“Jones-ing” for a Solution: Commercial Street Surveillance and Privacy Torts in Canada

Abstract

While street surveillance technologies such as Google Street View are deployed with no discriminatory intent, there is selective scrutiny applied to the published imagery by the anonymous crowd. Disproportionately directed at women and members of ethnic minority groups, this scrutiny means the social risks of street surveillance are not equal. This paper considers the possibility of invasion of privacy actions in tort brought against the commercial service provider as a possible solution. Analysis suggests that Canadian law has evolved in a way such that it is exceedingly difficult to make a claim for “privacy” in tort when the plaintiff is located in public space. This evolution exists in order to ensure that innocuous behavior not be rendered actionable. Furthermore, conceptual reasons exist to suggest that actions in tort are unlikely to be the best solution to the problems posed by commercial street surveillance. While any individual case of embarrassment or nuisance matters, broader “macro-harms” that impact entire communities reflect perhaps the most serious problem associated with the selective scrutiny of street surveillance imagery. Yet, it seems difficult to justify attaching liability for those harms to the commercial providers. While limits need to be placed on the operation of these street surveillance programmes, it is unlikely that invasion of privacy actions are the most effective way to achieve that goal.

1. Introduction

Commercial public street surveillance, as distinct from the surveillance of private space for commercial needs (e.g., CCTV), is a growing phenomena in Canada. The best-known example is Google’s expansive “Street View” project, which has digitally photographed almost all populated areas of Canada (indeed, of the world) and then made available those images online for “virtual exploration”. While this new form of digital mapping is technologically impressive and has proved popular with many Canadians, it brings with it certain risks, driven by a form of scopophilia—the love of looking. Although the technology is facially neutral and is deployed by Google et al. without discriminatory intent, the gaze of the anonymous audience that accesses the imagery is not directed in such an equal manner. As a result, despite the benign intentions on the part of the service providers, the effects of these projects are likely to be experienced in different ways by different groups. Take, for instance, the range of third-party websites in which images deemed “notable” or “interesting” are copied from the original imagery published the commercial service providers, archived and re-posted for comment [1,2,3]. The individuals that participate in these online communities tend to be young, male, and digitally well-connected—and the targets of their scrutiny and subsequent commentary are predominantly drawn from racialized groups and women. Gender, ethnic, and economic biases are all revealed and reinforced by the way in which the scrutiny is applied.

In particular, the scrutiny applied to street surveillance imagery by the anonymous audience has a heavy focus on sexualized imagery of women. Images of women sunbathing [4], dressed in “revealing” clothing [5], or in swimwear [6] are particularly popular with the anonymous audience, who appear to devote considerable energies to finding such images and then rating the attractiveness of the women they feature. The website “doxyspotting.com” is perhaps the most overt example: users scour Google Street View for images of women believed to be sex workers, and then record them in a competitive fashion. The “frequently asked questions” portion of their website states:

Why spot prostitutes on Google StreetView?

Google Maps Street View is the ideal voyeuristic tool. People are voyeuristic in nature—men especially. It is very easy to spot a girl with nice big boobs on Google Street View. Spotting prostitutes may seem a bit more difficult. But if you know what to look for, it actually becomes quite easy. For example you can drop the StreetView Icon pretty much anywhere in Madrid’s Zona Industrial de Villaverde and you will be able to spot a hooker on Google Street View. Doxy Spotting is almost a sport. It is the perfect and insuspicious [sic] way to kill some time at work or if you are bored. But be warned—doxy spotting is highly addictive.

[7]

This digital replication of the “male gaze” is by no means limited to doxyspotting.com. Across all the websites that archive images, invariably young women form a significant portion of what is deemed by the viewers to be notable or interesting, even though all they are doing is simply “existing” as women in public space. Once the images are uploaded and shared, the women tend to be the subject of sexualized commentary. The scrutiny applied to street surveillance imagery is by no means limited to this kind of sexual voyeurism, however. Images that feature members of ethnic minority groups are often selected and shared, and then interpreted by online commentators as showing criminality, despite there not being any clear criminal behaviour in the images. Other images can be best understood as examples of “poverty porn”, in which assumptions about members of less well-off communities are made and the images are used for the gratification of the viewer. For instance, anonymous viewers discovered a Street View image of young black men gathered around a car, and instantly concluded that it that it revealed a “drug deal from every angle”, despite a complete absence of narcotics in the image [8]. Similarly, another image discovered appeared to show a black man getting out of his car at a stop-light. A police car was elsewhere in the image, and the anonymous crowed instantly took this to mean that the man in the first vehicle was “getting arrested”, despite absolutely no evidence to support such a claim [9]. Images of what has been termed “poverty porn” are often also frequently shared and commented upon the users of these third-party websites. In one instance, a photo of three girls outside a working-class estate in England was taken to be an example of “urban poverty, drug crime, and decay” [10], while a photo of four black men walking down the street was cited as “one of the most dangerous neighbourhoods in America” [11]. In all these cases, by being photographed in these images and associated with disparaging commentary, subjects of street surveillance are subject to the phenomena of the synecdoche—a small slice of their life is taken to represent the entirety of their being.

While in extreme cases actions in tort for harassment, nuisance, or defamation might conceivably be brought against the anonymous users that re-share, scrutinize and comment upon the imagery online, any single piece of commentary likely does not rise to this level—but the cumulative effects of the commentary are nonetheless significant. This also means that, short of outright “hate speech”, actions against anonymous commentators will frequently run into (entirely legitimate) counter-balancing concerns regarding freedom of expression. Moreover, the potential for a successful action against an online harasser is greatly complicated by the fact that the putative defendant is almost certainly operating under an anonymous identity, and may not even be located in the same jurisdiction. These are not insurmountable problems, of course, but they significantly increase the costs and difficulties associated with bringing a claim against a member of the anonymous crowd, correspondingly lowering its chance of success. In this paper, therefore, I want to consider whether an alternative action brought against a well-known service provider for “invasion of privacy” would be a viable means of reducing the likelihood of serious social harms arising from the use of commercial street surveillance technologies in Canada.

This paper therefore considers the potential of bringing such an action under the statutory torts that protect privacy in several Canadian provinces (British Columbia [12], Manitoba [13], Saskatchewan [14], Newfoundland and Labrador [15]) and under the common law “intrusion upon seclusion” tort that was recently adopted in Ontario [16,17]. In either form, however, I suggest the likelihood of success remains low. The jurisprudence under the statutory torts relies upon an a priori distinction between “public” and “private” space that appears to doom any claim related to being photographed in public. While there is not yet a highly developed body of Ontario jurisprudence based on the new common law tort, the American law on which it is based may serve as a reasonable predictive guide. Under that jurisprudence, meeting the constituent elements of the tort in the context of public photography is also extremely difficult. This, though, is less a “failure” of the jurisprudence and more a necessary limitation on the scope of any privacy tort. While commercial street surveillance programmes can be understood as causing a privacy loss, that is a separate question from whether that loss ought to be actionable. A privacy tort that renders being photographed in public space actionable would likely also attach liability to a whole host of innocuous behaviour. I conclude by suggesting it may be the case that the most serious kind of harm associated with a diffuse commercial surveillance programme is not best remedied by tort law at all, since the injuries caused are not limited to just those suffered by an individual captured in a photograph and subject to unwanted scrutiny and online commentary. Rather, the selective selection of imagery for scrutiny and sharing results in broad dignitary damage to entire socio-economic groups, which harms all members whether or not any one individual is actually photographed or otherwise recorded by a street surveillance programme. It is these harms that are perhaps the most serious, but actions aimed at remedying discrete/isolated intrusions into personal privacy are not an effective solution to them.

2. The Provincial Statutory Torts for Invasion of Privacy

The contours of the four provincial statutorily-created private law actions for invasion of privacy are generally similar. The Privacy Act in British Columbia, for instance, creates a tort, “actionable without proof of damage, for a person, wilfully and without a claim of right, to violate the privacy of another” ([12], s. 1(1)). While the B.C. Act does not provide an exhaustive definition of what a violation will entail, it does note that it includes “eavesdropping and surveillance, whether or not accomplished by trespass” ([12], s. 1(4)). Individuals are deemed to be entitled to a degree of privacy that is “reasonable in the circumstances”, and the “nature, incidence and occasion of the conduct in question” as well as the relationship between the parties will assist the court in determining whether a violation of that privacy has occurred ([12], s. 1(2), (3)). Statutory defences include consent; that the conduct was in the exercise of a lawful right in defence of property or person or was authorized or required by law; or that the conduct was that of a peace or public officer engaged in their official duties ([12], s. 2(2)). There is also a defence for publication of a private matter if the matter was in the public interest or fair comment on a matter within the public interest ([12], s. 2(3)). The other provincial Acts are much the same, with one notable variation being the Manitoba variant which, unlike the others, requires that there be a “substantial” violation of privacy in order to ground a claim, and also lacks the “wilful” requirement [13]. No action against a commercial street surveillance provider has been brought under any of these Acts, and so the analysis that follows will be based on other cases of surveillance, tracking, or other forms of privacy invasion that relate to the revelation of publicly available information [18].

2.1. Wilfully

In Peters-Brown, the plaintiff sought compensation after the defendant hospital created a list of patients around whom special precautions ought to be taken; somehow, that list ended up in the hands of her co-workers, who were led to believe she was suffering from AIDS [19]. While her claim for negligence against the hospital succeeded, her claim for invasion of privacy under the Sask. Act failed, with the court reasoning that “wilful” in the context of the Act meant an “intentional” invasion of privacy ([19], para. 35). Since the hospital had not intended that the patient information be circulated beyond the hospital staff, there could be no liability under the Act. It is true that the photography that happens to capture an identifiable individual is merely an unintended consequence of the broader project to record public space—problematic as it may be, the purpose or intent behind photography by Google Street View and similar services is not to invade individual privacy or to collection personal information, but rather to create digital maps. Thus, in some sense the invasion of privacy is collateral to this purpose. However, per Hollinsworth, the issue of intent turns not on the direct intent to invade privacy, but rather the intent to do an act that the defendant “knew or should have known would violate the privacy of another person” ([20], para. 24]. On this standard, it seems that the actions of the service providers can fit within the “wilful” requirement as it has previously been interpreted under the provincial Acts. A much greater challenge, however, turns on whether or not an individual in public space has a reasonable expectation of privacy.

2.2. A Reasonable Expectation of Privacy

On the one hand, as a practical question Google seems to accept that individuals in public places have some kind of expectation of privacy vis-à-vis being photographed—after all, Google attempts to automatically blur both faces and license plates under the guise of a “privacy protection” [21]. However, as a legal question, the answer to whether or not individuals retain an expectation of privacy is significantly different. As the jurisprudence has developed under the provincial statutory torts, the courts have placed significant importance on an initial spatial or locational question in determining the reasonableness of an individual’s expectation of privacy. In circumstances where the subject is in an area typically understood as “private” and the watcher has invaded that same area in order to conduct surveillance, then the reasonable expectation of privacy afforded to the subject is considered high. Thus in cases involving sexual voyeurism [22] via hidden peepholes or cameras, courts have been quick to find that an individual’s reasonable expectation of privacy has been violated, often awarding significant punitive damages [23]. In contrast, when both the subject and the watcher are in public space, the expectation of privacy of the former is deemed to be much lower. Courts have accepted that “there is no reasonable expectation of privacy for actions taking place in public” ([24], para. 77).

A person’s reasonable expectation of privacy in his or her own home is ordinarily very high whereas… in a public place [it is] substantially less so… overt actions and behaviours occurring in public are not really ‘private’ at all.

([25], para. 21)

However, when the subject of the surveillance is in an area typically understood to be private but can easily be seen from public space without special devices, the judicial calculation is slightly more complicated. For instance, where events on openly viewable private property have been recorded from the vantage point of a public street, one court found that the individuals recorded had no recourse under the British Columbia version of the tort.

The salient feature… is the location in which the filming took place. Events transpiring on this parking lot could hardly be considered private in the sense of being shielded form observation by the general public.

([26], para. 17)

More surprisingly, perhaps, this logic has occasionally been extended to interiors of private homes that are viewable from the outside—in Milner it was held that while “a person’s expectation of privacy [is] highest in [their] home”, it was nonetheless lowered since the plaintiff had left “the blinds open and the lights [on]”, meaning anyone could see her ([24], paras. 76, 83). Other decisions appear to reject this contention, however. In Wasserman, the court found that the plaintiff had a reasonable expectation of privacy in the interior of his home, even though he had initially declined to take any steps to block the view of the camera his neighbour had set up to monitor their mutual fence following a property line dispute [27]. Wasserman also held that individuals have a reasonable expectation of privacy in the immediate vicinity of their homes, even when those areas are visible to passers-by (in Wassserman, the plaintiffs were also recorded sitting on their outdoor patio, which was viewable from the street). This was consistent with the earlier decision of Heckert, in which the complaint related to a video camera ostensibly installed for security purposes in an apartment building [28]. The image generated by the camera showed the entire common hallway, the elevator doors, and the door to the plaintiff’s unit in the immediate foreground; the images did not reveal any activities inside the unit ([28], paras. 37–39). Walker J. accepted that the plaintiff nonetheless enjoyed a reasonable expectation of privacy in the hallway, even though it was open to anyone who lived in the building and their guests and was thus a quasi-public area (or in the words of Walker J., “not truly private”) ([28], para. 90).

Now, it is true that commercial street surveillance often captures images of individuals in the vicinity of their homes, and, though rarely, might also capture images of people within their homes (for instance, through a window or open door). Distinctions can still be drawn, however, between this and the way in which the “vicinity” cases operate. In both Wasserman and Heckert, the cameras used were continuously active video cameras with a relatively narrow field of vision, trained on a fixed area, located adjacent to an a priori private place. In contrast, street surveillance cameras irregularly photograph broad areas with wide-angle lenses, using still imagery rather than continuous video recording. Of course, it is possible that future versions of the service might adopt different techniques of photography, including high resolution video [29]. Indeed, there are some services that already offer such a product, though not in Canada. Recall that the Acts all note that the reasonableness of the expectation of privacy in any given situation is altered by the “nature, incidence and occasion of the conduct in question”; Heckert and Wasserman are thus probably best interpreted as edge cases related to particularly intrusive forms of surveillance of quasi-private areas. As a general rule, when an action under one of the provincial torts has been brought for a claimed violation of privacy thanks to visual surveillance, once the spatial question about the location of the claimant has been answered by the courts as being “public” or “viewable from public space without special measures”, that effectively ends the chances for a successful claim. The influence of this spatial question is therefore most likely fatal to any claim brought against a commercial street surveillance provider brought under any of the provincial Acts that provide statutory protections against invasions of privacy.

3. The New Ontario Common Law Tort for Invasion of Privacy

Under the common law, privacy interests in Canada have historically been protected not through a distinct “privacy” tort, but rather others such as nuisance, trespass, defamation, injurious falsehood, deceit, and passing off [30]. In the 1990s and the early twenty-first century, however, the influence of s. 8 the Charter [31] began to be felt in this area, as it “underscore[d] that privacy is a societal value to be protected” ([32], pp. 226, 230). Lower courts in Ontario began mulling the protection of privacy through a specific tort more explicitly, though they continued to prefer to rely on more well-accepted torts to resolve the cases before them [33]. In Tran, however, the Superior Court of Justice reviewed a number of those early cases and suggested that together they established that an independent tort of invasion of privacy ought to be considered a “valid” cause of action in Ontario ([34], para. 38). In 2005, the Ontario Court of Appeal confronted the possibility in Euteneier [35], in which the plaintiff claimed, inter alia, that her privacy had been invaded when she was placed in a video-monitored cell and stripped of her clothing by two police officers subsequent to a (lawful) arrest and detention. The defendant police officers were successful at trial and partially successful at the divisional court. The cross-appeal of the plaintiff at the Court of Appeal was an argument that the trial judge erred by failing to take into account the plaintiff’s privacy and dignity interests when formulating the duty of care standard to be applied. In dismissing this, the Court noted that counsel for the plaintiff “properly conceded in oral argument before this court that there is no ‘free-standing’ right to dignity or privacy… at common law” ([35], para. 63). They did not, however, rule that such a tort could not exist.

Released immediately on the heels of Eutenier (and so not citing it) came Somwar [36], in which the plaintiff sued a well known fast-food chain for unlawfully invading his privacy after they conducted a credit check on him without permission. The defendants brought a motion for summary dismissal, arguing that since there was no recognized tort of invasion of privacy at the common law, the plaintiff’s claim disclosed no reasonable cause of action [37]. Stinson J. concluded, however, that:

The traditional torts such as nuisance, trespass, and harassment may not provide adequate protection against infringement of an individual’s privacy interests. Protection of those privacy interests by providing a common law remedy for their violation would be consistent with the Charter values and an incremental revision’ and logical extension of the existing jurisprudence… the time has come to recognize invasion of privacy as a tort in its own right.

([36], paras 29, 31)

Somwar, though only a ruling on a motion and therefore of effectively no precedential value, was the clearest acceptance then yet made by an Ontario court of a freestanding invasion of privacy tort in the common law, and was subsequently cited with approval in two other decisions at the Superior Court of Justice [38]. This evolution appeared to culminate with the appellate decision of Jones [17], a claim relating to the improper accessing of the plaintiff’s financial records by the defendant. Though they did not know one another personally, both parties worked for the Bank of Montreal and the defendant had a relationship with the plaintiff’s former husband. After becoming embroiled in a financial dispute with that individual, the defendant accessed the plaintiff’s financial records on at least 174 occasions over two years in an attempt to confirm that the ex-husband was in fact making child support payments to the plaintiff ([16], paras. 12–14).

After reviewing the relevant jurisprudence in both Ontario and other common law jurisdictions, and considering the impact of the Charter, Sharpe J.A. (writing for the entire court) concluded that “it [was] appropriate for [the Court of Appeal] to confirm the existence of a right of action for intrusion upon seclusion” ([17], para. 65). In recognizing this new tort, Sharpe J.A. accepted that the American four-part approach to protecting privacy found in the Second Restatement of Torts was a useful starting point:

652A General Principle.

(1)

One who invades the right of privacy of another is subject to liability for the resulting harm to the interests of the other.

(2)

The right of privacy is invaded by:

(a)

unreasonable intrusion upon the seclusion of another; or

(b)

appropriation of the other’s name or likeness; or

(c)

unreasonable publicity given to the other’s private life; or

(d)

publicity that unreasonably places the other in a false light before the public [39].

Sharpe J.A. suggested that the facts in Jones lent themselves to consideration under the “intrusion upon seclusion or into private affairs” branch ([16], para. 21). Much like the American approach to that branch [40], the requirements of the new action in Ontario are now (1) intentional conduct by the defendant, (2) which results in the invasion of the plaintiff’s private affairs without lawful justification, and that (3) a reasonable person would consider the invasion to be highly offensive ([17], para. 71). Could a claim against a commercial street surveillance provider fit within these contours? Given the recent vintage of Jones there is not yet any jurisprudence provided more examples of how the three factors might be met, so it is difficult to know with certainty [41]. In the following section, then, I consider both the jurisprudence under the provincial torts (given the similar subject matter) and that under the American version of the tort (given the explicit way in which the new tort in Jones is modelled upon it) in an attempt to develop a predictive guide as to how each of the three requirements of the new tort might be applied.

3.1. Intentional Conduct by the Defendant

The service providers do not intend to actually record images of people insofar as they do not set out to photograph them—it is merely is a collateral consequence of the broader operation aimed at photographing vast areas of public space. The fact that individuals are captured in the images is neither necessary to the project nor apparently desired by the service providers. However, if the “intent” requirement under the Jones tort develops in the same way as the “wilful” requirement under statutory torts, it is probable that an action against street surveillance providers will meet the necessary standard—the surveillance project as a whole is intentional, and (assuming it can be ultimately proven that the plaintiff’s privacy was invaded) so long as the hypothetical defendant knew or ought to have known that this would violate the plaintiff’s privacy, then this threshold will be met. Indeed, the efforts of the service providers to electronically blur faces (though only effective roughly 90% of the time [42]) are reflective of their acceptance that public photography presents a threat to an individual’s subjective sense of privacy.

Interestingly, a different approach to “intent” in the context of public photography can be found in the decision of Aubry [43], which though decided under civil law principles was cited by the Alberta Court of Queen’s Bench as being “instructive” for recognizing the importance of balancing off “the mutual rights and obligations… protected by the common law and privacy legislation” ([44], para. 22). Aubry claimed her privacy was invaded after a photograph of her sitting on the steps of a building was published in magazine. While the Supreme Court of Canada accepted that this was prima facie a violation of the right to privacy protected under the Quebec Civil Code [45], they also argued that it was a right that “[came] into conflict with freedom of expression, which includes artistic expression” ([43], para. 55). In such cases, according to the majority,

[The right to free expression] prevails when a person appears in an incidental manner in a photograph of a public place—an image taken in a public place can then be regarded as an anonymous element of the scenery, even if it is technically possible to identify individuals in the photograph.

([43], para. 59)

In Aubry, it was clear that the photographer intended to capture the image of the plaintiff specifically and through his technique he ensured that she was the prime subject. This led the Court to find that the invasion of Aubry’s privacy was not justified, and award her $2000 for the loss she had suffered. If the Ontario courts look to Aubry as they develop Ontario’s new “intrusion upon seclusion” tort, they may adopt a similar approach in determining the issue of intent in the context of a claim over public photography [46]. If they do so, it would appear to limit the chance of a successful claim against commercial street surveillance providers, at least in the majority of cases—street surveillance is invariably taken with a wide-angle lens, and individuals are rarely the prime focus of the image initially published by the provider. Of course, thanks to the ability to digitally crop an image, the anonymous viewers can effectively “create” a new image in which a single individual is the prime or only subject, but that cannot be connected to the initial image collection and therefore cannot necessarily be ascribed as the “intent” of the service provider. But this is not definitive, of course, and it is possible that a street surveillance image can be focused on only a single individual, without any subsequent editing. In any event, the greatest challenge under the Jones formulation in the context of public photography relates not to the question of intent, but rather to whether or not a state of seclusion actually exists to be invaded, and if so, whether or not that invasion is “highly offensive”.

3.2. Invasion of Seclusion or Private Affairs without Judicial Authorization

The American jurisprudence indicates that “intrusions” into seclusion are not limited only to the “physical” variety, but can be “physical, electronic or mechanical” ([47], p. 349). This might include the use of parabolic microphones located in public space to record the activities of a couple in their bedroom [48], or wiretapping a telephone conversation even when the “bug” is not located within the target’s private space [49]. However, much like the approach of the Canadian courts under the statutory provincial torts, the American courts have been extremely reluctant to find that any kind of privacy interest has been invaded or intruded upon when the plaintiffs themselves are in public space or otherwise easily viewable by others without special means; in general, one is not in a state of “seclusion” when in public, and therefore there is nothing to be intruded upon. In Gill, the plaintiffs sued a newspaper for intrusion upon seclusion after the paper published a photograph of them in an “affectionate pose” while at an ice-cream stand at the Los Angeles Farmers’ Market ([50], para. 227). By virtue of being in public, said the court, the couple had “voluntarily exposed themselves to public gaze in a pose open to the view of any persons who might then be at or near their place of business… [and thus had] waived their right of privacy” ([50], para. 230).

A similar principle can be gleaned from the jurisprudence that surrounds the third branch of the tort—publicity given to private facts. Though not adopted in Jones, the American approach to it may nonetheless influence the development of the intrusion upon seclusion tort in Ontario, given that it requires the existence of “private facts”, much the way the first branch requires an initial state of “seclusion” or “private affairs”. In the Restatement’s commentary on the third branch, “there is no liability for giving further publicity to what the plaintiff himself leaves open to the public eye” ([40], § 652D, comment b). Thus, a dancer at a strip club whose image was broadcast as part of a news segment on a particularly notorious bar in Tennessee failed in her claim that by revealing to her unknowing friends and family that she was an exotic dancer the broadcast invaded her privacy through the public revelation of private facts [51]. In dismissing her appeal, the Sixth Circuit Court of Appeals did not find it necessary to make recourse to any constitutional arguments regarding freedom of the press and attempt to balance off any competing interests. Instead, the court simply concluded that since the “plaintiff's activities at the club were open to the public, [her] claim for public disclosure of private life was properly dismissed by the district court as a matter of law” ([51], p. 6). The same principle was applied to deny remedy to a plaintiff who sued private investigators for using a hidden GPS device to track his movements by car on the orders of his wife as part of their divorce proceedings [52]. The court concluded that even though the placement of the device was done without the plaintiff’s knowledge, since he had never driven the car to a “secluded” location or one that was out of public view, permanent tracking of his (public) location did not amount to a violation of his privacy—no “private facts” were revealed ([52], p. 1). The general principle in the American jurisprudence regarding privacy torts (broadly, not only the third branch) is that “public matters [cannot] provide a plaintiff with a cause of action” ([53], p. 920). Given the decision in Jones to model the Ontario tort on the American framework, it seems probable (though not definitive) that this may end up the case in Ontario too.

3.3. “Highly Offensive” to the Reasonable Person

In Jones, Sharpe J.A. stated that conduct that would meet the “highly offensive” threshold would be that which the reasonable person could see “causing distress, humiliation, or anguish” ([17], para. 71). He went on to provide a (presumably non-exhaustive) list of examples, including intrusions into “financial or health records, sexual practices and orientation, employment, [or] diary or private correspondence” ([17], para. 72). Jones, of course, was a case regarding an informational invasion rather than a visual one, and this may account for his examples. But the American jurisprudence suggests that in visual cases, plaintiffs will struggle to establish that public photography reaches the highly offensive threshold. Indeed, the sole reported American case involving a lawsuit against Google for invasion of privacy as a result of their street surveillance programmes turned largely on this issue [54,55].

After discovering photos of their residence on Google Street View despite it being at the end of a private, unpaved road, Aaron and Christine Boring filed suit claiming that Google “significantly disregarded their privacy interests” ([54], p. 699). As the Borings failed to state precisely in their brief which privacy interest (under the four-part framework) it was that was violated, the District Court reviewed the possibilities and concluded that only alternatives were “intrusion upon seclusion” or “publicity given to private life” ([54], p. 699). Under both, as now is the case in Ontario, there is a requirement that the intrusion or revelation be “highly offensive” for the claim to be made out. The court concluded that while many people might resent some of the privacy implications of Street View, only “the most exquisitely sensitive... would suffer shame or humiliation” ([54], p. 699). On appeal, the Third Circuit Court of Appeals reached the same determination with regard to both branches of the tort, finding that:

No person of ordinary sensibilities would be shamed, humiliated, or have suffered mentally as the result of a vehicle entering into his or her un-gated driveway and photographing the view from there… the alleged conduct would not be highly offensive to a person of ordinary sensibilities.

([55], p. 3)

The implication is that it is unreasonable to be highly offended by the capture of “everyday” activities that are publicly viewable. Indeed, in other cases under the intrusion upon seclusion branch of the American tort, the spatial question (the location of the surveillance) is often considered not only as relevant to whether there is a state of seclusion that can be invaded, but also as a relevant to the question of “offensiveness”. This effectively privileges the spatial question in the context of photography cases, as it “counts twice” under the legal analysis—once to prove there is no invasion, and then again to prove that if there was in fact an invasion, it is not offensive. Returning to Gill, the California Supreme Court ruled that while the plaintiffs were photographed linked arm-in-arm in a fashion that might be considered an intimate gesture, revealing it in the photograph did not meet the “highly offensive” threshold, as it was the portrayal of “an incident which may be seen almost daily in ordinary life” ([50], p. 231). In so doing, the court appears to at least in part link the location of the incident and its possible offensiveness. While in Jones this issue of spatial location was not addressed since it was a question of “informational” intrusion rather than visual surveillance, given the relative closeness to which Jones already tacks towards the American tort and the importance given to spatial considerations under the prior Canadian privacy-related jurisprudence, it is probable that the questions of spatial location of the plaintiff and the question of offensiveness will ultimately be linked in Ontario too. This would seem to be a fatal blow to the chances of a successful action being brought against a commercial street surveillance provider.

In Jones, much as in Boring, the question of the offensiveness of the intrusion to the “reasonable person” serves to limit scope of the new tort by excluding from its ambit “claims from individuals who are sensitive or unusually concerned about their privacy” ([17], para. 71). But there is necessarily a difficulty in relying upon a “reasonable person” standard when the effects of street surveillance can be experienced so differently by different groups of people; it is doubtful whether a truly objective standard can be developed. I am much less likely to have my image set upon by the anonymous crowd and shared for titillation and pleasure or mockery and shame than I would if I was a young woman or member of a visible minority. Should those who feel dehumanized and objectified by this process therefore be dismissed as “unusually concerned” or “sensitive” about their privacy? I would argue not. But the problem from the perspective of a claim against a commercial service provider is that even if we could agree that most people would find the sexualized commentary or mockery “offensive”, the same cannot necessarily be said for the initial image collection, upon which a successful invasion of privacy action would turn. The standard in Jones and the American jurisprudence is whether the invasion is highly offensive, and if we are considering an action against a service provider, then the invasion in question must be the initial photography. And yet, the photography at issue is by and large incredibly mundane, revealing nothing more than can be gleaned by walking down the street and observing one’s surroundings. As a result, absent a societal belief that any photography in a public place is “highly offensive”, this element of the new Jones tort puts another very high hurdle in front of a successful claim against the service providers for invasion of privacy.

4. The Necessity and Difficulty of Limiting the Scope of Privacy Torts

It seems then that the likelihood of a successful action against commercial street service providers brought under a Jones-style tort is low, thanks to the limits on the scope of the tort created by the court. Austin describes the judicial search for appropriate limits as “containment anxiety”, in which the courts resist extension of tort protections for invasions of privacy out of a fear of unduly restricting the legitimate interests of other parties ([56], p. 3). Such resistance should not be interpreted as a failure on the part of the courts, however: a right to claim an actionable privacy interest over any activity in public would be far too broad. For instance, without limitations to its scope, a tort that captured the operation of street surveillance technologies would also capture an individual tourist taking photographs on a visit to a new city; both are based on the digital “recording” of publicly viewable spaces which incidentally records people. Clearly, this would be a draconian and unacceptable limit on the expressive rights of individuals, and explains the limits seen in this paper placed by various courts when confronted with claims for privacy interests in public space. The question is, then, whether or not it is possible to come up with other limitations than the ones adopted in Jones, such that the tort that could capture the operation of commercial street surveillance technologies, but not the camera-toting tourist. It seems unlikely. For instance, one limitation might be to only allow claims where the photography has been hidden or the subject otherwise did not consent. But adding this element of “surreptitiousness” still fails to resolve the “tourist issue”—tourists generally do not actively advertise they are taking photographs, and thus many individuals captured in their images may be unaware. Likewise, street photography for the purposes of art is common and generally unobjectionable, but often incidentally captures individuals in a streetscape image without their knowledge or consent.

Another alternative might be to acknowledge a reasonable expectation of privacy not over the image capture, but with regards to the kind of distribution of images taken of them in public. For instance, an individual might reasonably expect that only those present in the area can observe their behaviours. Widespread digital distribution changes that calculus, by potentially rendering their activities viewable by billions. A claim for a reasonable expectation of privacy in public centred around the nature of distribution can account for the fact that individuals do accept a lower expectation of privacy when in public but without eliminating that expectation entirely. But unfortunately problems still exist with this conception. Distribution of an image to an audience of millions or even billions is no longer the preserve of large corporations—the digital revolution has put the power of that technology in the hands of anyone with a camera and access to a computer. As a result, we would still find that a “distribution-focused” privacy tort would treat the tourist photographer who subsequently uploaded her images to a photo-sharing website in the same fashion as the commercial operators of street surveillance technologies. It would do the same for those who took photos of a friend’s birthday party in a park and uploaded them to a Facebook account without securing the consent of every person at the party. Once again, we are unsuccessful in being able to draw the boundaries of the tort tight enough to include public street surveillance but exclude behaviours for which we do not wish to attach legal liability.

The problems go beyond protecting the tourist photographer and it easy to think of many instances in which a claim for privacy vis-à-vis being photographed in public will conflict with other no less valid rights. It is not necessary to consider them all here, but the point is simply that this struggle to limit the scope of privacy claims such that other socially-beneficial behaviour does not create valid causes of action has been faced by courts in a variety of jurisdictions. Generally, common law legal systems appear to be coalescing around “offensiveness” as the best option to limit the scope of privacy claims made in public. New Zealand, for instance, has adopted the third branch of the American tort and likewise has adopted the requirements that the matters not already be “public” and that the disclosure be “offensive” [57]. In Australia, the latest lower court decisions appear to have, much like Jones, accepted the first branch of the American framework into the Australian common law, along with the requirement of “high offensiveness” [58]. An outlier remains England, whose courts have largely continued to dismiss [59] the idea of introducing a freestanding privacy tort, suggesting the creation of any such action is better suited to Parliament [60]. Instead, they have been more receptive to extending protections for certain kinds of breaches that would fall under the rubric of “publicity given to private facts” by expanding the action of breach of confidence. However, they have still had to struggle with the best way in which to restrict privacy interests protected by such an action so that the claims do not serve to unduly restrict the legitimate expressive interests of others. One line of cases suggests that the former can only outweigh the latter when revealing “private” information would cause significant harm to the claimant [61]. The European Court of Human Rights has also relied upon “serious harm” as a method of limiting the scope of a claim that can be made by individuals photographed in public space who feel their privacy has been invaded [62]. Though not examples of freestanding privacy torts, Campbell [61] and Peck [62] are also instances in which courts have recognized that individuals can retain some kind of privacy interest while in public or over materials they bring into the public realm. If the Jones line of jurisprudence adopted “significant harm to the plaintiff” as a limiter in place of the intrusion being “highly offensive”, could that allow the successful grounding of an action for invasion of privacy in the context of street surveillance? After all, by focusing on the issue of harm, we might avoid the problem of capturing the actions of the innocent tourist photographer within the boundaries of the tort.

On the one hand, it avoids the problem that street surveillance primarily captures the mundane aspects of life—if the issue is whether or not the intrusion causes serious harm to the subject, then it should not matter whether the intrusion is into mundane affairs or not. We are confronted, however, with the problem that even though the service providers are responsible for the intrusion through unwanted photography and this creates the potential for serious harms to later arise, the former is not the direct cause of the latter. Yet, invasion of privacy torts have always been considered a species of intentional torts—those which require a deliberate act on the part of the defendant, in contrast to negligent behaviour. It is difficult to justify an invasion of privacy action against a service provider under Jones if the jurisprudence were to evolve in a way that looked to the “serious harm” caused by the intrusion, since that harm cannot be directly tied to the actions of the service provider [63]. Thus, even though a tort that evolved in this manner would no longer capture the innocuous behaviour of the camera-toting tourist, likewise it probably no longer captures the behaviour of the commercial service provider either, thus defeating the point.

5. The Nature of the Harm and Private Law Actions Generally

Above, I have suggested that there are particular jurisprudential reasons that an invasion of privacy tort would struggle to account for street surveillance without also capturing innocuous or socially beneficial behaviour. In this section, I want to further suggest that there are also conceptual reasons why an action in tort, no matter how its scope is drawn, may have limited usefulness in the context of commercial street surveillance. My initial claim in this paper was that despite the benign intentions of the service providers, commercial street surveillance subjects women to an intense gaze for the purposes of (apparently) male sexual curiosity or gratification. Likewise, that gaze has also been directed intensely to scrutinize and bring forward for anonymous commentary the everyday activities of the poor or ethnic minorities. Experiencing such unwanted scrutiny, commentary, mockery, and shaming, are perhaps best interpreted as “micro-harms”, which while undeniably problematic for the individuals concerned, nonetheless pale in comparison to the dignitary “macro-harms” suffered by entire socio-economic groups; the latter are more than the sum of the former. When an image of a woman innocently sunbathing with her friends or simply walking down the street in a revealing outfit is surreptitiously taken without her consent and shared widely online, she may legitimately feel embarrassment, anxiety, and distress. But that action also—and perhaps more damagingly—reinforces a set of acceptable behaviours and norms towards women generally; whether or not those women are ever captured by street surveillance. Normative acceptance of selective “image sharing” of (primarily) young women sends the message that it is right that they be treated objects for the sexual gratification of men, that it is normal for their bodies to exist as the subject of debate and discussion, that it is valid a woman going about her everyday business to be forced into a conversation or to otherwise be expected to account for her sexuality. Likewise, propagating a false assumption that a black man photographed talking to a police officer is being arrested is problematic not only for the individual concerned, but also for a much broader community if it feeds into stereotypes about preponderance towards criminality or social deviance. This means that the “harms” associated with commercial street surveillance are suffered by far more than those actually captured in a given image and subject to derision or mockery online.

Though these “macro harms” represent the fundamental problem with unrestricted commercial street surveillance, they do not seem amenable to remedy though actions in tort. As Solove notes, “tort law looks to isolated acts, to particular infringements and wrongs” and is therefore ill suited to remedying such complaints ([64], p. 64). It would surely be challenging to successfully argue before a court that an individual who feels a loss of dignity thanks to being a member of a group subjected to longstanding stereotypes and poor social treatment has suffered an injury that could be connected to an identifiable tortfeasor. Moreover, the diffuse (yet no less serious) harms for those not caught on camera but belonging to socio-economic groups who suffer as result of their individual members being repeatedly “targeted” by anonymous viewers for selective scrutiny simply cannot be addressed by invasion of privacy actions at all. In such cases, the necessary connection between the party responsible for the initial visual intrusion and the individual complaining of an impact on their privacy is entirely missing.

A further hurdle to any kind of private law action against street surveillance providers relates to the issue of remedies. As noted, invasion of privacy torts are intentional torts, and so the issue of remoteness would not be relevant. If causation of the claimed loss could successfully be established as a result of the intentional acts of a defendant service provider, then a damage award is likely to be the remedy. However, the quantum of such an award is likely to be minimal. Under the statutory torts, even the grossest of invasions of privacy (i.e., hidden cameras inside an individual’s bedroom or bathroom) resulted in maximum awards for harm suffered at around $50,000 (Malcolm, for instance, resulted in an award of $15,000 for general damages and $35,000 for punitive damages) [23]. The award in Aubry, for an invasion of privacy via photography in public that revealed no intimate or embarrassing behaviour, was only $2000—an amount described by the Supreme Court of Canada as “high”, though not rising to the level of a reversible error [43]. Given this judicial history, it is safe to assume that future awards for similarly innocuous public photography will be low. If we rely on actions in tort as quasi-regulatory tools, then they must have a reasonable chance of actually altering broader corporate behaviour. Assuming for the sake of argument that all the difficulties I have identified in mounting an action under an invasion of privacy tort could be overcome, it is far from certain that an award of $2000 would be sufficient to alter that behaviour, given the immense resources that corporations like Google and Microsoft can draw upon. It would be surprising if the threat of paying out the occasional five-figure damage award gave them significant pause in the way they operate the programmes [65]. There is, of course, also the possibility of equitable relief being obtained against the company engaged in the surveillance, if a plaintiff could demonstrate that monetary damages were inadequate for some reason, or that irreparable harm would continue. But courts might find it difficult to craft an injunction in a fashion that would not have the effect of requiring street surveillance providers to cease operations in vast geographic areas. Courts may be reluctant to take such a step, given that these are undoubtedly popular technologies and most Canadians probably do not want to see them eliminated entirely.

6. Conclusions

Invasion of privacy actions, whether under one of the existing statutory torts in the provinces that have them or under Ontario’s new common law version, are unlikely to successfully provide a remedy for an individual photographed without their consent by a commercial service provider. In both cases, the spatial question appears to dominate the legal analysis, meaning that once someone is located in public space, they lose any reasonable expectation of privacy in terms of being photographed. Moreover, it seems difficult to draw the boundaries of the tort in a fashion that would capture commercial street surveillance without also attaching liability to socially beneficial behavior. And yet, I’ve argued that these technologies are something quite different than the typical kinds of “street photography” experienced in the past. By creating a virtually-explorable archive of imagery that can be viewed and shared by an anonymous audience of billions, commercial street surveillance fundamentally changes the possible exposure that someone has simply by stepping outside their door. Significantly, this increased exposure is unlikely to be felt equally—the most harmful social effects are likely to be experienced by women and minority groups. This means that while the solution is unlikely to be found in private law actions, some kind of remedy remains necessary. Future work, then, must consider the possibility of developing a regulatory framework designed specifically to govern the use of street surveillance technologies in a manner that limits their potential for creating serious social risks (particularly for women and minority groups), whilst still preserving them as useful technologies for Canadians.