Semantic Stopsigns

by Eliezer Yudkowsky2 min read24th Aug 2007110 comments


Confirmation BiasParadoxes

And the child asked:

Q: Where did this rock come from?

A: I chipped it off the big boulder, at the center of the village.

Q: Where did the boulder come from?

A: It probably rolled off the huge mountain that towers over our village.

Q: Where did the mountain come from?

A: The same place as all stone: it is the bones of Ymir, the primordial giant.

Q: Where did the primordial giant, Ymir, come from?

A: From the great abyss, Ginnungagap.

Q: Where did the great abyss, Ginnungagap, come from?

A: Never ask that question.

Consider the seeming paradox of the First Cause. Science has traced events back to the Big Bang, but why did the Big Bang happen? It’s all well and good to say that the zero of time begins at the Big Bang—that there is nothing before the Big Bang in the ordinary flow of minutes and hours. But saying this presumes our physical law, which itself appears highly structured; it calls out for explanation. Where did the physical laws come from? You could say that we’re all a computer simulation, but then the computer simulation is running on some other world’s laws of physics—where did those laws of physics come from?

At this point, some people say, “God!”

What could possibly make anyone, even a highly religious person, think this even helped answer the paradox of the First Cause? Why wouldn’t you automatically ask, “Where did God come from?” Saying “God is uncaused” or “God created Himself” leaves us in exactly the same position as “Time began with the Big Bang.” We just ask why the whole metasystem exists in the first place, or why some events but not others are allowed to be uncaused.

My purpose here is not to discuss the seeming paradox of the First Cause, but to ask why anyone would think “God!” could resolve the paradox. Saying “God!” is a way of belonging to a tribe, which gives people a motive to say it as often as possible—some people even say it for questions like “Why did this hurricane strike New Orleans?” Even so, you’d hope people would notice that on the particular puzzle of the First Cause, saying “God!” doesn’t help. It doesn’t make the paradox seem any less paradoxical even if true. How could anyone not notice this?

Jonathan Wallace suggested that “God!” functions as a semantic stopsign—that it isn’t a propositional assertion, so much as a cognitive traffic signal: do not think past this point.1 Saying “God!” doesn’t so much resolve the paradox, as put up a cognitive traffic signal to halt the obvious continuation of the question-and-answer chain.

Of course you’d never do that, being a good and proper atheist, right? But “God!” isn’t the only semantic stopsign, just the obvious first example.

The transhuman technologies—molecular nanotechnology, advanced biotech, genetech, artificial intelligence, et cetera—pose tough policy questions. What kind of role, if any, should a government take in supervising a parent’s choice of genes for their child? Could parents deliberately choose genes for schizophrenia? If enhancing a child’s intelligence is expensive, should governments help ensure access, to prevent the emergence of a cognitive elite? You can propose various institutions to answer these policy questions—for example, that private charities should provide financial aid for intelligence enhancement—but the obvious next question is, “Will this institution be effective?” If we rely on product liability lawsuits to prevent corporations from building harmful nanotech, will that really work?

I know someone whose answer to every one of these questions is “Liberal democracy!” That’s it. That’s his answer. If you ask the obvious question of “How well have liberal democracies performed, historically, on problems this tricky?” or “What if liberal democracy does something stupid?” then you’re an autocrat, or libertopian, or otherwise a very very bad person. No one is allowed to question democracy.

I once called this kind of thinking “the divine right of democracy.” But it is more precise to say that “Democracy!” functioned for him as a semantic stopsign. If anyone had said to him “Turn it over to the Coca-Cola corporation!” he would have asked the obvious next questions: “Why? What will the Coca-Cola corporation do about it? Why should we trust them? Have they done well in the past on equally tricky problems?”

Or suppose that someone says, “Mexican-Americans are plotting to remove all the oxygen in Earth’s atmosphere.” You’d probably ask, “Why would they do that? Don’t Mexican-Americans have to breathe too? Do Mexican-Americans even function as a unified conspiracy?” If you don’t ask these obvious next questions when someone says, “Corporations are plotting to remove Earth’s oxygen,” then “Corporations!” functions for you as a semantic stopsign.

Be careful here not to create a new generic counterargument against things you don’t like—“Oh, it’s just a stopsign!” No word is a stopsign of itself; the question is whether a word has that effect on a particular person. Having strong emotions about something doesn’t qualify it as a stopsign. I’m not exactly fond of terrorists or fearful of private property; that doesn’t mean “Terrorists!” or “Capitalism!” are cognitive traffic signals unto me. (The word “intelligence” did once have that effect on me, though no longer.) What distinguishes a semantic stopsign is failure to consider the obvious next question.

1 See Wallace’s “God vs. God” ( and “God as a Semantical Signpost” (


109 comments, sorted by Highlighting new comments since Today at 1:00 PM
New Comment
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings
Q. Where do priors originally come from?
A. Never ask that question.


from the before time. from the long long ago.

Interesting post. It strikes me that semantic stopsigns join adoration of mystery and non-falsification as survival tricks acquired by story memes when curiosity -- and other stories -- threatened their existence.

Oops, that should be "non-falsifiability," not "non-falsification."

Eliezer: "How could anyone not notice this?"

Because the human brain -- like many simpler programs -- generally finds basic beliefs more practical than an infinite regress?

Infinite regress is still a semantic stopsign. If all chickens came from eggs, and all eggs came from chickens, the obvious next question is "Why is there an infinite regress of chickens and eggs?"

There are certainly possible infinite regressions that don't exist, so it can't exist simply because of an infinite regress.

1MrMormon9ySometimes, the question "why?" is meaningless. If "all chickens came from eggs, and all eggs came from chickens" is a premise, that is the "why". Asking the why of the why is a tautology.
2DanielLC9yBut why do you assume all chickens came from eggs and all eggs came from chickens? How is this different than just having "God exists" as a premise?
1jopt28yI don't think it is different in itself. I think some premises are more useful than others, an the anticipating-experience-sense.

Synonyms for "I don't know".

There might be more of these than we think. Two candidates:

Aesthetics. People have a lot of understandable preferences: they prefer bigger houses to smaller ones, longer vacations to shorter ones, air conditioned rooms to hot and humid ones, and so on. What these have in common is that we can easily understand and explain the preferences. Aesthetic preferences, however, are generally characterized by it being hard, maybe impossible, to explain what it is about one thing that makes it more aesthetically pleasing than another. This suggests when we say "aesthetically pleasing", we almost mean, "pleasing, but if you asked why I wouldn't really be able to give you a satisfactory explanation."

Intelligence. Many have observed that each new success in reproducing, in machinery, the capabilities of the human mind, has in turn led to a narrowing of what is considered "intelligent" - a narrowing that excludes whatever it is that machines can now do. This phenomenon is explained if "intelligent behavior" is a subclass of some larger class of behavior (a larger class that includes the things that we have gotten machi... (read more)

Constant, good points both - though no word is a stopsign of itself, the question is whether one uses it that way. There are definitely people out there who use "aesthetic" and "intelligent" as stopsigns.

Doug, "Never ask that question" is an Ambassador Kosh quote.

Programming. Its all in the program. Why do we think time is linear? Here in the States we control movement with stopsigns. Elsewhere they use round-a-bouts.

I can think of a few semantic roundabouts as well. Postmodernism comes to mind.

The entire function of God seems to be as a multi-purpose philosophical semantic stop sign. It isn't just the horror of thinking about the beginning of the universe he protects us from. Consider for instance morality (an atheist's morality is empty if they just make it up, so where does God get his morality from?), and free will (can't see how a material being can have free will? It's controlled by an eternal soul with God-given free will. How does a soul - or for that matter a God - have free will? Still any aspect of its behaviour which is not related to... (read more)

Even so, you'd hope people would notice that on the particular puzzle of the First Cause, saying "God!" doesn't help. It doesn't make the paradox seem any less paradoxical even if true. How could anyone not notice this?

Thinking well is difficult, even for great philosophers. Hindsight bias might skew our judgment here.

"About two years later, I became convinced that there is no life after death, but I still believed in God, because the "First Cause" argument appeared to be irrefutable. At the age of eighteen, however, shortly befo... (read more)

0Mets7yIf I remember correctly, Russell thought that if anything could exist timelessly, without a cause, it might as well be the universe and not God. But the problem is, we now know (with some reasonable certainty) that the universe began with the Big Bang. The universe could have been timeless, but it isn't. Postulating that there is a timeless First Cause outside the universe solves this problem, since there is no similar theory that shows God has a cause, though it's not a falsifiable or very satisfying solution.
0gjm7yWe really don't know with much certainty that the universe began with the Big Bang. For instance, if "eternal inflation" is right then what we call the Big Bang was the beginning of "our" universe but not of the universe.

Whatever happened to "I don't know"

Another semantic stopsign often used, unfortunately: "that's biased".

Whatever happened to "I don't know"

Hear, hear! One of the most baffling things I've had a theist say to me is 'I don't really know where existence came from, but I need to believe something so I believe in God.' If you can't stand to say "I don't know", that's a serious bug.

2pnrjulius9yIf we interpret it as an adaptation of religious memes, it's not a bug---it's a feature.

From one point of view, all metaphysics is a semantic stopsign. I've always been sympathetic to the basic anti-metaphysics argument of Ayer, i.e. that if it's outside of the verifiable then discussion of the subject is literally meaningless, since we have no language or medium for such a discussion. I suppose Eliezer would say that belief in God is a proposition that does not control expectations, i.e. the God that explains everything is a God that explains nothing.

I hope that pre-Big Bang history doesn't remain a (literally) metaphysical subject. I'd really like to know how all this happened.

From one point of view, all metaphysics is a semantic stopsign.

Not that I necessarily disagree, but, what is the obvious next question which metaphysics prevents you from asking?

I think some theists would say that the "who made God" question is a semantic stop sign, but that this is OK. That is, they would say that they are not capable in probing into the question any further, but that the leaders of their religion (with the help of the sacred texts) are capable of doing so, and they bring back from the other side the answer that the religion is true and everything is OK.

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 no... (read more)

-2pnrjulius9yYeah, I found the "liberal democracy" example somewhat unfair too. It may be a stopsign for some people, but it's also a pretty darn good idea in general. How well has liberal democracy done on similarly tricky problems? Might it be fair to say... better than anything else yet devised? I guess I would agree it's not a complete answer. But answering "liberal democracy!" seems about as valid as answering "rationality" or "science", which we actually do give as answers sometimes (at least when pressed for time).

Hmm... "Love" is also often used as a semantic stopsign, which may contribute to the cynicism with which some people regard it.

And "I love you, but..." is the start of an argument.

David J. Balan, you write: "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."

That sounds like a contestible claim. There often seems to be a "no true scotsman" element to arguments buttressing that claim.

"What distinguishes a semantic stopsign is failure to consider the obvious next question."

I disagree. The distinguishing event is a refusal (not just a failure) to consider it, for reasons other than something like "I don't have the time right now." One cannot ask all questions in an average 70+ year lifetime, so one picks which avenues of questioning to pursue most fervently. Sometimes, one simply has to say "I choose to avoid thinking too much about what came before the big bang, because I have to spend more time thinking abou... (read more)

Before the Big Bang is beyond the universe. Beyond the universe are other laws of physics. Which laws? All self-consistent laws. What are sets of laws of physics? They're mathematics. What is mathematics? Arbitrary symbol manipulation. And there you've reached a final stopping point. Because it isn't even intelligible to ask why there are symbols or why there is mathematical existence. They are meta-axiomatic, and there is nothing beyond or beneath them. More importantly, there is no meta-level above them because they are their own meta-level.

This is merely a semantic stop-sign; it appears wise because it uses modern vocabulary. The structure of the argument is identical to the structure of a neo-Platonist argument that most LW readers would instantly reject:

Before God's Creation is beyond the universe. Beyond the universe are other Creations. Which Creations? All Creations that God created. What are God's Creations? They are instantiations of God's will. What is God's will? Arbitrary manipulation of Form. And there you've reached a final stopping point. Because it isn't even intelligible to ask why there is Form or why God wills one thing and not another. Form and Will are eternal, and there is nothing beyond or beneath them. More importantly, there is no form from which they derive their pattern because they are their own perfect Form.

0Brilliand5yThis looks like equivocation between the math-like structure of the universe and mathematics itself - mathematics proper is something invented by humans, which happens to resemble the structure of the universe. Whatever is outside the universe is unknown, but probably can be discovered with considerable difficulty (and will probably be describable by mathematics, but will not be mathematics itself).

Can we even ask the question about what comes before the big bang and hope for a scientific answer? We can see light from the universe that took about 14 billion years to reach us, but that doesn't mean we see the beginning. If it took that light so long to reach us, then we cannot say definitively whether or not there is light that is 15 billion years old which has not yet reached us. We don't even know the shape of the universe or where we are in that shape. Are we in the middle? On the edge? Is there even an edge? I doubt we will ever be able to answer these questions. I'd say the only answer we could really give to the question of "What came before the big bang?" is "I don't know."

This is a disingenuous (though not uncommon among modern theists) interpretation of the first cause argument. The justification for stopping with God is that, supposedly, his existence is necessary. If a creative God had to exist, then this explains why there is so much stuff about when seemingly, there might not have been.

This is a defence that does not rest easily in the theists hands though, for it relies upon there being an argument that gods existence is necessary, and it is difficult enough to make out what that claim amoutns to let alone what reason, if any, there is for believing it true.

It's still a semantic stop sign, because it attempts to stop you from asking the next question, which you mentioned:

Why is God necessary?

The point is to never stop asking. You might be stuck at "I don't know", but that just means you need to find more information.

A "stop sign" is any answer that automatically causes you to stop asking the next question. It can be "God", but it can just as easily be "the big bang" or "evolution". If your intent in making the statement is to prevent further questions, it's a stop sign statement in that case.

If you believe there must be a God, there isn't really anything wrong with that as long as you aren't using the idea of God to keep you from asking the next question. I find the belief ridiculously hard to maintain though.

By the way, Eliezer, I think I have to officially recognize my atheism. I've been clinging to the last vestiges of my monotheistic upbringing up until now, but this post hits the nail on the head, so to speak.

1pnrjulius9yWe do have a pretty good understanding of some necessary claims, like "It is necessarily true that 2+2=4." Asking "why is that necessary?" has a fairly good answer: "Because it's what the symbols mean, and if you deny it you soon find yourself in incoherence." Whereas, one does not seem to fall into incoherence when asserting "There is no such thing as an omniscient, omnipotent being whose son died by crucifixion." (I dare say one teeters upon that chasm when asserting the opposite!)

Regarding the Big Bang, I don't know, you really don't know what you're talking about. The scientific estimate of the age of the universe is not based on how long light took to reach us. Nor is it based on whether there are parts of the universe which are not visible to us; it is assumed (reasonably) that there are.

C'mon, no theist arrives at God after 5 questions regarding existence. Nor after 4 or 6. It happens when we realize that there's an infinite number of existential questions before we can know God. We believe in God because we see there IS NO semantic stop sign.

For a metaphor, review your integral calculus. Belief is the Riemann sum of the existential questions as they approach infinity.

0pnrjulius9yInteresting theology. Completely irrelevant to 98% of believers I'm fairly sure. (I mean, how many people even know calculus, much less use it to justify religion?)

It happens when we realize that there's an infinite number of existential questions before we can know God. We believe in God because we see there IS NO semantic stop sign.
And you feel compelled to create one.

Religion is not a search for truth. It's a way to short-circuit the search.

3Mass_Driver10yHang on; it depends what your religion consists of. If, having concluded that "there is a God," you stop thinking and stop asking existential questions, then, fine, you're short-circuiting out of something like fear. But if you use religion to fast-forward a bit, to skip a few existential questions here and there and ask others further down the line, it's not clear why that would be ethically inappropriate. If existential inquiry is a journey without an end, why privilege the first several questions on that journey above all others?
1pnrjulius9yDoes religion actually help us move ahead in the list of questions? I don't see how it does.
2Mass_Driver9yIt did for me for a while. Having decided that the world was purposefully created by a benevolent deity who left much of it unfinished because he wanted humans to have a chance to complete the world so that the god would have fellow creators that he could relate to, I then moved on to ask (a) what parts of the world are unfinished, (b) what can I do to improve them, (c) what does it mean to be a creator, (d) what parts of the world are already set up as if by a friendly-to-humans force, and (e) how, if at all, can I engage in a personal relationship with creative forces so abstract and mighty as to be impersonal? To me, anyway, these questions seem more interesting than endless variations on the theme of "Oh, hey, what's this life stuff all about?" and "Is there any point in trying to accomplish things?" At the moment, I'm an atheist, and don't have much use for religion. I wouldn't say religion is so useful to existential journeys that it's worth trying to cram thoughts into your head that you don't believe -- but while I did 'naturally' believe in a religion, it offered me some useful benefits, which I enjoyed.

Eliezer: partway through your essay you make the claim that when someone hits their semantic stopsign (eg, starts to say "God" or "Liberal Democracy", full-stop), that their statement at that point is better classified as a statement of tribal membership (or, perhaps, a tribal ritual to ward off discomfort?) than as an actual semantic statement addressing the question at hand.

Or, rephrased, if I ask "from whence came those physical laws" and you say "from God!", then under this theory the fairest re-statement of the ... (read more)

0[anonymous]11yYou didn't need to phrase all that as a question; it makes a nicely plausible thesis when plainly stated.
0pnrjulius9yWell, hopefully not ALL our statements amount to tribal membership... then we couldn't even assert that our statements amount to tribal membership, now could we? And clearly some of the things you might ask along the chain are normal, unproblematic questions, like "Why is there a dog in the yard?" and "Will it rain tomorrow?"; so I don't think it's a big stretch to presume that our discourse is properly semantic until we're given reason to think otherwise. (Truth be told, I'm dubious of non-cognitivism in general; obviously it does work in your ATM example, and maybe even in the "God did it" example... but lots of people try to apply it to things where there really is an obvious propositional interpretation.)

I wonder if a religious bulletin board linked to this page or something? Clearly a lot of commenters here who haven't read anything else on Overcoming Bias.

Passing Through, Sam Harris discusses this in The End of Faith - even when beliefs are tribal, they can still act as beliefs and control behavior. E.g. suicide bombers.

0pnrjulius9yThough then they do have to have propositional content, so maybe we should say that they are stopsigns---but not merely stopsigns?

Why would we need to postulate new laws of physics to avoid a beginning?

it might be more productive to simply ignore god.

IMO we extend (scientific) enquiry by ignoring questioning taboos - e.g. the 'Copernican revolution.' For me a stop-sign is usually a direct display of power (sometimes it's indirect, when the speaker is deferring to 'its' authority.) i.e. the difference between 'As a close personal friend of Herr Furrer, I know he will not be pleased with your question' AND 'God, what would the Furrer think if he heard you say that?'

The obvious next question is, why should we care what people believe so long as they conform to act within the range of behavior standards agreed upon by our now-global tribe? E.g. not suicide bombers.

The obvious next question, somewhat after that, is, is there an end to next questions?

N.B. not sarcastic here, seriously asking both questions. Forgive me if I've not done enough coverage of this site to encounter such discussion. Pointers welcome.

2pnrjulius9y1) Because beliefs matter; a world where people are epistemically rational is, ceteris paribus, better than a world in which they are not. 2) Because you can't just shield off all your beliefs from the real world. You can try, and thereby make them almost not beliefs at all; but as long as they reside in your brain, they are moving neurotransmitters around in ways that they should not be moved. For example, consider how the idea of immortality affects the way people view existential risks: If you believe, or even suspect, that you might live forever in Heaven after you die, you'll be just a little bit less inclined to worry about that plague, that asteroid, that cascade of Von Neumann machines. Maybe not a lot less---but a little less, and that could make all the difference.

From the Bṛhadāraṇyaka Upanishad (~6-700 BC), Olivelle translation:

Then Gargi Vacaknavi began to question him. 'Yajnavalkya', she said, 'tell me - since this whole world is woven back and forth on water, on what, then, is water woven back and forth?
'On air, Gargi.'
'On what, then, is air woven back and forth?'
'On the worlds of the intermediate region, Gargi.'
'On what, then are the worlds of the intermediate region woven back and forth?'
'On the worlds of the Gandharvas, Gargi.'
'On what, then, are the worlds of the Gandharvas woven back and forth?'
'On the worlds of the sun, Gargi.'
'On what, then, are the worlds of the sun woven back and forth?'
'On the worlds of the moon, Gargi.'
'On what, then, are the worlds of the moon woven back and forth?'
'On the worlds of the stars, Gargi.'
'On what, then, are the worlds of the stars woven back and forth?'
'On the worlds of the gods, Gargi.'
'On what, then, are the worlds of the gods woven back and forth?'
'On the worlds of Indra, Gargi.'
'On what, then, are the worlds of Indra woven back and forth?'
'On the worlds of Prajapati, Gargi.'
'On what, then, are the worlds of Prajapati woven back and forth?'
'On the worlds of brahman, Gargi.'
'On what, then, a

... (read more)
2pnrjulius9ySo here we have the literal and Ur-example of semantic stopsigning.

Eliezer writes: "What distinguishes a semantic stopsign is failure to consider the obvious next question."

I think that you are being unfair here, because you are not reading the fine print. The signs actually read "Stop, then proceed with caution." The subtext is that if you proceed beyond this point, you are going to need a different kind of explanation than the one which has served so far. Ask an atheist where life today came from and he will tell you that it came from past life (and definitely from more than 6000 years ago). Pursue the chain of causality and eventually you will come to a stop sign. "Abiogenesis" says the smart atheist. "Prebiotic soup" catechises the stupid one. These are stop signs, but they are necessary stop signs. They don't so much fail to consider the obvious next question as warn that the obvious next question may be a misleading question. What is going on in the vicinity of the stop sign may be completely different in kind from the kind of thing you are familiar with. Or so claims the stop-sign poster. It is not an explanation, to be sure. But it is something of a hypothesis. Like a notation in the une... (read more)

5Jeff_Rubinoff10yWhat if you are trying to explain evolution to someone and he states "Evolution is just another religion." Is that a stop sign? To me it is, in the sense that the only reason to continue at that point would be to enjoy the sound of your own voice. The person has just signalled his membership in a tribe; you recognize that you are not in that tribe; and you recognize that he will not consider anything further you have to say on the subject, because that would be disloyal to the tribe. Global warming is a religion, taxation is theft, property is theft, healthcare is not a right (I'm not sure if the reverse is used as a flag, too), there is no peace without justice, "allopathic medicine"; there are a lot of them. I'm old enough to remember "The Soviet Union is a state in transition," too. (Sadly, it was in transition to total collapse.) All these statements are what Eliezer calls "Green and Blue" (I think--those are the two chariot racing team colors, right?) markers. I'm not sure if these statements are also semantic stop signs. Anyway, I think that class of statement is very different than statements like abiogenesis or prebiotic soup, because the latter statements indicate that the original topic has been exhausted. That line of reasoning has gone to its logical end, and to continue conversing, we must switch to a different discussion. Not quite the same thing as saying that "tribal loyalty dictates that I do not use reason to consider anything further you say on this subject."
2PaulAlmond10yNot necessarily. Maybe you should persist and try to persuade onlookers?
7Mass_Driver10yI'd be careful before writing off otherwise polite and thoughtful people as irrational loyalists simply on the basis of a single semantic stop sign. For one thing, you might be getting a false positive -- sometimes I say things like "there is no peace without justice," but I don't mean to cut off debate about political science; I'm just trying to call people's attention to the possibility that the people they see as violent troublemakers may simply be responding to a perceived injustice. For another, even an intentional semantic stop sign might not indicate unthinking loyalty; it may simply indicate that your listener has erroneously concluded that there is nothing more to say about a particular topic. Some libertarians might think that taxation is theft, not because they refuse to listen to your counterarguments, but because they can't imagine a morally legitimate real-world government. Finally, attempting to identify semantic stop signs with the motive of screening out those who are unworthy of further conversation will inevitably lead to improper rationalization; you will feel subjectively that someone is unworthy of conversation and then concoct a story for yourself about why the other person has been using stop signs. Thus, it is better to ask people what kind of argument might convince them than to assume that people are irrational.
4Perplexed10yOh, I agree that it is a tribal slogan, signaling tribal membership. But before interpreting it as a stop sign, I'd want to ask myself just why he had come up with that particular slogan at that point in our conversation. Did he somehow perceive that I wasn't really explaining evolution, I was preaching it? That I wasn't just trying to correct misconceptions regarding a set of ideas, that I was trying to convert him? That I wasn't really interested in his opinions, but that I wanted him to enjoy the sound of my voice? Well, yes, if that seems to be the reason that slogan just happened to pop into his head, then stopping is probably the best option. But if it seems that he said that simply because it was his turn to speak and that is one learned slogan he hadn't used yet, then I would treat the statement as a question, "Evolution is just another religion, isn't it?" And I would answer, "No not like a religion at all. Evolution only deals with the subject matter of two chapters of Genesis; it doesn't even attempt to answer the questions that the rest of the Bible deals with. Some evolutionists are Catholics, some are Jews, some are Protestants, some are atheists. What kind of crazy religion is that?" Who knows? Maybe that will make him think a bit. But yes, beyond trying to clarify what evolution is and is not, I wouldn't try to get him to leave his tribe. I would feel rather silly if I had been trying to do that in the first place. Evangelizing, rather than simply explaining, about science is pointless. You cannot succeed unless the person is open to learning. And if they are open, then explaining should be all that is needed.

What’s the alternative? Say I can’t solve something, like I couldn’t solve 3n+1 (aka Collatz conjecture).

1) Accept that this is something I can’t solve, and give up? Should I live with the frustration of an open question, rather than take comfort in deferring to a semantic stopsign?

2) Try to figure it out on the off chance that I can do better than the 6.5 billion living people plus the scholars of the past? (Almost drove myself mad with 3n+1). There’re other problems that I want to tackle, ones I CAN possibly solve, ones with greater applicability to real... (read more)

1JohnH10yThere have always been questions that are unanswered. What those questions are change over time. If one believes that it is impossible to answer a question then for that person it most likely is, but not because it actually is impossible to solve or even that it is impossible for them to solve it. If problems were unsolvable because no one has solved them yet then where did any solutions of any problems ever come from? Economic growth can only happen by solving unsolved problems or solving solved problems in unknown ways. If the problem bothers you and you believe that you might possibly be able to solve it by studying the problem and working on possible solutions then you should try and solve it. It might be worthwhile to dedicate set amounts of time to the problem so that it doesn't cut into productivity in other areas of life. This would allow you to not worry much about the problem during the rest of the day, which is often helpful in solving difficult problems.

I must say I resent the allegation that all readers of this blog must be atheist - is it not permitted for me to be interested in rationality just because I am one of these 'obviously deluded' religious types.

And should you not, as a rationalist, accept the explanation that God created the universe, which is an explanation that fits the facts, and makes predictions about the future (even if you do not believe that the results can be observed), than accept that something happened (the Big Bang) which your worldview has no explanation for?

And why is God creating the universe paradoxical? Outside of this universe, with the physical laws that require causality, why does He require a beginning?

2wedrifid10yNo. Basically not. You lost all your righteous momentum here and started sounding nutty.
0theflyingfrogfish10yCare to explain my error? I'm somewhat new to this - why is a theory that has no explanation of the facts be placed above that which does?
3Plasmon10y1) The universe itself is also not something that exists inside the universe. Hence, if you think causality doesn't apply to your god, you shouldn't think causality applies to the universe as a whole either. (thus, your characterization of the big bang as a "theory that has no explanation of the facts" vs. theism as a theory which does explain the facts, is wrong) 2) You claim your religion predicts the future ("makes predictions about the future (even if you do not believe that the results can be observed)"). Presumably you refer to various afterlife-scenarios. The big bang theory predicts far more things which can be observed. Hence there is no reason for your theory (theism) to rise to the status of a hypothesis in the first place. Read Privileging the Hypothesis [] 3) The notion of "causality" in physics has changed a lot since the days of classical physics. It no longer corresponds to the naive notion of causality that might naively seem to require an uncaused cause.
0theflyingfrogfish10y1. An interesting point - all my learning in astrophysics up to now had basically said that 'we don't know how it can have happened given that the laws of entropy and thermodynamics seem to prevent it'. Although the universe as a whole seems to obey at least some physical laws, e.g. expanding at a constant rate, etc. I happen to believe that there is a scientific explanation to be found for the Big Bang - I doubt that God will have created a perfectly cohesive set of scientific laws until you get to a certain point where he says 'Oh, all right, you got me there'. My point was more directed at what seems to be 'any scientific theory, even if it can't explain all the evidence, is better than any other theory, even if it can't be disproved'. 2. The Big Bang and theism are in no way mutually exclusive - I myself have no problem that the Big Bang is probably the origin of the universe. As for predictions, there are various prophecies and suchlike in the old testament that were fulfilled in the new testament - though you will probably call doubt onto the reliability of these sources. 3. My understanding of quantum physics is not as good as I wish it were - what is different?
1Plasmon10yI agree Those are predictions about the past, not about the future. If you accept this as evidence, have you studied a large amount of other religions to see if they make similar claims? And if you had lived before the events of the new testament, would you have been able to distinguish the (supposedly true) "old testament" predictions from the (false) predictions made by thousands of other prophets of thousands other religions/sects? First, radio-active decay is uncaused (or so it appears, to the extent that quantum mechanics is correct). But that is not the point I was trying to make. In (special) relativity, "causality" basically refers to "all observers agree about light cones" or "if information flows from space-time point a to space-time point b, all observers will agree that the information went from a to b instead of vice versa", or simply "No one observes information travelling to the past". No notion of "all effects have a cause" or anything like that. In mechanics, say the description of a billiard table, (naive) causality seems obvious: why did this ball change its direction? Because it was hit by another ball! all the way back to a player hitting the first ball. But, there is the principle of least action [] (the related principle of least time [] is better known but describes light only). The principle of least action says that the trajectory of all those billiard balls is a stationary point of some functional (the action). In this view, there is nothing special about ball collisions. We ask "why did the ball change its direction (when it was hit by another ball or by the wall)" but we might as well ask "why didn't the ball change its direction (when it wasn't hit by another ball or by the wall)" and the first question is no more natural than the second. And the answer to both questions is the same: "because that path is a stationary point of
-4JohnH10y"Those are predictions about the past, not about the future." So Deuteronomy 28 predicted the scattering of the Jews to all nations of the earth. Then in Deuteronomy 30 it is predicted that they will be gathered from all nations of the earth (as well as if they are driven to the outmost parts of heaven they will be gathered from their). The Jews during the Roman times were scattered (documented fact) to all nations and have more recently begun to be gathered again to their ancient lands. (see Isaiah 11:11 as well). That was a prediction that up until this last century was about the future, and a prediction that the century before was considered impossible. Take from it what you will.
0Brilliand5yPeople who knew about that prophecy deliberately helped to fulfill it. That destroys much of its value as evidence.
5TheOtherDave10yOf course it's permitted. We couldn't prevent your interest in rationality even if we wanted to, which we don't. Nor could we prevent your reading the blog, even if we wanted to, which we don't. Whether theism is permitted is an uninteresting question. I can understand why you would resent that. Do you find it any more objectionable than your implication that we should all be theists -- that is, that we should "accept the explanation that God created the universe"? I agree that "as a rationalist" I should, in general, prefer to accept an explanation for an event that fits the facts rather than accept the existence of an event that has no explanation. I don't agree with your implication that I am actually forced to choose between "God created the universe" and "the Big Bang happened" (indeed, I know many theists who believe both), nor with your assertion that my "worldview" has no explanation for the Big Bang happening. Observing the results really ought to count for a lot... that's not something to casually throw away. I mean, think about it: if I propose a theory that predicts certain results to certain tests, and you go out and perform those tests and you don't observe those results, what conclusion would you come to about that theory?
1theflyingfrogfish10yI did not intend to imply anything of the sort - as I said above, I was more challenging the general attitude and querying my understanding of rationality than attempting to directly challenge anyone's theism or lack of same. I agree completely - I also believe that the Big Bang occurred. My point was more about why and how it happened, not if. I don't mean that performing the tests will not give visible results, I mean that performing the test leaves you with difficulty in reporting your findings. Large chunks of the Bible is about predicting either the future or what was the future when it was written. And most religions make a pretty big prediction of and event that will definitely happen to all of us.
4TheOtherDave10yOK, fair enough. For future reference, phrasings like "should you not, as an X, do Y?" will frequently be interpreted by native English speakers as implying that Xes should do Y. (In this case, that rationalists should accept "God created the universe" as an explanation.) Ah. Thanks for clarifying that. For my own part, I have no clear idea why or how the Big Bang happened. Neither am I very clear about why and how stars were formed, or why and how the state borders of Louisiana were established, or why and how the Connecticut state constitution was ratified. So I suppose you could say that I have a "worldview" that has no explanation for these things. It's hard to know for sure, since I'm not quite sure what "my worldview" refers to. I certainly believe that there is an explanation for how and why those things happened (several explanations, actually), if that clarifies anything. But in none of these cases does my ignorance of why and how that thing happened strike me as particularly compelling evidence for anything particularly significant, and it certainly doesn't seem to be evidence that God created the universe, or the stars, or the state of Louisiana, or the Connecticut state constitution. Now you've just confused me. Can you describe more concretely the test you have in mind, and what I should expect to experience after performing that test if God created the universe, and what I should expect to experience if the universe came into being some other way? Again, I'm unsure what you mean. Can you be clearer about what event most religions predict that will definitely happen, and how that prediction serves as evidence about how and why the universe came into being?
2theflyingfrogfish10yThe test I was referring to was dying - if the afterlife is as a religion says it is, then it can probably be accepted that the rest of the religion's doctrine is correct - at least the essentials. Or if not, you could ask the Supreme Being what IS correct. Conversely, if there is no afterlife, then if can be accepted that the religion is incorrect. Obviously this does not apply to all religions, but server the purpose here, I believe.
5TheOtherDave10yAh, I see. Thanks for clarifying. Sure, I agree: if, upon my death, I find myself in an afterlife consistent with religion X's teachings about the afterlife, and/or able to ask questions of some entity who claims to be the Supreme Being, I should update my beliefs about the likelihood of such an afterlife/Being.
3pnrjulius9yBut of course, we're pretty sure this won't happen. Indeed, let's consider two alternatives: 1. Afterlife exists, but God set it up so that you can't report back because... uh... I'll get back to you? 2. Afterlife doesn't exist, which is why you can't report back (there's nothing to report back from). In more explicitly Bayesian terms, which is larger: P(~report|afterlife) or P(~report|~afterlife)? Pretty clearly the latter, right? So the lack of reports is therefore evidence against an afterlife. (Maybe not conclusive evidence, but evidence.)
1TheOtherDave9yAgreed that this is evidence against an afterlife.
0[anonymous]10yTheflyingfrogfish means "everybody's going to die;" and he doesn't view that as the end of each person's ability to sense and evaluate. I would recommend to TFFF, on the subject of what religion claims regarding provability, to read this Yudkowsky post [].

I find most of this article extremely enlightening on the foundation of many problems with modern life. I also, however, have issues with your examples concerning government and other semantic stop signs. Liberal democracy is not necessarily a stop sign. It is easily countered by asking what that has to do with anything, as no current country in the world has a true democracy. They have republics due to the sheer size of countries rendering direct democracy pointless. Also, governments are reliant on the intelligence of their leaders and on those who ... (read more)

4Swimmer96310yIf only everyone had the same definition of what is right and what is wrong...
2Nornagest10yRestricting "true democracy" to direct democracy essentially renders the category meaningless.
0pnrjulius9yHow does it help us to put a stop sign there, and not say slightly before or slightly after? Indeed, what is "there" exactly (the proposition "we ought to do what is right" strikes me as obviously true, but also sort of trivial).

So the logical conclusion is that there is no beginning to time. *nods*

I don't think Stop Sign is the best metaphor here.

People like God as an answer because they dislike uncertainty and thinking. It's useful precisely because it predicts nothing, but explains everything. "God did it" acts as a Finish Line more than a Stop Sign. It says the race is done, and grants license to stop running.

Am I the only one who thinks of those Family Circus cartoons with the ghostly "Not Me" and "I Dunno" anytime someone says that "God did it"?

You know, that First Cause problem really is kinda a big one. Maybe we should be working on that?

Okay, there needs to be a semantic stop-sign after "But why doesn't s/he like me?" taught to all children in middle school.

Why wouldn't you automatically ask, "Where did God come from?"

I asked that a long time ago, in Sunday School. I don't think anyone has a good answer to that (at the time I came up with a recursive answer that relied on time travel; I have not yet found a better answer).

Speaking for a moment as a discourse analyst rather than a philosopher, I would like to point out that much talk is social action rather than reasoning or argument, and what is said is rarely all, or even most, of what is meant. Does anyone here know of any empirical discourse research into the actual linguistic uses of semantic "stopsigns" in conversational practice?

0imaginaryphiend8yTelms, it seems you are looking to tread in the path of the logical positivists where they sought to sort this out within a context of early Wittgenstein. Taken to the logical extreme with regard to a logical epistemic foundationalism, they tend to be generally dismissed, but in the context of an semantics relavant to general, meaningful discourse, i thought they tended to make a lot of good sense. Ironically, i keep going back to positivism. The irony being in the potential paradoxes of me seeing my self as essentially an epistemic nihilist. lol... I see it all metaphorically as an epistemology modelled visually to appear as ever expanding circles of reasoning, looking like an outward moving psychedelic spiral. If we try to deconstruct that psychedelic, perpetually moving spiral, we reach further and further towards an propositional foundation, but find we can only approach it as an infinitessimal, proposed, but not actually realizable, absolute beginning.
7telms7yMmm, that's not really where I'm coming from. There is an aggressively empirical research tradition in applied linguistics called "conversation analysis", which analyzes how language is actually used in real-world interaction. The raw data is actual recordings, usually with video so that the physical embodiment of the interaction and the gestures and facial expressions can be captured. The data is transcribed frame-by-frame at 1/30th of a second intervals, and includes gesture as well as vocal non-words (uh-huh, um, laugh, quavery voice, etc) to get a more complete picture of the actual construction of meaning in real time. So my question was actually an empirical one. It's one thing to guess at an analytical level that "God" might be a stop-signal in religious debates or in question chains involving children. But is the term really used that way? Has anyone got any unedited video recordings of such conversations that we could analyze? After making very many errors of my own based on expectation rather than actual data, I tend to be skeptical of any statement that says "language IS used in manner X", when that manner is not demonstrated in data. Language CAN be used in manner X, yes, but is that the normative use in actual practice? We don't know until we do the hard empirical work needed to find out.

Another thing that can act as a semantic stopsign is not just a word, but sometimes an image. What I mean is actually something quite similar to the emergence phenomenon.

For example, I am just learning the basics of Economics. I just came across the rules of Supply and Demand. Instead of analysing and observing how these play out in the real world, I am just content to form an image in my head of prices moving up and down (based on moving graphs and numbers), and deciding that I know all there is to know and not bother finding out more for myself.

I agree... (read more)

Q: Where did the great abyss, Ginnungagap, come from?

A: From the rock I chipped off the big boulder.

In reading these Sequences, I am noting that it is sometimes difficult to tell when you are building on an older body of work and when you are unaware of the older body of work and are independently deriving an equivalent concept. Semantic stopsigns is a particularly good example of this. Are you aware of the existence of another term for this: the thought-terminating cliché? (Sometimes thought-stopping cliché.) There is some fascinating literature on the subject of their use in cults, which may be directly applicable to understanding Dark Side techniq... (read more)

Good morning Eliezer,

Someone on another website posted a link to this interesting blog of yours. Of course you've always had an interesting perspective on things, esp. AI. But I will confine my comments to this thread.

As far as I can tell this entire thread assumes 'facts not in evidence'.

It assumes that existence requires a beginning

People who assume an entity like God created everything try to solve the problem of First cause by claiming God had no beginning.

Of course people searching for First Cause ignore the fact the the universe itself may have had no beginning.

First cause is an assumption like God. We have no evidence of either.

2gjm3yAre you saying that Eliezer's original post assumes that existence requires a beginning? (I can't see that it does.) Or that subsequent discussion in comments here assumes that? (Maybe some people here do, but who cares?)
1gridflash3yI'm impressed that someone actually responded this quickly to an 11 year old thread. Anyway gjm, in response to your question. No. Eliezer specifically avoided that issue. As I stated in my first sentence, my post was about the comments in the thread. People searching for First cause are like the people who believe in a God. They're both searching for something that has no beginning. If you don't care about the issue why respond?
2gjm3ySorry, it wasn't clear to me whether you were talking about the comments or about the OP; thanks for the clarification. So, you're talking about the comments rather than the original post, and specifically about Eliezer's comments. Except that I can't find any comments from Eliezer in this thread that show (to me) the slightest sign of assuming that existence requires a beginning. Nor, actually, do the other comments seem to me like they're making such an assumption. There are a lot of them and I might have missed some, but I'm not seeing a general trend of making that assumption. I think it would help if you were more specific. Could you point to a few specific things in this discussion that show the assumption in question being relied on? (As to "If you don't care about the issue, why respond?": whether I care depends on exactly what the issue is, and I'm not yet sure what it is. The specific point I was making there is that if you were just saying "one or two people in an 11-year-old discussion made a dubious assumption" then I don't see why you'd care about that or why anyone else should. If you're saying that Eliezer made that assumption, or that the whole LW community did, that surprises me more because it doesn't seem to me like the sort of assumption I would expect either Eliezer or the whole LW community to make.)

In a couple of Paul Graham's essays about neural network computing he suggests that Semantic Stopsigns are a necessary part of the design for general-purpose, parallel-computing intelligences to keep them from getting stuck in infinite loops attempting to solve infeasibly large problems.

The key is learning to recognize it as an "overflow error" flag and not a "this problem is solved" flag.  Internally they feel almost the same.