No One Can Exempt You From Rationality's Laws

My blog happens to be absolutely packed with rational proof of Orthodox Judaism and proof against atheism.

Let anyone who ever thinks that I am attacking a strawman visit the above blog.

Let anyone who thinks I am mistaken take the trouble to explain why. Unless they cannot.

Fantastic post Eliezer, many of your recent posts have been articulating thoughts that I've been mulling about in my head over the last year or so. This one especially since I had an argument with a friend on this very subject not even a week ago haha. When you get around to publishing all of this as a book I will definitely buy a copy for myself and for my friend.

Also, Jacob, I glanced at your blog and saw the post on human evolution, if that is any indication of the quality of proof you have for Orthodox Judaism I don't think I need to read any further.

Savagehenry, I would happy to hear you enlighten with your thoughts. If you indeed have any.

Jacob - going into detail about why atheists are evil, violent, pornography-loving, science-worshiping people doesn't disprove their worldview. (And I find it interesting that you claim that atheists go into science, rather than scientists choosing atheism - but then, you don't seem to know what science is, so this shouldn't surprise me.)

Incidentally, out of eight models for quantum mechanics, at least two continue to permit determinism, which, notably, is another thing you erroneously attribute to atheism. One, neo-realism, of which Einstein was a follower. The other, multiverse theory. One of many matters on which you get your facts entirely wrong.

Quantum mechanics has posed a very serious challenge to determinism. http://en.wikipedia.org/wiki/Determinism#Determinism.2C_quantum_mechanics_and_classical_physics

However, I would not expect someone as deeply in denial of reality as an atheist to be able to admit that.

Eliezer,

Given that Jacob Stein claims the reason Darwinism is accepted by scientists is that it provides justification for raping children, I think it's fair to say that he is a straw man.

Anyway, great post. As usual.

I guess we should just accept Darwinism and abandon monotheism because scientists say so. That appeal to authority is SO convincing. I can't imagine why it hasn't occurred to me?

I suppose 300 years ago, when every professor at Oxford would have told me that Jesus Christ was my Lord and Savior, I would have been a fool for denying that as well.

Not even a vote of the whole human species can obtain mercy in the court of Nature.

Because the Vote was corrupted in some way: conglomerates own the media; racism and sexism persist; the Voters have internalized their oppression; violence is institutionalized in Western culture.

Hollerith, was that a joke? Sarcasm sometimes doesn't carry well over text.

Between appeal to valid authority and guilt by association, I choose argumentum ad logicam.

Ah, very Latin of you. Is argumentum ad logicam something like argumentum ad webicam?

Jacob, please ensure that at any given point, no more than two of the comments listed on the "Recent Comments" display are from you. I would also suggest that anyone interested in engaging Jacob on this, do so on his blog, rather than here.

I am looking forward to reading any comments, however many, which anyone may wish to leave on my blog. I am sure I will find them interesting.

Your point would be so much stronger, Eliezer, if you were allowed to ignore the role of models in rationality. But in all cases an infinity of alternative models may also account for what you think you have proven rationally. In your terms, no one can revoke the law that any belief in "accurate beliefs" rests on a priori assertions about what can exist and what constitutes evidence. It rests on a priori structures in your brain, designed to notice some things and not others.

Rationality is heuristic. In the case of waiting for water to spontaneously freeze at room temperature, it may be a marvelous heuristic not to hold your breath, but that's a straw man. What I'm worried about as a post-modern skeptic is what ways of organizing the world you and I have systematically failed to consider in our rational analyses. Because many internally consistent constructions of the world may be incommensurable, and yet lead not only to different predictions, but incommensurable predictions.

When you write about rationality as a way to defeat self-certainty, I'm excited and grateful. That's also how I use it. I'm more nervous when you write as if rationality is a tool that inevitably to accurate beliefs.

When you write about rationality as a way to defeat self-certainty, I'm excited and grateful. That's also how I use it. I'm more nervous when you write as if rationality is a tool that inevitably to accurate beliefs.

I think, if you'll look at what's said above, the idea is that to get anywhere we must follow a course that partakes of the rationality-pattern, of evidence-processing. Nothing is said of all problems being solvable. Nothing is being said of humans becoming perfect through following, with their finite computing power and noisy brains, some rationalist's apprehension of the Way.

But it is equally an error to say, "This Way will not always give you a solution; therefore, it must not be the true Way; therefore, to draw maps without looking must be the true Way, and will always yield a solution."

More on the concept of "a priori" tomorrow.

I suspect that if I tried to argue with Jacob Stein, the discussion would eventually turn into something like this.

Me: "X is so." Jacob: "No, ~X is so." Me: "But experts say X is so." Jacob: "But these other experts say ~X is so." Me: "Your experts are wrong and incompetent." Jacob: "No, your experts are wrong and incompetent." ad infinitum.

We'd just contradict each other and get nowhere.

Yes, joke. Irony often yields wonderful conciseness, but can make the person I am replying to feel attacked, so maybe I should refrain.

Although you display a healthy scepticism towards America's civic religion, many of your colleague-bloggers here are pickled in it, as evidenced by the fascination with the agreement theorem and with flimsy evidence for the genius of crowds for estimating weights and numbers of marbles, so I get eager to counter their pernicious influence.

I was happy to see evidence that you have looked at Unqualified Reservations, which IMO is second only to Nick Szabo's blog on our legal and political institutions.

Doug, I hope I'm not exceeding too much my comment quota. But why be so pessimistic? The Talmud states "A wise person is someone who learns from everyone." Even if we do not reach complete agreement, we make pick up a few good ideas from each other.

Why be so pessimistic? Well, right now this minute the latest article on your blog suggests that if an eminent scientist declares that evolution is a fact then one should consider seriously that "quite possibly" he "may well" be saying it only because he's a child-molesting predator.

Whether this indicates that you're incredibly stupid, or that you simply don't care whether or not what you say is true, or that you think there's nothing wrong with slandering thousands of people in this way, or that you're living in some separate reality from the rest of us, doesn't really matter; it's already sufficient to make my personal estimate of the probability of any worthwhile interaction well below 1%. (And also to make it very difficult for me to believe that you have any sincere intention of "picking up good ideas from each other".)

(Incidentally, I think it very unlikely that incredible stupidity is the explanation. That is not, in this context, to your credit.)

g, I am merely suggesting that we take everything with a grain of salt. I'm sorry if you have a problem with that. Perhaps you are accepting atheism as uncritically as you previously accepted Christianity.

No, Jacob, you aren't merely saying that and it's transparently obvious that you aren't merely saying that. To think that a good way of merely saying that is to do as you did -- or even to think that any sane person would believe you when you claim it -- would require that incredible stupidity I mentioned, and I don't think you're actually incredibly stupid.

Anyway, I shall now take Eliezer's advice and stop attempting to discuss things with you here. If I'd noticed that an anonymous commenter here had already drawn attention to your odious comments about scientists and believers in evolution, I wouldn't have bothered in the first place. My apologies to readers here for wasting bandwidth.

Greatest OB discussion thread ever.

g, I would also like to point out that the fact that you previously embraced Christianity and currently believe in evolution with equal enthusiasm does not demonstrate that you are a foolish dupe. I would merely say that it's a possibility one should consider when debating with you.

Here is another comment on democracy, and I warn the reader of highly provocative and unorthodox opinions ahead.

Here is a quick example of what I mean when I say that some of the the bloggers here are pickled in our civic religion:

As for liberal democracy, it's clearly an error to assert without further argument that liberal democracy will solve all future problems. But it is not a mistake to say that it is far and away the most successful thing that humans have ever come up with, and so that it is the best framework in which to try to address future problems.

Note that it does not say, "the most successful political system," which is IMO a reasonable assertion, but rather, "the most successful thing," e.g., more successful than our science and technology.

If liberal democracy is more potent or more successful than science and technology, why have there been no significant innovations (or improvements) in that tradition since the Jeffersonian- Washingtonian- Hamiltonian- Franklin innovation? (The civil-rights movement was not a significant innovation IMO, but merely a more consistent holding of the system to promises already made by the original (Jeffersonian et al) innovation. Also, even if we count it as a significant innovation, the makes only one significant innovation in 200 years. The parliamentary innovation is an innovation only if you disbelieve the Jeffersonian argument about the separation of powers and about the advisability of intentionally making it hard to institute sweeping changes.)

Since the last important innovation in liberal democracy, our (Baconian- Galilean- Newtonian) scientific tradition has seen vigorous improvement. Namely, we have seen the Frege- Hilbert- Russell innovation in mathematical logic, the Bayesian- Laplacean- Goodean- Jaynesian innovation (well, the Goodean- Jaynesian part of it anyway: Bayes predates the American revolution and Laplace was 32 when the war ended), the Darwinian- Wallacean innovation, the Lorentz- Poicaire- Einstein innovation, the innovation that is quantum physics, the Williams- Maynard Smith- Hamilton- Trivers innovation in evolutionary psychology, the Kahneman- Tversky innovation.

Similarly, since the last important innovation in liberal democracy, we have witnessed impressive improvement in our technology: the Industrial Revolution, the Eli Whitney- Henry Ford- Frederick Winslow Taylor- Peter Drucker- Deming innovation for running manufacturing firms, the von-Neumann- Turing- Aiken- Backus innovation of the digital computer, the Licklider- Roberts- Cerf- Engelbart- Saltzer- Reed- Clark innovation in packet-switched networking, the Sutherland- Nelson- Engelbart- Berners-Lee- Cunningham- Sanger- Wales innovation in hypertext and communications, etc.

The idea that majority votes tend to produce correct or ethical decisions (either directly or via elected representatives) and the idea that if they do not, then they can be made to do so by promoting social justice, social, racial or economic equality, nonviolence or wider participation in elections and in the public discourse leading up to elections -- that is a really silly and poisonously false idea.

In reality, the capacity to make correct and ethical decisions in our complex modern world is distributed extremely unevenly in the population, and no New Deal or Great Society program or progressive agenda is going to change that fact in any relevant time frame.

At the Singularity Summit, a woman in the audience asked, "I am an artist." How can I participate in the implementation of the singularity (not verbatim)? Well, the answer is, Unless you are an extremely unusually rational artist, you can't. Your best course of action is not to try to add your voice to the conversation. If you want to help, send money or if you seek a more personal involvement, befriend a singularitarian and share with him the knack for pleasure and delightful experience that many artists have.

Is Eliezer going to tell me that my answer to the artist is wrong?

Well, maybe he is. After all, if I understand correctly he depends for his living on donations to the Singularity Institute. Agreeing with me will alienate most prospective donors. They say that it is impossible to convince a man of a truth if his livelihood depends on his not understanding it. Although I believe Eliezer to be much less prone to bias than most, maybe this bias ("livelihood bias"? "bias towards wanting to eat and make the car payment"?) is too much for him to overcome.

Arrogance is not a very nice trait, but does believing oneself to be more rational than most people always entail arrogance? Is there no chain of experiences a human being can undergo under which it is rational to conclude that one is much more rational than the average person?

One important reason liberal democracy and later elaborations involving social justice, equality, nonviolence, universal sufferage have so few thoughtful critics is because those with the skill and knowledge to critique it tend to be employed as scholars, scientists or at least as professionals of some sort, and the ideologues will get you fired from these sorts of jobs if your critique is too successful at finding an audience. Also, it can be quite costly to one's career or one's social standing to create the perception among prospective friends and partners of being recklessly impolitic or dangerously heterodox.

Admittedly, another (quite sensible) reason liberal democracy has so few thoughtful critics is that the two (nondemocratic) political innovations of the 20th Century that were tried on a large scale in the industrial world went disasterously wrong and killed many people. So let me stress that I am not advocating another large-scale experiment in government. I explain the reason for my critiquing of democracy below.

Again, I repeat: most of our received political wisdom is really silly and poisonously false.

It would take a concerted campain of organized violence to correct the problem, however, because you have to divest the ideologues from their power (particularly from the universities and the school system), and if you were to succeed, then human nature being what it is, some new political or religious orthodoxy would take its place in a few generations. Also, there is nothing that the ideologues are likely to do that cannot be corrected over the course of a few human generations. Russia, Central Asia and Eastern Europe were damaged by their 70-year experiment with Marxist-Leninist ideology, but there is no reason to think that they will not recover from the damage in a few generations. I do not mean to dismiss the real suffering and loss of human life and human potential caused by that experiment in Marxism-Lenninism, or of the (less severe) suffering caused by the political system of the Western democracies, but since you and I have only very course control over the political environment, we cannot (without ultratechnologies) prevent it from doing significant damage. It is important not to lose track of the most important consideration, which in this case is as follows: Ever since Kepler, Galileo and Newton, our civilization has seen almost continuous progress in science, technology and the amount of wealth that is available to the average person. As a result, a young prospective rationalist has available leisure hours, bodies of knowledge and tools (like the internet) far better suited to making rapid progress learning rational skill than that available to previous generations. Our political system is not the ideal one to nurture the continuation of this progress (e.g., it has a worrisome tendency to enforce an orthodoxy on social scientists), but it is not too bad, and is probably the best out of all the systems that have been tried. And changing it would be very disruptive and, well, bloody. So, it is probably best to leave it standing as long as it continues to allow scientific, technological and economic progress.

The one exception I know of to the general rule that our silly and poisonous political culture cannot do any damage that cannot be corrected in a few generations is the project Eliezer started, namely, the deliberate creation of an intelligence explosion through AI programming. A mistake there can persist for billions of years and indeed might even persists in ways that cannot be adequately described by the passage of time. Moreover, in 2004 or so, Eliezer authored a document that was titled Collective Volition and is now titled Coherent Extrapolated Volition, which in my humble opinion shows Eliezer to have been as of 2004 too credulous and too enamoured of this idea that important decisions come out better when as many people as possible have a say or a vote or an influence. Note the similarity between the title "Collective Volition" and the fatuous platitude "the Will of the People".

Individuals like Eliezer who have mastered a great deal of science at a young age will tend to have come from loving and saner-than-usual families and to be able to attract saner-than-usual friends and colleagues, so it is possible that they find it hard to imagine fully the mendacity, fatuousness and zealotry of the idealists in our "opinion-making" professions and the ruthless careerism and casual butchering of truth of the realists in our opinion-making professions. It is of course our opinion-makers who create and refine our political culture.

Since I have lived my life on the margins of our society and have depended often on ordinary professionals (in my case, doctors, other health-care providers, a few lawyers) with no particular distinction in rationalist skill, maybe I can convey to the reader just how shoddily ordinary professionals treat evidence and treat hypotheses.

It would serve no purpose IMO to challenge the democratic ideologues on, e.g., Daily Kos or indeed in the vast majority of public forums. It would merely ignite a nasty flamewar and it is unlikely to change anyone's opinion.

In contrast, the presence of Eliezer and young Eliezer wanna-bes on this blog makes it worthwhile for somebody (me if no one else steps up) regularly to criticize democratic political ideals, to try to neutralize the flood of democratic ideology and ideals (as exemplified by the quote above).

Again, I ask the reader to consider the proposition that most of our political traditions are silly, fatuous and false -- the modern equivalent to the Medieval Catholic Christianity. Our political culture enjoys its near monopoly on political opinion, like I said, by making life miserable for dissenters and enacting quite serious and severe punishements on dissenters.

Eliezer has over the last few months presented an accomplished technical explanation of how an individual human being makes good decisions (and how that skill might be improved). His explanations are free from slogans. Everything reduces to the non-mysterious operation of physical laws. When his explanations depend on or refer to received wisdom (about physics or neuroscience, for example), there is no reason to believe that that received wisdom is maintained by the punishment of dissenters, the removal of dissenters from positions of visibility (e.g., academic and journalistic positions), or the denouncing of dissenters as immoral or hateful.

Dear reader, ask yourself, Where can I find a corresponding detailed technical explanation of how my favorite political system makes a good decision? What is the causal mechanism in the political system that produces the correct or ethical decisions? How many decibels of evidence support the hypothesis of the existence of the causal mechanism? Has anyone calculated that quantity? The reader should demand detailed answers free from mysteries and from trite slogans that derive their persuasive power from endless repetition. Whey the voting scheme is run in a "controlled experiment", meaning one in which the voting system is asked to answer a mathematical, scientific, technological or economic question to which the answer is already known, does it in fact produce the correct answer?

Richard: I think that you misunderstand the reason for interest in the agreement theorem. The theorem is not seen as evidence in favor of America's civic religion, but rather as a particularly important proof of how far almost everyone departs from mutually recognized rationality almost all the time.

Vassar, maybe I misunderstand. I always thought informing someone about the agreement theorem will decrease the probability that that person will dare to dissent from a widely-held consensus. The belief that Majority Rule is an effective and reliable way to make correct or ethical decisions is of course a widely-held consensus. I would be obliged if those who write about the agreement theorem would periodically disclaim or at least profess agnosticism towards the notion that it applies to pursuits such as religion and politics in which dissenters face widespread ostracism and other sanctions. Nor do I buy that it applies to markets with strong network effects (meaning markets with a large "first-mover advantage" or in which an incumbent enjoys a large advantage over upstart competitors) e.g., the market for operating systems for personal computers, e.g., the market for undergraduate education at highly competitive colleges.

If I may be allowed a short tangent from the topic of this post, my strongest objection to the enthusiastic application of Majority Rule and related ideas is directed not so much at the governance of nation-states as at, e.g., the important Wikipedia project and, e.g., the extremely important singularitarian project. Rather than elections, Wikipedia would better serve its public IMO by scrapping elections and making it as easy as possible for groups to fork Wikipedia. Putting the content under a permissive, open-source license was a major step in that direction. The 2 major remaining steps IMO are a technical provision by which every competing encyclopedia's software may be notified of every change to every Wikipedia page as soon as the change is saved and the development of search engines able to lead the surfer through the bewildering array of world views and editorial approaches nurtured by the governance structure I just described.

Richard, I claim the agreement results apply fully to most topics in religion and politics, and to markets with network effects. I suspect you misunderstand what is being claimed; it applies to honest beliefs, not words or actions.

People often consistently profess some belief, which their actions belie. But that is not what you mean. Do I understand you to maintain that there is a third aspect to every person's model of reality, namely, their "honest belief", that has no particular relationship to the model they profess or the model that can be inferred from their actions? What would you consider evidence that this "honest belief" does not exist?

What I most want to know: Do you believe the results imply or strongly suggest that if a large group of people share some model of reality which differs in an important detail from models shared by much smaller groups, the model of the large group usually contains more true information and less false information than every model shared by every smaller group? Even in academic psychology, academic sociology, principles of governance and how the government should intervene in the economy?

If yes, do you require the large group to consist only of experts or those with training in the domain to which the models pertain?

Richard, honest belief is the belief they would have if they were being honest. Averaging over all possible groups large and small, the large group average belief must tend to be more accurate. But if we condition on a small group having some feature X, then of course it becomes an open question, depending on X.

Vassar, I think I get what you are trying to tell me: the agreement theorem says that if everyone were a perfect Bayesian reasoner (and a few other conditions hold) then everyone would be in perfect agreement. But surely no one believes the converse! Surely everyone agrees that perfect agreement can come about by other means, e.g., a program of indoctrination and suppression of dissenters.

I can see two cases where tu quoque arguments could have some utility:

• if they make the recipient realize "hey, I'm doing it wrong too", or
• if they lead the recipient to explain her valid reason for what looks like a wrong behavior, which might either enlighten her accuser to see that his own behavior can't be excused in this way, or enlighten the recipient herself into realizing that her excuse also excuses the accuser's behavior.

This probably doesn't happen often, though.

So what if I'm a hypocrite? You're a hypocrite too!

While the general argument is valid, I'm not sure how these accusations of socially-derived rules making up traditional rationality. There were many mathematicians and scientists before Bayes was born, and they derived their beliefs from logic and evidence, not social norms. Take Galileo as an extreme and famous example. Is there any evidence behind these unflattering descriptions of traditional rationalists?