Skip to main content


Reset A A Font size: Print

United States Court of Appeals, First Circuit.

Hayat SINDI, Plaintiff, Appellee, v. Samia EL-MOSLIMANY and Ann El-Moslimany, Defendants, Appellants.

No. 16-2347

Decided: July 11, 2018

Before Barron, Selya and Stahl, Circuit Judges. John A. Kiernan, with whom Bonner Kiernan Trebach & Crociata, LLP was on brief, Boston, MA, for appellants. Eugene Volokh, pro se, on brief for Eugene Volokh, amicus curiae. David H. Rich, with whom Suzanne Elovecky and Todd & Weld LLP were on brief, Boston, MA, for appellee.

This case implicates a plethora of issues arising in the shadow of the First Amendment. Most notably, it requires us to address the power of a court to impose a prior restraint in the form of a permanent injunction forbidding the publication of words—words that the court believes have been used to defame the plaintiff in the past and are likely to be repeated. The case also presents issues as to whether, consistent with the First Amendment and state law, the evidence adduced at trial allowed the jury to find defendant-appellant Samia El-Moslimany (Samia) liable for intentional infliction of emotional distress and to find Samia and her mother, defendant-appellant Ann El-Moslimany (Ann), liable for defamation, tortious interference with contract, and tortious interference with advantageous relations. Finally, it presents issues as to whether the damages awarded on these claims, totaling in the millions of dollars, are excessive.

After careful consideration, we conclude that the district court's permanent injunction cannot survive the strict scrutiny that the Constitution demands for prior restraints on speech. Thus, we vacate the injunction. We affirm the jury's findings of liability on most (but not all) of Dr. Sindi's tort claims and affirm the corresponding money judgments (some that represent the jury's assessment of damages and some that represent the district court's remittitur of jury awards). Not so the claim for tortious interference with advantageous relations: finding the evidence insufficient, we vacate the jury awards on that claim and direct the entry of judgment for the appellants.


We offer only a sketch of the relevant events and travel of the case, reserving a fuller elaboration for our discussion of specific issues. For these purposes, we take the facts in the light most hospitable to the jury verdict, consistent with record support. See Casillas-Díaz v. Palau, 463 F.3d 77, 79 (1st Cir. 2006).

In November of 2010, Samia and her husband, Fouad Dehlawi, hosted a Thanksgiving dinner at their Seattle-area home. Their guest list included the plaintiff, Dr. Hayat Sindi, a prominent Saudi scientist and entrepreneur who was then a visiting scholar at Harvard University. Several months later, Samia came to believe that her husband and Dr. Sindi were engaged in a meretricious relationship. For the next five years, Samia and Ann published a series of web posts pertaining to Dr. Sindi in a variety of forums, including, Facebook, the Washington Post website, and various blogs. They also sent e-mails regarding Dr. Sindi to members of the scientific community and to investors in Dr. Sindi's Institute for Imagination and Ingenuity (i2 Institute). Among other calumnies, the appellants accused Dr. Sindi of fraudulently obtaining her doctorate by paying a colleague to ghostwrite her dissertation, repeatedly lying about her age in order to obtain awards meant for younger scientists, and inflating her resumé by falsely touting her role in Harvard's Diagnostics for All initiative.

Dr. Sindi did not take this campaign of vilification lightly. On January 25, 2013, she sued Samia and Ann in a Massachusetts state court. Her complaint alleged defamation, intentional infliction of emotional distress, tortious interference with contract, and tortious interference with advantageous relations. Citing diversity of citizenship and the existence of a controversy in the requisite amount, Samia and Ann removed the case to the federal district court. See 28 U.S.C. §§ 1332(a), 1441(a). Following some pretrial skirmishing (not relevant here) and extensive discovery, the case went to trial on July 11, 2016.

The trial lasted seven days (exclusive of jury deliberations). At the close of all the evidence, the district court denied the appellants' motion for judgment as a matter of law, see Fed. R. Civ. P. 50(a), and sent the case to the jury. In the course of its jury instructions, the court encouraged the jurors to consult a nine-page document (referred to as a “chalk”), which listed approximately 132 allegedly defamatory statements attributed to Samia and/or Ann.1 Neither Samia nor Ann objected to this portion of the instructions.

The jury returned a general verdict in Dr. Sindi's favor on all but one of the submitted claims. It found Samia liable for intentional infliction of emotional distress; absolved Ann of that charge; and found both Samia and Ann liable for defamation, tortious interference with contract, and tortious interference with advantageous relations. The jury awarded damages totaling $3,500,000.2

The jury verdict generated a flurry of post-trial activity. Samia and Ann renewed their motion for judgment as a matter of law, see Fed. R. Civ. P. 50(b), and moved alternatively for either a new trial or a remittitur, see Fed. R. Civ. P. 59(a), (e). For her part, Dr. Sindi moved for a permanent injunction, seeking to enjoin Samia and Ann from uttering or otherwise publishing a multitude of described statements. On August 18, 2016, the district court granted Dr. Sindi's motion and enjoined the appellants from publishing “orally, in writing, through direct electronic communications, or by directing others to websites or blogs reprinting” six statements that the district court concluded were defamatory.

Some six weeks later, the district court denied the appellants' motion for judgment as a matter of law. At the same time, the court denied their alternative motion for a new trial or a remittitur, with two exceptions. First, the court granted a remittitur of the damages awarded against Samia for tortious interference with contract (directing Dr. Sindi to remit all of the $2,000,000 verdict on that claim in excess of $576,000). See Sindi v. El-Moslimany, No. 13-cv-10798, 2016 WL 5867403, at *6 (D. Mass. Oct. 6, 2016). Second, it granted a remittitur of the damages awarded against Ann for tortious interference with contract (directing Dr. Sindi to remit all of the $400,000 verdict on that claim in excess of $144,000). See id. The court proceeded to enter an amended final judgment, which included prejudgment interest, see Mass. Gen. Laws ch. 231, § 6B, costs, and the permanent injunction.3

This timely appeal ensued. Following oral argument, we directed the parties to submit supplemental briefs designed to answer certain questions affecting the validity vel non of the permanent injunction. We have received those supplemental briefs, along with a thoughtful amicus brief, and the appeal is now ripe for decision.


We review the district court's denial of a motion for judgment as a matter of law de novo. See Trainor v. HEI Hosp., LLC, 699 F.3d 19, 26 (1st Cir. 2012). In conducting this tamisage, we examine the record in the light most favorable to the nonmovant and will reverse “only if reasonable persons could not have reached the conclusion that the jury embraced.” Sanchez v. P.R. Oil Co., 37 F.3d 712, 716 (1st Cir. 1994).

Our review of the district court's denial of a motion for a new trial under Rule 59 “is even more circumscribed.” Id. at 717. A trial court may “set aside a jury's verdict and order a new trial only if the verdict is against the demonstrable weight of the credible evidence or results in a blatant miscarriage of justice.” Id. When a movant attacks an award of damages as excessive, a court may remit the award only if “the award exceeds any rational appraisal or estimate of the damages that could be based upon the evidence before it.” Trainor, 699 F.3d at 29 (quoting Wortley v. Camplin, 333 F.3d 284, 297 (1st Cir. 2003) ). We review the district court's adjudication of a motion for either a new trial or a remittitur for abuse of discretion. See id.; Sanchez, 37 F.3d at 717.

Since this case comes to us by means of our diversity jurisdiction, we must look to state law for the substantive rules of decision. See Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938); Sanders v. Phoenix Ins. Co., 843 F.3d 37, 42 (1st Cir. 2016). In this instance, we—like the court below—follow the parties' lead and look to the substantive law of Massachusetts. See Shay v. Walters, 702 F.3d 76, 80 (1st Cir. 2012).


We begin our analysis with the defamation claims. In Massachusetts, a defamation plaintiff must establish that “[t]he defendant made a statement, concerning the plaintiff, to a third party”; that such “statement could damage the plaintiff's reputation in the community”; that “[t]he defendant was at fault in making the statement”; and that “[t]he statement either caused the plaintiff economic loss ․ or is actionable without proof of economic loss.” Ravnikar v. Bogojavlensky, 438 Mass. 627, 782 N.E.2d 508, 510–11 (2003). “A false statement that ‘would tend to hold the plaintiff up to scorn, hatred, ridicule or contempt, in the minds of any considerable and respectable segment in the community,’ [is] considered defamatory.” Phelan v. May Dep't Stores Co., 443 Mass. 52, 819 N.E.2d 550, 553 (2004) (quoting Stone v. Essex Cty. Newspapers, Inc., 367 Mass. 849, 330 N.E.2d 161, 165 (1975) ).

The First Amendment, made applicable to the states through the Fourteenth Amendment, overlays state defamation law and imposes a number of constraints on a plaintiff who seeks relief for defamation.4 See N.Y. Times Co. v. Sullivan, 376 U.S. 254, 276–77, 283–84, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964). This is as it should be: “it is essential that the First Amendment protect some erroneous publications as well as true ones” in order “to insure the ascertainment and publication of the truth about public affairs.” St. Amant v. Thompson, 390 U.S. 727, 732, 88 S.Ct. 1323, 20 L.Ed.2d 262 (1968). It follows that a public figure may recover for defamation only if she proves actual malice by clear and convincing evidence. See Gertz v. Robert Welch, Inc., 418 U.S. 323, 342, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974). That is, such a plaintiff must demonstrate with convincing clarity that “the defamatory falsehood was made with knowledge of its falsity or with reckless disregard for the truth.” Id. This requirement applies both to plaintiffs whose “pervasive fame or notoriety” makes them “public figure[s] for all purposes and in all contexts” and to plaintiffs who are public figures with respect to the “limited range of issues” surrounding the claimed defamation. Id. at 351, 94 S.Ct. 2997.

In proving actual malice, a defamation plaintiff must shoulder a heavy burden. The Supreme Court has underscored that “[a] reckless disregard for the truth ․ requires more than a departure from reasonably prudent conduct.” Harte–Hanks Commc'ns, Inc. v. Connaughton, 491 U.S. 657, 688, 109 S.Ct. 2678, 105 L.Ed.2d 562 (1989) (internal quotation marks omitted). Thus, a public-figure plaintiff must point to clear and convincing evidence that the defendant made the challenged statement with a “high degree of awareness of [its] probable falsity,” Vascular Sols., Inc. v. Marine Polymer Techs., Inc., 590 F.3d 56, 60 (1st Cir. 2009) (per curiam) (quoting Garrison v. Louisiana, 379 U.S. 64, 74, 85 S.Ct. 209, 13 L.Ed.2d 125 (1964) ), or “entertained serious doubts as to the truth of his publication,” id. (quoting St. Amant, 390 U.S. at 731, 88 S.Ct. 1323).

Of course, a statement is not actionable “unless in a given context it reasonably can be understood as having an easily ascertainable and objectively verifiable meaning.” Levinsky's, Inc. v. Wal–Mart Stores, Inc., 127 F.3d 122, 129 (1st Cir. 1997). Statements that are merely “ ‘imaginative expression’ ” or “ ‘rhetorical hyperbole’ ”—in other words, statements that “no reasonable person would believe presented facts”—are not actionable. Id. at 128 (quoting Milkovich v. Lorain Journal Co., 497 U.S. 1, 17, 20, 110 S.Ct. 2695, 111 L.Ed.2d 1 (1990) ).

We caution, however, that the First Amendment does not command “a wholesale defamation exemption” for statements that “might be labeled ‘opinion[s].’ ” Milkovich, 497 U.S. at 18, 110 S.Ct. 2695. Rather, “[a] statement couched as an opinion that presents or implies the existence of facts which are capable of being proven true or false can be actionable.” Levinsky's, 127 F.3d at 127.

The First Amendment imposes yet another safeguard with respect to awards of damages for defamation. It requires an appellate court to review the supporting evidence independently. See Bose Corp. v. Consumers Union of U.S., Inc., 466 U.S. 485, 510–11, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984). Thus, we must afford plenary review to “mixed fact/law matters which implicate core First Amendment concerns,” such as the jury's conclusions regarding falsity and actual malice. AIDS Action Comm. of Mass., Inc. v. MBTA, 42 F.3d 1, 7 (1st Cir. 1994); see Hurley v. Irish-Am. Gay, Lesbian & Bisexual Grp. of Bos., 515 U.S. 557, 567, 115 S.Ct. 2338, 132 L.Ed.2d 487 (1995). Put another way, we must ensure that the jury's verdict “does not constitute a forbidden intrusion on the field of free expression.” N.Y. Times Co., 376 U.S. at 285, 84 S.Ct. 710.

Withal, “[i]ndependent review is not a limitless ransacking of the record as a whole.” Mandel v. Bos. Phoenix, Inc., 456 F.3d 198, 208 (1st Cir. 2006). The usual deferential Rule 50 standard applies to mixed fact/law questions that do not implicate First Amendment concerns. See Bose Corp., 466 U.S. at 514 n.31, 104 S.Ct. 1949. Causation is such a question. See Fiori v. Truck Drivers, Local 170, 354 F.3d 84, 89 (1st Cir. 2004). So, too, deference is due to the jury's assessment of witness credibility. See Hurley, 515 U.S. at 567, 115 S.Ct. 2338; Mandel, 456 F.3d at 208.


With this backdrop in place, we proceed to examine the vitriol-soaked comments that fueled the defamation claims at issue here. Our starting point is clear: Dr. Sindi, an appointee of Saudi King Abdullah to his government's Shura Council and a goodwill ambassador of the United Nations Educational, Scientific and Cultural Organization, concedes that she is at least a limited-purpose public figure. We must, therefore, independently mine the record to determine whether Dr. Sindi proved by clear and convincing evidence that Samia and Ann maliciously defamed her. See Gertz, 418 U.S. at 342, 94 S.Ct. 2997.

Following a thorough appraisal, we conclude—without serious question—that the defamation verdicts pass constitutional muster. While the record reflects a grotesque number of false statements that hold Dr. Sindi up to public scorn and contempt (including a majority of the statements memorialized on the chalk), the law of the case, as exemplified by the district court's unchallenged jury instructions, requires only that Dr. Sindi show that one or more defamatory statements were made.5 Therefore, no useful purpose would be served by evaluating separately each of the approximately 132 allegedly defamatory statements listed on the chalk. Given the law of the case, it suffices for us to shine the light of our inquiry on three categories of statements that were primary focal points of the trial. No more is exigible to validate the defamation verdicts under the district court's jury instructions.6


We start with Samia's repeated accusation—variously phrased and published in myriad web postings and in e-mails to members of the scientific community, journalists, investors in the i2 Institute, and State Department officials—that Dr. Sindi fraudulently obtained her Ph.D. from Cambridge University.7 Representative of this category of statements is a February 12, 2014, e-mail to the i2 Institute's board members and sundry journalists that:

[Dr. Sindi's] research was allegedly conducted and her dissertation written, by Adrian Stevenson, a postdoctoral and very intimate friend of Sindi. According to Sindi's live-in boyfriend from 2001 to 2005, throughout the writing of her dissertation, Stevenson was allegedly financially compensated by Sindi's father to act as her “bodyguard.” [Cambridge University Professor Christopher] Lowe confirmed that the writing style of her dissertation was clearly that of Stevenson, and that they were “very, very intimate friends.” Furthermore, Lowe believes that “money definitely changed hands.” Myer Berlow ․ also confirmed that she did not have the basic scientific or technical knowledge to have conducted the research or to have written her dissertation.

These statements have an easily decipherable and verifiable meaning, present the existence of specific facts that are capable of being proven false, and are more than mere rhetorical flights of fancy. See Levinsky's, 127 F.3d at 127–28. In addition, they are plainly defamatory: they impugn Dr. Sindi's professional competence while accusing her of fraud, notwithstanding the utter absence of any probative evidence contradicting Dr. Sindi's testimony regarding the elaborate research and writing process she undertook to complete her dissertation and obtain her degree. See Phelan, 819 N.E.2d at 553.

The question reduces, then, to whether the statements were made with actual malice, that is, either with knowledge of their falsity or with a reckless disregard for the truth. See Bose Corp., 466 U.S. at 513, 104 S.Ct. 1949. This inquiry is both subjective and time-sensitive, turning on “the defendant's state of mind at the time of publication.” Kahl v. Bureau of Nat'l Affairs, Inc., 856 F.3d 106, 118 (D.C. Cir. 2017). Since “direct evidence of actual malice is rare,” we have permitted actual malice to be proved through inference and circumstantial evidence alone. Levesque v. Doocy, 560 F.3d 82, 90 (1st Cir. 2009); see Connaughton, 491 U.S. at 668, 109 S.Ct. 2678. For example, actual malice “may be found where a publisher fabricates an account, makes inherently improbable allegations, relies on a source where there is an obvious reason to doubt its veracity, or deliberately ignores evidence that calls into question his published statements.” Levesque, 560 F.3d at 90. Although motive alone cannot suffice to prove actual malice, it is a highly relevant consideration. See Connaughton, 491 U.S. at 665, 667–68, 109 S.Ct. 2678; Vascular Sols., 590 F.3d at 61.

With respect to the “doctoral dissertation” statements, the jury was entitled to find that Samia fabricated material facts. Although Samia declared that the well-known entrepreneur and scientist, Myer Berlow, “confirmed” that Dr. Sindi lacked the prerequisite scientific or technical prowess to have written her dissertation, Berlow testified unequivocally that he had never made such a statement. Such a gross fabrication is powerful evidence of actual malice. See, e.g., St. Amant, 390 U.S. at 732, 88 S.Ct. 1323; Tosti v. Ayik, 394 Mass. 482, 476 N.E.2d 928, 936 (1985). To cinch the matter, Samia admitted during cross-examination that she had “no confirmed facts” to support her claim of fraud.

Nor was this all. The jury heard evidence that Samia deliberately ignored facts that called her public statements into question. For example, she admitted that she had no proof that any academic institution had ever investigated possible improprieties in connection with Dr. Sindi's doctorate. She also admitted that she had contact information for Dr. Stevenson (an academic who had publicly lauded Dr. Sindi's dissertation), yet she never reached out to him. On this record, the jury reasonably could have inferred that Samia deliberately chose not to contact Dr. Stevenson out of a concern that he would vouch for the legitimacy of Dr. Sindi's degree and thereby undercut Samia's criticisms. Refusing to take easily available steps that could confirm or refute a claim may constitute probative evidence of a reckless disregard for the truth. See Connaughton, 491 U.S. at 682–84, 109 S.Ct. 2678; Desnick v. Am. Broad. Cos., 233 F.3d 514, 517 (7th Cir. 2000).

Casting a further pall over Samia's statements is the fact that she had an obvious motive to besmirch Dr. Sindi's reputation: she believed that Dr. Sindi had engaged in an extramarital affair with her husband. In an e-mail dated December 17, 2011, Samia admonished Dr. Sindi that “you will rue the day you took advantage of my hospitality, came into my home, seduced [and] then tried to steal my husband.” In another e-mail, Samia informed Dr. Sindi that she and Ann had prayed that God would “expose[ ] [Dr. Sindi] and deliver[ ] justice.” Samia's vengeful motive, while insufficient on its own to establish actual malice, furnishes cogent evidence supporting such a finding. See Connaughton, 491 U.S. at 668, 109 S.Ct. 2678.

To be sure, Samia testified that several people had told her that Dr. Sindi obtained her Ph.D. through various sorts of chicanery and sleight of hand. But Samia did not produce any of those third parties as witnesses, and the jury was not required to credit Samia's second-hand and uncorroborated account. See id. at 688, 109 S.Ct. 2678 (noting that a jury's credibility assessments are reviewed for clear error, even in First Amendment cases).


The next group of statements involves Samia's accusations that Dr. Sindi (who was born on November 6, 1967) lied about her age in order to secure awards meant for younger scientists. Representative of these accusations is Samia's blog post on April 21, 2012, in which she wrote that Dr. Sindi “misrepresent[ed] her age” in order to win the 2007 Arab-American Science and Technology Young Professional Award, the 2009 PopTech Social Innovation Fellowship, and the 2011 National Geographic Emerging Scholar Award, thus “rob[bing] opportunities for recognition, public relations support, funding ․ and career advancement” from younger scientists. Similarly, in a letter to State Department officials dated February 12, 2014, Samia claimed that Dr. Sindi had misrepresented her age by some eleven years in connection with each of these awards.

We have scant difficulty in concluding that these statements are actionable. To begin, each statement about Dr. Sindi's age has “an easily ascertainable and objectively verifiable meaning.” Levinsky's, 127 F.3d at 129. Viewed in context, such statements had the undeniable potential to prejudice Dr. Sindi's professional and business endeavors. See Ravnikar, 782 N.E.2d at 511. What is more, the statements were demonstrably false: Dr. Sindi testified that she had never lied about her age to an award-granting entity, and Samia conceded that she had no competent evidence to the contrary.

Dr. Sindi also showed that these statements were made with actual malice. Samia confessed that she had never spoken to anyone with authority to award the prizes that she identified. In fact, she had done nothing even remotely resembling due diligence to verify her claim of mendacity. For aught that appears, Samia simply plucked the accusation out of thin air. On this record, the jury had ample room to find that Samia's age-related statements were total fabrications and, thus, actionable. See St. Amant, 390 U.S. at 732, 88 S.Ct. 1323.


The last category of statements clusters around Samia's comments about Dr. Sindi's inflation of her resumé through apocryphal boasts that she was involved in founding Diagnostics for All (DFA). Some background facts help to put these comments in perspective.

DFA was created to disseminate affordable diagnostic tools developed in the laboratory of a Harvard professor, Dr. George Whitesides, for use in third-world countries. The effort was widely acclaimed, and DFA won a $100,000 prize in an MIT entrepreneurship competition. Dr. Sindi was a visiting fellow in Dr. Whitesides' laboratory at the time DFA took shape, and she frequently touted her role in its creation. At times, she described herself as a cofounder and/or coinventor.

After a laudatory column regarding Dr. Sindi was published on the Washington Post website on January 18, 2013, Samia posted a comment urging readers to “ask [Dr. Whitesides] about [Dr. Sindi's] non-existent role in the founding of DFA.” Samia proceeded, at various times, to make further statements of this nature alleging in substance that Dr. Sindi had either invented or at least wildly exaggerated the importance of her efforts vis-à-vis DFA.

At the outset, we note that Samia, in disseminating the original statement, urged readers “to [s]peak to Professor Whitesides of Harvard.” Although this statement implies that Samia had herself interviewed Dr. Whitesides prior to commenting, she had never so much as exchanged a word with him. That Samia misrepresented the information gleaned from her sources strongly suggests actual malice. See St. Amant, 390 U.S. at 732, 88 S.Ct. 1323; Levesque, 560 F.3d at 90.

Nevertheless, Samia doggedly insists that these statements were true or, at least, mere hyperbole. She leans heavily on the fact that Dr. Whitesides downplayed Dr. Sindi's role in creating the specific diagnostic tools used by DFA, testifying that he and Dr. Carmichael Roberts were the technology's coinventors. But this emphasis on a single snippet of testimony distorts the picture: Dr. Whitesides made pellucid that, from “the very beginning,” Dr. Sindi was “part of the team” involved in the development of the overall DFA technology. He further testified that Dr. Sindi played an integral role in constructing the business plan for DFA and credited her with helping DFA win the MIT competition. In the same vein, Berlow—an early leader of DFA—lauded Dr. Sindi's important contributions in launching DFA. As Samia's own notes revealed, Berlow told her as much during a conversation in April of 2012. Thus, it is evident that Samia was aware of facts flatly contradicting her statement. Yet, she continued to shout from the rooftops (figuratively speaking) that Dr. Sindi had nothing to do with DFA's success.

Samia's statements, which falsely claimed that Dr. Sindi's role in the DFA endeavor was nonexistent when in fact it was significant, held Dr. Sindi up to public scorn and opprobrium. The statements also characterize Dr. Sindi's truthful claims as lies. Especially in light of the history of acrimony between the two women, the jury was entitled to find that Samia's DFA-related statements about Dr. Sindi were false, defamatory, and made with actual malice.


The same three categories of statements, at a bare minimum, are actionable against Ann. For the most part, Ann simply regurgitated Samia's falsehoods regarding Dr. Sindi's Ph.D., age, and relationship to DFA, authoring a host of derogatory Facebook posts and e-mails to Dr. Sindi's professional associates. As we have shown, see supra Parts III(A)(1)-(3), all of these animadversions were false and defamatory (as were many others memorialized on the chalk but not analyzed in depth here).

This leaves only the question of actual malice. To begin, Ann—as Samia's mother—harbored ill will towards Dr. Sindi. Moreover, she conceded at trial that she had done nothing in the way of serious research to verify Samia's spectacular allegations before broadcasting them wholesale. Significantly, Ann was keenly aware that her daughter was not a neutral source of information: she had full knowledge of Samia's antipathy toward Dr. Sindi. When a speaker relies on a single source notwithstanding the existence of obvious reasons for skepticism about that source's accuracy, a jury may infer actual malice. See St. Amant, 390 U.S. at 732, 88 S.Ct. 1323; Celle v. Filipino Rep. Enters. Inc., 209 F.3d 163, 190 (2d Cir. 2000). So it is here: though Dr. Sindi's defamation claim against Ann is less robust, it is hardy enough to survive independent review.


Represented by new counsel on appeal, Samia and Ann have a fallback position. They assert that the court erred in instructing the jury that a defendant could be held liable as long as that defendant had published at least one defamatory statement with actual malice. In their view, the court should have instructed the jury to specify which of the statements on the chalk were maliciously defamatory and, thus, formed the basis of its verdict. For support, they rely principally on our decision in Levinsky's, in which (as here) the jury returned a general verdict for the defamation plaintiff. See 127 F.3d at 136. We vacated that judgment, explaining that the plaintiff had charged the defendant with making two statements, only one of which we found to be actionable. Consequently, the verdict could not stand because it did not specify the statement on which liability was premised. See id. Extrapolating from this decision and from a similar decision in Simon v. Navon, 71 F.3d 9, 19 (1st Cir. 1995), the appellants argue that we must order a retrial if so much as a single statement displayed on the chalk fails to satisfy the requirements for a defamation claim.

Here, however, there is a rub. Samia and Ann failed to request a jury instruction along these lines in the district court. To compound the problem thus created, they did not object to the instruction about which they now complain prior to jury deliberations. See Fed. R. Civ. P. 51(c)(1) (requiring parties before a case is sent to the jury to “state[ ] distinctly the matter objected to and the grounds for the objection”). Nor did the appellants raise this issue in either their motion for judgment as a matter of law or their motion for a new trial.

Just as actions have consequences, omissions too have consequences. It is black-letter law that claims of instructional error not seasonably advanced in the district court can be broached on appeal only for plain error. See DeCaro v. Hasbro, Inc., 580 F.3d 55, 60 (1st Cir. 2009); Ferrara & DiMercurio v. St. Paul Mercury Ins. Co., 240 F.3d 1, 13 (1st Cir. 2001). To establish plain error, a party must show “(1) that an error occurred (2) which was clear or obvious and which not only (3) affected the defendant's substantial rights, but also (4) seriously impaired the fairness, integrity, or public reputation of judicial proceedings.” United States v. Duarte, 246 F.3d 56, 60 (1st Cir. 2001). The party claiming plain error must carry the devoir of persuasion on all four facets of this test. See United States v. Bramley, 847 F.3d 1, 5 (1st Cir. 2017). Not surprisingly, then, reversals for plain error are “hen's-teeth rare” in civil cases. Teixeira v. Town of Coventry, 882 F.3d 13, 18 (1st Cir. 2018); see Amicas, Inc. v. GMG Health Sys., Ltd., 676 F.3d 227, 235 (1st Cir. 2012).

Samia and Ann cannot clear this high hurdle. Even if we assume, for argument's sake, that some of the roughly 132 statements limned in the chalk are not actionable, the trial focused primarily on the three categories of statements discussed above (that is, false statements pertaining to Dr. Sindi's Ph.D., age, and connection with DFA). Seen in this light, the chances are virtually nil that the jury premised its liability determination on protected speech. See Van Liew v. Eliopoulos, 92 Mass.App.Ct. 114, 84 N.E.3d 898, 913 (2017) (affirming verdict where three of twenty-nine allegedly defamatory statements were non-actionable but were not the focus of trial and did not “add measurably” to plaintiff's injuries). Plain error is plainly absent.

Nothing more need be said. Even if the appellants are correct in suggesting that the jury instructions were infected by an obvious strain of error (a matter on which we take no view), there is Buckley's chance that the verdicts on the defamation claims rested exclusively on any of the few arguably non-defamatory statements. Consequently, the appellants cannot satisfy the third prong of the plain error test. See Bramley, 847 F.3d at 7 (explaining that proponent of plain error must show, at a minimum, a reasonable probability that but for the alleged error, the outcome of the trial would have been different).


The issue of damages remains. Samia and Ann characterize the damages awarded by the jury on the defamation claims ($400,000 against Samia and $100,000 against Ann) as excessive and entreat us to either grant a new trial on damages or to reduce the awards. Their main argument is that the damages are too high because Dr. Sindi offered insufficient evidence of economic loss resulting from their libels.

The court below was tasked with assaying the damages awarded by the jury, and its decision to deny the appellants' motion for a new trial on damages or for a remittitur is reviewed for abuse of discretion. See Trainor, 699 F.3d at 29. We discern none here.

To recover damages, Massachusetts does not require a plaintiff to prove that economic harm resulted from defamatory statements alleging “that the plaintiff lacks a necessary characteristic of [her] profession.” Ravnikar, 782 N.E.2d at 511. In such circumstances, the plaintiff may recover for wholly noneconomic losses, including “impairment of reputation and standing in the community, personal humiliation, and mental anguish and suffering.” Draghetti v. Chmielewski, 416 Mass. 808, 626 N.E.2d 862, 868 (1994).

Samia's and Ann's statements regarding Dr. Sindi's Ph.D. and previous accomplishments impugn Dr. Sindi's scientific aptitude and her professional integrity, which are necessary characteristics of her vocation. Here, moreover, Dr. Sindi introduced evidence of reputational harm flowing from the appellants' defamatory statements, including Berlow's testimony and the testimony of Joi Ito (the director of the MIT Media Lab). She also introduced evidence concerning the humiliation that she experienced as a result of the appellants' campaign of vilification. Given the quantity and quality of this evidence, we hold that the jury's awards of damages for defamation were not so exorbitant as to exceed any reasonable appraisal of damages sustained. Nor were they so extravagant as to shock the conscience. It follows inexorably that the district court's refusal to order either a new trial on damages or a remittitur fit comfortably within the realm of its broad discretion.8


The next leg of our journey takes us to Dr. Sindi's claim for intentional infliction of emotional distress. The jury found Samia liable for this claim and awarded damages against her in the amount of $100,000. At the same time, the jury exonerated Ann on a counterpart claim, and Dr. Sindi has not appealed this finding.

Samia challenges the liability finding, the damages awarded, and the district court's denial of her post-trial motion seeking either to set aside the verdict or to reduce the award. These challenges are unavailing.

Under Massachusetts law, a plaintiff claiming intentional infliction of emotional distress must show that the defendant “intended to inflict emotional distress or that [she] knew or should have known that emotional distress was the likely result of [her] conduct”; that the defendant's “conduct was extreme and outrageous,” such that it transgressed “all possible bounds of decency and was utterly intolerable in a civilized community”; that the conduct caused the plaintiff to suffer emotional distress; and that this distress “was severe and of a nature that no reasonable [person] could be expected to endure it.” Agis v. Howard Johnson Co., 371 Mass. 140, 355 N.E.2d 315, 318–19 (1976) (internal quotation marks omitted). Samia contends that her conduct was not sufficiently extreme or outrageous to come within this framework.

It is common ground that liability for intentional infliction of emotional distress cannot be predicated upon the ordinary vicissitudes that mar human relationships: “mere insults, indignities, threats, annoyances, petty oppressions, or other trivialities” are not enough. Roman v. Trs. of Tufts Coll., 461 Mass. 707, 964 N.E.2d 331, 341 (2012) (quoting Foley v. Polaroid Corp., 400 Mass. 82, 508 N.E.2d 72, 82 (1987) ). But neither a factfinder nor an appellate court is obliged to balkanize the defendant's course of conduct, isolating its component parts and, in the bargain, minimizing their net effect. See Boyle v. Wenk, 378 Mass. 592, 392 N.E.2d 1053, 1055 (1979). “Repeated harassment ․ may compound the outrageousness of incidents which, taken individually, might not be sufficiently extreme to warrant liability for infliction of emotional distress.” Id. at 1056. Nor can a defendant demand the benefit of every conceivable doubt. Rather, a jury is “entitled to put as harsh a face on the actions of the [defendant] as the basic facts would reasonably allow.” Richey v. Am. Auto. Ass'n, Inc., 380 Mass. 835, 406 N.E.2d 675, 678 (1980).

In the case at hand, the evidence, taken in the light most favorable to Dr. Sindi, shows beyond hope of contradiction that Samia transmitted a series of vicious and extraordinarily disturbing e-mails and text messages to Dr. Sindi. By way of illustration, these missives included a December 17, 2011, e-mail expressing thanks that Dr. Sindi's deceased father was not “alive to witness the truth about his sinful, selfish, coniving [sic] Munafika [an Arabic word for hypocrite] of a daughter” as well as a series of text messages referring to Dr. Sindi as “Hoota [an Arabic word for little whale] the Sinful Liar,” predicting that Dr. Sindi would “get cancer” because of “the number of people praying against [her],” declaring that Dr. Sindi would be “exposed” as a “hypocrite & fraud,” and denigrating Dr. Sindi's appearance. After Dr. Sindi blocked Samia from her telephone in late 2011, Samia began to travel from her Seattle home to conferences around the globe where Dr. Sindi was scheduled to speak, handing out leaflets containing a demeaning image of Dr. Sindi and urging conference-goers to visit a blog dedicated to besmirching Dr. Sindi's reputation. Samia even called upon Dr. Sindi's disabled mother in Saudi Arabia for the purpose of confronting her about her daughter's misbehavior.

Given these and other incidents, and the more than four-year long war of vituperation waged by Samia against Dr. Sindi, we think that the jury supportably could have concluded that Samia's course of conduct amounted to far more than mere insults, indignities, and petty oppression. So, too, the jury could supportably have concluded that Samia, over a long period of time, displayed a strain of deliberate malevolence that easily qualified as extreme and outrageous conduct. See Conway v. Smerling, 37 Mass.App.Ct. 1, 635 N.E.2d 268, 273 (1994).

Samia next contends that Dr. Sindi failed to prove that her emotional distress was severe. In evaluating this contention, we recognize that Massachusetts law sets a high bar for proof of severity. See Kennedy v. Town of Billerica, 617 F.3d 520, 530 (1st Cir. 2010) (noting that “mere ‘emotional responses including anger, sadness, anxiety, and distress’ ․ are ‘often not legally compensable’ ” (quoting Quinn v. Walsh, 49 Mass.App.Ct. 696, 732 N.E.2d 330, 338 (2000) ) ). But the length of time that a plaintiff is forced to endure emotional distress is a highly relevant datum in determining whether that distress is sufficiently severe to be compensable. See Homesavers Council of Greenfield Gardens, Inc. v. Sanchez, 70 Mass.App.Ct. 453, 874 N.E.2d 497, 504 (2007); Brown v. Nutter, McClennen & Fish, 45 Mass.App.Ct. 212, 696 N.E.2d 953, 957–58 (1998). One more wrinkle is worth noting: emotional distress may be deemed severe even if it does not produce any physical manifestations. See Cady v. Marcella, 49 Mass.App.Ct. 334, 729 N.E.2d 1125, 1131 (2000) (citing Nancy P. v. D'Amato, 401 Mass. 516, 517 N.E.2d 824, 827 (1988) ).

Here, the relentless nature of Samia's pernicious attacks and the duration of her onslaught weigh heavily in favor of a finding of severity. Dr. Sindi testified that—beginning in late 2011 and continuing up to the time of trial—she suffered great anguish as a result of Samia's harassment. That anguish manifested itself in divers ways including lost sleep, blinding headaches, heart palpitations, and fears for her safety. This constellation of symptoms limited her ability to function.9 On this record, the jury reasonably could have concluded that Dr. Sindi's emotional distress was sufficiently severe to justify recovery.

Samia counters that the verdict must nonetheless be overturned because Dr. Sindi failed to introduce any medical testimony in support of her claim. She is wrong: Massachusetts law allows recovery in emotional distress cases based exclusively on lay testimony. See, e.g., Poy v. Boutselis, 352 F.3d 479, 485–86 (1st Cir. 2003) (applying Massachusetts law).

Battling on, Samia asserts that the evidence was insufficient to establish causation. In this regard, she emphasizes evidence indicating that Dr. Sindi had been treated for stress-related conditions prior to 2011. This assertion is fruitless: “[c]ausation is a factbound issue and, as such, is normally left to the trier.” Limone v. United States, 579 F.3d 79, 99 (1st Cir. 2009) (applying Massachusetts law). This case falls within the general rule, not within the long-odds exception to it. For one thing, there was proof of causation-in-fact: given the duration and persistence of Samia's attacks, the jury had ample reason to infer that her conduct caused Dr. Sindi's emotional distress. See Cady, 729 N.E.2d at 1132. For another thing, the record supports the jury's determination that Dr. Sindi's emotional distress was the foreseeable result of Samia's years-long pattern of vilification, thus establishing proximate cause. See Limone, 579 F.3d at 100.

That is game, set, and match. Beyond her allegations that Dr. Sindi's harm was not severe and that no causal connection was sufficiently proven, Samia makes no developed argument that the damages awarded on this claim are excessive. Consequently, we treat any such argument as waived. See United States v. Zannino, 895 F.2d 1, 17 (1st Cir. 1990). We therefore conclude that Samia, in mounting her challenge to the jury verdict on the intentional infliction of emotional distress claim, is swinging an unstrung racquet.


The jury found both Samia and Ann liable for tortious interference with contract and awarded Dr. Sindi jackpot verdicts: $2,000,000 against Samia and $400,000 against Ann. On post-trial motions, the district court reduced these awards to $576,000 against Samia and $144,000 against Ann. Dr. Sindi does not take issue with the reduction of the awards. Samia and Ann, though, challenge the sufficiency of the evidence supporting the liability findings and also claim that even the reduced damages amounts are excessive.

To prevail on a claim for tortious interference with contract, a plaintiff must prove that she “had a contract with a third party,” which the defendant “knowingly induced the third party to break.” Abramian v. President & Fellows of Harvard Coll., 432 Mass. 107, 731 N.E.2d 1075, 1088 (2000). The plaintiff also must prove that this interference “was improper in motive or means” and caused her harm. Id. For this purpose, “improper means” may include the commission of certain common-law torts, such as defamation. Cavicchi v. Koski, 67 Mass.App.Ct. 654, 855 N.E.2d 1137, 1142 (2006). Relatedly, proof of malice directed toward the plaintiff may serve to establish an improper motive. See id.

Samia and Ann do not seriously contest the majority of these elements. They acknowledge that Dr. Sindi had an employment contract with the i2 Institute, which entitled her to a $10,000 monthly salary. Given what we already have said, the jury had more than enough evidence to find that the appellants' interference with this contract was deliberate—for example, they e-mailed a stream of defamatory statements about Dr. Sindi to board members and investors of the i2 Institute—and that the appellants, sparked by improper motives, employed improper means.

Mindful of these damning facts, Samia and Ann train their fire on the issue of causation. They point out that Dr. Sindi had difficulty in recruiting investors for the i2 Institute even before they began their avalanche of vituperation in 2012, and they suggest that the Institute would have struggled quite apart from their meddling. They also suggest that Dr. Sindi stripped the i2 Institute of financial resources by mismanaging its affairs and insisting that it pay some of her legal expenses.

These suggestions lack force. In the present posture of the case, we are required to weigh the facts in favor of the verdicts, and we have no authority to set those verdicts aside merely because some evidence in the record cuts the other way. See Sanchez, 37 F.3d at 716. Moreover, our deference to jury verdicts, great in any event, is magnified where, as here, the attack on the verdicts relates to causation (which is a matter “peculiarly within the competence of[ ] the factfinder”). Peckham v. Cont'l Cas. Ins. Co., 895 F.2d 830, 837 (1st Cir. 1990) (applying Massachusetts law and quoting Swift v. United States, 866 F.2d 507, 510 (1st Cir. 1989) ). In this instance, there was more than enough evidence to ground a reasonable inference that the appellants' defamatory statements drove supporters away from the i2 Institute and thus caused its financial woes.10

Samia and Ann also argue that the damages awards, even as reduced by the district court, are excessive. Their principal point is that the awards should be further reduced to reflect the i2 Institute's payment of certain of Dr. Sindi's legal bills.

This argument will not wash. While the appellants introduced evidence that, in 2014, the i2 Institute paid 73,125 Saudi Riyals (approximately $20,000 at the time) to cover certain of Dr. Sindi's legal expenses, the appellants cited this evidence to the district court in support of their requests for remittiturs. We have no reason to believe that the district court did not take this payment into account when it granted those remittiturs. When (as in this case) the district court has granted a remittitur, the scope of judicial review—narrow in any event—becomes even narrower. See Wagenmann v. Adams, 829 F.2d 196, 215 (1st Cir. 1987). After all, a challenge for excessiveness to an already trimmed jury award requires an appellate court “not merely to grade the essay, but to grade the teacher's grading of the essay.” Id. The evidence showed that Dr. Sindi was not paid her $10,000 monthly salary for at least three years and was never reimbursed for certain i2 Institute expenses that she paid out of her own pocket. And as the district court observed, the evidence supported a reasonable inference that Dr. Sindi's “contract with i2 would have continued for a number of years,” thus entitling her to future lost earnings. Sindi, 2016 WL 5867403, at *6.

In setting the remittitur amounts, the district court found that the evidence warranted recovery for Dr. Sindi's past lost earnings from her employment with the i2 Institute (totaling $360,000), payment of certain out-of-pocket expenses associated with that employment (totaling roughly $70,000), and her future lost earnings from the Institute (totaling roughly $290,000). The court then apportioned the damages to reflect the jury's finding that Samia was responsible for approximately 80% of Dr. Sindi's losses. Giving this reasoning due weight, the awards as remitted are nowhere near “so extravagant as to shock the appellate conscience.” Sanchez, 37 F.3d at 724.


Samia and Ann next challenge the adverse jury verdicts on Dr. Sindi's claim for tortious interference with advantageous relations. To prevail on such a claim, a plaintiff must show that she had “a present or prospective contract or employment relationship,” that “the defendant knowingly induced a breaking of the relationship,” and that such interference “was improper in motive or means” and caused her harm. Blackstone v. Cashman, 448 Mass. 255, 860 N.E.2d 7, 12–13 (2007). Although the plaintiff need not prove the loss or diminution of a fully formed contract, she must, at a bare minimum, prove harm to a “probable future business relationship from which there is a reasonable expectancy of financial benefit ․” Owen v. Williams, 322 Mass. 356, 77 N.E.2d 318, 322 (1948); see Singh v. Blue Cross/Blue Shield of Mass., Inc., 308 F.3d 25, 48 (1st Cir. 2002) (applying Massachusetts law).

Mere speculation regarding potential future business opportunities is insufficient to prove this element. See Singh, 308 F.3d at 48. Rather, there must be competent evidence of a specific business relationship, the consummation of which was reasonably likely. See id.; see also Am. Private Line Servs., Inc. v. E. Microwave, Inc., 980 F.2d 33, 36 (1st Cir. 1992) (applying Massachusetts law and holding that plaintiff may prevail by showing that she was engaged in promising contract negotiations that were knowingly disrupted by defendant's tortious interference).

Samia and Ann maintain that the evidence on this claim was so sparse that the district court was obliged to grant their motions for judgment as a matter of law. In their view, Dr. Sindi failed to offer probative evidence of a reasonably likely relationship between herself and any identified third party with which they knowingly interfered. We test this premise against the record.

To be sure, Dr. Sindi testified that certain potential business partners ceased communicating with her after Samia and Ann began disseminating their libelous statements. Dr. Sindi failed, however, to introduce any competent evidence concerning the content of her negotiations with these third parties, the details of any potential arrangement, or the likelihood that (absent tortious interference) such a relationship would come to pass. When all is said and done, her claim of tortious interference with advantageous relations is woven entirely out of gossamer strands of speculation and surmise. It follows that Dr. Sindi's professed expectancy of financial benefits from these wholly conjectural relationships was little more than wishful thinking. Certainly, any such expectancy was not objectively reasonable. See Singh, 308 F.3d at 48.

There is a further flaw in Dr. Sindi's argument. A plaintiff who sues for tortious interference with an advantageous relationship must prove not only that the defendant interfered with that relationship but also that the defendant did so knowing of the existence of the relationship. See Bennett v. Saint-Gobain Corp., 507 F.3d 23, 33 (1st Cir. 2007) (applying Massachusetts law). Dr. Sindi has not pointed to a shred of evidence showing that either Samia or Ann was aware of her discussions with any of the third parties alluded to in her testimony. Because any such prospective business relationships were unknown to the appellants, they cannot form the basis for a finding of tortious-interference liability. See id.; Comey v. Hill, 387 Mass. 11, 438 N.E.2d 811, 816 (1982).

Dr. Sindi has a fallback position. She posits that the verdicts on this count can be sustained on the basis that Samia and Ann knowingly interfered with her relationship with the i2 Institute and, thus, with her expectancy of future financial benefits from that relationship. The district court seized upon this rationale: in upholding the jury verdicts on this count ($400,000 against Samia and $100,000 against Ann), the court theorized that Dr. Sindi had proven an expectancy of future lost earnings from the i2 Institute. See Sindi, 2016 WL 5867403, at *6 & n.4.

In the circumstances of this case, the district court's rationale is untenable. It is black-letter law that a plaintiff's recovery under one tort theory precludes her from “duplicative recovery for the same damages under some other tort theory.” Borden v. Paul Revere Life Ins. Co., 935 F.2d 370, 383 (1st Cir. 1991); accord Calimlim v. Foreign Car Ctr., Inc., 392 Mass. 228, 467 N.E.2d 443, 448 (1984). This salutary principle ensures that a plaintiff injured as a result of the defendant's tortious conduct is made whole, but is not made more than whole. See Dopp v. HTP Corp., 947 F.2d 506, 517 (1st Cir. 1991); Szalla v. Locke, 421 Mass. 448, 657 N.E.2d 1267, 1271 (1995).

That principle is pertinent here. Dr. Sindi prevailed against Samia and Ann on her claim for tortious interference with contract. See supra Part V. The damage awards on that count, post-remittitur, encompassed all of the damages flowing from the appellants' interference with Dr. Sindi's relationship with the i2 Institute (past, present, and prospective). Indeed, in ordering remittiturs for tortious interference with contract and capping the recoverable amounts at a total of $720,000, the district court made pellucid that these awards included Dr. Sindi's lost earnings from the i2 Institute both for the period between 2013 and 2015 and for future years (in which her contract ostensibly would have continued but for the appellants' interference). See Sindi, 2016 WL 5867403, at *6.

That ends this aspect of the matter. Massachusetts law, as we understand it, will not countenance allowing a plaintiff to salvage a tort claim by double-counting. Damages already recovered on one theory cannot be recovered again on another theory. See Fox v. F & J Gattozzi Corp., 41 Mass.App.Ct. 581, 672 N.E.2d 547, 552 (1996); see also United States v. Poole, 545 F.3d 916, 920 (10th Cir. 2008) (Gorsuch, J.).

We summarize succinctly. Dr. Sindi's proof on her claim for tortious interference with advantageous relations is deficient in major respects. Most notably, she has failed to prove that she had a reasonable expectancy of financial benefit from a potential third-party relationship (other than her relationship with the i2 Institute), with which Samia and/or Ann knowingly interfered. We therefore reverse the judgments on this count.


This brings us to the pièce de résistance: the district court's post-trial grant of a permanent injunction. We set the stage.

Although the jury found Samia and Ann liable for defamation, see supra Part III, it returned only general verdicts on those claims and did not identify any specific statements as defamatory. During the post-trial proceedings, Dr. Sindi moved for the entry of a permanent injunction barring Samia and Ann from republishing, in any medium and in any context, a compendium of statements.

Based on the evidence adduced at trial, the district court made some further findings of fact. First, the court found that six specific statements were false, defamatory, and made with actual malice and that, absent an injunction, the appellants were likely to repeat them. The court further stated (albeit without making any meaningful findings) that Dr. Sindi had shown that she faced the prospect of irreparable harm. Finally, the court concluded that the balance of harms favored the issuance of an injunction and that the public interest would not be threatened by a grant of injunctive relief. Based on those determinations, the court entered an order broadly enjoining Samia and Ann from republishing the six statements in any medium or for any purpose. Specifically, the injunction (reprinted as part of Appendix B) enjoined the appellants from “repeating—orally, in writing, through direct electronic communications, or by directing others to websites or blogs reprinting”—any of six particular statements, namely:

1. That Hayat Sindi is an academic and scientific fraud;

2. That Sindi received awards meant for young scholars or other youth by lying about her age;

3. That Sindi was fraudulently awarded her PhD;

4. That Sindi did not conduct the research and writing of her dissertation;

5. That Sindi's dissertation was “ghost researched” and “ghost written”;

6. That Sindi's role in the founding of Diagnostics For All was non-existent, and that Sindi did not head the team of six people that won the MIT Entrepreneurship Competition.

On appeal, Samia and Ann question the district court's authority to issue such an injunction, the breadth of the injunction, the court's supplemental factfinding, and a miscellany of other matters incidental to the grant of injunctive relief. Dr. Sindi submits that the appellants have waived or forfeited certain arguments pertaining to the injunction's validity and enforceability. In addition, she defends the injunction in all its particulars. To sort out these competing claims, we delineate the scope of our appellate review and thereafter turn to the appellants' challenges.


In mounting their attack on the injunction, the appellants rely on conclusory argumentation and, in many respects, fail to develop relevant points. When a party's contentions “lack both coherence and development,” we ordinarily deem them procedurally defaulted. Marek v. Rhode Island, 702 F.3d 650, 655 (1st Cir. 2012) (citing Zannino, 895 F.2d at 17). This principle, sometimes inexactly called the “raise-or-waive rule,” is “founded upon important considerations of fairness, judicial economy, and practical wisdom.” Nat'l Ass'n of Soc. Workers v. Harwood, 69 F.3d 622, 627 (1st Cir. 1995). It is not to be taken lightly. In the end, though, “[r]ules of practice and procedure are devised to promote the ends of justice, not to defeat them.” Hormel v. Helvering, 312 U.S. 552, 557, 61 S.Ct. 719, 85 L.Ed. 1037 (1941). Since the application of the so-called raise-or-waive principle is discretionary and non-jurisdictional, an appellate court may, under exceptional circumstances, elect to reach unpreserved issues in order to forestall a miscarriage of justice. See Chestnut v. City of Lowell, 305 F.3d 18, 21 (1st Cir. 2002) (en banc) (per curiam); United States v. La Guardia, 902 F.2d 1010, 1013 (1st Cir. 1990).

While recognizing that this exception to the raise-or-waive principle must be applied sparingly and with great circumspection, we have not hesitated to invoke it where the equities of a particular case counsel strongly in favor of such a step. See Nat'l Ass'n of Soc. Workers, 69 F.3d at 627. In assaying those equities, we have given substantial weight to considerations such as whether the inadequately preserved arguments are purely legal, are amenable to resolution without additional factfinding, are susceptible to resolution without causing undue prejudice, are highly convincing, are capable of repetition, and implicate matters of significant public concern. See id. at 627–28. So, too, we have taken into account whether the failure to advance an argument was deliberate or inadvertent. See id.

In the case at hand, the propriety of the challenged injunction turns on purely legal questions. Those questions can be answered without further factfinding and without causing unfair prejudice to any party. Moreover, the critical issues are virtually certain to arise in future defamation cases. See, e.g., McCarthy v. Fuller, 810 F.3d 456 (7th Cir. 2015); Kinney v. Barnes, 443 S.W.3d 87 (Tex. 2014); Balboa Island Vill. Inn, Inc. v. Lemen, 40 Cal.4th 1141, 57 Cal.Rptr.3d 320, 156 P.3d 339 (2007). To cinch matters, the arguments against allowing the injunction to stand are quite persuasive; those arguments touch upon matters of significant public concern; and the appellants' failure to develop them was apparently careless rather than deliberate. These factors counsel strongly against a mechanical application of the raise-or-waive principle. See Gencarelli v. UPS Capital Bus. Credit, 501 F.3d 1, 8 (1st Cir. 2007).

Our dissenting brother questions this conclusion, noting that the Supreme Court has never directly addressed the constitutionality of a post-trial injunction involving a previously defamed public figure. See post at 42–43. He seems to suggest that the absence of a Supreme Court opinion directly on point somehow militates against considering the appellants' defaulted arguments. This suggestion overlooks that the answer to a legal question may be clear even without a precedent on all fours. Cf. United States v. Morales, 801 F.3d 1, 10 (1st Cir. 2015) (stating that a court may plainly err, even in the “absence of a decision directly on point”). And in any event, the constitutional question that we confront is virtually certain to be litigated in future cases—a factor that weighs in favor of reaching the merits. See La Guardia, 902 F.2d at 1013.

The dissent also suggests that the appellants' failure to develop certain arguments against the legality of the permanent injunction was deliberate rather than inadvertent. See post at 41–42. We do not agree. Although the appellants were admittedly careless in framing their objections, they never expressly abandoned arguments such as the patent failure of the injunction to satisfy strict scrutiny; they simply overlooked these objections while challenging the injunction on other grounds. This was not good lawyering—but a lawyer's failure to articulate an argument does not amount to a deliberate abandonment of that argument. See United States v. Ortiz, 741 F.3d 288, 293 (1st Cir. 2014) (finding forfeiture, not waiver, when appellant failed “to articulate his best argument” and left the trial court “in the dark as to that argument”).

Nor is there any real risk of unfair surprise. Both in her initial brief and in her oral presentation to this court, Dr. Sindi anticipated virtually all of the arguments against the injunction and attempted to explain why those arguments lacked merit. In addition, she has had the opportunity in her supplemental briefing to address our concerns about the injunction. Since Dr. Sindi has fully availed herself of the chance to expound upon whatever legal arguments she may wish to pursue, no cognizable prejudice would flow from excusing the appellants' procedural default. See Singleton v. Wulff, 428 U.S. 106, 120, 96 S.Ct. 2868, 49 L.Ed.2d 826 (1976).

The fact that the appellants are challenging an injunction is itself a factor that cuts in favor of relaxing strict rules of preclusion and considering inadequately preserved arguments. After all, it is well-settled that, upon due notice, a court may dissolve an injunction sua sponte (even in the absence of objections from the party enjoined) when the injunction is no longer equitable or consistent with the public interest. See Moore v. Tangipahoa Par. Sch. Bd., 864 F.3d 401, 407 (5th Cir. 2017); Armstrong v. Brown, 768 F.3d 975, 980 (9th Cir. 2014). Because an injunction is “an extraordinary remedy never awarded as of right,” Winter v. Nat. Res. Def. Council, 555 U.S. 7, 24, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008); see Weinberger v. Romero-Barcelo, 456 U.S. 305, 311–12, 329, 102 S.Ct. 1798, 72 L.Ed.2d 91 (1982), no one can expect that the terms of an injunction will persist in perpetuity. Indeed, any such expectation would be inconsistent with the verity that courts have the “continuing duty and responsibility to assess” an injunction's “efficacy and consequences.” Brown v. Plata, 563 U.S. 493, 542, 131 S.Ct. 1910, 179 L.Ed.2d 969 (2011). Consistent with this imperative, courts have excused procedural defaults and grappled with arguments against injunctions that implicate issues of “constitutional magnitude,” even when those arguments were unpreserved. Real Estate Bar Ass'n for Mass., Inc. v. Nat'l Real Estate Info. Servs., 608 F.3d 110, 125–26 (1st Cir. 2010); see Schlesinger v. Councilman, 420 U.S. 738, 740, 743, 95 S.Ct. 1300, 43 L.Ed.2d 591 (1975) (considering unpreserved arguments against injunction that touched upon “proper relationship between the military justice system” and Article III courts); Younger v. Harris, 401 U.S. 37, 40–41, 46, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971) (vacating injunction that violated “fundamental policy against federal interference with state criminal prosecutions,” notwithstanding petitioners' failure to raise argument in opening submissions).

The challenged injunction falls squarely into this category of cases. The omitted arguments implicate a court's limited authority, consistent with its equitable jurisdiction and the First Amendment, to enjoin speech. This is an area of considerable constitutional concern, and one that has major institutional implications for the federal judiciary. Moreover, our ongoing duty to review the efficacy and consequences of an injunction takes on special importance in the First Amendment context: because such an injunction carries significant “risks of censorship and discriminatory application,” the Supreme Court has directed judges to scrutinize injunctions restricting speech carefully and ensure that they are “no broader than necessary to achieve [their] desired goals.” Madsen v. Women's Health Ctr., 512 U.S. 753, 764–65, 114 S.Ct. 2516, 129 L.Ed.2d 593 (1994).

The bottom line is that this case calls for an exception to the usual rule: it arrives on our doorstep in a posture that allows us, in the exercise of our discretion, to consider inadequately preserved arguments against the challenged injunction. Given the special importance of the issues surrounding the injunction and the other factors that we have mentioned, we conclude that a mechanical application of the raise-or-waive principle would work a miscarriage of justice. Under these exceptional circumstances, we look past the infirmities in the appellants' briefing and proceed to consider all the available arguments affecting the validity and enforceability of the injunction, regardless of whether some of those arguments may have been forfeited.


As a general matter, the First Amendment forbids the government, including the Judicial Branch, “from dictating what we see or read or speak or hear.” Ashcroft v. Free Speech Coal., 535 U.S. 234, 245, 122 S.Ct. 1389, 152 L.Ed.2d 403 (2002). The question that remains in this case is whether the district court offended the First Amendment by enjoining the appellants from republishing, orally or in writing, any of six statements that they previously had employed to defame Dr. Sindi. Some courts have adopted the view that an injunction against future speech following a defamation trial may be consistent with the First Amendment. See, e.g., Lothschuetz v. Carpenter, 898 F.2d 1200, 1208–09 (6th Cir. 1990) (Wellford, J., for the court in part); Lemen, 57 Cal.Rptr.3d 320, 156 P.3d at 349. Others, though, have expressed deep skepticism, suggesting that such a remedy is per se unconstitutional. See, e.g., Fuller, 810 F.3d at 464–66 (Sykes, J., concurring); Kinney, 443 S.W.3d at 89, 94; see also Erwin Chemerinsky, Injunctions in Defamation Cases, 57 Syracuse L. Rev. 157, 158 (2007). Although the Supreme Court once granted certiorari to resolve this conundrum, it disposed of the case on less controversial grounds, leaving the constitutional question open. See Tory v. Cochran, 544 U.S. 734, 737–38, 125 S.Ct. 2108, 161 L.Ed.2d 1042 (2005).

We need not decide today the broader question of whether the First Amendment will ever tolerate an injunction as a remedy for defamation. In all events, “courts should not rush to decide unsettled issues when the exigencies of a particular case do not require such definitive measures,” Privitera v. Curran (In re Curran), 855 F.3d 19, 22 (1st Cir. 2017)—and this is such a case. Consistent with our prudential practice of forgoing broad constitutional holdings unless such holdings are unavoidable, see Hudson Sav. Bank v. Austin 479 F.3d 102, 106 (1st Cir. 2007); El Dia, Inc. v. Hernandez Colon, 963 F.2d 488, 494 (1st Cir. 1992), we decide the issues concerning the validity and enforceability of the challenged injunction on narrower grounds. The injunction cannot survive the strict scrutiny required to legitimize a prior restraint, principally because of its failure to account for contextual variation. Therefore, the injunction must be vacated.11

We start this phase of our analysis by rehearsing abecedarian principles of equity. A court may not issue a permanent injunction unless, among other things, “remedies available at law, such as monetary damages, are inadequate to compensate for” an “irreparable injury.” eBay, Inc. v. MercExchange, L.L.C., 547 U.S. 388, 391, 126 S.Ct. 1837, 164 L.Ed.2d 641 (2006).12 Moreover, such an injunction must be “no more burdensome to the defendant than necessary to provide complete relief to the plaintiffs.” Madsen, 512 U.S. at 765, 114 S.Ct. 2516 (quoting Califano v. Yamasaki, 442 U.S. 682, 702, 99 S.Ct. 2545, 61 L.Ed.2d 176 (1979) ). Although we review the issuance of a permanent injunction for abuse of discretion, see eBay, 547 U.S. at 391, 126 S.Ct. 1837, we perform this task mindful of our unflagging “obligation to ‘make an independent examination of the whole record’ in order to make sure that ‘the judgment does not constitute a forbidden intrusion on the field of free expression,’ ” Bose Corp., 466 U.S. at 499, 104 S.Ct. 1949 (quoting N.Y. Times Co., 376 U.S. at 284–86, 84 S.Ct. 710); accord Metro. Opera Ass'n, v. Local 100, Hotel Emps. & Rest. Emps. Int'l Union, 239 F.3d 172, 176 (2d Cir. 2001).

The injunction issued in this case, which prohibits the appellants from republishing six particular statements, is a paradigmatic example of a prior restraint: it is a “judicial order[ ] forbidding certain communications ․ issued in advance of the time that such communications are to occur.” Alexander v. United States, 509 U.S. 544, 550, 113 S.Ct. 2766, 125 L.Ed.2d 441 (1993) (emphasis in original) (citation omitted). As such, it is subject to even more exacting requirements under settled First Amendment doctrine.13 See Tory, 544 U.S. at 738, 125 S.Ct. 2108 (treating post-trial injunction against republication of previously defamatory statements as prior restraint).

There is a strong presumption that prior restraints on speech are unconstitutional. See N.Y. Times Co. v. United States, 403 U.S. 713, 714, 91 S.Ct. 2140, 29 L.Ed.2d 822 (1971) (per curiam). So drastic a remedial device may only be imposed when it furthers “the essential needs of the public order.” Carroll v. President & Comm'rs of Princess Anne, 393 U.S. 175, 183, 89 S.Ct. 347, 21 L.Ed.2d 325 (1968). A prior restraint cannot be imposed when those needs can be achieved through less restrictive means. See id. at 183–84, 89 S.Ct. 347; see also Tory, 544 U.S. at 738, 125 S.Ct. 2108. And even when a prior restraint may theoretically be permissible, the decree that embodies it must be precisely tailored both to meet the exigencies of the particular case and to avoid censoring protected speech. See Carroll, 393 U.S. at 183–84, 89 S.Ct. 347. In the last analysis, a party who seeks a remedy in the form of a prior restraint must establish that the “evil that would result from” the offending publication is “both great and certain and cannot be mitigated by less intrusive measures.” CBS, Inc. v. Davis, 510 U.S. 1315, 1317, 114 S.Ct. 912, 127 L.Ed.2d 358 (1994) (Blackmun, J., in chambers) (citing Neb. Press Ass'n v. Stuart, 427 U.S. 539, 562, 96 S.Ct. 2791, 49 L.Ed.2d 683 (1976) ); see In re Goode, 821 F.3d 553, 559 (5th Cir. 2016); Cty. Sec. Agency v. Ohio Dep't of Commerce, 296 F.3d 477, 485 (6th Cir. 2002); Levine v. U.S. Dist. Ct., 764 F.2d 590, 595 (9th Cir. 1985). Consequently, a prior restraint on speech must survive the most exacting scrutiny demanded by our First Amendment jurisprudence. See Stuart, 427 U.S. at 559, 96 S.Ct. 2791.

Such intensive scrutiny is warranted because an animating purpose of the First Amendment was to create a bulwark against previous restraints upon speech. See Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713, 51 S.Ct. 625, 75 L.Ed. 1357 (1931). Since “the line between legitimate and illegitimate speech is often so finely drawn,” we “prefer[ ] to punish the few who abuse rights of speech after they break the law than to throttle them and all others beforehand.” Se. Promotions, Ltd. v. Conrad, 420 U.S. 546, 559, 95 S.Ct. 1239, 43 L.Ed.2d 448 (1975) (emphasis in original). Thus, prior restraints are regarded as “the most serious and the least tolerable infringement on First Amendment rights.” Stuart, 427 U.S. at 559, 96 S.Ct. 2791.

The operation of the collateral bar rule compounds the grave perils posed by prior restraints. This rule requires that an injunction be followed upon pain of contempt until modified or vacated, and the unconstitutionality of the injunction typically does not justify a refusal to obey it. See Metro. Opera Ass'n, 239 F.3d at 176 (citing Walker v. Birmingham, 388 U.S. 307, 314–21, 87 S.Ct. 1824, 18 L.Ed.2d 1210 (1967) ). It follows that once an injunction in the nature of a prior restraint issues, the harm is “immediate and irreversible.” Stuart, 427 U.S. at 559, 96 S.Ct. 2791.

In this case, Dr. Sindi argues that the challenged injunction comports with the First Amendment because the six statements were previously employed to defame her and, thus, no longer constitute protected speech. This argument has some superficial appeal: an injunction against speech sometimes may pass constitutional testing if it follows an adjudication that the expression is unprotected, and the injunction itself is narrowly tailored to avoid censoring protected speech. See Pittsburgh Press Co. v. Pittsburgh Comm'n on Human Relations, 413 U.S. 376, 390, 93 S.Ct. 2553, 37 L.Ed.2d 669 (1973); Paris Adult Theatre I v. Slaton, 413 U.S. 49, 55, 93 S.Ct. 2628, 37 L.Ed.2d 446 (1973). For instance, the Supreme Court has approved a permanent injunction against the distribution of specific booklets “found after due trial to be obscene,” where the injunction did not extend to “matters not already published and not yet found to be offensive.” Kingsley Books, Inc. v. Brown, 354 U.S. 436, 437, 445, 77 S.Ct. 1325, 1 L.Ed.2d 1469 (1957); cf. Auburn Police Union v. Carpenter, 8 F.3d 886, 903 (1st Cir. 1993) (setting forth similar proposition in dictum). The analogy that Dr. Sindi draws to Kingsley Books is tempting because (in the idiom of the First Amendment) obscenity—like defamation—is a category of unprotected speech. See Free Speech Coal., 535 U.S. at 245–46, 122 S.Ct. 1389.

In the end, though, Dr. Sindi's proffered analogy glosses over significant distinctions between obscenity and defamation that make injunctions of obscene communications less problematic in constitutional terms. The obscenity doctrine proscribes specific expressive works (such as books or movies) that appeal to prurient interests, depict sexual behaviors in patently offensive ways, and lack “serious literary, artistic, political, or scientific value.” Miller v. California, 413 U.S. 15, 24, 93 S.Ct. 2607, 37 L.Ed.2d 419 (1973). Works adjudged obscene—such as the booklets in Kingsley Books—are immutable forms of expression. Hence, the permanent injunction there could be carefully crafted to ensure that it applied only to the specific publications found obscene without exposing the bookseller to contempt sanctions for distributing other publications that might be protected under the First Amendment. See Kingsley Books, 354 U.S. at 445, 77 S.Ct. 1325.

An injunction that prevents in perpetuity the utterance of particular words and phrases after a defamation trial is quite a different matter. By its very nature, defamation is an inherently contextual tort. See Greenbelt Coop. Publ'g Ass'n v. Bresler, 398 U.S. 6, 13–14, 90 S.Ct. 1537, 26 L.Ed.2d 6 (1970); Piccone v. Bartels, 785 F.3d 766, 772 (1st Cir. 2015); cf. United States v. Alvarez, 567 U.S. 709, 719, 132 S.Ct. 2537, 183 L.Ed.2d 574 (2012) (plurality opinion) (noting that defamation entails not merely a “false statement,” but a “legally cognizable harm associated with a false statement”). Words that were false and spoken with actual malice on one occasion might be true on a different occasion or might be spoken without actual malice. What is more, language that may subject a person to scorn, hatred, ridicule, or contempt in one setting may have a materially different effect in some other setting.14 Cf. Pittsburgh Press, 413 U.S. at 390, 93 S.Ct. 2553 (sustaining injunction where court was not required “to speculate as to the effect of publication”).

The cardinal vice of the injunction entered by the district court is its failure to make any allowance for contextual variation. Refined to bare essence, it enjoins Samia and Ann from repeating certain words, regardless of their purpose in employing them. Consequently, the injunction “sweeps ․ more broadly than necessary” by prohibiting the appellants from engaging in speech about a public figure “before an adequate determination that it is unprotected by the First Amendment.” Id.

For instance, the injunction precludes the appellants from restating that Dr. Sindi “is an academic and scientific fraud.” Although the appellants have in the past used those words with actual malice (or so the district court supportably found), there are a number of future contexts in which their repetition of this statement might be protected speech. We offer three examples:

• If, say, Samia or Ann learns in the future of fraud actually perpetrated by Dr. Sindi and accurately reports it, the speaker would face contempt sanctions under the injunction even though the right to disseminate truthful information about public figures lies at the core of the First Amendment. See N.Y. Times Co., 376 U.S. at 270, 84 S.Ct. 710.

• If, say, Samia or Ann were interviewed by a reporter and asked what speech the challenged injunction prevented them from repeating, a reply to the effect that, “I am not allowed to state that Dr. Sindi is an academic and scientific fraud” would subject the speaker to contempt sanctions notwithstanding the truth of the reply.

• Perhaps most remarkably, the appellants would face contempt sanctions for disseminating a letter describing their accusations and apologizing for them.

The list of contextual permutations is virtually endless. The situations that we have described are but a few of the possible examples that show, beyond hope of peradventure, that the challenged injunction is neither narrowly tailored nor precisely fitted to the circumstances of the case.

As framed, the injunction is so wide-ranging and devoid of safeguards that it plainly contravenes the First Amendment's limitation of liability for speech about public figures to false assertions of fact made with actual malice. See Hustler Magazine, Inc. v. Falwell, 485 U.S. 46, 56, 108 S.Ct. 876, 99 L.Ed.2d 41 (1988). We conclude, therefore, that the injunction punishes future conduct that may be constitutionally protected, see Conrad, 420 U.S. at 559, 95 S.Ct. 1239, and thus fails the First Amendment requirement that it be “tailored as precisely as possible to the exact needs of the case,” Carroll, 393 U.S. at 184, 89 S.Ct. 347.

In an effort to blunt the force of this reasoning, our dissenting brother defends the injunction on the ground that, should the appellants choose to republish any of the six statements for a non-defamatory purpose, they may move to modify the injunction in light of changed circumstances. See post at 47–48. To support this defense, he relies on the California Supreme Court's dictum surmising that a defamation defendant's ability to move to modify an injunction alleviates any concern that the injunction may penalize or chill constitutionally protected speech. See Lemen, 57 Cal.Rptr.3d 320, 156 P.3d at 353. But this is little more than a hopeful improvisation: neither our dissenting brother nor the California Supreme Court identifies any other First Amendment precedent supporting this extraordinary proposition. In light of a court's power to levy contempt sanctions (up to and including imprisonment) for disobedience under the collateral bar rule, see Walker, 388 U.S. at 314–21, 87 S.Ct. 1824, “the right to free speech should not lightly be placed within the control of a single man or woman,” Madsen, 512 U.S. at 793, 114 S.Ct. 2516 (Scalia, J., concurring in part and dissenting in part). A decree that requires a judicial permission slip to engage in truthful speech is the epitome of censorship. See Near, 283 U.S. at 713, 51 S.Ct. 625; Kinney, 443 S.W.3d at 98; see also Chemerinsky, supra, at 172. To make a bad situation worse, the appellants would bear the burden of pointing to changed circumstances in any proceeding to modify the injunction. See Horne v. Flores, 557 U.S. 433, 447, 129 S.Ct. 2579, 174 L.Ed.2d 406 (2009). Such a circumstance would be repugnant to the First Amendment, which requires a public-figure plaintiff, not the defendant, to prove actual malice and falsity. See Gertz, 418 U.S. at 342, 94 S.Ct. 2997.

The dissent attempts to analogize this case to Madsen and Schenck v. Pro-Choice Network of Western New York, 519 U.S. 357, 117 S.Ct. 855, 137 L.Ed.2d 1 (1997). See post at 44–47. With respect, this attempted analogy does not work. In those cases, the Supreme Court partially sustained injunctions against protest activities near abortion clinics. The Court concluded that neither injunction was a prior restraint and, therefore, neither was presumptively unconstitutional. See Schenck, 519 U.S. at 374 n.6, 117 S.Ct. 855; Madsen, 512 U.S. at 763 n.2, 766, 114 S.Ct. 2516. The Court's rationale is instructive. It emphasized that the injunctions were content-neutral and left “alternative channels of communication” available to the antiabortion protesters. Schenck, 519 U.S. at 374 n.6, 117 S.Ct. 855. The protesters “remain[ed] free to espouse their message,” so long as they were outside the buffer zone delineated by the injunctions. Id. at 385, 117 S.Ct. 855; see Madsen, 512 U.S. at 763 n.2, 114 S.Ct. 2516.

The injunction here is quite different. As Dr. Sindi acknowledges, it is not content-neutral. This is significant because the Supreme Court has found Madsen inapposite when—as in this case—the defendant was exposed to liability based on “what [it] said.” Snyder v. Phelps, 562 U.S. 443, 457, 131 S.Ct. 1207, 179 L.Ed.2d 172 (2011). What is more, the challenged injunction forbids the appellants from ever republishing the six statements about Dr. Sindi, regardless of the forum or the purpose. As such, it does not leave open alternative channels of communication. Seen in this light, the injunction must withstand strict scrutiny (as Dr. Sindi concedes) and, thus, is presumptively unconstitutional.

When all is said and done, we need not answer the vexing question of whether a federal court may ever permanently enjoin republication of ad hoc oral or written statements on the ground that those statements will be defamatory if made anew. Similarly, we take no view of the legality of an injunction ordering “the removal or deletion of speech that has been adjudicated defamatory,” such as a decree requiring the erasure of a statement from a website after an adjudication that the statement was “unprotected in the context in which it was made.” Kinney, 443 S.W.3d at 89, 93, 99 (upholding such an injunction and explicating the “legally cogent division between mandatory injunctions calling for the removal of speech that has been adjudicated defamatory and prohibitive injunctions disallowing its repetition”).

To say more would be to paint the lily. The First Amendment requires that less intrusive remedies be unavailable before injunctive relief can be considered and that any injunction be as narrowly tailored as possible to avoid censoring protected speech. See Carroll, 393 U.S. at 183–84, 89 S.Ct. 347. Because the challenged injunction cannot conceivably survive this strict scrutiny, it must be vacated.15


We summarize succinctly. The evidence in this case tells a tawdry tale of two women who let their antipathy for a third woman lead them into inexcusable behavior. The jury supportably found that this course of conduct was tortious in several respects, and its assessment of damages on those counts, as refined by the able district judge, passes muster. The post-trial injunction, though, is a bridge too far: it cannot survive the strict scrutiny that the First Amendment demands of prior restraints on speech. Even the bad behavior exhibited by the appellants cannot justify crossing well-established constitutional lines.

We need go no further. For the reasons elucidated above, we affirm the judgment of the district court with respect to the claims of defamation, intentional infliction of emotional distress, and tortious interference with contract. We reverse the judgment with respect to the claim for tortious interference with advantageous relations. Finally, we vacate the post-trial injunction improvidently issued by the district court. All parties shall bear their own costs.

So ordered.

There is no more basic First Amendment principle than that the government may not restrain speech in advance of its expression simply because it may cause offense. We must be cautious, therefore, before we uphold an injunction, like the one before us, that bars the expression of certain specific statements due to the harm that they may cause. At the same time, there are few more basic principles of adjudication than that “if it is not necessary to decide more, it is necessary not to decide more.” PDK Labs. Inc. v. DEA, 362 F.3d 786, 799 (D.C. Cir. 2004) (Roberts, J., concurring). In my view, the majority, by seeking to vindicate the first principle, gives insufficient attention to the second.

The result is that the majority strikes down this injunction by unnecessarily announcing a broad constitutional rule. Under that rule, it would appear that a lower court may not enjoin a recidivist defamer from using particular words, even when he has been properly found to have repeatedly used those same words in the past to defame the party that seeks the injunction and even when he has been found to be likely to do so again absent the injunction.

This result follows from the majority's decision to subject such an injunction to strict First Amendment scrutiny. The majority applies that demanding form of review because it treats such an injunction as a presumptively unconstitutional prior restraint. And the majority then strikes down this injunction under such scrutiny because it fails to require proof of actual defamation in order to show its violation. See Maj. Op. at 30–35.16

In adopting this constitutional rule, the majority makes the following equation. It treats a specific, tailored means of stopping the recurrence of speech that the First Amendment does not protect as if it were a regulation designed to stop the initial expression of protected speech due to the offense that it may cause.

The decision to make this equation fits uncomfortably with our own circuit's precedent. It also conflicts with the only precedents that have decided the issue under the First Amendment. And, finally, it creates tension with Supreme Court rulings that afford lower courts significant discretion to enjoin parties from resuming their unprotected and unlawful expressive conduct.

The majority's rule also gives rise to significant practical concerns. We live in a world in which defamation campaigns may reach millions in an instant and essentially for free. Injunctions crafted in general terms to conform to the majority's rule risk inviting obstinate and proven defamers to resume their defamatory campaigns by wagering that their victims will lack the energy to enforce an injunction that requires them to prove actual defamation all over again.

In light of these concerns, I cannot sign on to the majority's rule, whatever its ultimate merits. And that is because, in my view, there is no need to announce it. The enjoined parties never timely made the debatable federal constitutional arguments on which the majority relies.

Nor can I sign on to the decision to vacate this injunction. The only other argument that the majority suggests could be a ground for vacating it, which challenges the District Court's finding that Sindi would suffer irreparable harm absent this injunction, see id. at 35-36 n.15, also was not properly raised by the defendants either below or on appeal. And the arguments that the defendants did properly present to us in challenging the injunction lack merit.

For these reasons, although I fully join the majority's thorough and persuasive analysis in Parts I through VI of its opinion, I dissent from Part VII.


To explain my concerns, I first review how this injunction came to be. I then describe the limited reach of the grounds for striking it down that the defendants timely made. Finally, I explain that this is not a case in which we should make an exception to our usual “raise-or-waive” requirement. Nat'l Ass'n of Soc. Workers v. Harwood, 69 F.3d 622, 627 (1st Cir. 1995).


The plaintiff, Hayat Sindi, came to federal court to seek relief from the defendants' five-year defamatory campaign. She successfully made her defamation case to a jury, which awarded her a multi-million dollar verdict.

Nevertheless, Sindi was concerned that the defendants would not be deterred. She therefore sought a permanent injunction to prohibit them from making the statements that she alleged they had been making to defame her in the five years preceding her suit. Absent such an injunction, Sindi argued, she would suffer irreparable harm to her “reputation, business dealings, and emotional well-being” because the defendants would pick up where they had left off.

After hearing from the parties, the District Court issued a narrowed injunction that encompassed only six of the twenty-six statements that Sindi initially had sought to enjoin. In doing so, the District Court found that the defendants clearly had used the six statements to defame Sindi, resulting in irreparable harm to her, and that, absent the narrowed injunction, the defendants would likely continue to do so.17

Significantly, the defendants never made a peep—either below or in their opening and reply briefs on appeal—that indicated that they wanted to use the six statements in different contexts from those in which they had used them in the past. The defendants also did not meaningfully dispute—either below or on appeal—that they were likely to repeat the statements in that same way.18

Nor did the defendants argue that Sindi had failed to show that only an injunction, as opposed to a damages award, would be a sufficient remedy for any harm that she would suffer from the defendants' continuing to use the statements as they had. Rather, below, the defendants initially argued that she would not suffer such irreparable harm because their past communication of the statements had not actually harmed Sindi. And, on appeal, the defendants then abandoned even that limited challenge to the finding of irreparable harm that the District Court had ended up making.

The defendants did contend throughout this litigation that the proposed injunction violated the First Amendment. But, they did so by contending only that an injunction that barred the future expression of the six statements could not possibly be a valid prophylactic means of stopping their defamatory conduct going forward because: (1) the jury had returned a general verdict and thus did not expressly find that each of the statements encompassed by the injunction had been made in a defamatory manner in the past and (2) the evidence presented to the jury was, in any event, too weak to have permitted a jury to have so found.

The defendants thus never suggested at any point that strict scrutiny (or even heightened review) applied to the injunction insofar as it was properly predicated on findings that the defendants had engaged in prior defamation through their use of those statements.19 Nor did the defendants argue that such a properly predicated injunction would be an impermissible prior restraint under the First Amendment.

At oral argument on appeal, the defendants' counsel (who was not trial counsel) did attempt to argue for the first time, and despite the defendants' previous assertions to the contrary, the following: The injunction was a prior restraint that violated the First Amendment because it enjoined the defendants from repeating the six statements regardless of the context in which they might be communicated in the future. But, even then the defendants' counsel did not directly contend that strict scrutiny applied. And, when asked if the defendants had made that argument in their briefs on appeal, he conceded: “No, not the contextual argument.”

Thus, based on the only two arguments that the defendants properly presented to us—namely, that the injunction encompassed statements that no jury had found to be defamatory and that, on the record established at trial, no adjudicator could so find—we have no reason to vacate this injunction. As Sindi persuasively shows, and the majority does not dispute, neither argument has merit.

The defendants do not adequately explain why the First Amendment bars the District Court from issuing this injunction simply due to the absence of a special verdict by the jury, given that the District Court supportably found that the defendants defamed Sindi in the past with each of the six enjoined statements and that the defendants would likely continue to do so.20 The defendants also fail to show that the record cannot support the District Court's finding that the defendants had used each enjoined statement to defame Sindi.


Nonetheless, the majority does vacate the injunction. It does so by relying on a ground that the defendants did not properly raise either below or on appeal: that this injunction is a presumptively unconstitutional prior restraint that is subject to strict scrutiny, which it flunks because it enjoins particular statements regardless of the context in which they are used. See Maj. Op. at 30–35.

The majority acknowledges that its decision to rely on this defaulted argument is most unusual. See id. at 27-28. Under our “raise-or-waive” rule, which we ordinarily apply with “a near-religious fervor,” the defendants would have to “forever ․ hold their peace” with respect to that argument. Nat'l Ass'n of Soc. Workers, 69 F.3d at 627 (failure to raise below); accord United States v. Zannino, 895 F.2d 1, 17 (1st Cir. 1990) (failure to raise on appeal). And, the majority recognizes, see Maj. Op. at 27-28, there is good reason to enforce that “raise-or-waive” requirement strictly, as it serves important “systemic ends,” Nat'l Ass'n of Soc. Workers, 69 F.3d at 627, by incentivizing parties to make arguments in a timely way and by ensuring that like cases are treated alike.

The majority nevertheless suggests that because this case involves a regulation of speech crafted by a federal court, there is reason to relax our usual “raise-or-waive” rule. See Maj. Op. at 28–29. But, the majority does not rule that the District Court lacks equitable jurisdiction to impose this remedy. See id. at 35. And I see no reason to encourage parties to assume that, in general, they need not be as diligent in pressing their personal constitutional rights in challenging court-crafted injunctions as we routinely require criminal defendants to be in challenging court-crafted sentences.21

Moreover, to the extent that the majority is inclined to relax our “raise-or-waive” rule in this case, I certainly see no reason to reach the prior restraint/strict scrutiny issue. The majority itself suggests that the injunction could be independently invalidated on the far narrower ground that the record does not support the District Court's finding that the injunction is necessary to protect Sindi from irreparable harm. See id. at 35-36 n.15.

To be sure, that argument, too, was defaulted by the defendants. But, by resting its vacatur solely on the irreparable harm argument, the majority at least would be issuing a relatively narrow, record-dependent ruling, with no broad federal constitutional implications.22

Nor am I persuaded by the majority's conclusion that it makes sense to decide this case on the basis of the forfeited constitutional argument in light of our precedent recognizing that “ ‘an appellate court has discretion, in an exceptional case, to reach virgin issues,’ that is, to relieve a party of a prior forfeiture.” Chestnut v. City of Lowell, 305 F.3d 18, 21 (1st Cir. 2002) (en banc) (per curiam) (quoting United States v. La Guardia, 902 F.2d 1010, 1013 (1st Cir. 1990) ). Under this exception to the “raise-or-waive” rule, we may exercise discretion to decide a case based on a forfeited argument after considering a variety of factors, such as whether the underlying issue is of “constitutional magnitude” and “great public moment”; the party's failure to address it was “entirely inadvertent rather than deliberate”; its proper resolution is sufficiently clear that the lower court can be said to have plainly erred; and deciding it will not result in “special prejudice or inequity” to the non-defaulting party or “deprive[ ] the court of appeals of useful factfinding.” Nat'l Ass'n of Soc. Workers, 69 F.3d at 627–28; accord Chestnut, 305 F.3d at 21.

Here, of course, we are dealing with a failure to properly raise an argument in appellate briefing as well as in the district court. But, insofar as the exception to our “raise-or-waive” rule on which the majority relies applies to such a situation,23 I cannot see why it applies here. The prior restraint/strict scrutiny issue is of “constitutional magnitude” and, at least arguably, of “great public moment.”24 But, the other factors that we have held bear on deciding whether to excuse a forfeiture weigh against doing so.

To begin with, it would be extremely generous to characterize as “entirely inadvertent,” Nat'l Ass'n of Soc. Workers, 69 F.3d at 628 (emphasis added), the defendants' years-long strategy of training their fire solely on the supposedly inadequate predicate finding that the defendants used the six enjoined statements to defame Sindi in the past. Indeed, at the hearing on the proposed injunction, the District Court, quite conscientiously, sought to make sure that the defendants' federal constitutional challenge to Sindi's proposed injunction was as limited as it appeared to be. And, in response, the defendants' counsel made clear that it was: “I think there would not be a prior restraint, your Honor, if there had been a final adjudication as to certain statements” finding that they were defamatory. That counsel also confirmed that same position repeatedly at that same hearing. And he did so without ever suggesting that the injunction, as drafted, might be unconstitutional under the First Amendment if it were properly predicated.

Consistent with those representations, moreover, the defendants also declined the District Court's express invitation to suggest that “the language [of the injunction] should be tweaked one way or another to not create a prior restraint.” And that was the case even though the District Court soon thereafter had, prudently, circulated for comment a narrowed version of the proposed injunction that targeted just six of the twenty-six statements that Sindi initially had sought to enjoin.

It also seems to me that prejudice does result from our willingness to revive this never-before-raised constitutional argument. The parties were given a chance to provide supplemental briefing to address it. But, we have never suggested that the provision of that opportunity is a panacea. And here it is not.

If the defendants had given Sindi some indication below that they actually wished to use the enjoined statements in new contexts, she potentially could have further developed the record regarding just how the defendants did intend to use those statements and why the injunction—in whatever form it would then take in such circumstances—was necessary to prevent the defendants from nevertheless using the statements to defame her. Had that happened, the District Court could have then evaluated that more developed record and either scaled back the injunction in some calibrated manner that might still protect Sindi or issued this same injunction after making findings on the key disputed points concerning the defendants' likely future conduct.

What the case then would have looked like we cannot know, precisely because we are raising these constitutional issues on our own and are thus deprived of that “useful factfinding.” Id. at 627. But, we are not the only ones who lose out by short-circuiting this normal adjudicative process. Because we have transformed what had been a concrete dispute into an abstract one, Sindi finds herself stripped altogether of the protection that she had secured. And she is stripped of it based on a speculative expressive interest that we have assumed the defendants must have, even though the defendants themselves never gave her (or the enjoining court) any indication that they actually do.25

Perhaps the fact that we are deciding this case in this artificial posture does not matter. Perhaps, under the majority's rule, there is no showing that Sindi could make about her proven defamers' likely future conduct that would entitle her to an injunction of this kind. Perhaps, in fact, she would not be able to make such a showing even if the defendants had been found to have been in violation of an earlier injunction that did require Sindi to prove defamation to enforce it.

But, if, as appears, that is what the majority means to hold, then, in my view, it is especially clear that we have no good reason to make an exception to our “raise-or-waive” rule here. For, as I will next explain, such a broad federal constitutional holding hardly rests on a legal conclusion that is so plainly right that it is of the kind that “often inclines a court to entertain a pivotal argument for the first time on appeal.” Id. at 628 (quoting La Guardia, 902 F.2d at 1013).


As the majority recognizes, there is no on-point precedent—from either our court or the Supreme Court—that dictates the federal constitutional rule that it announces. See Maj. Op. at 30. Of course, the absence of such precedent is not conclusive as to whether the rule that the majority adopts is so plainly right that the party that would benefit from its announcement may be excused for having failed to raise the issue properly.

But, here, the problem with finding the law so clear that no argument about it need be timely raised is more fundamental. For, in this case, there is not merely a dearth of controlling supportive precedent, but also substantial (though not controlling) opposing precedent and not a single case of any court that actually holds what the majority now does.


To begin, as the majority acknowledges, there is no controlling Supreme Court precedent that makes clear what the majority holds: that an injunction that bars the expression of certain statements is a presumptively unconstitutional prior restraint under the First Amendment even when it rests on findings that the enjoined party had engaged in prior unprotected, unlawful uses of the enjoined statements and will likely use those statements in that same unprotected and unlawful manner going forward absent the injunction. And the fact that there is no such precedent should give us pause.

This injunction—like any that bars a party from making any statement—does preclude expression before it is expressed. But, we have no reason to conclude that the absence of Supreme Court precedent treating an injunction like this one as a presumptively unconstitutional prior restraint should be chalked up to the fact that the Court simply has not yet gotten around to doing so, because, once it does, the result will be obvious. To the contrary, the Supreme Court has expressly cautioned that “[t]he phrase ‘prior restraint’ is not a self-wielding sword. Nor can it serve as a talismanic test.” Kingsley Books, Inc. v. Brown, 354 U.S. 436, 441, 77 S.Ct. 1325, 1 L.Ed.2d 1469 (1957); see also Madsen, 512 U.S. at 764 n.2, 114 S.Ct. 2516 (explaining that “[n]ot all injunctions that may incidentally affect expression ․ are ‘prior restraints’ in the sense that that term was used in New York Times Co. [v. United States, 403 U.S. 713, 91 S.Ct. 2140, 29 L.Ed.2d 822 (1971) (per curiam) ]”).26

Our own precedent, moreover, has been sensitive to this guidance. That precedent involved a statute that authorized injunctive relief to be ordered on the basis of a finding that a defendant had engaged in unprotected charitable solicitation. See Auburn Police Union v. Carpenter, 8 F.3d 886, 902 (1st Cir. 1993). And we explained there that “[a]n injunction that is narrowly tailored, based upon a continuing course of repetitive speech, and granted only after a final adjudication on the merits that the speech is unprotected does not constitute an unlawful prior restraint.” Id. at 903.

Further, a number of courts, including the Sixth Circuit and the California Supreme Court, have actually approved, in the face of First Amendment challenges, injunctions just like this one. See, e.g., Lothschuetz v. Carpenter, 898 F.2d 1200, 1208–09 (6th Cir. 1990) (Wellford, J., for the court in part); Balboa Island Vill. Inn, Inc. v. Lemen, 40 Cal.4th 1141, 57 Cal.Rptr.3d 320, 156 P.3d 339, 342–53 (2007); cf. McCarthy v. Fuller, 810 F.3d 456, 462 (7th Cir. 2015) (observing that “[m]ost courts would agree” with the Sixth Circuit on this issue).27 And no precedent, so far as I am aware, has struck a similar one down under the First Amendment.28

The majority does rely on one Supreme Court precedent that invalidated an injunction that was a remedy for past defamation: Tory, 544 U.S. 734, 125 S.Ct. 2108, 161 L.Ed.2d 1042. See Maj. Op. at 31-32. But, the Court held there that the injunction was an “overly broad prior restraint” only because the defamation victim died while the case was pending before the Court. Tory, 544 U.S. at 738, 125 S.Ct. 2108. The Court then explained that, in consequence of the defamation victim's death, even though the case was “not moot,” it was both “unnecessary” and “unwarranted” to further “explore” the enjoined parties' claims there, including the claim that “the injunction (considered prior to [the defamation victim's] death) was not properly tailored and consequently violated the First Amendment.” Id. at 736–38, 125 S.Ct. 2108; see also Carroll, 393 U.S. at 180, 89 S.Ct. 347 (noting that the Court need not decide the “thorny” problem of whether an injunction against a white supremacist organization's rally could be justified based on findings that the organization had engaged in unprotected conduct at a prior rally because the injunction could be invalidated on the narrower ground that it was issued ex parte without notice or an opportunity to be heard).29


This body of precedent suggests to me that, at the very least, there is good reason to tread cautiously in the face of the defaulted prior restraint/strict scrutiny argument, just as the Court chose to do in Tory itself. The majority may be right that the courts that have upheld injunctions as prophylactic means of preventing the likely recurrence of defamation, like the one before us, have been wrong to rely on the Kingsley Books line of Supreme Court precedent. See, e.g., Balboa Island, 57 Cal.Rptr.3d 320, 156 P.3d at 346–47. That line of precedent may be distinguishable due to defamation's more “mutable” nature. See Maj. Op. at 33. But, the briefing in Tory indicates that we also need to address a different line of precedent, which cannot be similarly distinguished.

Specifically, the Tory briefing points to Madsen and Schenck v. Pro-Choice Network of Western New York, 519 U.S. 357, 117 S.Ct. 855, 137 L.Ed.2d 1 (1997), each of which remains good law. Those cases upheld portions of injunctions, a permanent one in Madsen and a preliminary one in Schenck, that restricted defendants from “demonstrating” on public rights of way within fixed buffer zones outside abortion clinics—activity that was, of course, otherwise constitutionally protected. Schenck, 519 U.S. at 380–85, 117 S.Ct. 855; Madsen, 512 U.S. at 768–71, 114 S.Ct. 2516.

The Court reasoned that those portions of the injunctions survived First Amendment review because they “burden no more speech than necessary to serve a significant government interest.” Schenck, 519 U.S. at 372, 117 S.Ct. 855 (quoting Madsen, 512 U.S. at 765, 114 S.Ct. 2516 (citing Carroll, 393 U.S. at 183–84, 89 S.Ct. 347) ). Thus, neither case required the application of strict scrutiny, which demands that a regulation of expression be “the least restrictive means of achieving a compelling state interest.” McCullen, 134 S.Ct. at 2530.

The Court held that this less exacting form of review applied, moreover, even though the injunctions “restrict[ed] only the speech of antiabortion protesters.” Madsen, 512 U.S. at 762, 114 S.Ct. 2516. And the Court explained that this less demanding form of scrutiny applied because each injunction, in relevant part, issued “not because of the content of [the protesters'] expression, ․ but because of their prior unlawful conduct.” Schenck, 519 U.S. at 374 n.6, 117 S.Ct. 855 (quoting Madsen, 512 U.S. at 764 n.2, 114 S.Ct. 2516) (alteration and omission in original). The Court then went on to explain that those injunctions, in relevant part, survived that review because of the issuing court's supportable findings that the enjoined parties would likely continue to engage in that same conduct absent the injunction, id. at 380–82, 117 S.Ct. 855; Madsen, 512 U.S. at 769–70, 114 S.Ct. 2516, which had involved impeding access to the clinics and harassing those clinics' patients in violation of, respectively, a prior injunction in Madsen and state law in Schenck. Schenck, 519 U.S. at 375, 117 S.Ct. 855; Madsen, 512 U.S. at 763, 114 S.Ct. 2516.30

The Court in Schenck neatly described the underlying logic for permitting courts to impose such speech-restrictive prophylactic injunctive relief in rejecting the argument that the injunction must be struck down because “a ban on ‘demonstrating’ within the fixed buffer zone is ‘a ban on peaceful, nonobstructive demonstrations on public sidewalks or rights of way’ ”:

This argument ․ ignores the record in this case. Based on defendants' past conduct, the District Court was entitled to conclude that some of the defendants who were allowed within [a certain distance] of clinic entrances would not merely engage in stationary, nonobstructive demonstrations but would continue to do what they had done before: aggressively follow and crowd individuals right up to the clinic door and then refuse to move, or purposefully mill around parking lot entrances in an effort to impede or block the progress of cars.

519 U.S. at 381–82, 117 S.Ct. 855 (emphasis added).

The injunction here is no different. It, too, was imposed as a prophylactic means of ensuring that proven unprotected and unlawful expression would not be repeated. And it, too, rests on unchallenged findings that the enjoined parties likely would continue to do what they had been doing absent the injunction.

The majority nevertheless attempts to distinguish Madsen and Schenck on the ground that this particular injunction is content-based and so for that reason must be subjected to strict constitutional review. See Maj. Op. at 34-35; Reed v. Town of Gilbert, Ariz., ––– U.S. ––––, 135 S.Ct. 2218, 2227, 192 L.Ed.2d 236 (2015); Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713, 51 S.Ct. 625, 75 L.Ed. 1357 (1931). But, strict scrutiny would ordinarily apply to a speech regulation that, like the ones in Madsen and Schenck, “covered people with a particular viewpoint,” Madsen, 512 U.S. at 763, 114 S.Ct. 2516, and yet the Court did not apply strict scrutiny in either of those cases.

The Court nicely laid out the reason why in Madsen. In rejecting the argument that the injunction there was “necessarily content or viewpoint based” simply because the face of it restricted “only the speech of antiabortion protesters,” the Court explained:

To accept [that] claim would be to classify virtually every injunction as content or viewpoint based. An injunction, by its very nature, applies only to a particular group (or individuals) and regulates the activities, and perhaps the speech, of that group. It does so, however, because of the group's past actions in the context of a specific dispute between real parties. The parties seeking the injunction assert a violation of their rights; the court hearing the action is charged with fashioning a remedy for a specific deprivation, not with the drafting of a statute addressed to the general public.

Id. at 762, 114 S.Ct. 2516 (emphasis added).

That same reasoning suggests to me that it is hardly clear that this injunction is subject to strict scrutiny just because it targets specific statements. The District Court included the six statements in the injunction for the entirely content-neutral reason that the record showed with unusual clarity that the defendants had used these particular statements to defame Sindi in the past and that they would likely use them to do so again. Accordingly, there is no reason to think that the District Court singled out these statements for any reason other than the content-neutral one for which the lower courts permissibly singled out certain abortion protesters in issuing the injunctions in Madsen and Schenck—namely, to ensure that the enjoined parties would not continue unlawfully to harass their targets through the resumption of unprotected expressive activity.

For this reason, Snyder v. Phelps, 562 U.S. 443, 131 S.Ct. 1207, 179 L.Ed.2d 172 (2011), does not show—let alone clearly—that Madsen and Schenck are beside the point. See Maj. Op. at 35. In Snyder, the Court distinguished Madsen on the ground that, in Snyder, a state had imposed tort liability for prior protected speech because of both “the content and viewpoint of the message conveyed” and not as a prophylactic check against the recurrence of prior unprotected speech. 562 U.S. at 457–58, 131 S.Ct. 1207. Thus, by treating this case as if it is the same as Snyder, the majority, in my view, makes the basic mistake that generally underlies its analysis: It equates an injunction that has been crafted as a prophylactic means of stopping the likely recurrence of speech that has already been found to have been expressed in an unprotected manner with a regulation to restrict the expression of offensive but protected speech from ever being uttered at all.


Turning to the issue of whether this injunction is sufficiently narrowly tailored, I do not deny that, as written, it precludes statements that can be expressed in ways that would be protected. And, I cannot deny that, notwithstanding what the undisputed record shows, the defendants may at some point choose to use the enjoined statements for some reason other than to continue their defamatory campaign against Sindi.

But, the defendants would not then be forced to choose between contempt and silence. If, as the majority speculates, Maj. Op. at 34, they happen to have a surprising change of heart that leads them to want to, say, apologize to Sindi by renouncing—by means of repeating—their prior statements, they would need only to call upon the District Court's unquestioned responsibility to modify the injunction. See Balboa Island, 57 Cal.Rptr.3d 320, 156 P.3d at 353 & n.13.

I recognize—as the majority rightly notes—that a regulation of speech ordinarily may not be justified on the ground that it permits restricted speakers to obtain a court's permission to speak. See Maj. Op. at 34-35. But, as Madsen and Schenck recognized, cases like the one before us arise in the wake of a party's having engaged in prior unprotected conduct. And, in cases of that type, per Madsen and Schenck, lower courts have been afforded room to craft particularized, prophylactic injunctions to prevent the recurrence of irreparable harm based on supportable findings that the parties to be enjoined will resume their prior pattern of unprotected, unlawful conduct absent the injunction.

All of that said, I do not dispute that this injunction could be more narrowly drawn—just as the ones at issue in Madsen and Schenck also could have been. It could, for example, have included a coda that enjoined the listed statements only insofar as the defendants use them to defame Sindi, just as each of the injunctions in Madsen and Schenck could have included a coda that limited the protesters' presence in the buffer zone only to the extent that they behaved in an unprotected manner.31

But, such a coda comes at the expense of the specificity and clarity of the prohibition and thus at the ease of its enforcement. And because such codas invite enjoined parties to press their luck, a constitutional requirement to impose one amounts to a constitutional requirement that victims of unlawful campaigns of defamation—such as Sindi—tolerate a greater risk of suffering irreparable harm.

There is no clear precedent, however, that requires proven defamation victims to bear that risk. In fact, Madsen permitted the imposition of a prophylactic ban on some otherwise protected demonstrating in part because a more tailored prior injunction banning “blocking or interfering with public access to the clinic, and from physically abusing persons entering or leaving the clinic” had “proved insufficient.” 512 U.S. at 758–59, 114 S.Ct. 2516. And Schenck then clarified that a court may proceed with imposing “a ‘speech-restrictive’ injunction” that is found necessary to avoid irreparable harm “without first trying a ‘non-speech restrictive’ injunction.” 519 U.S. at 382, 117 S.Ct. 855.


The majority's First Amendment ruling limits a defamation victim's right to secure protection from the harm that her obstinate defamers are likely to inflict. But, this ruling may have even broader implications, as I do not see why its logic applies only to remedies for defamation. See, e.g., Aguilar v. Avis Rent A Car Sys., Inc., 21 Cal.4th 121, 87 Cal.Rptr.2d 132, 980 P.2d 846, 853–59 (1999) (holding that enjoining a defendant's use of racial epithets at the defendant's workplace was not an unconstitutional prior restraint because it was based “on [his] continuing course of repetitive conduct” that violated employment discrimination law).

By discussing the merits of this ruling at length, however, I do not mean to resolve the underlying constitutional issues. I mean only to explain my disagreement with the majority's assertion that its conclusions are so firmly rooted in basic First Amendment principles and precedents that we have good reason to depart from our usual “raise-or-waive” rule. Nor do I see any reason for the majority to address these debatable and defaulted First Amendment arguments when the majority suggests that the much less consequential, albeit still defaulted, argument that the record did not show that an injunction was necessary to prevent irreparable harm could on its own suffice to justify the invalidation of the injunction.

The majority itself counsels that “courts should not rush to decide unsettled issues when the exigencies of a particular case do not require such definitive measures.” Maj. Op. at 30 (quoting Privitera v. Curran (In re Curran), 855 F.3d 19, 22 (1st Cir. 2017) ). That counsel commands special attention, it seems to me, when its disregard risks causing irreparable harm to a proven victim of a years-long defamation campaign for reasons first brought to her attention—if even then—only at oral argument in our court.

For these reasons, I respectfully dissent from Part VII, while otherwise fully joining the majority's excellent analysis.




1.   The chalk, prepared by Dr. Sindi's counsel, purported to encapsulate evidence presented at trial. It had been referred to by Dr. Sindi's counsel during closing argument, without objection. A copy of the chalk is reprinted as Appendix A.

2.   Specifically, the jury found Samia liable for damages in the amount of $100,000 for intentional infliction of emotional distress, $400,000 for defamation, $2,000,000 for tortious interference with contract, and $400,000 for tortious interference with advantageous relations. The jury found Ann liable for $100,000 for defamation, $400,000 for tortious interference with contract, and $100,000 for tortious interference with advantageous relations.

3.   A copy of the Amended Final Judgment is reprinted as Appendix B.

4.   Samia and Ann also invoke the protections of Article 16 of the Massachusetts Declaration of Rights. They have not developed, though, any separate analysis under this provision. And in any event, “the criteria which have been established by the United States Supreme Court for judging claims arising under the First Amendment ․ are equally appropriate to claims brought under cognate provisions of the Massachusetts Constitution.” Doe v. Sex Offender Registry Bd., 459 Mass. 603, 947 N.E.2d 9, 28 (2011) (quoting Ops. of Justices, 387 Mass. 1201, 440 N.E.2d 1159, 1160 (1982) ).

5.   Absent plain error, we treat the relevant jury instructions as the law of the case because neither Samia nor Ann interposed any timely objection to them. See Moore v. Murphy, 47 F.3d 8, 11 (1st Cir. 1995); see also United States v. Hussein, 351 F.3d 9, 18 (1st Cir. 2003) (noting that unobjected-to jury instruction becomes binding unless plainly erroneous).

6.   Although the appellants make passing mention of their plaint that the defamation verdicts are against the weight of the evidence, they do not accompany that plaint with any developed argumentation. Consequently, we deem any such challenge abandoned. See United States v. Zannino, 895 F.2d 1, 17 (1st Cir. 1990).

7.   Although Samia disclaimed responsibility for some of these e-mails and posts, the jury supportably could have found that she authored all of them.

8.   We need not linger long over the appellants' exhortation that we should order a new trial because of allegedly inflammatory statements made by Dr. Sindi's counsel during closing argument. These statements drew no contemporaneous objection at trial; and since the claim of error is made for the first time in the appellants' reply brief, we deem it too little too late. See United States v. Eirby, 515 F.3d 31, 36 n.4 (1st Cir. 2008); Sandstrom v. ChemLawn Corp., 904 F.2d 83, 86 (1st Cir. 1990).

9.   Samia points out that some evidence in the record suggests that Dr. Sindi's functioning was not impaired. This evidence includes Dr. Sindi's ambitious travel schedule, her service as a Saudi government official, and her continued work as a scientist and entrepreneur during the relevant time frame. In the end, though, this suggestion boils down to an invitation that we should weigh conflicting evidence differently than the jury—and that is an invitation that we must decline. See Trainor, 699 F.3d at 26 (making clear that, on Rule 50 motion, reviewing court must draw “all reasonable inferences” from the evidence favorably to nonmovant). It is for the jury, in the first instance, to resolve conflicts in the evidence and to decide factbound issues on which reasonable minds may differ. See id.; Agis, 355 N.E.2d at 319.

10.   While the appellants make passing mention that the verdicts are against the weight of the evidence, they offer no developed argumentation on point. Thus, we deem their motion for a new trial on liability abandoned. See Zannino, 895 F.2d at 17.

11.   Although the appellants have not adequately developed a separate argument concerning the legality of the injunction under the Massachusetts Declaration of Rights, see supra n.4, it is worth noting that Massachusetts courts have harbored doubts regarding the appropriateness of injunctions in defamation cases, see Krebiozen Research Found. v. Beacon Press, Inc., 334 Mass. 86, 134 N.E.2d 1,6 (1956) (“It is apparent that the constitutional protection of free speech and public interest in the discussion of many issues greatly limit ․ the power to give injunctive relief ․ in defamation cases.”); cf. Nyer v. Munoz-Mendoza, 385 Mass. 184, 430 N.E.2d 1214, 1217 (1982) (suggesting, in dictum, that “even allegedly false and defamatory statements are protected from prior injunctive restraint by the First Amendment and art. 16” of the Massachusetts Declaration of Rights).

12.   The amicus posits that Massachusetts law, not federal law, should govern with respect to the motion for a permanent injunction. This point of view raises a nuanced question implicating the Erie doctrine, see Erie, 304 U.S. at 78, 58 S.Ct. 817, but it is a question that we can safely bypass. For one thing, no party has objected to the district court's decision to apply the federal standard. For another thing (and relatedly), it is settled that an amicus “cannot introduce a new argument into a case.” United States v. Sturm, Ruger & Co., 84 F.3d 1, 6 (1st Cir. 1996). Finally, nothing appears to turn on this point: Massachusetts law and federal law seem to place substantially similar burdens on a party seeking a permanent injunction. See Kenyon v. City of Chicopee, 320 Mass. 528, 70 N.E.2d 241, 244 (1946).

13.   The district court, relying on precedent from the California Supreme Court, see Lemen, 57 Cal.Rptr.3d 320, 156 P.3d at 343, concluded that the challenged injunction was not a prior restraint because it followed a finding of defamation liability at trial and, therefore, was not presumptively unconstitutional, see Sindi v. El-Moslimany, No. 13-cv-10798, 2016 U.S. Dist. LEXIS 110021, at *1–2 (D. Mass. Aug. 18, 2016). We do not agree. The California Supreme Court's approach impermissibly conflates “the question of whether the injunction is a prior restraint with the issue of whether the injunction should be allowed.” Chemerinsky, supra, at 165; accord Kinney, 443 S.W.3d at 93. Consistent with this view, Dr. Sindi (in her supplemental briefing) concedes that the challenged injunction is a prior restraint. She also concedes that the appropriate level of scrutiny is strict scrutiny.

14.   For example, a criminal suspect once sued a newspaper for defamation over its report that he had been arrested “after assaulting a police officer ․” Foley v. Lowell Sun Publ'g Co., 404 Mass. 9, 533 N.E.2d 196, 196 (1989). Though the plaintiff insisted that this amounted to a false accusation that he had committed assault, the Massachusetts Supreme Judicial Court disagreed after reviewing the allegedly defamatory sentence in the context of the entire article. See id. at 197. Among other things, the headline made clear that the plaintiff had only been “charged with assaulting [the] officer,” and the story repeatedly employed cautionary language. Id. (emphasis in original); see id. at 199 (reporting that the plaintiff committed the assault, “according to police”). Once the statement was “read in the context of the article as a whole, its clear meaning [was] to report” the plaintiff's arrest, not to accuse him of committing assault. Id. at 197. Since it was undisputed that the plaintiff had been arrested, the statement was not actionable.

15.   For the sake of completeness, we note that the injunction appears to suffer from other defects, including the absence of any detailed findings regarding the adequacy of remedies at law (a sine qua non for injunctive relief). See eBay, 547 U.S. at 391, 126 S.Ct. 1837. This omission is especially troublesome in light of the strong presumption that damages are an adequate remedy for a defamation plaintiff. See Metro. Opera Ass'n, 239 F.3d at 177; Organovo Holdings, Inc. v. Dimitrov, 162 A.3d 102, 117 & n.67, 119 (Del. Ch. 2017).We also note that our holding eliminates any necessity for us to pass upon the appellants' other challenges to the injunction, including their contention, clearly articulated for the first time at oral argument in this court, that the Seventh Amendment bars the injunction because the jury returned only a general verdict.

16.   The majority at times relies on precedents that apply something less than strict scrutiny, which requires that a regulation of speech be “the least restrictive means of achieving a compelling state interest.” McCullen v. Coakley, ––– U.S. ––––, 134 S.Ct. 2518, 2530, 189 L.Ed.2d 502 (2014). For example, the majority relies on the constitutional test described in Carroll v. President & Commissioners of Princess Anne, 393 U.S. 175, 89 S.Ct. 347, 21 L.Ed.2d 325 (1968). See Maj. Op. at 34, 35-36. But, in Madsen v. Women's Health Center, Inc., 512 U.S. 753, 114 S.Ct. 2516, 129 L.Ed.2d 593 (1994), the Supreme Court described the Carroll test as no different from the one applied in Madsen, id. at 767, 114 S.Ct. 2516, which the Supreme Court distinguished from strict scrutiny, id. at 762–64, 114 S.Ct. 2516, and which the majority here distinguishes from the form of “strict scrutiny” that it asserts applies. Maj. Op. at 35.

17.   To support this finding, the District Court pointed to Sindi's evidence showing that the defendants “continued their libelous campaign even up to the night before trial began” and that at trial they then “both admitted under oath that they intended to continue their defamatory campaign in the future.” In addition, the District Court reasoned that, “[e]ven following a jury award of $3,500,000 in damages, [the defendants'] opposition to the motion for [a] permanent injunction speaks only to their purported right to make the statements and the court's purported lack of authority to enjoin the conduct, but offers no assurances that they will voluntarily stop their tortious conduct.”

18.   The defendants' counsel did represent, in response to a question from the District Court at the hearing on the proposed injunction, that “it is not their intention to continue making these statements.” But, he offered no evidence, and he conceded that the testimony at trial was to the contrary.

19.   In the District Court, the defendants did appear to attack the enjoining of a libel under Massachusetts law. But, whether this injunction may issue under Massachusetts common law and the Massachusetts Constitution is among the issues that the defendants have failed to preserve, as the majority recognizes. See Maj. Op. at 30 n.11. In any event, any such state-law-based argument is not clearly supported by the state precedent brought to our attention by the amicus, which precedent the majority observes merely raises “doubts” about the propriety under Massachusetts law of an injunction of the kind before us but does not categorically preclude its issuance. See id.

20.   As the majority notes, the defendants did not develop a timely Seventh Amendment challenge to issuing the injunction absent a special verdict by the jury. Maj. Op. at 35-36 n.15.

21.   The injunction cases on which the majority relies do not suggest otherwise. See Maj. Op. at 29. In two of them, the Supreme Court on its own raised arguments that the parties had not pressed only because the equitable jurisdiction of the federal courts was at issue. See Schlesinger v. Councilman, 420 U.S. 738, 743–44, 753–61, 95 S.Ct. 1300, 43 L.Ed.2d 591 (1975); Younger v. Harris, 401 U.S. 37, 40, 43–54, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971). But, here, the majority assumes (without deciding) that the District Court did have equitable jurisdiction to issue the injunction, see Maj. Op. at 35, and strikes it down based solely on the defendants' personal rights under the First Amendment. The only other injunction case that the majority cites reached an arguably unpreserved argument that clearly had been raised on appeal. See Real Estate Bar Ass'n for Mass., Inc. v. Nat'l Real Estate Info. Servs., 608 F.3d 110, 125–26 (1st Cir. 2010).

22.   There is yet another way to issue a narrower, non-constitutional ruling in this case. It is by no means clear that “whatever equitable remedy is available in a State court must be available in a diversity suit in a federal court,” given the precedent that suggests that “[e]quitable relief in a federal court is of course subject to restrictions”—including that “the suit must be within the traditional scope of equity as historically evolved in the English Court of Chancery”—and “[t]hat a State may authorize its courts to give equitable relief unhampered by any or all such restrictions cannot remove these fetters from the federal courts.” Guar. Tr. Co. v. York, 326 U.S. 99, 105–06, 65 S.Ct. 1464, 89 L.Ed. 2079 (1945). Thus, even if there were a reason to decide this case based on defaulted arguments, I do not see why it is clear that the right defaulted argument to rely on is one that restricts the authority of not only federal courts but also state courts. And that is especially so given the care with which state courts seem to be grappling with the longstanding question concerning the scope of their own equitable jurisdiction to remedy defamation. See Roscoe Pound, Equitable Relief Against Defamation and Injuries to Personality, 29 Harv. L. Rev. 640 (1916).

23.   In every case that the majority cites concerning this doctrine, see Maj. Op. at 27-28, the issue we reached had been, unlike in this case, timely raised by the defaulting party on appeal. See Gencarelli v. UPS Capital Bus. Credit, 501 F.3d 1, 8 (1st Cir. 2007); Chestnut, 305 F.3d at 19–21; Nat'l Ass'n of Soc. Workers, 69 F.3d at 627–30; La Guardia, 902 F.2d at 1012–13. Nor do the Supreme Court cases that the majority cites with respect to excusing procedural defaults address the circumstances in which appellate courts may reach issues never timely raised on appeal. See Maj. Op. at 27-28, 28-29. In Hormel v. Helvering, 312 U.S. 552, 61 S.Ct. 719, 85 L.Ed. 1037 (1941), the Supreme Court merely acknowledged a reviewing court's authority to reach an unpreserved issue that had been argued before it on appeal. Id. at 554–59, 61 S.Ct. 719. And, in Singleton v. Wulff, 428 U.S. 106, 96 S.Ct. 2868, 49 L.Ed.2d 826 (1976), the issue was whether a reviewing court may pass on a properly presented merits argument, as opposed to remanding, after reversing a lower court's dismissal for non-justiciability; the Court had no occasion to address a reviewing court's discretion to address an argument that no party had properly presented to it. Id. at 120–21, 96 S.Ct. 2868.

24.   We have explained that the “great public moment” factor concerns whether the defaulted argument “touches upon policies as basic as federalism, comity, and respect for the independence of democratic institutions.” Nat'l Ass'n of Soc. Workers, 69 F.3d at 628. It is at least not obvious to me that the defaulted strict scrutiny/prior restraint argument implicates policies of that sort, unlike the defaulted arguments about the immunities enjoyed by state legislators and municipalities in National Association of Social Workers, 69 F.3d at 627, and Chestnut, 305 F.3d at 19–20.

25.   Nor do the interests of third parties make the First Amendment interests potentially at stake in this case any less theoretical. This injunction expressly applies only to the two defendants, and they have not challenged the District Court's findings that they likely want to use the statements only as they had used them before, which necessarily means that they are unlikely to communicate the statements to any third parties for any protected purpose.

26.   The only precedents involving injunctions targeted at unprotected speech that the defendants cite in their supporting brief for the view that strict scrutiny applies here did not in fact apply strict scrutiny. For example, the defendants cite language from Tory v. Cochran, 544 U.S. 734, 125 S.Ct. 2108, 161 L.Ed.2d 1042 (2005), that “[a]n ‘order’ issued in ‘the area of First Amendment rights’ must be ‘precis[e]’ and narrowly ‘tailored’ to achieve the ‘pin-pointed objective’ of the ‘needs of the case.’ ” Id. at 738, 125 S.Ct. 2108 (second alteration in original) (quoting Carroll, 393 U.S. at 183–84, 89 S.Ct. 347). They also cite language from Pittsburgh Press Co. v. Pittsburgh Commission on Human Relations, 413 U.S. 376, 93 S.Ct. 2553, 37 L.Ed.2d 669 (1973), that an injunction against unprotected commercial speech should “sweep[ ] no more broadly than necessary.” Id. at 390, 93 S.Ct. 2553. The Supreme Court, however, has expressly stated that a test requiring that an injunction “ ‘burden no more speech than necessary’ to accomplish its objective” is no different from the Carroll test, Madsen, 512 U.S. at 767, 114 S.Ct. 2516, and that neither test amounts to strict scrutiny. See id. at 762–64, 114 S.Ct. 2516.Sindi does say in her supplemental brief that strict scrutiny applies here. But, it would be ironic to conclude that we are bound by her acceptance of the defendants' asserted standard of review, given that she does so in a supplemental brief that she submitted only because we chose to disregard her counsel's quite justified contention at oral argument (and repeated in her supplemental brief) that we have no reason to overturn the injunction on grounds that the defendants had not timely raised.

27.   See also Retail Credit Co. v. Russell, 234 Ga. 765, 218 S.E.2d 54, 62–63 (1975); Advanced Training Sys. v. Caswell Equip. Co., 352 N.W.2d 1, 11 (Minn. 1984); Flint v. Hutchinson Smoke Burner Co., 110 Mo. 492, 19 S.W. 804, 806 (1892); Nolan v. Campbell, 13 Neb.App. 212, 690 N.W.2d 638, 652 (2004); O'Brien v. Univ. Cmty. Tenants Union, Inc., 42 Ohio St.2d 242, 327 N.E.2d 753, 755 (1975); cf. Wagner Equip. Co. v. Wood, 893 F.Supp.2d 1157, 1161–62 (D.N.M. 2012) (adopting a constitutional rule that such an injunction may issue); Hill v. Petrotech Res. Corp., 325 S.W.3d 302, 309 (Ky. 2010) (same).

28.   The only precedents of which I am aware that have struck down injunctions in defamation cases as prior restraints did so under state constitutions. See Kinney v. Barnes, 443 S.W.3d 87, 101 (Tex. 2014); Willing v. Mazzocone, 482 Pa. 377, 393 A.2d 1155, 1157–58 (1978); see also Kramer v. Thompson, 947 F.2d 666, 669–80 (3d Cir. 1991) (applying Pennsylvania law under Willing, despite finding the authorities upholding such injunctions under the First Amendment to be “quite persuasive”).

29.   Significantly, the Supreme Court stayed its hand in Tory even though the injunction there was even broader than the one here, insofar as it permanently enjoined Ulysses Tory “and his employees, agents, representatives, and all persons acting in concert, cooperation or participation with him” from, among other things, “orally uttering statements about [the plaintiff]” in a public forum. Pet'rs' Br. at 5–6, Tory, 544 U.S. 734, 125 S.Ct. 2108, 161 L.Ed.2d 1042.

30.   The Court also relied in both cases on the fact that “alternative channels of communication were left open to the protesters.” Schenck, 519 U.S. at 374 n.6, 117 S.Ct. 855 (citing Madsen, 512 U.S. at 764 n.2, 114 S.Ct. 2516). That is, the protesters were “not prevented from expressing their message in any one of several different ways” so long as they were outside the buffer zone. Madsen, 512 U.S. at 764 n.2, 114 S.Ct. 2516. Likewise, the defendants here may still express any protected message through “different ways.” For example, while the majority speculates that the defendants might one day wish to apologize by repeating the enjoined words, Maj. Op. at 34, the defendants could still apologize without repeating the enjoined words. Of course, that might not be a satisfactory alternative to a defamer who actually wants to apologize by using enjoined words, though it would seem to be the best way of doing so sincerely. But, where a defamation defendant objects to a proposed injunction on that ground, the district court could easily accommodate the concern by fashioning the injunction to permit the apology. Again, it is only due to the artificial posture of this case that we are concerning ourselves with the potential infringement of the expression of messages that the defendants have never said they want to express.

31.   As explained in his thoughtful brief, the amicus would go one step further and say that even a coda would not be enough to save the injunction before us because the injunction “threatens criminal punishment [for violating the injunction] without providing the important procedural safeguards that criminal libel law provides.” In my view, however, this argument, not raised by the defendants, mistakenly equates criminalizing defamation as primary conduct (as in the case of criminal libel) with criminalizing the violation of an injunction that has been issued as a properly predicated prophylactic protection against the future expression of unprotected speech found likely to recur. Certainly there were no criminal safeguards provided for in the injunctions in Madsen and Schenck. See Pro-Choice Network of Western N.Y. v. Project Rescue Western N.Y., 799 F.Supp. 1417, 1440–41 (W.D.N.Y. 1992); Operation Rescue v. Women's Health Ctr., Inc., 626 So.2d 664, 676–82 (Fla. 1993) (per curiam). But, the Court was not troubled by that fact, even though the underlying harassing conduct could be criminalized only by respecting those safeguards.

SELYA, Circuit Judge.

Copied to clipboard