Scott Aaronson suggests that Many-Worlds and libertarianism are similar in that they are both cases of bullet-swallowing, rather than bullet-dodging:

    Libertarianism and MWI are both are grand philosophical theories that start from premises that almost all educated people accept (quantum mechanics in the one case, Econ 101 in the other), and claim to reach conclusions that most educated people reject, or are at least puzzled by (the existence of parallel universes / the desirability of eliminating fire departments).

    Now there's an analogy that would never have occurred to me.

    I've previously argued that Science rejects Many-Worlds but Bayes accepts it.  (Here, "Science" is capitalized because we are talking about the idealized form of Science, not just the actual social process of science.)

    It furthermore seems to me that there is a deep analogy between (small-'l') libertarianism and Science:

    1. Both are based on a pragmatic distrust of reasonable-sounding arguments.
    2. Both try to build systems that are more trustworthy than the people in them.
    3. Both accept that people are flawed, and try to harness their flaws to power the system.

    The core argument for libertarianism is historically motivated distrust of lovely theories of "How much better society would be, if we just made a rule that said XYZ."  If that sort of trick actually worked, then more regulations would correlate to higher economic growth as society moved from local to global optima.  But when some person or interest group gets enough power to start doing everything they think is a good idea, history says that what actually happens is Revolutionary France or Soviet Russia.

    The plans that in lovely theory should have made everyone happy ever after, don't have the results predicted by reasonable-sounding arguments.  And power corrupts, and attracts the corrupt.

    So you regulate as little as possible, because you can't trust the lovely theories and you can't trust the people who implement them.

    You don't shake your finger at people for being selfish.  You try to build an efficient system of production out of selfish participants, by requiring transactions to be voluntary.  So people are forced to play positive-sum games, because that's how they get the other party to sign the contract.  With violence restrained and contracts enforced, individual selfishness can power a globally productive system.

    Of course none of this works quite so well in practice as in theory, and I'm not going to go into market failures, commons problems, etc.  The core argument for libertarianism is not that libertarianism would work in a perfect world, but that it degrades gracefully into real life.  Or rather, degrades less awkwardly than any other known economic principle.  (People who see Libertarianism as the perfect solution for perfect people, strike me as kinda missing the point of the "pragmatic distrust" thing.)

    Science first came to know itself as a rebellion against trusting the word of Aristotle. If the people of that revolution had merely said, "Let us trust ourselves, not Aristotle!" they would have flashed and faded like the French Revolution.

    But the Scientific Revolution lasted because—like the American Revolution—the architects propounded a stranger philosophy:  "Let us trust no one!  Not even ourselves!"

    In the beginning came the idea that we can't just toss out Aristotle's armchair reasoning and replace it with different armchair reasoning.  We need to talk to Nature, and actually listen to what It says in reply.  This, itself, was a stroke of genius.

    But then came the challenge of implementation. People are stubborn, and may not want to accept the verdict of experiment.  Shall we shake a disapproving finger at them, and say "Naughty"?

    No; we assume and accept that each individual scientist may be crazily attached to their personal theories.  Nor do we assume that anyone can be trained out of this tendency—we don't try to choose Eminent Judges who are supposed to be impartial.

    Instead, we try to harness the individual scientist's stubborn desire to prove their personal theory, by saying:  "Make a new experimental prediction, and do the experiment.  If you're right, and the experiment is replicated, you win."  So long as scientists believe this is true, they have a motive to do experiments that can falsify their own theories.  Only by accepting the possibility of defeat is it possible to win.  And any great claim will require replication; this gives scientists a motive to be honest, on pain of great embarrassment.

    And so the stubbornness of individual scientists is harnessed to produce a steady stream of knowledge at the group level.  The System is somewhat more trustworthy than its parts.

    Libertarianism secretly relies on most individuals being prosocial enough to tip at a restaurant they won't ever visit again.  An economy of genuinely selfish human-level agents would implode.  Similarly, Science relies on most scientists not committing sins so egregious that they can't rationalize them away.

    To the extent that scientists believe they can promote their theories by playing academic politics—or game the statistical methods to potentially win without a chance of losing—or to the extent that nobody bothers to replicate claims—science degrades in effectiveness.  But it degrades gracefully, as such things go.

    The part where the successful predictions belong to the theory and theorists who originally made them, and cannot just be stolen by a theory that comes along later—without a novel experimental prediction—is an important feature of this social process.

    The final upshot is that Science is not easily reconciled with probability theory.  If you do a probability-theoretic calculation correctly, you're going to get the rational answer.  Science doesn't trust your rationality, and it doesn't rely on your ability to use probability theory as the arbiter of truth.  It wants you to set up a definitive experiment.

    Regarding Science as a mere approximation to some probability-theoretic ideal of rationality... would certainly seem to be rational.  There seems to be an extremely reasonable-sounding argument that Bayes's Theorem is the hidden structure that explains why Science works.  But to subordinate Science to the grand schema of Bayesianism, and let Bayesianism come in and override Science's verdict when that seems appropriate, is not a trivial step!

    Science is built around the assumption that you're too stupid and self-deceiving to just use Solomonoff induction.  After all, if it was that simple, we wouldn't need a social process of science... right?

    So, are you going to believe in faster-than-light quantum "collapse" fairies after all?  Or do you think you're smarter than that?

    New Comment
    136 comments, sorted by Click to highlight new comments since:
    Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

    "Libertarianism secretly relies on most individuals being prosocial enough to tip at a restaurant they won't ever visit again. An economy of genuinely selfish human-level agents would implode."


    A thoughtful article about this and related issues is:

    "Even if you want no state, or a minimal state, then you have to argue point by point. Especially since the minimalists want to keep the economic and police system that keeps them privileged. That’s libertarians for you — anarchists who want police protection from their slaves. No! If you want to make the minimum-state case, you have to argue it from the ground up." - Kim Stanley Robinson,

    "Or do you think you're smarter than that?"


    ZMD, is that the answer you think I'm looking for, your personal answer, or the correct answer?

    Eliezer, why are you concerned with untestable questions?

    Statistically it would seem unlikely that I am anything near approaching "rational".

    Gray Area: Eliezer, why are you concerned with untestable questions?

    Questions you can easily test experimentally are hard for Science to get wrong.

    There are numerous questions that are hard to test experimentally right this minute but are extremely important because of their future consequences. I bet you can think of one or two.

    I chose quantum physics as my point of departure because the case is mathematically clear-cut.

    Incidentally, it looks to me like you should be able to test macroscopic decoherence. Eventually. You just need nanotechnological precision, very low temperatures, and perhaps a clear area of interstellar (intergalactic?) space.

    Has anyone tried to actually DO Solomonoff induction against the real world? If I understand, it's incomputable, and even the idea of encoding reality as a program... well, it would be a very big program. So except as a pointer to and clarification of Occam's razor, does it have a real world use?

    Generalizing: is it actually possible to use pure Bayesianism in any non-contrived, non-trivial context? And if purity can't be attained, is there an optimal impure approximation?

    Sure. The difficulty of actually doing Solomonoff induction is irrelevant, because SI isn't actually part of Bayesianism as everyone in the world except Eliezer Yudowsky defines it. Cox's Theorem gives us the basic laws of probability, but there is nothing comparable telling us that algorithmic probability is the correct prior we should be using. A prior is an encoding of one's prior knowledge / state of information before seeing the experimental data, and we have no a priori reason to expect simple explanations for everything.
    I seem to recall Eliezer writing a post on this - and did not seem to disagree with the above passage.
    But we could establish an a posteriori one

    I always tip because it's the convention and I'm afraid of being confronted about not doing it. If I didn't have that fear I most certainly would tip far less.

    You should drop the stuff about selfishness. It's self-interest. Everyone agrees if you define "selfish" as not doing anything that could possibly aid anyone else then human society wouldn't work. Clearly you'd live in a cave and have a subsistence existence until you're dead at 15. I don't know anyone who actually advocates that. So the definitions must be off.

    "That’s libertarians for you — anarchists who want police protection from their slaves. No! If you want to make the minimum-state case, you have to argue it from the ground up." - Kim Stanley Robinson,

    There is a problem here that you do not seem to recognize. If any meta-level approach is better, i.e., will yield a more correct model of the universe than the current scientific method, then the scientific method will, over time, devour it, make it a part of itself. This is, because in the end, the "better" alternative approach will at some point yield a theory, no matter how small, but still, perceptibly better prediction. It may not do so for QM - it may yield only a different "interpretation", but it will, somewhere along the line, ... (read more)

    Or do you think you're smarter than that?

    Yes I do, since you're asking. Is that correct? Probably not. But that doesn't bother me. I have to trust my rationality before I can get anything out of science, right? As far as I'm concerned, my own, personal, internal logic is The King.

    Still not convinced that Science (capital S) wants me to believe in anything it can't provide evidence for though. Logic and induction might postulate certain unprovable beliefs, but there's no reason why Science should flat out disagree with them. Still not feeling the dilemma.

    A lot of people have a problem with Kolmogorov complexity and Solomonoff induction being "ideals". Sure, you can't build a working perfect compressor in order to compute the Kolmogorov complexity of a binary string. The best you can do is to approximate it. Furthermore, the ways in which your compressor fails to achieve the perfect compression of Kolmogorov complexity are weaknesses of your compressor that a more powerful compressor could overcome... and so on and so on. It's only in the limit that you get a completely general compressor that... (read more)

    That doesn't sound like a very serious problem with these things being "ideals". Ideals don't have to be attainable.

    Incidentally, it looks to me like you should be able to test macroscopic decoherence. Eventually. You just need nanotechnological precision, very low temperatures, and perhaps a clear area of interstellar (intergalactic?) space.

    Short of that, building a scalable quantum computer would be another (possibly easier!) way to experiment with macroscopic coherence. The difference is that with quantum computing, you wouldn't even try to isolate a quantum system perfectly from its environment. Instead you'd use really clever error-correction to encode quantum information in nonlocal degrees of freedom, in such a way that it can survive the decoherence of (say) any 1% of the qubits.

    Science is built around the assumption that you're too stupid and self-deceiving to just use Solomonoff induction.

    Now you appeal to our pride, and our vanity.

    If it bothers you to be accused of trying to start a cult, why do you persist in trying to start one?

    we are talking about the idealized form of Science, not just the actual social process of science.

    What is the point of this ideal if it is not actually implemented to a substantial degree? And of course the answer to the bottom question is that most us think we personally are too smart to need this ideal constraint, even if we think most others are not that smart. Of course most of us must be wrong to think ourselves so much smarter than the rest.

    Libertarianism secretly relies on most individuals being prosocial enough to tip at a restaurant they won't ever visit again. An economy of genuinely selfish human-level agents would implode. Similarly, Science relies on most scientists not committing sins so egregious that they can't rationalize them away.<<

    Libertarianism rarely exists as a dominant paradigm, except when certain religions, with Protestantism as an example, are dominant, and when religion is fading in strength, such that material concerns become greater than spiritual ones.

    An economy of genuinely selfish human-level agents would implode.

    Wouldn't the implosion leave all the selfish agents worse off? If they were even rudimentarily rational, wouldn't they then act in a way to prevent that inward collapse?

    Of course I think I'm smarter than that. Anyone who actually contributes does, or they wouldn't bother to contribute. And sometimes they are right, which is why our knowledge keeps moving forward. Of course, sometimes we're wrong, and the structure is what keeps everything from moving backwards.


    The core argument for libertarianism is historically motivated distrust of lovely theories of "How much better society would be, if we just made a rule that said XYZ." If that sort of trick actually worked, then more regulations would correlate to higher economic growth as society moved from local to global optima.

    Only if economic growth was the only indicator of "how good a society is."

    But when some person or interest group gets enough power to start doing everything they think is a good idea, history says that what actually happens i... (read more)

    Libertarianism secretly relies on most individuals being prosocial enough to tip at a restaurant they won't ever visit again. An economy of genuinely selfish human-level agents would implode.

    In other words, libertarianism could only ever work with real people. It would never work with the fictional creatures that people both for and against libertarianism philosophize themselves into imagining.

    Where are the witty critical posters? I'm surprised to be the first to observe that this post (favorably comparing Science with Libertarianism) reads kind of like a self-parody of the OvercomingBias blog. Is one libertarian if one holds up each claim of libertarianism and says "Well, that's an empirical question. Let's look at the data". Because that's the scientific, empirical approach, it seems to me. I think libertarianism starts to look sill when viewed in that light. To be fair, so do the claims of any political party that size or larger, of which I'm aware.

    Caledonian - not sure if this is what was originally alluded to, but the Prisoner's Dilemma / Tragedy of the Commons scenario is one where agents acting in their best interest get screwed. Of course, that is why we have governments in the first place (i.e. to get around those problems).

    M - How do you figure Somalia is libertarian? Libertarianism requires a stable government (i.e. a monopoly on force) which Somalia definitely does not have.

    H.A. - I don't think the point was that Libertarians are more scientific than others, but that Libertarianism and Science are similar in the sense that they put more faith in processes than in people.

    Of course, that is why we have governments in the first place (i.e. to get around those problems).

    It doesn't get around those problems, DaveInNYC, it just changes the conditions under which the problems arise. Having one, really powerful actor that can dominate the Commons doesn't solve anything - if there aren't enough reasonable and enlightened people to maintain control of the government, it begins abusing the Commons itself, and anyone not completely controlled by the government lose any disincentives to act for their immediate short-term interests... (read more)

    What guards institutions is other institutions, eg a free press. You are tacitly assuming there is just a Government and the People and that is that. But healthy systems have multiple institutions, and even split the government, eg into executive and legislature.

    Added small-'l' to "libertarianism" to hopefully make it clear that I'm talking about pragmatic distrust of governmental solutions, not the American political party. Also added: "People who see Libertarianism as the perfect solution for perfect people, strike me as kinda missing the point of the "pragmatic distrust" thing."

    "Libertarianism and Science are similar in the sense that they put more faith in processes than in people."

    Science doesn't seem to me to need or benefit from a "libertarian" connection. It's more reputable than libertarianism, and for good reason, in my observation. If one wants to bring science into the public policy space, then one should scientifically determine what we want to do with public policy (maximize HA's persistence odds, of course), and scientifically determine the best way to accomplish that (an empirical question). Not s... (read more)

    I am not smarter than that. But you might (just might) be. "Eliezer says so" is strong evidence for anything. I'm too stupid to use the full power of Bayes, and I should defer to Science, but Eliezer is one of the few best Bayesian wannabes - he may be mistaken, but he isn't crazily refusing to let go of his pet theory. Still not enough to make me accept MWI, but a major change in my estimate nonetheless.

    As a side note, what actually happens in a true libertarian system is Europe during the Industrial Revolution.

    As far as I can see the main problem with libertarianism (versus any variety of freebie-ism for any favored group, left or right) is the classic monkey trap problem. Freebie-ism delivers you a candy now. Libertarianism lets you work like stink and pass it on to the kids. It gives you industrial revolutions, which blacken the air - and result in modern communication, travel, medicine, computers, materials, and an upraised middle class to appreciate them. The trouble is that the "jam tomorrow" of libertarianism is quite obvious from an theorist's a... (read more)

    I don't see libertarianism as being able to jump outside the (legal/social) system in quite the way described here. It is not an escape from "how much better society would be if we just made a rule...". It is, rather, a very specific implementation of that principle: how much better society would be if we just lived by libertarian ideals, private property, courts where we can sue for fraud and coercion, etc. And then, due to its failure to jump outside the limitations of systems composed of imperfect humans, it fails for the same reasons. People ... (read more)

    Actually, I was kind of joking; sorry.


    Regarding Somalia, it actually improved under anarchy.

    HA, you may be interested in the "post-libertarianism" of Jeffrey Friedman.

    I've been a sometime educator to a steady stream of Somali immigrants into my city whose ongoing diaspora suggests you shouldn't read too much into the above essay. Some of the claims in that essay are just surreal distortions of reality (like that the violence in Somalia largely vanished in 1994 -- actually, most stability that there has been is to be found within the regions controlled by the transitional federal government around Mogadisho-- outbreaks of violence and warlordism are not things of the past here. It's jarring to contrast this account of relative prosperity compared to the Barre government with those of my more recently-arrived students, one of whom had an unborn child literally sliced out of her womb with a bayonet five years ago, or the one who lost an arm to random gunfire sometime in the last three, or the teacher who held on for a decade and a half in Mogadisho before finally fleeing in the face of warlord crossfires (even under the aegis of what passes for a government). There are hundreds of these people in the ESL classes I taught alone.
    Which system are you saying produced less violence? I'm not sure I follow.
    What I'm saying isn't which system produced less violence, it's that the framing in the article of things being mostly stable and peaceful and violence being an isolated, localized phenomenon is misleading at the very best -- the article says that by 1994 it had largely petered out. Given what I've heard about daily life there from people who've actually, you know, lived there in the timeframe the article cites, including in the areas generally agreed to be the most stable, that seems like comparing a sucking chest wound from a bullet to a 6-inch long knife wound that's penetrated the abdominal cavity and saying that the latter is much better. Well, it's not as immediately life-threatening, sure (in a merely-relative sense), but it's still intolerably bad, enough that trumpeting it as a triumph of some social policy is just disingenuous.
    Let us compare with Ivory coast democracy, where most of the coastal population was ethnically cleansed to the less desirable inland areas, or Nigerian democracy that led to the the Biafran genocide. Black African governance is apt to be extremely bad under all systems. Anarchic Somalia not only does considerably better than the outstandingly dreadful Barre government, but arguably better than some governments beloved by the UN, the world bank, and western agencies. Faced with Ivory coast democracy, it is reasonable to fight. Had the locals fought, and done OK, you would be getting similar reports about how dreadful things were in the Ivory coast.

    I don't think this was yet encompassed by the discussion or post: Another similarity between Science and libertarianism is that they both follow Bayes "usually, but not always":

    -Science rejects late-coming simpler theories that offer no new predictions. -libertarianism rejects certain government interventions proposed by a superintelligent FAI that follows Bayes and values what humans do.

    Or am I off the rails here? ;-)

    -libertarianism rejects certain government interventions that it doesn't take a super-intelligent FAI to figure out are wins all around...
    citation needed. edit: no seriously. I would like to see evidence of positive effects from government intervention. Sorting through the poor quality economic analysis on the internet is not a good use of my time if someone has links to high quality material.

    Science and Eliezer both agree that evidence is important, so let's collect some evidence on which one is more accurate.

    Echoing Hal Finney...

    It seems like you (Robin Hanson) are arguing that Libertarianism (small or big L) is some kind of alternative to rule making, or as I would say it 'believing in your theory'. But my impression -- not extravagantly well informed theoretically, but fairly informed by looking at actual, self styled libertarians -- is that Libertarianism is precisely an anti-theory theory. Terrified of the failures of other rules/models/belief systems, they create a new rule which says that all rules are wrong. The obvious tail chasing is well, what abo... (read more)

    On the other hand, I believe that the mortgage crisis was partly caused by the government failing to follow the minimum libertarian rule set-- it didn't punish fraud by the banks.

    Science is built around the assumption that you're too stupid and self-deceiving to just use Solomonoff induction.

    We can't use Solomonoff induction - because it is uncomputable.

    We don't have any good quality computable approximations to it either. That is indeed because we are too stupid. That is more fact than assumption, though.

    What do you mean by this? Surely not that it is uncomputable in the theoretic sense? Minimum Description Length and Kolmogorov Complexity are most definitely computable for a wide range of problems and are highly used for model selection, especially model order selection, in pattern theory, decision theory, and machine learning. These are equivalent to Solomonoff Induction. Is there some special case problem to which you refer? These things may not be computable in some cases, but they are generally computable.
    Solomonoff induction is uncomputable. So is Kolmogorov complexity. Actually it's equivalent to the halting problem. Could you give some references for your claims?
    Yes, you can't compute the specific value of K. But I'm asking why anyone thinks that is relevant to the use of Kolmogorov complexity for distinguishing hypotheses. If someone specifies the Mandlebrot set graphically, and I specify it with a short algorithm, then my explanation for the set wins based on minimum description length. Maybe my algorithm for it doesn't achieve the "true" value of K for the Mandlebrot set, but why does this matter? When I say that these things are computable for a wide range of problems, I don't mean that they violate the equivalence with the halting problem. I mean that you can compare hypotheses and distinguish them by noting which has smaller description length. The example that comes to mind is the classic example of a Gaussian mixture model. How many components should it have? It's a trade-off between having a small number of components and being able to match the data. If several hypotheses can all match the data equally well, then the one with the fewest components wins. It is in this sense that the OP above mentions how Science doesn't trust us to use Solomonoff induction. It doesn't mean anything to say something like: "We can't use Solomonoff induction - because it is uncomputable." To me, this is like saying we can't distinguish between which algorithm for a given problem is more complex because we don't know if P = NP, or something along these lines. The practical use of Solomonoff induction (i.e. doing computations with it) has little to do with its non-computability in the theoretical sense. See Bayesian information criterion.
    Sure it does: that was just stating a true fact.
    I disagree that the statement of that fact is relevant to the OP above. Further, it's not a true fact. We can use Solomonoff induction despite the fact that it's not computable. Even just its mere conceptual analysis is useful for driving the understanding of approximations. I'm not critical of the noncomputability part of the statement; I am critical of the logical leap in going from its noncomputability to the claim that we can't use it.
    I recommend distinguishing between "using Solomonoff induction" and "using the idea of Solomonoff induction". We can do the latter, but not the former. If you want to talk about "computable approximations to Solomonoff induction" it is probably best to spell that concept out as well - or else use a term like "general-purpose forecaster".
    In a general discussion about Solomonoff induction, I'd agree with you. But based upon Eli's claim in the OP: I don't think it's reasonable to bring in the noncomputability of Solomonoff induction into this. That has nothing to do with why we build things around the assumption that individuals won't correctly utilize Occam's razor. It's a curmudgeony nag to critique this statement by dragging computability theory into this and doesn't address the spirit of the arguments about Science. If you have Hypothesis A and Hypothesis B and B is simpler and fits the data just as well (all else being equal), then (by appeal to Solomonoff induction) it's rational to choose B. But here in Science we don't see this. We cling to old theories out of stigma, like clinging to Copenhagen and requiring that MW bring forth overwhelming new experiments to refute Copenhagen. The point of the OP is that yes, we need to distrust that others will correctly put aside biases and use Occam's razor correctly when distinguishing hypotheses. However, one can get carried away with this and transform it into a sort of bias itself; one in which demonstrably reasonable arguments are not paid attention to simply because they are new or because they address interpretations of prior results rather than presenting new experiments that visibly distinguish between themselves and the old hypotheses. The tendency to cling to old hypotheses is not grounded in the accusation that Occam's razor (Solomonoff induction) is itself a deficient way to view the problems. It's grounded in what might be called scientific inertia. If science is in motion in the direction of Theory A, there's an irrational sluggishness in suddenly jumping ship to Theory B. Solomonoff induction or Occam's razor or Kolmogorov complexity or Minimim Description Length are attempts to be truly optimal about jumping ship between ideas. You shouldn't easily accept any reasonable sounding argument. But you shouldn't dismiss reasonable sounding a
    There were two criticisms: * We can't use Solomonoff induction - because it is uncomputable. * The "scientific assumption" that we are too stupid to use a sophisticated approximation of Solomonoff induction is more like a true fact: we can't do that - and that is essentially because we are too stupid to know how to do it. The idea of "Science" as incapable of using Occam's razor seems like a bit of a straw man. I learned to use Occam's razor by studying science. Distinguishing between scientific theories is listed as the first application of the razor here.
    In my experience, it is rare that someone has a legitimate background in the appropriate use of Occam's razor. I do not agree that this is a straw man. I think you're also conflating two issues. I see it as an issue that individuals are not able to overcome biases well enough to reliably use Occam's razor when promoting solutions to problems. The scientific community as a whole is much more successful in doing this, and no one (neither me nor the OP) disagrees. But an alternate issue arises which is that the scientific community tends to simply fail to evaluate whether or not a proposed theory wins (in the Occam's razor sense) unless there is a tremendous stack of easily visible experimental evidence to motivate such an evaluation. This is a major reason why single-world views have persisted for so long. Few cling to single-world views because they are "favorite pet theories" (which would classify such an error into the appropriate-use-of-Occam's-razor type). More often it is just that alternative explanations will simply not even be considered just because they don't have the temporally aggregated endorsement of the scientific community. If Eliezer walked up to Sir Roger Penrose and presented a great argument about why the explanation of consciousness due to quantum gravity was just a mysterious answer to a mysterious question, and Penrose replied with something like, "Come back and talk to me when you've got 20 years worth of experimental evidence on your side... I don't want to hear about your retro-active interpretations... it's not worth my time if there's not a mountain of evidence to persuade me to update to any new position", this would be the type of mistake that the OP is trying to point out. And as a grad student at an R-1 university, I can tell you this is anything but a straw man. People go around not updating their maps all the time and their reasoning is just that until some new interpretation is overwhelmingly salient in terms of a flurry of brand n
    You are saying reasonable things, yet at the same time engaging in a stupid argument resulting from ambiguous use of words. Please don't do that.
    Generating hypotheses is uncomputable. However, once you have a candidate hypothesis, if it explains the observations you can do a computation to verify that, and you can always measure its complexity. So you'll never know that you have the best hypothesis, but you can compare hypotheses for quality. I'd really like to know if there's anything to be known about the nature of the suboptimal predictions you'll make if you use suboptimal hypotheses, since we're pretty much certain to be using suboptimal hypotheses.


    I've been reading LW sequences sine a few months, and I find them very interesting, but I think you made a mistake in mixing politics (libertarianism, french/american revolutions, ...) into this post.

    I won't go into explaining why I think economical libertarianism is deeply flawed and not similar at all to the process of Science (for once, I don't think it degrades well at all), but above that, by calling into very complicated and very debated concepts, you're just making following your core reasoning harder to follow.

    I also think you make some factual ... (read more)

    Interesting! Welcome to Less Wrong!
    1) Welcome! 2) Libertarianism was being contrasted with Science, non-libertarianism was presented as analogous to it. 3) Policy debates should not appear one-sided

    You claim to be critiquing "economical libertarianism", but in fact you are critiquing microeconomics. For instance, you critique the familiar critique of price-fixing by presenting a purported counterexample. But the idea that price-fixing has certain predictable perverse consequences comes, not from libertarians, but from standard microeconomics, since it's a simple deduction from basic theory of supply and demand.

    Libertarians do, to be sure, make heavy use of microeconomic theory, but this does not warrant calling microeconomics "economical libertarianism", any more than the use of a bicycle by Mao to commute to work would warrant calling bicycles "transportational communism".

    So, to reinterpret your post, taking you to be attacking microeconomics, you are saying that the science of microeconomics is not in fact a science, since it is immune to empirical refutation, such as by the purported success of price-fixing.

    Well, I think my comment was misunderstood - I didn't want to start a full debate on economical libertarianism, economics or politics. To be done seriously, it would require much longer posts than a small comment on an article about science and rationality. My point was mostly that the political issues about libertarianism and about the French and American Revolutions are highly debatable, and shouldn't be sorted out in a few bold sentences as Eliezer did on the post, and that by doing so, he's more making is core post about the differences between Science and Rationality harder to follow, because he's dragging into it a very heated and complicated debate. For that, I pointed a few examples of things done by the French Revolution which were (in my opinion) very successful, but it was just an example to illustrate my core point which was : "don't drag politics in such a bold way in a post about rationality, you'll commit factual errors and antagonize people". A bit a variation over the "don't take QM as en example", that's all. Sorry for the noise ;)

    I pointed a few examples of things done by the French Revolution which were (in my opinion) very successful

    The worst policy has good consequences, the best policy has bad ones.

    The successes you cited would only be relevant if one understood Eliezer to be claiming that every consequence was bad or ephemeral from the French Revolution. While that is how politicians speak and how others speak much of the time, it's not charitable to interpret arguments as if they were from politicians.

    In the French Revolution, they were really, really confident that things would be best if they could decide more or less ad hoc to kill tens of thousands for interfering with it. In the American Revolution, they didn't trust themselves so, they tolerated more anti-revolutionary behavior, and things turned out better. That's all.

    Even if the French do make fantastic bread, the Reign of Terror was still not a good idea.

    Well, even like that I don't agree with your analysis of the Reign of Terror. It was a bad idea, but first you can't resume the French Revolution to it, and then, it was not that they "were really, really confident that things would be best if they could decide more or less ad hoc to kill tens of thousands". Robespierre, the head of state during the Terror, was against death penalty : . But he was against the death penalty in time of peace, not in time of war. That's they key to understand it : the situation of French Republic was terrific. The League of Kings (that is, the rest of Europe) was in war against it, and there were many people inside France trying to destroy the Revolution from inside and go back to Old Regime. The Terror was a survival policy, much more than a policy of overconfidence. If you look at the core texts of the French Revolution, like the Declaration of Human Rights of 1793 ( translation available at ), which was inside the Constitution of the First Republic, they were really showing a huge distrust about themselves. Just a few quotes : « Art 9. The law ought to protect public and personal liberty against the oppression of those who govern. At 12. Those who may incite, expedite, subscribe to, execute or cause to be executed arbitrary legal instruments are guilty and ought to be punished. Art 28. A people has always the right to review, to reform, and to alter its constitution. One generation cannot subject to its law the future generations. Art 31. The offenses of the representatives of the people and of its agents ought never to go unpunished. No one has the right to claim for himself more inviolability than other citizens. Art 35. When the government violates the rights of the people, insurrection is for the people and for each portion of the people the most sacred of rights and the most indispensable of
    I'm unimpressed with claiming that one has higher ideals generally yet one's present situation is an exception in which they don't apply. I want to avoid a false dichotomy: they didn't have to act like the American revolutionaries as they had a unique situation, however, they should have killed fewer people for treason, and with more rigorous trials, etc. They ought to have had a Constitution that was appropriate for reality and wasn't a suicide pact for the state if followed. To have such a thing seems like signalling to the world one's noble values and goodness, while its impracticality demands its violation and one is no longer bound by legality.
    This is almost always true, but war usually is the exception in which they don't apply. Thus, terror against farmers and the revolution devouring its children is better evidence for the terroristic tendencies of the French Revolution than terror against aristocrats.

    In general, what is in a written Constitution doesn't matter much if institutions and attitudes don't support what is there. A fair number of authoritarian countries have strong free speech and similar rights enshrined in their Constitutions, See for example the Syrian constitution. Classically authoritarian regimes either ignore such provisions or in the case of Syria use a combination of ignoring the provisions, a favorable judiciary, and using potential loopholes to minimize the actual impact of those rights. On the other hand, some countries with little to no formally documented rights are quite democratic and functional. There isn't a great correlation between what people say in their Constitutions and what they do or intend to do.

    In general, what is in a written Constitution doesn't matter much if institutions and attitudes don't support what is there.

    For some grimly comic reading, see the declaration of rights in the 1936 constitution of the U.S.S.R., especially the articles 124-128.

    This constitution was ratified a few months before the climax of the Great Terror.

    This reminds me of the People's Republic of Tyranny Trope (TV Tropes warning). All sorts of self-interest, repression, and tribalism gets justified by the ideals of freedom, justice, and equality. It seems that a large amount of aggression has been promoting by groups styling themselves as anti-oppression movements. I recently wrote an article about the modern notion of "social justice", which is beginning to show similar sorts of newspeak, in my view.
    This account of the French Revolution fails to explain their alarmingly drastic price control measures, which measures prefigured the liquidation of the kulaks. They killed people frivolously, because they could, not because they were threatened. Doubtless they were threatened by plots from aristocrats. Where they also threatened by farmers reluctant to sell their grain and young men reluctant to be conscripted?
    How did France manage to have subsidized and/or price controlled bread without degrading the quality?
    In a free market, price, weight, and quality would all fluctuate, so the system would be flexible at those points. Legislating two of those variables to be fixed would seem to leave the third responsible for reflecting economic changes, despite diminishing returns on changing it when the other two haven't been adjusted at all. Avoiding degrading quality wouldn't be hard, one would set the price higher than the free market would have borne, so bakers have to compete for what inelastic demand there is by improving quality alone. The cost to society is the absence of cheaper, lower quality bread, which would be a reasonable policy choice. If the legislated price were too low, bakers would compete on low manufacturing costs, and quality should spiral downward. It seems controllable whether the cost of regulation is quality or affordability.

    I hope someone shows up with knowledge of the actual history.

    I'm assuming it wasn't a price floor-- that would make bread less affordable. I believe that subsidy to manufacturers + price controls leads to decline in quality because the incentives become doing just enough to meet the regulations and competing to get permissions and subsidies from the government.

    It's possible that France exists to annoy libertarians. Essence of Style: How the French Invented High Fashion, Fine Food, Chic Cafes, Style, Sophistication, and Glamour (a book I've heard an interview about but not read) tells the story of Louis XIV making France into a world center of style. He picked winners, or created them. He promoted companies which continued to make high quality goods for centuries.

    By libertarian standards, this should be just about impossible. Or maybe it's like winning the lottery-- you might succeed, but the odds are so low that it's a very bad strategy.... and, of course, the French aristocracy's spendthrift ways and insulation did lead to the French Revolution, and a utilitarian might say that any amount of delightful food and fashion just isn't worth it.

    Still, picking winners on that scale is am... (read more)

    How much could one pick winners by simply throwing the force of your big monarchy behind them and then just relying on inertia? If a sufficiently powerful organization is set up as being culturally considered best at something that gives them a possible long-term advantage even if there's no objective standard by which theirs is better.
    Good question. One way of testing it would be to see whether other monarchies have tried anything like it, and if so, whether they succeeded.
    Googling turns up some sources, so if you're really interested this sounds like the type of project to hone (y)our scholarship skills on, rather than passively wait for someone knowledgeable. The book linked to suggests that the rise in quality coincident with price controls (I don't know about subsidies but I can confirm the price controls) is a recent phenomenon, so data should be readily available. This is a great "applied rationality" question IMO - you've noticed a clash between some doctrine that you hold and some external evidence. Suggestion: make a Discussion top-level post, making a little more precise where the clash lies, to coordinate the finding and reporting of relevant evidence.
    It's telling that the Quatorze winner-picking you single out are in "high fashion, fine food, chic cafes, style, sophistication, and glamour." It's notoriously difficult to find objective measures of quality in those sectors. 17th century England was far more "libertarian" than 17th century France. And more prosperous too, right?
    You agree that the effect of a high floor would be higher quality, but doubt such a law wold be enacted because of the unpopularity of raising food proces? If a higher than free market price floor raises quality, and subsidies per unit lower it, and a high price floor is politically unpopular as it would raise food prices, a subsidy lowering the free market price and a floor above that new price would still raise quality, depending on the variables. Much better to be able to change what other people want. Guessing is so hard!
    No, I'm not sure what the effect of a price floor would be on quality (it seems to me it could be positive, negative, or neutral), but I don't have to care because I don't think a price floor would be politically possible under the circumstances. ---------------------------------------- I'm not convinced it's possible to reliably change what people want.
    If you think subsidies politically possible, and subsidies would lower food prices, wouldn't subsidies and a price floor be politically possible together? Then, if the floor raised quality more than subsidies lowered it, couldn't there be a net increase in quality?
    The point of not having a price floor is that the political pressure is to keep bread affordable. I've ordered the book about the history of French bread, though I don't have tremendous faith that there will be enough about politics and economics to answer the question of how a controlled market maintained high quality. I'm fantasizing that there were panels of bakers doing blind tests of the flour, but really I'm guessing. Meanwhile, something to contemplate from a fine essay about samovars: I believe competence happens when there's enough pressure for accomplishment, but not too much.
    An analogy: The point of having a minimum wage is to help low wage earners. Would it be politically difficult to lower the minimum wage by 15 cents? I assume so, and resistance would be motivated by direct concern for the poor. Would it be politically difficult to simultaneously, in a single bill, lower the minimum wage by 15 cents, slightly raise all taxes, and provide a transfer of hundreds of dollars per month to every person in your country? Also yes, but for different reasons entirely, having nothing proximately to do with direct concern for the poor. Combining the measures would eliminate the previous political objections, like a strong wave swamping weak one. In practice, some laws have establish fixed prices. A fixed price law is the same as a bill with two laws establishing a floor and a ceiling. Such laws have not been politically impossible in history, and a compromise bill needn't satisfy various constituencies by containing laws of parallel structure (i.e. limits to a price range). I don't think one can say a measure (especially one strongly supported by a minority) would not be politically feasible alone and consequently conclude it would not be the outcome of a compromise political process.
    The Führerprinzip. Committees will always produce crap. It is entirely unsurprising that prerevolutionary France should be the leading market for high style, fashion, and luxury goods, and entirely unsurprising that it should be the leading producer. I am inclined to doubt that Louis XIV created winners, though doubtless he picked them, the latter being much less surprising.
    It comes from two different main reasons to me. The first one is asymmetry of information in economics : the information on price in much easier to have, and with more confidence, than the information on quality. So customer have a more complete information about pricing (when we speak of bread at least, it may not be the case with complex debt schemes) than on quality, which makes competing on the price more efficient (in term of dragging customer) than competing on the quality. The second one is in economies of scale : by fixing the price of bread to a value low enough for people to afford it, but high enough for bakers to not go bankrupt, it'll increase the demand for bread, decreasing the production cost of each single bread. Economies of scale can often counterbalance the law of supply and demand (or more exactly, they violate the hypothesis on which the law of supply and demand is built, making it inapplicable), and are badly underestimated in the usual formulations of economical liberalism.
    My understanding is that France at one point (Paris in charge) fixed the price of bread artificially low, to benefit Paris at the expense of the peasantry, with utterly disastrous consequences (famine, urban riots, and terror against the peasantry), and much later, after the Paris commune was overthrown and the leadership of the rural militias was in charge, fixed the price of bread artificially high, to benefit the farmers at the expense of the city folk. I conjecture that it is this latter regime that produced competition on quality.
    When the maximum was first applied, the price of bread was controlled down, there were food riots, so it would seem that they did degrade the quality. However, when the price of bread was controlled up, rather than controlled down, that did not affect the quality, and arguably improved it.

    I also think you make some factual errors : saying the "american revolution" is a success but the french one a collapse is a great mistake. Most of the progress of the French Revolution lasted for very long and still last.

    With all due respect, your account of the French Revolution is just cartoonishly biased. The "progress of the French Revolution" included, among other things:

    • The introduction of total war fought with mass conscript armies, for which all the resources of the nation are requisitioned, in place of the 18th century limited and professional warfare regulated by strict codes and financed mostly from monarchs' private purses.

    • This invention leading to two decades of Europe-wide mass slaughter and destruction that left an unknown number of millions of people dead. It also left the recurring idea of spreading the national glory and ideology (as opposed to mere interests of rulers, which may be vicious but are at least limited and sane) by war and conquest.

    • Overall, the nationalist ideology born in the Revolution and the Revolutionary Wars, both in France and elsewhere as a reaction to it, had subsequent historical consequences for which "c

    ... (read more)
    If you think legal documents are important, this is backwards. Both were written in August 1789 and probably had little influence on each other. But both were largely based on George Mason's 1776 Virginia Declaration of Rights. Mason went on to write the US list, after refusing to sign the Constitution in 1787 and demanding a national list. In particular, his 1776 document gave freedom of press. I don't know how they expanded from press to speech; maybe one of the 1789 documents copied the other in this expansion, but Milton bundled the two in the 1644 Areopagitica as did many later people. As to the Enlightenment concept of human rights, it is pretty clear in Jefferson's Declaration of Independence. But it owes a lot to pre-Enlightenment English documents, particularly the Petition of Right of 1628. But why should we trace these concepts to their endorsement by governments? The documents did not create the ideas, but adopted them from a long train of authors, particularly the French Enlightenment. The documents are important as signposts in the triumph of the Enlightenment, but they are very crude measures because talk is cheap.

    Science first came to know itself as a rebellion against trusting the word of Aristotle.

    Digressing, this is PC history.

    Science first came to be as Roger Bacon writing up the scientific method. His approach was to not trust anyone, but to trust Aristotle more than most. Unsurprisingly, he was put in solitary confinement on bread and water. The Church then issued a list of forbidden thoughts, with Aristotle prominently on the list. That science started with a revolt against Aristotle is a whitewash of the conflict between the theocratic state and Science. Science, science in the sense of the scientific method, not science in the sense of a state anointed priesthood ceremonially wearing labcoats as white robes, is inherently revolutionary, a defiance of authority, but it was not the authority of Aristotle that they were revolting against. Rather, all beliefs were subject to empirical scrutiny, including the beliefs of the authorities of Roger Bacon's day, which was revolt against present authority, not Aristotle.

    We do not know what the charges were against Roger Bacon (most likely the nominal charges were irrelevant, and the real charge was having a bad attitude), but it was more likely he was imprisoned for respecting Aristotle, than disrespecting him.

    Indeed, Aristotle was in many ways the first Empricist, and fell into/ out of fashion several times in the history of the Church

    Libertarianism secretly relies on most individuals being prosocial enough to tip at a restaurant they won't ever visit again.

    I'm puzzled that you gave that specific example, given that it's obviously wrong. Most countries do not have a culture of tipping, and their economies don't implode. They just have less headaches at bill time. And in many cases (a long way from libertarianism) their wait staff get paid a living wage.

    I'm also not sure what it means for libertarianism to rely on something, since libertarianism is not an actual functioning thing in e... (read more)

    I think that was a metonym. Basically, don't prudently predate.
    Fair enough, and taking it that way, I think the reasoning does hold up.