Slippery slopes are themselves a slippery concept. Imagine trying to explain them to an alien:

"Well, we right-thinking people are quite sure that the Holocaust happened, so banning Holocaust denial would shut up some crackpots and improve the discourse. But it's one step on the road to things like banning unpopular political positions or religions, and we right-thinking people oppose that, so we won't ban Holocaust denial."

And the alien might well respond: "But you could just ban Holocaust denial, but not ban unpopular political positions or religions. Then you right-thinking people get the thing you want, but not the thing you don't want."

This post is about some of the replies you might give the alien.

Abandoning the Power of Choice

This is the boring one without any philosophical insight that gets mentioned only for completeness' sake. In this reply, giving up a certain point risks losing the ability to decide whether or not to give up other points.

For example, if people gave up the right to privacy and allowed the government to monitor all phone calls, online communications, and public places, then if someone launched a military coup, it would be very difficult to resist them because there would be no way to secretly organize a rebellion. This is also brought up in arguments about gun control a lot.

I'm not sure this is properly thought of as a slippery slope argument at all. It seems to be a more straightforward "Don't give up useful tools for fighting tyranny" argument.

The Legend of Murder-Gandhi

Previously on Less Wrong's The Adventures of Murder-Gandhi: Gandhi is offered a pill that will turn him into an unstoppable murderer. He refuses to take it, because in his current incarnation as a pacifist, he doesn't want others to die, and he knows that would be a consequence of taking the pill. Even if we offered him $1 million to take the pill, his abhorrence of violence would lead him to refuse.

But suppose we offered Gandhi $1 million to take a different pill: one which would decrease his reluctance to murder by 1%. This sounds like a pretty good deal. Even a person with 1% less reluctance to murder than Gandhi is still pretty pacifist and not likely to go killing anybody. And he could donate the money to his favorite charity and perhaps save some lives. Gandhi accepts the offer.

Now we iterate the process: every time Gandhi takes the 1%-more-likely-to-murder-pill, we offer him another $1 million to take the same pill again.

Maybe original Gandhi, upon sober contemplation, would decide to accept $5 million to become 5% less reluctant to murder. Maybe 95% of his original pacifism is the only level at which he can be absolutely sure that he will still pursue his pacifist ideals.

Unfortunately, original Gandhi isn't the one making the choice of whether or not to take the 6th pill. 95%-Gandhi is. And 95% Gandhi doesn't care quite as much about pacifism as original Gandhi did. He still doesn't want to become a murderer, but it wouldn't be a disaster if he were just 90% as reluctant as original Gandhi, that stuck-up goody-goody.

What if there were a general principle that each Gandhi was comfortable with Gandhis 5% more murderous than himself, but no more? Original Gandhi would start taking the pills, hoping to get down to 95%, but 95%-Gandhi would start taking five more, hoping to get down to 90%, and so on until he's rampaging through the streets of Delhi, killing everything in sight.

Now we're tempted to say Gandhi shouldn't even take the first pill. But this also seems odd. Are we really saying Gandhi shouldn't take what's basically a free million dollars to turn himself into 99%-Gandhi, who might well be nearly indistinguishable in his actions from the original?

Maybe Gandhi's best option is to "fence off" an area of the slippery slope by establishing a Schelling point - an arbitrary point that takes on special value as a dividing line. If he can hold himself to the precommitment, he can maximize his winnings. For example, original Gandhi could swear a mighty oath to take only five pills - or if he didn't trust even his own legendary virtue, he could give all his most valuable possessions to a friend and tell the friend to destroy them if he took more than five pills. This would commit his future self to stick to the 95% boundary (even though that future self is itching to try to the same precommitment strategy to stick to its own 90% boundary).

Real slippery slopes will resemble this example if, each time we change the rules, we also end up changing our opinion about how the rules should be changed. For example, I think the Catholic Church may be working off a theory of "If we give up this traditional practice, people will lose respect for tradition and want to give up even more traditional practices, and so on."

Slippery Hyperbolic Discounting

One evening, I start playing Sid Meier's Civilization (IV, if you're wondering - V is terrible). I have work tomorrow, so I want to stop and go to sleep by midnight.

At midnight, I consider my alternatives. For the moment, I feel an urge to keep playing Civilization. But I know I'll be miserable tomorrow if I haven't gotten enough sleep. Being a hyperbolic discounter, I value the next ten minutes a lot, but after that the curve becomes pretty flat and maybe I don't value 12:20 much more than I value the next morning at work. Ten minutes' sleep here or there doesn't make any difference. So I say: "I will play Civilization for ten minutes - 'just one more turn' - and then I will go to bed."

Time passes. It is now 12:10. Still being a hyperbolic discounter, I value the next ten minutes a lot, and subsequent times much less. And so I say: I will play until 12:20, ten minutes sleep here or there not making much difference, and then sleep.

And so on until my empire bestrides the globe and the rising sun peeps through my windows.

This is pretty much the same process described above with Murder-Gandhi except that here the role of the value-changing pill is played by time and my own tendency to discount hyperbolically.

The solution is the same. If I consider the problem early in the evening, I can precommit to midnight as a nice round number that makes a good Schelling point. Then, when deciding whether or not to play after midnight, I can treat my decision not as "Midnight or 12:10" - because 12:10 will always win that particular race - but as "Midnight or abandoning the only credible Schelling point and probably playing all night", which will be sufficient to scare me into turning off the computer.

(if I consider the problem at 12:01, I may be able to precommit to 12:10 if I am especially good at precommitments, but it's not a very natural Schelling point and it might be easier to say something like "as soon as I finish this turn" or "as soon as I discover this technology").

Coalitions of Resistance

Suppose you are a Zoroastrian, along with 1% of the population. In fact, along with Zoroastrianism your country has fifty other small religions, each with 1% of the population. 49% of your countrymen are atheist, and hate religion with a passion.

You hear that the government is considering banning the Taoists, who comprise 1% of the population. You've never liked the Taoists, vile doubters of the light of Ahura Mazda that they are, so you go along with this. When you hear the government wants to ban the Sikhs and Jains, you take the same tack.

But now you are in the unfortunate situation described by Martin Niemoller:

First they came for the socialists, and I did not speak out, because I was not a socialist.
Then they came for the trade unionists, and I did not speak out, because I was not a trade unionist.
Then they came for the Jews, and I did not speak out, because I was not a Jew.
Then they came for me, but we had already abandoned the only defensible Schelling point

With the banned Taoists, Sikhs, and Jains no longer invested in the outcome, the 49% atheist population has enough clout to ban Zoroastrianism and anyone else they want to ban. The better strategy would have been to have all fifty-one small religions form a coalition to defend one another's right to exist. In this toy model, they could have done so in an ecumenial congress, or some other literal strategy meeting.

But in the real world, there aren't fifty-one well-delineated religions. There are billions of people, each with their own set of opinions to defend. It would be impractical for everyone to physically coordinate, so they have to rely on Schelling points.

In the original example with the alien, I cheated by using the phrase "right-thinking people". In reality, figuring out who qualifies to join the Right-Thinking People Club is half the battle, and everyone's likely to have a different opinion on it. So far, the practical solution to the coordination problem, the "only defensible Schelling point", has been to just have everyone agree to defend everyone else without worrying whether they're right-thinking or not, and this is easier than trying to coordinate room for exceptions like Holocaust deniers. Give up on the Holocaust deniers, and no one else can be sure what other Schelling point you've committed to, if any...

...unless they can. In parts of Europe, they've banned Holocaust denial for years and everyone's been totally okay with it. There are also a host of other well-respected exceptions to free speech, like shouting "fire" in a crowded theater. Presumably, these exemptions are protected by tradition, so that they have become new Schelling points there, or are else so obvious that everyone except Holocaust deniers is willing to allow a special Holocaust denial exception without worrying it will impact their own case.


Slippery slopes legitimately exist wherever a policy not only affects the world directly, but affects people's willingness or ability to oppose future policies. Slippery slopes can sometimes be avoided by establishing a "Schelling fence" - a Schelling point that the various interest groups involved - or yourself across different values and times - make a credible precommitment to defend.

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

There are also a host of other well-respected exceptions to free speech, like shouting "fire" in a crowded theater.

I wonder how many people would use this example nowadays if they knew that it comes from a WW1-era U.S. Supreme Court opinion upholding a ten year prison sentence for sedition against an anti-war activist -- whose crime was to distribute pamphlets arguing that military conscription is unconstitutional under the 13th Amendment, which prohibits "involuntary servitude."

(By the way, speaking of slippery slopes, the following year the same court upheld another ten year conviction for a speech whose content was carefully crafted to remain within the bounds of the sedition laws, but which was still judged to be illegal on the grounds of intent and indirect implication.)

As an interesting side note, the grimly ironic pinnacle of these American WW1-era sedition laws was the ten-year sentence (another one!) against a certain Robert Goldstein for defying the censorship of his film about the American Revolution. The film was deemed to be seditious on the theory that it incited hostility towards Britain, a war ally.

Not much relevant to the discussion but I never knew that George Orwell in his novel 1984 was making a reference to a person that existed when talking about Goldstein, the leader of the revolution in the novel (or just mere coincidence perhaps?)
1984 (like animal farm) is an analogue to the USSR - Goldstein may represent Leon Trotsky. I’m not entirely sure if Emmanuel Goldstein corresponds to anyone, though - we don’t know a lot about him except that propaganda considers him a traitor.

You may be correct, but the speaker in that case was probably referencing the Italian Hall Disaster, a genuine precedent. A rationalist can't just chose to ignore a data point like just because they don't like the way it's normally used.

5Ben Pace
No, but the point is that what appear to be good examples are often used when they are inappropriate, so you should be wary of agreeing more with someone's position just because they say their case is analogous.

I wonder how many people would use this example nowadays if they knew that it comes from a WW1-era U.S. Supreme Court opinion upholding a ten year prison sentence for sedition against an anti-war activist -- whose crime was to distribute pamphlets arguing that military conscription is unconstitutional under the 13th Amendment, which prohibits "involuntary servitude."

Wow. The 'fire' thing doesn't even fit well as an analogy in that context. Your country scares me!

My country?! I'm not American. I haven't even been to the U.S. in several years.

(That said, you're Australian, right? I've definitely seen stuff written on LW that is technically illegal in Victoria -- google for "serious contempt for, or revulsion or severe ridicule," with the quotes.)

The fire analogy must be understood in the context of the legal test of "clear and present danger" that the court was upholding. The theory is that just as the shouting "fire" in a theater creates a clear and present danger of a stampede in which people get hurt, so does the anti-war agitation create a clear and present danger of subverting the war effort, in this case by inciting resistance to conscription.

Of course, you may reply that this "clear and present danger" stuff can be stretched this way without limit, but that is the point. (By the way, in U.S. jurisprudence, this standard has in the meantime been superseded by a much clearer one of imminent lawless action, which, whatever its overall merits and faults, has shown in practice to provide for a very strong Schelling point.)

My country?! I'm not American. I haven't even been to the U.S. in several years.

My apologies. I assumed too much from your evident awareness of US internal politics.

I've definitely seen stuff written on LW that is technically illegal in Victoria -- google for "serious contempt for, or revulsion or severe ridicule," with the quotes.

Without at all claiming that examples don't exist it isn't technical illegality that concerns me but actual instances of punishments, along the lines of the two cases of 10 year incarceration.

As you have illustrated, things being legal or otherwise is not too important. Far more important is the inclination of the judge (and the surrounding political incentives) to punish any given behavior. Both of the cases you cite illustrate how easily laws can be ignored by introducing a fully general excuse along the lines of "except when it is really politically expedient to do so!"

Without at all claiming that examples don't exist it isn't technical illegality that concerns me but actual instances of punishments, along the lines of the two cases of 10 year incarceration.

The Victorian law is one of those laws whose real meaning is quite different from its formal one. Read literally, it is a sweeping law that criminalizes anyone who says anything unkind about any religion. (It's literally criminal to "behav[e] in a way that encourages... revulsion or ridicule of another person, because of the other person's... religion," or to "encourage[]... severe ridicule of... [a] class of persons, because of their... religion.") Yet in practice, of course, it's tacitly clear what kinds of people should watch their mouths or else fear prosecution under this law, and who can in turn safely ignore it. (Generally, throughout the Western world, with the U.S. as a lone exception, modern speech restrictions commonly have this form, providing effectively a broad mandate for ideological censorship.)

Note also that a century ago people were tougher and had less to lose, so that much more severe penalties were necessary to get them in line. Also, a criminal or ... (read more)

Edit: By the way, none of these Americans (as far as I know) actually served a full ten-year sentence. They were all pardoned in the early 1920s by Warren Harding under his policy of "return to normalcy."

I take pleasure in reading this. Sanity is out there somewhere!

Yes, but it certainly hasn't won out when you look at the way Woodrow Wilson and Warren Harding are remembered nowadays...

All else being equal (taking only the potential influence of these particular actions on popularity) that isn't encouraging. I don't know enough about US politics to know whether dislike of Harding is at all related to him having a nature such that he would make these (desirable) decisions.

How they're viewed has little to do with these decisions. Harding is seen as a weak President who was manipulated by those around him and who allowed corruption to flourish in his administration. The most famous example of this was The Teapot Dome Scandal. He may also, much like his successor Coolidge, have become associated with the boom that preceded the Great Depression.

Wilson is, unfairly I think, rewarded for leading America during WWI. This is in spite of the fact that he promised to keep America out of the war and managed to lose the peace that followed. Besides upholding sedition laws he did a number of other thoroughly un-liberal things.

How they're viewed has little to do with these decisions.

That is true, literally speaking. The point, however, is that the qualities that got Wilson celebrated and Harding forgotten or despised in the century since then were actually the same ones that led the former to fill jails with political prisoners and the latter to consider it unacceptable to preside over a country whose jails are filled with political prisoners. Besides, compared to the way things were done under Wilson, holding these petty corruption scandals as a heavy sin against Harding shows a complete lack of perspective.

But here we are getting into a historical and ideological discussion for which this forum is probably not a good place.

Harding also didn't join the League of Nations or forgive Germany's war debts, both of which seem likely to have increased the odds of WWII.
I heard somewhere that for the civil rights protesters of the 60s, going to jail had a very different meaning than it does for protester types today. Respectability was a much bigger deal then than now - those people dressed up in suits to protest and be arrested. At the time, the only marginally well-known precedent for being arrested for a good cause was Thoreau. I know people now who are proud of their protest-related arrest records, but that wasn't really a thing before the 60s.
Yes and no. In the last several decades there has certainly been a troubling trend of disregard for free speech in civil law - i.e. "cyberbullying" kinds of things, where people are now basically able to sue because their feelings got hurt. And you're absolutely correct about the cost of litigation having a strong multiplier effect on the "chillingness" of such suits. But in criminal law, things are definitely MUCH better today than they were in the WW1 era. For one thing we now have a de facto Shelling fence around punishment of speech which is explicitly political in nature (like protesting the draft).
That era, and those rulings, are pretty widely regarded as the low point of American first amendment jurisprudence. Things have rebounded significantly since then, and, as Vladimir points out, those cases have long since been superseded as binding case law. Edit: spelling.
I believe the theory was that he was promoting illegal activity, i.e., draft dodging.
I'm following posts up the chain to 2012 but I think I'll comment on this anyway: I don't think it's correct to say "its content was crafted to remain within the bounds of the laws, but was judged to be illegal on the grounds of intent and indirect implication". The law doesn't say "using X phrase is illegal, and if you don't use X phrase, you're in the clear". It actually uses the word "intent" to describe what is illegal; intent and implication is inherently a part of the law, and someone who is convicted based on intent and implication isn't convicted despite the law, he's convicted according to the law. Furthermore, even if it didn't say "intent", lots of laws are interpreted according to intent and implication. If a bank robbery law says you can't threaten to kill someone to get money, and the bank robber points a loaded gun at the bank teller, and the teller gives him money without him having to say a word, he doesn't get to say "the law requies a threat and I didn't say anything, so there's no threat"--he made an implicit threat.
Wow, man, awesome kudos for this link. The slope indeed slips the other way; once you allow the supreme court to make fire in the theatre analogies, you can start prosecuting people for 'sediction', which is in direct contradiction with the principle of free speech as necessary for the liberty of a free state.

Thinking about this article a bit more, you're missing a very important point. Namely, a "Schelling fence" is often of supreme importance even when there aren't multiple interest groups involved. (Nor even time- or value-inconsistent versions of the same party.)

This is because in most cases of conflict, pre-commitments are difficult to communicate. If you're maintaining some line of defence against an opponent and you're credibly pre-committed to defend even if the cost of defence is higher than the value of the prize fought over, you're safe against a rational opponent who perceives this pre-commitment correctly. However, if you have no such pre-commitment, the opponent has the incentive to mount an overwhelming attack even if the cost of the attack is higher than the prize -- since in that case you'll find it rational to retreat and avoid fighting.

The trouble is, communicating pre-commitment credibly is difficult, since it's always in your interest to assert pre-commitment to defend as credibly as possible, but secretly plan to withdraw if your bluff gets called. So this is where the Schelling points come into play: a conspicuous Schelling point helps create a meeting... (read more)

I know that discussing politics is Dangerous, but this is too tantalizing to ignore. Do you have any examples?

Try this thread.

Another approach to (or rather away from) slippery slopes is to see the entire slope as a single thing à la TDT. Gandhi, contemplating his willingness to make the trade to become 95% Gandhi, can also foresee that 95% Gandhi would make a similar trade to 90% Gandhi, and so on. So his first decision is acausally linked to the whole of the slope, and to decide to take one step is to decide to go all the way.

The concept predates explicit TDT and can be found in popular wisdom: how often have I heard "there is no just once" in fiction, whether a policeman asked to break the rules just this once, an alcoholic offered just one drink, etc. Kant's Categorical Imperative is similar.

Cf. the maxim "Everything you do is a decision about who you want to be", or the outside-view version, "The way a person does one thing is the way they do everything."

So his first decision is acausally linked to the whole of the slope, and to decide to take one step is to decide to go all the way.

(emphasis added)

No no no. His first decision is causally linked to the whole of the slope. If you draw out the DAG of causation, there's an arrow going right from "became 95% Gandhi" to "became 90% Gandhi", and an arrow going from "became 90% Gandhi" to "became 85% Gandhi", and so on (with some intermediate nodes depending on resolution).

I get the impression that "acausal" is an applause light here.

Or possibly a typo.
I doubt that, given that he also said TDT instead of CDT, etc.
I think "acausal" (or logical) linkage was used as a generalization of "causal" linkage, as 'dependences that TDT would take into account'.

You are correct.

8Nathan Rosquist
[de-jargoning for newcomers] TDT, CDT, ADT := models of Decision Theory DAG := Directed Acyclic Graph
Thanks for noting the connection to Kant; ADT's minimally-metaphysical yet deontological approach, i.e. its Kantian approach, is its hallmark and triumph. (TDT goes heavier on the metaphysics and is weaker for it.)

It's a funny sort of deontology that is justified by the agent's preferences over its consequences ...

That's true, it is a funny sort of deontology. Note that Kant actually claimed that he was not preferring one consequence over another, he was finding a self-contradiction in one consequence, and no self-contradiction in the other. That is, "you should not steal" because, at the end of the slippery slope, there is a self-contradiction, something like "if everybody ought to steal, what does theft even mean?". (I'm trying to give Kant a fair shake, though I think he's wrong.)
I'm really not very familiar with Kant, but I guess I always thought to frame it like, after an infinity of time certain norms will defeat themselves, and other norms will reinforce themselves, and if we are to support a self-defeating norm then that's like choosing to become counterfactual. Then endorsing a self-defeating norm seems metaphysically but not phenomenologically possible, and since Kant didn't care about metaphysics, it's straight up impossible. Or something, agh. But anyway it looks ADT-esque, except that Kant had a hard time with meta levels for some reason.
I don't think it's right to say that Kant didn't care about metaphysics. Given that his work on ethics was Metaphysics of Morality and that was based on the more-metaphysical Groundwork for the Metaphysics of Morality. Whether one can choose a norm that is inconsistent, is a question for the noumenal world, not the phenomenal world.

I'd like to mention that I had an entire family branch hacked off in the Holocaust, in fact have a great uncle still walking around with a number tattooed on his forearm, and have heard dozens of eye witness accounts of horrors I could scarce imagine. And I'm still not okay with Holocaust Denial laws, which do exist where I live.

In part, this is just my aversion to abandoning the Schelling point you mention; but lately, this is becoming more of an actual concern: My country is starting to legislate some more prohibitions on free speech, all of them targeting one side of the political spectrum, and one of the main arguments touted for such laws is "well, we're already banning Holocaust denial, nothing wrong came of that, right?".

The slope can be slippery indeed...

Given the discussion, strictly speaking the pill reduces Ghandi's reluctance to murder by 1 percentage point. Not 1%.

Glad I'm not the only person who noticed. Pedantry for the win! (Upvoted.)

A true pedant would insist on correct spelling of Gandhi's name.


Touché. I'll have to tear up my Pedantry Association membership card.

What is the difference?

Couldn't the alien argue that slippery slopes slip both ways? If we start allowing Holocaust denial, soon it will be legal to yell "fire!" in a theater.

You know... if I actually yelled "Fire!" in a theater I think I'd just end up with people looking at me like I was a dick.

The example when coined almost certainly was a reference to the famous Italian Hall Disaster in which it seems 73 people died because of someone falsely yelling "fire".

If you shout from the middle of the crowd - probably so. If you enter - visibly exhalirated - just to shout "Fire!", there is a risk.


Fire suppression systems have gotten better, we don't overcrowd theaters anymore, and we now use electricity for lighting instead of burning things, so fires are less common and not as dangerous.

This. This this this. I feel like the "fire in a crowded theater" example is a pretty painfully outdated one. If you imagine a more crowded theater made of dry wood with no sprinklers/fire extinguishers, it becomes a lot more reasonable to expect people to panic at the thought of a fire.
I think the prevalence of the "falsely shouting fire in theater" quote is a product of usage of that quote in an important freedom speech decision (that ultimately ruled against free speech). Thomblake suggests convincingly that the quote was referencing a then recent disaster, but you are correct that the circumstances in theaters today are so different that the quote lacks the same impact as when it was written. In other words, the same legal reasoning today would probably be supported by a more culturally relevant example. For what it is worth, the legal standard on liability for possible harm from speech is different (and more speech protective) than the decision in the Schenk case from which the quote originates.
Yes, but the danger of a false call of "Fire!" lies in the subsequent stampede (and, even if no one is hurt, the inevitable disruption of everyone's prior activity).
Wedrifid was saying that people aren't that likely to stampede or even evacuate if someone yells "fire" now. I think he's right. If I started shouting "fire" right now I think the people around me would think I'd snapped and would at the least ask what I was seeing that they weren't. (If I actually wanted people to listen to me I would start by asking "do you smell fire?" to the person next to me. Two people shouting fire is much less likely to be someone hallucinating.)
Real life, 2007 -- 100 dead. Real life, 2013 -- 242 dead.
Reality is more complicated than that. Grandparent is right about deaths from fire in many parts of the world. The following deals with the fire death rate from 1979 to 1992. -- "Fire Death Rate Trends: An International Perspective" For the US in particular, this trend has continued into the present. We leave as an exercise to the reader what exactly the flaw in your argument was.
(By convention there the reference is to 'grandparent', not 'parent'. Context is sufficient here to correct the meaning but in less clear circumstances it would be misleading.)
Single datapoints rarely matter with these kinds of things. The overall statistics strongly support the grandparent's point.
"These kinds of things" are public perception and, in particular, public perception of risks and what to be afraid of. For these kinds of things single datapoints matter a great deal. Prime example: 9/11. Public perception of risks notoriously does not care about statistics.
I see, I took you to be responding only to jkaufman's comment. Since rare dangers typically get more publicity than common dangers, we might even expect that under many circumstances public angst may increase as the actual danger decreases (due to the remaining incidents getting overreported).
Nowadays you might also get charged with making terrorist threats. Speaking of nowadays, screaming "It's a bomb, it's going to blow" is probably going to be much more effective at creating stampedes...
Or, for that matter, held without charge or trial. Or perhaps "Look! It's Justin Bieber and he's taking his shirt off!".
Any theatre offering access to the public should be able to evacuate all patrons safely in the event of a real fire, thus someone falsely yelling "fire!" simply inconveniences people and should be banned by the owner. There is no need to get the police involved and have this ridiculous example as legitimisation on limits on free speech in a public arena. There are limits that should be upheld, but not necessarily by police (e.g. someone being slandered could sue rather than have the slanderer arrested.) Incitement to violence may be banned as a pragmatic measure, but that would be my Schelling point (and even then I'm not 100% sure...)

Being able to evacuate all patrons with sufficient safety is not necessarily the same thing as being able to evacuate all patrons with perfect safety.

A theater than can protect its patrons against a real fire doesn't necessarily equal a theater that can protect its patrons against a false fire. Nor do I find it more reasonable to ban businesses that are incapable of protecting all patrons against the repercussions of all types of lying. I find it far more reasonable, and much of a "Schelling point" to ban spreading knowingly false information. That's effectively a matter of degree -- you're just saying that the punishment should be a fine instead of a prison sentence. It's not really a difference in kind.
That is a patently false statement. And thus the slippery slope becomes a teflon cliff. One is criminal and the other is civil? One is a dispute between two (equal under the law) individuals in front of a (supposedly) impartial judge and the other is the state versus an individual? Quite a lot of difference really.
How do you know this? In the case of a true fire, fire extinguishers may activate, fire exits might automatically open, or perhaps the theater's army of flying robots would come and carry everyone to safety. Or perhaps the theater is designed from such materials that it can't catch fire. Excuse me? The concept of "knowing lie" isn't a good enough Schelling Point for you? I'm in favour of banning the act of falsely shouting "Fire" in a theater, for the exact same reason that I'm in favour of banning the act of giving poison to other people while calling it "orange juice" I don't think you'd have any opposition to treating the latter an act of murder, why are you against treating the former likewise an act of murder? This is a legalism, a difference in labelling, nothing else. Again mere legalism. In practical terms, it's the same, at least from the point of view of the accused -- you get accused for something, someone is accusing you, if you're found guilty you get punished. Simple as that.
I believe there actually is a salient distinction between the two. Namely, whereas the harm in the poisoning is the act of placing the poison in the drink and giving it to another person, the act in shouting "Fire!" is in the believing that there is no fire and the harm resulting from the shout. While the distinction would not amount to much for an all-knowing god, in reality it would probably prove difficult to establish what the person shouting expected the result to be, and to determine whether he had believed there was a fire or not. Punishing alarmist statements is indeed a potential slippery slope - if someone shouts "Earthquake!" in the middle of a crowded street and pedestrians panic and someone ends up in the traffic, shall we prosecute this shouter too? Or what if someone proclaims that the nation is under imminent threat of nuclear attack and this results in massive economic disruption as people try to hoard survival-type goods? The problem here is that it is difficult for a court to assess whether the person actually believed what they said, and whether they anticipated (perhaps a priori rationally) that there would be no negative consequences of their proclamation. It may also interest you to know that the original "shouting fire in a crowded theater" claim was part of a judicial ruling written by Oliver Wendell Holmes, Jr. in which the Supreme court unanimously ruled that it was a violation of the Espionage Act of 1917, to distribute flyers opposing the draft during World War I, which I think is a suitable example of its slippery slope potential. My general impression is that while there have been real instances of deaths caused by such shouts, most of these occurred before recent times, although I am not too confident in this.
Courts do this all the time. Further, there's a well developed set of doctrines about when a person is held responsible for consequences that the person did not actually anticipate. More generally, the quote about falsely shouting fire is not intended to support a particular test for the limits of freedom of speech. Rather, the quote makes the point that speech is not unlimited.I can agree that Schneck's conviction should have been overturned without believing that no pure speech can have legal consequences. As an aside, the current test in the United States is imminent lawless action, a much more pro-speech standard. Nonetheless, a knowing false statement likely to lead to multiple injuries is almost certainly punishable by the criminal law.
Thanks, that helps update my knowledge of the current standards. I certainly agree that what you say is reasonable, however I do not think that a potential for a slippery slope effect can be eliminated through carefully formulated test or doctrines (merely substantially reduced). For example, it is conceivable that in emergencies the fact that we now accept some restrictions on free speech will make it much easier to accept further restrictions and thus make unreasonable restriction more likely. Whether that is a reasonable compromise depends on the actual danger of "pure speech" causing negative consequences, on alternative ways of mitigating them, and on the actual degree of the slipperiness of the slope. Personally, based on my current knowledge, I would prefer to pursue various alternatives before resorting to criminal legal action, however that's dependent on actual facts and has little to do with the main topic being discussed.
This distinction breaks down very quickly. Consider Hustler v. Falwell, which limited the scope of civil remedies because enforcement of those remedies violated freedom of speech.
Or, more recently, Phelps v. Snyder, where the Court overturned a jury verdict for intentional infliction of emotional distress based on Fred Phelps's First Amendment protected funeral protest.
Limiting the scope of a civil remedy is somewhat removed from the distinctions between civil and criminal, no?
No. If you really believe the public/private distinction is solid, then the First Amendment has nothing to do with private (i.e. civil) disputes. So Hustler should have come out the other way. In general, American law has really struggled with the public/private distinction. I would say this is because the distinction is not rigorously meaningful - although I doubt most judges would frame it quite that way. Regardless of the framing, American law recognizes that the situation of "two equal individuals before an impartial judge" includes the fact that the judge is an arm of government, exercising government power. What that means in practice is less clear. Compare Shelley v. Kraemer ("That the action of state courts and judicial officers in their official capacities is to be regarded as action of the State within the meaning of the Fourteenth Amendment is a proposition which has long been established by decisions of this Court.") with Schiavo v. Schiavo (holding that a state judge ordering the removal of a feeding tube is not state action).
I've heard it argued this way. I've also started an argument on some other forum, to test some ideas, with regards to the 'harm principle' (in law). Most of the laws of free countries do conform to harm principle, even though it is not in constitution and a few laws don't (usually for quite bad reason and with very ineffective enforcement). But if you try to bring it up you get slippery slope arguments along the lines of "if we start legalizing harmless behaviours we'd have to legalize driving on other side of the road". Try talking to authoritarians and you can see them argue the slippery slopes the other way. It must be noted though that there's much longer way to slip in the direction of disallowing speech, than in the direction of allowing speech.

Very good post! I feel like I understand the concept of slippery slopes much better now. The breakdown into different categories of "slippery slope argument", and cases where they would be valid, is one I find especially useful.

It does make me wonder whether the existence of "slippery slopes" is based on the flawed design of human brains, i.e. hyperbolic discounting, and just the general tendency to alter beliefs for self-consistency or because they sound/feel better. (An example of this would be a real-life version of Gandhi's pill, where starting out with a small not-very-bad act which someone can justify within their moral system, like stealing candy, could lead someone to be more comfortable with stealing in general and end up stealing much bigger things.) Would a non-human 'perfectly rational' alien, one that didn't hyperbolically discount and had no tendency to automatically update their beliefs/moral system so that their past actions weren't evil, still need to worry about slippery slope arguments?

Not from hyperbolic discounting or value drift. Maybe from other sources, like the coalition argument presented by Yvain.
You could still hit them with large rewards for making themselves less rational, and thus recreate the slippery slope argument along that axis.
Or large rewards for changing their utility function.

I think you are dismissing the abandoning the power of choice argument unfairly. I'll try to give it a better formulation than you did.

There is a decision to make. The question is not just "A or B?" It is "How shall this type of choice be made?" and "Who shall make this type of choice?"

Changing how decisions of this type are made and who makes them are dangerous. "Holocaust denial is verboten," implies "There are authorities who decide what beliefs are acceptable." and "Individuals cannot be trusted to seriously consider all alternatives for belief." The second is true, because by removing the ability to reject the holocaust removes the ability to consider if the histories are reliable.

For the Civilization example. "I should decide on when to go to sleep based on a prior rule I accepted." is a different principle than "I should decide on when to go to sleep based on my current enjoyment of whatever I am doing now."

Admittedly, I rarely make arguments of the slippery slope form. I prefer to point out the toxic principles being accepted.

A thought inspired by an exchange upthread with Vladimir, but more general than what we were immediately discussing:

It seems that (e.g., political) extremism may be a result of a certain perception of where the Schelling points are and an application of meta-level thinking to the slope, even where basic preferences aren't changed much. Suppose that it is 11:30 and you are playing Civ. You want to play until 11:40 and then stop, but you know that if you do you'll want to play until 11:50 and then stop, etc.

If you think everything is a nice round number, you can just say "I'm going to stop at 11:40" and then do it. This is great if you're not lying to yourself, which you probably are. If you think :00s and :30s are nice round numbers, you can compare how you feel about stopping now, 10 minutes earlier than your current ideal, to stopping at midnight, 20 minutes after it. Trivially so for whatever other round numbers you might see, &c.

But suppose to yourself: "but why should my night be dictated by what I think at 11:30?" Well, because that's who you are at this point in time, and thus whose desires reason consists in fulfilling, obviously. But: you are in a s... (read more)

Meh. I just say "20 more minutes" and set up an alarm timer. Pomodoro technique FTW. (And when the alarm goes off, if I still want to do what I'm doing, I halve the time and restart the timer. This way, the total time I waste can't exceed twice the original time I used, in this case 40 minutes.)

In parts of Europe, they've banned Holocaust denial for years and everyone's been totally okay with it.

You should look closer, they're banning speech left and right all across Europe under variety of trivial reasons.

And let's not forget the speech being banned under American influence.

That does not follow. As Giles suggests, the UK does not have a law against Holocaust denial, and the UK is the relevant country for libel tourism, which is the only phenomenon you linked to. Furthermore, the UK standards for libel are very old, and there has been some success in the UK in reducing them. This is partially in response to pressure from the US, which no longer extradites for speech violations that are not illegal in the US, in direct response to libel tourism. I'd link to a reference with details on all this, but you already did.
I thought that libel tourism was basically a UK thing, which doesn't have a law against Holocaust denial. I guess I'm saying: * Different nations can have different Schelling fences * Some nations can establish Schelling fences and others can be on slippery slopes. The strategy of the nations with Schelling fences would depend on the extent to which they care about the other nations.

like shouting "fire" in a crowded theater.

Pedantic-lawyer Tim says the exception is for falsely shouting fire in a crowded theater.

Edit: But the general point is right. Lots of law is some implementation of: "Better decided clearly now than decided correctly some vague time in the future."


I am surprised that pedantic-lawyer Tim does not also point out the origin of that phrase.

Specifically, the case Schenck v United States in which "falsely shouting fire in a crowded theater" was used as an analogy for protesting the draft.

I have therefore always been troubled by the origin of the phrase, even though it feels like a reasonable exception to make.

Edit: And now I see that Vladimir_M has already posted this.

Reminds me of the "best defence against libel is truth."
I doubt it. It doesn't seem likely that for some reason declarations that happen to be true will be the optimal declarations to make.
Sorry I may have been unclear, the quote means that when accused of libel a defendant can be exonerated if they prove the statement to be true, however damaging it might have been shown to be by the prosecution. So to count as libellous a statement must be both false and have done provable damage to reputation. As such the easiest way to prove a statement is not libellous is to show it is true. Rather than, how I think you interpreted it as meaning, that when libelled one should tell the truth in response.
Unfortunately, at least in certain jurisdictions, the burden is on you to prove that what you say is true.
A good burden requirement would be more qualified than either of those approaches. I don't want to have to prove I do not spend my nights fantasizing about eating babies in the privacy of my own thoughts.
"I'm not saying my fellow candidate is a terrorist, but ladies and gentlemen, has he presented any conclusive evidence that he is not?"
"I just want to make him deny it."
And I would argue you should have the right to make that statement.
Ahh, that makes sense! That's right.

Great post. Very clear and concise as usual. I recently read an interesting article by Eugene Volokh on slippery slopes focused specifically on gay marriage, which you can find in pdf form here. (If you don't like pdf's, the title is "Same-Sex Marriage and Slippery Slopes.") Interestingly, he also discusses the US's first amendment as something like a Schelling point, though he doesn't use the same terminology.

In parts of Europe, they've banned Holocaust denial for years and everyone's been totally okay with it. There are also a host of other well-respected exceptions to free speech, like shouting "fire" in a crowded theater

From what Volokh says in the article, it seems that many countries aside from the US don't have as strong of protections for freedom of speech. E.g., in Canada and Sweden ministers have been prosecuted for comments condemning homosexuality.

Volokh actually also has an older, more general article, "Mechanisms of the Slippery Slope. It's well worth reading.
To include links, use [link text](http://url.example). Click on the “Show help” button under the edit box for more info about markdown.
Thanks. I always screw that one up, but I usually notice and correct it. I wish LW would just accept HTML.
Well, the posts do. (It's the HTML button in the toolbar in the WYIWYG edit box.)
The "Political Momentum Slippery Slopes" section in that link seems to make some features of US political discourse a lot clearer to me.

I am new to the website. So new that this is my first comment and I didn't even particularly want to sign up. I found it interesting having just come from reading Eleizer's post about 0 and 1 not being probabilities to here I immediately had a question form in my mind.

How certain of a thing do we have to be in order to prescribe that the state be able to end someone's life for their speaking it?

There are several points in this question that require some unpacking. The most prominent being about the state being able to end someone's life. Though people are generally self-seeking and somewhat rational in being so (so taking a prison sentence over engaging in a shootout with police), any action we ask the state to police is then done so with the threat that the state has total authority to use lethal force in the case of serious non-compliance. If we grant that the state has the right to put people in prison for an action, we grant that the state has the right to kill someone for refusing to comply with the states right to put them in prison.

So how certain of a thing do we have to be in order to grant the state the right to kill someone who questions it? Can any... (read more)

Firstly, welcome!

Beyond a certain probability (say 99.9% confidence that a story is true in its generality, even if one is less sure of some of the specifics), it seems to me that the truthfulness of the story is no longer the main consideration in whether to instigate such a law. In that case I would be more interested in how such a law would alter the incentives of society and the knock-on effects of such. Not making a decision due to imperfect information can often be a mistake.

The point about granting the state authority to end a life for breaking any law isn't something I'd thought about before and is a very interesting one. I feel like it possibly proves too much if relied on too heavily to make decisions about which laws to implement - I can apply the same argument to speeding but that isn't a strong argument against speeding laws. The strength of the argument depends on how often one would expect such a death to occur. Assuming this list is typical, the argument is much stronger in e.g. the US than in Europe (where it can be all but ignored).

I do think that Holocaust denial laws are problematic if you're not allowed to say anything even slightly less sev... (read more)

Thank you for your response. I don't think you are misunderstanding me. I can't present certain evidence or even write too extensively on this topic (which is in part why I phrased everything as a non-specific question in the first place) as there are two publicly known cases where people in the country I live have ended up facing court for writing in a public forum on the internet - one was sent to prison for being the owner of a revisionist website that was hosted in the U.S. In a more general discussion of these examples (as specifics are legally tenuous), were there evidence to the contrary regarding the use of gas chambers or the number of the dead, what is the probability that the discoverer of such would be chilled in their speech from reporting this evidence, compared to were these convictions not in place? I can think of an experiment to test this - how many people question these facts where the questions themselves are not illegal (the US for example) compared to where they are (Germany)? The second question is that what is the probability the evidence will be judged on its own merits when people have been convicted for presenting it, compared to when people have not? I can't think of any way to quantitatively evaluate this since we live in a world where people /have/ been convicted for presenting evidence to the contrary. Leaving aside evaluations of truthfulness, or even the specific topic at hand, what is the effect on encouraging rationalism to society at large when some conclusions are a priori prohibited no matter what the evidence may find? Eleizer writes in at least one (though it feels like many) of the introductory guides of the dangers of making conclusions before evaluating the evidence - and that this leads humans to reject evidence that doesn't fit the conclusion and seek out evidence that does. Doesn't using legal systems to prosecute conclusions set up a society-wide bias that forces true evidence to be rejected?
In the most general sense, any law which impinges on free speech has the potential to be detrimental to accuracy of beliefs. For example, if I make a defamatory claim about someone and they take me to court, the onus is on me to prove that what I said was true (at least in the UK). This will discourage me from making a claim that I believe to be true but don't have strong evidence for and so I cannot publish some true information. In the US the burden of proof would be more on the person who I defamed to show that what I said was false (I'm not a legal expert, I got this from an episode of The Good Wife!). This is a lesser brake on free speech and allows me to say things which I am confident are true, even if my proof would be insufficient for a UK court. However, there is a flip side. Completely free speech is not beneficial for truth seeking unless all members of the society can be trusted to communicate in good faith. Without any defamation laws everyone can say whatever they like about anyone else and no-one knows what to believe. I can imagine circumstances where if the burden of proof is overly on the defamed then people can make up things which are very hard to disprove and again the truth can suffer. Another example would be that hate-speech laws discourage racism but also make it more difficult for people to discuss the possibility of differences between races. So the choice of where to draw the line on free speech includes a play-off between allowing accurate evidence to be presented and preventing bad faith communication. In the case of Holocaust denial I don't think it would be too controversial to suggest that most revisionist theories constitute bad faith communication (I'll be honest, I haven't looked at any myself). My personal preference on this wouldn't be to ban holocaust denial, as the social norms where I am from are sufficiently strong that they constitute enough of a barrier to Holocaust denial entering the main stream but I can certain
Thank you. I think you are right. I did not sit down and think through this idea before proposing it. Such an experiment would not just be useless, it would probably be worse than useless. I think it would give meaningless data that could easily be confused for meaningful. I appreciate the correction. I think that this is wise, but perhaps it just comforms to my own opinions. You made not about libel and slander - both of which I agree with your positions on. The difference, as I can fathom it, between being able to sue someone for saying something that is untrue (whomever the burden of proof is on), and making questioning parts of an event illegal, is in the quest for truth. In the case of a legally protected event, you can never prove whether or not what you are saying is true. In the particular case we are speaking of, even to present true evidence that is 1/googleplex against being false is itself a crime, and the evidence inadmissible in your defense. I can't comprehend how rationality can hope to propagate in an environment that values social nicety over truth.
The point I was trying to make was that social nicety is a prerequisite for truth, or if not social nicety per se, at least good faith communication. In general I'd agree that society values nicety more highly than is strictly healthy. To propagate rationality in such circumstances you focus on the battles that you can win. I'm not optimistic about rationality propagating fast but I don't think focusing on extreme and emotionally charged hypotheticals will get us there any faster. Maybe give it another 30 years and we'll see where we are! (Of course if this is less hypothetical then this discussion would be a very different one.)

>Give up on the Holocaust deniers, and no one else can be sure what other Schelling point you've committed to, if any...
>...unless they can. In parts of Europe, they've banned Holocaust denial for years and everyone's been totally okay with it.

I'd say this has aged poorly, except it was evident at the time this was written as well, if not quite as much.  Europeans have managed to slippery-slope from banning Holocaust denial to banning political speech that could be painted as related to an oppressed group in any way, so you get people prosecuted for objecting to immigration or to trans politics.

As a European, there aren't any problems with free speech. I've read this twice already from anonymous people online some years ago, and when I asked them for the source, I never got a response. Do you have a source that in some European country, you can get prosecuted for objecting to immigration?
It's not hard to google this. (paywall, but the synopsis is enough to prove that they exist) (ditto) Here's a trans politics one:
The first two are identical. The third is a xenophobic portal and the fourth is, in fact, transphobic (not merely "against transgender politics"). What I had in mind was speaking up against immigration and also not being hate speech.
That's like saying "I wanted an example of someone being put in jail for just marijuana possession, and also not being arrested." "Hate speech" is the excuse used by Europe to prosecute people for objecting to immigration. It's something you guys made up in order to stop free speech.
Obviously green_leaf's claim is not merely that the people being prosecuted for objecting to immigration are being accused of hate speech, but that they are doing things that a reasonable person would consider hate speech rather than merely "objecting to immigration" and that the latter, if done not-hatefully, doesn't get anyone prosecuted. The paywalled article at your first and second links proves that some people are objecting to immigration and getting prosecuted for hate speech, but (at least on this side of the paywall) doesn't make it clear exactly what they're saying, hence doesn't let us tell whether they're just "objecting to immigration" as opposed to, say, calling immigrants vermin or suggesting that they should be thrown into the sea or something. The third link gives several purported examples but again doesn't do much to clarify exactly what was said in each case. * Markos Vasilakis: I don't have the text of his sermon or anything, but the Gatestone Institute article about it links to a Greek-language article that quotes some fragments and (from Google's translation, whose accuracy I can't vouch for) it doesn't seem as if the Gatestone Institute is being very honest when e.g. it says that he merely "express[ed] accurately the percentages of refugees and illegal migrants entering the country". Also, was he actually prosecuted? The GI article says that the government "asked the district attorney to prosecute" Vasilakis, but that doesn't seem like it's the same thing. * British man arrested "for tweeting about a conversation he had with a Muslim woman about terrorism": he tweeted that he had "confronted" a Muslim woman and demanded that she "explain" the terrorist attacks that had recently happened in Brussels, and followed it up by saying "Britain is for Britain's" (presumably he meant "Britons") and "Who cares if I insulted some towelhead?". Whatever you might think about that, I don't think "objecting to immigration" is a reasonable description o
If you think it's possible to object to immigration in Europe in a manner similar to other political views, in a way which can't easily be painted as 'hate speech', I'd like to know some examples of things which can be said, without watering down objections to immigration.
Well, since your earlier claim was that people are being prosecuted "for objecting to immigration" and so far as I can tell none of the people in the FairUS article were actually prosecuted, one answer to that would be "all the things the people you used as examples said". But here are a few other concrete examples of people saying anti-immigration things without being prosecuted. The examples I've given are of politicians because they tend to be the people who say these things very visibly and get reported on. They're mostly from the UK because that's where I am and English is the language I know best. * Nigel Farage: "I think that if you have an immigration rate that runs too high, with people who speak different languages, have a totally different culture, and in particular attitudes towards women and their place in society, then I think you're storing up for yourself an enormous problem which then leads to cities with divided communities." Complains that after a period with ~30k immigrants per year "Blair opened the doors, the Tories have kept them open, and that's why this has now become a big political issue". * Suella Braverman (the UK's "Home Secretary"): "The British people deserve to know which party is serious about stopping the invasion on our southern coast and which party is not. Some 40,000 people have arrived on the south coast this year alone, many of them facilitated by criminal gangs, some of them actual members of criminal gangs." She got some criticism for this, unsurprisingly, but equally unsurprisingly she did not face prosecution or official censure by the House of Commons or losing her job near the top of the UK government. * Richard Tice of the "Reform Party" talking about a "betrayal of Brexit" represented by current levels of immigration to the UK. "They've betrayed us on taking back control of lawfu
That's not what I asked for. What I asked for was And there's a reason why. "Hate speech" censorship works by having vague, broad, rules that let the governments censor anything they want related to the subject. The fact that the government hasn't actually chosen to use this unlimited discretion all the time, yet, doesn't mean that they don't have it. Are you claiming that not only were those statements not prosecuted as hate speech, but they can't be prosecuted as hate speech? If he wanted to claim that such immigrants are merely disproportionately criminals, thus not using the literally false word "all", would he be permitted to do that? Would it matter if they actually are disproportionately criminals?
This seems like obvious goalpost-moving. What you originally said is that people are being prosecuted for objecting to immigration. I think you should provide some examples of people being prosecuted for objecting to immigration. I looked at the things you cited, and not one of them seems to be that. But now you're demanding examples of anti-immigration sentiments that can't possibly be painted as "hate speech", that couldn't possibly result in prosecution. That seems like an obviously impossible demand; if I give what I claim is an example, you can just say "nah, they'd totally call that hate speech" or "so it didn't get prosecuted this time, but it totally could have". Do you have examples of people being prosecuted just for objecting to immigration? Or did you claim that that's happening without having any evidence? Reminder: what you said was: I think you should (1) show us some compelling examples where the thing you claimed happens has actually happened, preferably enough and clear enough to show that it's an actual pattern and an isolated weird one-off; or (2) explain what good reason you have for believing that "you get people prosecuted for objecting to immigration" that doesn't enable you to provide a compelling list of examples; or (3) admit that you made that claim without actual grounds for thinking it to be true. As for the weaker claim you are making now: it may well be true that laws around "hate speech" are too vague and too broad. However, I invite you to show me a law valid in (let's say) the UK, France, or Germany, that makes it a criminal offence for someone to say, for instance, "In my opinion, societies with high immigration rates are less stable, and I therefore think that my country should reduce its immigration rate by a factor of 10" or "Immigration is expensive and many immigrants cost the country they move to more than any economic benefit they provide, and I therefore think my country should only permit immigration by people with
They are. A subset of them prosecuted is still people being prosecuted. Using "hate speech" as an excuse to prosecute them doesn't mean it doesn't count, since "hate speech" is so broad it can apply to anyone. The original claim is directly related to the "weaker" claim, because if "hate speech" is broad, prosecuting them for it is just an excuse for prosecuting them for something else; they're really being prosecuted because they don't like immigration. But that waters down the previous statement. It no longer claims that immigrants have high crime rates, and uses the vague term "stable". If you were asked to clarify and you said "by stable I mean low crime, since immigrants are disproportionately criminals", that's still hate speech. (And of course the government could say "he said 'stable', but that's a dogwhistle for 'high crime'. Illegal hate speech.")
Suppose I say "these days there's so much racism that you can get arrested for being black". You express some slight skepticism about this. I present five examples of black people getting arrested. You point out that they were arrested for assault, tax fraud, libel, and theft. I say "A subset of them arrested is still people being arrested". You would, I take it, not be impressed. If you say that people are being prosecuted for objecting to immigration, you need to show us some examples where that is what they are being prosecuted for. (Though ... a first step would be showing us some people being prosecuted at all, for anything, which so far you haven't bothered to do.) No, it's completely separate from the previous statement. The two anti-immigration propositions I mentioned are not intended to be equivalents of any other anti-immigration things that have been discussed elsewhere in the thread. They are examples of ways in which one could "object to immigration" while facing no, zero, none, zilch, zip, nada, risk of prosecution for it. Which (if I am right about the risk being negligible) shows that it isn't true that objecting to immigration puts you at risk of prosecution. It's other things that do that. You made the claim that "you get people prosecuted for objecting to immigration". You have presented no examples of this. You have in fact presented no examples of people getting prosecuted at all, and it's not clear that any of the examples you've (indirectly) given are of people who were merely "objecting to immigration". You have given no specific examples of relevant laws. You have given no specific examples of things someone could do that (1) could reasonably be described as merely "objecting to immigration" and (2) would pose any real risk of prosecution. What is asserted without evidence can be dismissed without evidence. It is not my job to give you examples of people who couldn't be prosecuted. You claim that "you get people prosecuted for objecting
I'd respond "is assault defined in such a way that it can be used against most black people, and it just isn't because of government whim? If so, then being arrested for assault could be described as being arrested for being black." Of course, you would answer "no, assault isn't defined that way". Then I'd say that since assault has a narrow definition, a black person who's arrested for assault was arrested specifically for assault, not arrested for being black with assault as the excuse. Needless to say, I'd answer the corresponding question differently for hate speech.
So far you have given zero evidence that "hate speech" is defined in such a way that it can be used against most people who want less immigration. And, I repeat, you've shifted the goalposts (from "people are prosecuted for objecting to immigration" to "people could be prosecuted for objecting to immigration") and not acknowledged having done so. Note: I'm not sure that's actually the right way to operationalize the distinction you're trying to make. Suppose there's some drug, let's call it cake, that 51% of people use. Then a law against using cake could be used against most black people, but I think it would be wrong to say that it means people can be arrested for being black. (For the avoidance of doubt, I also think it would be a bad law even if cake is very harmful; laws that criminalize very large sections of the population usually are; in that case the laws should target the dealers.)
Those two things are connected. The fact that the laws are broad enough that they could be used to prosecute almost anyone opposed to immigration implies that, for the subset of people who are prosecuted, "violated the law" is an excuse and "arrested for opposing immigration" is what's actually going on. If most people use cake, but aren't arrested for it, then it's entirely plausible that the people who get arrested "for using cake" are really being arrested for some other reason with cake as the excuse. Governments love to do this--create laws that lots of people violate, and selectively prosecute when politically expedient.
Of course they're connected and no one has suggested otherwise, but they are not the same and you started claiming one and have switched to demanding that I disprove the other. Do you have any intention of giving any actual examples of the thing you claim is happening?
The examples from before are actual examples. The vagueness of the laws makes them examples of people arrested for opposing immigration, rather than people arrested for hate speech only.
They are not actual examples. You said: You offered three links in support of this. One is behind a paywall and from your own description it doesn't sound as if you have read the paper any more than I have, so we don't know what the people in question were actually prosecuted for (other than that officially it was "hate speech"). The second is the same paper as the first. The third is the thing we have already discussed, which so far as I can tell includes no examples of anyone being prosecuted for anything. (And only one example of anyone suffering any consequences for "objecting to immigration", where (a) the consequence was police warning him that if he stirs up violence he could be legally responsible, and (b) the police allegedly referred to "your tweets" but we were only shown one tweet, so we don't know whether there was more and if so what, and (c) so far as I can tell we only have that one person's word for what happened.) (I do not know for sure that the Greek bishop was never actually prosecuted. But I do know that the article says only that someone in government had asked for him to be prosecuted, and that six years later I cannot find any sign on the internet that any such prosecution ever happened. But I don't know Greek well enough to e.g. search for possibly relevant Greek-language terms to find local news that might have covered it, so maybe it did happen?) You have (without acknowledging that it's a change) shifted from "prosecuted" to "arrested" but it doesn't appear that any of the people involved in these stories were arrested any more than they were prosecuted. You keep talking about "the vagueness of the laws" but this is another thing you have been curiously unwilling to give examples of. Show us some actual examples of hate-speech laws that could be used to punish people for objecting to immigration. It seems as if you feel that * you don't need to give examples of people actually being punished for objecting to immigration, because
That's the whole point fo having a vague law--so you can selectively prosecute with plausible deniability.
So show us the vague laws, and how they're vague, and what scope they give for that sort of selective prosecution. And show us some specific cases where people have been prosecuted for doing things that clearly shouldn't be illegal and that got them in trouble only because the laws are too vague. None of which you've done, even though it seems like minimal epistemic hygiene around this mindkilling stuff, and even though I've politely asked for examples several times. Less and less politely, I admit, because it's becoming harder and harder to believe that you're actually making any sort of honest effort to distinguish truth from untruth on this topic. Can't you see how your mode of argumentation here insulates your position from all actual contact with empirical observation? You've apparently found a way of believing that "you get people prosecuted for objecting to immigration" that isn't actually causally linked to whether or not people are being prosecuted for objecting to immigration.
It's not hard to falsify empirically at all. Take a broad set of anti-immigration arguments that people actually use, and see whether the laws can be twisted to cover them. If they can't, they're not vague (in a relevant way). I suspect the real answer is that you think that most anti-immigration arguments are hate speech anyway, so using the laws to cover them isn't twisting them. What exactly do you think is hate speech? For instance, if the example about criminals was modified so he just said that too many immigrants are criminals, would that be hate speech? What if he mentioned specific crimes, or said that the immigrants are poorly educated, or even low IQ? What if he said that immigrants should be kept out based on his personal bad experiences with them, without being able to prove that those experiences are representative? And does being true matter for whether something is hate speech?
I don't think falsification is nearly as straightforward as you're trying to argue. I can already see how it would go. "Here's an argument. I don't think it could reasonably be called hate speech according to typical hate speech laws." "Of course it could! Those laws are so vague, anything can count as hate speech." And, I repeat for what seems like the dozenth time but probably isn't actually more than the sixth, you made a specific claim which you have so far neither given any evidence at all for nor retracted, namely that people are being prosecuted for objecting to immigration. If they are being prosecuted for attempting to stir up hatred of immigrants, maybe that's a good thing or maybe it's a terrible thing, but it is not the same thing as being prosecuted for objecting to immigration. And even if you claim that the laws are drafted in a way that could in principle have someone going to prison who hasn't actually attempted to stir up hatred, I think we need to ask who is actually getting prosecuted, and who is actually getting convicted, under these laws, and so far as I can tell it is generally people that any reasonable person would agree were trying to stir up hatred. I haven't looked carefully into this. I am willing to be corrected. By actual evidence. ---------------------------------------- Now, on to your latest attempt at distraction, which asks: am I arguing against you only because actually I think that objecting to immigration is generally hate speech? I do not think that most anti-immigration arguments are hate speech. I do think some anti-immigration arguments as actually made can reasonably be called hate speech; I do think some opponents of immigration are in fact motivated by something akin to hate, and some people arguing against immigration are doing so in a fashion that they intend to stir up hatred against immigrants. But certainly not all. I gave, above, four examples of European politicians saying very negative things about immigr
Saying "haven't given evidence" over and over again doesn't make it true. I gave evidence; you just didn't like it. The argument about the vague laws determines whether the evidence is good, and you haven't settled that yet. You can't say I haven't provided evidence when we're still arguing something that determines whether the evidence counts. I would agree that the law isn't vaguer than that standard of hate speech, but that's just because they're equally terrible and equally prone to abuse. "Banning hate speech" according to your standards--that is, making something illegal because it's said "with the intention of fomenting resentment and hatred"--grants unbridled discretion to the law. Anyone opposing immigration could be found guilty under that standard. It's an example of the kind of vagueness I was talking about. (That's why your example of "less stable" doesn't work--if the law claimed that the word "stable" was trying to insinuate something that foments hatred, you have no defense.)
The claim you originally made, once again: Regardless of the merits of your arguments about how we should decide what someone has really been prosecuted for, it seems obvious that evidence for this would have to be evidence of people actually being prosecuted for something. Could you please tell me, let's say, two people you have shown us were prosecuted for something at any point in this discussion? It seems to me that however vague the laws about "hate speech" are -- and I remind you that this is another thing for which you have given no actual evidence: not one instance of a specific law that is vague in a specific way and permits some specific sort of abuse -- you cannot reasonably claim to have supported your original claim without showing us someone prosecuted for something. I am not claiming that hate-speech laws -- either any specific ones that actually exist, or the broad concept -- are good laws. I am only pointing out that if the actual situation is that "people are prosecuted for objecting to immigration in ways designed to foment hatred and resentment against immigrants" then it is not correct to describe that situation as "people are prosecuted for objecting to immigration". (For clarity: I was not claiming that "with the intention of fomenting resentment and hatred" is language suitable for writing into law. It might not be, even if one wanted to criminalize fomenting resentment and hatred. Vagueness in my description of an action does not equate to vagueness in the actual laws that apply to such actions. Maybe the actual laws are in fact too vague, but let's see some examples.) You keep making claims about what hate-speech laws permit. I doubt those claims. (I repeat: that does not mean that I endorse those laws. They might be bad laws. But I am not convinced that you are describing their consequences accurately.) Could you please give some examples of particular countries' hate-speech laws, and of things that don't go beyond "objecting to immig
Come on. One of the links' titles is "Prosecuted, yet popular? Hate speech prosecution of anti-immigration politicians in the news and electoral support". I admit that the summary (which is all I can read, since it's in a paywall) doesn't contain names, but it's pretty obvious that such people exist.
Nope. It makes it obvious that some people are anti-immigration and have been prosecuted for hate speech. Very likely the things they were prosecuted for aren't independent of their opposition to immigration, but we don't know anything about what those things were. Oooh. You can get the actual paper (or maybe it's an earlier pre-publication version of it) here. Let's take a look at the first few concrete examples in it, assuming there turn out to be some. This looks like it might be informative -- it seems as if we have two cases near the borderline between what gets actual legal penalties and what doesn't. (For the avoidance of doubt: I appreciate that merely getting prosecuted is unpleasant, time-consuming, potentially expensive, etc. I think there are a few key thresholds: what do you have to do to get prosecuted, what do you have to do to get convicted, and what do you have to do to get a substantial sentence imposed?) * 2009-2011. I think this document quotes the things he said that got him in trouble. (Though not enough trouble for an actual hate-speech conviction.) A few I-hope-representative samples: "Islam wants to control, subdue and is out for the destruction of our Western civilization." "The demographic composition of the population is the biggest problem of the Netherlands. [...] We must stop the tsunami of the Islamisation." "The core of the problem is the fascist Islam, the sick ideology of Allah and Mohammed as laid down in the Islamic Mein Kampf: the Quran." "One out of five Moroccan youngsters is registered with the police as a suspect." "Former chief of the Mossad Efraim Halevy says that the Third World War has already started. I am not putting these words in my mouth, but it is correct though." "I have had enough of the Islam in the Netherlands: no more Muslim immigrants. I have had enough of the adoration of Allah and Mohammeed in the Netherlands: no more mosques. I have had enough of the Quran in the Netherlands: prohibit this fascist book
You are shifting between objections. does not leave room for "it doesn't count because they're prosecuted for the wrong thing".
At the point when I said "Regardless of the merits ..." the nearest you had come to providing actual examples of prosecutions was gesturing vaguely towards an article neither of us had read, the information about which we had gave us no information about any prosecutions. I concede that in some sense this constituted evidence for the proposition "some people have been prosecuted for something", but it did not tell us anything of the form "someone was prosecuted for X" where X could credibly be "objecting to immigration". Then I found the actual paper (note: I did, you didn't; all of the actual intellectual work in this discussion is being done by me) and now the situation is no longer "we don't have any examples of anyone prosecuted for anything". It turns out that this doesn't actually help your case, since the examples in the paper don't appear to include anyone who can credibly be described as having been prosecuted for objecting to immigration. But it does mean that my previous simpler objection (no examples of prosecutions provided, nothing else to say) needed to be replaced by something more complicated (now we have some examples of prosecutions, so let's see what they are). Yes, I change what objections I'm making when the argument I'm addressing changes. In this case it changed because I discovered I was able to perform a part of your job that you hadn't bothered to do, namely finding some concrete examples. So. Got any specific examples of people prosecuted for objecting to immigration, yet? Or any specific examples of hate-speech laws for which you can explain how they would criminalize objecting to immigration, as opposed to attempting to foment hatred and resentment against immigrants? (For the sake of explicitness: if you wish to scale your claim back to "... now you get people prosecuted for attempting to foment hatred and resentment against immigrants", I will readily agree that that happens, but not that there's no real difference between forbid
The "only" is about the reach of the law, not the person's actions. It's like having a law that criminalizes breathing, and the law is being used to prosecute a thief. Even though the person's actions were not only breathing, I'd still call that "prosecution for breathing".
The law in this case criminalizes stirring up hate (at least, that is what "hate speech" laws generally do, and you have not seen fit to clarify what specific laws you have in mind). Your analogy would be valid if there were laws against objecting to immigration and they were being used to prosecute someone who is stirring up hatred. That is not the case. Rather, it's as if there were a law against theft, and they were used to prosecute people who were breathing, and you said "look, people are getting prosecuted for breathing". If the anti-theft laws were broadly enough drafted, or being applied in an unprincipled enough way, you might have a case, but the onus would very much be on you to show that that was what was happening, because prima facie there's a law against theft and it's being used to prosecute a thief. (In this case: prima facie there are laws against stirring up hatred and they're being used to prosecute people who are stirring up hatred.)
I was initially going to reply to Jiro's last comment to me, but you grasped the nettle so firmly that I think I'd be just superfluous here, so I just wanted to say I almost completely agree with everything you write and that it's awesome you put in so much effort.
Thank you for the kind words! It seems that Jiro has lost interest, though.
As I keep pointing out, who counts as being prosecuted for immigration depends on whether prosecuting someone selectively under a broad law, while officially calling it something else, counts as prosecuting them for immigration. We haven't settled that yet, and you shouldn't be acting as though we've settled that.
If I say "there are people being prosecuted for being black", and present as evidence a bunch of black people prosecuted for theft, fraud, murder, etc., and claim that this justifies my claim because the laws against theft, fraud, murder, etc., are too broad, then it is my responsibility to show how broad the laws are, and how policing and prosecution are effectively looking at people's race rather than their actual behaviour. If you say "there are people being prosecuted for objecting to immigration", and present as evidence a bunch of anti-immigration politicians prosecuted for hate speech, and claim that this justifies your claim because the laws against hate speech are too broad, then it is your responsibility to show how broad the laws are, and how policing and prosecution are effectively looking at people's opinions on immigration rather than the incendiariness of their rhetoric. But you aren't doing any of the work required to justify your (prima facie implausible) claim. You're behaving as if everyone's default assumption should be that (1) the hate speech laws are so broad that they make basically everyone who objects to immigration a criminal, and that (2) what on the face of it are prosecutions for hate speech are really motivated by wanting to punish everyone who objects to immigration. I do not think either of those is a reasonable default assumption. I think both are implausible, and I think that if you want us to believe them you should give some evidence for them. However, I'm kinda losing hope of your ever producing any evidence for any of the claims you make, so I guess I'll have to do your job again. ---------------------------------------- Consider four possible worlds. In world A1, hate-speech laws are extremely broad and criminalize a lot of things that don't really involve stirring up hate and resentment and whatnot. This fact is exploited by the law-enforcement system to persecute people who object to immigration, so that such people ar
Not everyone objecting against immigration legally commits hate speech (or is claimed by the police/prosecution to commit hate speech), while everyone who is put in jail has been arrested. The legal concept of hate speech is made up, of course (just like all legal concepts), but the phenomenon of hate speech is very, very real. (Or maybe you mean that xenophobic hate speech is made up but other kinds of hate speech aren't, in which case I have to say that specific kind of hate speech is, unfortunately, very real as well.)
To hopefully be a little clearer: Calling something hate speech, and banning it for that reason, is the method by which Europe bans this kind of speech. Saying "it doesn't count because they're just banning hate speech" is equivalent to "it doesn't count because they're banning it in the way they usually do". If the typical method of doing X automatically makes it not count--just because it's using the typical method--then of course you'll have trouble finding examples of X.
So hate speech is real (independently of whether someone puts the phrase "hate speech" into law or not), and the typical way of being anti-immigration involves hate speech. I agree. I don't see what the problem is - it's up to the anti-immigration activists to find a way to protest that wouldn't be hate speech. The fact that until now, they usually used hate speech, is purely their problem, and purely their fault. Alternatively, what you're saying could also be interpreted as claiming that Europe abuses the phrase hate speech to ban non-hate-speech discourse (which becomes "hate speech" upon the speech being legally banned)... but that's not true.
The typical way of being anti-immigration "involves hate speech" because "hate speech" is so broadly defined that it's just about impossible not to. European governments get to define hate speech, so activists can't find such a way, if the governments don't want them to have one.

After I read Wikipedia’s article on Schelling points (a.k.a. “focal points”), this article made much more sense. I recommend reading it – it’s only three paragraphs at the moment.

In short, a Schelling point is a point that everybody agrees is an “obvious” cutoff point. That knowledge helped me understand this article’s point that Schelling points make good fences to precommit to because you can’t justify moving the fence anywhere else later – the other points are not nearly as “obvious”.

Slippery slopes legitimately exist wherever a policy not only affects the world directly, but affects people's willingness or ability to oppose future policies.

Isn't this all the time, though? Whenever a policy changes, if nothing else, the incumbent policy changes, which will affect people's willingness or ability to oppose future policies. Looking at the history of politics and law, it looks like slippery slope arguments are right more often than they're wrong.

That's really vague. Examples would help, preferably ones from long ago so as to avoid the mind-killer: for instance, the Reign of Terror could be an example of a slippery slope. The other side, of course, is that not all slippery slopes lead to bad outcomes- consider the civil rights movement. Anyway, of course every policy change affects the likelihood of other policy changes, but only occasionally is it a runaway effect- usually it stops there or peters out.

That's really vague.

Senses of history are vague, but still informative. It's not clear to me there's much value in digging up examples.

preferably ones from long ago

I don't consider 1950 to be long ago.

that not all slippery slopes lead to bad outcomes

No, and I did not mean to imply that they did. Many intentionally begin slippery slopes to lead to outcomes they like; foot-in-the-door techniques can be seen as an example. The takeaway is that the slippery slope meme doesn't appear to actually be fallacious- if you think that A will increase the chance of B, and you dislike B, it's often the correct strategic move to oppose A, even if you think A in isolation is a good thing. The challenge is getting the correct model of how A will impact the chances of B.

I don't think we disagree here. As far as 1950 being "long ago", my point was that I picked examples that I really don't expect to be live issues for Less Wrong readers; there were other issues being discussed in 1950 that are still subjects of disagreement between LW-type people, and those I shouldn't use.
Many very smart people wouldn't unreservedly agree that this wasn't a bad outcome; perhaps a different example could be found though? (Ideally one that isn't as politically charged?)

Many very smart people wouldn't unreservedly agree that the Terror was bad, either. If you're far left enough to cheer the Terror or far right enough to boo civil rights or rebel in a clown suit enough to do both, you should be used to not being the default audience and practiced at separating such "obvious" examples from the formal role they play in the argument.

You are ignoring the fact that the historical developments commonly known as "civil rights" have in fact led to a progression of ever more extreme policies (i.e. a slippery slope) whose present outcome is controversial even in the mainstream. This is indisputable no matter what position (if any) you happen to support in these controversies.

This slippery slope can be roughly described with the following progression:

1: Government-mandated discrimination across racial/ethnic/religious groups.

2: Libertarian/classical liberal position: procedural equality for everyone as far as the government is concerned, freedom to discriminate (or not) for private parties.

3: Prohibition of overt discrimination even for private businesses and organizations.

4(a): Affirmative action -- the government (and private parties under its influence and pressure) actively try to equalize statistical outcomes across groups by favoritism towards members of groups that do worse on average.

4(b): Disparate impact doctrine -- even if there is no overt discrimination, unequal statistical group outcomes are considered as evidence of discrimination by themselves, and any institution that produces such outcomes can be held legally liable on that basis alone.

While 1-3 are no longer controversial in the mainstream, 4(a) and 4(b) are still matters of intense public controversy. (Admittedly, for unclear reasons, 4(b) gets far less publicity than 4(a), despite its arguably even greater impact in practice.)

Yes, you're right. (And if we like we can extend this slope further out until it includes the Terror, and tada, we've constructed the classical one-axis political spectrum.) Obligatory face-saving : I still think my claim stands that people (especially "very smart people" like LW members consider themselves) should be able to read the comment, recognize the formal nature of the argument, substitute in slopes they themselves thing are good or bad, and then evaluate it on those corrected terms, even if we generalize it to include normal people and not just nutcases like me or you. But maybe I shouldn't think it stands, because Will (who is very smart, and a nutcase to boot) apparently can't think of any self-reinforcing social changes that he thinks are good. (Or maybe he can, but lays emphasis on universal agreement for some reason? Idunno.)
(I'm very, very bad at this sort of coming up with examples, so I don't think my inability to come up with any is much evidence for anything. I'm also very, very bad at finding physical objects amongst other objects, e.g. looking in the fridge for a certain jar. I strongly suspect that those two skills are strongly related. Eliezer also claims to be very bad at coming up with examples and has told an anecdote about his inability to find things in the fridge (which he then ascribed to males in general---there are many reasons to be skeptical of the generalization). I suspect something interesting is going on here, and I tentatively wonder if it has to do with damage to, or atrophy of, the dorsolateral prefrontal cortex.)
Hmm, its curious. I am pretty good at coming up with examples or picking out items from crowded environments. Well, for the social changes... what's about gradual abolition of religious fundamentalism? It can be self enforcing (just as introduction of fundamentalism is; instability implies self-enforcing effects both ways). In general if you can come up with some self-reinforcing social change that you think is bad, the same change, starting from the bad state (assuming that it can start at all), would be self-reinforcing in the good direction. A steel object falling off from under a magnet is a self-reinforcing process - the further it falls, the lower is the attraction - and so is the steel object snapping onto a magnet - the closer it gets, the stronger is the attraction force. edit: ahh, i looked up in comment thread. Indeed, the problem is that it is hard to keep unstable process in equilibrium, and the self-reinforcing processes go too far. At same time, if you take a terribly religious population where we burned witches, or where they stone the rape victims to death, it is easy to imagine that on the other end of the slippery slope - if the slope is at all inclined in the other direction - the life is massively better. I made a magnetic levitation device once - it would suspend iron nail under electromagnet. It seems deceptively simple - when nail goes up, turn off magnet, when nail goes down, turn it on - but if you do so you get rapidly increasing oscillations - you have to have a circuit that blends in first and second derivatives of the position into the control signal. A great deal of complexity for a very simple unstable system.
Tangent: A similar problem was described in Sebastian Thrun's Udacity CS373 couse with respect to steering a self-driving car. It seems the similar principles should apply, except that the distance from the magnet will change the effect of gravity on the nail, which makes the problem more complicated.
This is the best one-paragraph technical argument against "small-c conservatism" (social, cultural, etc) that I've ever read. Edit: "the best" doesn't necessarily mean "totally overwhelming"; I just find it a very good illustration of the inherent problems and opportunity costs.
If this is the case, i.e., if this really is the best argument against conservatism you can come up with, it strikes me that you should become a conservative. Notice that this type of argument is even stronger against anything else, i.e., what makes you think you can manage social change?
Um, I really don't think so. This argument appears to imply that, if "conservatism" is a costly, complex and unreliable effort to keep a system somewhere between two or more "attractor" end-points, then one should seriously try to predict what those end-points could be like, and whether facilitating a gradual "slide" towards one of them could not be a better use of time and resources then keeping up the unstable equilibrium. E.g. consider how increasingly open and permissive Western culture has been growing in regards to sex, or how privacy norms have been eroding in the last decades, or how "democracy" has been losing substance and legitimacy in favour of de facto oligarchic or bureaucratic rule in the post-war world order. It might make more sense even for a person who dislikes some particular accompanying changes to consider the range of probable outcomes for such trends - and whether trying to ensure a better, less destructive transition to such an outcome could be more worthwhile than a hopeless defense of the current state or trying to launch a counter-trend. (With the above examples: improving sex ed and correcting biased feminist/etc dogma instead of preaching puritanism/monogamy. Providing citizens with ways to spy back on their governments and corporations instead of trying to curb the ominpresent surveillance. Developing better expert-driven, liberty-preserving political systems, like futarchy, instead of clinging to the facade of elected officials driving policy.) We see that similar kinds of social change can be managed or steered in better or worse ways (e.g. the denazification of Germany versus its treatment after WW1 or the West's handling of USSR's collapse). Meanwhile, attempting to keep a society, its politics or culture in a stasis has had ended with an uncontrollable shift (late USSR) or an explosion (Chinese Empire, Japanese Shogunate) every single time.
Um no, conservatism is an attempt to keep society at a relatively stable point. What makes societies stable is being at Schelling points, and one way a Schelling point can be stable is to have an established tradition behind it. Another way to have a stable Schelling point is to move it to a maximally extreme position, the problem with this approach is that it's nearly always possible to become more extreme and the society will collapse before you can become extreme enough. Ok, a counter movement frequently occurs when society starts to collapse pushing the trend in the other direction, but you seem to be arguing against such movements. Looking at history there are also many inevitable-seeming trends that failed or were even reversed, e.g., the trend towards absolute monarchy in the 18th century, the rise of eugenics in the early 20th century, the expansion of communism in the late 20th century. I have another example for you: with the trend towards a wider acceptance of torture (e.g., war on terror) we should manage it my formulating rules for when torture is and isn't acceptable rather than keeping to a no torture policy. It only seems this way because the instances when a society failed to change are less memorable.
FWIW, me too, on both counts. Though I don't classify myself so much as very bad at finding physical objects anymore, since I default to systematic search pretty quickly, and while it's slow, it almost always works. Interestingly, I noticed I can do it the 'normal' way when on Adderall - it feels almost like having a HUD that's highlighting relevant objects for me (very useful when driving a car).
4b needs "statistical significance of discrimination" and "concentrated responsibility" clauses, and then it becomes an antitrust law.
Good point, but still: does anyone know of any slippery slope [ETA: by which I mean a cascade of self-reinforcing changes in laws or social norms] that most everyone can agree was clearly not-bad? I ask because there are various theoretical reasons why one should almost never expect slippery slopes to have good consequences, but if empirically that's not the case then I need to revise my sociological and historiographical models. (ETA2: My bad, I confused levels of abstraction; I agree with the criticisms that such an analysis is unfeasible even if possible.)

Aren't you smuggling in the conclusion? Incremental change that builds on previous changes is generally called a "slippery slope" only when the consequences are undesired. Is the gradual increase in homosexual rights good? Then it won't be called a slippery slope.

More generally, there aren't many changes that "most everyone can agree was clearly not-bad." If everyone thought it would be a good idea, society wouldn't have been doing things some other way.

This runs into the general problem of determining whether moral progress exists. Namely, after your morals change the change is always good as judged by your (new) morals.
Could you elaborate on these theoretical reasons? Because obviously the desirability of a slippery slope isn't a relation between the slope and possible desirers, not a property of the slope as such, and it's difficult to see what dynamics of slopes in action could be affected by the attitudinal relations of later persons towards them. Bad from the perspective of those who initiated them, perhaps? Since history tends to ebb and flow even when it does have a secular direction, examples of self-reinforcing changes in laws or social norms can be found sloping in both directions across any dimension, so if e.g. you think that civil rights was bad, it is not hard to find periods and places where movements in the opposite direction had a self-reinforcing nature. So leaving aside very strong formalist conservatives who oppose changes in laws or social norms a priori, even if there are no slippery slopes that everyone considers good, I would expect that every person would be able to see at least one slippery slope that is by their lights good.
I doubt it, but because of the difficulty of citing any such social or political change, rather than because of some special property of slippery slopes. Edit: oops, didn't see TimS beat me.
Going further back in history, you have the process by which the King of England lost power and Parliament gained it...
As others have said, this seems like a confused question. The legalization of interracial marriage in the US would seem like the obvious example, but I don't know what you want to count or not count as part of the trend. And I'm just guessing at how to interpret "most everyone".
reduced levels of violence in society? e.g. the reverse of this slippery slope
Seems to me that rationalism as a living ideal is a slippery slope with a positive outcome. Once someone takes the initial steps to use rationalism, they then seek to learn more about rationalism, they practice it more and they become more effective and efficient at utilising it. That looks like a slippery slope to me, but obviously one that has a different outcome type than a traditional negative outcome orientated slippery slope.

I think I can describe the 'slippery slope' to an alien and I think this description offers an obvious (if difficult) solution.

The 'slippery slope' is what we call a situation in which good judgement and a certain algorithmic response to something correspond in one case, but not in others, such that following that algorithm in similar cases will cause us to diverge from our good judgement at some point. 'Slippery slopes' come up when our good judgement about something is partially but not entirely captured by a particular algorithm.

Our solution to slippery... (read more)

[comment deleted]

[This comment is no longer endorsed by its author]Reply
It might work if you assume he gives it all away to charity.

Or could precommit: he takes the deal every time, sending something like 100m to charity, saving hundreds of thousands of lives. He unfortunately kills one person and goes to prison for life, but that's still net hundreds of thousands of lives.

(This is too utilitarian for the real Ghandi, of course, who told the Jews to just let Hitler kill them.)

By the time Gandhi had taken the deal 99 times, though, don't you think he might be less inclined to behave so altruistically? 1%-Gandhi would additionally know that he was on his last chance to become a millionaire, which might help sway him further.
Well, yeah, hence the mention of precommitting.
But 1% Gandhi has no reason to honor the precommitment.

But 1% Gandhi has no reason to honor the precommitment.

I don't think that word means what you think it means.


A precommitment you can dishonor isn't much of a precommitment!

Or find a way to pay in constant amounts of utility, rather than constant amounts of money - which would avoid the distraction of diminishing rate of return.

Regarding what you call abandoning the power of choice, here is an example from HPMoR that looks more like a slippery slope.

The point being, said his inner monitor, it's getting worse literally by the minute. The way spies turn people is, they get them to commit a little sin, and then they use the little sin to blackmail them into a bigger sin, and then they use THAT sin to make them do even bigger things and then the blackmailer owns their soul.

Didn't you once think about how the person being blackmailed, if they could foresee the whole path, would just

... (read more)
This is somewhat more than a slippery slope. It isn't just that large decisions seem easier to make, there is more force to ensure they will happen. The incentives are such that without any slippery slope reasoning and each breach seeming to be just as bad at the end as it did at the start the steady corruption would still occur.

I think your idea about a Schelling point deserves further thought. But, why (and how) select the point arbitrarily? Why now presume that rational 100% Gandhi would perform an optimisation, according to his personal utility function, calculating the good the offered money could do against the harm done to the world by becoming less pacific?

Since you've already posited a third party, in your example, engaged to destroy Gandhi's prized possessions for deviations, why not just have Gandhi charge the man to shoot him as soon as he shows any sign of ... (read more)

And Ghandi spoke, "I will pay you a million dollars to invent a pill that makes me 1% more pacifist."

There is (entirely fictional) evidence that this may have disastrous) consequences.
He'd lose money. 1.01 * 0.99 = 0.9999 < 1. Or in general, (1+x)(1-x)=1-x^2 < 1
And having never taken the first pill, he'd be glad to lose it to take the second pill.

"Refuse to adjust your utility function because you will no longer be you, unless the adjustment improves you in terms of your own values" seems to be an important general principle, and it should be enough for Gandhi to turn down the pill.

There is a potential negative consequence, when groups are together against a reasonable project, but can not agree on the alternative they actually prefer. That can be viewed in action in Germany with big infrastructure projects. In particular at the moment with the reconstruction of the Train Station in Stuttgart. Many groups are against the project, but have not compatible ideas on what to do instead. If contrarian groups get better at coordinating their protests you might end up in a situation where no productive project is possible any more.

In politics it leads to a part of the population to always vote for the other guy.

Maybe Gandhi's best option is to "fence off" an area of the slippery slope by establishing a Schelling point - an arbitrary point that takes on special value as a dividing line. If he can hold himself to the precommitment, he can maximize his winnings. For example, original Gandhi could swear a mighty oath to take only five pills - or if he didn't trust even his own legendary virtue, he could give all his most valuable possessions to a friend and tell the friend to destroy them if he took more than five pills. This would commit his future self to stick to

... (read more)

Here's an argument I found that "hyperplastic agents" (i.e.,Strong AI) cannot make use of Schelling Fences:

"One evening, I start playing Sid Meier's Civilization (IV, if you're wondering - V is terrible)" THANK YOU. ;D

The ban on holocaust denial undermines the concept of free speech - there is no agreed upon schelling point and arguments start. Many people don't really understand the concept of free speech because the example they see is actually a counterexample.

Not everyone is totally okay with it, I certainly am not.


Maybe Gandhi’s best option is to “fence off” an area of the slippery slope by establishing a Schelling point—an arbitrary point that takes on special value as a dividing line. If he can hold himself to the precommitment, he can maximize his winnings. For example, original Gandhi could swear a mighty oath to take only five pills—or if he didn’t trust even his own legendary virtue, he could give all his most valuable possessions to a friend and tell the friend to destroy them if he took more than five pills. This would commit his future self to stick to the

... (read more)

Man, this article hits different now that I know the psychopharmacology theory of the FTX crash...

[block]<quote>"There are also a host of other well-respected exceptions to free speech, like shouting "fire" in a crowded theater."</quote>[/block]

The case that this quote came from was overturned 40 years ago. If you intend to continue using the analogy, please read the attached link first. It makes a good argument for retiring the phrase.


Orthogonal to the main point, but I think the banning of holocaust denial seems a really stupid policy. Not for freedom of speech reasons, but because the policy is "aimed badly", failing to cover some of the people we would presumably like to restrict, and restricting (at least if the law is taken literally) some viewpoints that seem harmless.

If there is a person with Neo-Nazi politics running around in uniform intimidating people then I have a serious problem with that person and can see the point in a law designed to limit their neo-Nazi-ing. But, that ... (read more)

I am a hyperbolic discounter. Now I can see it clearly. I always default on my own promise to myself that I will stop at doing things that just waste my time (like reading fanfiction, or just plainly procrastinating) at some specific point in time. Then when that time point finally arrives I just continue wasting my time, and I make the same promise as before. I tell myself the same lie as before. However, now I know the problem and I have the tool - the Shelling fence. Now I can face the problem straight ahead and concur it. Wish me well.

I've read the comments, and nobody seems to have mentioned the different functions of speech, and the different goals of limitations. Yvain's examples of Holocaust denial and "shouting fire" are not the same sort of thing.

Speech as an act causally linked to a desired consequence unrelated to communication, like falsely shouting "fire" in a crowded theater, or commanding a subordinate to shoot someone, or etc., isn't really protected (I'm not very familiar with US Law, but I can't think of counterexamples). These limitations don't usuall... (read more)

I don't understand how you distinguish the two. Shouting "Fire!" in a crowded theater looks like "transmission of information" to me. On the other hand, publishing some compromising material during an election campaign seems to be "an act causally linked to a desired consequence".
Sorry, that was supposed to be "falsely shouting fire". I'll be surprised if someone is punished for shouting "fire" if there really was a fire and they were only transmiting this information. For that reason, I also expect that "I thought there was a fire!" would be an effective defense.
I reiterate that I don't understand how do you distinguish the two.
By the presence of this additional element, the intent for the speech to cause a concrete effect besides transmission of information, and its ability to cause it. ETA: For some reason I only now noticed the "other hand" on your first response. Yes, all speech transmits information (even if just the information that I think A or desire B), and most speech intends some goal (even if just the goal that people think like me). When I think of the problem of putting a speech act in the first or second boxes my mind follows paths similar those used in determining causation in Law, with similar difficulties. There won't always be a bright line, but most instances will be closer to one end than the other.
A great deal, might even be most, of speech has intent "to cause a concrete effect besides transmission of information". The freedom of speech laws do not aim to protect transmission of information -- they aim to protect precisely the right to speak in order to produce tangible consequences. Consider e.g. whisteblowing. Or pretty much any political speech -- are you saying engaging in political speech specifically in order to influence the elections "isn't really protected"?
No, I'm saying that causation is sufficiently less direct in this case (than in cases like shouting "fire" and ordering a murder) that it's more reasonable to put it in the "intent to transmit information" box.
I really don't see that. Take a plain-vanilla election poster consisting of an ugly mug and "Vote for X!" This is a pretty direct attempt at causation and I don't really see much of information being transmitted.
Looking at my thought process, I think I'm using this differentiating test: Look at the probability of the outcome, given the speech - if it's high enough that you can ignore the receiver of the message as an independent agent whose response generates uncertainty, the causation looks pretty direct. But if the outcome is dependent on people freely considering the information and acting on their own conclusions (as they would if the information was known by other means), then it looks indirect enough that I consider "transmiting information" as the function of the speech.
Example 1: shouting "Fire!" in a crowded theatre. Are you quite sure the crowd will stampede? or they'll look at you like you're an idiot and tell you to shut up? Example 2: Russel and Hugh are two best mates living in Australia. Russel says: "An election is coming and I have to vote. I don't care about them slimy politicians and I'll vote for whoever you tell me". Hugh says "Sure, mate, this time vote for the Wombat!". Is Hugh's speech protected?
1- I am not sure it would happen, but I think that someone who does shout "fire!" is indeed quite sure people will run. 2 - I don't know Australia's laws, so I don't know what would be protected. But Hugh's speech goes in my first box (the only information being transmited is Hugh's preferences. Also, by analogy: if it were "should I kill him?", both would be responsible).
In the US you can't directly command someone to shoot someone but you can advocate that someone deserves to be shoot (see Brandenburg v. Ohio)
Thanks, this is what I thought. That would be the difference between my first and third types of speech, and an example of a controversy about how to draw the line.

I feel like the following game-theoretic model might be illustrative. Suppose that you have N people ordered by radicalness. These people will repeatedly vote on the proposition of killing the most radical among them (with a simple majority needed to succeed). Assume that each person's preferences are such that they would prefer to have people who are more radical than they are killed but not at the cost of being killed themselves.

This is a game of perfect information and is not hard to analyze. The result is that people will be killed until a power of... (read more)

i really enjoyed reading this thank you. best article i've read on here so far. fantastic examples and the examples really 'make' it - i shall try not to plaguerise but from now on when arguing about this stuff i will definitely give examples that show the progression.

the big one here in europe at the minute is 'hate speech' - arguments between whether it's an exception that's dangerous and damaging enough to be worth an exception or whether it's a dangerously low point on that slippery slope.

Economist Jeff Ely recently blogged an interesting example of a slippery slope.

Is "demonstrably optimal policy" a shelling point or not?

Depends in the inferential distance to the "demonstrably" part. But otherwise yes, frequently even when the policy is "demonstrably optimal" with respect to something that's not quiet the right utility measure.
One hopes so, unless there is some other shelling point that clouds it. Perhaps a status quo within a local minima.

I never thought I would hear a plausible defence of slippery slope arguments.

An interesting analogy is with the Sorites or 'heap' paradox, and mathematical induction. In the paradox you show that one grain of sand is not a heap, and that two grains are not a heap, and three.... so you generalise that for if N grains of sand is not a heap then N+1 grains is also not a heap. Therefore 10^1000 grains of sand cannot be a heap, and there are no heaps!

Obviously the problem is that the premise isn't true for any arbitrary N, (unlike cases of mathematical inducti... (read more)

You may want to look at this.

Thanks, I've seen that before. What interested me about the reaction to it was every commentator decried them for suggesting babies should be killed, said that it would give weight to the arguments of anti-abortionists or that it showed how out of touch academics were with public opinion. But no-one gave an argument in response about why an 8 month abortion and a born baby are different in a morally relevant way. I had underestimated how much in general public discourse even discussing a morally condemned act was itself condemned.

In the context of slippery slopes, again this is moving between two adjacent points not showing you can just as easily move to any point on the scale.

Yes, and then we move to a point adjacent to the new point, and then to a point adjacent to the next point. This is how slippery slopes work.
He means adjacent without ascension/descension: lateral movement, i.e. change in non-morally relevant variables.
This is a tangent, but... This is certainly not obvious (I submit that if it were, the Greeks would have noticed) and I don't think it's true either. Treating heapness as a boolean, the Sorites paradox indeed implies that there are no heaps, or that any number of grains is a heap. If N grains of sand together are not a heap, then N+1 grains of sand also are not; the piles are visually the same, and so anyone who knows what 'heap' means would look at them and classify them as both "heap" or "not heap". I submit that if you ran that experiment for different values of N (showing each respondent only one instance of N and N+1), you would find no N for which most respondents classified one as a heap and the other as not a heap. Since this is a paradox, we know there's a problem with one of our assumptions. These days, the obvious one is that heapness is not a boolean. There's a cluster in thingspace we call "heap", and some things are clearly part of the cluster, like a pile of 2000 grains of sand, and some things are clearly not part of the cluster, like a single grain of sand. Then to do our mathematical induction, we must agree that for any arbitrary N, if N is clearly part of the cluster then N+1 is also clearly part of the cluster. But there is some point at which you become uncertain. I submit that there is some N such that N grains of sand is clearly not a heap, and N+1 grains of sand may or may not be a heap; also, there is some N such that N grains of sand may or may not be a heap, and N+1 grains of sand is a heap. Though really if we wanted to formalize this, we'd probably set different thresholds depending on whether we're adding or removing grains of sand; the N at which something stops being clearly a heap is lower than the N at which something starts being clearly a heap, and we might even cut out the "uncertain" category from this model. And this would probably match people's intuitions if they watched people adding or removing grains of sand and were
Or, you can still treat "heapness" as a boolean and still completely clobber this paradox just by being specific about what it actually means to have us call something a heap.
Any feasible number (free PostScript version) of grains of sand is not a heap.

Great post Yvain!

By the way, seems to be small typo:

One evening, I start playing Sid Meier's Civilization (IV, if you're wondering - V is terrible). I have work tomorrow, so I want to go to stop and go to sleep by midnight.

[This comment is no longer endorsed by its author]Reply

Upvoted; enjoyable anecdotes.

Same here, utterly sacrilicious.
Upvoted; new vocabulary.
I tried to tell my husband about murder-Gandhi but I was laughing too hard.

You don't really need to figure out who the Right Thinking People are. You just need to figure out who the Horrendously Completely Almost-Unassailably Wrong Thinking People are.