O'BRIEN v. Borowski

961 N.E.2d 547, 461 Mass. 415, 2012 WL 255785, 2012 Mass. LEXIS 19

• Court: Massachusetts Supreme Judicial Court
• Decided: January 31, 2012
• Docket: SJC-10866
• Status: Published

Opinion

Gants, J.

After a hearing, a judge of the District Court on September 3, 2010, issued a harassment prevention order under An Act relative to harassment prevention orders, G. L. c. 258E, inserted by St. 2010, c. 23 1 (act or c. 258E), that directed Robert O’Brien (O’Brien) not to abuse or harass Alan Borowski (Borowski), to stay fifty yards away from Borowski, and to remain away from Borowski’s residence. O’Brien filed a petition to vacate and dismiss the order under G. L. c. 211, § 3, before a single justice of this court, who reserved and reported the case for decision by the full court. O’Brien argues that c. 258E is unconstitutionally overbroad on its face because it regulates protected speech, and unconstitutional as applied to him because the conduct complained of was protected speech. 2 We conclude that c. 258E is not unconstitutionally overbroad under our interpretation of the statute. Because we vacate the now-expired harassment prevention order on other grounds, we do not reach the question whether the application of the statute in his case was unconstitutional. 3

Background. Borowski is a police officer with the Northampton police department, who knew O’Brien before he became a police officer thirteen years ago. He had “charged [O’Brien] with a crime” in 2006, and knew O’Brien to be “a fighter.”

On the evening of May 15, 2010, Borowski entered a bar with his girl friend, saw O’Brien inside, and decided immediately to leave. After Borowski left the bar, O’Brien followed him out of the bar, yelled Borowski’s name, and when Borowski turned to look at him, raised both of his middle fingers in the air and said, “[F]uck you.”

On the afternoon of August 8, 2010, Borowski was at his home moving his truck from the driveway when he saw O’Brien in the front passenger seat of a truck traveling past his house. As the truck drove past the house, O’Brien put his hand outside the window and again “flipped [him] off” with his middle finger. After Borowski got out of his truck, the truck O’Brien was traveling in stopped in the middle of the street approximately seventy-five yards from Borowski’s house, remained there for several seconds, and drove off.

Approximately ninety minutes later, Borowski was standing on his deck when he heard a horn sound in front of his house. He saw O’Brien lean forward in the passenger seat of the truck and again extend his middle finger in the air; the driver then drove away.

Borowski applied for a harassment prevention order on August 23, 2010, and a temporary order issued. A hearing was conducted on September 3, 2010; Borowski was the lone witness. After the hearing, the judge extended the order until September 2, 2011. On September 30, 2010, O’Brien filed the instant petition under G. L. c. 211, § 3, for relief from the harassment prevention order. In reserving and reporting the case to the full court, the single justice urged the court to consider, apart from the constitutional issues, “whether the case is moot, and if so, whether it nonetheless should be decided; and whether the use of G. L. c. 211, § 3, shall continue as the avenue of review of an order entered pursuant to G. L. c. 258E.”

Discussion. We first address the procedural questions raised by the single justice. The harassment prevention order expired on September 2, 2011, and Borowski did not move to extend the order, so the case is now moot. Even though there is no longer a live dispute among the parties, we shall reach the merits of this case because it is fully briefed and raises issues of public importance regarding the constitutionality of a recently enacted statute that will likely arise again but, if we dismiss on grounds of mootness, evade review. See Aime v. Commonwealth, 414 Mass. 667, 670 (1993), and cases cited.

As to the proper avenue of review, the parties agree that, because the act does not expressly provide any other means to appeal from a harassment prevention order, the only alternative presently available is to seek relief from a single justice of this court under G. L. c. 211, § 3, and then, if necessary, from the full court on appeal. See Flynn v. Warner, 421 Mass. 1002 (1995); Callahan v. Boston Mun. Court Dep’t, 413 Mass. 1009 (1992). In Zullo v. Goguen, 423 Mass. 679, 681 (1996), we recognized that G. L. c. 211, § 3, had provided the only avenue to appeal from a restraining order issued by a District Court judge under G. L. c. 209A. But we concluded that “[ujniformity of treatment of litigants and the development of a consistent body of law” would be better accomplished if the appeal from all c. 209A orders were directed to the Appeals Court, and so ordered. Id. at 682, quoting Department of Revenue v. Jarvenpaa, 404 Mass. 177, 181 (1989). For similar reasons, we reach the same conclusion here with respect to harassment prevention orders under c. 258E. For example, where someone alleges abuse from a former spouse and the spouse’s current partner, and seeks an order against the former under c. 209A and against the latter under c. 258E, it would not serve the interests of justice for an appeal from the c. 209A order to be heard in the Appeals Court and an appeal from the c. 258E order to be heard by a single justice of the Supreme Judicial Court. After the date of the rescript in this case, all litigants seeking judicial review of such orders are directed to the Appeals Court.

1. Statutory framework. Before we turn to the constitutional questions raised by O’Brien, we consider first how the act fits within the over-all statutory framework protecting victims of abuse and harassment. General Laws c. 209A, inserted by St. 1978, c. 447, § 2, enables a person “suffering from abuse from an adult or minor family or household member” to obtain a protective order directing the defendant, among other things, to refrain from the abuse. G. L. c. 209A, § 3. “Abuse” is defined as “the occurrence of one or more of the following acts between family or household members: (a) attempting to cause or causing physical harm; (b) placing another in fear of imminent serious physical harm; (c) causing another to engage involuntarily in sexual relations by force, threat or duress.” G. L. c. 209A, § 1. The violation of an abuse prevention order is a crime, punishable by a fine or imprisonment in a house of correction. G. L. c. 209A, § 7.

A person who is abused by someone other than a “family or household member” does not qualify for a protective order under c. 209A and could obtain a restraining order only by seeking injunctive relief in the Superior Court under Mass. R. Civ. P. 65, 365 Mass. 832 (1974). Violation of such a restraining order may constitute a contempt of court, but is not a crime. Mass. R. Civ. P. 65.3, as appearing in 386 Mass. 1244 (1982).

Chapter 258E was enacted in 2010 to allow individuals to obtain civil restraining orders against persons who are not family or household members, and to make the violation of those orders punishable as a crime. See An Act Relative to Harassment Prevention Orders, Formal House Session January 28, 2010 (statements of Representatives O’Flaherty, Atkins, Jones, Swan). But while a protective order under c. 209A requires a finding of “abuse,” a protective order under c. 258E requires a finding of “harassment,” defined in G. L. c. 258E, § 1, as “[three] or more acts of willful and malicious conduct aimed at a specific person committed with the intent to cause fear, intimidation, abuse or damage to property and that does in fact cause fear, intimidation, abuse or damage to property.” 4

The definition of “[h]arassment” in c. 258E, § 1, evolved from the criminal harassment statute enacted in 2000, G. L. c. 265, § 43A, inserted by St. 2000, c. 164, which makes it a crime to “willfully and maliciously engageQ in a knowing pattern of conduct or series of acts over a period of time directed at a specific person, which seriously alarms that person and would cause a reasonable person to suffer substantial emotional distress.” 5

Both civil and criminal harassment require proof of three or more acts of wilful and malicious conduct aimed at a specific person. See Commonwealth v. Welch, 444 Mass. 80, 89 (2005) (Welch) (“phrase ‘pattern of conduct or series of acts’ [in G. L. c. 265, § 43A,] requires the Commonwealth to prove three or more incidents of harassment”). But the definitions of civil and criminal harassment differ in three respects. First, there are two layers of intent required to prove civil harassment under c. 258E: the acts of harassment must be wilful and “[m]alicious,” the latter defined as “characterized by cruelty, hostility or revenge,” and they must be committed with “the intent to cause fear, intimidation, abuse or damage to property.” G. L. c. 258E, § 1. Only the first layer of intent is required for criminal harassment under c. 265, § 43A. Second, the multiple acts of civil harassment must “in fact cause fear, intimidation, abuse or damage to property,” while the multiple acts of criminal harassment must “seriously alarm[]” the targeted victim. Third, criminal harassment requires proof that the pattern of harassment “would cause a reasonable person to suffer substantial emotional distress,” but civil harassment has no comparable reasonable person element.

2. Constitutionality of G. L. c. 258E. In Welch, we considered the constitutionality of the criminal harassment statute, G. L. c. 265, § 43A, and concluded that, while the harassing conduct punishable under the statute may include harassing speech, the statute passed constitutional muster because it was carefully crafted by the Legislature to apply “solely to constitutionally unprotected speech.” Id. at 99. We recognized that we had “narrowly construed statutory language in the past to avoid constitutional problems,” but we declined “to narrow the statute by engrafting onto it a savings clause or other limiting construction at this time.” Id. at 99-100 & n.16. We added that “we would have no hesitation in reading into the statute such a narrowing construction to ensure its application only to speech that is accorded no constitutional protection” if the Commonwealth were to attempt to prosecute a person for speech that is constitutionally protected. Id. at 100.

Having concluded in Welch that the criminal harassment statute is not constitutionally overbroad, the issue we confront here is whether the differences in the definition of civil harassment so significantly broaden the scope of potentially prohibited speech as to render the civil harassment statute constitutionally over-broad. Before we embark on that analysis, we explore first the overbreadth doctrine and the relevant categories of unprotected speech.

a. Overbreadth. We recently noted:

“The overbreadth doctrine allows an individual whose speech may be constitutionally regulated to argue that a law is unconstitutional because it infringes on the speech of others. See [Trustees of the State Univ. of N.Y. v. Fox, 492 U.S. 469, 482-483 (1989)]. See also United States v. Stevens, 130 S. Ct. 1577, 1591-1592 (2010). Overbreadth has thus been described as an exception to the general principle that litigants only have standing to assert their own rights and not the rights of others; in the free speech context, such challenges have been permitted in order ‘to prevent [a] statute from chilling the First Amendment rights of other parties not before the court.’ Secretary of State of Md. v. Joseph H. Munson Co., 467 U.S. 947, 957, 958 (1984). See Eisenstadt v. Baird, 405 U.S. 438, 445 n.5 (1972). . . .

“The Supreme Court has recognized the overbreadth doctrine as ‘strong medicine’ and has limited its application to instances where a law ‘prohibits a substantial amount of protected speech.’ United States v. Williams, 553 U.S. 285, 292, 293 (2008), quoting Los Angeles Police Dep’t v. United Reporting Publ. Corp., 528 U.S. 32, 39 (1999). Substantial overbreadth demands a ‘realistic danger that the statute itself will significantly compromise recognized First Amendment protections of parties not before the Court.’ City Council of Los Angeles v. Taxpayers for Vincent, 466 U.S. 789, 800-802 (1984) (declining to entertain overbreadth claim where such danger not shown). Further, the overbreadth must be ‘substantial, not only in an absolute sense, but also relative to the statute’s plainly legitimate sweep.’ United States v. Williams, supra at 292.”

Bulldog Investors Gen. Partnership v. Secretary of the Commonwealth, 460 Mass. 647, 676-677 (2011).

In determining whether a statute is facially overbroad, we look not only to the language of the statute but to any limiting construction we have grafted on that language in interpreting the statute. See Commonwealth v. A Juvenile, 368 Mass. 580, 585 (1975). And we have not hesitated to construe statutory language narrowly to avoid constitutional overbreadth, especially where we discern a legislative intent that the statute prohibit only constitutionally unprotected speech. See Welch, supra at 99, 100 n.16; Commonwealth v. Templeman, 376 Mass. 533, 538 (1978).

b. Classes of unprotected speech. While most speech is protected from government regulation by the First Amendment to the United States Constitution and art. 16 of the Massachusetts Declaration of Rights, as amended by art. 77 of the Amendments to the Massachusetts Constitution, there are “certain well-defined and narrowly limited classes of speech” that are not protected because they are “no essential part of any exposition of ideas, and are of such slight social value as a step to truth” that whatever meager benefit that may be derived from them is “clearly outweighed” by the dangers they pose. Chaplinsky v. New Hampshire, 315 U.S. 568, 571, 572 (1942). See Shackelford v. Shirley, 948 F.2d 935, 938 (5th Cir. 1991), quoting L. Tribe, American Constitutional Law § 12-8, at 836-837 (2d ed. 1988) (“It is not plausible to uphold the right to use words as projectiles where no exchange of views is involved”). Among these well-defined and narrowly limited classes of speech are “fighting words” and “true threats.” See Welch, supra at 94-95 (First Amendment allows restrictions on speech that are “fighting words” or “true threats”). See also Madsen v. Women’s Health Ctr., Inc., 512 U.S. 753, 774 (1994) (“Absent evidence that the protesters’ speech is independently proscribable [i.e., ‘fighting words’ or threats], or is so infused with violence as to be indistinguishable from a threat of physical harm, . . . this provision cannot stand” [citation omitted]).

i. Fighting words. The “fighting words” exception to the First Amendment is limited to words that are likely to provoke a fight: face-to-face personal insults that are so personally abusive that they are plainly likely to provoke a violent reaction and cause a breach of the peace. See Cohen v. California, 403 U.S. 15, 20 (1971) (fighting words are “personally abusive epithets which, when addressed to the ordinary citizen, are, as a matter of common knowledge, inherently likely to provoke violent reaction”); Chaplinsky v. New Hampshire, supra at 572 (fighting words are words whose “very utterance . . . tend to incite an immediate breach of the peace”); Commonwealth v. A Juvenile, supra at 591. Fighting words thus have two components: they must be a direct personal insult addressed to a person, and they must be inherently likely to provoke violence. As such, the fighting words exception is “an extremely narrow one.” Johnson v. Campbell, 332 F.3d 199, 212 (3d Cir. 2003).

ii. True threats. In Virginia v. Black, 538 U.S. 343 (2003) (Black), the Supreme Court defined true threats:

“ ‘True threats’ encompass those statements where the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals. . . . The speaker need not actually intend to carry out the threat. Rather, a prohibition on true threats ‘protects] individuals from the fear of violence’ and ‘from the disruption that fear engenders,’ in addition to protecting people ‘from the possibility that the threatened violence will occur.’ ” (Citations omitted.)

Id. at 359-360, quoting R.A.V. v. St. Paul, 505 U.S. 377, 388 (1992). See United States v. Cassel, 408 F.3d 622, 633 (9th Cir. 2005) (Black Court’s definition of “true threat” binding on lower courts). See also Commonwealth v. Robicheau, 421 Mass. 176, 183 (1995) (“First Amendment does not protect conduct that threatens another”). “The term ‘true threat’ has been adopted to help distinguish between words that literally threaten but have an expressive purpose such as political hyperbole, and words that are intended to place the target of the threat in fear, whether the threat is veiled or explicit.” Commonwealth v. Chou, 433 Mass. 229, 236 (2001) (Chou).

A true threat does not require “an explicit statement of an intention to harm the victim as long as circumstances support the victim’s fearful or apprehensive response.” Id. at 234. See United States v. Fulmer, 108 F.3d 1486, 1492 (1st Cir. 1997) (“use of ambiguous language does not preclude a statement from being a threat”); United States v. Malik, 16 F.3d 45, 49 (2d Cir.), cert. denied, 513 U.S. 968 (1994) (“absence of explicitly threatening language does not preclude the finding of a threat”). Nor need a true threat threaten imminent harm; sexually explicit or aggressive language “directed at and received by an identified victim may be threatening, notwithstanding the lack of evidence that the threat will be immediately followed by actual violence or the use of physical force.” Chou, supra at 235. See Black, supra at 359-360 (defining true threats without imminence requirement); Doe v. Pulaski County Special Sch. Dist., 306 F.3d 616, 622 (8th Cir. 2002) (“serious expression of an intent to cause a present or future harm” is true threat).

For example, in Chou, supra at 230-231, the defendant produced flyers with the word “MISSING” printed in large type across the top, beneath which was a photograph of a former girl friend who had broken up with him and offensive sexual remarks about her, and he then hung the flyers in several places in her high school. He was convicted under G. L. c. 272, § 53, id. at 230, which criminally punishes “persons who with offensive and disorderly acts or language accost or annoy persons of the opposite sex.” We concluded that the defendant’s criminal conviction for this communication did not violate the First Amendment because it was a true threat that “had no expressive purpose but was, instead, intended to ‘get back’ at the victim by placing her in fear that she might suffer some sexual harm or wind up among the ‘missing.’ ” Id. at 237. We recognized that “[sexually explicit language, when directed at particular individuals in settings in which such communications are inappropriate and likely to cause severe distress, may be inherently threatening.” Id. at 234.

In Black, the Supreme Court upheld the constitutionality of a Virginia law that prohibited the burning of a cross with intent to intimidate, concluding that, in view of the Ku Klux Klan’s “long and pernicious history” of using the burning of a cross “as a signal of impending violence,” it constituted a true threat, unprotected by the First Amendment, when committed with an intent to intimidate. Black, supra at 363. 6 The Supreme Court noted that “[ijntimidation in the constitutionally prescribable sense of the word is a type of true threat, where a speaker directs a threat to a person or group of persons with the intent of placing the victim in fear of bodily harm or death.” Id. at 360. Taken together, Chou and Black demonstrate that the “true threat” doctrine applies not only to direct threats of imminent physical harm, but to words or actions that — taking into account the context in which they arise — cause the victim to fear such harm now or in the future and evince intent on the part of the speaker or actor to cause such fear.

c. Construction of G. L. c. 258E. We conclude that the Legislature crafted the civil harassment act, G. L. c. 258E, with the intent that the definition of harassment exclude constitutionally protected speech, and we interpret G. L. c. 258E to effectuate that legislative intent. See Welch, supra at 99 (finding that Legislature took similar care in crafting criminal harassment act, G. L. c. 265, § 43A). Because the definition of “civil harassment” is substantially broader than the definition of “fighting words,” we discern no legislative intent to confine the meaning of harassment to fighting words, but we do discern an intent to confine the meaning of harassment to either fighting words or “trae threats.” 7

Looking carefully at the elements set forth in the definition of civil harassment, we conclude that, with one exception, the Legislature accomplished that intent through its crafting of G. L. c. 258E. To establish harassment, a complainant must prove that the defendant, motivated by cruelty, hostility, or revenge, wilfully committed three or more acts aimed at a specific person, each with the intent to cause that person to experience fear or intimidation, or to cause abuse or damage to property, which, considered together, did in fact cause fear, intimidation, abuse, or damage to property. G. L. c. 258E, § l. 8

The intent requirements in the act plainly satisfy the “true threat” requirement that the speaker subjectively intend to communicate a threat. See Black, supra at 360; United States v. Cassel, supra. See also Shackelford v. Shirley, 948 F.2d 935, 937, 940 (5th Cir. 1991) (specific intent requirement in statute punishing making of telephone call “with intent to terrify, intimidate, or harass, and threaten to inflict injury or physical harm” limited punishable speech only to “a class of ‘true threats,’ and not social or political advocacy”).

The requirement that the pattern of harassment in fact cause fear, intimidation, abuse, or damage to property satisfies the “true threat” requirement that the threat be regarded as a serious expression of intent and not mere hyperbole. See Black, supra at 359; Chou, supra at 236-237.

Where the acts are aimed at a specific person, an intent to cause “abuse” is certainly consistent with a true threat, because abuse is defined in the act as “attempting to cause or causing physical harm to another or placing another in fear of imminent serious physical harm.” G. L. c. 258E, § 1. An intent to cause property damage is also consistent with a true threat because, as with intent to cause abuse, it constitutes an intent to commit a criminal act. So, too, is an intent to intimidate, which the Supreme Court specifically concluded was punishable when coupled with an intimidating communication. See Black, supra at 363.

An intent to cause fear, however, is less certain to be consistent with a true threat unless we narrow what it is that must be feared. As the United States Court of Appeals for the Seventh Circuit noted, there are types of threats that may “contain ideas or advocacy, such as a ‘threat’ to picket an organization if it does not yield to a demand to take some social or political action.” United States v. Velasquez, 772 F.2d 1348, 1357 (7th Cir. 1985), cert. denied, 475 U.S. 1021 (1986). Such a threat may cause a fear of economic loss, of unfavorable publicity, or of defeat at the ballot box, but such fears cannot be enough to make the threat a “true threat” that may be prohibited as civil harassment under the First Amendment and art. 16. There is nothing in the language of the act or in its legislative history to suggest that the Legislature intended “fear” to mean more than fear of physical harm or property damage, and such a limiting interpretation would effectuate the legislative intent to confine the prohibited speech in the act to constitutionally unprotected speech. Therefore, we narrow the meaning of “fear” under the act to fear of physical harm or fear of physical damage to property.

O’Brien contends that, even with this interpretation of “fear,” the civil harassment act is still constitutionally overbroad because, in contrast with the criminal harassment act, it does not require that the pattern of harassment would cause a reasonable person to suffer substantial emotional distress. The legislative history of c. 258E suggests that the Legislature intended to omit this reasonable person element, because it, along with other language from the criminal harassment act, was included in the initial version of the bill, but this element was removed from later versions and from the bill ultimately enacted. Compare 2009 Senate Doc. No. 1611, An Act relative to sexual assault and stalking restraining orders (filed Jan. 14, 2009), with 2009 Senate Doc. No. 2185, An Act to prevent harassment at § 1 (filed Oct. 26, 2009); 2009 Senate Doc. No. 2212, An Act relative to harassment prevention orders at § 1 (filed Nov. 17, 2009); and c. 258E, § 1. In its place, the Legislature inserted the requirements that the defendant intend to cause fear, intimidation, abuse, or damage to property, and that the pattern of harassment actually cause fear, intimidation, abuse, or damage to property. We conclude that these requirements adequately ensure that protected speech will not be found to be civil harassment in violation of c. 258E.

There is no appreciable amount of protected speech where the speaker both intends to cause intimidation, abuse, damage to property, or fear of physical harm or property damage, and does in fact cause one of these alternatives. A reasonable person standard may have been necessary to uphold the constitutionality of the criminal harassment act, which otherwise required only that the knowing and malicious pattern of conduct “seriously alarm” the person harassed, see Welch, supra at 97, but the additional requirements in the civil harassment act more carefully limit the scope of prohibited speech to constitutionally unprotected “true threats.”

Therefore, we conclude that, with our interpretation of the meaning of “fear” in G. L. c. 258E, § 1, the civil harassment act is not constitutionally overbroad because it does not prohibit a substantial amount of protected speech, either in an absolute sense or in relation to the statute’s legitimate sweep. See Bulldog Investors Gen. Partnership v. Secretary of the Commonwealth, 460 Mass. 647, 676-677 (2011).

3. Application of G. L. c. 258E to O’Brien’s conduct. We now turn to O’Brien’s claim that the act was unconstitutionally applied to him because the finding of harassment was based on his raising of his middle finger, which is constitutionally protected speech. We recognize that the raising of the middle finger as a form of insult has a long, if not illustrious, history dating back to ancient Greece. See Robbins, Digitus Impudicus: The Middle Finger and the Law, 41 U.C. Davis L. Rev. 1403, 1413 (2008). 9 Like its verbal counterpart, when it is used to express contempt, anger, or protest, it is a form of expression protected by the First Amendment. See, e.g., Sandul v. Larion, 119 F.3d 1250, 1255 (6th Cir. 1997) (passenger yelling “fuck you” and extending middle finger while passing group of protestors entitled to First Amendment protection); Duran v. Douglas, 904 F.2d 1372, 1374, 1378 (9th Cir. 1990) (“obscene gesture” and profanities directed to police, while “[i]narticulate and crude,” “represented an expression of disapproval toward a police officer” that “fell squarely within the protective umbrella of the First Amendment”).

But, in certain limited circumstances, when accompanied by other less expressive and more threatening conduct, raising the middle finger may constitute fighting words or a true threat. Compare United States v. Poocha, 259 F.3d 1077, 1082 (9th Cir. 2001) (clenching fists and yelling “fuck you” to police officer not fighting words), and Cook v. County Comm’rs of Wyandotte County, 966 F. Supp. 1049, 1051, 1052 (D. Kan. 1997) (raising middle finger at police officer parked in patrol car not fighting words), with Gower v. Vercler, 377 F.3d 661, 664, 670 (7th Cir. 2004) (repeatedly yelling “fuck you” and insults, and attempting to humiliate, constitutes fighting words where parties had brandished weapons at each other previous night), and State v. Groves, 219 Neb. 382, 384, 386 (1985) (calling police officer “motherfucker” and challenging him to make arrest alone after learning that officer had called for assistance were fighting words).

Borowski does not contend that O’Brien’s use of his middle finger constituted fighting words that incited him to violence. Rather, he argues that, when viewed in context, where Borowski had earlier arrested O’Brien and knew O’Brien to be a “fighter,” and where O’Brien followed Borowski out of a bar to raise both middle fingers in the first alleged act of harassment and later raised his middle finger as he drove past Borowski’s home in the second and third alleged acts of harassment, O’Brien’s raising of his middle finger reflected a threat of physical harm and therefore is not protected by the First Amendment or art. 16.

We need not reach the issue whether there was sufficient evidence for the judge to conclude that the totality of O’Brien’s conduct, including the raising of the middle finger, met the standard for civil harassment because there is another issue that, if the case were not moot, would require a remand for further factual findings. We noted earlier that, under the act, the plaintiff must prove at least three wilful and malicious acts of harassment directed to a specific person, each committed with the intent to cause intimidation, abuse, damage to property, or fear of physical injury or property damage. Because the judge did not make any factual findings, orally or in writing, we cannot know whether the judge made this finding as to each of the three alleged acts of harassment, and the evidence is not so strong as to permit us to infer such a finding, especially with respect to the first act committed outside the bar. In such circumstances, we would generally remand the case to the judge for further factual findings, but such a remand makes no sense where the case is moot, the harassment prevention order having expired. Because we cannot affirm the order of harassment without further findings, and because the case is now moot, the order of harassment must be vacated.

Conclusion. We interpret the word “fear” in G. L. c. 258E, § 1, to mean fear of physical harm or fear of physical damage to property. With that narrowed construction, we conclude that the civil harassment act, G. L. c. 258E, is not constitutionally overbroad because it limits the scope of prohibited speech to constitutionally unprotected “true threats” and “fighting words.” Because the case is moot as a consequence of the expiration of the harassment prevention order and because further findings from the judge would be needed to affirm the finding of civil harassment, we remand the case to the county court and direct the single justice to vacate the order of harassment. We also order that, after the date of the rescript in this case, all litigants seeking judicial review of harassment prevention orders under G. L. c. 258E are directed to the Appeals Court.

So ordered.

1 General Laws c. 258E, inserted by St. 2010, c. 23, became effective on May 10, 2010.

2 G’Brien also claims on appeal that the harassment prevention order against him constituted a prior restraint on speech. O’Brien, however, did not make this claim before the District Court judge or the single justice, and therefore we will not consider it on appeal. See, e.g., Century Fire & Marine Ins. Corp. v. Bank of New England-Bristol County, N.A., 405 Mass. 420, 421 n.2 (1989).

3 We acknowledge the amicus briefs submitted by the American Civil Liberties Union of Massachusetts, and by the Victim Rights Law Center, Boston Area Rape Crisis Center, and Jane Doe, Inc.

4 General Laws c. 258E, § 1, provides an alternative definition of “[h]arassment” that does not apply to the facts of this case and is not claimed by O’Brien to be unconstitutionally overbroad:

“[A]n act that (A) by force, threat or duress causes another to involuntarily engage in sexual relations; or (B) constitutes a violation of section 13B [indecent assault and battery on child under the age of fourteen], 13F [indecent assault and battery on a person with an intellectual disability], 13H [indecent assault and battery on a person fourteen years or older], 22 [rape], 22A [rape of a child], 23 [rape and abuse of a child], 24 [assault with intent to commit rape], 24B [assault of a child with intent to commit rape], 26C [enticement of a child under sixteen], 43 [stalking] or 43A [criminal harassment] of chapter 265 or section 3 of chapter 272 [drugging a person for sexual intercourse].”

5 The crime of harassment, in violation of G. L. c. 265, § 43A, is a lesser included offense of criminal stalking, in violation of G. L. c. 265, § 43, which has the additional element of proof of a “threat with the intent to place the [victim] in imminent fear of death or bodily injury.” See Commonwealth v. Kulesa, 455 Mass. 447, 451 & n.6 (2009); Commonwealth v. Welch, 444 Mass. 80, 88 (2005) (Welch).

6 A plurality of the Supreme Court also concluded that cross burning was protected speech if not done with an intent to intimidate. Virginia v. Black, 538 U.S. 343, 365-367 (2003) (plurality opinion).

7 We recognize that this constitutes a departure from our analysis in Welch, supra at 94 n.14, where we declared that the “ ‘true threats’ exception is not applicable to the criminal harassment statute because the statute was enacted to criminalize those acts of harassment that do not rise to the level of threats.” The threat element in the criminal stalking statute that was omitted from the criminal harassment statute, however, required a “threat with the intent to place the person in imminent fear of death or bodily injury” (emphasis added). G. L. c. 265, § 43. Because a “true threat” need not produce an imminent fear of harm, we erred in concluding that “true threats” analysis did not apply to criminal harassment. Welch, supra. Having unnecessarily excluded “true threats” from our constitutional analysis, we similarly erred in concluding that the criminal harassment statute was limited in its reach to “fighting words” rather than both “fighting words” and “true threats.” See id. at 98-99.

8 We interpret G. L. c. 258E, § 1, to require proof that the defendant had the requisite intent in committing each of the three required acts of wilful and malicious conduct. A plaintiff seeking protection through a civil harassment order must show that the defendant engaged in at least three wilful and malicious acts, and that for each act the defendant intended to cause fear, intimidation, abuse, or damage to property. But, in determining whether these three acts did “in fact cause fear, intimidation, abuse or damage to property,” the fact finder must look to the cumulative pattern of harassment, and need not find that each act in fact caused fear, intimidation, abuse, or damage to property. General Laws c. 258E, § 1, defines harassment as “[three] or more acts of willful and malicious conduct. . . that does in fact cause fear, intimidation, abuse or damage to property” (emphasis added). The use of the singular “does” rather than the plural “do” indicates that it is the entire course of harassment, rather than each individual act, that must cause fear or intimidation.

9 The gesture’s history in this country dates back to at least 1886, when its first recorded appearance “occurred in a professional baseball team photograph, where a pitcher for the Boston Beaneaters gave the middle finger while posing for a joint team picture with the New York Giants.” Robbins, Digitus Impudicus: The Middle Finger and the Law, 41 U.C. Davis L. Rev. 1403, 1415 (2008).