David R. Henderson  

Caplan, Kahneman, Bastiat, and the First Amendment

What is a Structurally Impaire... Tevi Troy on Think Tanks...

Bryan Caplan's post earlier this week, "Eureka! Economic Illiteracy as Mental Substitution," is one of his best ever. And that's a high bar.

Bryan applies the insight from Kahneman--that people answer the question they want to answer rather than the question that was asked--to explain the difficulty most people have in understanding economics. It also applies to many other political discussions, which should come as no surprise.

Specifically, it fits in with my discussion of Bastiat last week. I'll requote a section of the Bastiat quote that I used last week:

When we oppose subsidies, we are charged with opposing the very thing that it was proposed to subsidize and of being the enemies of all kinds of activity, because we want these activities to be voluntary and to seek their proper reward in themselves. Thus, if we ask that the state not intervene, by taxation, in religious matters, we are atheists. If we ask that the state not intervene, by taxation, in education, then we hate enlightenment.

Basically, it comes down to whether you're willing and able to make distinctions. I don't know about most people's ability; but we see ample evidence of most people's lack of willingness.

We are used to this from undergrads in economics classes. But we also see it in newspaper reports. One news item in today's Monterey County Herald is the story about a First Amendment case that the Supreme Court is considering. The story is about TV broadcasters asking the Supreme Court to throw out "decency standards" for TV and radio. Recall that the entire content of the First Amendment is:

Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.

Each day, the Monterey County Herald chooses one news item to which to attach a "Herald Question of the Day." What would be a good question for such a news item? Oh, I don't know, maybe, "Do you think broadcasters should be free to broadcast indecent language or indecent images?" That's what's at issue.

But what does the Herald choose as the question? Here's what:

Are you offended by the amount of nudity and curse words on TV?

I know my answer to that question: absolutely, I find it disgusting. I hate it when louis ck, for example, spices up his great, relatively-clean routine with disgusting language. I could give many more examples.

But my answer to that question has nothing to do with the case before the Supremes. The Herald has substituted, to use Kahneman's and Caplan's language, an easy-to-answer question for the real question at hand.

It gets worse. It's not only the Herald that substitutes; it's also many of the Supreme Court justices who are hearing the case. Here's what Chief Justice John Roberts said:

All we are asking for, what the government is asking for, is a few channels where ... they are not going to hear the S-word, the F-word, they are not going to see nudity.

Now there's some scary content. All we are asking for? Roberts apparently caught himself saying that he's not just taking the government's side in a contested case but is the government's side.
"Asking for?" I do not think the word "asking" means what Roberts thinks it means. No one is asking. The issue is whether the government gets to tell. If all Roberts, oops, I mean "the government" was doing was asking, this would not be a Supreme Court case. In fact, it wouldn't be a legal case at all.
And, finally, note that Roberts discusses the case entirely in terms of what's decent, what our kids will see, etc. and not at in terms of the actual principle that he swore to uphold. Now that's Kahneman-style substitution.

Comments and Sharing

COMMENTS (26 to date)
Silas Barta writes:

Regarding the issue of "asking" vs "legally requiring" and heuristics that blur the distinction, I thought I might link this recent Daniel_Kuehn post that has a picture joke regarding the same distinction.

Someone sprays the graffiti, "spread anarchy", and someone else crosses it out and sprays "don't tell me what to do" -- obviously misinterpreting what is being asked, which is the joke.

This is another case of the same thing: treat the heuristic question "do I like being told what to do?" as (somehow!) equivalent to "is anarchy better than the current political system?". Quite a leap!

Chris Koresko writes:

Chief Justice John Roberts: All we are asking for, what the government is asking for, is a few channels where ... they are not going to hear the S-word, the F-word, they are not going to see nudity.

Agreed, that is scary. I hope he said it without thinking it through. My impression of Roberts is that he is a deep and careful thinker, very smart, and relatively conservative. If that impression is right, his judicial opinion probably won't reflect the words quoted above.

I think it takes a very disciplined mind to avoid dropping into "fast-thinking" mode at inappropriate times.

This discussion relates to some earlier threads about the value of math in economics. I think one of the most important contributions math can make is to force you into slow-thinking mode, making assumptions and logic very explicit so that it's harder to base your conclusions on assumptions you don't realize you're making.

Norman Pfyster writes:

You realize that you just engaged in the same kind of thinking you were decrying, right? If you knew anything about First Amendment jurisprudence, you would be aware of the many distinctions that exist in that body of law.

Hume writes:

What Norman said is correct. Precedents matter to Supreme Court justices.

Also, the "asking" vs. "telling" distinction is not as clear cut as some may think it is, at least not in the example made by Silas above. This point is made pretty clear by, for example, Gerald Gaus. When we make a moral claim on someone else, we are not "asking" that they, e.g., refrain from lying or cheating on their girlfriends. We are telling them that they have no choice in the matter, that they are morally required or obligated to refrain from the action. When we "ask" someone not to do something, we are presenting the issue as an option that the actor is normatively free to accept or decline.

The demandingness of moral claims gets muddled because people sometimes associate you "must" do X with "if you dont do X, I will use force/coercion to make you do X." This, of course, is not the case. I can "tell" you to do X, I can claim that you morally "must" do X, yet I am not thereby committed to the claim that I will force/coerce you to do X.

Silas Barta writes:

@Hume: Are you saying you don't understand the joke in the picture I linked? Or the difference between "spread anarchy" and telling someone what to do (in the standard usage of the term)?

Hume writes:


(1) Yes I get the joke and I understand your interpretation of it.

(2) Yes I understand your interpretation of the difference between "spread anarchy" and telling someone what to do (in the standard usage of the term).

My point is this. When one interprets "spread anarchy" as an instance of "asking" others to do something, you are interpreting that statement as non-normative. This interpretation, however, is not exclusive. There are at least four ways to interpret "spread anarchy":

(1) the person is "asking" others to spread a certain ideology or political program. This is non-normative and is akin to saying "I subjectively prefer it if you would spread anarchy, so I would like you to do so."

(2) the person is making a normative claim on others. She is claiming that you have a moral obligation to spread anarchy. From the moral point of view, you must do so whether or not you are inclined.

(3) the person is making a normative claim that you must spread anarchy coupled with the claim that she is morally permitted to coerce you if you fail to live up to your obligations.

(4) the person is making a normative claim coupled with a claim to authority. This is a content-independent claim and looks something like this: "I have the moral right to issue morally binding directives; therefore, when I promulgate a directive you are morally obligated to obey. My directive is to spread anarchy, so you are thereby bound."

Socal Bill writes:

So David, do you think broadcasters should be free to broadcast indecent language or indecent images?

David R. Henderson writes:

@Socal Bill,

David R. Henderson writes:

@Norman Pfyster,
I don’t think you’re right. At no point in my post did I suggest or implicitly assume that there is no body of law making various distinctions.
Your comment would be more helpful if you can quote anything in my post in which you think I did.

Hume writes:


Do you think people should be able to stand on the road in front of your home and throughout the night, all night, play Michael Bolton cover songs on amplifiers that go up to 11?

I realize that this is disanalogous, and I agree with your moral/political stance, but in the context of a discussion of Kahneman, the analogy exemplifies that the First Amendment's proscription of laws violating "freedom of speech" has been interpreted to make exceptions to such an all-or-nothing concept. This is what SCOTUS has done with "obscene" speech, so it is misleading not to mention the "Miller Test" in this context, instead offering an account of cognitive biases. The Miller Test, as part of positive U.S. constitutional law, defines a work as "obscene" if, taken as a whole, it (a) appeals to the prurient interest in sex, (b) portrays, in a patently offensive way, sexual conduct specifically defined by applicable state law; and (c) taken as a whole, lacks serious literary, artistic, political, or scientific value. So asking whether someone is "offended" seems relevant, from the legal point of view, as to whether or not something is "offensive."

Whether you find all of this politically illegitimate is irrelevant from the legal standpoint (unless, of course, you believe in some sort of Dworkinian natural law theory of the concept of law. If so, good luck with that).

Silas Barta writes:

@Hume: So you don't understand how the "spread anarchy" is intended to mean something like 1-2, while it is being (deliberately) misinterpreted to mean 3-4?

That would go against your claim to understand the joke, is all.

David R. Henderson writes:

Hume asks:
Do you think people should be able to stand on the road in front of your home and throughout the night, all night, play Michael Bolton cover songs on amplifiers that go up to 11?
I’m not sure why you ask that, but my answer would depend on who owns the road.
I don’t see much of a connection with the case at hand. To make the analogy, you would need to add that I can turn the volume down to zero. If I didn’t like Bolton songs or liked them but had enough of them, that’s what I would do. Remember that people can always turn off a TV set.
And what I’m saying is relevant or, of course, I wouldn’t have said it. Pointing to a bad decision made 39 years ago doesn’t persuade me.

Hume writes:


I completely agree with from a moral/political standpoint. But as a lawyer (and now graduate philosophy student) and a subscriber to legal positivism, I can tell you that in the American legal system, SCOTUS precendent is legally relevant. If you ignore it, you are a bad lawyer and will give bad legal advice.

Remember, this is a discussion of cognitive biases and not political philosophy or ideal constitutional interpretive methodology. It is a social practice among judges and SCOTUS justices to give weight (sometimes over-riding weight) to precedent. To say that they are illustrating cognitive bias by asking "what is offensive" instead of some other standard, and thus are not paying attention to the correct legal standard is simply false or greatly misleading.

"I don’t see much of a connection with the case at hand." I said above that the analogy is bad, and I will add that it is bad if we are talking moral/political philosophy. The purpose of the analogy was simply to exemplify that "shall make no law" does not actually mean in practice "no" law. Exceptions have been made. If you object and claim that my example is bad because it depends on who owns the road, I think you are either (1) being difficult (most roads are in actuality publicly owned), or (2) exhibiting cognitive biases by replacing the question "are there laws that forbid playing loud music at night? with the question "should there be laws that forbid playing loud music at night?"

Joel writes:


I think the point in the visual joke above is that anarchy by definition involves not telling people what to do. So the individual’s response of “don't tell me what to do” misinterprets the original statement, as if the vandal truly wanted to spread anarchy, he would be against anyone telling people what to do. At least that’s how I see it.

James writes:


Your last comment illustrates one of the biggest problems with legal positivism. As you point out, in practice "shall make no law" means something just the opposite, more like "shall make some laws." How can you then base a legal analysis on the assumption that the law is whatever the state says when we can't even tell what the state is actually saying?

I'm sure it would not be successful, but I'd love to see a broadcaster defend against obscenity charges by pointing out that if "shall make no law" really means "shall make laws," a reasonable person might take this as precedent for interpreting "shall utter no profanity" to mean "shall utter profanity."

Brandon Berg writes:

Apparently Roberts has never actually watched television. There are already many channels which don't allow swearing or nudity. The ones that do are actually in the minority.

Brandon Berg writes:

Apparently Roberts has never actually watched television. There are already many channels which don't allow swearing or nudity. The ones that do are actually in the minority.

David R. Henderson writes:

@Brandon Berg,
I think you misunderstand Roberts. He says there are many such channels (that don’t allow swearing or nudity) but sees that as due to current government regulations rather than as simply a marketplace outcome.

David R. Henderson writes:

@Hume and James,
Interesting thoughts. I’ll have to noodle them both. Right now I’m leaning in James’s direction.

Hume writes:

"How can you then base a legal analysis on the assumption that the law is whatever the state says when we can't even tell what the state is actually saying?"

There are numerous problems with this question/claim as an objection to legal positivism.

First, a practical point. Laws change. At time T1 the law can require actions x, y, and z. At time T2, laws can require actions -x, y, and q. A fundamental fact about "law" is that it is amendable. Thus, at one point in time the law regarding speech required x, y, and z (perhaps at this time, "shall make no laws" meant "shall make no laws"). But at a later point in time, the law regarding speech has changed, and we the law-subjects can recognize this. We do it all the time and lawyers giving legal advice know where to look in order to advise their clients as to what the law is at a particular time. Included within this "legal material" is judicial precedent. To the extent that it is impossible to know what "the law" is, many legal positivists (see, e.g., Joseph Raz, Scott Shapiro's Practical Difference Thesis) argue that there is simply no law on the subject. This does not mean absolute discretion if a case comes before a court, for the body of law can guide decision-making within predictable routes. But to the extent that the previous body of legal material cannot be figured out, there simply is no law and the judge legislates (this is not to say that the judge acts like a legislature, that is the point of saying that the judge is guided in certain directions).

Second, legal positivists do not claim that "the law is whatever the State says it is." Courts can be wrong (Hart spends time discussing this against the early "realist" claim that the law is whatever judges say it is), the police can be wrong, etc. A legal system is a system of rules, standards, etc., and whether some norm is a legal norm is determined by whether it is in certain relations with the other legal norms of the system (Raz's Practical Reason and Norms is very helpful in analyzing this aspect of legal systems, as is his The Concept of a Legal System, although the latter is extremely tedious and painful to read).

That being said, legal norms have to come from somewhere. So this is where the rubber hits the road. Legal positivism gains its credibility from recognizing that laws exist and exert influence on people. Why is this the case? Where do laws come from? How can we tell whether a norm is a norm of a legal system or a norm of some other system? On my account, the concept "law" is intimately related to the institutions that possess the ability to enforce their norms. I differ from Raz and many other contemporary positivists on this point (Raz has us imagine the "society of angels" in arguing that coercion is not a necessary concept for understanding the nature of law), for they tend to emphasize legislative institutions and adjudicative institutions. I believe that such norm-creating institutions (legislatures) and norm-applying institutions (courts) are essential, but so are enforcement institutions.

Either way, when we think of "the law" we think of legislatures and courts. So to the extent that these institutions are salient in our concept of "the State" (as in your objection "the law is whatever the State says it is"), then "the State" is extremely important in determining what "the law" in any given legal system actually is. This presupposes that "the State" is made up of norm-creating and norm-applying institutions, it presupposes that there are people doing things, that is, creating rules and applying them either in their day-to-day life as a standard of behavior or in court during dispute resolutions.

Hart's theory of the rule of recognition is fundamental in understanding why it is both correct and incorrect to say that the law is whatever the State says it is. This rule of recognition provides Hart’s explanation of propositions that assert legal authority. When one is confronted with a rule claiming legal validity (“rule R”), a question arises: “what gives rule R legal authority?” A common response (perhaps by an official demanding conformity to the rule) is to point to some other rule: “statute X confers legal authority on rule R.” But again, one may ask “well, what gives statute X legal authority?” Such a question usually generates a similar response: “statute Y confers legal authority on statute X (which in turn confers legal authority on rule R).” This process could go on ad infinitum. The rule of recognition, for Hart, is the backstop to this process. It provides validity conditions the possession of which authoritatively mark a norm as a law of the community. At some point, if one were to ask “why is this rule legally authoritative?” the answer would be that it just is. The rule of recognition is simply accepted as appropriate for determining the legal validity of other norms. Thus, it is an “ultimate” rule that is neither valid nor invalid; it is the measuring stick for determining the validity of other norms.

On this account, at the foundation of every legal system is a social rule that is manifested in the general practice of identifying rules as the laws of a community. Therefore, the existence of a rule of recognition in a given community requires that certain social facts obtain: certain individuals or groups must take an internal point of view towards the rule by accepting it as a norm and using it to assess the validity of other norms. That is, certain individuals and groups believe that certain criteria is the appropriate criteria for determining whether a norm is a law of the system. This belief manifests itself when these individuals or groups use these criteria in validation of rules as laws and apply them accordingly. Thus, Hart states that “the rule of recognition exists only as a complex … practice of the courts, officials, and private persons in identifying the law by reference to certain criteria. Its existence is a matter of fact.”

Thus, it is in some sense correct to say that "the law is whatever the State says it is" in the sense that the State is fundamentally important in locating those officials we recognize as "legal" officials and thus whose practice is constitutive of the rule of recognition. It is incorrect because the acts and legal interpretations of particular legal officials that do not bring about a change in this practice and are not followed by anyone else in the syatem are not accorded legal status. Similarly, if the supposed "law subjects" never use the rules claimed to be "laws" and are never punished for violating these rules, we are likely talking about something other than law (perhaps the ramblings of a law professor's model legal system).

Liam McDonald writes:

@ Hume.

You had me at "a Practical Point". (in other words I stopped reading around there)

Liam McDonald writes:

Here's the thing.

The Constitution states "Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech." which is exactly what they do when they say you can't say "shit" on Sesame Street. But who in their right mind would say use obscenities on a Children's Show or a Family Show? If you did you would lose advertisers and then show would no longer air. All of that without Government intervention!

But since these are broadcasters they presumably know which audiences are watching what and they are just following the Benjamins. They will have channels that suit various people and leave it up to the people to decide for themselves. Why should Comedy Central have to censor words that millions and millions of people use everyday because the Government’s “decency standards”? Because a child might watch them? Because Grandma doesn’t like hearing them cuss words?

So Hume, when you start quoting legal precedent and Hart’s Theory and Kramer vs. Kramer, you are missing the fundamental argument that the Government infringes Freedom of Speech when they impose their own standards of decency as well as making a perfect case to why lawyers have no sense of humor and should never be invited to parties.

Oh, and the "Spread Anarchy" joke is funny

Hume writes:


Thanks for the response. This discussion started with David's claim that discussions of whether something "offends" another exemplify cognitive biases (substitution and "whether you are willing to make distinctions"). In other words, these people are substituting a question that it irrelevant to the task at hand (whether X is offensive) for something that is relevant ("'Do you think broadcasters should be free to broadcast indecent language or indecent images?' That's what's at issue"). My point was simply that the law on the subject, as per SCOTUS precedent, instructs one to explore what is or is not offensive. As such, this is not a case of cognitive bias.

In an ironic twist (to me anyway), it appears David has substituted a legally irrelevant question of political philosophy ('should' be free to broadcast indecent language) for the legally relevant question ("offensive" as per SCOTUS precedent). If a willingness to make distinctions is relevant to this discussion of cognitive biases, then some here are unwilling to make the distinction between "Freedom of Speech" as a moral/political concept and "Freedom of Speech" as a technical legal concept.

In a reply, David made a statement that can be interpreted as either (1) "I dont care about some stupid old case" (political philosophy), or (2) "that old case is irrelevant" (legal claim). If his claim is interpreted as (1), then he has substituted political philosophy for law. If his claim is (2), then he has made what I believe an implausible claim that precedent is legally irrelevant. It is the latter claim that required me to delve into legal philosophy against the charge that legal positivism naively asserts that the law is whatever the courts say it is (I apologize for boring you with details and not being more festive when engaging in a discussion on a blog that fancies itself more than just off-the-cuff talk of folk economics).

Ken B writes:

I think Bastiat's seen and unseen fits the economics case better, and more specifically. The consequnces of the incentives created, and usually the incentives too, are often quite, quite unseen by policy advocates. That fits economics, but it doesn't fit math. The simpler question substitution fits teaching probably every subject.

David R. Henderson writes:

I think you stated it really well in your comment at 10:47 AM. I get it now. As one of my professors, Leland Yeager, used to say when one of us won an argument with him, touche.

Liam McDonald writes:


Yes, you're point of the legal viewpoint on determining the definition of "offensive" is apparent. However my point is that a determination of what is offensive should be moot since the concept of free speech is that it is the individual's (or group as the case may be) ability to police themselves in regard to what is and is not offensive and not the Government's.

Case in point, my previous comment contained a profanity. The group which monitors this site withheld my comment until they reviewed it and determined whether to allow it based on it's context. This is exactly how it should be. I am free to use profanity just as they are free not to publish what they consider offensive.

(TOTH to Lauren Landsberg)

I believe that broadcaster's are seeking the same freedom. Also note that I do not say, "seeking the same right" as there is a fundamental difference (IMHO) between a right and a freedom. For example, it is your right to work in a smoke free environment but I am free to smoke.

Oh, and I don't think you were being boring. I was just taking some lighthearted jabs, as is my wont.

Comments for this entry have been closed
Return to top