see there is political stability that I would care about (high levels of political violence/civil wars) may be rare enough in countries capable of attracting immigrants to not come to strong conclusions on.

        How I see this, if low skill immigration does cause corruption/solvency/user satisfaction issues with government services, then that’s potential evidence that large issues might arise too. I personally don’t think that those less severe issues will be meaningfully associated with low skill immigration. But we may not be sufficiently adversarial for this case.

    • ZachJacobi says:

      If you think I’m adequately adversarial, I’d be willing to do the other side of this.

      I’m Canadian and strongly believe that a point system that favours highly skilled immigrants is important to avoiding the difficulties in integration we’ve seen in Europe. I’m very much in line with Joseph Heath and agree with the immigration points he makes here

      • christhenottopher says:

        While we likely would have areas of disagreement, I’m worried they’d be a bit too subtle to come across as truly adversarial. Two people who agree “western countries should embrace immigration from many non-western countries” may seem to be pulling punches compared to highly anti-immigration views that are common.

        • Scott Alexander says:

          I don’t know, pro-low-skilled immigration vs. only-accept-high-skilled-immigrants seems like a pretty important disagreement to me.

          • reasoned argumentation says:

            The anti-low skilled immigration argument is that you lower your society’s average IQ and introduce a permanent underclass that’s genetically lower IQ so can never do better than a life of marginal employment and petty crime.

            The anti-high skilled immigration argument is that you introduce low trust alien cultures which then proceed to practice ethnic networking to the detriment of the host society even if individuals were beneficial before they were able to do that.

            The former counter-position is taboo. The latter is even more taboo.

            This is not a subject where you (you meaning Scott Alexander specifically – not the general “you”) are willing to host an honest debate.

          • Aapje says:

            @reasoned argumentation

            A genetic argument is not necessary. AFAIK, most anti-low skilled immigration advocates argue that the cultural and social gap is so large that letting these people migrate causes burdens way beyond the benefits & that these gaps take many generations to close.

          • albatross11 says:

            Another argument against lots of high-skilled immigration is that it can make it uneconomical for Americans to go into some high-skill fields. If we import enough doctors from foreign countries that the expected income for a doctor stops making up for the expense of medical school and the long period of deferred or reduced income during training, then we may set up a situation in which there are no Americans going to medical school. Whether that’s a major problem is arguable, but it’s not crazy to be unhappy with such a situation.

          • Randy M says:

            A genetic argument is not necessary.

            It is not the only argument worth addressing, but if it is off the table the debate is going to be constrained in one direction.

          • Aapje says:


            Yet another argument is that it closes off the pipeline for some natives, blocking them from getting the low-skill jobs that they can use to climb the ladder.

        • Reasoner says:

          My general impression is a lot of alt-right people would be placated by a Canada-style immigration policy. Check it out: Canada doesn’t appear to have much of an alt-right.

          It’s rather frustrating actually, because a Canada-style immigration policy seems like the perfect compromise to me. It should appeal to liberals because liberals love Canada. It should appeal to conservatives because economic nationalism and America needs the best people. But our national conversation around immigration is too dysfunctional for folks to see this. (I also do place some credence in the theory that the left is engaging in motivated reasoning because it knows immigrants will vote Democrat.)

          • albatross11 says:

            Won’t highly-skilled immigrants also mostly vote Democrat? And probably be more likely to vote?

          • hdem says:

            Lots of businesses and farmers who are more likely to be conservative do benefit from having lots of low-skilled illegal immigrants. They are much more likely to comply with illegal requests, unpaid overtime, and much less likely to sue or unionize. A system that stops low-skilled illegal immigrants would not be desirable for these groups.

          • WarOnReasons says:

            a Canada-style immigration policy seems like the perfect compromise to me

            I share the same view, but it does not seem to be politically feasible.

            There are many Republicans who in principle would like increasing the number of high-skill immigrants in return for decreasing the low-skill immigration. However, based on the experience with the past amnesty deals when illegal immigrants were granted citizenship in return for better border security (which failed to materialize), most may not trust any such deal with the Democrats.

            For the Democrats, switching to a Canada-style system is also unappealing, since low-skill immigrants vote for them in much greater numbers than the high-skill ones.

          • Iain says:

            …in return for better border security (which failed to materialize)

            This is false. (See also.)

          • WarOnReasons says:

            To me better border security means less illegal border crossings, not higher budget for ineffective government agencies.

          • WarOnReasons says:

            Have you actually read the articles you’ve linked? The Pew survey chart shows that the number of illegal immigrants has roughly quadrupled in the 25 years since the first amnesty.

  17. anglevice says:

    Moderate, regular alcohol consumption is good for your health. The research I’ve done seems to agree with this, but I’ve probably missed the other side, looking for someone who has thoroughly researched counter arguments to this statement. I feel it’s a commonly stated sentiment in certain circles.

    • Douglas Knight says:

      As far as I can tell, this is only mildly controversial. I think the medical consensus is that it’s true, at the weak level of evidence that is applied to many dietary claims, but the controversial question is whether the intervention of telling people is a good idea. As far as I can tell, a lot of people seem to say that they don’t think there is enough evidence to believe it, but they really mean that there isn’t enough evidence to promote it. I don’t think that makes for a good argument.

      I think it’s true and I think you shouldn’t promote it. I believe in the specific mechanism that it is a blood thinner. Having a mechanism that matches the epidemiology leads me to have strong belief in the claim. But there are other blood thinners, so there’s no point in encouraging alcohol.

      A more controversial question would be drinking while pregnant.

      • Scott Alexander says:

        I think it is controversial; wasn’t there a study that came out recently which pretty strongly endorsed the opposite claim?

        • Yosarian2 says:

          Yes; I think you’re talking about this one?

          It’s also gotten a lot of press coverage.

          • Douglas Knight says:

            But, no, it’s just claims to be quibbling over the precise value of the minimum. But it isn’t even doing that.

          • pdbarnlsey says:

            I think the claim, from reading several summaries but not the study itself, is that previous correlation studies showing health gains from moderate consumption were flawed, because they used a zero alcohol control group which included people who avoided alcohol due to pre-existing health conditions and/or recovering alcoholics who had done themselves significant harm by that point.

            The new study uses levels of consumption within the drinking population to show that relatively low levels of consumption correlate with poorer health outcomes. By design, it can’t show the effects of nonconsumption, but you could at least argue that the dose response would remain similar down to zero.

            IMHO, the problems with dissimilar groups drinking different amounts are just as severe as those between nondrinkers and drinkers. My suspicion is that people who are careful with their alcohol intake are likely to be careful elsewhere in life, and I can’t think of any good confounders running the other way.

            I think this would be a good topic, but probably lack the time/expertise to do it myself.

          • Douglas Knight says:

            previous correlation studies showing health gains from moderate consumption were flawed, because they used a zero alcohol control group

            Yeah, the paper says that. It is a lie.

          • pdbarnlsey says:

            Sorry Douglas, could you clarify – are you saying that previous studies didn’t (exclusively?) use non-drinkers as their control?

          • Douglas Knight says:

            This study finds an optimum at 1 drink/day for men. Earlier studies found an optimum at maybe 1.5 drink/day (although I don’t think it cites any earlier studies, only government guidelines). How else could they possibly get such precision without comparing to lower consumption?!

          • pdbarnlsey says:

            Ok, so you’re saying that earlier studies had at least some lighter drinkers in their sample?

            That certainly sounds plausible, but doesn’t lead me to discard the results of a large study which might (at least in theory) have removed a source of bias in the form of unhealthy nondrinkers.

            And, to be clear, as a moderate drinker, I’d like to thinking I’m not doing myself too much harm, so I’d be pleased to agree with you.

            My view is that the unhealthy nondrinkers thing sounds pretty specious – depending on the level of abstemiousness required to qualify, I feel like there are a lot more t-totalers who are doing it for religious/bodily purity reasons than ones doing it because their liver is failing.

          • albatross11 says:

            Is there survey data on the question of what fraction of people rarely/never drink and their reasons?

          • Douglas Knight says:

            1. I’m not asking you to discard this study! On the contrary, I’m asking you not to discard earlier studies just because this study lied about them.

            2. But even if you did rely on just this study, it is still saying that moderate drinking, 1/day, is the optimal level. Just because it lowers the number doesn’t mean that it “pretty strongly endorsed the opposite claim,” as Scott described it.

      • A1987dM says:

        I believe in the specific mechanism that it is a blood thinner.

        The one I think I heard is that it works by reducing anxiety and helping socialization.

      • ConnGator says:

        An economist did an analysis of drinking while pregnant and found that one drink per day _improved_ fetal outcomes (due to relaxing the mother) and even two drinks per day was not negative. This is a shorter version of a longer podcast I heard about it:

    • skef says:

      I fear that a debate on this subject would devolve into “Which of the recent large meta-studies are less flawed?”

    • BJH says:

      I am potentially very interested in taking the other side of this. For personal reasons, I’d already started collecting literature to review, and my read is the opposite of yours.

      So as to make sure that we can speak the same language, can I ask — what’s your background? My background is in economics and causal inference. In particular — and this would 110% be something we would need to discuss in a write-up — in the land of applied microeconomics, observational studies without any form of identification carry essentially zero scientific weight.

      We can also discuss over email.

    • Terence says:

      my extensive research with collaborators Cuervo & Daniels confirbms thiss

    • I’m inclined to take the other side, but I haven’t researched it thoroughly. I’m unsure how much time I’m willing to devote to this. Contact me at

    • ohwhatisthis? says:

      I’m down to argue against that, and there are no health benefits, barring a few isolated psychology benefits and edge cases(sleep aid(

      • dndnrsn says:

        Alcohol is a pretty dreadful sleep aid. It’ll get you to sleep, but the sleep is awful.

      • Douglas Knight says:

        You want to argue not only that there is no net benefit, but that there are no individual benefits? OK, alcohol is a blood thinner.

        • ohwhatisthis? says:

          I should have included a few more lines expounding on edge cases, like other medical uses.

          • Douglas Knight says:

            By “individual benefit” I don’t mean useful to some individuals, I mean useful in the general population for the single largest cause death. The standard recommendation is that most men 45-80 and women 55-80 should take a blood thinner. That population is not an edge case.

  18. bernie638 says:

    I work as an instructor at a nuclear power plant and I would volunteer to take the pro-nuclear energy side of a collaboration

    My position is that Nuclear energy is extremely safe. I will concede that it has a high initial cost, but it is cost competitive to operate.

    • Michael Arc says:

      I think that Nuclear Energy is safe, but not economical.
      It’s worth continuing to run existing plants, but not to build be ones or to do more research, given price trends for nuclear and got batteries. Is that a different enough opinion?

      I notice that the challenge of sharing an ontology is the usual challenge for adversarial collaborations. This seems like an exception.

      • bernie638 says:

        Thank you, but unfortunately, I’m not confident I could give a good account of accounting. Is it worth it involves too many predictions of the future cost of storage and other power sources along with interest rates. I’d rather stick with something that has actual research. Thank you though.

    • skef says:

      I don’t think I could muster enough research time to take the other side of this, and anyway I lack the sort of qualifications a reader would look for. But, a question that I would think would need to be a focus in such a write-up:

      Is your position that nuclear energy is extremely safe with indefinite on-site storage of waste products, or is some sort of processing and/or storage needed to complete the picture? And if the latter, isn’t actual safety in part a political problem, which shows no sign of being resolved after many decades?

      • bernie638 says:

        Yes. I’m not at all concerned with storage. Recycling would be awesome, but permanent disposal or on site storage are all fine by me. The dry cask storage used now is much safer than any other hazardous waste storage.

    • pdbarnlsey says:

      I think that’s a good topic. I am not qualified to contribute, but will say, as someone who does cost benefit analysis professionally, that discount rates are going to be a huge deal when determining the economic viability of building a new plant today. You’ll need to have a good strategy for researching/defending best practice in this area – any sort of commercial discount rate (8+%) or arbitrary time horizon for considering benefits is likely to kill nuclear investments dead.

      Perhaps a reasonable outcome would be a conclusion that new plants are economic at a discount rate of, say 3% or less.

      • bernie638 says:

        I’m worse than unqualified to discuss economics. I’d just rely on the people who are actually building. It’s not looking good in the US right now, only Vogtle is currently building, however Turkey Point just got a COL, but they may not build. Other countries are building and it looks like that will continue. Plans For New Reactors Worldwide

        • pdbarnlsey says:

          I don’t think it necessary follows that if a government is pursuing a policy, then that policy generates net benefits. Some of those decisions might be wrong from the point of view of a benevolent social planner.

    • eccdogg says:

      I was an economist for a Nuclear trade association in DC about 13 years ago. I currently work in the energy industry.

      I have not looked at the data in quite some time, but I believe

      Nuclear energy is safe, but new nuclear plants are not economical given western regulatory standards and current natural gas prices

      I am not sure how much time I have to devote to this, but I could participate.

    • thad says:

      I would love to see a nuclear debate. I don’t believe the opposition position, but I would be willing to play devil’s advocate, and I think I could do an ok job. I’ve worked at research reactors, one as an undergrad and one as a full-time operator for two years.

      Alternatively, is there a different nuclear topic? Something about current level of regulation? I’m willing to argue against ALARA as a useful guideline. Is there anyone here who thinks low dose radiation is beneficial?

  19. tlrs says:

    I don’t know whether this has been discussed before on SSC, since I haven’t read that many of Scott’s articles, however: Skills are generally non-transferable. They are only transferable in begin-phases of skill acquisition. Furthermore, talent plays a minor role in many domains of expertise.

    This isn’t an absolute statement, because obviously genetics excludes the possibility of expertise in some things (i.e. sports)

    • quanta413 says:

      Hmmm… could you be more specific? I would agree mostly that skills don’t transfer well. Generalizing is hard or humans are bad at it depending on your point of view. I’m not even totally sure what the theory is of what should be gained by transference though. Like, is transference when if X is related to Y and I do Z years of X then I should reach W years of skill at Y in less than W years?

      On your second question though, I think talent (or genetics or something we don’t know how to control) plays a major role in setting what is possible to attain for a particular individual for many skills. Athletics, mathematical ability, social charisma, etc. You still have to actually do something to build up your skills, but the rate of skill attainment and the ceiling can differ by orders of magnitude.

      • albatross11 says:

        What experimental or observational data would we use to tell us whether success in a given area was mainly driven by innate talent vs domain-specific work/practice?

        Would the general IQ results (the ones showing that IQ positively correlates with performance on just about any job) contradict this claim?

        • quanta413 says:

          Both things are absolutely required for a human to get anywhere. There’s an obvious point past which without prior knowledge you just can’t do a lot of things. For example, someone from the 1700s is just not going to invent a transistor with 1700s knowledge. They are two centuries away from having the appropriate industrial base and domain specific knowledge and no amount of IQ can cover for that in one human’s lifetime.

          One way to look at it is that lack of domain specific knowledge is kind of how like we can lower someone’s IQ by clubbing them in the head hard enough. Environment drops off a lot in importance in a relative sense because in the U.S. there’s not a lot of traumatic cranial injuries and everyone is well fed. I think there are a lot more obvious differences in domain knowledge and expertise etc. between different professions then there are things left that cause differences in individual IQs though. So we should expect domain knowledge to explain a lot more variance about current job performance than say, shared environment does about future monetary earnings.

          Another way to look at it is how well do things like the IQ of a companies workers predict its success vs. the cultural history of the company, it’s domain knowledge, trade secrets, etc.

          Attaining a certain level of domain specific ability may often require possessing a certain amount of innate talent.

          I think the general IQ results show that if you want someone to do a job for you a year or two from now (shorter or longer depending on the job) and that person is training in the meantime, then maybe you don’t care much about their domain knowledge right now. But if you need your car fixed this week, it’s obvious you should pick a 100 IQ mechanic over a 145 IQ professor of mathematics who’s never looked under the hood of a car before.

    • Michael Pershan says:

      I would take the other side of this, but much of it would start with offering proper definitions of “skill” and “transferable.” I think the non-transferability of skills is obviously true or obviously false according to entirely reasonable definitions of both terms. (I’m a k12 math teacher professionally.)

  20. edd91 says:

    I would be interested in picking up the adult brain neurogenesis debate if anyone is interested. I believe in neurogenesis FWIW

  21. tailcalled says:

    I am interested in doing an adversarial collaboration on Blanchard’s Two-Type Male-to-Female Transsexual Taxonomy.

    … assuming it isn’t too taboo? It might be. Is it? I’m coming from the side of believing that the taxonomy holds.

    Summary of my views:
    – There are two and probably only two kinds of trans women, homosexual and autogynephilic
    – The somewhat-confusingly-named Homosexual Transsexuals are heterosexual and transition because they’re very feminine which makes transition a huge overall social-success win
    – The Autogynephilic Transsexuals transition because they have a paraphilic sexual orientation that makes them attracted to being women

    • skef says:

      Unlike the other comments made here, I might be interested in actually taking the other side on this one. It depends on:

      Other than the existence of the two categories (and perhaps its necessity? can the lines blur in particular cases?), what do you see as at stake or relevant in the taxonomy?

      [To give an idea of what I’m asking, suppose there were only transsexuals of the first type. By stipulation, these are people transitioning because they are “very feminine”. What in Blanchard’s research supports an only-outward-facing understanding of “femininity”, such that we should attribute transitioning to “social success”?]

      If there’s nothing to the observation other than “transsexuals are usually of one of two types”, then I myself don’t see any strong objection, assuming the research itself is accurate.

      • tailcalled says:

        There’s not much other stuff at stake. I’m generally pro-trans, I just don’t think you can understand how transgender stuff works without having the taxonomy.

        Most likely, not all trans women fit into either category, but I think these would be extreme rare exceptions rather than something that’s useful to include in the general considerations. I doubt there’s much blurring of the lines between the categories, but I do see some ways in which there could be a little bit. In practice, though, this blurring can also be disregarded.

        I’m less confident about the exact reasons HSTSs transition, and I could imagine that this is one of the cases where we would conclude that there is a lack of evidence. The HSTSs I know tend to emphasize the social aspects of dysphoria, but I’m not sure how much research supports that.

        I guess I should share more background:

        I’m very AGP but not socially transitioning (because my dysphoria is mostly about my body; I’m not opposed to social transition in general). I have a trans girlfriend and a lot of trans friends, and I’m running a number of discussion places focused on transness and sexology.

        My interest in the topic of transness of course started out with my own gender issues, which lead me to start doing gender-related surveys. At the time, I didn’t believe in the taxonomy, but after learning more about the subject, seeing my own survey results, and being introduced to some theory by a guy I met, I’ve come around to thinking that the taxonomy is definitely true.

        • skef says:

          I personally think that most of the controversy about Blanchard’s work stems from perceived implications (accurate or not) that it: 1) calls the status of transsexuals as women into question and 2) carries the suggestion* that it might be better for society to discourage transitions.

          I don’t think I would bother delving into the debate unless it touched on those issues. Just the bare existence of two categories doesn’t seem very surprising or relevant to much of anything else.

          * Which is compatible with no overt claim, or even overt claims to the contrary. “The data suggest X, even if the researcher doesn’t have the courage to say so.”

          • tailcalled says:

            A lot of people seem to claim to disagree with the taxonomy, not just the implications but also the causal claims. This might mostly just be a thing that other trans people do, though – I sometimes get the impression that a lot of cis people more or less implicitly believe it. I don’t know.

          • skef says:

            I also disagree with the taxonomy the way it is described. For example:

            The Autogynephilic Transsexuals transition because they have a paraphilic sexual orientation that makes them attracted to being women

            This way of putting things invites certain comparisons and corresponding dismissals. “Should we invite diaper fetishists to just walk around in their diapers in public? Because that’s really what is going on with this kind of ‘transsexual’. They’re getting off on this!”

            Lots of people have non-standard sexual interests, including various paraphilias. Almost no one restructures their entire life around them. The strength and pervasiveness of the feelings that lead to transition, with all of its social, material, and subjective costs (e.g. physical pain) suggest (to me, at least) that a bit more is going on. Most people are relatively content to keep their sexual business to sexy-time.

          • tailcalled says:

            Most people aren’t relatively content to keep their sexual business to sexy-time. People’s sexual orientation tends to have a large influence on who they live with (most people tend to live with their sexual partners), who they display affection to, who they most closely share resources with, etc.. I don’t think it’s unfair to say that people do to a large degree structure their life around their sexual preferences.

            These sorts of things are often attributed to “love”, but “love” is a very vague concept to work with. A lot of the things people do for “love” seem to have functional purposes, for example to secure the access to the mate, to pool resources for raising children, and similar. I’m not sure to what degree these are the direct motivations (i.e. we have an innate desire to secure access to the mate and socially justify this desire using made-up concepts like “love”) versus evolutionary forces that have created a more-direct motivation (i.e. we have instincts that make us develop some form of attachment to our sexual target because it is evolutionarily adaptive), but either explanation seems like it could very well apply to paraphilic interests too.

            My impression is that a lot of people who are very paraphilic tend to have “love-like” feelings relating to their paraphilia, but this impression might be wrong. I probably haven’t looked into it as much as I should because I find the topic vague to work with, but I think it could probably be brought up during the collaboration. That said, if you don’t really disagree with the fundamentals of the typology, I think I’d rather collaborate with someone who does, since that feels like the more important part.

          • skef says:

            People’s sexual orientation tends to have a large influence on who they live with (most people tend to live with their sexual partners), who they display affection to, who they most closely share resources with, etc.. I don’t think it’s unfair to say that people do to a large degree structure their life around their sexual preferences.

            I would characterize the feelings you’re referencing here as romantic rather than sexual, and I claim the two are often, if not always, separable. Sexual and romantic preferences can and often do come apart. As a dichotomy, it’s at least as plausible as the original topic.

            So if we wanted to say instead:

            The Autogynephilic Transsexuals transition because they have romantic feelings entwined with seeing themselves as women

            That seems quite plausible. But it’s also quite a different spin on the situation. (There is an interesting but very complicated adjacent question, which is whether this or a statement very close to this applies to most women.)

            My impression is that a lot of people who are very paraphilic tend to have “love-like” feelings relating to their paraphilia, but this impression might be wrong.

            My impression is that this does happen, but is less common than paraphilias generally. “I’m in love with my car” is a small subset of “I get off on pictures of cars.”

          • tailcalled says:

            I mean, there’s no question that these sorts of topics are relevant to bring up in a “gender dysphoria: much more than you wanted to know”-style post. However, I don’t think we disagree much on this topic, and even if we do, I think the “adversarial” aspect is supposed to be more than this smallish thing.

          • albatross11 says:

            This is a tangental question, but is it common for people to separate their romantic and sexual interests very widely? How widely? Like are there a lot of men who are sexually attracted to men but romantically attracted to women? Or the other way around?

          • skef says:

            This is a tangental question, but is it common for people to separate their romantic and sexual interests very widely? How widely? Like are there a lot of men who are sexually attracted to men but romantically attracted to women? Or the other way around?

            It’s probably least useful to start with one of the broadest possible categorizations (that being a difference in genders).

            Is it common for people to intensely crave sex with people they have no interest in maintaining a relationship with, and even actively dislike in other ways? Yes.

            Is it common for people to prefer sexual activities or have paraphilias that they take to have no independent role, or even a negative role, in their relationships? Yes.

            Do some people maintain romantic relationships with little or no sex while seeking sex outside of the relationship? Yes.

            Are there romantic asexuals? Yes. Do such people have “romantic preferences”? Often.

          • albatross11 says:


            Okay, those examples all seem solid. I think I was just not thinking the issue through beyond my own experience.

          • aristides says:

            I’m involved in a paraphilic community, and they actually took a poll on this issue. 10% of participants said they were motivated only by love, 34% said they were motivated only by lust, and 56% said they were motivated by both.

  22. poignardazur says:

    Oh man, I’ve been thinking about this for a long time. I’d be interesting in doing an adversarial collaboration on whether the FCC’s recent Open Internet Order is good for US internet consumers on the long term. The general opinion is that it’s really bad, my opinion is that it’s probably slightly good or at least not that bad.

    Fair warning if you’re interested: I already have a potential partner in mind, but I’d still be potentially interested in making a team of more than two people.

    • mw says:

      I’m interested. You can reach me m matthi coffee.

      I would want to very slightly rephrase the thesis as “[…] good for the general public […]”. I believe this is functionally identical to your phrasing because we would probably assume almost every member of the public to also be an “internet consumer”, at least in the long term. But I abhor how consumer has replaced citizen, and I think it puts undue weight on strictly economic factors to the detriment of other dimensions of well-being.

      We’d also need to agree to good/bad as the standard of measurement, because “at least not that bad” just opens it up to completely arbitrary judgement of “badness”.

  23. Truism says:

    Looking for a partner on gun control. Happy to work on framing the question.

    About me:
    Late twenties Australian military logistician. Strongly pro-gun ownership. Strongly anti-gun control. Formerly pro-gun control.

  24. textor says:

    Okay, what about it: adversarial collaborations are unlikely to produce better knowledge representation, because, for example, they redirect attention from data to sophistry, modeling the opponent and maneuvering around his/hers blind spots. Some cases to explore: court trial vs. criminal investigation, scientists’ replies to critique vs. published research, future essays replying to this post vs. traditional comments.

    The opposite hypothesis is obvious.

    (I’m kidding)

    • realwelder says:

      Additionally, adversarial collaborations are unlikely to produce better knowledge representation, because they will be insufficiently adversarial.

      That is, they may overemphasize less relevant points of consensus while eliding the more meaningful points of disagreement.

  25. gvprtskvnis says:

    I’d be interested in doing an adversarial collaboration on whether qualia are real/fundamental or an emergent phenomenon. I believe that they are. I am also potentially open to adversarial collaborations in other areas of metaphysics. This is fairly different from the other topics in that the evidence is largely from logical arguments rather than published research; however, I think it would still be a practical topic for an adversarial collaboration. I am an undergraduate student studying philosophy (and math), so I’m not an expert, but I have read a significant amount of literature on this topic. Although I’d be able to start immediately, I wouldn’t be able to spend more than a couple hours per week until June.

    • orin says:

      You believe that qualia are what? I think your description of your position is currently ambiguous.

    • amaranth says:

      what’s the short version of your argument for why they’re fundamental? i’ve heard a lot of bad ones

      • gvprtskvnis says:

        Our only evidence for physical facts comes from our qualia, so the qualia must be at least as fundamental as the physical facts.

        • Protagoras says:

          That qualia just are experiences is an interpretation of our experiences, it is not something the experiences themselves tell us. And while some try to evade this by saying qualia are the experiences by definition or stipulation, that is untenable, as qualia theory makes claims about the nature of experience which, again, go beyond anything the experience as such can provide, and one cannot establish such a contentful theory by stipulation.

          • gvprtskvnis says:

            How does qualia theory make claims about the nature of experience which go beyond what the experience can provide? (Also, I feel like it’s less definition/stipulation so much as pointing to a thing that the listener can only see for themselves.)

          • Protagoras says:

            Well, that there is a thing the listener can only see for themselves is already an assumption that goes beyond the evidence.

          • gvprtskvnis says:

            Sure, but it doesn’t really matter that you might be a p-zombie since, if you aren’t, you can replicate my argument for yourself.

          • Protagoras says:

            You seem to have reversed the intent of my comment. I do not believe that p-zombies are possible. What I think is possible is that someone other than the experiencer can know that a quale is being experienced.

          • gvprtskvnis says:

            How could they do so? By looking at the physical structure of the brain?

          • Protagoras says:

            By finding out things that are ultimately physical facts about the brain, certainly. I suppose looking is one of the ways of doing that.

          • gvprtskvnis says:

            So, you think that it’s *logically* necessary–not just (meta-)physically necessary–that our subjective experiences arise from physics in the way that they do? If so, why?

          • Protagoras says:

            Because the experiences and the physical states are identical, and identity works like that.

          • gvprtskvnis says:

            The experiences and physical states being identical is a very strong claim. We only have access to the experiences–what convinces you that physical states are the same?

          • I think it would be possible to argue to that position by excluding the alterantives.

          • Protagoras says:

            At some point it becomes impossible to be brief. I mostly agree with what Lewis has had to say on this issue (perhaps most to the point in “Should a Materialist believe in Qualia?”, but that builds on his earlier papers on philosophy of mind). Dennett also has some pointed analysis of many relevant thought experiments. A few of my own thoughts are here.

        • Our only evidence for physical facts comes from our qualia, so the qualia must be at least as fundamental as the physical facts.

          That seems to conflate and epistemic and an ontological definition of “fundamental”.

          • gvprtskvnis says:

            Hmm. I’d be willing to claim that only things that are ontologically fundamental can be epistemically fundamental. (At least for “fundamental” meaning “real”, not necessarily “most simple” or something like that.)

          • Protagoras says:

            Now that sounds like an extremely strong claim to me. Why exactly would being ontologically fundamental (or real) have anything to do with being epistemically fundamental (and what does “real” have to do with epistemology)?

          • gvprtskvnis says:

            cogito ergo sum

          • Protagoras says:

            Descartes bundles an awful lot of assumptions about the nature of the self into that one. I find its epistemic priority suspect, and its ontological priority much more so.

          • gvprtskvnis says:

            Yeah, I don’t agree with the assumptions about the self. However, it seems impossible to argue that my current experience does not exist, since then there’s no reason to believe that *anything* exists.

          • Protagoras says:

            So “there is a thought.” One could try to go all Kantian, and say this establishes that we exist in the kind of universe that makes thought possible, but that requires us to have theories about what actually is required for that, and whatever theories of that kind we have, they do not share the certainty of “there is a thought.” I certainly don’t see how the move from the epistemic status of “there is a thought” to any conclusions about metaphysical priority.

          • gvprtskvnis says:

            I would have defined “ontologically fundamental” as just “is a thing that exists”. How are you defining it?

          • Protagoras says:

            I wouldn’t define “ontologically fundamental;” it does not strike me as a useful expression. It certainly strikes me as a tendentious way to say “existing.” But perhaps you mean something different by “exist” than I do; some people do pack a lot of strangeness into that expression.

          • gvprtskvnis says:

            I guess I’ll just elaborate then. If thoughts exist, they must be a thing that exists in their own right. They can’t just be an illusion created by physical processes, although they can be a causal result of them. The only way to unify them with physical processes is to accept idealism.

          • Protagoras says:

            Why can’t the thought be a physical process?

          • gvprtskvnis says:

            It can be, but then you have to use a pretty different definition of “physical process” than people usually do. It is not a result of e.g. the standard model. I think that definition alters what it means to be a physical process so much that it’s more accurate to call that view idealism, although of course it’s just a matter of what words you use and doesn’t really matter.

          • Protagoras says:

            How do you know it is not a result of the standard model?

          • gvprtskvnis says:

            Because the standard model doesn’t even try to describe subjective experiences. It can predict that I’ll say that there’s a difference between red and green, but it can’t predict what “red” or “green” (i.e. my subjective experience of them in a particular moment; I make no claims as to whether my subjective experiences of red or green are even kind of similar to yours) actually are.

          • Protagoras says:

            I do not know what it means for the standard model to “try to describe” something. Very little of what happens constitutes obvious consequences of the standard model, and I am not sure where you derive your apparent confidence in your ability to distinguish between things that may be unobvious consequences of the standard model and things that could not be unobvious consequences of the standard model.

          • gvprtskvnis says:

            It’s the same reason I can be sure that math doesn’t prove anything about friendship. Math doesn’t try to describe friendship. Math doesn’t even have the language to talk about things like friendship (at least without translating “friendship” to a toy model that no one agrees captures the entire phenomenon). The standard model doesn’t even have the language to describe qualia, so although it may have many unintuitive consequences, none of them can be about qualia.

          • Protagoras says:

            OK, so you know it in the same way you know something else that I’m pretty sure you don’t actually know. Good to know.

          • gvprtskvnis says:

            …do you actually think math can prove things about friendship? (I think there’s a case to be made that the two cases I proposed are different because “friendship” as a meaningful category doesn’t actually exist–but that seems to be almost the opposite of what you’re saying.)

          • Protagoras says:

            I may have been distracted by your example. However, I also think it’s an amazingly strong claim to say that math can’t prove anything about something. Still, I’ll be more cautious; if math can prove anything about about physics, it can prove something about friendship. And if it can’t prove anything about physics, the math/friendship relationship is a poor analogy.

          • gvprtskvnis says:

            It really isn’t a strong claim, since you have to “import” the thing you’re trying to describe into the language of math. And indeed, math can’t prove anything about physics–it can prove things about a system that has many similarities to what we observe, and is probably intimately connected since physics has such a simple mathematical description, but it still has to take some things as axiomatic. (In other words, you can build a model that you then use to predict the behavior of electrons, but you can’t actually put a reference to real electrons in mathematical equations. You can only point at the concept.)

            (Tangentially, I’m not sure why it makes the math/friendship analogy a bad one. My hope was to pick an example that made it clearer what I mean, and since physics is so closely tied to math it makes it harder to explain.)

  26. tribsantos says:

    Definitely possible. Kahneman and Klein have an article which is a work of beauty.

  27. promotoriustitiae says:

    Would be interesting in arguing that not only is Utilitarianism not the correct ethical system, promoting aims actively opposed to the good, but it is properly considered to not be an ethics at all. Naturally, done from a rationalist position. I’ll be free after this weekend to argue it as then my current writing projects end.

    • gvprtskvnis says:

      I am potentially interested in arguing in favor of utilitarianism as the correct ethical system.

      • promotoriustitiae says:

        Contact me at farnxobhg@tznvy.pbz (after some rot13) so we can set out some starting points and negotiate what we want to cover, suspect we will end up narrowing the focus.

        Also on the thread below, I won’t be claiming to do a complete job but I would be surprised if any of these could do that. It’s a collaboration, that’s all.

    • amaranth says:

      here’s the entire argument necessary in favor of this: utilitarianism doesnt say anything about what good *is*

    • Philosophisticat says:

      For reasons I mention in a comment below. I don’t something like this would be illuminating for the community from people who are not experts, and would be likely to just spread confusion. It is already painful to read some of the amateur philosophy in rationalist circles.

      • A lot of rationalists seem to view U. as obviously correct. Spreading some doubt on the issue, at least, would move them to a more correct position, in my view.

        But a lot depends on whether the “rationalist position” means “using facts and logic” or “assuming the correctness of The Sequences”.

        • Philosophisticat says:

          It would do good only if were done by someone who understood the issues well. If done badly, it might either reinforce irrational confidence in utilitarianism, or it might raise doubts for bad reasons, neither of which would be very good outcomes. Perhaps worse, if it ended up being a point of reference for rationalists, it would set a confused framework for future thinking, and give people a misguided sense of understanding, already major problems with the rationalist community re: philosophy.

          • Philosophisticat says:

            Scott, for instance, is intelligent, careful, an excellent writer, and IIRC has an undergraduate degree in philosophy (?) and his Consequentialist FAQ is full of errors. I think that’s about the upper bound of what we could expect from a nonprofessional.

          • I think you might be letting the perfect be the enemy of the good, there,

          • Philosophisticat says:

            I think you might be underestimating how much an amateurish account presented as thorough and/or treated as a touchstone by the community can be worse than just linking people to the stanford encyclopedia article, noting the lack of expert consensus, and reciting the virtues of epistemic humility.

          • There’s nothing to stop rationaists looking at SEP except lack of motivation. The lack of motivation presumably comes from believing that Uism is obviously true. So summarising why it isn’t obviously true might motivate them to research it properly.

          • Philosophisticat says:

            Part of the effect, intended or not (and I think it should be intended!) of these sorts of posts summarizing the state of research on an issue, is to serve as a trustworthy-if-imperfect substitute for the readers doing the research themselves. The effect (on me) of Scott’s posts on wheat and marijuana was not to whet my appetite to do my own independent research on wheat and marijuana. It was to take the conclusions of his post as a reasonable guide to form an opinion, on the assumption that the author was a thorough and competent evaluator of the field. I formed opinions on the basis of those posts that I did not have before, and there is research I might otherwise have done that I did not do because I believed, I hope reasonably, that doing it would not be likely to lead me much closer to the truth, without substantial effort, than just taking Scott’s analysis on board.

            If Scott were, for example, statistically incompetent in a way I could not recognize, or if he were not actually looking at the best work, the post might easily have done more harm to me than good. And I think this would be the position of most readers here with respect to an amateur presentation of a difficult philosophical issue.

          • Philosophisticat says:

            I think utilitarianism is false and would love to see someone in the community present a thorough case against it. But the original poster says, for example, that utilitarianism is “not even an ethics”. This is just false. Hearing that someone who holds that view is going to represent the correct side in a primer on utilitarianism is like being a libertarian and hearing that my side in a primer on the costs and benefits of tariffs is going to be defended by someone who thinks that wealth is measured in seashells. The difference is that it’s very hard for someone to be that wrong about economics, but very easy for someone to be that wrong about philosophy.

            And when that happens, very often, one of two things will happen. Either people will not notice how wrong the take is, and be fooled, which is bad, or they will notice how wrong the take is, and decide that the (correct!) position is defended by stupid arguments. Which is even worse.

          • Michael Cohen says:


            Also, can you link me to a philosophical paper that captures your objection to consequentialism? I am a foundational consequentialist, I was until recently a factoral consequentialist, and now I can’t decide about the latter anymore; I’ve become very confused about decision theory.

          • Philosophisticat says:

            @Michael Cohen

            It’s hard to say something very general because of the great range of possible consequentialist views. I think the main reasons to reject the standard versions of consequentialism are pretty well known – they have radically counterintuitive implications for things like promising and other agent-relative reasons, cases where the number and identity of who exists depends on your action, etc.

            Most confusion I think is about what positive reasons there are to accept consequentialism. In short, I don’t think there are many.

          • Michael Cohen says:


            Can you clarify what you mean by

            cases where the number and identity of who exists depends on your action

            Do you mean for average-case, your action depends on the number and identity of who exists? I don’t see how facts about what exists could depend on your action.

            Also, to clarify a more precise position, bracketing decision theory for a second and supposing CDT, I buy
            -welfarism (so the utilitarian version of consequentialism)
            -total not average
            -objective list theory of welfarism, in which good mental states, genuine relationships with others, virtue, etc. compose a good life; this should suffice to cover what creatures count here–they count insofar as they can experience mental states, have relationships, etc.

          • Philosophisticat says:

            When you’re deciding whether to have children, who ends up existing depends on your action. That is all I meant.

            Anyway, I think your view has the standard problems I mentioned – it fails to capture agent-relative reasons (for instance, to keep one’s own promises, rather than make sure other people’s promises are kept). It runs into the repugnant conclusion. It also runs into the utility monster. I also think objective list theories run into some worries about disunity of the good and anthrocentric bias in selection of the goods, but I don’t think those are knock down problems.

          • It’s beginning to look like Philosophisticat and Scott should be adversaries.

    • Philosophisticat says:

      Also, “naturally, done from a rationalist position” is a red flag.

  28. JohnBuridan says:

    I am planning on entering. I have found a partner, but we are trying to find the area were we both have more than light disagreement. I will return in two days to post our proposed controversial topic.

    • scherzando says:

      If we don’t hear back from you, I will assume that you’ve found two equally-attractive topics to write about and are unable to decide between them.

    • JohnBuridan says:

      Hahaha! Perfect! Unfortunately our argument about what to argue over did end in an agreement. 😛

      We have agreed on arguing whether or not contemporary Islamic polities have the tools to adopt liberal democracy, and the conditions required for that to be possible. Liberal here means general buy in to the declaration of the rights of man and citizen from 1789 (or something significantly similar) and democratic means collective self-rule, non-tyranny, resistance to elite capture, and resistance to tyranny of a simple majority.

      • JohnBuridan says:

        Islamic majority countries have been and some currently are “liberal democracies”, but our conversation, I think, concerns whether the tension between Sharia and liberal democracy can be resolved from within contemporary Islamic teaching.

  29. Prussian says:

    Question on taboo topics: I would love to have an adversarial collaboration on the subject of whether or not Islam represents a threat to modernity, largely because there is so little honest debate on that. However, I get that that is quite taboo today. Would that be on? Else, I’ll have to think again.

  30. Immortal Lurker says:

    If these are successful, will there be a way besides Scott’s patreon to fund future ones?

    I do have an idea for a website, where people could post bounties for certain topics, and authors could offer rates on certain topics. I am publicly precommitting to putting actual work into that idea, if this contest produces work that I find valuable.

  31. flylo says:

    I am interested in debating whether tokens and ICO’s will have a substantial impact on business practices in the next 10 years. I believe that they will. By “substantial” I mean the average person on the street will be aware of their influence on business practices.

    • defteq says:

      I might be interested in a collaboration – I am a bit of a crypto skeptic (especially where it pertains to tokens and ICOs) – and a practitioner in the current financial system. I think we would need a more rigorous definition of “substantial”, but we could discuss.

  32. Ski says:

    Anyone want to take either side of something like foreign aid/charity has been a major driver of large scale improvements in human development and/or health. I work in the field but would be up for taking either side.

    • Smith EE says:

      Hi Ski, I’d definitely be interested in doing this. I’m an undergrad economics student with no claim to any expertise, but I’d be willing to do as thorough and in-depth an investigation as you’d be up for. I’m inclined to argue in favor of the proposition but I’m not too partial to either side.

      I’ll be able to devote significant time to this from mid-May through the end of July. I really am a novice in the area, but if you’re willing to commit to the project, I’d be willing to put a lot of time into research–I could read plenty of relevant books and studies over the two and a half months. I’m very interested in effective altruism and feel that a thorough and nuanced summary of the research could be fairly useful for EA organizations.

      If anyone else wants to join, particularly people with more anti-aid/charity views, I think another researcher would absolutely make sense.

      How can I contact you?

    • Mark V Anderson says:

      I’m not impressed. He spends most of the essay giving Caplan’s point of view, which was pretty convincing, IMO. When he finally got to his objections, he seemed to have three reasons to be against Caplan: 1) Scott Ar himself liked school and didn’t value credentials over learning, 2) The usefulness of universities to society for inventing things like the transistor and the Internet, and 3) He doesn’t like the politics of the people he suspects would implement the de-education of society. Very weak arguments against the billions we now spend on educating people who aren’t interested in learning for its own sake and will not make those fabulous inventions. He gives the impression that Caplan is 95% correct. I’ll agree with that assessment.

  33. TracingWoodgrains says:

    I’m interested in an adversarial collaboration about K12 education. I strongly support ability-based grouping, early specialization, gamification and game-based learning, and the critical priority of developing and using high-quality online education tools with an explicit goal of being entertaining. My stance is that the current culture and focus of US/western education is actively damaging many of the brightest and most eager students within the system and that the bulk of the conversation around it is focused in the wrong direction. Further, that this topic is placed as much too low a priority within the rationalist sphere.

    That has potential to be pretty broad, so I’m happy to narrow it down with an interested collaborator.

    • Michael Pershan says:

      I’m a k12 math teacher. My read of edu research is that it won’t support the “actively damaging” claim at all, first because school doesn’t have enough power to really harm anyone so significantly. But also I think your pedagogical assumptions are backwards, and there’s a huge research literature on online learning, game-based learning, learning from media, learning while being entertained, etc. Our disagreement might be too broad and the evidence base too thin for a productive collaboration, though.

      • TracingWoodgrains says:

        I’d be happy to at least run through the basics of where our views differ and the evidence we see, and see if we can make something productive out of it. If you’re on reddit, my username there is the same as here. I’m interested in hearing more of your thoughts particularly about the differences in our pedagogical assumptions. One topic in particular where we could focus is on ability-based grouping, since that is generally looked at with severe (and, I think, misplaced) distrust in the current educational culture.

    • markus says:

      I don’t think I have time for a full adversarial collaboration, at least I’m not sure enough to be able to commit to participating.

      What I do have time for is some kind of double crux (if I have understood the term) around the area. Trying to gain insight into where ways part and what evidence that might close some roads down rather then going after that evidence as well.

      I´m a high school teacher doing a phd in education (and I realize that is counter-signaling) and have a lot of sympathy for the things you bring up, but pressed to an standpoint I view them as mostly unsubstantiated.

      Contact me using myusername.stoor followed by googles mail solution.

    • maintain says:

      If you gamify a waste of time it’s still a waste of time. Debate me.

    • JohnBuridan says:

      I’m professionally interested in all these things, and am up for some double-crux on certain educational issues. Namely, what makes good gamification, what is an education game anyway, and what can’t gamification do?

      I’m against “Early specialization” if that implies the tradeoffs I think it does.

      I believe high-quality online tools are important, but only under certain, relatively narrow, circumstances can they be effective.

      Message me on the Discord Server.

      • JohnBuridan says:

        Note: I have made and implemented 3 educational games. I’m generally committed to a liberal arts view of education.

        My view of human nature is that bad motives for becoming educated are acceptable, so long as one actually tries to become educated, just as mixed motives for acting like a good person are acceptable to society at large.

  34. Tracy W says:

    Central planning and collective ownership of property, attempted at an economy wide-scale, have an internal logic that lead to dictatorships. I won’t say inevitably – a government might change their mind first. Or some outside force might intervene, like a war or a coup.

    • melboiko says:

      > Central planning and collective ownership of property

      One of those things is not like the other, though. That’s a bit like saying “unregulated free markets and inheritance laws have an internal logic that lead to erosion of the commons and ecological collapse”. Even if that’s correct, unregulated free markets isn’t the same thing as inheritance laws.

      So do you mean:

      a) Both central planning and collective ownership of capital (I take it you mean ownership of the means of production, not of personal property), independently, will lead to dictatorships? or,
      b) Only a combination of the two will lead to dictatorships? or,
      c) You believe it’s impossible to have collective ownership of the means of production without central planning? (This would, I think, need an argument on its own.)

      Note that depending on how you define central planning and collective ownership, they might actually preclude one another. E.g. Lenin’s definition of the Soviet Union as “State Capitalism”, since the state, as an independent entity, owned the capital, and the workers (=the actual collective) had no power on how to use it nor its profits.

      If all you mean is that “centralization of the means of production into the hands of one agent leads to dictatorships”, then you might have trouble finding an argument partner in SSC (of all places), since even rabid socialists like Bakunin, Emma Goldman and George Orwell have been arguing the same from the beginning.

      • shenanigans24 says:

        There’s not much of a distinction. Central planners telling property owners what they can and cannot do with their property means that by definition the central planners are the real property owners.

      • Tracy W says:

        Good point. Let’s start with an “or” then.Central planning or collective ownership of property, attempted at an economy wide-scale, have an internal logic that lead to dictatorships.

        • albatross11 says:

          This is basically Hayek’s _Road to Serfdom_.

        • pjiq says:

          oh I’d argue against the central planning thesis or the collective ownership thesis. Either or.

          But the “have an internal logic that lead to dictatorships” is somewhat weaker of a claim than something more specific like “lead to dictatorships” or “lead to less speech freedoms” or something else like that.

          But I’m down with the way you worded it too, let me know (via a comment reply here)


          • Tracy W says:

            Tracyw77 – at – gmail.

            Wording is because life’s uncertain, as I mentioned in my starting comment.

  35. Inside a semicircle of displays says:

    I’d like to argue that machine translation will not replace professional translators (until the advent of AGI). I’m a professional translator who’s built a view on this topic over the past 15 years of being told that my job will be gone within five years. I vaguely remember we’ve got a few guys working on the software side of machine translation around here, so that might be interesting. Any takers?

    • amaranth says:

      how could it? language can encode AGI-complete subtleties, so you’ll always have texts that encode subtleties that the latest weak AI can’t translate well

      (this doesn’t preclude weak AI replacing *most* professional translators)

      • Inside a semicircle of displays says:

        >(this doesn’t preclude weak AI replacing *most* professional translators)

        I’d argue against that as well, and even leave literary translation completely out of the argument.

    • A1987dM says:

      Do you mean that it won’t replace any professional translators, that it won’t replace most of them, or that it won’t replace all of them?

      • Inside a semicircle of displays says:

        The vast majority (90%+) of human translators won’t be replaced by machine translation.

        • Adam Berman says:

          The differing opinions on this topic on both sides of the aisle are really interesting, and I’d love to read the result of a collaboration. Notably, I’d like to bring up that Victor Mair at Language Log (a blog founded by academic linguists at UPenn; Mair is one contributor of many) and some others there have seemed alternately optimistic and pessimistic about the use of translation software in the short to mid term for at least some applications.

          I don’t see software displacing the UN translators anytime soon; the stakes are too high. But a lot of things, like ad copy, are often translated by humans and the results are less than stellar. When it comes to software replacing human translators, I’d expect it to really have an effect at the low end.

          That said, it may really be that the vast majority of applications for professional human translators (which I assume is what you mean, rather than “secretary who learned English in university and is asked to translate a document”), the stakes are high enough that we won’t see much displacement.

          • Inside a semicircle of displays says:

            The differing opinions on this topic on both sides of the aisle are really interesting, and I’d love to read the result of a collaboration.

            That’s why I’d like to do one!

            That said, it may really be that the vast majority of applications for professional human translators (which I assume is what you mean, rather than “secretary who learned English in university and is asked to translate a document”), the stakes are high enough that we won’t see much displacement.

            There’s a number of interesting aspects beyond just methods of translation. Corporations love shifting liability, that’s why I’m insured for unreasonably high sums of damages. Corporations also love their confidentiality and are sometimes legally required to enforce it, for instance with price-relevant information like merger announcements. So, no matter if e.g. a cloud-based machine translation tool becomes better than me, it’s not going to take those jobs off me.

    • JohnBuridan says:

      I am always struck by the way people talk about ordinary language in computer programming;
      it’s often like they are not fluent in any foreign language at all…

      The sheer quantity and complexity of idiom formation, slang, word choice, and “grammar exceptions” makes me gape in awe. I find it beautiful.

      As for translation. There’s enough philosophical complexity there to span volumes, and the requirement of “good judgment” is ever present.

      We will have machines doing translations of Cervantes, when we have machines interpreting human legal codes.

      What language(s) do you translate?

      • Inside a semicircle of displays says:

        English and German, running my own small firm, mainly doing medical technology/IT stuff. At my last dentist appointment, I noticed I had translated a fair share of the UI and the manuals for their dental X-ray setup, that was pretty cool.

  36. Scott Alexander says:

    People who have posted things and other people have said they’re interested – are you currently continuing the discussion via email, or is everyone floundering around not knowing what to do once it looks like there are some takers?

    • Erusian says:

      The latter, in my case. Specifically, I’ve made a few proposals and no original person hasn’t replied yet. Perhaps they will though.

      • Truism says:

        Similarly, I made a proposal and had interest, but haven’t been contacted after posting contact info.

    • christhenottopher says:

      @NoRandomWalk and I have already agreed on basic parameters we’re looking for, time commitments, and starting check in dates. So, for the early stage we’ve gotten to so far, we’re progressing well on my question!

    • Douglas Summers-Stay says:

      Our conversation about the likelihood of AGI is continuing by email.

    • anglevice says:

      I’ve reached out to BJH as they seem the most serious about pursuing my inquiry but haven’t heard anything yet. I’ll make a throwaway email later that people can reach me at

  37. melboiko says:

    I think it would be worthwhile doing something on transgender science, but I currently do not have the time or wherewithal to collaborate with anyone. (I fear adversarial collaboration might be too distressing for me anyway). Anyway, there are certain issues for which I would like to read arguments from the transgender-critical side—as long as they’re good, empirical, data-supported arguments (unlike, say, the crappy data of that one “autogynephilic/homosexal” silly dichotomy, which have failed to replicate numerous times and at this point is basically horoscope).

    Some topics I think would be worth discussing:

    – Feminist Janice Raymond says that “the problem of transsexualism would best be served by morally mandating it out of existence.” Pressed for clarification, she explains that “what this means is that I want to eliminate the medical and social systems that support transsexualism.” Meanwhile the evidence-based medical establishment (or the sexist patriarchal technomedicine, depending on your side) prescribes transition as the most effective treatment for transgender people. How strong is the evidence that a) transgenderism is a persistent, damaging condition and b) transition is the best treatment for it?

    – A related argument: Transgender people are 4× more likely to be poor, and have difficulty with finding employment etc. This can be especially troubling in countries with dysfunctional health insurance systems, where people have to pay thousands of dollars for simple treatments. Should transition be covered by insurance? (This is a moral argument). For example, suppose someone is really distressed over going bald, to the point of becoming depressed. Baldness can be treated by hormones (in fact by some feminizing hormones that can also be used in transition). Is there any reasonable argument for providing medical insurance to transgender people, but not to balding people? Should both of them get medical insurance? (If it’s cheaper to treat baldness than baldness-induced depression, does this effectively put into question its “cosmetic” status)? Neither?

    – Recently there’s been controversy over childhood gender dysphoria. One study purports to show that most children with gender dysphoria will have gotten over it by adulthood. Transgender-critical advocates conclude from this that childhood transgenderism goes away and gender-dysphoric teenagers should not be given puberty-delaying medicine. Pro-transgender people contend that there’s a difference between ephemeral gender dysphoria and a persistent one, so they don’t frame this as “most trans kids cure themselves” but as “most gender-dysphoric kids don’t turn out to be trans”; and furthermore that if someone reaches puberty and is still dysphoric and insisting they want to delay puberty, their choice should be respected. Anti-transgender advocates contend that young teenagers are not fully autonomous decision makers, and transgender ideology can bias their choice in ways that will turn out damaging for them. In my view the important question is the practical one: If a dysphoric teenager wants to take puberty delayers, should they be allowed to? What are the real risks involved in taking puberty-delaying medicine? What are the real risks involve in not taking them (how likely is a dysphoric teenager to be negatively affected by pubertal development, and how does this compare with the opposite risk of taking blockers for ephemeral dysphoria, and how are these two risks weighted by the chance of young-teenager dysphoria being persistent)? How does this relate to general ethical arguments about the autonomy of minors?

    (And then there’s the always-popular question: is transgender identity intrinsic/biological, and how much? But I don’t think there’s much to be gained by arguments here; we still need more hard data. There’s been some interesting results so far, and the scientific establishment appears to be tending towards genetic factors being significant, and social ones less so (thus the Endocrine Society position statement); epigenetic factors appear to play a role too, cf. the case of DES fœtal exposure; but causality remains little understood, and at this point I don’t think we’ll uncover the truth through arguments. The question of etiology is interesting, but I don’t think it’s very urgent; the most urgent question is “should transgender people be supported in their desire to transition?”, and related concerns like “is dysphoria a grave condition? Is it widespread? Is it persistent? Is transition advisable?” There’s been a good bunch of studies on the latter questions, so an adversarial collaboration would have concrete data to chew on.)

  38. Philosophisticat says:

    I thought about whether anything like this would be worth doing on a philosophical topic, but I can see some worries (some I think reasonable, some I think unreasonable)

    – Some people think, in my view wrongly, that philosophical matters are not matters of fact, or that there can be no evidence one way or another for philosophical propositions. What is true is that there is less consensus on how to assess evidence for philosophical claims than for some other disciplines, less common ground, and more interaction between the assessment of evidence for one claim and stances on other very controversial claims.

    -For some fields, an intelligent amateur with a basic grasp of general statistical concepts is in a reasonably good position to form judgments by scanning the literature, but I don’t think any amateur would be in a good position to judge the state of the evidence on philosophical questions by scanning the literature, so you’d need experts for it to be worthwhile.

    -It’s not clear how much of an improvement the result would be over things like

    So I dunno.

    • gvprtskvnis says:

      I think those concerns are valid, but I also think doing an adversarial collaboration in philosophy would be interesting, and I might want to do one if someone else is interested.

      • Philosophisticat says:

        What is your level of expertise in philosophy?

        • gvprtskvnis says:

          I’m an undergrad studying philosophy, so I’m certainly not an expert, but my thesis is in a topic related to the question I proposed above. (Since I’ve talked about this with professors, I doubt I’m completely off-base in my understanding of things, though obviously I can expect to make some mistakes.)

    • says:

      I agree with Philosophisticat.

      However, maybe one solution is to have collaborations in which at least one partner has formal training in philosophy? I am one such person and I’d be interested in doing collaborations on several philosophical topics. Here is a nonexhaustive list:

      –Consciousness (I think physicalism is probably false and phenomenal idealism is probably true)
      –Ethics (I think hedonistic utilitarianism is the devil and preference utilitarianism is OK, but only because Kant saves it)
      –Decision theory (I think CDT is very wrong, EDT or FDT might be right, for more or less the reasons MIRI gives)
      –Anthropics/Induction/Bayesianism (Lots of stuff to say about this…)

      • gvprtskvnis says:

        I’m a non-expert (undergrad) in philosophy, and I would be willing to support hedonistic utilitarianism. (I’d also like to hear your thoughts on anthropics, since I’m particularly interested in that.)

        • says:

          OK, well why don’t we continue this via email? Even if we don’t end up doing a collaboration we could at least have a nice conversation.

      • Michael Cohen says:

        I would be interested in defending Parfit’s account of wellbeing (taking pleasure in things of real value) against a preference theory of well being, if as I take it, you think wellbeing consists in one’s preferences being satisfied. Would enjoy discussing over email, but probably don’t want to do an adversarial collaboration format.

      • Philosophisticat says:

        I think physicalism is true (but I’m not really an expert on this), all forms of utilitarianism are false, and CDT is correct. I’m still a bit skeptical about the value of this kind of collaboration but I’m happy to talk about them.

        • says:

          I agree that all forms of utilitarianism are false. I’d love to talk about CDT or physicalism with you. Email me if you are interested!

    • JohnBuridan says:

      I agree with everything you’ve been posting. I imagine, you like me, feel that there is an extreme knowledge imbalance between yourself and any collaborators? And in Guided By The Beauty of our Weapons, Scott, like Aristotle, mentions the need for the discussion to be between people who consider themselves equals.

      Peerless Philosophisticat, you probably could find some other philosopher with a similar knowledge base as you. I am not up for writing philosophy right now, though, unless to defend some concepts traditionally found in virtue ethics, which I believe are horrifically absent in the rationalsphere.

    • JohnBuridan says:

      I think philosophical discussion in this case should not be limited to the general academic concerns, since, as you state, no one here is likely to do better than SEP. However, we here are likely to provide salient corrections to common philosophical mistakes in the rationalsphere.

      I see you complaining about these mistakes not infrequently.
      So maybe what is the biggest philosophical misunderstanding held in common among LessWrongers, Eliezer, Scott, Tetlock, Caplan, Hanson, etc?

      • Philosophisticat says:

        I think philosophical misunderstanding is a bit of an “unhappy families are each unhappy in their own way” kind of thing, and I doubt there’s any philosophical misunderstanding that all of those people hold in common. Some particular idiosyncratic philosophy memes of the rationalist community that I think are mistaken – the obviousness of consequentialism, the clear superiority of functional decision theory, the rational significance of calibration exercises.

        This is only true of a smaller subset of the community, but Scott and some others have been seduced towards an especially implausible version of utilitarianism – negative average preference utilitarianism, for, as far as I can tell, extremely bad reasons (assuming that the reference that they provide – a paper by Roger Chao, represents the reasons they have).

      • JohnBuridan says:

        I’d also add that there might be widespread misunderstanding of the role of literature and art and aesthetics in human life. (Since I started hanging out around here 4 years ago, I find myself constantly slipping into teaching poetry as though it’s philosophy.)

        It seems to me that one of the roles of philosophy is to gadfly. Perhaps, you are more a grumpy Meletus, than a gadfly by temperament?

        • Philosophisticat says:

          Perhaps, you are more a grumpy Meletus, than a gadfly by temperament?

          Them’s fighting words.

      • Answer: you can solve philosophical problems easily, and, in particular, without learning much philosophy.

  39. fion says:

    Think this is a really fun idea, and I hope it produces something interesting/useful!

  40. Froolow says:

    I remember Scott mentioned something about this a couple of years ago so I tried a couple of adversarial collaborations as an experiment (the WordPress site I hosted them at is named in honour of this one).

    I stopped because they were insanely time consuming and I had just started a PhD, and I found it really difficult to recruit people – I was trying to recruit through subreddits with a particular political slant (ie recruit renewable advocates from the environmental subreddits) and was basically told to stop blogspamming them. I guess its fair enough for them to enforce whatever rules they want in their walled garden, although I think what I was trying to do would have been of interest to them if they’d given it a go. Anyway I basically lost interest after I realised I was spending longer trying to recruit people than actually running the collaborations!

    I have a gun control one written up somewhere I could polish up too – I’d be interested to see how different/similar the gun control one developing above looks when it is done.

  41. richardberg says:

    There is no social utility and/or economic value in blockchains. Every proposed application would be better implemented with a conventional append-only database, or at worst, an N-of-M consensus algorithm between semi-trusted parties.

    • flylo says:

      Are you interested in the topic I suggested above?

      • richardberg says:

        My background is primarily (1) CS theory (2) FX market making. My perception is that tokens are more of a marketing or fundraising strategy, i.e. that their overlap with the database or banking sectors is much more limited than typical cryptocurrencies.

        But sure, to the extent your thesis overlaps my areas of expertise, I’m happy to help / refute. ars@[username].net

  42. spork says:

    I just want to remark on the equanimity with which we’re all accepting the whole premise of undebatable taboos. I’m not judging. Maybe I’m just not used to this level of openness about it. Is it something new?

    This refreshing kind of openness makes me want to pry a bit more into the topic. Clearly a taboo debate is different from a settled debate. Sure, we have some basically settled debates in which the partisans of the losing side are still sore and possibly in denial. When we know that all they would bring to the debate are points that have already been discredited, it makes sense that we might not want to go through the ritual of rehashing the evidence for the n+1st time. So settledness is one reason to not want to open a certain topic of debate.

    A different reason is that some controversies are undebatable because reasons that stand up to community standards of rigor are not available for either side. Or maybe the partisans of undebatable topics rely on reasons from non-overlapping sets of assumptions, so they are doomed to talk past each other. Fine, we’re right to not want to host their arguments about whether Jesus was really resurrected – but not because the topic is taboo.

    This tells me that a taboo topic is neither settled, nor epistemically undebatable. So are they just topics too scary to think about with the sobriety that an open debate requires? I notice that though I can guess some taboo debate topics, I’m not eager to even name them in a list. I notice that Scott did not do this either. But if braver people than me care to produce one, it would be really interesting to examine if there are patterns in how we justify allowing a taboo keep us from debating a topic. The list would help because I suspect that taboos shroud different debates for slightly different reasons.

    Or is the debate about which debate topics deserve to be tabooed and why itself a taboo debate?

    • bean says:

      It’s a practical matter on Scott’s part. He gets in hot water frequently enough, and doesn’t want more by posting an adversarial collaboration about Horrible Banned Discourse or the like. He didn’t say he’d shun anyone involved, just that he wouldn’t post it here.

    • skef says:

      One of our host’s frequently offered reasons for tabooing topics is to avoid the negative attention they bring to the site. So you can think of taboo topics as those that enough people consider bad to even discuss, and taboo positions those that enough people consider bad to even entertain.

      This being an open site, there’s no tractable way to alter what topics and positions are taboo on this view. And a list of such topics would self-qualify.

      • spork says:

        …taboo positions those that enough people consider bad to even entertain

        That can’t be right. Most positions that people consider too bad to entertain are not taboo. That’s because we’re comfortable with openly and confidently stating the reasons why those positions are so bad. For a position to be taboo, we need to consider it too bad to entertain and also lack reasons for thinking so that we would be willing to state in public.

        • skef says:

          I meant “bad” in a moral sense, not merely mistaken.

          • albatross11 says:

            In some times and places, open discussion of some ideas is liable to bring an angry mob to your doorstep demanding blood. At that point, the rightness or wrongness of your arguments becomes less relevant than the size and violence of the mob baying for your blood.

            This is, as best I can tell, independent of the rightness/wrongness of defensibility of your claim. There have been times and places where contending that man evolved from other animals, or that slavery is morally wrong and should be abolished, would get that angry mob to show up.

            It’s a very bad thing overall for the best available answers to moral and factual questions to be affected by angry mobs, but that’s the world we live in, and it’s prudent to take notice of that fact.

          • spork says:

            I meant “bad” in a moral sense, not merely mistaken.

            Yeah, me too. I was saying that it’s usually pretty easy to explain openly why bad positions are bad (in a moral sense). The distinctive thing about taboo positions is that for some reason, these are bad and yet we cannot articulate in public what makes them bad. It’s a philosophically interesting category of positions.

          • albatross11 says:

            Are there different categories of bad positions? Like

            a. Probably untrue but seems convincing to the uninitiated, leading to widespread error.

            b. Probably true but likely to lead to socially damaging consequences if widely known.

            c. Probably untrue but likely to lead to socially beneficial consequences if widely believed.

            d. Mindkilling idea that corrupts the very moral/intellectual integrity of anyone who engages with it.


            Maybe at least thinking about what category things are in is a starting point?

          • LadyJane says:

            @albatross11: I think a lot of taboo topics fall under A or B, or are a mix of both. There may be some grain of truth to the controversial argument in question, but to a far lesser degree than people think; nonetheless, because the argument sounds convincing and/or confirms people’s existing biases, people will exaggerate the degree to which it’s true, often in a way that leads to negative social outcomes.

            To use a very thinly-veiled analogy, let’s say that Garmoks have been saying for centuries that Zobos are inherently less intelligent, and using that claim to justify the oppression of Zobos. Then some study proves that, even accounting for all other factors, Zobos have an average IQ that’s 3 points lower than Garmoks. It’s doubtful that 3 point difference would be enough to cause any real difference in life outcomes for Zobos, and it certainly isn’t enough to justify treating Zobos any differently. Unfortunately, a lot of Garmoks still believe that Zobos are significantly dumber than them, and actively want to believe that, so Garmok newspapers run catchy headlines like “Zobos proven to have lower IQs than Garmoks” and leave their readers to assume that it’s a 30 point difference instead of a 3 point difference (which fits their intuitions better). Suddenly, Garmoks start believing that they have a right and an obligation to control the lives of Zobos again, and start pushing to bring back all those oppressive policies that were repealed a few years ago, leading to severe racial tensions that make life worse for both groups. Wouldn’t everyone have been better off if the results of that study hadn’t been so widely publicized?

            I’m not saying the comparable real-life situations are anywhere near that extreme, but I chose a deliberately extreme analogy to highlight the type of effect I’m talking about.

          • albatross11 says:

            Lady Jane:

            Suppose we wanted to estimate how many taboo ideas were either:

            a. False

            b. Socially destructive

            How would we go about it?

            One idea that occurs to me is to look at the taboo ideas in other times/places and see what the distributions look like, assuming that we can probably look at other cultures’ taboos without being mindkilled or going off into a spiral about whether we should even be discussing this stuff in public. The interesting question is how we’d figure out a representative list of taboo ideas/topics. Availability bias is a big danger here–we can all probably think of taboo ideas of the past that we now think of as either obviously right (example: evolution) or at least perfectly respectable opinions held by well-intentioned and serious people who may be wrong (example: atheism).

          • spork says:

            @LadyJane – No doubt you’re right, but notice that if your fictional universe is like ours, it would be taboo to say something as constructive and levelheaded as:

            “People, the mean difference between the Garmoks and the Zobos is a measly 3 IQ points, which, in the small samples that individuals encounter, is indistinguishable from statistical noise. So can we just please accept this scientific finding, acknowledge its irrelevance to any issue that matters, and move on?”

            What I’m trying to say is that other things being equal, taboos are bad and need to be periodically challenged. (Think about former taboos on discussing homosexuality, polyamory, masturbation etc.) I fully understand why Scott wouldn’t want his blog to be hijacked by the heat that’s inevitably generated when we collectively work on digesting a taboo and moving forward into a post-taboo world where the sun still rises and the birds still sing. Scott wants the discussion to focus on another topic, and I respect that. But if I had to express my view of what we ought to do generally, it’s that we should seek opportunities to move issues from the territory of unmentionable taboos to either the territory of the openly debunked (as we are doing with racist views about the economic harms of immigration) or the territory of the openly defanged, as we did with taboos on discussing various issues surrounding transsexuality.

            Edit: So since I’ve been drawn out to express a view, let me say that I’d consider collaborating with a thoughtful adversary who thinks that taboos should, in general, be left alone and unpoked.

          • LadyJane says:

            So since I’ve been drawn out to express a view, let me say that I’d consider collaborating with a thoughtful adversary who thinks that taboos should, in general, be left alone and unpoked.

            Good luck finding someone around here with that view. I certainly think that taboos, on general principle, are a very bad thing.

          • albatross11 says:


            To decide how comfortable you are with that scenario, you also need to think through the case where the tabooed topic has facts that actually do support some of the bad stereotypes you’d like to avoid.

            Suppose it turns out that some Garmok stereotype about Zobos is substantially borne out by the data. Maybe Garmoks never invite Zobos over to their homes because Zobos are rumored to suffer from terrible flatulence, and it turns out Zobos really do have way more flatulence than Garmoks.

            Now, the truth is that Zobos really are a lot gassier than Garmoks, just like the stereotypes say. But publicizing that is not going to help *at all* with the lack of dinner invitations the Zobos are getting to Garmok homes. So should we push back on people publicizing that data?

          • spork says:

            But if they could just talk it through in the open, they’d have a chance to work toward a solution, or at least some kind of accommodation. Without that, Garmoks who snub Zobos may be forever suspected of discrimination, when really all they’re doing is trying to avoid finding themselves in a fart cloud. Garmoks might also misinterpret the behavior of other Garmoks as being somehow more than just fart avoidance, mistaking it for something deeper. That kind of thing could feed a spiral that drives the two groups further apart, and no remedy can be contemplated if the farting can’t be discussed. If it could then maybe they’d all agree to at least hang out together on windy beaches. Down the line, maybe homes can be equipped with powerful, partially subsidized ventilators. Whatever it takes to prevent mere farting from leading to unhealable social division.

        • Jiro says:

          For a position to be taboo, we need to consider it too bad to entertain and also lack reasons for thinking so that we would be willing to state in public.

          First, a position counts as taboo if people can’t *support* a position in public, not if they can’t name the position. Giving reasons why the position is bad is acceptable.

          Second, most people’s reasons why the things are bad will be pretty terrible, since they have no incentive to actually come up with good arguments (since the opposite position is never argued in public so that they would have to dispute it).

          • Aapje says:

            Good counterarguments can even be taken as evidence of sympathy with the other side (since they are often more charitable and less extreme claims than the bad counterarguments).

          • albatross11 says:

            Also, because both the writer and reader know what conclusion they want to reach (taboo idea = bad), the average quality of anti-taboo arguments is very low. In muggle-realist circles, I sometimes find myself arguing against some conclusion which I think is wrong, but which seldom gets a high-quality argument against it because low-quality arguments and pointing and sputtering in outrage are all that’s ever produced against the idea in public.

        • realwelder says:


          Most positions that people consider too bad to entertain are not taboo. That’s because we’re comfortable with openly and confidently stating the reasons why those positions are so bad.

          I read this as saying that a position is not taboo if one is free to publicly condemn it. I disagree with that.

          For a position to be taboo, we need to consider it too bad to entertain and also lack reasons for thinking so that we would be willing to state in public.

          That may be true of some taboos, but does not seem to me to be a good guiding principle.

          There are taboo positions which can be supported by credible evidence from orthodox sources.

          Here’s an example:

          A certain negative trait is found in a greater frequency in group X,

          where X is a race, ethnicity, sex/gender, sexuality, or religion.

          (This is especially taboo if suggesting the trait(s) are intractable, and even more so if calling it inherent.)

          Even presenting both sides of the case without denouncing the taboo side as false and/or evil is taboo.

          Or, if not claiming data, take one’s personal experience. Compare, for instance, the following paragraph applied to two groups: people from a neighboring town vs people from a distant part of the world.

          “A bunch of them went to my school growing up. They would start fights with us all the time. I was friends with a couple, but most of them were jerks. When I started running a business, I hired whoever was the most qualified and seemed the best fit. But every single time I’ve hired one of them, it’s been a disaster. So I just try not to have anything to do with those people anymore.”

          Same statement, same reasons (stated publicly), but one a taboo position, one not.

          • spork says:

            Ok, there’s certainly something right about this. I was thinking at first about taboo topics, but maybe there are just taboo positions. Expressing your view on a topic is not taboo as long as it falls on the “safe” side, like “In all my years in HR, I’ve found Filipino employees to be every bit as capable as any other.”

            So I guess we shouldn’t say that it’s taboo to broach a certain topic, only to debate that topic (or otherwise suggest that the orthodoxy regarding it could be challenged). In retrospect that seems obvious, but this conversation helped me get some clarity on this small point.

          • albatross11 says:

            I’d say both taboo topics and positions exist. And that sometimes, you see someone trying to move the taboo from the whole topic to a specific position, as with the first Vox article on Murray and Harris.

    • albatross11 says:

      Some discussions of questions of fact or morality attract a lot of angry attention by people who are outraged by the very existence of anyone taking one of the sides in the discussion. As I understand the Vox/Harris/Murray kerfluffle, this is essentially what happened w.r.t. Murray’s belief that the black/white IQ gap is largely genetic.

      On one hand, there are discussions that cause genuine distress to some people reading them, and some of those will cause an organic outcry.

      On the other hand, there is pretty clearly a strategy being employed by some groups of causing a loud outcry when some views are expressed, in order to shut those views out of the public sphere. (To my mind, this is exactly why institutions (universities, employers) should not bow to this kind of mass-outcry. Yielding to the heckler’s veto encourages that which you want less of.)

      And on the gripping hand, if the activists manage to get you fired or get your inbox flooded with credible death threats and your blog DDOSed, it may not matter to you personally whether that was done by authentically outraged people or strategically outraged people. Their strategy works, in the same way that it may be prudent to pay the mafia guys for fire insurance even though they’re just extorting money from you.

    • Deiseach says:

      Seeing over on the sub-reddit how one particular debate about A Taboo Topic went, I’d be as glad if it wasn’t raised on here. Plenty of heat, not much light, and the end was as murky as the beginning. And in my own personal opinion, some tendency to slide towards “population X is clearly inferior genetic material and it is up to us, superior genetic stock that we are, to decide their fate” (where X need not necessarily correlate with race – there’s an alarming enthusiasm for ideal solutions to the problem of the poor and what we should be doing with/to them).

      A proper adversarial debate would be great if it could be well-done, but the “well-done” is the sticking point.

  43. Caliban says:

    I think this is extremely interesting and I’d love to participate (long-time lurker, made an account just to post).

    I have a question riffing on @RC-cola-and-a-moon-pie comments. My “day job” is closely related to empirical research (I have one top publication in my field, hurray!), and as a hobby I’ve “dug deep” into data/empirics on controversial topics. BUT something I know from both my “day job” and my “hobby empirical tinkering” is that … I definitely don’t have time to give something like this the “serious go” that I’d prefer to give it. I’m still young in my field and my time right now is largely devoted to working my ass off to get my next “day job” publications.

    How would you feel about someone like me offering to “be an adviser” to any group who *can* devote reasonable time to a topic I am interested in? I wouldn’t want to “split the prize” or even necessarily have my name on it (maybe one of those “thanks in a footnote” you see on papers). Would just be interested in seeing that I can “donate the thinking time I’ve done” to topics I think are interesting/important. Ah, and to be clear, those controversial “hobby topics” would be gun control (both pro and con — once mildly pro, currently leaning con), likelihood coronal mass ejection re-creating Carrington event in our lifetime (yeah pretty specific — a while back I saw a paper I just didn’t believe the numbers in, so did some minor back-of-envelope calculations myself, ended up “not rejecting” the numbers from the paper)

    I also have to imagine that there are others in my position.

  44. Matthias says:

    Scott, I’m open to giving 500 USD in extra prize money to distribute as you see fit. (Either more money for first prize or something else.)

  45. rahien.din says:

    @Hence, should we do something on NHST? Specifically :

    Science should utterly abandon NHST, in favor of other statistical measures.

    I would (shocking!) start from some counter position.

    You’ve already done a ton of legwork (and publicly) so I would offer to begin by reading your tract and the papers therein in full. Furthermore, I only have limited time. You would have to pre-grant me some general slowness. But I feel like we can reach satisfying convergence, and this discussion can only benefit my practice.

    (If someone with more time wants to take up this mantle, I would accede and look forward to the result.)

  46. aasitus says:

    I’d be tentatively willing to argue that social media has not caused political polarization. Also, if there is an increase in polarization attributable to social media, it has more to do with how mainstream media are interacting with social media than social media itself.

    This is related to the common claims that social media is doing something bad to the political environment in Western countries. What exactly is meant by increased polarization can be discussed and defining it would probably be a main point here, but it could refer to a polarization in opinions or increased controversy in online political discussions.

    I’m not entirely sure yet that I actually have time for this, but if anyone’s interested, let’s talk about it. I’m a social science PhD student interested in the subject, but while I’ve done some reading on it, I’m not intimately familiar with it.

    • Gazeboist says:

      Increased polarization would probably be defined as one or more of:

      – Decreased willingness to engage in debate with political opponents.
      – Decreased belief that opponents are arguing in good faith and/or increased willingness to argue in bad faith.
      – An increase in the strength of the filter-bubble effect.
      – A shift away from goal-oriented politics and towards political tribalism.

      Plausible positions on whether this increase in polarization is happening at all include “yes”, “no”, and “yes, but the prior era was unusually de-polarized”.

      • albatross11 says:

        I think it’s going to be easy there to find evidence that there’s increasing political polarization at the same time as increasing social media/internet use. But I’m not sure how you’d work out whether this was causal or not. Maybe look at uptake of Facebook in different regions/states/countries vs political strife in those regions/states/countries?

        • aasitus says:

          My understanding is that there’s limited evidence of increased polarization, at least when polarization is defined as divergence in attitudes or opinions. But I base this on a small number of studies (such as this) and may be wrong. Comparing social media usage to levels of political strife sounds like a sensible approach; another would be looking at demographics, which I know has been attempted.

      • aasitus says:

        Political polarization has sometimes been defined as divergence in a distribution of opinions, which can itself be defined in several ways; this is one example of what I meant by polarization of opinions.

        However, that’s just one example, and not necessarily the most meaningful one. I like your proposals, too. There’s a fairly extensive literature on filter bubbles and echo chambers. To me, the rest sound (even) more difficult to study, but may be closer to what people are actually worried when they’re worried about the internet and polarization. An adversarial on this should perhaps focus on picking some relevant aspects that have been studied extensively, and review the literature on those?

    • tayfie says:

      I would be interested in collaborating with you. My current view is that social media encourages and inflames ideological polarization and this damages discourse in all contexts. It bears significant responsibility for the current climate of division.

  47. episcience says:

    I hold standard left-wing feminist views, and would be interested in defending these. I’m not an economist, so am probably not going to be much use on wage gap issues, but much else is probably fair game.

    • says:

      Can you be more specific about “standard left-wing feminist views?” That could be something I agree with, or not, depending on how you cash it out. It also could be something I have no expertise in, or some.

    • ksvanhorn says:

      Not sure if I’m willing to put in the effort on this one, but how about this thesis: Modern-day mainstream feminism is deeply sexist and misandric.

      • Gazeboist says:

        Biggest problem there is defining “mainstream”. You’d probably need to agree that some certain set of propositions constitutes “feminism”, or that a select list of authors/groups does the same.

        • oconnor663 says:

          Isn’t defining “sexist” at least as problematic? It seems like the two sides of this one have a strong tendency to prefer different definitions:

          – The “feminism is sexist” side prefers to refer to things like “discriminating on the basis of gender” as sexism. (So for example, affirmative action for women or men would be sexist.)
          – The “mainstream feminism” side prefers to refer to things like “causing or exacerbating gender disparities” as sexism. (So for example, affirmative action for men would be sexist, but not for women.)

          There are some potentially factual questions around sexism (like “how big is the wage gap?”), but even those tend to break down into definitional fights in practice (“should we control for childraising or not?”). When the majority of a debate centers around what question to ask, it doesn’t seem like there’s much value on collaborating on the answers to questions.

          • Aapje says:

            Mainstream feminism does both though. For example, portraying domestic violence as being perpetrated nearly exclusively by men against women is “causing or exacerbating gender disparities” in how victims are treated by the legal system, the media, people in society, etc.

      • episcience says:

        I’d be willing to tackle this one — I suppose agreeing on definitions would be the hardest part, since a lot of this is value-laden rather than empirical. But could be interesting trying to tackle it.

        • Aapje says:

          To me it seems like too broad a topic. You’d have to cherry pick what part(s) of mainstream feminism to discuss. Otherwise you’d be tackling many huge topics (including the gender earnings gap, which is one of the main issues that get brought up).

          • albatross11 says:

            My intuition is that it’s going to be very different to try to discuss an claim of fact vs a moral argument. The claim of fact may or may not be something we can resolve with current information, but you can at least see how we might all agree on it given the right kind of evidence. But the moral argument might not even be possible to agree on, since we don’t all share the same moral premises.

      • RalMirrorAd says:

        I do not believe you will get two people who agree/disagree with that statement [respectively] to be able to provide definitions of mainstream feminism, sexist, and misandric that they could agree upon.

    • LadyJane says:

      I’m basically on your side on social/cultural issues (economically, I’m more centrist than leftist, and even right-leaning in some respects), but this really doesn’t seem workable. “Standard left-wing feminist views” is far too vague to be falsifiable in any meaningful way, plus it’s almost certainly going to be one of those taboo topics that Scott warned about. Maybe try to pinpoint a particular feminist idea that you believe is highly contested right now and would be worth defending?

    • Aapje says:


      One issue with this is that Scott has already addressed many of the views that are suitable. He will probably also have very high standards for a such a topic.

      Perhaps this one is suitable: I think that the mainstream feminist claim is that men are socialized/encultured to use violence to control women, but not vice versa.

      I claim that this is false for Western men.

    • AG says:

      The objection that most have come around to in the rat-ratadj parts is that standard left wing feminists spend way too much time frolicking in the bailey, to even a point where consequentialism may demand rejecting their mottes in order to limit the damage they would do if they continue to hold the mainstream. Another variation is the usual “communism/marxism is doomed to fail by its own foundations” style argument that common social justice mottes will inevitably be executed in their bailey form in practice.
      That they are made up of majority weakman/eggmans (who any remaining steelmans nonetheless continue to defend), to the point that the the original valid points regrettably must be abandoned, and to find another way to achieve the desired end results.
      Ironically, this is like the arguments commonly used by the left against certain topics, as demonstrated by the recent Klein/Harris dustup, except applied to SJ as the topic in question.

      The pro-feminist position would have to argue that the damage done by the baileys is sufficiently minimal to be far outweighed by the benefits of promoting the mottes, and possibly that continued mainstream adoption of feminist thought will eventually turn around and deal with said bailey damage. I’m thinking “damage done by baileys is good” would be the taboo pro-feminist conclusion.

      (Not offering to collaborate, and man does this comment completely mangle the original metaphor.)

  48. drunkfish says:

    I would be interested in a collaboration on whether planetary colonization is a productive goal to pursue in the near(ish). Specifically, I don’t believe that colonizing mars/the moon in the next few decades is a productive goal to pursue. I believe that humans eventually becoming an multistellar species is very desireable, but that highly Earth-dependent bubbles on mars/the moon are unproductive ways to develop the relevant technology. The main point that comes to mind to the contrary is that generation ships are probably the way to populate stars, and colonies on mars/the moon are ways to practice for that, but I don’t find this all that compelling.

    I’m a graduate student in planetary science but only an interested layperson when it comes to manned spaceflight. There is definitely a risk that my mind is changed during the project, which probably invalidates it? Still, if somebody is interested I think it’d be worth discussing whether we can come up with a specific question on this front that we disagree on and want to research. I think this would focus more on technology than on empirical data, but that there are still legitimate conclusions that can be drawn.

    • drunkfish says:

      Replying because I left my computer and can’t seem to edit from my phone. I’d also be open to discussing the potential for asteroid mining in the near future. Again, I don’t believe it will be profitable in the next few decades. This is a lot closer to my expertise, but also something I’ve spent less time thinking about. This has the benefit of having a very straightforward metric (dollars) to try to pin down, and it’s less caught up in flashy news, but also probably somewhat less fun to work on.

      • Douglas Knight says:

        What’s your order of magnitude cost for mining and returning to earth 1000 tons of metallic asteroid over the course of 2040-2050?

    • Gazeboist says:

      I would be interested to hear what you do think a productive near-term project is, given a longer-term goal of interstellar colonization.

      • drunkfish says:

        Yeah that’s an important question. I guess first, I view interstellar travel as something happening in ~centuries not ~decades. I think the primary limitations are propulsion and power for a generational ship. The obvious solution to this is nuclear fusion, in the absence of which I’m not convinced this project is possible at all, so I think research on that should be a priority. Given that we can work on multiple things at a time, if one believes we should start working on how the colony ship should work internally (presumably it would require a functioning ecosystem instead of just storing huge amounts of food), I think that would be much better served by Earth-based experiments. The costs of moving things between planets are enormous, so putting the colonies on mars or even the moon siphon huge amounts of money from research to life support. The only real benefit I can see coming from putting them on other planets is you can guarantee no external influences from life, but I don’t buy that as sufficiently important (and it seems like something that we’d want to figure out how to control anyway).

        Something I should have mentioned in my main comment is that I think a lot of people view colonizing mars as setting up a sort of backup planet, and I think this is incredibly ill conceived. I don’t believe that terraforming mars is remotely viable with modern technology (if we can’t mitigate a few hundred ppm CO2 we don’t have a chance on mars). I also believe that the idea of a backup planet has the potential to make people worry less than they should about taking care of the only legitimately habitable planet we have (this concern is very speculative, I doubt many people are thinking “humans will make it so who cares if I die”, but it’s something I’d want to investigate because it’s important if true).

        • Evan Þ says:

          if we can’t mitigate a few hundred ppm CO2 we don’t have a chance on mars

          What if the reasons we can’t do it are political or economic, not technological? Those barriers would be greatly reduced on Mars.

          • drunkfish says:

            Economic barriers are massively increased, not reduced, on Mars. Political I guess is a fair point, but I think you only have political barriers when something is already very difficult. If the technology to mitigate climate change (lifestyle changes aren’t relevant when considering Mars) existed, I’d expect political barriers to evaporate. Nobody is opposed to preserving the earth, people are opposed to quality of life changes trying to do it, especially when they don’t believe there’s a problem.

          • Evan Þ says:

            Economic barriers are massively increased, not reduced, on Mars.

            Not necessarily. We could trivially reduce carbon emissions by banning all internal combustion engines, coal and biomass burning, etc. Of course, we won’t here on Earth, because that would drastically harm the economy. But on Mars, there isn’t an economy to harm, so a whole lot of dramatic interventions become – at least in one sense – much more possible.

          • albatross11 says:

            Also, outdoor pollution on Mars probably doesn’t need to be very strictly regulated, at least not far from habitats–there’s nothing living outside, neither people nor plants nor animals nor even bacteria. We probably care about super-toxic crap that will somehow get tracked back into the habitats, but almost all toxic wastes could just be dumped 10km from the nearest habitat and nobody would care.

            On the other hand, indoor pollution on Mars is going to be *extremely* tightly regulated, since you’ll be living in an artificial habitat that can’t count on having a new atmosphere just blow in from time to time. There’s some kind of air circulation going from habitats to greenhouses and back, probably with some filtering for dust and any toxins you need to get rid of. Anything the plants and filters can’t take out sticks around until it becomes toxic and people or plants you need for air/food start dying.

            Similarly, recycling requirements on Earth are maybe a little silly, but on Mars they’d be super-serious. Nobody’s worried about landfill space, but most of that stuff cost a gazillion dollars to send to Mars in the first place, so you want to get every bit of use you can from it.

            It’s quite possible that long-term, Martians will have a better life than Earthers, even more personal freedom. But for a long, long time after Mars is colonized, the Martians (and Loonies) will have very constrained lives where their personal freedom and privacy and standard of living is comparable to that you’d find on Earth among sailors living on a submarine.

          • Gazeboist says:


            The preferred term is Earthicans.

    • The Pachyderminator says:

      I’d be very, very interested to read the result of this collaboration. (Don’t have the expertise to participate, and I mainly agree with drunkfish anyway.)

    • John Schilling says:

      I’ll take the other side of this one if you like. You know my name, and it’s no secret I’m an alumni of USC (an educational institution).

      • drunkfish says:

        Alright I think I made a mistake, I’m sorry. I can’t picture a productive collaboration between us because of how different our levels of expertise were. I guess they way I’m viewing this is there are two sorts of collaborations, layperson with layperson where they both try to become well read and write something useful to an interested layperson, and expert with expert where the goal is to write something that can have conclusions meaningful to the field itself. As someone very non-expert, I was picturing the layperson-layperson collaboration. I was probably just being unreasonable proposing a collaboration at all. I don’t think we could productively have an adversarial collaboration because of the enormous gap between our levels of competency. I’d love to see you do this with someone at your level of expertise though, maybe they’ll show up when they see me backing out…? Sorry if I wasted your time.

        • John Schilling says:

          No problem, and my offer stands if anyone else wants to take the opposing side.

          • Athrithalix D says:

            I’m happy to take the anti-colonisation side (I understand you want to defend colonisation?) if you’re still looking for someone?

  49. Kelley Meck says:

    Here, here, here, I staked out and made an effort to find out why some people who I respect disagree strongly with me about whether political action to lessen and prepare for climate change is not clearly appropriate. I specifically went after understanding @DavidFriedman’s (I don’t understand how to indicate a space in a user name… should I include the space?) view that we don’t even know the sign for whether our greenhouse effect will be, on net, good or bad.

    I mostly got what I was looking for, which was to resolve my confusion about how our disagreement was even possible. My own priors were not changed, or not much, unless maybe it’s that I’m even more eager than usual to stress the principal of being charitable, in the sense of assuming people you disagree with or don’t understand are basically about as decent and righteous as you. Some possible climate topics, although I am in a busy month and would cede my spot on these if someone else on the “climate is real and bad” side wanted to weigh in.

    Here’s some things I could offer:

    1. There exist multiple parallel sources of temperature records which are quite detailed and generally very reliable.
    2. The error bars on how bad ocean acidification *might be* are alone enough to justify aggressive mitigation. (at least, would be if aggressive mitigation came with a clear price tag. I still don’t know how to argue with, and don’t want to engage with, people who are simply fatalistic that anything can be done, and expect all efforts to have no effect, but to waste political capital or whatnot)
    3. The cost to a state or national economy that moves from being at the back of the pack, to being a leader, for aggressively switches to renewables is near-zero.

    Post your email or how I reach you if you are interested.

    • Glen Raphael says:

      @ Kelley Meck :

      My own priors were not changed, or not much

      Wait, really? You started out with a collection of separable independent reasons why you thought quick action was necessary on global warming. I mean, we’ve all seen longer lists of that sort 🙂 , but yours was definitely a big list, not just ocean acidification. Various people (including me) provided evidence that several specific climate-related claims you had made might be overblown, to which you mostly admitted you weren’t sure and said you’d have to go check. So…are you sure your priors “haven’t changed, or not much“?

      For instance, do you still believe your original estimates regarding the magnitude of the impending worldwide climate refugee problem? Do you still believe what you said about the expected cost of sea level rise [creating] existential risks for coastal cities[…]? One of your posts had 8 points that you implicitly considered inarguable – do you still believe all 8 of those points to be valid?

      If your priors on the older issues haven’t changed…why not? Did you do more research and decide all the arguments made against you weren’t valid and you were still right all along? (If so, I’d really like to see you support THAT argument.)

      On the other hand, if you have changed your priors about some of the points people have objected to here – if you realized that many climate claims you used to argue for were false or poorly-supported – then shouldn’t that have changed your priors in general on the subject? That is to say: if your sources originally led you to believe some set of 8 points was inarguably true and now you think, say, that 4 of those points are bogus, shouldn’t that reduce your confidence in your sources and thus reduce your confidence in the remaining points?

      • Kelley Meck says:

        @Glen Raphael

        I guess saying my priors haven’t changed muddies things a bit. My views have both estimate and confidence level. Mostly, the estimate hasn’t changed. (My original statement re: refugees, that it could potentially involve flows like 1/3rd of Bangladesh leaving Bangladesh in a given decade or things like that–those weren’t responded to. I could still be right about that, and my guess for whether I’m right probably hasn’t changed much. I elsewhere in the discussion got some really good info about how there were exaggerated reports of already-extant refugee flows, and I want to learn more about that, because I probably have too little evidence to justify my sense that a lot of the political turmoil of the aughts and teens of this century related to climate-related famine.)

        My confidence interval, or whatever we call the range of true values that wouldn’t surprise me much, has grown quite a bit for refugees (especially right now, less so for whether the future is potentially pretty bad, and enough so to motivate action now), and somewhat for cost of sea level rise in the next 60ish years for coastal cities. (I also still think “existential” is not that likely to be wrong for a lot of cities. Cities come and go in history quite a bit, small but unchangeable advantages/disadvantages add up, because cities exist because of network effects interacting with small benefits.) The original 8 part list had been intended to stake out a strong position to get people to argue with me. I’m not saying Boston faces an existential risk. But if Boston is building a sea wall, then *some* set of cities are going to screw up their sea wall plans, or be geographically disadvantaged, or too poor, or otherwise get screwed enough that they go detroit because of climate change. I think probably one or more of those cities is in Florida, although I don’t know which one(s). (link).

        So, for my 8, I wasn’t 100% confident I was right, nor expecting to really persuade people who disagreed. I just didn’t understand where people who didn’t generally tend to agree with the cut of my jib, as it were, were coming from. And, probably this explains more why my priors haven’t changed that much–I still have a lot of work to do to work through some of the things people suggested. Actually, one of the clearest things I took from the whole discussion was that the explanation for the difference may relate to something like “semantic stop signs” (link), but more like semantic resting points. It’s like we’re all equally smart, but we’re working just hard enough to have something really, really smart to say about why [the act-on-climate crowd / the let’s-not-act-on-climate crowd / what-have-you] are where we currently stand, and once we find the smart thing to say, we take a rest. I’m in a busy period, April/May this year, and I admit I have several things I plan to look into more. But if I’m going to put as much effort into getting information that isn’t just intended to re-justify my existing priors, I need to continue to work to summon the effort needed for that before I jump into some of these issues. (Unless it’s adversarial collaboration, which if I understand right, is a structure where it’s fine if I’m advocating strongly for what I already think.)

        In case that stupidly long paragraph didn’t answer your question… I don’t think any of my original 8 were bogus. Doesn’t mean I’m right, just means that I haven’t got a basis to change my view on those yet.

        • Douglas Knight says:

          My original statement re: refugees, that it could potentially involve flows like 1/3rd of Bangladesh leaving Bangladesh in a given decade or things like that–those weren’t responded to.

          What could it possibly mean to respond to an unsourced claim? Glen put a lot of work into figuring what this claim even meant. How is that not a response?

          • Kelley Meck says:

            @Douglas Knight, thanks for the clear call out.

            I don’t like my phrase “weren’t responded to.” @Glen Raphael did a find job responding to my original point re: refugees, and I seem to have been a bit lax about acknowledging that. Simply replying “not responsive” makes me seem deaf or troll-like or just rude, which I regret. Plus, Glen was also one of the more thorough commenters on my long thread about the “consensus” so the least I’d like to manage is a thoughtful admission that I haven’t had time to follow links and think through things as I’d like.

            I’ve overrun my budget for comment time this week, already, but maybe sometime soon I’ll find time to formulate a cogent description of what my view of climate refugees as a problem was pre- and post-Glen’s comments, or marshall some links of my own.

        • Glen Raphael says:

          Bangladesh is growing. It is growing in physical size, in population, and also experiencing economic growth – as more people show up the GDP/capita keeps improving. Bangladesh had a really bad storm season in 2009 and so far as I can tell the doom-and-gloom stuff was largely based on the assumption that 2009’s experience would become typical for the area rather than the extreme outlier it actually was. (so, sort of like how An Inconvenient Truth used Katrina.)

          The first UN prediction was for 50 million climate refugees by 2010. When that failed to happen, the prediction was changed to 50 million by 2020. When you brought up the question here, I noted that 2020 is not far away and it doesn’t look at all likely for that second prediction to happen either, so naturally it kept growing and getting pushed farther out; the latest I can find is 200 million by 2050.

          I’d call this nonsense on stilts, but since being on stilts is a good way to avoid high water, I’ll just say it’s nonsense and leave it at that. 🙂

          I think probably one or more of those cities is in Florida, although I don’t know which one(s).

          The sea level rise situation you’re referring to there is a local phenomenon – climate change doesn’t explain it and that sort of rapid change (much faster than normal apparent sea level rise through either water “piling up” or land subsiding) would still happen in some locations even had we entirely stopped emitting CO2 30 years ago. Um, try here.

    • Glen Raphael says:

      As for the original proposal…

      I think you’ll have trouble finding anybody here who disagrees with your point #1.

      Regarding point #2, whether aggressive mitigation is a good idea mostly depends on (a) how much it costs, (b) how much benefit it produces. If you could put good numbers on both of those things and demonstrate some action passes a decent cost-benefit calculation (including, yes, both a reasonable discount rate and some amount of padding to account for political wastage), then you’d have something interesting. Alas, I’m not aware of any proposals that do that – are you? Last time around I believe you spoke of spending a trillion dollars without really saying what you planned to spend it on or how you knew doing so would produce more than a trillion dollars of benefit.

      [side note: at that time I pushed back a bit on acidification, but did so very late so you might not have seen it. In particular, I thought you might be overlooking papers like this that think global warming will on-net help coral and meta-analyses like this that argue the impact has been overstated.]

      I’d probably willing to argue against your point #3, but I don’t think I have enough to say on it to justify doing a serious dual paper – the topic seems more suitable for a brief discussion in the comments. Maybe the next Open Thread?

      • Kelley Meck says:

        You are right, I’d missed your push-back on how decreasing ocean pH is a problem. I’ll look at those.

        I think you are right that I haven’t thrown out the right topics for climate for an adversarial-type project. The truth is, I’m probably too busy this May to put the time in to do an adversarial-style go at climate anyway. I was hoping others would chime in, and I could be supportive of others taking on something climate-related. When I can, I’ll start another mini-topic in the open thread.

        Actually, I’ve been reading Kasparov’s “Winter is Coming” and I was thinking about writing about how much money some small set of Russian oligarchs must have on the line w/r/t/ the world *not* achieving Jim Hansen’s goals for climate action, and how that kind of guarantees that there’s likely upwards of a billion dollars of now-deployed dark money ready to attack any western institution, up to and including western elections or even governments, that seem to have their big-boy pants on w/r/t/ climate change to the point where they might actually succeed. (And that is irrespective of whether Hansen is right or wrong.) I guess at this point, that’s a teaser for somewhere down the road. That’s my #8–it’s probably the most tenuous of everything on my list. Anyway, probably I can’t marshall the time to do this well this month, so just as well to just be poking at it a bit here and there.

        • Glen Raphael says:

          @Kelley Meck:

          I think you are right that I haven’t thrown out the right topics for climate for an adversarial-type project.

          I asked about earlier exchanges because I have the sense your views on climate might be overdetermined. You have a conclusion (WE MUST ACT NOW ON CLIMATE) which you say is justified by a premise (OCEAN ACIDIFICATION IS REALLY BAD), right? So just as a thought experiment let’s suppose you and I investigated the premise carefully and at the end of investigation we determined to your satisfaction that ocean acidification isn’t such a big problem after all. Would that change your view about the conclusion? Which response from you is more likely in that circumstance:

          (a) “Gee, I guess I was wrong and ACTING NOW ON CLIMATE isn’t important after all!”


          (b) “Okay, I guess acidification is less important than I thought but I still have a long list of reasons why WE MUST ACT NOW ON CLIMATE so it doesn’t change my conclusion, it just lowers the priority of one specific argument for it.

          If the best-case response we could hope to get from you is (b) then acidification is just a placeholder. It’s not load-bearing; it’s not your crux. Maybe you started with 50 arguments and this was one that people here seemed least interested in making an effort to rebut so it feels stronger than others but that doesn’t mean rebutting it would change anything – you’re still holding the other 49 in reserve if this one fails. Right?

          If so, I’d recommend (when you have the time/energy) digging a bit deeper to see if you can find some set of premises that would matter to your conclusion if they were overturned.

          [Full disclosure: my own belief that ACTING NOW ON CLIMATE IS PROBABLY A BAD IDEA is also overdetermined. There’s at least a half-dozen premises supporting it, many of which boil down to “thinking more like an economist/engineer than an environmentalist/political activist”]

          • Axiomata says:

            @Glen Raphael @Kelley Meck

            If the deadline were August 1 instead of July 1, I would consider participating in a collaboration on the side of “We should act on climate now”. In fact I’m an active volunteer with Citizens’ Climate Lobby ( I have a well-read layperson’s understanding of climate science.

            As it is, I have a question I’m interested in for Glen Raphael–one which for me at least, is central to how I think about climate action. Do you have an idea (point estimate and/or confidence interval) of where you would place the social cost of carbon? I know official and academic estimates vary a lot–would you say your point estimate is close to zero? Or if not, are there other reasons you’d oppose the plan of experts coming up with a consensus estimate (as has in fact already been done during Obama years) and taxing CO2 equivalent emissions at that rate?

          • Glen Raphael says:

            @ Axiomata:

            Do you have an idea (point estimate and/or confidence interval) of where you would place the social cost of carbon? I know official and academic estimates vary a lot–would you say your point estimate is close to zero?

            I have low confidence as to any particular value but my suspicion is that the current net social value of CO2 emission is positive. (It’s possible it might become negative at some point in the future.)

            I do not claim to know what average planetary temperature is ideal for the flourishing of human civilization but it seems unlikely – on par with believing our world is the center of the universe – that the temperature in 1990 was precisely optimal and all warming since then has been unambiguously worse. A warmer world with more CO2 in it is a wetter world with more vegetation and more productive agriculture. Early IPCC reports strongly suggested the next 1-3 degrees of warming would make it easier to feed the world. Feeding the world more cheaply is good; I am in favor of that. Given that we can’t stop world temperature from changing, I’d rather it be changing warmer than changing colder. Ice ages frighten me; being further away from the next one seems like a win.

            People don’t experience the average world temperature. They experience the temperature in a particular place and pretty much every particular place we look, we find evidence of that place having been warmer than it is now at some point during the last couple thousand years. So we know that temperatures somewhat warmer than now are compatible with human flourishing.

            With regard to a carbon tax, I recommend this Cato Institute paper: The Case Against a US Carbon Tax

            are there other reasons you’d oppose the plan of experts coming up with a consensus estimate (as has in fact already been done during Obama years) and taxing CO2 equivalent emissions at that rate?

            When the OMB evaluates policies it uses a discount rate of 7% and compares costs to Americans with benefits to Americans. Among other errors, the Obama-era consensus estimate was based on a discount rate of 3% and used benefit estimates across the whole world rather than just our share of world GDP.

            Their argument for using a discount rate of 3% seems to have been that it was “the middle estimate”…but they picked the numbers to consider. So here’s my answer to that: let’s try a discount rate of 7% (because that’s what OMB always does) and also a discount rate of 10% (because that’s a Big Round Number close to average stock market returns since 1970). Now 7% is a middle estimate so we can justify using that…but we should also keep track of the 10% number as an outside possibility.

            (One reason for a high discount rate is that if we don’t ruin their economy today via bad climate policies, future generations are expected to be MUCH RICHER AND BETTER INFORMED AND MORE TECHNOLOGICALLY ADVANCED than we are which makes it easier for them to solve big technical problems in the future than it is for us to solve them right now. With a high enough discount rate, the social cost of carbon correctly becomes negligible-to-negative.)

            So if we had to do something like the Obama-era cost calculation, at a minimum I’d want them to apply a reasonable discount rate – at least 7%. And also update the sensitivity estimate range using more recent studies (it should be lower). But I’d rather just not do it at all. Near as I can tell, even granting most of their terrible assumptions the benefits of a carbon tax don’t exceed the costs. I’m afraid that if we put technocrats in charge of this what comes out of the sausage factory will be more stuff like the Biofuels mandate – policy that sounds plausibly sensible at the time but ends up making the world worse off and is hard to fix later due to path dependence.

            (see the Cato paper for other problems with the assumptions behind applying the tax once agreement has been reached. Incidentally, an issue they call “leakage” is understated in that essay: if any set of countries voluntarily uses less oil and coal that makes oil and coal cheaper so any countries that don’t join in will be incentivized by the price to use even more. We could hypothetically avoid this by instituting a massive One World Government that can enforce its iron will on every country on earth and force them ALL to institute the same tax/quota policies…but then you’re stuck with a massive worldwide totalitarian government which has its own problems that might be bigger than the problem you’re trying to fix. There’s also a fairly hefty principle/agent problem hiding in those weeds – even if countries like Mexico pretend to agree in principle, will they look the other way in practice? Does illicit oil smuggling become the next drug war?)

          • Glen Raphael says:

            @ Axiomata:
            UPDATE: The result of merely plugging a 7% discount rate into the FUND model (FUND was one of the models the Obama administration’s Working Group chose as reflecting academic consensus) is apparently to make the “Social Cost of Carbon” negative (meaning carbon emission is on-net helping humanity) at least through 2030. Source (second table on page 16).

            (The reason why the OBM recommends including 7% in all such reports is that this number is the average before-tax rate of return to private capital investment.)

          • Glen Raphael says:

            Could you drop me an email? I’d like to ask a couple questions about . Thanks! (I’m graphael at gmail dot com )

          • Kelley Meck says:

            @Glen Raphael re: overdetermined… Re As versus Bs… I’m an A. But see the Chesterton quote, “found difficult, and left untried.” As a younger person I read several books on this topic, with hunger and energy, and then argued with people who seemed like B (except on the ‘climate change isn’t real’ side) for a while… until (in some cases anyway) they were suddenly clearly As, because they’d changed their minds. I didn’t originally start comments on climate because I expected to change everyone’s mind–I just realized I’d gotten so far down the path I’m on that I couldn’t even explain how people I respect were so far removed from my own view.

            You are right that ocean acidification is one of a few places where I would have to change my mind before I’d think it makes sense for electeds or other institutional power-holders to be generally less activist in terms of building support for and taking action on mitigating climate change. Basically, I’m far enough to the “do more about climate now” side of things that I’d have to move pretty far to be calling for “do less about climate, for now.” That doesn’t mean I don’t move, just that it’s a big distance.

            Another run at some cruxes:

            * Ecologies we don’t understand are clearly affected in ways that are big but not understood, and that means there are big unknown risks and points of no return here (this includes ocean pH).
            * Stateless people have terrible outcomes (I don’t think very many people disagree with me here, but some hard line libertarians might).
            * Climate change is likely to mean many more stateless people. (I admit the flow of refugees now was not very central, as cruxes go.) If I become convinced that there will not be many climate-caused refugee flows or otherwise stateless people in the future (My current view puts a pretty high likelihood on climate-related major food insecurity creating state-failure and/or mass migration) that’ll be very close to becoming convinced that climate isn’t the huge human problem I currently think it is. I’m pretty far from that, at the moment.
            * Climate change is back-loaded. If I thought we were presently experiencing all or nearly all of the climate change we’ve committed ourselves to, that would probably go as far or further to relax my urgency for climate action than a definitive proof that the ocean’s ecology will not be negatively affected by atmospheric CO2 increases. Something I just thought of is to think of it as like an interest rate… if I thought we could just put the money we might want to use on climate change in a savings account at 5% interest, and spend it in 2050 if it turns out to be worthwhile, I think (interest notwithstanding) we’d find our money goes much, much less far in 2050 than it does today. Enough so that maybe we’d need a safe 20% annual interest or something silly-large like that before it’d make sense to take no action now. (I know some technologist arguments run the other direction, that action will be cheaper later, and I admit there is a good amount of uncertainty here… but uncertainty on this scale cuts against doing nothing)
            * Paid disinformation / rational ignorance are extremely serious overlapping problems, and ones that we can reasonably expect to be very heavily weighted in favor of the interests of mature industries and authoritarian states.
            * A consensus of climate scientists exists and is worthy of significant respect. I think career climate scientists are generally very risk-averse in claiming any particular finding and that any general consensus among climate scientists likely deserves more respect than health advice from a panel of five randomly chosen high-prestige MDs but less respect than the level of respect due to, say, a general consensus among rocket engineers about a well-studied type of rocket engine. In other words, climate science is a physical science an a tricky, more-political-than-average problem space… but it’s still a physical science, and less tricky/political than medicine. I have a big thread of comments on this crux, and don’t really want to re-open it now.

  50. Brian Bleakley says:

    I will debate/collaborate with anyone who thinks that a cure for death, as championed by Bostrom, is a good idea. My specific claim is that indefinite human lifespans would be a social and ecological disaster for humanity.

    • drunkfish says:

      I’m in no position to debate, but I’d love to hear you elaborate

    • yaolilylu says:

      I think you should make that in bold so others will find your topic more easily.

    • Adam Berman says:

      Per yaolilylu, bolding Brian Bleakley’s claim to encourage collaboration:

      indefinite human lifespans would be a social and ecological disaster for humanity.

      If you wish to discuss this, see parent.

    • Bugmaster says:

      I am not competent enough to formally debate on any topic, but can you elaborate ? As far as I can tell, the technology we’d need to potentially cure death is so far in the future, that by the time we get there we would no longer be recognizably human, anyway — so, what’s the point of debating it now ?

  51. inachodladh427986167 says:

    Scott, is “Cause/malleability of racial/demographic voting patterns” a taboo topic?

  52. JRM says:

    OK. Let me try a thing in my field of expertise:

    The legalization of heroin is a bad idea and a net negative to society.

    I might be persuaded to do a collaborative piece about marijuana, but the problem is that I don’t have as strong a conviction on that topic.

    Caveats: I can do one of these. My worktime for this would have to be primarily between June 6-June 23, though I’d be happy to do an hour or two a week before then. I have considerable legal expertise which may be helpful; my guess is that there are people here with interest in the topic. I am prepared to be wrong on this, but I do not think I am.

    Other potential topics are bail and “mass incarceration,” but I do not think they fit nicely into this framework. If you’ve got a thing that needs an SSC-style prosecutor, email me at my initials adding ayne to the end with a gmail address.


    • Alex Zavoluk says:

      I’ll be e-mailing you when I get back to my personal computer.

    • yodelyak says:

      I want to know what you think about bail and mass incarceration, even if you don’t do an adversarial collaboration.

      • toBoot says:

        Me too!

      • JRM says:

        VERY short version:

        1. The increase in incarceration rates both in the 1980’s and the 2000’s led to lower crime. However, the system that increased sentences for the really bad people also brought in substantially less dangerous folks. Competence uber alles; if we argue “increased incarceration is good,” vs. “increased incarceration is bad,” we’re missing the point.

        2. Bail: California in a case called In re Humphrey changed the bail system judicially in a manner spectacularly divorced from the facts and the court came just short of ending the opinion with a picture of an upraised middle finger. It said that there would have to be lots more resources spent on bail, but if you don’t want to give the judges more money, that’s on us. (It less explicitly discusses the burdens on other levels of government.)

        That doesn’t mean that bail in California (where I practice) is working. The first and most plainly obvious fix is to have bail companies be able to charge whatever they want to post bail. Right now, if I have a $50K bail, the bond company has to charge $5K. Maybe I’m not a risk of flight – let the market sort it out. There are states where this happens, and those states provide all kinds of interesting data – the market can actually help solve this.

        Unfortunately, this doesn’t fit on the “keep everyone in,” vs. “let everyone out,” metric. Also, there’s no market to compete. (While I am here ranting about market: Jail phone services are wildly overpriced and scammy and that should stop.)

        The second fix is to have pretrial services for those released. If you have very low-level offenders who can’t make nominal bail, the solution may be to let them out and monitor them. Yeah, that homeless guy is a risk not to return, but there may be other solutions. Our local public defender’s office just hired a social worker, and this is a really good idea.

        Caveats: I am probably more expert than you are on bail, but I am not a true bail expert. I have read several studies on bail, and most of them are of almost unbelievably poor. I have a good grasp on California bail and a few other jurisdictions which I have taken an interest in, but I don’t know how every jurisdiction works and there may be some near-perfect model somewhere I am unaware of. Because most studies are awful, it is much easier to accept conclusions I like and reject those I don’t; I think I am mostly immune to that, but that’s what a non-immune person would also think.

        SSC’ers are my peeps, so if you have more questions, hit me up on an open thread and email me to make sure I see it. Man, this wasn’t short. Gotta get back to running for office and fixing a broken system one county at a time.

  53. allyscully says:

    I’d be interested in doing an adversarial collaboration on cybersecurity (e.g., corporate spending on cybersecurity is/is not directly correlated with a decreased risk of being breached) or olfaction (e.g., olfactory receptors expressed in tissues outside of the olfactory epithelium are/are not functional). For the former, I lean toward “is not”, and for the latter, I lean toward “are”. I’m open to other questions in these areas, too.

    • (tongue-in-cheek) I believe the politically correct term for ‘cybersecurity’ these days is ‘infosec’.

      (Although as far as I can tell, it really is true the cyber- prefix annoys a lot of people in the field because it’s historically been such a source of confusion about what the field even does. To be clear about it: I am not one of those people.)

      I’d be interested in seeing this adversarial collaboration, since I am in the field, and can’t reconcile your position with my practical observations, but also know that my experience so far is limited to a handful of companies, without much insight into practical effects.

      Specifically, I have some (thankfully slim, but existent) practical experience with breach reactions from management. The ones I’ve witnessed seem to always serve as a wake-up call and make them start spending money on security to, bluntly, patch up gaping holes in their infrastructure, such as… well, even just using TLS everywhere on the local network, which apparently isn’t done in the wild unless you have a security-minded professional to point out how unfathomly stupid what a poor choice it is not to encrypt your traffic.

      This is a short-term benefit, but it seems this would possibly already invalidate the claim as stated. If you agree that this benefit happens (you might not! e.g. if you find evidence that these security projects are rarely completed, and the attack surface is only negligibly decreased, but attack vectors remain the same, then the risk of being breached may not necessarily have decreased), perhaps a different hypothesis to test would be:

      1) “corporate spending on cybersecurity does not continually decrease the risk of being breached, but confers at most a ‘one-time benefit’ over a period of one year”


      2) “the cost of corporate spending on cybersecurity is inefficiently allocated – average breach damage is not reduced by even 50% of security spending”?

      I plugged in an arbitrary percentage for the second claim, in the off-chance you want to use that one, I recommend picking a percentage you feel is correct.

      Either way, I would love to see someone take you up on this topic and very interested in the results! Can see this going either way and would be interested in the result, and in the case of negative outcome, the causes of the result (e.g. I could picture a finding like “spending on security does not perceptibly change the corporate culture” be a big issue in actually producing security gains; or “established security teams on average are too invested in looking good over doing good due to the high social status security knowledge conveys” (though I hope not! XD); or even just “increased security spending attracts proportionally more interest from criminals, resulting in a net negative effect” et cetera).

      I don’t have time, unfortunately (nor am I good opponent – I almost pathologically avoid conflict, I would almost certainly err on the side of just agreeing with my opponent). Just wanted to mention my interest. <3

      If you end up doing either: Good luck!

      • Brad says:

        (Although as far as I can tell, it really is true the cyber- prefix annoys a lot of people in the field because it’s historically been such a source of confusion about what the field even does. To be clear about it: I am not one of those people.)

        My impression is that “cyber-” connotes the government, especially the military, and government contractors, especially military contractors. And downstream of that some of the very large non-tech companies that like to hire from those organizations.

        Whereas the other side of the industry runs as fast and as far away from the prefix as it can.

    • warrel says:

      Just curious, are you taking a Robin Hanson-like stance that most cybersecurity efforts are about “signalling caring” rather than actually preventing breaches/damage?

      • allyscully says:

        Yes, I think so. I’m not sure I’ve read Robin Hanson’s writing on this, but signalling is a good way to describe it. I agree that companies will need to spend more after a breach to fulfill their damage control/breach response obligations, but I think in companies that have not been breached, higher than average spending on security indicates a larger attack service (through more products, perhaps more consultants and third-party vendors) and fear or uncertainty over their security (which probably means they do have something to worry about)

  54. ChelOfTheSea says:

    I’d be very interested in taking the pro-trans side on the question of “is transition beneficial for trans people and otherwise non-harmful such that it is the clear and unambiguously proper treatment?”

    • ohwhatisthis? says:

      I believe I can argue against that. Though its going to be a subgroup based off of specific traits and an argument against “clear and unambiguously”

  55. tailcalled says:

    If this works out well, will you consider doing it regularly? I could imagine that one limiting factor in the creation of adversarial collaborations is that while many people want to have their view shared, most people don’t tend to have a platform to share it on, and so having an influential person present good ones creates an incentive to create adversarial collaborations.

  56. ZachJacobi says:

    I’m not sure if I’ll find anyone here to take the opposite, but I would be interested in collaborating something along the lines of:

    “Common sense conservatism” is uniquely bad compared to other major mainstream political ideologies (social democracy, neoliberalism, economic conservatism, centrism, libertarianism) and is likely to lead to much lower capacity for collective action and less ability to respond to crises.

    I’d be taking the “Is bad” side.

    • Patten says:

      I’m interested, but can you define common sense conservativism? My Google search made it look like it was Trump’s way of adding on positions that were popular with his base but went against conservative ideology. It’s hard to even define it as an ideology, if we can’t identify a central theme.

    • Don_Flamingo says:

      It seems to me, you’re equating ‘good’ with better at taking collective action and responding to crises.

      A collective action however can be homicidal (e.g. genocide, witch burnings, struggle sessions, intellectual purges) and a high ability to respond to crises also means the ability to respond energetically to something that isn’t a crisis at all or badly overreacting to a real one (e.g. satanic kindergartens, rainbow parties, reefer madness, post 9/11 security [maybe this is controversial, but the way, I see it: of all the security implemented in the aftermath, the only thing that was even remotely worthwhile was to lock all cockpits with bulletproof steel doors, which could have been done with only couple of 100 Million Dollars spent in total;;; ignore that point, if (you think) I’m totally wrong about this]).
      Aren’t most forms of Libertarianism/conservatism/neoliberalism skeptical about this definition of good and wanting to explicitly reduce this capacity and ability (so how can it be in this definition not be worse than whatever CSC is)?

      I’m not interested in doing a collab (don’t have the stat skills yet to produce or even fully comprehend clever SSC-types of analyses), but could you elaborate on your definitions of good and bad here (maybe, that’s already something that needs a collab all of it’s own to hash out)?

      • Glen Raphael says:

        Right, if we think of “capacity for collective action” as a bazooka, the question is how optimistic we are that those directing it are likely to aim in the right direction, fire only when necessary and minimize harm, cost, and blowback? If the answer is “not very”, we might not want to maximize the bazooka.

        Regarding “ability to address crises” there’s also a bad-drives-out-good thing whereby relying on government too much for too long damages or destroys self reliance by weakening all the private institutions we might otherwise use to address the same problems. Our collective ability to respond can become fragile when solutions need to go though a single government bureaucracy instead of a bottom-up private discovery process.

        (I can’t take the adversarial position on this one because I have no idea what “common sense conservativism” is – I’m more libertarian-ish.)

    • RalMirrorAd says:

      Common sense conservatism has the same problems as debating any ‘ism’ — it’s too broad.

      I could just as easily say that neoliberalism is uniquely bad because the deterioration of social capital prevents collective action and lowers the ability to respond to crises. Such a debate would come down entirely to definitions and ability [or lackthereof] to prove attribution.

  57. jkranak says:

    I’d be interested in taking on philosophical topics. I’d like to take the position of opposition to moral realism. It would probably be good to focus just on something like whether moral values (good, bad, neutral) are real attributes of the world rather than questions about moral rules or deontic categories, just for the sake of not taking on too much.

    I should also say I’m also interested in the free will debate. I personally support libertarian incompatibilism, so it’d be good to hear from a compatibilist.

    • yodelyak says:

      I am interested to see how this goes for you, if you find a partner. My prior is that I don’t think an adversarial collaboration will be that productive. My experience has been that people who are really careful about thinking about how problems of language can corrupt this debate reach “agree to disagree” very quickly, or else claim that there may not even be any “real” difference whatever between even the most wildly different-sounding positions about whether free will “exists.” Meanwhile people who are less careful find this to be an area where huge moral judgments about individuals in society and agency in history and etc., turn on agreeing that freewill “exists” and that the future is “indeterminate.”

      • jkranak says:

        There definitely a bit and seemingly intractable problem around what counts as “free will.” If that’s what you’re referring to, I think you’re right that it may be impossible to make progress. But I’m more interested in the question of whether we have the ability to do otherwise. As far as I know, the terms of that debate or clearer.

        • Philosophisticat says:

          The two debates are tied together. Philosophers identify many different notions of “ability to do otherwise” some of which are compatible with determinism and some of which are not. The debate, then, is over in which sense we need to be able to do otherwise for us to have free will (which is what everyone cares about).

    • Philosophisticat says:

      I don’t think I’ll be up for a collaboration, but I am happy to talk about both of these issues.

  58. pjiq says:


    This is a VERY fun idea, and I agree that adversarial research should be expanded to political issues more. However, I kind of want to submit something in a different style.

    So I know this wouldn’t be eligible, but an adversarial project where you work together just seems kinda, I dunno, not that adversarial? If I sent you something in a slightly different format- more of an evidence-focused debate- would you read it? Something like:

    A argues this with this many words
    B responds with this
    A responds with this
    B concludes.

    I think that would be more fun to read and feel less like I was being brainwashed into *thinking* that the TRUTH has been found through conflict, where really it’s just two people pretending to argue from different sides just to emphasize the “absolute facts” they feel like emphasizing. Not that I’m part of the post-truth thing, and I hope I’m not being too snarky here against the greatest blog writer the internet has ever known- I’m just, ya know, taking an adversarial position on your approach to adversarial research submissions ;)-



  59. pjiq says:

    Also I’ll argue with any libertarian on here against any libertarian thing.

    (respectfully, using facts, historical examples, policy results, etc. I love libertarians, I just have some different opinions/ contradicting theories)

    • LadyJane says:

      Drug prohibition causes vastly more harm than good.

      • pjiq says:

        any drug in particular? Or all illicit drugs, regardless? If the second, I’ll definitely do some research and debate that with you.

        • LadyJane says:

          Prohibition in general. Proving that cannabis prohibition causes more harm than good would be too easy. 🙂

  60. oli cairns says:

    I have an economics background. Would anyone be interested in working on the following topics?

    The socially optimal rate of corporation tax in America. I think it’s likeky to be lower than the current levels, and that Trump’s recent tax reform is likely to be a step in the right direction.

    The regulation of tech firms. I think the scrutiny of tech firms by competition authorities, and proposed additional data/privacy regulations are likely to lead to net harm.

    The short term rate of technological progress. I think living standards in developed economies have progressed less in the past 10 years than any non-war period in modern history. I think that this is likely to continue for the next 10 years.

    • ohwhatisthis? says:

      The short term rate of technological progress. I think living standards in developed economies have progressed less in the past 10 years than any non-war period in modern history. I think that this is likely to continue for the next 10 years.

      I can argue against this one. Note, that this may involve different definitions of some terms, but I think I can make a case you’re not thinking about it quite properly.

    • Jiro says:

      I am unaware that there was a non-war period in human history.

  61. Callum G says:

    Perhaps these would be better if the issues were phrased moot style e.g. This House Would Provide All Police Officers With Firearms with one side supporting the motion and the other opposing it. I’m worried we will get intelligently written essays that aren’t adversarial because the pair didn’t establish a common framework for debate.

  62. weaselword says:

    I am interested in charter schools, and I am willing to support the following positions:

    1. Charter schools have a positive effect on overall educational outcomes within school district.

    2. Charter schools have a positive effect on educational outcomes within the traditional (that is, non-charter) public schools in the school district.

    I have some skin in the game on this issue, so this is a topic I genuinely want to learn more about. I also know that there are reasonable positions that take a negative stance towards charter schools, and I would like to correct my pro-charter bias by learning more about those.

    If anyone is interested, my handle on Reddit is the same, or email me at a1v1a1r1v1a1k (remove the numerals) at

  63. toBoot says:

    I’m interested in engaging in an adversarial collaboration about effective altruism. I would take a somewhat anti-effective altruism stance:
    (1) The goal of effective altruism is to reduce unnecessary suffering, and not the kind of suffering that is a necessary part of growth. Too much context is required to predict with any accuracy what kind of suffering is a necessary component of growth (both personal and sociological), and what kind of suffering is unnecessary. This makes scalability of charity based on suffering focused ethics really difficult, and increases the likelihood that relieving suffering in a familiar/local context or relieving suffering that may be mundane but is highly unlikely to lead to growth (e.g. tedium) is just as/more beneficial than the usual approaches advocated by effective altruists. Counter-argument: even though the likelihood that you are relieving unnecessary suffering per charitable act decreases as you increase in scale, the increase in number of charitable acts increases so dramatically that overall, unnecessary suffering will still be reduced.
    (2) People are more likely to engage charity if they get positive emotional feedback from it. That kind of positive feedback doesn’t come at a distance. I predict that people who engage in feel-good charity are more likely to engage in more charity over a lifetime, such that it compensates for any gap in suffering reduction left over from the counterargument to my first point.
    I’m not sure if I have enough expertise to do this topic justice. It’s also got some really difficult empirical questions with a lot of uncertainty, and my data analysis skills are not up to the task of dealing with massive uncertainties. So if folks wanna just school me up here (or send me readings) rather than engaging in a collaboration, that’d be welcome too!

    • rathdane says:

      I would like to see something similar on the quality of life while lived vs. total quality of life, or the link or lack of one between altruism and immortality. Some proponents of effective altruism want to end the cycle of life and death, while others want to address overpopulation and healthy renewal. A meaningful collaboration across that divide would be a revelation.

    • maintain says:

      One thing that immediately comes to mind: You’re talking about the effective altruism movement. They believe in hardcore autistic debate on every topic. If you throw a criticism at them, and your criticism turns out to be valid, they should be able to take that criticism into account, and modify their process accordingly.

      You’re not going to be able to go up to EA and be like “Oh yeah, well you didn’t take X into account! Checkmate!”

  64. warrel says:

    suffering that may be mundane but is highly unlikely to lead to growth (e.g. tedium)

    Except that Buddhist/etc. meditation turns tedium into a growth-opportunity..

  65. mtl1882 says:

    I’m trained as a lawyer and like this sort of thing, but lately have trouble narrowing my focus onto a topic. I am not particularly attached to certain political positions, and I love to research. But a good idea isn’t really coming to mind – I have trouble seeing things as two-sided versus very multidimensional in most cases. I’ve been really into Civil War/Victorian era research, especially about the press at that time. Not sure if anybody has any ideas based on that. One thing of interest to me is the idea of whether we should judge people by the norms of their time. That used to seem intuitive to me, but now it doesn’t. I read old newspapers and books all the time and it’s very clear that for most major controversies, a sizable minority was very aware that a moral and logical issue existed and discussed it intelligently. At every point, some people educated their daughters and trained them in professions (I do give people a pass for thinking that women have a generally home-based role by nature, because biology pre-birth control dictated that). At every point, some people agitated against slavery. At every point, some people question dubious medical knowledge and religious knowledge. That doesn’t mean we demonize people who went along with norms, but people who had the ability and courage to question should get credit – it shouldn’t be shrugged off as something that “obviously” couldn’t happen back then. And brilliant people who did question but then kind of dodged the issue should get a side eye. Thomas Jefferson agonized over racial differences and slavery and if he had been wrong about his beliefs, and was attacked for his relationship with Sally Hemings at the time – that’s not a modern SJW controversy. That’s another thing I think about – the idea that we only suddenly inserted minorities into history or that there were no women equal in significance to some of the men who were written about, or that criticism of the historical heroes is new. There were women and other minorities playing significant roles who were simply written out of history and are merely being recovered, and many of the modern debates about our heroes happened at the time. The press and controversies are not that different than now. Old newspapers are quite an interesting read. These sort of historical issues are things I’d like to explore further.

    I also work in the test prep industry, which raises some interesting issues.

  66. Axiomatic Doubts says:

    The topics I’d be interested in making an adversarial collaboration about:

    -Effect of macronutrient intake as a ratio of total caloric intake in weight gain or loss

    -Effects of higher education in society

    -Factors that most influence present-day human well-being

    -Utilitarian value of modernity

    -Legal restrictions on suicide

    Email me at

    • maintain says:

      Which sides do you take on those issues? It’s pretty important to know that beforehand.

      • Axiomatic Doubts says:

        My side on those topics:

        – Ratio of fat to carbohydrate intake does not affect weight gain or loss
        – Insulin does not predict and/or cause weight gain or loss

        – Higher education is beneficial not only for individuals but also for society at large, by encouraging a signaling arms race and higher parental investment in children

        – Psychological/neurological/organic factors (and by extension, sociometric status) affect present-day human well-being more than relative wealth (aka socioeconomic status)

        – Modernity is good from a utilitarian point of view

        – People should not be involuntarily committed for suicidal ideation, and physician-assisted suicide should be legal (where it isn’t)/more accessible (where it is)

        • maintain says:

          > Higher education is beneficial not only for individuals but also for society at large, by encouraging a signaling arms race and higher parental investment in children

          I’m somewhat skeptical of this. I read Caplan’s book, where he said education is 20% learning, and 80% signaling, and I took it to mean that education really is 80% waste.

          Well, I guess signaling helps employers determine which potential employees are competent, and our economy would be less efficient without that, but I still tend to think of any signaling arms race as inherently being a waste.

          You can email me (rot13) ulqebtraohvyqvat at gmail.

  67. Callum G says:

    SlateStarCodex comments section is not arranged as best as possible

  68. Patten says:

    Since I haven’t gotten a response on my queries on other people’s conversation topic, I will propose my own. What level of occupational licencing is beneficial to society? My own position is the only profession that should be subject to occupational licencing is MDs. We can refine the subject as needed if you are interested.

    • ksvanhorn says:

      You should check out Milton Friedman’s arguments against licensing doctors. He wrote that “I am myself persuaded that licensure has reduced both the quantity and quality of medical practice.”

      • Patten says:

        I’ve read several compelling arguments against medical licensure, but I remain unconvinced. If someone wants to argue against medical licensure, I can argue the pro side?

  69. Prussian says:

    I’d like to do one on:

    Is Islam a credible threat to modernity / civilisation?

    If that’s not taboo

  70. Joyously says:

    I’m torn as to whether I have the time/energy/dedication to do this right, but I would be interested in arguing:

    Restrictions on abortion reduce abortions more effectively than birth control subsidies.

  71. Aron Wall says:

    Here’s a suggestion at the intersection of religion and medicine. If I weren’t so busy with other things, I would propose doing an adversarial collaboration on Some modern-day religious healing claims are very difficult to explain by ordinary naturalistic causes (e.g. chance, fraud, mistake, placebo, etc.). Maybe one of my fellow Christians who reads SSC would be willing to take this side?

    If so, I highly suggest not phrasing the dispute as anything like “Do miracles happen?”, since that would inevitably get bogged down in evaluating the philosophical priors and I would prefer seeing an attempt to agree on how to evaluate the likelihood of these events given Naturalism.

    Suggested resources to get started:

    Pro: Miracles by Craig Keener (2 volumes)
    [Ignore the misleading subtitle, this scholarly tome is almost exclusively about modern day miracles (mostly healings) and has a gazillion references in the back if you want to follow up on any of the cases he mentions.]

    Con: The Faith Healers by James Randi
    [I haven’t actually read this book yet, I know about it because it’s mentioned in the Keener book—I assume because of the author that the investigations were competently done, but I think it’s more popular and limited in scope than the Keener book, so not an ideal counterweight to Keener,, just a place to begin.]

  72. MugaSofer says:

    I’m interested in collaborating on effective abortion reduction. Specifically, I believe that the most effective methods are laws restricting abortion, distribution of and education regarding contraceptives, and improving social conditions for mothers.

    If we can get a person who believes legalization is the best path to decreasing abortion from the left, and a person who believes contraceptives increase abortion rates from the right (both common claims), might be able to make it a three-way collaboration.

    • albatross11 says:

      Your phrasing seems ambiguous there. Do you mean:

      Possibility one is all of:
      a. Laws restricting abortion
      b. Distribution of contraceptives
      c. Education regarding contraceptives
      d. Improving social conditions for mothers

      Possibility two is all of:
      a. Laws restricting abortion
      b. Laws restricting distribution of contraceptives
      c. Laws restricting education regarding contraceptives
      d. Improving social conditions for mothers (or maybe laws restricting improving social conditions for mothers?)

  73. yaolilylu says:

    I’m interested in arguing that the benefits of breastfeeding are minor when controlling for confounders. I know someone posted relating to this earlier but I’m looking for someone to argue the pro-breastfeeding position.

  74. Swami says:

    Would anybody be interested in the topic of US Median Income Stagnation? I would argue that there has been no median income stagnation in the US for the past forty years. Thus, I would need someone to argue the side that there has been median income stagnation.

    Alternatively, we could be more specific, for example, whether median incomes have increased more or less than ten percent or such (effectively considering “stagnation” to be anything less than 10%).

    For the record, I am not an economist, nor do I play one on TV. I just think it is an interesting topic.

    • thad says:

      Also not an economist, but I might be willing to do this thaddeus reich @ g mail

  75. maintain says:

    I want to do this

    -People under 18 should have more rights, including, possibly, the right to choose their guardians, the right to choose their schools, and the right to choose not to go to school at all.

    -People stop existing when they die.

    -The noporn movement is not a bunch of crackpots: Internet porn can be bad for you.

    • aristides says:

      I would like to argue for porn. I was planning on researching this anyways, so it’ll be even better in this format. To clarify, can we argue Whether porn is good for society or the average consumer? I don’t want to argue that porn can never be bad for anyone, when all you need is one example to prove me wrong. If you have a better way to phrase it let me know.

      • maintain says:

        Yeah, it’s difficult to come up with an actual phrasing. You don’t want to argue that porn can never be bad for anyone, and I wouldn’t want to argue that porn can never be good for anyone.

        If we phrased it as is porn good for the average consumer, we could end up, for instance, finding that porn literally kills 1% of the population, but leaves the rest better off, and thus we would conclude that porn is good for the average consumer.

        Maybe we can figure something out.

        You can email me (rot13) ulqebtraohvyqvat at gmail.

    • ohwhatisthis? says:

      -People stop existing when they die.

      I’ll argue against this. Email?

      • maintain says:

        I already agreed to collaborate with aristides. You can email me anyway. Maybe we can collaborate on something outside of the contest after I’m done.

        (rot13) ulqebtraohvyqvat at gmail.

  76. jericho says:

    Scott, I love the general concept of this. Some of the topics being proposed seem quite esoteric. Wouldn’t it be better to take up the most important issues of our time? It might work better if you chose valuable topics and then invited people to “apply” to represent one side or the other, by sharing their starting opinions and qualifications to hold those opinions.

  77. grendelkhan says:

    I have an interest in housing and housing markets. I’d like to test something like the basic YIMBY thesis–that building market-rate housing in California is very likely to reduce the cost of rent there. I’d especially be open to trying to explore the various tradeoffs in property tax revenue, displacement, home equity, and so on.

  78. Michael Crone says:

    I would be interested in an article about abortion where I am the collaborator who believes abortion should be illegal. I think a collaboration would be a good project on its own merits, but I’m not interested in the contest because (1) I want to be able to be slow to get work done (see the date on this comment) and (2) Scott may view it as taboo anyway.

    If interested, I have a Gmail account with the handle drmichaelcrone. That should get my attention, while responses to this comment probably won’t.

  79. J Mann says:

    I’ll post on the latest open thread, but if someone wants to take a different position, I’d love to do some variation of:

    The allied intervention in the Libyan civil war was based on unjustified assertions of an impending massacre in Benghazi, and was illegal under international law. Whether through negligence or intention, allied leaders misled the public about the imminence of the threat and whether the mission was aimed at toppling Qaddafi and winning the civil war for the rebels.

    Note – by “allied,” I’m intending focus mostly on Britain and the US, because I don’t speak French, but I’m open to expanding the scope.