An estimated ten million Americans have had sexually explicit images of themselves posted or threatened to be posted without their consent,1 but courts have struggled to determine which among these acts of so-called revenge porn2 constitute criminal privacy violations. Recently, in State v. VanBuren,3 the Vermont Supreme Court affirmed the dismissal of a prosecution under Vermont’s revenge porn statute, holding that a woman had no reasonable expectation of privacy regarding naked photos she sent to a man with whom she was not in a relationship.4 By defining privacy in terms of relationships, the court constructed an underprotective standard that departs from the statutory text and threatens to distort the privacy analysis at victims’ expense.
In October 2015, Dana,5 a woman in Vermont, sent photos of herself naked via Facebook Messenger to Anthony Coon, an ex-boyfriend.6 At the time, Coon was allegedly involved with another woman, Rebekah VanBuren, who accessed Coon’s account and saw the photos.7 The next day, VanBuren posted the images on Coon’s public Facebook page and tagged Dana in the photos, which multiple people saw.8 After Dana left a voicemail for Coon asking him to delete the photos, VanBuren called back from Coon’s phone,9 saying she was going to “ruin” Dana.10 She called Dana a “moraless [sic] pig”11 and said she would tell Dana’s employer, a childcare facility, about “what kind of person work[ed] there.”12
After Dana reported the incident to police, the State charged VanBuren with a misdemeanor under its revenge porn statute, section 2606.13 The statute prohibits “knowingly disclos[ing] a visual image of an identifiable person who is nude or who is engaged in sexual conduct, without his or her consent.”14 The defendant must act “with the intent to harm, harass, intimidate, threaten, or coerce the person depicted,” and the disclosure must objectively cause harm.15 Section 2606 carves out several exemptions from liability, including for “images involving voluntary nudity or sexual conduct . . . in a place where a person does not have a reasonable expectation of privacy.”16 In February 2016, VanBuren filed a motion to dismiss on constitutional and statutory grounds.17 Under the First Amendment, she brought a facial challenge to the statute.18 Under the terms of the statute, VanBuren argued that her disclosure was exempt because Dana had no reasonable expectation of privacy.19
In July 2016, the Vermont Superior Court in Bennington dismissed the charge, invalidating the statute as facially unconstitutional under the First Amendment.20 The court held that the statute was an impermissible content-based regulation of protected speech.21 Section 2606 regulates content, the court reasoned, because it applies to images based on their depictions of nudity or sexual conduct.22 The court concluded that since the statute did not qualify for the categorical exclusion from full First Amendment protection for obscenity, it was subject to strict scrutiny, which requires that the law be narrowly tailored to serve a compelling state interest.23 Section 2606 failed the test because less restrictive alternatives such as civil penalties might have served the government’s interest in protecting individual privacy.24 After ruling the statute unconstitutional, the court dismissed the case without reaching VanBuren’s statutory arguments.25
The State appealed the decision to the Vermont Supreme Court,26 which in August 2018 upheld the statute.27 Writing for the majority, Justice Robinson28 recognized that, although the statute did not qualify as a regulation of obscenity under existing First Amendment doctrine, revenge porn would be a strong contender for a new categorical exemption.29 She identified two reasons for this: First, revenge porn is a private matter, and speech on private matters generally receives less First Amendment protection than does speech on public matters.30 Second, revenge porn is an invasion of privacy — a longstanding right that is not necessarily subordinate to free speech.31 However, Justice Robinson ultimately concluded that the U.S. Supreme Court’s aversion to new categorical exemptions required the court to refrain from predicting the addition of a new exemption for revenge porn.32 Thus, the court reviewed the statute under strict scrutiny.33
The statute survived strict scrutiny since the government’s interest was compelling and the statute was narrowly tailored.34 With revenge porn, Justice Robinson noted, the government’s interest in protecting privacy is “at its highest,” given the severe emotional and professional consequences that regularly follow disclosure.35 Moreover, Justice Robinson perceived little distinction between section 2606 and commonplace restrictions on disclosures of health and financial information because both types of limits govern speech conveyed in the context of private relationships.36 The statute was narrowly tailored because it covered a restricted set of images and contained intent and objective harm requirements.37 To ensure narrow tailoring, the court provided a “narrowing construction” of the statute, reading its exclusions from liability to exempt any images that individuals either recorded or distributed “in a manner that undermines any reasonable expectation of privacy,” even though the statute did not clearly address distribution.38
Justice Skoglund dissented, maintaining that section 2606 could not withstand strict scrutiny.39 The State lacked a compelling interest since it could not in the digital age purport to protect individuals who send such photos from “their own folly.”40 Nor was the statute narrowly tailored, as civil remedies offered a less restrictive alternative.41
In June 2019, after additional briefing, the court ruled on the second issue of statutory application. It held that the prosecution had failed to allege that Dana had a reasonable expectation of privacy and dismissed the claim.42 Writing for the court once again, Justice Robinson43 held that the State had not met its burden because the State stipulated that Dana and Coon were not in a relationship when she sent him the photos, nor did any other evidence indicate they had a relationship that could give rise to a reasonable expectation of privacy.44 Finding that a reasonable expectation of privacy was a prima facie element of the crime, not a defense,45 the court proceeded to give further indications about what a reasonable expectation of privacy entailed. The court would not “precisely define” a reasonable expectation of privacy but said in this context that it required “the exclusion of everyone but a trusted other or few.”46 Tort standards for invasion of privacy were instructive.47 Fourth Amendment privacy expectations were not relevant, however, because revenge porn required no balancing of law enforcement interests.48
Section 2606 assesses privacy expectations based on where images were recorded, a standard that is too broad to withstand strict scrutiny. The court attempted to repair the constitutional weakness by attaching privacy expectations to the way images are distributed, which in VanBuren hinged on the relationship between the subject and the recipient. This narrowing construction causes trouble of its own, however. As a matter of statutory interpretation, the text does not support a distribution-based standard and fails to explain how such a standard would apply. Even if the court’s interpretation had textual support, it would create practical problems for courts tasked with judging relationships and would risk underprotecting victims with legitimate privacy interests outside of relationships.
Given the constitutional infirmity of the statutory language, some divergence from the text may have been warranted. As the court recognized, a privacy standard that depends only on the place of recording would protect many images that are in fact public, such as photos that an individual took in her bedroom but then posted in a park or on a blog.49 It is difficult to understand how the State could assert a compelling privacy interest in such images. The wide sweep of the text implies that less restrictive alternatives are available for protecting the State’s interest. Since the statute is not narrowly tailored, section 2606 as written fails strict scrutiny. As a result, per Vermont precedent, the court was obliged to read the statute to avoid invalidation if a “readily apparent construction” could resolve the constitutional problem.50
Yet the construction that the court chose is far from obvious. The court recognized that the text did not explicitly cover distribution, but it believed its interpretation coincided with the legislature’s “clear intent to protect people[’s] reasonable expectations of privacy in intimate images.”51 However, just because the legislature sought to protect reasonable privacy expectations does not mean that the legislature understood privacy to attach based on the manner of distribution, particularly not when the statute refers to the place of recording. Privacy is a notoriously ambiguous inquiry,52 perhaps especially in revenge porn, where it is saddled with jurisprudence from other contexts.53 One court has even described a reasonable expectation of privacy as “implicitly inherent from the nature of the act depicted.”54 In the absence of a settled doctrinal definition, the best signal of what the legislature intends is what it says.
In light of the statutory text, the court’s understanding of privacy expectations is unpersuasive for three reasons. First, the court’s reading strains the syntax of section 2606. The statute exempts from liability “[i]mages involving voluntary nudity or sexual conduct in public or commercial settings or in a place where a person does not have a reasonable expectation of privacy.”55 Under the most natural reading of the exception, the preposition “in,” which appears twice in the exemption, modifies the words that precede it: “nudity or sexual conduct.”56 Thus the privacy expectation attaches to the place where the depicted nudity or sexual conduct occurred. But the court maintained that the exemption stretches further, to shelter images “recorded in a private setting but distributed by the person depicted to public or commercial settings or in a manner that undermines any reasonable expectation of privacy.”57 That is, under the court’s reading, “in” would oddly take two referents, both the “images” and the nudity or sexual conduct depicted in the images. To bring the place of the image within the reach of the exception, the court stretched the grammatical function of the text.
Second, the court’s reading raises questions about the place of distribution that the statutory text does not answer. While it is relatively simple to identify the place where the nudity or sexual conduct occurred, it is more challenging to identify the place of the image, particularly in cyberspace,58 where many of the images that the statute implicates are distributed. In VanBuren, for example, the photos could be characterized as residing within the Facebook messaging inbox, on the Facebook platform, or on the device that displayed them, and the privacy expectations associated with each location may differ.59 Perhaps in an attempt to avoid delineating the uncertain boundaries of digital place, the court replaced the statutory term “place” with the term “manner.”60 Instead of asking where the image was distributed, the court now asked how, an inquiry that is at least as open ended. This shift in language further disregards the physical constraint that the text set forth.
Third, a distributional standard is incomplete because the statute does not specify a distributor. The court read the statute to shield defendants from liability where the image is distributed in a public place. Without additional detail, an image could in theory lose protection if anyone, including the defendant, distributed the image to the public. The court avoided this loophole by considering only those locations where the subject, the “person depicted,” distributed the image,61 but nothing in the language of the statute supports this restriction to a particular individual. An image is “in a place” regardless of who put it there. Because a distributional standard is coherent only when it is limited to particular distributors, and because the statute provides no such limitation, a distributional standard does not comport with the text.
Nevertheless, the court applied the standard it crafted to suit a particular vision of privacy that is framed around relationships. Under the place-of-recording standard that the statute prescribes, the court would have had little cause to investigate the relationship between the subject and the recipient. But under the court’s manner-of-distribution standard, the court had reason to analyze the nature of their relationship. In the court’s view, relationships are at the heart of the State’s interest in protecting people’s privacy.62 The court took issue with disclosures of personal information — whether related to finance, health, or sex — because the information was obtained in the context of an intimate or confidential relationship.63 Since Dana and Coon had no such relationship when she sent him the photos, the court determined that reasonable privacy expectations and the State’s interest in upholding those expectations disappeared.64
This focus on relationships as the crux of the privacy analysis causes two problems. First, it creates interpretive and factfinding trouble for courts, which often struggle to characterize personal relationships.65 Without a clear definition of private relationships, judges may insert their own preferences.66 More basically, the people involved may themselves disagree about the nature of their relationship. In VanBuren, these difficulties are not apparent, since the State stipulated that Dana and Coon were not in a relationship when she sent him the photos.67 Yet even here, conflicting narratives surface just to the side, in the relationship between VanBuren and Coon. While VanBuren “described herself as Mr. Coon’s girlfriend,” Coon told Dana that VanBuren was “obsessed with him and that he had never slept with her.”68 Such conflicts are to be expected in revenge porn cases since many photos are sent on the fringes of relationships, such as in flirting or cheating.69 The exchange of photos itself may foster and thus to some extent attest to intimacy.70 Standing alone, privacy is already difficult to untangle.71 Piling on the mess of personal relationships does little to clarify and much to complicate the analysis.
Second, a privacy analysis that centers on relationships excludes victims with legitimate injury since relationships are an inadequate proxy for privacy. Even where there is no private relationship, there may be other reasons to find a privacy interest in sexual images.72 Relationship-based definitions of privacy are usually too narrow because they fail to recognize a person’s self-interest in privacy, which can exist separate from the status of their relationship: “Individuals not intimately related may nevertheless assert that their relation or activity is a private one in the sense that it is not the proper concern of the community . . . .”73 Sexual activity is commonly understood to be that.74 Privacy in sexual images is valuable not just for its own sake but also because it furthers individuals’ interest in self-expression.75 Through her photos, Dana forged an identity for herself as a sexual being separate from her identity as an employee at a childcare facility. This space for exploring multiple versions of the self is the area VanBuren may have sought to invade when she threatened to tell Dana’s employer “what kind of person work[ed] there,”76 bringing Dana’s private sexual persona to her public professional role without her consent. The court failed to grapple with such harms to the privacy and concomitant self-expression of the individual.
Ostensibly, section 2606 seeks to update the law and protect those involved in the increasingly common practice of sharing sexual images. The court’s desire to preserve that protection in spite of constitutional infirmity is to be commended. But its method leaves much to be desired. By redefining the standard for privacy around relationships, the court muddled and cramped the statutory text. Its interpretation reinforces narrow conceptions of privacy that leave many victims out of bounds.