The Volokh Conspiracy

I’m writing an article that indirectly touches on this question, and I thought I’d ask our readers for their take on it. I’d particularly like to hear from people who are knowledgeable about privacy law, and who (unlike me) support information privacy speech restrictions, such as the disclosure-of-private-facts tort.

Minnesota has an interesting statute that allows courts to enjoin speech that “ha[s] a substantial adverse effect ... on the ... privacy” of a person, Minn. Stat. Ann. § 609.748. Five months ago it was used to issue an injunction banning online speech by a person about his ex-girlfriend, Johnson v. Arlotta (2011), but it has been used before as well.

I’m curious what people think of this — again, especially people who generally support the disclosure tort — given the lack of a statutory or judicial definition of what constitutes “privacy” for purposes of the statute, and the criminal penalties for violating the order. How should the statute be read, and is it constitutional? Should “privacy” be read as tortious invasion of privacy, with all the common-law twists on that (e.g., the exception for newsworthy speech, and the requirement that the speech be said to the public and not just as gossip within a circle of friends)? Is that sufficiently clear for an order that can be enforced through criminal penalties? Also, are temporary restraining orders under the statute — which may be issued ex parte — unconstitutional prior restraints?

Here’s the relevant excerpt:

(a) “Harassment” includes:
(1) a single incident of physical or sexual assault or repeated incidents of intrusive or unwanted acts, words, or gestures that have a substantial adverse effect or are intended to have a substantial adverse effect on the safety, security, or privacy of another, regardless of the relationship between the actor and the intended target;
(2) targeted residential picketing; and
(3) a pattern of attending public events after being notified that the actor’s presence at the event is harassing to another....

(c) “Targeted residential picketing” includes the following acts when committed on more than one occasion:
(1) marching, standing, or patrolling by one or more persons directed solely at a particular residential building in a manner that adversely affects the safety, security, or privacy of an occupant of the building ....

Subd. 4. Temporary restraining order. (a) The court may issue a temporary restraining order ordering the respondent to cease or avoid the harassment of another person or to have no contact with that person if the petitioner files a petition in compliance with subdivision 3 and if the court finds reasonable grounds to believe that the respondent has engaged in harassment. When a petition alleges harassment as defined by subdivision 1, paragraph (a), clause (1), the petition must further allege an immediate and present danger of harassment before the court may issue a temporary restraining order under this section....

(b) Notice need not be given to the respondent before the court issues a temporary restraining order under this subdivision....

(c) The temporary restraining order is in effect until a hearing is held on the issuance of a restraining order under subdivision 5. The court shall hold the hearing on the issuance of a restraining order if the petitioner requests a hearing. The hearing may be continued by the court upon a showing that the respondent has not been served with a copy of the temporary restraining order despite the exercise of due diligence or if service is made by published notice under subdivision 3 and the petitioner files the affidavit required under that subdivision.

(d) If the temporary restraining order has been issued and the respondent requests a hearing, the hearing shall be scheduled by the court upon receipt of the respondent’s request. Service of the notice of hearing must be made upon the petitioner not less than five days prior to the hearing. The court shall serve the notice of the hearing upon the petitioner by mail in the manner provided in the Rules of Civil Procedure for pleadings subsequent to a complaint and motions and shall also mail notice of the date and time of the hearing to the respondent. In the event that service cannot be completed in time to give the respondent or petitioner the minimum notice required under this subdivision, the court may set a new hearing date....

Subd. 5. Restraining order. (a) The court may grant a restraining order ordering the respondent to cease or avoid the harassment of another person or to have no contact with that person if all of the following occur:
(1) the petitioner has filed a petition under subdivision 3 [and the order has been served on the respondent];
(2) the sheriff has served respondent with a copy of the temporary restraining order obtained under subdivision 4, and with notice of the right to request a hearing, or service has been made by publication under subdivision 3, paragraph (b); and
(3) the court finds at the hearing that there are reasonable grounds to believe that the respondent has engaged in harassment.
A restraining order may be issued only against the respondent named in the petition; except that if the respondent is an organization, the order may be issued against and apply to all of the members of the organization. If the court finds that the petitioner has had two or more previous restraining orders in effect against the same respondent or the respondent has violated a prior or existing restraining order on two or more occasions, relief granted by the restraining order may be for a period of up to 50 years. In all other cases, relief granted by the restraining order must be for a fixed period of not more than two years. When a referee presides at the hearing on the petition, the restraining order becomes effective upon the referee’s signature....

(c) If the court orders relief for a period of up to 50 years under paragraph (a), the respondent named in the restraining order may request to have the restraining order vacated or modified if the order has been in effect for at least five years and the respondent has not violated the order....At the hearing, the respondent named in the restraining order has the burden of proving by a preponderance of the evidence that there has been a material change in circumstances and that the reasons upon which the court relied in granting the restraining order no longer apply and are unlikely to occur. If the court finds that the respondent named in the restraining order has met the burden of proof, the court may vacate or modify the order. If the court finds that the respondent named in the restraining order has not met the burden of proof, the court shall deny the request and no request may be made to vacate or modify the restraining order until five years have elapsed from the date of denial. An order vacated or modified under this paragraph must be personally served on the petitioner named in the restraining order.

Subd. 6.... (b) Except as otherwise provided in paragraphs (c) and (d), when a temporary restraining order or a restraining order is granted under this section and the respondent knows of the order, violation of the order is a misdemeanor.
(c) A person is guilty of a gross misdemeanor who knowingly violates the order within ten years of a previous qualified domestic violence-related offense conviction or adjudication of delinquency.
(d) A person is guilty of a felony and may be sentenced to imprisonment for not more than five years or to payment of a fine of not more than $10,000, or both, if the person knowingly violates the order:
(1) within ten years of the first of two or more previous qualified domestic violence-related offense convictions or adjudications of delinquency;
(2) because of the victim’s or another’s actual or perceived race, color, religion, sex, sexual orientation, disability as defined in section 363A.03, age, or national origin;
(3) by falsely impersonating another;
(4) while possessing a dangerous weapon;
(5) with an intent to influence or otherwise tamper with a juror or a judicial proceeding or with intent to retaliate against a judicial officer, as defined in section 609.415, or a prosecutor, defense attorney, or officer of the court, because of that person’s performance of official duties in connection with a judicial proceeding; or
(6) against a victim under the age of 18, if the respondent is more than 36 months older than the victim....

A pleasant story of the law being fixed as a result of public criticism — much of it online — just as in Arizona last month:

Last year, the Tennessee Legislature enacted a statute that essentially banned the online posting of images that cause “emotional distress” “without legitimate purpose.” As I blogged on June 6, the law made it a crime to

(4) Communicate[] with another person or transmits or displays an image in a manner in which there is a reasonable expectation that the image will be viewed by the victim by [by telephone, in writing or by electronic communication] without legitimate purpose:

(A) (i) With the malicious intent to frighten, intimidate or cause emotional distress; or

(ii) In a manner the defendant knows, or reasonably should know, would frighten, intimidate or cause emotional distress to a similarly situated person of reasonable sensibilities; and

(B) As the result of the communication, the person is frightened, intimidated or emotionally distressed.

The law therefore applied not just to one-to-one communication, but to people’s posting images on their own Facebook pages, on their Web sites, and in other places if (1) they were acting “without legitimate purpose,” (2) they caused emotional distress, and (3) they intended to cause emotional distress or knew or reasonably should have known that their action would cause emotional distress to a similarly situated person of reasonable sensibilities. So,

1. If you had posted a picture of someone in an embarrassing situation — not at all limited to, say, sexually themed pictures or illegally taken pictures — you’d likely have been a criminal unless the prosecutor, judge, or jury concludes that you had a “legitimate purpose.”
2. Likewise, if you had posted an image intended to distress some religious, political, ethnic, racial, etc. group, you too could have been sent to jail if a government decisionmaker thought thinks your purpose wasn’t “legitimate.” Nothing in the law required that the picture be of the “victim,” only that it be distressing to the “victim.”
3. The same would have been true even if you hadn’t intend to distress those people, but reasonably should have known that the material — say, pictures of Mohammed, or blasphemous jokes about Jesus Christ, or harsh cartoon insults of some political group — would have “cause[d] emotional distress to a similarly situated person of reasonable sensibilities.”
4. And of course the same would have applied if a newspaper or TV station had posted embarrassing pictures or blasphemous images on its site.

After — I can’t say whether because of — the June 6 post, there was a good deal of criticism of the new law, and it looks like the Tennessee Legislature listened. Last week, it passed (unanimously in the Senate, 76-14 in the House) and sent to the Governor a bill that basically limited the statute to threats; if the Governor signs the bill, then the law would be limited to making it a crime to

(4) Communicate[] with another person or transmits or displays an image without legitimate purpose with the intent that that the image is viewed by the victim ... [when the communicator]:

(A) Maliciously intends the communication to be a threat of harm to the victim; and

(B) A reasonable person would perceive the communication to be a threat of harm.

This seems to be limited to speech that fits within the “true threats” exception to First Amendment protection (at least if “harm” is reasonably interpreted as a threat of illegal physical harm or vandalism, rather than a threat of, say, boycott or social ostracism, which are generally constitutionally protected).

So it looks like public criticism of speech restrictions, even ones that seem to target supposed “bullying” or “harassment,” has worked in this instance; I’m very glad to see that. For more on the change in the law, see this TN Report article posted yesterday.

Minnesota law defines “harassment” to include,

repeated incidents of intrusive or unwanted acts, words, or gestures that have a substantial adverse effect or are intended to have a substantial adverse effect on the safety, security, or privacy of another, regardless of the relationship between the actor and the intended target.

If someone complains about “harassment,” a court can issue an order banning such “harassment,” and violation of the order will then be a crime. Given this, what are you free to say about other people in Minnesota, assuming you want to say it several times, and they’re willing to go to court to stop you?

Say that you have been (accurately) telling your and your ex’s acquaintances that your ex cheated on you, or infected you with a sexually transmitted disease — is that a “repeated incident[] of intrusive or wanted ... words ... that ha[s] a substantial adverse effect on the [ex's] privacy”? What if you tell people that someone holds some publicly condemned religious or political beliefs that he has tried hard to conceal? The list could go on.

Note that there’s nothing in the statute that purports to limit this to speech that is a tortious disclosure of private facts, though that tort is itself constitutionally questionable, both on overbreadth grounds and vagueness grounds (given the lack of clear definition of the “newsworthiness” exception to the tort, an exception that does not appear in the statute).

Check out this 2009 D.C. law that purports to ban “stalking” (D.C. Code §§ 22-3132, -3133), and that provides, among other things:

“Emotional distress” means significant mental suffering or distress that may, but does not necessarily, require medical or other professional treatment or counseling....

“To engage in a course of conduct” means directly or indirectly, or through one or more third persons, in person or by any means, on 2 or more occasions, to:
(A) Follow, monitor, place under surveillance, threaten, or communicate to or about another individual;
(B) Interfere with, damage, take, or unlawfully enter an individual’s real or personal property or threaten or attempt to do so; or
(C) Use another individual’s personal identifying information [defined to include a person's name]....

(a) It is [a crime] for a person to purposefully engage in a course of conduct directed at a specific individual [to intentionally, knowingly, or negligently] cause that individual to ... [s]uffer emotional distress....

(b) This section does not apply to constitutionally protected activity.

So if you twice say something about someone, where a reasonable person should have known that this would cause the subject “significant mental suffering or distress,” you’re guilty of a misdemeanor — or a felony if the subject is under 18, and you are four or more years older than the subject — unless the court finds your speech to be “constitutionally protected activity.”

But what is constitutionally protected activity? For instance, is telling some friends about an acquaintance’s medical history or sex life constitutionally protected? The Court has had no occasion to decide this, partly because such private revelation is usually not even tortious under state law (in most states, the “disclosure of private facts” tort only covers speech to the public at large, or under circumstances where it’s likely to reach the public) and thus the matter rarely comes to court. How about marketing T-shirts or video games or comic books that contain a celebrity’s name or likeness? Lower courts are split on that. What about saying something that violates a promise not to say it? That’s constitutionally unprotected against a breach of contract lawsuit, but the Court has never decided it whether it can generally be criminalized.

Or how about residential picketing? Such speech is unprotected against a sufficiently narrow content-neutral residential picketing ban, but protected against content-based bans, and protected against even content-neutral bans that are too broad (e.g., apply to all picketing within 300 feet of a residence, rather than just picketing right in front of a residence). More broadly, the same speech is often protected against some sorts of restrictions but unprotected against others that are sufficiently narrowly tailored to some government interest.

To be sure, some courts have upheld laws that have such supposed “savings clauses,” but others have struck them down. And it seems that the latter view is the correct one; in the words of Long v. State (Tex. Ct. Crim. App. 1996) (paragraph break added):

While the provision would permit the defendant to introduce evidence before the jury regarding the constitutional nature of his conduct, it would relegate the First Amendment issue to a “case-by-case adjudication,” creating another vagueness problem. In essence, (a)(7)(A) [the Texas provision involved in Long -EV], as modified by the affirmative defense, would read something like “it is a crime to intentionally annoy someone unless by that conduct the actor engages in activity protected by the First Amendment.” Application of the affirmative defense to subsection (a)(7)(A) on a case-by-case basis would require people of ordinary intelligence — and law enforcement officials — to be First Amendment scholars. Arguably, people are always “on notice” that constitutionally protected conduct is exempt from prosecution, and law enforcement officials could always look to the First Amendment to determine when a law should not be enforced because it would interfere with constitutionally protected activity. But, the mere existence of the First Amendment has never been held automatically to cure vagueness problems implicating First Amendment freedoms.

Because First Amendment doctrines are often intricate and/or amorphous, people should not be charged with notice of First Amendment jurisprudence, and a First Amendment defense cannot by itself provide adequate guidelines for law enforcement. [Footnote: Charging ordinary citizens and law enforcement officials with knowledge of constitutional law seems especially inappropriate in an area of law, such as stalking, that is relatively new.] Moreover, an attempt to charge people with notice of First Amendment caselaw would undoubtedly serve to chill free expression.

UPDATE: I should note that I think the Texas court’s argument is correct whether or not the “constitutionally protected activity” proviso creates an affirmative defense as to which the defendant must introduce evidence, or creates an element of the crime (the lack of constitutional protection for the activity) that the prosecution must prove. Even if the prosecution must prove that the activity is constitutionally unprotected, the provision remains so unclear that it’s unconstitutional, either under the void-for-vagueness doctrine as such or (as in Reno v. ACLU) because the vagueness of the law leads it to deter even constitutionally protected speech.

Twenty-three of the forty-nine New York Assembly Republicans, plus one Independent and one Democrat, introduced this bill last Fall but just “unveiled” it yesterday:

1. Definitions. As used in this section, the following words and terms shall have the following meanings:

(a) ["]Anonymous poster["] is any individual who posts a message on a web site including social networks, blogs forums, message boards or any other discussion site where people can hold conversations in the form of posted messages.

(b) “Web site administrator” means any person or entity that is responsible for maintaining a web site or managing the content or development of information provided on a web site including social networks, blogs forums, message boards or any other discussion site where people can hold conversations in the form of posted messages, accessible via a network such as the internet or a private local area network.

2. A web site administrator upon request shall remove any comments posted on his or her web site by an anonymous poster unless such anonymous poster agrees to attach his or her name to the post and confirms that his or her IP address, legal name, and home address are accurate. All web site administrators shall have a contact number or e-mail address posted for such removal requests, clearly visible in any sections where comments are posted.

It’s not clear what it means to “confirm” that one’s IP address, legal name, and home address are accurate; but at the very least, this bill would require a Web site administrator — me, for instance, if I were found to subject to New York jurisdiction — to remove any comment unless the commenter signs his name to it.

Nor would this be limited to comments that allegedly libel someone, or even insult someone (though that would be bad enough), despite all the talk of preventing cyber-bullying by the bill’s backers. Rather, the law would apply any time anyone makes a “request” that a comment be removed, even if the comment doesn’t mention anyone by name but is simply religiously or politically offensive to the “request[er].” The same would apply to anonymous material added to Wikipedia, if Wikipedia were found to be subject to New York jurisdiction, anonymous videos posted to YouTube, and so on.

The bill is unconstitutional, see Talley v. California (1960) and McIntyre v. Ohio Elections Comm’n (1995); the First Amendment, the Supreme Court has held, protects anonymous speech (except in limited conditions related to election campaigns). The proposal is thus a fitting bookend to the four Democratic New York Senators’ paper on, among other things, how “[p]roponents of a more refined First Amendment argue that this freedom should be treated not as a right but as a privilege — a special entitlement granted by the state on a conditional basis that can be revoked if it is ever abused or maltreated,” though more New York Republicans are on board this bandwagon than New York Democrats were on board the other one.

Thanks to Steven Jens for the pointer.

UPDATE: By the way, say that a Web site with tens of thousands of comments gets a batch of demands from a political opponent of the site: “[R]emove any comments posted on [your] web site by an anonymous poster” — defined as “any individual who posts a message on a web site including social networks, blogs forums, message boards or any other discussion site where people can hold conversations in the form of posted messages” — unless that individual “agrees to attach his or her name to the post and confirms that his or her IP address, legal name, and home address are accurate” (emphasis added). The time and effort it takes to get such “confirm[ation]” from all the commenters, including ones who had signed their names in the first place (recall that “anonymous poster” is defined to mean “any individual who posts a message on a web site,” even if the message is signed with what is ostensibly the poster’s name) might well be prohibitive for many Web site operators, whose only option at that point would be just to delete all the comments.

The Arizona cyber-harassment bill, which I blogged about March 31, has now been narrowed in the Arizona Legislature. The original proposal — which had been passed in nearly identical forms by both houses of the Arizona Legislature — read,

It is unlawful for any person, with intent to terrify, intimidate, threaten, harass, annoy or offend, to use a telephone ANY ELECTRONIC OR DIGITAL DEVICE and use any obscene, lewd or profane language or suggest any lewd or lascivious act, or threaten to inflict physical harm to the person or property of any person.

This, as I argued, would have posed serious First Amendment problems. Telephones are basically one-to-one devices, so a phone call that uses profane language to offend is likely meant only to offend the one recipient, rather than to persuade or inform anyone; but computers used to post Facebook messages or send Twitter messages or post blog items can offend some listeners while persuading and informing others.

So, under the proposal, posting a comment to a newspaper article — or a blog — saying that the article or post author is “fucking out of line” would have been a crime: It’s said with intent to offend, it uses an electronic or digital device, and it uses what likely will be seen as profane language (see, e.g., City of Columbia Falls v. Bennett (Mont. 1991)). Likewise if a blog poster were to post the same in response to a commenter’s comment. Likewise if someone posts something in response to an e-mail on an e-mail-based discussion list, or in a chatroom, or wherever else. (Note that if “profane” is read to mean not vulgarly insulting, but instead religiously offensive, see City of Bellevue v. Lorang (Wash. 2000), then the statute would have been unconstitutional as well.)

The same would have been true if someone posted something lewd in one of these places in order to annoy or offend someone, for instance if he posted a comment on a police-run public discussion page that says something like “the chief of police can suck my dick,” to borrow subject matter from a prior Arizona telephone harassment case. And, given that case, the speech need not even have been about one of the recipients, so long as it had been intended to annoy or offend one of the recipients.

Fortunately, the Arizona Legislature’s House-Senate conference committee has dramatically narrowed the proposed statute. The new version — which I expect will become law — reads,

A. It is unlawful for any person, with intent to terrify, intimidate, threaten or harass a specific person or persons, to do any of the following:

1. Direct any obscene, lewd or profane language or suggest any lewd or lascivious act to the person in an electronic communication.

2. Threaten to inflict physical harm to any person or property in any electronic communication.

3. Otherwise disturb by repeated anonymous, unwanted or unsolicited electronic communications the peace, quiet or right of privacy of the person at the place where the communications were received....

C. This section does not apply to constitutionally protected speech or activity or to any other activity authorized by law....

This is much more likely to be read as limited to one-to-one messages, such as unwanted targeted e-mails, instant messages, text messages, and the like. The language is not completely airtight on that score, but I think it’s much more likely to be so interpreted. And though there are problems with the proposal even as to unwanted one-to-one messages — for instance, “harass” is not defined in the statute, and the word “harassment” is defined three different ways in three other Arizona statutes, which reflects its ambiguity — I think the new version is much better than the original draft.

Many thanks to Alan Solot for pointing me to the original version of the bill in the first place, to Charles Brownstein of the Comic Book Legal Defense Fund for writing the March 30 post that Mr. Solot had passed along to me, and to Media Coalition for its earlier criticisms (see here and here).

UPDATE: Thanks also to author Neil Gaiman; Alan Solot tells me that he learned of the bill through a Twitter message by Mr. Gaiman, which pointed to the Comic Book Legal Defense Fund post.

This is happening in § 1003 of the Violence Against Women Reauthorization Act of 2011, which is apparently being debated tomorrow. Here’s the deal: 47 U.S.C. § 223(a) currently criminalizes

(C) mak[ing] a telephone call or utilizes a telecommunications device, whether or not conversation or communication ensues, without disclosing his identity and with intent to annoy, abuse, threaten, or harass any person at the called number or who receives the communications; [or] ...

(E) mak[ing] repeated telephone calls or repeatedly initiat[ing] communication with a telecommunications device, during which conversation or communication ensues, solely to harass any person at the called number or who receives the communication.

This law already has some problems, I think, given the vagueness of terms such as “annoy” and “harass,” and it’s been held unconstitutional as applied in a case in which a man was prosecuted for leaving racist voice-mail for then-U.S. Attorney and now-Attorney General Eric Holder; it might therefore be unconstitutionally overbroad. But it has generally been upheld in most cases, and such a result may be defensible precisely because the statute deals with calls or messages to the person who is being harassed. Restricting such unwanted one-to-one speech (with a properly crafted and narrowed statute) should be permissible, because it leaves people free to communicate with willing listeners. See generally Rowan v. United States Post Office Dep’t (1970).

But now the Senate is considering replacing “harass any person at the called number or who receives the communication” with “harass any specific person.” This appears to be a deliberate attempt to remove the language that could be read as limiting the current statute to one-to-one communications. (The proposal would also remove “annoy” from item (C), but it would keep “harass.”)

So say that someone started to post anonymous (or pseudonymous) blog posts — or Twitter messages or online newspaper articles — criticizing Attorney General Holder, a local official, a police officer, a businessperson, a religious leader, or anyone else. The posts aren’t threatening or false, just harshly critical. And say a prosecutor decided that the speaker was partly motivated by a desire to “badger, disturb or pester” the target of his messages. The speaker could then be prosecuted, on the theory that he was

utiliz[ing] a telecommunications device ... without disclosing his identity and with intent to ... harass any specific person [added text italicized].

And this is so even though the speaker is talking to the public at large, not just to the target: As I said, the proposal would expressly delete the statutory language that could now be used to try to limit the statute to one-one unwanted messages.

The intent to harass would not have to be the sole intention. Subsection (E) specifically limits itself to situations where the speaker has the sole intention to harass, and while that’s problematic enough on its own terms (since it’s so hard to tell what’s a speaker’s “sole” intention), it makes clear that subsection (C), which lacks such limiting language, applies even to cases where the intent to harass is one of the speaker’s intention.

Say a speaker thinks Congressman Joe Schmoe (or Officer Joe Schmoe or Reverend Joe Schmoe) has done bad things and therefore (1) intends to communicate to the public why he thinks Joe Schmoe should be reviled, and (2) wants Joe Schmoe to feel reviled and embattled. Harsh and, especially, repeated anonymous criticism of Schmoe would then reasonably be seen as having an “intent to harass” as well as an intent to communicate to the public. Note also that the word “harass” is not defined in the statute, but the definition I gave above, which is to “badger, disturb or pester,” is taken from a case interpreting another criminal prohibition on things done with “intent to harass,” and reflects that court of appeals’ judgment of the dictionary meaning of the word. I have no reason to think that “harass” would be interpreted any more narrowly (or more clearly) than “badger, disturb or pester.”

And of course there’s every reason to think that the revised statute could be used not just to go after criticism of private individuals — though I think it would be unconstitutional even then — but also government officials. As you can see in these posts, these sorts of broad “harassment” statutes have recently been used to silence, prosecute, or try to unmask critics of prominent religious leaders, city commissioners, police officers, and candidates for elective office. Why is the Senate considering broadening federal speech restrictions to make such prosecutions easier?

Note also that the bill would also remove the limiting language from the provision (§ 223(E)) that covers even nonanonymous speech that’s said “solely to harass.” Under the bill, a signed blog post, online newspaper article, or Twitter message could lead to a prosecution if the prosecutor thinks the purpose of the message is “solely to harass” the person being criticized in the article. I would argue that statements to the public always have the intention to inform, persuade, entertain, or something else, even if they also have the intention to harass. But the bill is deliberately expanding the subsection to cover not just speech said to a person, but also speech said to the public about the person. And given the human tendency to assume the worst motivations in those whose views one disagrees with, it’s quite plausible that prosecutors, judges, and jurors might find someone guilty of publicly speaking “solely to harass” the person whom he is harshly criticizing, even when the speaker also has the purpose of informing or persuading the public.

As I noted last year, a federal prosecutor indicted a man for his campaign of insulting Twitter messages about a Buddhist religious leader, and a federal judge then dismissed the indictment on First Amendment grounds.

Some of the Twitter posts might have been seen as threatening, but the government’s theory wasn’t that they were threatening, but that they where posted “with the intent to harass and cause substantial emotional distress to” the religious leader, and actual caused such distress, in violation of the federal antistalking statute, 18 U.S.C. § 2261A. The district court correctly held that the statute was unconstitutional as applied, and left open the possibility that it might be unconstitutional on its face as well. The government appealed, but earlier this month decided to drop the appeal.

Now the Senate, rather than tightening the law to prevent its being applied to constitutionally protected speech, is considering expanding it. Section 107 of the Violence Against Women Reauthorization Act of 2011 (which I blogged about below, as to a different constitutionally troublesome provision) would take the existing statute and have it cover not just speech that causes substantial emotional distress, but also speech that “attempts to cause, or would be reasonably expected to cause substantial emotional distress,” so long as the speaker is intending to (among other things) “harass” the target. Anyone who

with the intent to ... harass ... uses the ... any interactive computer service ... to engage in a course of conduct that ... causes, attempts to cause, or would be reasonably expected to cause substantial emotional distress to [the person or the person's family member]

would now be committing a federal crime. If the statute were just limited to conduct and speech intended to threaten someone — which the rest of the statute does cover — there would be no constitutional problem. But trying to outlaw speech that is made “with the intent to ... harass” and “would be reasonably expected to cause substantial emotional distress” often violates the First Amendment, as the federal judge rightly held in the Twitter case. It seems to me that Congress should be fixing this constitutional problem rather than expanding it.

UPDATE: I accidentally quoted the wrong passage in the block quote; sorry about that.

That’s what would happen under a Monroe County (New York) bill, proposed by County Executive Maggie Brooks, County Legislature Vice President Michael G. Barker, and County Legislator Carmen F. Gumina. The bill reads, in relevant part (paragraph break added):

[It shall be unlawful for a person] to ... with intent to harass [or] annoy [a minor] ... engag[e] in a course of conduct or repeatedly commit[] acts of abusive behavior over a period of time by [electronic] communication.

Acts of abusive behavior shall include, but not [be] limited to: taunting; ... insulting; tormenting; humiliating; disseminating sexually explicit photographs, either actual or modified, of a minor; disseminating the private, personal or sexual information, either factual or false, of a minor without lawful authority.

So the following would all be criminals, punishable by up to a year in jail:

1. Someone who sends two e-mails over several days to friends — or posts two items on a Facebook page or a blog post — insulting a high school classmate, with the intent to annoy him.
2. Someone who twice (over several days) e-mails or posts something condemning her ex-boyfriend for cheating on her (since that would be revealing “sexual information” “of a minor”), with the intent to annoy the ex-boyfriend.
3. Someone who writes two newspaper articles or blog posts (over several days) insultingly condemning a juvenile criminal, with the intent to annoy him (perhaps because he wants the criminal to feel bad about the crimes).
4. Someone who twice (over several days) e-mails or posts something mocking a candidate for high school student government, with the intent to annoy the candidate.
5. Someone who twice (over several days) e-mails or posts something taunting a rival high school’s sports team, with the intent to annoy the players on that team.

Note that I say “twice” because the proposed law simply calls for “repeatedly committing acts over a period of time,” and “repeatedly” both seems to just mean “more than once,” and has been so interpreted in a closely related telephone harassment statute, see People v. Calderon (N.Y. trial ct. 1997). Note also that the proposal would cover threatening speech, which would be constitutional but likely already duplicative of state laws; I don’t quote that material because my objection is to the separate prohibitions on speech that’s said with the intent to “annoy” rather than threaten.

So argue two lawyers in a recent Chronicle of Higher Education article, warning of legal liability if colleges don’t take suitable steps to suppress such speech. The article is short on definitions of cyber-bullying, but calls for colleges to update their “anti-bullying and social-media policies to take into account the immediate and significant harm that can be inflicted when bullying behavior leaves the dormitory or the quad and goes online,” and to have administrative processes to “foster a safe and supportive” (and “more inclusive”) “environment for all of its students” by “[m]anaging the recent and exponential growth of social-media sites and digital forms of communication.”

This sounds to me like more than just a call for punishing constitutionally unprotected speech, such as threats of violence — though how much more is hard to tell. It will be interesting to see what new university speech codes aimed at preventing “cyber-bullying” are going to spring up in the coming years.

My post about the Minnesota trial court order forbidding a parent to urge a Catholic school to fire a gay teacher led to several commenters’ suggesting that such speech may indeed be restricted, at least if it consists of more than just one e-mail. Such speech, the theory is, may indeed constitute “harassment,” if it’s repeated (and perhaps if it’s aimed at not just the current employer but future employers, if this employer does fire the employee). Or, the argument goes, it could be punishable under the tort of intentional interference with business relations.

I think that’s wrong. The First Amendment, it seems to me, protects people’s rights to express their views, including when the views are aimed at persuading others to act, and including when the action is firing an employee, cutting off a contract with a contractor, and so on. If the speech falls within an existing First Amendment exception, for instance because it’s a threat of criminal conduct or a knowingly false accusation, then the speech can be punished. But outside those exceptions, speech urging the legally permissible firing of a schoolteacher — or a spokesman or a radio talk show host — is as protected as speech urging anything else.

The Supreme Court has expressly held that the tort of intentional interference with business relations is constrained by the First Amendment, and that speech aimed at producing economic pressure is constitutionally protected notwithstanding the tort. This happened in NAACP v. Claiborne Hardware (1982), where the NAACP organized a boycott of white-owned businesses aimed at pressuring them to hire black employees (and aimed at other things as well). The Court reversed an interference with business relations tort verdict, on the grounds that “Speech does not lose its protected character, however, simply because it may embarrass others or coerce them into action.”

Speech urging the firing of an employee is just as much advocacy as speech urging the hiring of an employee, and both can be “coerc[ive]” through the use of economic and social pressure. If the organization of a massive boycott aimed at coercing businesses is constitutionally protected, then the sending of a few e-mails — or even of many e-mails — aimed at pressuring a school to fire an employee is likewise protected.

Indeed, such speech urging the firing of employees (or the cancellation of contracts with contractors) is relatively common. Consider, for instance, the 1970s boycott against Florida orange juice aimed at getting the companies to stop using Anita Bryant as a spokeswoman; Bryant had been a prominent anti-gay activist. In my home town, there was a 1990s boycott of sponsors of black conservative talk show host Larry Elder’s radio show, aimed at getting the radio station to take him off the air. In 1990, there was public pressure that caused CBS to suspend 60 Minutes commentator Andy Rooney for allegedly making a racist comment; in 1988, public pressure that caused CBS to fire Jimmy “The Greek” Snyder on similar grounds. To be sure, there may often be ethical distinctions among these cases, and between these cases and calls for (lawfully) firing a Catholic school teacher because he’s allegedly gay (see generally my article on Deterring Speech: When Is It “McCarthyism”? When Is It Proper?, 93 Cal. L. Rev. 1413 (2005)). But I think the First Amendment equally protects all such advocacy.

Indeed, consider a hypothetical based on the case that started this discussion: Say a parent concludes that a schoolteacher is an anti-gay advocate, and he concludes that this makes the teacher a poor role model and a less effective educator. Say that the parent sends e-mails to the school urging the school to fire the teacher, tries to drum up support for this among other parents, writes letters to the editor about this, and so on. Should such speech be restrictable, on the grounds that it’s somehow “harassing,” or improperly interferes with the teacher’s business relations with the school?

I think the answer would be “no”; such speech is constitutionally protected advocacy, despite the harm it may cause to the teacher’s job prospects. Given the First Amendment requirement of viewpoint neutrality (and, generally, of content neutrality), the same must be true of speech aimed at getting a teacher fired for his supposed homosexuality. And that I think there’s a practical and moral difference between a teacher’s saying anti-gay things and a teacher’s being gay can’t justify the law’s restricting the speech of people who take a different view.

I’m writing an article on the First Amendment, criminal harassment law, and the constitutional distinction between speech said to a particular objecting person and speech said to the public; and in the process, I ran across this 2001 case that I thought I’d mention, since it could equally come up today.

Minnesota law provides, in relevant part,

“Harassment” includes: ... repeated incidents of intrusive or unwanted acts, words, or gestures that have a substantial adverse effect or are intended to have a substantial adverse effect on the safety, security, or privacy of another ....

The court may grant a restraining order ordering [a person] to cease or avoid the harassment of another person or to have no contact with that person if ... the court finds at the hearing that there are reasonable grounds to believe that the respondent has engaged in harassment....

A first-offense violation of the restraining order is generally both a misdemeanor and punishable as contempt of court, but when a person “knowingly violates the order ... because of the victim’s or another’s actual or perceived race, color, religion, sex, sexual orientation, disability ..., age, or national origin,” the violation is a felony.

Now note that the statute isn’t limited to traditional telephone harassment or stalking that only involves unwanted speech to the target. It potentially also includes unwanted speech about the target, so long as it is “repeated” and “[has] a substantial adverse effect or [is] intended to have a substantial adverse effect on the safety, security, or privacy” of the target. This means the law potentially cuts off not just speech to one unwilling listener (or a few unwilling listeners), but also speech to willing listeners as well. The Minnesota courts have read the statute as covering two categories of unprotected speech — “‘fighting words’ likely to cause the average addressee to fight or protect one’s own safety, security, or privacy” and “‘true threats’ evidencing an intent to commit an act of unlawful violence against one’s safety, security or privacy” — but also “speech or conduct that is intended to have a substantial adverse effect, i.e., is in violation of one’s right to privacy.” This latter category does not correspond to any settled First Amendment exception, and it’s not at all clear what exactly it means.

Now on to the case, Faricy v. Schramm (Minn. Ct. App. 2001):

Appellant Derek Schramm is a parent of children enrolled in a Roman Catholic grade school in Minneapolis. Respondent Zachary Faricy is a teacher at the school. In November 2001, Schramm sent a letter to the school principal and the parish pastor informing them of his suspicion that Faricy “might be a homosexual.” Schramm based his suspicion on the fact that Faricy’s car displays an Apple Computer decal [apparently, one of the rainbow ones -EV] and that Faricy “lives or has lived ... in a neighborhood that has a higher population of homosexuals.” Schramm had determined Faricy’s place of residence through an internet search. Schramm contended in the letter that “both wearing a rainbow sticker on your car [and] living in a neighborhood where more homosexuals live is enough to send up a red flag.” In closing, Schramm asked school officials

what can be done to learn if Mr. Faricy chooses a ... homosexual lifestyle and if he is a sodomite, then what is his future around children at [the] school?

Citing fear that Schramm’s “behavior will escalate into something possibly dangerous,” Faricy petitioned the district court to issue a harassment restraining order. After hearing testimony from both Faricy and Schramm, the district court found that Schramm had engaged in two “acts of harassment.” The first act consisted of sending a letter to Faricy’s employer suggesting that Faricy might be gay and that, if he were gay, he should not be allowed to teach in the school. The district court identified the second act of harassment as Schramm’s inclusion in his letter of information about Faricy’s place of residence and the decal on his car. On the basis of these findings, the district court issued a one-year restraining order prohibiting Schramm from contacting Faricy and from engaging in further acts of harassment....

Since the order banned “further acts of harassment,” it would have criminalized — indeed, made a felony, given that Schramm was motivated by Faricy’s perceived sexual orientation — any similar further communications, for instance to other officials at the school or the diocese. Presumably the court’s theory was that Schramm’s e-mail intruded on Faricy’s “privacy,” though the appellate opinion doesn’t make that clear.

The Minnesota Court of Appeals reversed, on the grounds that the statute requires “repeated incidents,” and the letter constituted just one incident. But the appellate court didn’t opine on the trial court’s underlying judgment that this is the sort of incident — perhaps if repeated at least once more, as per the statute — that could lead to a restraining order against further such speech. At this point, the only judge who considered that particular question (the trial judge) answered it “yes.”

Note, by the way, that though Minnesota law generally bans discrimination based on sexual orientation, it specifically exempts discrimination by religious institutions, including religious schools. So to the extent that Schramm was calling for Faricy’s being fired, he was calling for an action that was legal. I suspect that even calling for a firing that violates antidiscrimination law is constitutionally protected, given Brandenburg v. Ohio (though query what effect United States v. Williams‘ recognition of a “solicitation” exception to the First Amendment would have on this). But that question wouldn’t come up in this case in any event, because of Minnesota’s religious institution exemption.

Note also that the statute doesn’t require any finding that any allegedly “harassing” accusation is factually false. Schramm’s evidence for Faricy’s supposed homosexuality seems mighty thin, but even if he got that wrong, and that could lead to a civil lawsuit for defamation or false light, or even a criminal libel prosecution (Minnesota has a criminal libel law, though one that seems not to be used these days), the anti-harassment statute is not focused on falsehoods.

As I noted last week, both houses of the Arizona legislature passed a bill that would say,

It is unlawful for any person, with intent to[, among other things,] harass, annoy or offend, to use any electronic or digital device and use any obscene, lewd or profane language ....

Many people criticized the bill, as did I, but just a few days ago, a co-sponsor of the bill (UPDATE: Vic Williams) was having none of it, writing,

As the co-sponsor of HB2549 I can see the conspiracy have their tin-foil hats on tonight.

HB2549 is being chased down by the “black-helicopter” crowd. Their claims of internet restriction are unfounded and way off base!!

You can read the bill and full details @ http://www.vote4vic.com/index.cfm/article_58.htm

Yet the tin foil apparently worked: Another co-sponsor has now announced that the bill will be revised, before being sent to the Governor. According to CNN,

[Arizona Rep. Ted] Vogt said Wednesday that the bill would be amended to say those harassing communications must be directed at a specific person and must be “unwanted or unsolicited.”...

The bill will not apply to online comment sections or semi-public forums such as Facebook walls, Vogt said.

“With Facebook, you’ve got control over who your friends are,” he said. “So if somebody is threatening you and you never de-friend them then you’re not controlling it. You’re inviting people to comment freely on your Facebook page. You can de-friend them and you can end the problem there.”

Comments sections are the same, he said, since websites don’t have to invite people to comment and can take down those sections if they are worried about threats.

I haven’t seen any specific proposed text (none is posted yet on the Legislature’s site), so I can’t speak to how good the amendment will be. But I’m glad the legislators are at least rejecting the old, bad version of the bill.

I’m also glad that Connecticut legislators have killed a similarly bad recent Connecticut proposal.

That’s what Arizona HB 2549, which was just passed by both houses (though not yet signed by the Governor) provides, in relevant part:

It is unlawful for any person, with intent to terrify, intimidate, threaten, harass, annoy or offend, to use a telephone ANY ELECTRONIC OR DIGITAL DEVICE and use any obscene, lewd or profane language or suggest any lewd or lascivious act, or threaten to inflict physical harm to the person or property of any person.

I take it that this refers only to speech intended to offend someone who receives it, and not someone who hears about it indirectly. But note the significance of the shift from a telephone to “any electronic or digital device”: Telephones are basically one-to-one devices, so a phone call that uses profane language to offend is likely meant only to offend the one recipient, rather than to persuade or inform anyone; but computers used to post Facebook messages or send Twitter messages or post blog items can offend some listeners while persuading and informing others.

So, under the statute, posting a comment to a newspaper article — or a blog — saying that the article or post author is “fucking out of line” would be a crime: It’s said with intent to offend, it uses an electronic or digital device, and it uses what likely will be seen as profane language (see, e.g., City of Columbia Falls v. Bennett (Mont. 1991)). Likewise if a blog poster were to post the same in response to a commenter’s comment. Likewise if someone posts something in response to an e-mail on an e-mail-based discussion list, or in a chatroom, or wherever else. (Note that if “profane” is read to mean not vulgarly insulting, but instead religiously offensive, see City of Bellevue v. Lorang (Wash. 2000), then the statute would be unconstitutional as well.)

The same would be true if someone posts something lewd in one of these places in order to annoy or offend someone, for instance if he posts a comment on a police-run public discussion page that says something like “the chief of police can suck my dick,” to borrow subject matter from a prior Arizona telephone harassment case. And note that, given that case, the speech need not even be about one of the recipients, so long as it’s intended to annoy or offend one of the recipients.

Naturally, readers of this blog know that I am no fan of using obscene, lewd, or profane language with intent to annoy or offend people. But, given the First Amendment, the government may not restrict such speech on blogs, e-mail discussion lists, and newspaper Web sites. If the Arizona Legislature wants to apply the ban on telephone harassment to other one-to-one devices, such as text messaging or e-mails sent directly to a recipient, it may well be free to do so. (For more on this, see Volokh, Freedom of Speech in Cyberspace from the Listener’s Perspective: Private Speech Restrictions, Libel, State Action, Harassment, and Sex, 1996 U. Chi. Legal Forum 377, parts II.B-.C (1996).) But the just-passed bill has no such limitation, and thus poses the danger of restricting a great deal of speech that is protected by the First Amendment. Thanks to Alan Solot for the pointer.

The Raised Bill No. 456, sec. 2 would provide, in relevant part:

(a) A person commits electronic harassment when such person, with intent to harass, annoy or alarm another person, transmits, posts, displays or disseminates, by or through an electronic communication device, radio, computer, Internet web site or similar means, to any person, a communication, image or information, which is based on the actual or perceived traits or characteristics of that person, which: ...

(2) Has a substantial and detrimental effect on that person’s physical or mental health;

(3) Has the effect of substantially interfering with that person’s academic performance, employment or other community activities or responsibilities;

(4) Has the effect of substantially interfering with that person’s ability to participate in or benefit from any academic, professional or community-based services, activities or privileges; or

(5) Has the effect of causing substantial embarrassment or humiliation to that person within an academic or professional community.

So say Michael Bellesiles says things that are false, and my colleague Jim Lindgren posts a devastating critique of Bellesiles’ work. This rightly should “caus[e] substantial embarrassment or humiliation to [Bellesiles] within an academic or professional community,” and rightly should “substantially interfer[e] with [Bellesiles'] ... employment” (Bellesiles ultimately resigned under pressure from his tenured job because of what his critics, including Jim, pointed out about his work) as well as with Bellesiles’ “ability to ... benefit from any academic[ or] professional ... privileges.” [UPDATE: And the critique may well be based on Lindgren's judgment about Bellesiles' "perceived trait" of academic carelessness or even dishonesty.]

This means that the only thing that would stand between Jim and criminal prosecution — if Lindgren were in Connecticut, and the law were enacted — would be the question whether Lindgren had the “intent to harass, annoy or alarm another person.” Now knowing Jim Lindgren as I do, I’m sure he had no such intent. But some critics may well be so upset and offended by the behavior of the people they are criticizing that they might well have the intent to “annoy” with their criticism as well as the intent to inform the public. (Note that the statute requires just that there be an intent to harass, annoy, or alarm, not that this be the sole intent, likely because there is almost never a sole intent behind any human action.) [UPDATE: Note also that the law isn't limited to speech that is said to the person being criticized: It applies to any communication "to any person," not to "such person" (which would have referred back to the "another person" who is the target of the alleged harassment or annoyance).]

And of course that’s just one example. Customers who are publicizing what they see as business employees’ malfeasance could be criminals under this law. So could people who had been cheated on by their lovers, who inform their friends of the now ex-lovers’ misbehavior. So could crime victims who publicize their attackers’ misdeeds. So could newspaper reporters who are writing articles that criticize certain people — politicians, businesspeople, community activists, or others — and who are seen by prosecutors as being motivated by a desire to “annoy” the targets. And of course the law isn’t limited to threats, or libel, or any other First Amendment exception that has been recognized (or even hinted at) by the Supreme Court.

The law therefore strikes me as quite clearly unconstitutional, and I hope it will be defeated. I’m glad to see that the ACLU of Connecticut and the Connecticut Daily Newspaper Association have taken the lead in criticizing the law. If anyone knows more about who the political backers of the law are — a Hartford Courant article says it what proposed by “state prosecutors,” but doesn’t say which ones — please let me know.

UPDATE: The law talks about “perceived traits or characteristics” of a person, and I read that to include traits such as perceived dishonesty, incompetence, stupidity, and so on. There’s certainly nothing in the statute to limit it to, say, race, religion, sex, weight, and so on. But even if it is so limited, the statute is unconstitutional — it would criminalize criticism of people (including, say, politicians and academics) based on their religious beliefs, or for that matter based on claims that a person shouldn’t be in some job because of his or her race, sex, and such.