R v Jarvis, A Technologically Mindful Approach to the Meaning of Reasonable Expectation of Privacy

By: Emily Laidlaw

PDF Version: R v Jarvis, A Technologically Mindful Approach to the Meaning of Reasonable Expectation of Privacy

Case Commented On: R v Jarvis, 2019 SCC 10 (CanLII)

Last week the Supreme Court of Canada (SCC) released its long-awaited judgment R v Jarvis 2019 SCC 10 (CanLII) (Jarvis) and it is potentially a game-changer. The case focuses on a singular issue that is the core of privacy law: the meaning of the reasonable expectation of privacy (REP). What makes this case stand out from all the others is that it deals directly with frictions that have existed for a long time in how to conceptualize REP, namely the nature and extent to which we have a REP in public, how evolving technologies factor into conceptualizing REP in public, and issues of sexual integrity and privacy.

The accused was a high school teacher. He used a camera pen to surreptitiously record videos of female students as they went about their days at school. The students were unaware of the videos and at no time consented to being recorded. The videos focused on the students’ faces, breasts and upper bodies. There was a school board policy that prohibited this type of surreptitious recording and the school was subject to 24-hour security camera surveillance.

Mr. Jarvis was charged with voyeurism. Section 162(1)(c) of the Criminal Code provides, in part:

162 (1) Every one commits an offence who, surreptitiously, observes — including by mechanical or electronic means — or makes a visual recording of a person who is in circumstances that give rise to a reasonable expectation of privacy, if

…

(c) the observation or recording is done for a sexual purpose.

The facts of the case prompted visceral horror in most and highlighted the gap between the law and the world we are living in today. To put it another way, the experience of an invasion of privacy in this day and age and the concept of REP were laughably disconnected. The majority judgment in Jarvis helps bridge that gap. The question is, what will be the influence of Jarvis on the future of privacy law and REP?

There are many issues to explore related to Jarvis and I will examine only a few aspects of the case, drawing from my technology law background. My goal is to tease out the key privacy principles from the case and its potential influence in the future.

Judicial History

 At trial, the accused did not dispute that the videos were made surreptitiously and therefore the only issues were: (a) whether the videos were made in circumstances given rise to REP; and (b) whether the recording was done for a sexual purpose.

Mr. Jarvis was acquitted at trial, which was upheld on appeal. The SCC overturned the lower courts and entered a conviction. The trial judge held that the recordings gave rise to REP, but was not satisfied beyond a reasonable doubt that the recordings were made for a sexual purpose. On appeal, the Ontario Court of Appeal (ONCA) held that the trial judge erred in its findings concerning both issues (R v Jarvis, 2017 ONCA 778 (CanLII)). The trial judge erred in law by failing to find the videos were made for a sexual purpose, and erred in finding the students were in circumstances that gave rise to REP.

For the ONCA, the focus of the recordings on students’ breasts left no doubt in their minds that the videos were for a sexual purpose. However, on the issue of REP, the ONCA took a location-based approach to privacy, such that when one is out and about in the world, the REP is diminished (paras 92-98). In considering the proliferation of pictures and videos taken in public and posted online, the ONCA interpreted the concept of REP in the voyeurism offence as acting as a limit to the state’s ability to “to completely protect sexual integrity in our open society.” (ONCA, para 91). There was some wiggle room in s 162(1) with the language “circumstances that give rise to a reasonable expectation of privacy”, but in the majority’s view this gave a limited right to privacy in public, such as where parts of the body were hidden in a scenario of upskirting (recording up women’s skirts) (paras 95-96). According to the ONCA, since privacy is about a right to exclude others from private spaces and freedom from being observed, there could be no right to privacy here, because the students were in a common area of the school and were being recorded in any event by security cameras. In the ONCA’s view, the expectation that a teacher will not observe or record a student for a sexual purpose is based on the student/teacher relationship, not a REP (para 105).

Thus, Jarvis made its way to the SCC and the only issue on appeal was whether the ONCA erred in concluding that the recordings of the students were not in circumstances giving rise to a reasonable expectation of privacy (para 4).

SCC Reasoning

The Court was unanimous that the students had a REP and entered a conviction. However, the majority (delivered by Chief Justice Richard Wagner) and concurring judgments (delivered by Justice Malcolm Rowe) differed on two aspects: (a) whether s 8 Charter jurisprudence can inform interpretation of REP in s 162(1) of the Criminal Code; and (b) the appropriate contextual factors in assessing REP in s 162(1). My analysis of the case will not focus on a detailed comparison of the majority and minority judgments on these points. I encourage readers to examine Jarvis closely on these issues. Rather, my goal is to identify the key themes in the case, in particular the privacy principles that will potentially influence future cases, both criminal and civil.

Meaning of REP in the Context of s 162(1)

The majority made two points that will be influential in future cases. First, the majority rejected a location-based approach to REP under s. 162(1), preferring instead an analysis of REP based on context and circumstances. Second, they emphasized that privacy “is not an all-or-nothing concept” (para 41) and whether, or the extent to which, one has a right to privacy in public depends on context. The majority emphasized that the analysis should account of the entire context, and offered a list of non-exhaustive considerations:

1. The location of the observation or recording;
2. The conduct at issue, namely whether person was observed or recorded (a recording being more intrusive);
3. Details about the observation or recording, such as whether it was fleeting or sustained, the type of technology used etc.;
4. Whether the person consented or was aware of the observation or recording;
5. The subject matter of the observation or recording, such as incidental or targeted;
6. The purpose of the observation or recording;
7. The relationship between the parties, such as a relationship of authority or trust;
8. The personal attributes of the person observed or recorded, such as whether they are a child; and
9. Any rules or policies that regulate the type of observation or recording that took place; (para 5 and 29).

The concurring opinion accepted four of these principles as relevant to the analysis as they are required by ss 162(1) (1, 4, 5, 6) and rejected the other five factors as considerations at sentencing (paras 108-109).

In rejecting a location-based approach to privacy, the majority took the important step of sketching a framework for conceptualizing privacy in public. In their view, the right to privacy is generally higher when a person is in a private place, such as a home or washroom, but “a person does not lose all expectations of privacy, as that concept is ordinarily understood, simply because she is in a place where she knows she can be observed by others or from which she cannot exclude others.” (para 37) Taking a contextual analysis, the majority emphasized that a person can expect a type of observation or recording in one context but not another (para 38). For example, there is a difference between being captured incidentally in a photo and being the subject matter of a photo (paras 39-40). The court played out its thinking through technology examples, such as the difference between being captured incidentally and use of a telephoto lens, drone or other surreptitious methods of surveillance (para 40).

All of these examples are significant, because they demonstrate a contextual analysis at work, and reject the all-or-nothing concept that sometimes dominates privacy discourse. It brings to mind Elizabeth Paton-Simpson’s seminal work and criticism that “[m]ost people spend a great deal of their everyday lives in public places without imagining that they are being observed any more than casually and by a limited number of people.” (Elizabeth Paton-Simpson, “Privacy and the Reasonable Paranoid: The Protection of Privacy in Public Places” (2000) 50 UTLJ 305) She interrogates the assumptions that flow from a rejection of REP in public, namely the assumption that people do not expect any privacy when they are in public, and indeed, that people have a choice about what to do about it, namely stay home or take privacy protective measures.

Section 8 Charter Context

 The majority was of the view that since Parliament chose to use the term REP, a central concept in s 8 Charter jurisprudence, it followed that the Court could draw broadly from Charter jurisprudence on REP in its interpretation of the provision (paras 54-55). On this point, the concurring opinion disagreed, rather viewing the Charter and s 162(1) as conceptually distinct frameworks calling for different interpretive principles (see paras 93-106).

The majority’s interpretation of s 8 jurisprudence offers a powerful new paradigm for understanding the right to privacy under the Charter. It is not that the Court departed from earlier jurisprudence. Indeed, most of the principles emphasized by the majority were established s 8 principles, and the majority’s analysis highlighted how these principles supported their approach to REP under s 162(1). Namely, assessment of REP is contextual and takes account of the totality of circumstances (para 60); privacy is not an all or nothing concept (para 61); and analysis of REP is normative not descriptive (paras 68-70).

The power in the majority’s s 8 analysis is that it tweaked or re-framed our understanding of the jurisprudence in light of new technologies. This, I expect, will be the most influential aspect of Jarvis in future cases. For example, the Court stated that section 8 provides

a rich body of judicial thought on the meaning of privacy in our society. And far from being unmoored from our ordinary perceptions of when privacy can be expected, as Mr. Jarvis suggest, judgments about privacy expectations in the s. 8 context are informed by our fundamental shared ideals about privacy as well as our everyday experiences. (para 59)

Later, the majority acknowledged “the ubiquity of visual recording technology” (para 89), meaning that people reasonably expected to be incidentally recorded as they go about their day to day lives. What people do not expect is “to be the subject of targeted recording focused on their intimate body parts (whether clothed or unclothed) without their consent” (para 90). It is this locating of privacy in our day to day experiences that re-centres the discussion of the meaning of REP in light of new technologies, a point I will return to in examining the future influence of this case.

 Application

Applying the non-exhaustive contextual factors to the case at hand, the majority concluded that there was no doubt the students had a REP that they would not be recorded in the way that happened here (para 72). The contextual factors the majority considered in coming to that conclusion were:

• While REP in common areas of school is lower than traditional private places, a high school is not entirely public and have their own rules about behaviour (para 73).
• Students were being recorded, which is a greater privacy impact than mere observation (para 74).
• The recordings were done surreptitiously (para 75). Note here, the majority rejected the relevance of the security cameras. The presence of such cameras means there is a REP of being captured incidentally, but it does not follow that one reasonably expects to be filmed at close range by a teacher (para 76).
• The content of the recordings targeted students at close range or close-up, and the focus of the recordings was on their breasts, faces and upper bodies. As the majority noted, there is heightened REP concerning intimate body parts and sexuality (paras 78-82).
• The school board policy clearly prohibited the kinds of recordings Mr. Jarvis made (para 83).
• The recordings were a breach of trust between the teacher and students (“students should be able to reasonably expect their teachers not to use their authority over and access to them to make recordings that objectify them for the teachers’ own sexual gratification” (para 84)).
• This case involved young people, and they have a heightened expectation of privacy. Young people put a tremendous amount of trust in the adults around them that they are looking out for their best interest. Put into privacy terms, young people reasonably expect “the adults around them to be particularly cautious about not intruding on their privacy” (para 88).

Significance of Case Going Forward

An important sentence in the majority opinion is tucked away near the end of its reasoning. Chief Justice Wagner commented in obiter, “[i]ndeed, given the content of the videos recorded by Mr. Jarvis and the fact that they were recorded without the students’ consent, I would likely have reached the same conclusion even if they had been made by a stranger on a public street rather than by a teacher at school in breach of a school policy” (para 90). This sentence enables a wide reading of the case and minimizes the risks it will be read narrowly to apply to schools, young people or strictly cases of voyeurism or harms to sexual integrity.

 As identified above, the key point of significance in the case seems to be re-centering of REP in light of new technologies in our day to day experiences. On this point I wish to elaborate as I think it is crucial for future cases. All privacy cases, whether criminal or civil, rely on the REP concept, however loosely they might be influenced by s 8 jurisprudence. The readers are invited to inspect closely paragraphs 62 and 63 of the judgment.

In particular, paragraph 63 is influential because it recognizes that the technology changes the nature of a contextual analysis. To put it another way, a contextual analysis of REP is not, and should not be, blind to the qualities of the technology at issue, both descriptively and normatively. In contrast, the approach of the ONCA in essence treated all technology as equal and categorical – in function and capabilities. According to the ONCA reasoning, since there were security cameras in the school, this reduced the expectation of privacy in relation to all technology in that space, even targeted, sexual recordings by trusted authority figures.

Jarvis’ strength is in providing a roadmap for how to do a contextual analysis in a world of evolving technologies. Rather than being technologically blind, it is technologically mindful, if you will. Technology does things in the world and Jarvis analyses a reasonable expectation of privacy normatively and in light of the capabilities of technology, rather than as risk management (e.g. the idea that since there are security cameras and the hall is semi-public the expectation of privacy shrinks). As the Court stated in Jarvis,

[w]hether a person reasonably expects privacy is necessarily a normative question that is to be answered in light of the norms of conduct in our society. And whether a person can reasonably expect not to be the subject of a particular type of observation or recording cannot be determined simply on the basis of whether there was a risk that the person would be observed or recorded. (para 68) (emphasis in original)

The SCC in Jarvis locates privacy in our everyday experiences and connects technology with privacy based on its capability to transform what we reasonably expect as private into something public (see paras 62-63).

Let me give an example of how this analytical framework helps conceptualize a REP involving technology. For example, let’s say a woman is at a sports match and her team scores a goal. She jumps up and down screaming and her top goes up revealing, briefly, her breasts. Let’s say a photographer with a telephoto lens photographs her just at the moment her breasts are revealed. In such a scenario, there was likely a REP as against use of a telephoto lens for such a photo even before Jarvis. However, let’s say the photographer incidentally captures the woman in the background of a crowd shot. Under Jarvis, such a snapshot would be reasonably expected in the circumstances. However, let’s say that the photographer edits the image, zooms in on intimate parts of her body, then shares the new image online, or manipulates it in some other way (para 62). Jarvis would ask not whether we expect to be photographed when we are in public, as that would be the end of the matter. Rather, based on Jarvis we would ask whether we were in circumstances we would reasonably expect not to be photographed in this way. This flexibility is important as we go forward in a world where technological capabilities are evolving, and misinformation and reputational damage are becoming mainstays of privacy invasions. Consider the increasing social problems of deepfakes (superimposing real images or videos onto fake ones, such as celebrity faces onto pornography videos) and fake social media profiles, such as one man’s story of a fake Grindr profile. Or consider the increasing use of internet of things (IoT) devices in situations of domestic abuse to control victims.

In my view, Jarvis provides a foundation to argue more forcefully that privacy is about more than seclusion, but about enabling us to live the good life; that privacy is woven into the fabric of our social structures (see my discussion of public and social privacy in “Online Shaming and the Right to Privacy”, Laws 2017, 6(1)). As Lisa Austin has commented about privacy more generally, “[p]rivacy is not best understood as a state of social withdrawal but as a set of norms that enable social interaction.” (Lisa Austin, “Privacy, Shame and the Anxieties of Identity” (2012) working paper). A strange aspect of Jarvis and the reasoning of the ONCA is that these young women somehow lost any sense of ownership over their sexual identity the second they were in public. Teresa Scassa aptly described this issue in relation to Jarvis in a tweet as “about the publicness of women’s bodies”. The majority in Jarvis addressed this problem in terms of “unwanted sexual attention” (para 82) and the sacrosanct nature of “our sexual selves” (para 82).  Capturing privacy in this way links with Paton-Simpson’s argument that we generally do not expect to be observed more than casually when we are in public. Underlying this, I suggest, is the role of privacy as an enabler in setting the norms of social interactions in our world, both public and private.

The facts of Jarvis are particularly disconcerting and the majority judgment goes to great lengths to identify the heightened REP we have concerning our sexual identities and the targeted, sexualized harms that are central to what happened here. Indeed, sexual integrity is core to a voyeurism offence under s 162(1) and is the inflection point in the contextual analysis on which all things turn. The ONCA location-based approach to REP prevented the sexualized nature of the harm to figure as prominently, while the majority approach of the SCC allowed sexual integrity to take centre stage. However, I would have liked to see the majority more explicitly address the gender-based nature of this harm, because online abuse is increasingly gender-based, and more broadly, tends to target traditionally marginalized groups.

Here are a few statistics for readers concerning another type of technology-related abuse (most discussed in “Online Shaming”). Sexting is an increasingly common form of intimacy, with 57% of men and 45% of women having received an intimate image on their phone (Derek Bambauer, “Exposed”, (2014) Minn L Rev 2015, at 2034). And yet, women are the target of 90% of cases of non-consensual distribution of intimate images (known colloquially as revenge pornography) (Danielle Citron, Danielle Keats Citron, & Mary Anne Franks. “Criminalizing Revenge Porn”, (2014) 49 Wake Forest L Rev 345). A Pew Internet Centre study reported that while men were more likely to be harassed online, the nature of the harassment was different. Women were twice as likely to be targeted for online harassment based on their gender, and more likely to face sexualized harassment (Pew Internet Center, Online Harassment 2017). Technology has enabled an exacting price for women wishing to participate equally in the world. As we move forward, locating privacy in our everyday lives will require that we better account for the equality issues that inform the experience of privacy day to day.

Finally, I am struck by the widening gap between REP as now articulated by the SCC and our tort of privacy emerging out of Jones v Tsige, 2012 ONCA 32. While Charter values inform interpretation of private law, in my view the gulf between the contextual approach to REP in Jarvis and the descriptive, categorical formulation of the privacy tort is too wide and deep to bridge. I am currently working on a project funded by the Social Sciences and Humanities Research Council on developing the tort of privacy to address online abuse. Currently, the tort, where it exists, imports elements of the U.S. Restatement (Second) of Torts, which creates a taxonomy of harms caused by certain conduct (intrusion on seclusion, public disclosure of private, embarrassing facts, depiction in a false light and appropriation of personality for commercial gain). As Danielle Citron has criticized in the American context, by focusing on a narrow list of harms, the tort was calcified with the particular privacy concerns of that time (1960s to be precise when David Prosser proposed this tort) (Danielle Citron, Mainstreaming Privacy Torts, (2010) 98 Cal L Rev 1805). The problem is that the taxonomy boxes the privacy harm and limits a contextual analysis that might evolve and adapt with changing technologies. Put more simply, the tort is not fit for purpose. Contextual approaches to a privacy tort are evident in other countries, such as the United Kingdom, which might form a model for a Canadian tort of invasion of privacy. As it stands, however, I am hard-pressed to see how the Charter values enunciated by the majority in Jarvis can provide guidance in civil law without revisiting the fundamental underpinnings of the tort.

This post may be cited as: Emily Laidlaw, “R v Jarvis, A Technologically Mindfully Approach to the Meaning of Reasonable Expectation of Privacy” (February 26, 2019), online: ABlawg, http://ablawg.ca/wp-content/uploads/2019/02/Blog_EL_Jarvis_Reasonable_Expectation_Privacy_Feb2019.pdf

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg