Bryan Caplan reminds us of a great old puzzle: why are libel, slander, and blackmail illegal? Bryan and I find it easier to understand two extreme positions than the actual intermediate mixture we have. The extremes:
Institutions almost always punish falsehood although the manner differs. Religion punishes through doctrinal councils and excommunication, politics through parties and elections, science through peer review and rejection, the law through courts and penalties, markets through law and judgments, though there exist unsettled areas within them where listener beware operates until settled. Only inter institutional and interpersonal relations open themselves to arbitration in the absence of which listener beware rules.
I strongly agree with UlrichRoarke: this post (including the comments) has lots and lots of bad law.
It's a good thing it's not illegal to say untrue things about the law in a blog post, or a lot of people participating in this thread would be in trouble.
For instance, a commenter says: "in the USA factually true statements, as well as statements of personal opinion, are essentially immune." No. Truth is a defense to libel and slander, but there are other torts you can be liable for even based on making true statements about someone. If I procure your medical records and publish them in a newsletter, I may well have probably broken the law; you'd need to research the applicable state tort law.
More fundamentally: the post itself is assuming the law must be based on "bias." ("Surely there are some implicit bias theories behind these rules...") You're saying that because that's the theme of this blog. But there are innumerable pages worth of judicial reasoning in American case law explaining why we draw the lines where we do. Now, that reasoning may be open to criticism -- I'm not saying we have to go along with it. But why not give one single example of a Supreme Court case (or a law review article, etc.) on a specific 1st Amendment issue? There's been so much written about why you give special protection to political speech, why it's OK to heavily regulate commercial speech, etc.
When the blogger says, "until these theories are made more explicit it remains hard to evaluate how much sense these rules make," I wonder: "more explicit" THAN WHAT? Do you realize that when a 1st Amendment issue gets decided, it's frequently accompanied by a fairly nuanced written opinion going through the relevant concerns, often at great length? And even if you think judicial opinions are too focused on specific legal issues (though they can include lots of tangential material), what about scholarly articles, treatises, and law school casebooks? Have you seriously examined these before concluding that the theories underlying the regulation of speech needs to be "more explicit"?
Again, I'm not endorsing everything those theorists have written -- they disagree amongst each other sometimes, so they're not always right. But a more illuminating post would pick out something they've written and explain what it gets wrong, not simply assume that no one has given an adequate explanation.
There is a LOT of bad legal analysis going on here. Most of it is simply wrong. An example in the post itself is that you CAN be paid not to tell the truth, e.g. you can contract away your right to disclose that you were paid by X Corporation for your injuries. You can sign a non-disclosure agreement as to almost anything. Also, trial character witnesses may, in fact, testify to someone's general reputation in the community; so, even if it's true that a witness can't say, "Simon told me the Defendant kicked a puppy," he can say, "Defendant has a reputation for violence against animals."
I think you and Brian need a Con law expert, who will provide the real constitutional/First Amendment limits, both criminal and civil, for libel, slander, and blackmail. I think a great debate topic would be on "group libel." The Supreme Court has strongly indicated that a group cannot sue for libel, i.e. Jews cannot sue if someone publishes an article on how the blood libel is becoming popular again. See, e.g., http://supreme.justia.com/c.... Is that a good rule?
That is what was initially requested. Recall what Robin initially wrote:
Surely there are some implicit bias theories behind these rules, but until these theories are made more explicit it remains hard to evaluate how much sense these rules make.
Deceptions or attempt at deception is seen as immoral or unethical. People punish virtually all types of deception, even if it is just a lost of trust. Deception is seen as a public moral issue yet it is extremely common, and the consequences can be from negligible to very high. For authorities to regulate even a fraction of it would be impractical. So we usually see high consequences and easily identified deception regulated by law, like medical claims and truth in advertising. A more difficult identified from of deception in courts is also regulated because of the high consequences, courts usually have the power to imprison or expropriate assets according to testimony.
But they aren't special at all, but common to all market transactions. Misrepresentation is fraud. They apply equally to financial claims such as Madoff to selling tilapia as red snapper. If legal process is considered the search for truth, then falsehood cannot be tolerated, and as market disputes are settled legally, falsehood couldn't be tolerated there either or the we would have the defense 'the sucker believed me'. Science as the search for truth won't tolerate it either although the process is not a legal one, though the law will draw on science as a resource in adjudicating its disputes. Libel and slander are frauds on the market.
Still, they are in two areas that have already been identified as special (law and medicine). In particular, if we view the rules of medical speech as regulatory capture by a cartel, we should not expect those rules to be well-explained by avoiding harm to the general public.
Pointed direct harm to an individual is most relevant. One can certainly make all sorts of claims without punishment. 'Vaccines cause autism' has been one. Only identifying which vaccine would be actionable. Harm against an individual relying on it is less, but can come into play especially when there are commercial considerations. 'Laetril cures cancer' is alright for personal expression, but would not be tolerated commercially. There aren't, or shouldn't be, markets in libel, slander, and blackmail, but they have market implications. Punish falsehood applies very much to markets to reduce fraud and reinforce honesty. Opinions take place beyond markets and listener beware would typically apply. Opinions stated in court would take on aspects of libel and slander even if unintentional. Of course people have been sued for opinions even when they are presumably free to express them.
Mark Spottswood: Yes, I deliberately did not specify American law in that statement. It's my understanding (with only moderate confidence here), that US law is the outlier in this regard, and that in most European countries slandering someone with deliberate and malicious intent is potentially illegal, even if no falsehoods are stated (though it may strengthen the defendant's case). In contrast, in the USA factually true statements, as well as statements of personal opinion, are essentially immune. Or, at least, that's my understanding.
Blackmail seems to me rather simple to explain in terms of the incentives of powerful people. There's not much use in blackmailing Joe Average; even if you can prove he cheated on his wife, he doesn't have a lot of money to give you to avoid the consequences. So blackmail is an activity that, by and large, transfers money from rich (and/or powerful) people to poorer ones.
That said, it also seems to me that blackmail is a rather unusual sort of transaction, because it is very difficult to enforce. Suppose we agree that you will pay me $100 a year, and I will remain quiet about that time you got drunk at my pool party. How can this contract be enforced? As soon as the facts are laid before a judge, the secret is out! Even if the details are kept confidential, as Eliezer pointed out the other day, "the most important part of a secret is that a secret exists". Just knowing that you have a secret you are willing to pay to protect will cause most of the difficulties you are trying to avoid by protecting it, because people will insist on speculating. (And the speculations will likely be worse than your actual behaviour at the pool party!)
Now, if I could credibly and verifiably commit to actually forgetting about the pool party, in exchange for a one-time payment - editing my memories - that would be different.
there seems to be a lot of rationalisation of the status quo going on in the comments here
That is what was initially requested. Recall what Robin initially wrote:
Surely there are some implicit bias theories behind these rules, but until these theories are made more explicit it remains hard to evaluate how much sense these rules make.
Robin was asking for people to make explicit the implicit bias theories behind the status quo rules. By definition, those theories necessarily rationalize the status quo.
Why is this a puzzle? From either a monkey or a political theory perspective?
It's obvious from a monkey-tribe outlook - libel/slander speech acts (which by definition are untrue) attack & disrupt the social hierarchy and status machinery, a prized property of monkey society. If you lie about some element of my status, you've harmed one of my most important pieces of monkey capital and injured society overall by making it more difficult to rank status "correctly."
As for Rothbard, it is likewise obvious, yes? Rothbard's axiom is about violence & property. He often states it in a form similar to "the fundamental axiom of libertarian theory is that no one may threaten or commit violence ("aggress") against another man's person or property."
Until the invention of the consumer credit rating, which really was artifact of the 1960s as credit cards became widespread, one of a man's most important possessions was his public commercial reputation. Libel & slander laws - defamation laws - date back to Roman times and were created precisely to protect a man's property in his good public reputation.
Until the industrial revolution, most Western men had little or no tangible property - no land, no houses - perhaps only small livestock, some small furniture, and his chattel wife and children. In this environment, his intangible public reputation was actually his most valuable piece of property.
Thus it seems clear that defamation laws were an important piece of property protection, and remain so even to this day. Truth has long been considered a defense against defamation claims in the US.
Understanding this as background, Listener Beware seems appropriate when the economic burden would fall on the Listener, or when there is an ability to access accepted objective information, such as a credit score. Punish Falsehood seems appropriate when overall social harm is possible, the economic burden would fall on the person defamed, or on innocent third parties and there is no way to objectively check information.
Tom P. said: To simplify, in the US we have "listener beware" for political and religious claims and "punish falsehood" for other claims.
I appreciate to simplify, but I think this summary is misleading. We have, in fact, four primary regimes:
1. Listener Beware: Various non-verifiable statements, including religious and normative assertions. This applies to some political discourse, but not all; many political claims are in essence factual and can be the subject of defamation liability.
2. Punish Insincerity: This applies to factual assertions about political or other public figures, and (in essence, if not as a matter of formal doctrine) to most other speech on matters of public concern. (I would group the various "negligent misstatement" torts in this category for simplicity's sake, only because it will be rare for people to sincerely believe what they have no reason to believe, at least with respect to the verifiable matters that fall outside of the "listener beware" rule.) So understood, this rule also applies to perjury prosecutions, to almost all fraud claims and to many malpractice lawsuits premised on incorrect expert advice.
3. Punish Falsehoods (even if sincere): We apply this rule only to "commercial speech," a hard-to-define category that most often involves advertising or attempts to negotiate commercial transactions. Even here, there aren't many cases in which this rule is applied to its fullest extent (that is, to cases involving honest errors by a commercial speaker).
4. Mandate Disclosure of Underlying Data: we use this rule in the trial process and in securities law, for example. We often use it in conjunction with another rule; for instance, we apply rules 2 and 4 during trials (lies are punished as is the withholding of certain underlying information), while not punishing errors that involve neither a failure to disclose or an attempt to deceive.
So Robin is right to say we mostly have a blended regime. Our pluralism about normative and religious matters---combined with a desire not to stifle public debate on these issues---probably explains why we have category one. Mostly, we use category 2, which makes good sense if you think that (a) punishing honest errors will deter too much truthful speech, both directly (because even truthful speakers will decline to express themselves for fear of liability) and indirectly (because deterring false speech will also deter the truthful counterspeech that might better educate an audience), while also thinking that (b) deliberate lies cause more harm the honest errors (because they are harder to cure with effective counterspeech) but banning them will deter less truthful speech than would the "punish falsehoods" rule. In other words, our middle ground is designed to strike a balance between promoting as much truthful speech as possible, on the one hand, and deterring as much false speech as possible, on the other.
As to category 3: our use of this for commercial speech mostly stems from an empirically dubious assumption that commercial speakers are less likely to be deterred by liability than other speakers. I think this assumption is wrong (indeed, commercial speakers may be more likely to rationally weigh the costs and benefits of liability before speaking), so I generally advocate getting rid of the "punish falsehoods" rule except in a few narrow circumstances where the harms of false speech are very high and rebuttal speech is unlikely to occur. And even in those circumstances, a combination of 2 and 4 should normally work better than 3.
Most democracies do not have criminal defamation laws; rather these issues are raised in civil courts. In those cases they must be able to demonstrate that there was a damage created by a malicious intent - think negative-advertising where I just tell lies about your product loudly and frequently enough that many people believe it. Similarly as an individual I might attempt to discredit someone's reputation at work to get them fired so I can get their job.
I don't understand why people think words cannot create real harms or culpability.
In the US, alcohol companies may not buy TV ads truthfully saying most studies find people who drink more are healthier, and trial witnesses may not truthfully tell rumors they've heard about the accused.You disappoint me a bit here, Robin. The reasons for this are obvious, and unrelated to your main points - the first is because alcohol companies are allowed to advertise only on sufference anyway, and the second is because repeating hearsay in court is very pernicious to determining a correct sentence (especially if this is instiutionalised).
Blackmail typically increases total punishment, even of criminal activity.Of course; so blackmailing about crimes is a mixture of vigilanticism with concealment of the crime. All the more reason for the state to hate it.
Don't take this as rude, but there seems to be a lot of rationalisation of the status quo going on in the comments here. My "analysis" included, but at least my opinions have changed as a result of Robin bringing this up - I understand the banning of blackmail better, and (with the exception of witnesses at trials and scientists) would now favour less libel laws than I did before.
Have others had their opinions changed (even slightly) by this blog post?
Institutions almost always punish falsehood although the manner differs. Religion punishes through doctrinal councils and excommunication, politics through parties and elections, science through peer review and rejection, the law through courts and penalties, markets through law and judgments, though there exist unsettled areas within them where listener beware operates until settled. Only inter institutional and interpersonal relations open themselves to arbitration in the absence of which listener beware rules.
I strongly agree with UlrichRoarke: this post (including the comments) has lots and lots of bad law.
It's a good thing it's not illegal to say untrue things about the law in a blog post, or a lot of people participating in this thread would be in trouble.
For instance, a commenter says: "in the USA factually true statements, as well as statements of personal opinion, are essentially immune." No. Truth is a defense to libel and slander, but there are other torts you can be liable for even based on making true statements about someone. If I procure your medical records and publish them in a newsletter, I may well have probably broken the law; you'd need to research the applicable state tort law.
More fundamentally: the post itself is assuming the law must be based on "bias." ("Surely there are some implicit bias theories behind these rules...") You're saying that because that's the theme of this blog. But there are innumerable pages worth of judicial reasoning in American case law explaining why we draw the lines where we do. Now, that reasoning may be open to criticism -- I'm not saying we have to go along with it. But why not give one single example of a Supreme Court case (or a law review article, etc.) on a specific 1st Amendment issue? There's been so much written about why you give special protection to political speech, why it's OK to heavily regulate commercial speech, etc.
When the blogger says, "until these theories are made more explicit it remains hard to evaluate how much sense these rules make," I wonder: "more explicit" THAN WHAT? Do you realize that when a 1st Amendment issue gets decided, it's frequently accompanied by a fairly nuanced written opinion going through the relevant concerns, often at great length? And even if you think judicial opinions are too focused on specific legal issues (though they can include lots of tangential material), what about scholarly articles, treatises, and law school casebooks? Have you seriously examined these before concluding that the theories underlying the regulation of speech needs to be "more explicit"?
Again, I'm not endorsing everything those theorists have written -- they disagree amongst each other sometimes, so they're not always right. But a more illuminating post would pick out something they've written and explain what it gets wrong, not simply assume that no one has given an adequate explanation.
There is a LOT of bad legal analysis going on here. Most of it is simply wrong. An example in the post itself is that you CAN be paid not to tell the truth, e.g. you can contract away your right to disclose that you were paid by X Corporation for your injuries. You can sign a non-disclosure agreement as to almost anything. Also, trial character witnesses may, in fact, testify to someone's general reputation in the community; so, even if it's true that a witness can't say, "Simon told me the Defendant kicked a puppy," he can say, "Defendant has a reputation for violence against animals."
I think you and Brian need a Con law expert, who will provide the real constitutional/First Amendment limits, both criminal and civil, for libel, slander, and blackmail. I think a great debate topic would be on "group libel." The Supreme Court has strongly indicated that a group cannot sue for libel, i.e. Jews cannot sue if someone publishes an article on how the blood libel is becoming popular again. See, e.g., http://supreme.justia.com/c.... Is that a good rule?
That is what was initially requested. Recall what Robin initially wrote:
Surely there are some implicit bias theories behind these rules, but until these theories are made more explicit it remains hard to evaluate how much sense these rules make.
Ah yes. Me bad.
Deceptions or attempt at deception is seen as immoral or unethical. People punish virtually all types of deception, even if it is just a lost of trust. Deception is seen as a public moral issue yet it is extremely common, and the consequences can be from negligible to very high. For authorities to regulate even a fraction of it would be impractical. So we usually see high consequences and easily identified deception regulated by law, like medical claims and truth in advertising. A more difficult identified from of deception in courts is also regulated because of the high consequences, courts usually have the power to imprison or expropriate assets according to testimony.
But they aren't special at all, but common to all market transactions. Misrepresentation is fraud. They apply equally to financial claims such as Madoff to selling tilapia as red snapper. If legal process is considered the search for truth, then falsehood cannot be tolerated, and as market disputes are settled legally, falsehood couldn't be tolerated there either or the we would have the defense 'the sucker believed me'. Science as the search for truth won't tolerate it either although the process is not a legal one, though the law will draw on science as a resource in adjudicating its disputes. Libel and slander are frauds on the market.
The added examples are very good.
Still, they are in two areas that have already been identified as special (law and medicine). In particular, if we view the rules of medical speech as regulatory capture by a cartel, we should not expect those rules to be well-explained by avoiding harm to the general public.
Pointed direct harm to an individual is most relevant. One can certainly make all sorts of claims without punishment. 'Vaccines cause autism' has been one. Only identifying which vaccine would be actionable. Harm against an individual relying on it is less, but can come into play especially when there are commercial considerations. 'Laetril cures cancer' is alright for personal expression, but would not be tolerated commercially. There aren't, or shouldn't be, markets in libel, slander, and blackmail, but they have market implications. Punish falsehood applies very much to markets to reduce fraud and reinforce honesty. Opinions take place beyond markets and listener beware would typically apply. Opinions stated in court would take on aspects of libel and slander even if unintentional. Of course people have been sued for opinions even when they are presumably free to express them.
Mark Spottswood: Yes, I deliberately did not specify American law in that statement. It's my understanding (with only moderate confidence here), that US law is the outlier in this regard, and that in most European countries slandering someone with deliberate and malicious intent is potentially illegal, even if no falsehoods are stated (though it may strengthen the defendant's case). In contrast, in the USA factually true statements, as well as statements of personal opinion, are essentially immune. Or, at least, that's my understanding.
Blackmail seems to me rather simple to explain in terms of the incentives of powerful people. There's not much use in blackmailing Joe Average; even if you can prove he cheated on his wife, he doesn't have a lot of money to give you to avoid the consequences. So blackmail is an activity that, by and large, transfers money from rich (and/or powerful) people to poorer ones.
That said, it also seems to me that blackmail is a rather unusual sort of transaction, because it is very difficult to enforce. Suppose we agree that you will pay me $100 a year, and I will remain quiet about that time you got drunk at my pool party. How can this contract be enforced? As soon as the facts are laid before a judge, the secret is out! Even if the details are kept confidential, as Eliezer pointed out the other day, "the most important part of a secret is that a secret exists". Just knowing that you have a secret you are willing to pay to protect will cause most of the difficulties you are trying to avoid by protecting it, because people will insist on speculating. (And the speculations will likely be worse than your actual behaviour at the pool party!)
Now, if I could credibly and verifiably commit to actually forgetting about the pool party, in exchange for a one-time payment - editing my memories - that would be different.
there seems to be a lot of rationalisation of the status quo going on in the comments here
That is what was initially requested. Recall what Robin initially wrote:
Surely there are some implicit bias theories behind these rules, but until these theories are made more explicit it remains hard to evaluate how much sense these rules make.
Robin was asking for people to make explicit the implicit bias theories behind the status quo rules. By definition, those theories necessarily rationalize the status quo.
"why are libel, slander, and blackmail illegal"
Why is this a puzzle? From either a monkey or a political theory perspective?
It's obvious from a monkey-tribe outlook - libel/slander speech acts (which by definition are untrue) attack & disrupt the social hierarchy and status machinery, a prized property of monkey society. If you lie about some element of my status, you've harmed one of my most important pieces of monkey capital and injured society overall by making it more difficult to rank status "correctly."
As for Rothbard, it is likewise obvious, yes? Rothbard's axiom is about violence & property. He often states it in a form similar to "the fundamental axiom of libertarian theory is that no one may threaten or commit violence ("aggress") against another man's person or property."
Until the invention of the consumer credit rating, which really was artifact of the 1960s as credit cards became widespread, one of a man's most important possessions was his public commercial reputation. Libel & slander laws - defamation laws - date back to Roman times and were created precisely to protect a man's property in his good public reputation.
Until the industrial revolution, most Western men had little or no tangible property - no land, no houses - perhaps only small livestock, some small furniture, and his chattel wife and children. In this environment, his intangible public reputation was actually his most valuable piece of property.
Thus it seems clear that defamation laws were an important piece of property protection, and remain so even to this day. Truth has long been considered a defense against defamation claims in the US.
Understanding this as background, Listener Beware seems appropriate when the economic burden would fall on the Listener, or when there is an ability to access accepted objective information, such as a credit score. Punish Falsehood seems appropriate when overall social harm is possible, the economic burden would fall on the person defamed, or on innocent third parties and there is no way to objectively check information.
Tom P. said: To simplify, in the US we have "listener beware" for political and religious claims and "punish falsehood" for other claims.
I appreciate to simplify, but I think this summary is misleading. We have, in fact, four primary regimes:
1. Listener Beware: Various non-verifiable statements, including religious and normative assertions. This applies to some political discourse, but not all; many political claims are in essence factual and can be the subject of defamation liability.
2. Punish Insincerity: This applies to factual assertions about political or other public figures, and (in essence, if not as a matter of formal doctrine) to most other speech on matters of public concern. (I would group the various "negligent misstatement" torts in this category for simplicity's sake, only because it will be rare for people to sincerely believe what they have no reason to believe, at least with respect to the verifiable matters that fall outside of the "listener beware" rule.) So understood, this rule also applies to perjury prosecutions, to almost all fraud claims and to many malpractice lawsuits premised on incorrect expert advice.
3. Punish Falsehoods (even if sincere): We apply this rule only to "commercial speech," a hard-to-define category that most often involves advertising or attempts to negotiate commercial transactions. Even here, there aren't many cases in which this rule is applied to its fullest extent (that is, to cases involving honest errors by a commercial speaker).
4. Mandate Disclosure of Underlying Data: we use this rule in the trial process and in securities law, for example. We often use it in conjunction with another rule; for instance, we apply rules 2 and 4 during trials (lies are punished as is the withholding of certain underlying information), while not punishing errors that involve neither a failure to disclose or an attempt to deceive.
So Robin is right to say we mostly have a blended regime. Our pluralism about normative and religious matters---combined with a desire not to stifle public debate on these issues---probably explains why we have category one. Mostly, we use category 2, which makes good sense if you think that (a) punishing honest errors will deter too much truthful speech, both directly (because even truthful speakers will decline to express themselves for fear of liability) and indirectly (because deterring false speech will also deter the truthful counterspeech that might better educate an audience), while also thinking that (b) deliberate lies cause more harm the honest errors (because they are harder to cure with effective counterspeech) but banning them will deter less truthful speech than would the "punish falsehoods" rule. In other words, our middle ground is designed to strike a balance between promoting as much truthful speech as possible, on the one hand, and deterring as much false speech as possible, on the other.
As to category 3: our use of this for commercial speech mostly stems from an empirically dubious assumption that commercial speakers are less likely to be deterred by liability than other speakers. I think this assumption is wrong (indeed, commercial speakers may be more likely to rationally weigh the costs and benefits of liability before speaking), so I generally advocate getting rid of the "punish falsehoods" rule except in a few narrow circumstances where the harms of false speech are very high and rebuttal speech is unlikely to occur. And even in those circumstances, a combination of 2 and 4 should normally work better than 3.
Most democracies do not have criminal defamation laws; rather these issues are raised in civil courts. In those cases they must be able to demonstrate that there was a damage created by a malicious intent - think negative-advertising where I just tell lies about your product loudly and frequently enough that many people believe it. Similarly as an individual I might attempt to discredit someone's reputation at work to get them fired so I can get their job.
I don't understand why people think words cannot create real harms or culpability.
In the US, alcohol companies may not buy TV ads truthfully saying most studies find people who drink more are healthier, and trial witnesses may not truthfully tell rumors they've heard about the accused.You disappoint me a bit here, Robin. The reasons for this are obvious, and unrelated to your main points - the first is because alcohol companies are allowed to advertise only on sufference anyway, and the second is because repeating hearsay in court is very pernicious to determining a correct sentence (especially if this is instiutionalised).
Blackmail typically increases total punishment, even of criminal activity.Of course; so blackmailing about crimes is a mixture of vigilanticism with concealment of the crime. All the more reason for the state to hate it.
Don't take this as rude, but there seems to be a lot of rationalisation of the status quo going on in the comments here. My "analysis" included, but at least my opinions have changed as a result of Robin bringing this up - I understand the banning of blackmail better, and (with the exception of witnesses at trials and scientists) would now favour less libel laws than I did before.
Have others had their opinions changed (even slightly) by this blog post?