The Worst Way To Fight Fake News

There’s been a lot of talk about “fake news” lately, apparently because some people blame it for Donald Trump’s election. Over at Bloomberg View, columnist Noah Feldman, who’s also a professor of constitutional and international law at Harvard, thinks it’s time to do something about it. Basically, he doesn’t think our experiment with free speech is working out:

In the free marketplace of ideas, true ideas are supposed to compete with false ones until the truth wins — at least according to a leading rationale for free speech. But what if the rise of fake news shows that, under current conditions, truth may not defeat falsehood in the market? That would start to make free speech look a whole lot less appealing.

I’ll leave the legal analysis up to the lawyers (Scott Greenfield’s review of this article is scathing), but I think the professor has an interesting analysis of the problem. I think his solution is wrong, which I’ll get to later, but his analysis is still interesting.

But to take the marketplace metaphor seriously means admitting that sometimes, markets fail. Holmes himself gave us the most famous example of market failure when he said, in a different 1919 case, Schenck v. United States, that even “the most stringent protection of free speech would not protect a man in falsely shouting fire in a theatre and causing a panic.”This thought experiment in turn led Holmes to his most famous formulation of free speech doctrine: that the question in every case is whether the words “create a clear and present danger.”

Falsely shouting fire in a theater is a perfect example of market failure in the communication of ideas. The person shouting knows he is lying — but others don’t know, and won’t have time to check. The words will cause an immediate panic, even if everybody is acting rationally, because the only logical thing to do is to get out, and get out quickly.

The “marketplace of ideas” is mostly a metaphor, not a literal marketplace, but putting aside the questionable value of Holmes’s example, this is actually a fairly good analogy to the kind of market failure that can occur because of asymmetric information between the parties.

As the Nobel Prize winning economist George Akerlof showed in his classic 1970 article, “The Market for Lemons,” asymmetric information can systematically distort the quality of what’s available in the market. In his stylized example, if good cars and lemons are both for sale, and consumers know this but don’t know which are which, they will be willing to pay the average price. That will lead the sellers to withhold the good cars, which could fetch a higher price — but that in turn will lead consumers to lower the price they are willing to pay. The resulting spiral of adverse selection leads to market failure.

As it happens, it’s a lot more expensive to generate true news stories than false ones. News requires reporting and research and institutional structures like editors and fact checkers to support them. Fake news only takes one person’s imagination. And there is certainly information asymmetry between the person who writes a story and the person who reads it. Applying the Akerlof analysis suggests that fake news could conceivably drive out true news.

This is an interesting example of applying economic thinking to a problem that is not normally considered economic in nature. “The Market for Lemons” argued that in the presence of asymmetric information about quality, buyers wouldn’t know how to identify quality products and therefore the market would not reward sellers for quality. Sellers of high-quality products would therefore leave, and the process would spiral down until only very low-quality products were trading. In the worst case, the market would disappear completely.

Ackerlof’s paper takes asymmetric knowledge about product quality as a given, but in real-world markets it’s often not clear whether asymmetric information is a serious problem. That’s not a slam on Ackerlof. He was clear about his assumptions, and there’s strong evidence that asymmetric information does cause real problems in some markets. (It rules how insurance plans are designed and sold.) But there’s also evidence that many real-world markets have found ways to circumvent the problem.

The most common solution is for sellers to try to reduce the information asymmetry by establishing and maintaining a reputation. This makes use of the fact that buyers will want to purchase the same product over and over, so if a seller has a history of producing good products, buyers can rely on that history to guide their purchasing decisions. This sets up a positive feedback loop: The seller’s reputation gives buyers confidence in the product and therefore a willingness to pay more money, and that increase in potential future revenue makes it valuable for sellers to have a good reputation, which makes it worthwhile for sellers to build a good reputation by expending the effort to produce a quality product.

Another common solution is for buyers to try to reduce the information asymmetry by relying on third parties to provide reliable assessments of product quality. I had to buy a snowblower for the first time this year, and I relied on information from knowledgeable friends, Consumer Reports, a variety of web sites, and online buyer reviews at Home Depot, Lowes, and Amazon. In this age of the internet, information is easier to find than it’s ever been.

(Other solutions, such as offering easy returns and product warranties are effective as well, but I can’t see a way to apply them to news.)

Feldman, however, doesn’t seem interested in any of these solutions. He goes straight to stepping on necks:

The classic solution to market failure is regulation. Holmes, in his fire-theater example, certainly believed that was permitted by the First Amendment.

The question is whether government regulation of fake news would be justified and lawful to fix this market failure.

Justified? No, not even under Holmes’s example. If you’re in a crowded theater and someone shouts “Fire!”, your best move is to get out as fast as possible. You don’t have time to reflect on the shouter’s claim and debate it with your fellow theatergoers. The “market failure” in shouting fire in a crowded theater is that there’s no time for the “marketplace of ideas” to operate.

Reading the news doesn’t come with that kind of urgency. There’s plenty of time to research stories and read what other people are saying about them.

Obviously, it would be better if the market would fix the problem on its own, which is why attention is now focused on Facebook and Google. But if they can’t reliably do it — and that seems possible, since algorithms aren’t (yet) fact-checkers — there might be a need for the state to step in.

Here’s where Feldman tries to play a trick on his readers. He starts by saying the problem is that Facebook and Google are unable to function as fact checkers, which is fair enough, and then he says the state should “step in.” But he doesn’t just want the government to provide the fact checking that he says is needed. He wouldn’t need to write an article about that because fact-checking isn’t legally controversial: Government employees are free to research the statements of fact within a news story and publish their evaluation, and government press offices do that all the time already.

No, what Feldman wants is for the government step into the marketplace of ideas and pick winners by force, which is why he runs into concerns about constitutionality.

Under current First Amendment doctrine, that wouldn’t be allowed. The Supreme Court has been expanding protections for knowingly false speech, not contracting it. And it would be extremely difficult to separate opinion from fact on a systematic basis.

From there, Feldman’s argument dissolves into attacking a straw man with a flurry of hand waving:

But we shouldn’t assume that the marketplace of ideas works perfectly. And given that, we shouldn’t be slavishly committed to treating the marketplace metaphor as the basic rationale for free speech.

Perfection is not the standard. I don’t think anyone believes the marketplace of ideas is perfect. But if you propose to replace the free market with something else, you should at least do your audience the courtesy of trying to explain why your proposed solution would be better, and Feldman doesn’t even try. It’s like he thinks it’s just obvious that of course government can do this.

The current freakout over false news depends on two major items of concern: (1) That fake news is produced by liars, and (2) that fake news is believed by fools. Feldman’s proposal is utterly lacking in detail, but I’d love to hear why he’s so sure that his solution will not be created and carried out by more of the same liars and fools.

For Christ’s sake, we’ve spent the last year and a half watching a gruesome demonstration of how government leaders are chosen. What in God’s name makes anyone think those people should have the final word on what’s true in the news?

False news that hinders public discussion and encourages irrationality may have a role in the marketplace; but it doesn’t contribute to the good functioning of democracy.

Speaking of democracy, politicians are notorious liars. Unsurprisingly, so are a lot of the government functionaries who work for them. I’m not talking about crazy anti-government conspiracy theories, either, I’m talking about the routine lies that government employees tell to keep their jobs and make them easier: Cops lying about incidents, experts exaggerating their credentials, and department heads who refuse to recognize facts that would be inconvenient for the continued funding of their departments.

I’m talking about the Drug Enforcement Agency refusing to recognize the medical benefits of marijuana long after its acceptance by the medical community. I’m talking about the Tuskegee syphilis experiment. I’m talking about the numbing sameness of the lies that many cops tell on the stand. I’m talking about the legal fictions that label people as drug dealers when they don’t deal drugs, as pimps when they aren’t pimping anyone, and as money launderers when they aren’t laundering money.

I’m sure Professor Feldman imagines that when the government implements his fake news suppression program they will decide which news is real or fake with the help of wise and honest scholars (such as himself). I think it’s more realistic to assume it will be staffed by people like TSA agents, DMV clerks, and those public school administrators who call the cops when a kid makes a shooting gesture with his fingers.

Deceitful Words: Disqualifying

This is the first of what I hope will be a series of posts about words and phrases that should make you suspect that someone is trying to get something past you. I’m starting with one that we heard a lot of during this last election: “Disqualifying.”

People kept saying things like:

  • “Hillary Clinton’s risking American lives by mishandling classified emails should be disqualifying.”
  • “Gary Johnson’s ‘What’s Aleppo?’ moment should disqualify him as a serious candidate.”
  • “Donald Trump should be disqualified as President for accepting support from white supremacists.”

All of these comments cite bad behavior, but “disqualifying” implies that they were a special kind of bad. We normally encounter the concept of disqualification in situations where something is being judged, and disqualification usually means that some kind of balancing mechanism is being overridden by a hard rule.

For example, equestrian competitions are scored using a complex system of faults, timing, and subjective observations, but if any rider is found to have over-used the whip or spurs to the point that a horse is bleeding, they are simply kicked out of the competition. Their score no longer matters. Another example of disqualification occurs in competitive contract bidding. Normally, the bids are compared on the basis of things like price and compliance to requirements, but bidders can also be eliminated for missing critical milestones or failing to meet certain crucial requirements regardless of price. The disqualifying condition overrides all other considerations.

Elections don’t work that way. The scoring mechanism is very loose — everyone can vote for whichever candidate they want for whatever reason they want — and the vote is the only thing that matters. There are no disqualifying events because there are no rules for how voters should judge the candidates. There are technical disqualifications for conditions like not being a citizen or being unable to perform the duties of office (because of, say, illness or incarceration), but that’s not what these cries of disqualification are about. Thus, describing some politician’s behavior as “disqualifying” is usually just wishful thinking: We’re so appalled that we hope everyone else will feel the same way about it.

But labeling a politician’s behavior as “disqualifying” can also be a disingenuous attempt to smuggle in the overriding decisiveness of disqualification to win an argument: “Sure, you have a list of ten ways your candidate is better than mine, but yours did that awful thing, and that should be disqualifying.” The hope is that others agree the behavior in question is awful, and are thus unwilling to challenge your implicit argument that the awful behavior overrides other considerations, or that they can be convinced to think of the behavior as analogous to technical disqualification.

It’s temping to try to come up with exceptions to this. For example, surely outright racism should be disqualifying, shouldn’t it? Or are even Ku Klux Klan levels of racism not disqualifying? I don’t think they should be.

Don’t get me wrong, on the scale of evil, the KKK is pretty damned awful and should lose to almost any other contender. But the key word is almost, because there are a lot of things out there worse than the KKK. The Nazis, for example. It may not be a good idea to summarily disqualify a KKK Imperial Wizard if the alternative is a Nazi with plans to exterminate millions. The abyss of totalitarian evil is very deep. Sometimes the lesser of evils is all that’s possible.

Finally, there’s a difference between doing horrible things and wanting to do horrible things. The awfulness of a candidate is going to be limited by the fact that elected officials do not have absolute power. They are embedded in a system of checks and balances, so however awful their desires, depending on their role, they may not be able to do very much damage. Thus a crazy state governor who wants to re-institute slavery or nuke Mecca, for example, can’t do nearly as much real world harm as a governor who wants to drastically disrupt services to the poor in their own state.

(The converse is also true, so it’s a fair criticism of my KKK vs. Nazis comparison to point out that the KKK might have been worse than the Nazis if they’d ever had the power.)

The point is that even something as reprehensible as outright racism may not be as bad — might not be disqualifying — when viewed in context. Consider that when former Ku Klux Klan Imperial Wizard David Duke ran for Governor of Louisiana in 1991, four percent of black voters looked at his opponent, the notoriously corrupt Edwin Edwards, and decided they’d be better off with a Klansman. Presumably they figured that the legislature and the courts would keep his racism in check.

And if that wasn’t disqualifying, nothing else should be.

Some Observations on Halderman’s Concerns About Election Hacking

Word has been going around that some computer scientists have urged Clinton to challenge the election results because of possible hacking-related voter fraud in key states:

Hillary Clinton’s campaign is being urged by a number of top computer scientists to call for a recount of vote totals in Wisconsin, Michigan and Pennsylvania, according to a source with knowledge of the request.


The scientists, among them J. Alex Halderman, the director of the University of Michigan Center for Computer Security and Society, told the Clinton campaign they believe there is a questionable trend of Clinton performing worse in counties that relied on electronic voting machines compared to paper ballots and optical scanners, according to the source.

The group informed [the Clinton campaign] that Clinton received 7% fewer votes in counties that relied on electronic voting machines, which the group said could have been hacked.

Their group told Podesta and Elias that while they had not found any evidence of hacking, the pattern needs to be looked at by an independent review.

I have a few observations:

First, as of the time I’m writing this, nobody has found actual smoking gun proof that any election machines were hacked, let alone that a significant number were.

Second, Halderman is a legitimate computer scientist and an expert on computer security. That doesn’t necessarily mean that he’s not also some kind of political hack, but in that absence of evidence of dishonesty, we should probably make a rebuttable presumption that he knows what he’s talking about.

Third, Halderman’s statistical observation, as described here, is pretty weak. It’s based on a correlation showing that Clinton received fewer votes (compared to polls) in counties that used electronic voting machines. Like all correlations, whether it has found something meaningful depends on whether co-factors have been eliminated.

In this case, the problem is that use of electronic voting machines was not randomly assigned to counties. This makes it likely that both the decision to use electronic voting machines and the decision to vote for Trump are somehow related to a third factor. For example, Nate Silver has argued that the effect goes away when you control for race and education. This suggests that (I’m just guessing) perhaps affluent well-educated white people are more likely to lie and say they didn’t vote for Trump, and affluent counties are more likely to spring for electronic voting machines. You’d want to rule out things like that before declaring that the election had been fixed.

Fourth, the preceding does not mean that Halderman is (perhaps dishonestly) leaping to conclusions. Rather, this is how scientific investigations work. You begin by doing a quick and inexpensive investigation to see if it looks there might be something interesting going on. In this case, if they’d found no correlation whatsoever between electronic voting and a deviation from the polling data, they could have pretty much ruled out hacking and moved on to investigating something else. However, because the quick statistical analysis couldn’t rule out some kind of problem, the next step is to investigate further, perhaps by using more advanced statistics, or by examining the paper trail from the vote.

Fifth, Professor Halderman says pretty much the same thing in his post on the subject. He makes it quite clear that his best guess is that the election has not been tampered with. However, given the well-known security vulnerabilities in many of our electronic voting systems — Halderman’s team has hacked real voting machines in the lab — he thinks it would be nice to examine the paper trail, just to make sure. But random college professors have no legal standing to get this done. The legal request has to come from one of the candidates, and as a practical matter, only the losing candidate has an interest in a recount. That is pretty much all that Halderman is asking for: That someone in a position to do so asks to take a look at the evidence.

It’s worth noting that computer scientists as a group are unusually skeptical about the security of electronic voting machines. (I have two degrees in computer science, and I’m skeptical.) Before the first electronic voting machines appeared, computer scientists had spent decades researching how to build secure voting systems, yet it’s clear that few real-world electronic voting machines are based on that research. I’ve also heard that voting machines are not designed and operated in keeping with modern security practices. For all those reasons, a lot of computer scientists think it’s a good idea to stick to paper ballots.


Try to remember to be kind to people.

Try to remember to be kind to people.

Try to remember to be kind to people.

Try to remember to be kind to people.

Try to remember to be kind to people.

Try to remember to be kind to people.

Try to remember to be kind to people.


This isn’t for you. It’s for me.

Election Day 2016 – Cat Noses

Continuing with my plan to post only stress-reducing cat pictures this election day, here are closeup pictures of my cats’ noses.

Closeup of Ivy's Nose
Closeup of Ivy’s Nose


Closeup of Beezle's Nose
Closeup of Beezle’s Nose


My wife uses that second one as the background on her computer and that giant cat nose cracks me up every time.