"How many legs does a dog have, if you call a tail a leg?  Four.  Calling a tail a leg doesn't make it a leg."
        -- Abraham Lincoln

So I was at this conference.  Where one of the topics was legal rights for human-level Artificial Intelligence.  And personally, I don't think this is going to be the major problem, but these are the kinds of conferences I go to.  Anyway.  Brad Templeton, chairman of the Electronic Frontier Foundation, was present; and he said:

"The legal status of AIs is ultimately a legislative question, and in the American system of democracy, legislative questions are decided by the Supreme Court."

Much laughter followed.  We all knew it was true.  (And Brad has taken a case or two to the Supreme Court, so he was speaking from experience.)

I'm not criticizing the Supreme Court.  They don't always agree with me on every issue - that is not a politician's job - but reasoned cooperative discourse, compact outputs, and sheer professionalism all make the Supreme Court a far more competent legislative body than Congress.

Try to say aloud the color - not the meaning, the color - of the following letters:


Now say aloud the meaning of the letters, not their color:


Which task felt more difficult?  It's actually easier to pay attention to the labels of things than their substances.

But if you're going to be faced with several repetitions of the first task, there's a way to make it easier - just blur your eyes a little, so that you can see the color a moment before you're distracted by the meaning.  Try it - defocus your eyes slightly, and then say the following colors aloud:

BLUE        ORANGE        YELLOW

If you want to know what the Supreme Court really does, you should blur your eyes so that you can't see the words "Supreme" or "Court", or all the giant signs reading "judge", "Honorable", or "judicial branch of government".  Then how can you tell what they do?  Well, you could follow these nine people around for a few days and observe them at work.  You'll see that they dress in funny outfits and go to a building where they hear some people arguing with each other.  Then they'll talk it over for a while, and issue one or two short written documents, some of which tell other people what they are or aren't allowed to do.  If you were a Martian anthropologist and you had absolutely no idea that these people were supposed to be doing this or that or something else, you would probably conclude they were (among other things) making laws.

Do Representatives and Senators make laws?  Well, I've met one or two Congresspeople, and I didn't see them writing any documents that tell people what to do.  Maybe they do it when no one is looking?  I've heard their days are pretty heavily scheduled, though.

Some laws are made by Congressional staff, but the vast majority of legislation is written by professional bureaucrats, who - if you refocus your eyes and read their labels - are part of the "executive" branch.

What if you've got a problem with a bureaucrat getting in your hair?  You won't have much luck talking to the White House.  But if you contact your Representative or Senator's constituent service office, they'll be happy to help you if they can.  If you didn't know how the system works apart from a high school civics class, you might end up very frustrated - the people who help you deal with the executive branch of government have signs reading "legislative branch".

Your Congressperson would much rather help a little old lady deal with a lost Social Security check, than make potentially controversial laws about immigration or intellectual property.  That sort of thing may please some of your constituents, but it gets others very angry at you, and voters are faster to forget a favor than a disservice.  Keep making laws and you might just get yourself unelected.  If you know everyone is going to cheer you for a law, go ahead and pass it; but otherwise it's safer to leave legislation like e.g. school desegregation to the Supreme Court.

What Congresspeople prefer to do is write elaborate budgets that exert detailed control over how the government spends money, so they can send more of it to their districts.  That, and help their constituents with the bureaucracy, which makes friends without getting anyone mad at them.  The executive branch has no time for such matters, it's busy declaring wars.

So, bearing in mind that nothing works the way it's written down on paper, let's defocus our eyes and ask about the role of the voters.

If we blur out the label over your head and look at what you do, then you go to a certain building and touch one of several names written on a computer screen.  You don't choose which names will be on the screen - no, you don't.  Forget the labels, remember your actual experiences.  You walked into the building, you did choose which rectangle to touch, and you did not choose which little names you would see on the computer screen.

When Stephen Colbert wanted to register for a presidential run in South Carolina, the executive council of the South Carolina Democratic Party voted 13-3 to keep him off the ballot:  "He clearly doesn't meet the requirements.  It's a distraction and takes away from the seriousness of our primary here and takes attention from the serious candidates:  Clinton, Edwards, Barack Obama and the rest."

Hey, South Carolina Democratic Party executive council, you know who ELSE might be interested in determining whether someone is, or isn't, a "serious candidate"?  HOW ABOUT THE %#<!ing VOTERS?

Ahem.  But the psychology of this response is quite revealing.  "They want to prevent wasted votes" would be a polite way of putting it.  It doesn't even seem to have occurred to them that a voter might attach their own meaning to a vote for Stephen Colbert - that a voter might have their own thoughts about whether a vote for Stephen Colbert was "wasted" or not.  Nor that the voters might have a sense of ownership of their own votes, a wish to determine their use.  In the psychology of politicians, politicians own voters, voters do not own politicians.  South Carolina Democratic voters are a resource of the South Carolina Democratic Party, not the other way around.  They don't want you to waste their votes.

(Am I the only one on the planet to whom it occurred that the South Carolina voters could decide the meaning of a vote for Stephen Colbert?  Because I seriously don't remember anyone else pointing that out, at the time.)

How much power does touching a name in a rectangle give you?  Well, let me put it this way:

When I blur my eyes and look at the American system of democracy, I see that the three branches of government are the executive, the legislative, the judicial, the bureaucracy, the party structure, and the media.  In the next tier down are second-ranked powers, such as "the rich" so often demonized by the foolish - the upper-upper class can exert influence, but they have little in the way of direct political control.  Similarly with NGOs (non-governmental organizations) such as the Electronic Frontier Foundation, think tanks, traditional special interest groups, "big corporations", lobbyists, the voters, foreign powers with a carrot or stick to offer the US, and so on.

Which is to say that political powers do make an attempt to court the voters, but not to a noticeably greater degree than they court, say, the agricultural industry.  The voters' position on the issues is not without influence, but it is a degree of influence readily comparable to the collective influence of think tanks, and probably less than the collective influence of K Street lobbyists.  In practice, that's how it seems to work out.

The voters do have two special powers, but both of them only manifest in cases of great emotional arousal, like a comic-book superhero who can't power up until angry.  The first power is the ability to swap control of Congress (and in years divisible by 4, the Presidency) between political parties.  To the extent that the two allowed political parties are similar, this will not accomplish anything.  Also it's a rather crude threat, not like the fine-grained advice offered by think tanks or lobbyists.  There's a difference between the power to write detailed instructions on a sheet of paper, and the power to grunt and throw a big rock.

Possibly due to a coordination problem among individual politicians, the party currently in power rarely acts scared of the voters' threat.  Maybe individual politicians have an incentive to pursue their goals while the pursuing is good, since a shift in parties will not necessarily deprive them of their own seats in secure districts?  Thus, we regularly see the party in power acting arrogantly until the voters get angry and the vote swings back.  Then the minority-turned-majority party stops trying so hard to please the voters, and begins drinking the wine of power.  That's my best guess for why the balance of voters tends to be actively restored around a 50/50 ratio.

The voters' second hidden superpower can only be used if the voters get really, really, really angry, like a comic-book hero who's just seen one of their friends killed.  This is the power to overthrow the existing political structure entirely by placing a new party in charge.  There are barriers to entry that keep out third parties, which prevents the ordinary turnover visible in the history of US politics before the 1850s.  But these barriers wouldn't stop the voters if they got really, really mad.  Even tampering with electronic voting machines won't work if the vote is 90% lopsided.

And this never-used power of the voters, strangely enough, may be the most beneficial factor in democracy - because it means that although the voters are ordinarily small potatoes in the power structure, no one dares make the voters really, really, really angry.

How much of the benefit of living in a democracy is in the small influences that voters occasionally manage to exert on the political process?  And how much of that benefit is from power-wielders being too scared to act like historical kings and slaughter you on a whim?

Arguably, the chief historical improvements in living conditions have not been from voters having the influence to pass legislation which (they think) will benefit them, but, rather, from power-wielders becoming scared of doing anything too horrible to voters.  Maybe one retrodiction (I haven't checked) would be that if you looked at the history of England, you would find a smooth improvement in living conditions corresponding to a gradually more plausible threat of revolution, rather than a sharp jump following the introduction of an elected legislature.

You'll notice that my first post, on the Two-Party Swindle, worried about the tendency of voters to lose themselves in emotion and identify with "their" professional politicians.  If you think the chief benefit of living in a democracy is the hope of getting Your-Favorite-Legislation passed, then you might abandon yourself in adulation of a professional politician who wears the colors of Your-Favorite-Legislation.  Isn't that the only way you can get Your-Favorite-Legislation, the most important thing in the world, passed?

But what if the real benefit of living in a democracy, for voters, comes from their first and second superpowers?  Then by identifying with politicians, the voters become less likely to remove the politician from office.  By identifying with parties, voters become less likely to swap the party out of power, or cast it from government entirely.  Both identifications interfere with the plausible threat.  The power-wielders can get away with doing more and worse things before you turn on them.

The feature of democracies, of allowing heated color wars between voters on particular policy issues, is not likely to account, on average, for the benefits of living in a democracy.  Even if one option is genuinely better than the other, the existence of a color war implies that large efforts are being spent on tugging in either direction.

So if voters get wrapped up in color wars, identifying emotionally with "their" professional politicians and issues that put them at odds with other voters - at the expense of being less likely to get upset with "their" politician and "their" party - then shouldn't we expect the end result, on average, to be harmful to voters in general?

Coming tomorrow:  My shocking, counterintuitive suggestion for a pragmatic voting policy!  (Well... maybe not so counterintuitive as all that...)


New Comment
30 comments, sorted by Click to highlight new comments since: Today at 8:19 PM

Watch out, you're starting to sound like Mencius Moldbug.

Was just thinking the same thing.

Not much. This description is of your typical top-dog empire that got way too comfortable, not a theocracy with an uniquely insane clerical class. (For the record, I mostly agree with the former, and barely or even not with the latter).

EY: There should be a "one" in "met or two" :)

Hopefully tomorrow's suggestion will be communicable enough so I can tell it to my fellow voters, since I'm still not quite old enough myself.

"Then the minority-turned-majority party stops trying so hard to please the voters, and begins drinking the wine of power. That's my best guess for why the balance of voters tends to be actively restored around a 50/50 balance."

I think the disadvantage of the majority party is, that they are able to make mistakes. And this, rather that being corrupted by the power, drives the balance toward 50/50.

As it costs the South Carolina Democratic Party tens of thousands of dollars to place a name on the ballot, spending that money on a joke campaign would be a total loss from their perspective.

Why doesn't Colbert just run as an independent, if being a presidential candidate is so important to him?

Caledonian: I think that Steven Colbert would have paid the fee, considering that his purported reason for dropping his run as Republican candidate was the $35,000 ballot registration fee.

Also, I don't think that the point was "Does Colbert seriously want to be a presidential candidate?", but rather, "Why does the South Carolina Democratic party feel the need to dictate what a wasted vote is?"

You anger is refreshing, but did you consider that the nomination is not an election, and thus indeed belongs to the Democratic Party, not to voters? Voters have their say later. Colbert is free to register as an independent.

Eliezer, there are departments of political science at most major universities, full of academics who have been publishing on these issues for many decades. It is not clear to me that you are aware of the basic standard analyzes - you seem to be trying to reinvent it all from the ground up.


Having a computer scientist apply his narrow cognitive framework to larger questions is interesting in its way: after all, this isn't rocket science, and every thinking person has a chance of saying something insightful that the experts missed.

But what's most interesting about this engagement with the larger questions is that the author, really a near thinker, applies that thinking to far topics. This shouldn't disturb Robin Hanson, who specializes in defending the general superiority of near thought—the merits of the modes itself a far question—by means of near thinking.

The defects of the near impression that the billionaires don't wield political power is an example of the perils of modal mismatch (far subjects addressed with overly near tools). Today, when each candidate has his own billionaire keeping him afloat, even near thinking would repudiate the conclusion.

This application of near thinking to the farther reaches of politics characterizes the libertarianism both Hanson and EY espouse—and libertarianism generally. EY's inclination to exclude political issues as "mind killing" might be a safeguard against the risk of the nerds populating LW pretending they're eggheads, but for EY's gravitation to other far questions (e.g., planning for the "singularity"), which is subject to the same liability when handled by one gifted exclusively in near thinking.

Do you know where the Lincoln quotation comes from? I would be cautious about quoting it out of context. My guess is that it was a nasty remark about blacks. He made a similarly folksy riddle in the Lincoln-Douglas debates:

"Any system of argumentation that... argues me into the idea of perfect social and political equality with the negro, is a species of fantastic arrangement of words by which a man can prove a chestnut horse to be a horse chestnut."

Here's a link to my source.

Lee: Why does it matter if he made a good point in an argument that you disagree with? The quip rings true, even if he was flawed when he used it.

For example, if I say: "Most trees have leaves, therefore I hate black people." Total non-sequitur. Still, both statements might be completely true, and it would be unreasonable to disavow the fact that most trees have leaves or that I hate black people just because I used both points in a logically flawed argument.

Hanson: It is not clear to me that you are aware of the basic standard analyzes - you seem to be trying to reinvent it all from the ground up.

Can you briefly point to a specific example?

Igor and Maksym, see the next post.

Eliezer, parties and candidates may offer similar policies because they are competing with each other, and non-voters may have influence because voters listen to them and because they can inform candidates about voters. Your political posts have no citations, quite a contrast to your cognitive science posts.


Your political posts have no citations...


Paul: Tough-minded rationalism need not preclude considerations of tact. Some logical statements come with nasty historical baggage and should be avoided, especially in a political context like this one.

But again, I don't know that is the case with Eliezer's quotation. I am only urging caution about Lincoln-isms like that one in general.

On SCOTUS: My impression as a Brit is that this body is rather like the House of Lords - it's understood to act in a political way rather than just interpreting legal statutes and precedents, and indeed most of its members still very much wear their political colours, but they are appointed effectively for life by the executive, making them a long-run footprint of political control of the executive over the years (in the UK, the House of Lords is seen by some as a moderating influence because its composition doesn't change violently when there's an electoral aberration). The big differences though seem to be the much smaller number of SCOTUS members as opposed to Lords makes it a lot more variable as to how many members a given administration gets to appoint, and that SCOTUS is probably more powerful (several of their decisions are tantamount to constitutional amendments, a power SCOTUS has acquired due to the terseness of the US constitution and the aversion of the US system to amending it formally.)

On the power of the voter: I'm sure there are many out there who entertain fantasies of people going all "GRRR! ELECTORATE SMASH!" and booting out the demopublicratans. Unfortunately, if a Bayesian looked at world history he'd assign a high probability of victory in such a situation to a group based on some combination of xenophobia, religious fervour and class envy. The red mist can really bring out the worst in people when it comes to knee-jerk thinking.

On Colbert's candidacy: The Democratic Party is not an official body charged with upholding the general welfare of the American voter. It's up to the Democrats how they choose candidates. I think it's a dangerous path to start thinking of political parties as public institutions in their own right, and their primaries as being somehow the same kind of activity as a presidential election itself.


With delegates divided in accordance with the vote, South Carolina delegate slots for the Democratic convention are a scarce resource to represent the substantive preferences of the party. If Colbert were to somehow exceed the 15% minimum threshold, other state parties, possibly with different priorities, would predominate. Worse, from the state party's perspective, Colbert could encourage people who would otherwise not vote in the primaries to participate as a joke.

Primaries are not supposed to best express the popular will, but are creations of political organizations to select candidates who will advance certain policies. You could compare this to the Google founders retaining stock with multiple-voting rights in their company, with full disclosure, when selling equity to other parties.

"How many legs does a dog have, if you call a tail a leg?
Four. Calling a tail a leg doesn't make it a leg." -- Abraham Lincoln

This is the sort of quip that gives the speaker a cheap thrill of superiority, but underneath it is just a cheap trick.

In this case, the trick is that Lincoln (or whoever its real author is) has confused de dicto and de re. That is, he confuses assertions that are to be understood inside vs outside a quote-like context; in this case, in the context of the provision that we shall call a dog's tail a leg. He uses that to commit the fallacy of ambiguity. There is an undistributed middle term lurking in there, a modal operator that appears twice and needs to have the same semantics both times, and doesn't.

So I don't think this particular quote is a good illustration of "the map is not the territory". There's nothing about general semantics that forbids agreeing on or using some labelling scheme, even a variant labelling. The idea of GS is "the map is not the territory", not "use no maps" or "use no non-standard maps".

But to what degree can we really isolate -- in our minds -- the label itself from the meaning it possesses in every other instance? There's at least the danger of accidentally importing some other aspects of "legness" into our thinking about the dog's tail.

It is a cheap trick, though.

That's true, Benquo.

Eliezer proposed the idea that "if you looked at the history of England, you would find a smooth improvement in living conditions corresponding to a gradually more plausible threat of revolution".

According to Eric Hoffer, the above is true. However, the direction of causation is the reverse of what one would expect. Improved living conditions precede revolutionary sentiment.


Posting the link for "tomorrow's" post so that I can find it again next time I am here.

This article is witty and engaging, it deserves more love.

This was a really thought provoking article.

I will say that I think, in practice, the voter seems to have a larger impact on the outcomes of the democratic process then you would assume.

My hypothesis would be that if you temporally ignore all the ugliness and irrationality of the democratic process itself, and just consider it a big black box, that there is a high correlation between what voters as a whole want on any single narrow issue, and what the outcome of the black box produces. There is a significant lag time, but it does seem that if a majority of voters strongly want for something to change, and they are consistent about it for a number of years, that that outcome will change.

Just for a quick example, if you look at how voters views on gay rights in general and gay marriage have evolved over the past 20 years or so, you can see that the government's policies on the subject seem to follow the voter's will. 4 years ago, most voters were opposed to gay marriage, and so were most politicians (including President Obama). Now, a majority of voters are in favor of gay marriage, and it is happening in more states, and even Obama himself has changed his view on the issue. Other issues, like the legalization of marijuana, have gone through a similar trajectory.

You can go state-by-state as well on this. Take a contentious issue like gun control, and go across the board; states where the voters are generally more in favor of gun control tend to have more gun control (California, for example); states where voters are less in favor of gun control tend to have less gun control. To a significant extent, the output of the black box (the government policies) tends in the long run to mirror the inputs (voter opinions).

The exception is on big meta-issues, like "the size of government". The problem there is that while polls consistently say that people want a smaller government, whenever you ask them about specific issues ("should we cut the military." "Should we cut medicare." "Should we cut social security.") people consistently say "no, we shouldn't". So, from the irrationality of the voters, we get politicians who mirror them and say that they want a smaller government while refusing to cut any specific unpopular government programs.

there is a high correlation between what voters as a whole want on any single narrow issue, and what the outcome of the black box produces

I certainly agree that if I compare the results of the black-box output and reported public opinion in various U.S. jurisdictions, I'll find that shifts in the two are correlated on certain kinds of issues and not on others. I'll even find, often, that shifts in the latter predict shifts in the former (though, as you say, there's significant lag time).

I'm not convinced that the relevant distinction is "narrow" vs. "meta", but that's a longer conversation than I feel like having.

In those cases where such a correlation does exist, how would you go about distinguishing between the following theories to account for that correlation?

  • shifts in reported public opinion reflect changes in public opinion, and changes in public opinion drive changes in black-box output
  • shifts in reported public opinion drive changes in black-box output (e.g., because legislators take those reports seriously), regardless of actual public opinion
  • pollsters are able to predict shifts in black-box output, and alter their reporting of public opinion accordingly
  • the public is able to predict shifts in black-box output, and alter their opinions accordingly
  • some other set of factors we don't yet know about controls both public opinion shifts and black-box output shifts, and causes them to be correlated
  • some other set of factors we don't yet know about controls both shifts in public opinion reporting and black-box output, and causes them to be correlated

Interesting question.

I would say that a combination of the first two (Public opinion changes black-box output) and (changes in polls change black box output) are the most likely. I can think of some common political observations that seem to match those theories (for example: a President that wins an election by a large margin is considered to have more "political capital", it's easier for him to get things done in Congress in the near future.)

I guess the best way to distinguish between, say, "Public opinion changes cause political change" and "Political change causes public opinion change" is to look at which happens first. Looking at some historical examples, espcally on really big issues that people feel strongly about, it seems that there is a lead time where first people feel strongly about an issue and then, a few years later, the political outcome changes. For example, look at Prohibition. Popular opinion in favor of Prohibition grew for several years before it actually happened (it was a significant part of the progressive movement). Then, after it actually happened and people saw the flaws, popular opinion changed over time, but it took several years before Prohibition became unpopular before it was actually repealed. So there seems to be a lag time in most cases, which implies the direction of causality.

It's harder to rule out the possibility that there is one third factor that explains both. Perhaps in some cases there is some kind of general culture or media shift that affects both the popular opinion and the politician's opinion at the same time? Even if that is so, though, it still seems like a system that keeps the attitudes of the population and the end-result of the govenrment in synch is likely to be more stable then one that doesn't, so that still might not be a bad thing.

Anyway, the point of my post was just that, despite all the flaws of our democracy, that it appears that the attitudes and votes of the population as a whole seem to have a greater influence on the outcome then Eliezer was suggesting.

I agree that it's difficult to rule out the presence of common causal factors, especially in complex systems where there really are a lot of causal factors.

To my mind, that's a good reason not to assume simple causality from correlation between two factors in a complex system.

Necro - I know. However, I'd be willing to bet that few current readers have seen it and we're kind of hurting for new content, so it's probably fine to mine the archives a bit.

That being said, I really enjoyed this article. It seems to check with my own experiences reasonably well and shed some new light on the subject (for me at least). I hadn't really looked at the rack and stack of power to this level of detail nor considered closely where the power of voters really lies. It's also one of the few places where there's a good rational argument for why "Blue Team" "Green Team" is destructive (most of the other content on the site - including the fable of Green and Blue - seem to focus more heavily on the fact that it's annoying when people act irrationally rather than discussing a specific situation where that irrationality is actually harming them).

Interesting stuff.

It seems to be more noticeable in less democratic countries, I often hear the phrase "can you let the voters decide?" regarding the exclusion of a candidate or the raising of the barrier.

New to LessWrong?