1

Crowdsourced counter-surveillance: Examining the subversion of random breath testing stations by social media facilitated crowdsourcing

Mark Wood, The University of Melbourne

E: woo@student.unimelb.edu Presented at Rethinking Cybercrime 2016: UCLAN Cybercrime Research Unit

Mon 27th June 2016 Abstract

In this paper, I examine a social media facilitated phenomenon that may be termed crowdsourced counter-surveillance: the use of surveillance intelligence obtained by crowdsourced labour to subvert police operations. One of the most prevalent examples of this form of counter-surveillance are Facebook pages dedicated to revealing in real-time the locations of road-related police operations, such as speed camera traps and random breath testing stations. Given the illegality of the driving offences these pages facilitate, this use of crowdsourced counter-surveillance can be considered a form of what has recently been termed crime-sourcing: the employment of crowdsourcing techniques by distributed networks of criminals to orchestrate and undertake criminalized acts. Through engaging with crowdsourcing, surveillance studies, visibilities, and new media theory, this paper explores the socio-technical underpinnings of this new form of counter-surveillance and the potential for social media to broker collectives of strangers united in the goal of undermining police operations. Drawing specifically on Andrea Brighenti’s work concerning visibility, I argue that the affordances of social media platforms represent a potential threat to the efficacy of random breath test stations through altering the ‘regimes of visibility’ that enable this form of police operation to function. Introduction

In this presentation I’m going to discuss the emergence of a new social media-facilitated surveillance countermeasure that has been employed to avoid detection by traffic surveillance including speed cameras, red light cameras, and random alcohol and drug testing stations, know as random breath testing stations in Australia. Random breath test and speed trap pages are user-generated Facebook fan pages dedicated to providing their users real-time updates of the locations of forms of police surveillance to subscribers often numbering in the tens of thousands. In this presentation, I’ll trace the implications of these pages for random breath testing as a surveillance tool, and argue that RBT pages on Facebook exemplify a new form of social media facilitated counter-surveillance we term crowdsourced counter-surveillance.

Surveillance studies have tended to view social media as one of three things: (1) as platforms where institutions can surveil users through automated data collection, storage, and mining (see Fuchs 2011; Werbin 2011; van Dijck 2014; Trottier 2012; 2015), (2) as platforms where users can surveil one another (Trottier 2012), or (3) as platforms for distributing sousveillance footage/citizen journalism of misconduct (Albrechtslund 2008; Marwick 2012; Trottier 2012). Rarely have surveillance studies scholars examined the affordances social media provide for

2

enabling users to collectively coordinate and undertake surveillance or countersurveillance activities. The closest scholars have come to this issue are recent studies of Internet vigilantism or ‘digilantism’ where social media users, sometimes of their own accord and other times at the behest of law enforcement agencies, collectively attempt to identify perpetrators who have been recorded by other (visual) surveillance technologies (Graham, 2013; Nhan et al. in press). In his analysis of the remediation of CCTV footage online for investigative purposes, Trottier (2014) dubs this practice crowdsourced surveillance. In this article, we examine the inverse of this process: crowdsourced counter-surveillance.

In our research, we uncovered sixteen pages dedicated to publishing the location of police traffic surveillance measures in regions of Australia (see table 1). This list is, of course, far from exhaustive; there are undoubtedly many other smaller public pages dedicated to smaller regions in Australia, as well as many private ‘closed groups’ which can only be accessed by members. Further, pages dedicated to publishing the locations of portable traffic surveillance are not limited to Australia; in a relatively cursory search we encountered several established within the United Kingdom and the United States. Most of the pages we encountered had other ten thousand subscribers, though several had over 100,000. Of the pages we encountered, the earliest dated back to 2010, with the majority being founded in 2014-2015. Pagename: Region Likes FoundedPoliceintheArea Australiawide 180,294 23/08/12PerthWARevenueRaisersAlert Perth 111,700 30/05/11MelbourneBoozebusAndSpeedCameraLocations Melbourne 111,655 31/10/11PerthRBT&BoozeBusLocations Perth 71,767 27/05/13QLDSpeedCameraLocations Queensland 64,899 23/05/11AdelaidePoliceLocations Adelaide 50,387 6/9/12VictorianBoozeandDrugBuses Victoria 40,596 4/07/10PoliceIntheArea Australiawide 28,038 29/10/12WAspeedcameralocations WA 26,126 16/05/11PoliceinWerribeeandsurroundingareas Werribee 22,274 3/05/13RBT/CameraLocations Canberra 13,021 19/06/11RockhamptonandCQSpeedcameralocations Rockhampton 10,830 11/02/13NSWRBTLocationsLive NSW 5,264 14/05/14RBTBoozebus&SpeedCameraDriverAlertsVictoria Victoria 3,736 18/11/13RBTandSpeedCameralocationsinSydney Sydney 3,701 22/09/12TassieSpeedCameraAlerts Tasmania 3,005 30/01/13

Table 1. List of Australian speed camera/RBT location pages

Of course, these user-generated Facebook pages are not the only new media platforms dedicated to alerting individuals to the presence of police traffic surveillance. As I will briefly discuss later, smartphone apps providing real time updates on the location of portable traffic surveillance are also available, such as Trapster, DUIBlock (California), and Melbourne Speed

3

Camera Alert (Melbourne). These apps, as I will show, have more sophisticated capabilities for rendering traffic surveillance visible en masse.

Crowdsourced counter surveillance

At their core, RBT pages represent a form of counter-surveillance. Counter-surveillance has been defined by Monahan (2006: 516) as ‘intentional, tactical uses, or disruptions of surveillance technologies to challenge institutional power asymmetries.’ Further, as Marx (2003: 384) proposes, counter-surveillance entails ‘turning the tables and surveilling those who are doing the surveillance.’ The overwhelming majority of research concerning social media facilitated counter-surveillance has related to instances Mann et al. (2003) term sousveillance: the use of wearable or portable image-capture technologies for surveillance purposes (see Goldsmith 2010; Wilson & Serisier 2010; Wall & Linnemann 2014; Brucato 2015). When such portable devices are used to record authority figures or instruments of social control, sousveillance constitutes inverse surveillance.

Trapster and Facebook RBT pages represent a form of counter-surveillance that differs from the sousveillance or inverse surveillance proposed by Mann et al. (2003). Indeed, speed trap and RBT pages involve a wholly different form of counter-surveillance that may be termed crowdsourced counter-surveillance: the use of surveillance information obtained from a networked public of crowdsourced labour to engage in the inverse surveillance of law enforcement officials and technologies. The concept of crowdsourcing was first defined by Howe (2006):

Simply defined, crowdsourcing represents the act of a company or institution taking a function once performed by employees and outsourcing it to an undefined (and generally large) network of people in the form of an open call. This can take the form of peer-production (when the job is performed collaboratively), but it is also often undertaken by sole individuals. The crucial prerequisite is the use of the open call format and the large network of potential labourers.

Crowdsourced counter-surveillance is, by both definition and necessity, a collaborative undertaking; its effectiveness rests upon the involvement of a pool of motivated and dispersed contributors who, together, can provide a comprehensive map of police presence within an area. As the surveillance data provided by each contributor is invariably limited, the fewer contributors a page has, the greater the probability of gaps in its intelligence. In addition to a reliance upon a pool of motivated and dispersed contributors, a further precondition of this new form of counter-surveillance are technologies that enable the collation and communication of contributions. Crowdsourced counter-surveillance is, therefore, a distinct product of the web 2.0 and the collection of participatory social media domains that have emerged during this era of the Internet (see O’Reilly, 2007). Indeed, it is the possibility of collecting and communicating knowledge to large audiences, that separates crowdsourced counter-surveillance from previous forms of counter-surveillance deployed against social control institutions and agents. For whilst individuals have long used phone and other electronic communications technologies to warn others of the location of speed traps and other police measures the ability to disperse such intelligence via such

4

ICT’s was limited and typically restricted to interpersonal networks.

As Trottier (2014) notes, surveillance is ordinarily a top-down process undertaken by a

small population of specially designated operatives for a similarly small audience. The crowdsourcing of counter-surveillance inverts not only the typically top-down nature of surveillance, but also its intended audience. Where surveillance data is typically accumulated for a small but powerful audience of formal social control agents or corporate institutions, the crowdsourced counter-surveillance of RBT pages is conducted by the crowd, and for the crowd. In this way, RBT pages ostensibly resemble the synopticon proposed by Mathieson (1997), where rather than the few watching the many, the many (page subscribers) watch the few (surveillance agents). However, RBT pages depart from Mathieson’s model in a central respect: whilst synopticism is predicated upon a mass media that broadcasts its content – and therefore presupposes a non-fragmented audience - RBT pages narrowcast their content: that is they disseminate their content to a narrow audience, rather than the public at large. Not unlike that of Neighborhood Watch programs, crowdsourced counter-surveillance relies on ongoing voluntary action. Yet whilst Neighborhood Watch programs are predicated on geographically bounded communities, the RBT pages we examined were not. Further, whilst Neighborhood Watch programs are community based initiatives, the collectives that use and contribute to RBT pages are in this sense less a community than a public: a collective of strangers.

Surveillance and regimes of visibility

One productive theoretical lens through which to view RBT pages, is visibilities theory and specifically, the work of Andre Brighenti. In his work on the nexus between surveillance and visibility, Brighenti notes that ‘surveillance is a procedure for visibilising certain subjects and certain sites’ (Brighenti 2010, p.151), and moreover that ‘the management of visibilities lies at the core of all forms of social control, whether formal or informal’ (Brighenti 2010, p.148). In viewing the pre-conditions underlying the function of random breath testing as a social control method, it is clear that the management of visbility is imperative to maintaining the efficacy of this surveillance measure. As Homel et al. (1997) write, ‘the defining feature of RBT is that any motorist at any time may be required to submit to a preliminary breath test, and there is nothing he or she can do to influence the chances of being tested.’ The power of moveable speed cameras as surveillance devices is derived primarily from their invisibility. Without it, motorists are able to avoid detection through taking alternative routes. To use Brighenti’s (2010) term, random breath testing stations operate through a very particular ‘visibility regime’: that is, a systematic, routine and strategic set of practices that maintain the visibility or invisibility of particular phenomena for particular ends. To function as a general deterrent, random breath testing stations must be heavily mediated, yet geographically invisible. In other words, what must remain visible is the mediated sight of random breath-tests but not the site of random breath-tests.

Through mapping the locations of portable speed cameras and random breath testing stations, Facebook RBT pages subvert the carefully managed regimes of visibility that optimize

5

these surveillance measures. If, as Homel et al. (1997) posit, the success of RBT is contingent on the potential for any driver to be tested at any time, then the proliferation of new media domains dedicated to publishing their locations represents a threat to the effectiveness of this social control mechanism. RBT and speed trap pages, in short, remove the very conditions that underlie the success of the programs they target. Whilst to use Foucault’s (1979, p.200) phrase, ‘visibility is a trap,’ this goes for both speed traps and their targets: the more visible speed traps and RBT stations are in situ, the less effective they become as surveillance tools. Crowdmapping as a surveillance counter-measure

The pages we examined used three methods to convey the locations of RBT stations and speed cameras to their users. First, all pages informed their users through short entirely textual posts stating the location of an RBT station or speed camera. Second, several pages collated the crowdsourced information they received into tables detailing the locations of all RBT stations and speed cameras within an area on a particular day. Finally, a number of pages supplemented a number of their textual posts with screen shots taken of Google Maps pinpointing the exact location of a RBT station or speed camera (see figures 1 and 2).

Figure 1. Mapping RBTs: An update on Perth WA Revenue Raisers pinpointing the

location of a stationary RBT using Google maps: https://www.facebook.com/perthwarra/photos/a.172073389519507.44042.171302139596632/1045016672225170/?type=3&theater

6

Figure 2. Mapping speed cameras: An update on Block their Shot pinpointing the

location of a mobile speed camera using Google maps: https://www.facebook.com/www.blockthiershot.come/timeline

Through visualizing the locational data they receive from their users, these pages engage

in a rudimentary form of of crowdmapping. Crowdmapping has been defined by Qauintance (2011) as ‘the aggregation of crowd-generated inputs such as text messages and social media feeds with geographic data to provide real-time, interactive information on events.’ Though primarily associated with the mapping of disaster zones– a phenomenon that has been been specifically termed ‘crisis mapping’ (Poblet 2013) – crowdmapping has also been utilised to map state crimes (Williams 2013), and elections (Lokot 2015).

This mapping of speed cameras, RBT stations and other police surveillance measures is, of course, little akin to the advanced forms of crowdmapping enabled through software such as the original Crowdmap Application. Facebook’s user-generated fan pages provide limited affordances for undertaking such crowdmapping. Page administrators must repurpose another site, Google maps, in order to generate this content, and are unable to collate all the locational data they receive into a single map of an area that can be updated in real time. However, recently, applications with considerably more sophisticated crowdmapping capabilities have been developed to fill this void. One such example is the Melbourne Speed Camera Alert (see figure 3), a smartphone app available for download from Google’s app store that provides real-time updates on the locations of speed and red light cameras in the city. Co-opting Google Maps, Melbourne Speed Camera Alert’s interface pinpoints the locations of hundreds of speed cameras identified by users on a map of the city.

7

Figure 2. Crowdmapping speed cameras on Melbourne Speed Camera Alert: The

interface of the Melbourne Speed Camera Alert application, which provides real-time updates of speed cameras and red light cameras operating in Melbourne, and the ability for users to report and add new cameras to the map: https://www.facebook.com/MelbourneSpeedCameraAlert/photos/pb.383252001784596.2207520000.1466311720./513221775454284/?type=3&theater

8

Bibliography Albrechtslund A (2008) Online social networking as participatory surveillance. First Monday 13(3) Brighenti AM (2010) Visibility in Social Theory and Social Research. Palgrave Macmillan: New York Brucato B (2015) Policing Made Visible: Mobile Technologies and the Importance of Point of

View. Surveillance & Society 13(3): 455-473 Foucault M (1977), Discipline and Punish: The Birth of the Prison, London, UK: Penguin Books Fuchs C (2011) Web 2.0, Prosumption, and Surveillance. Surveillance & Society 8(3): 288-309 Goldsmith AJ (2010) Policing’s New Visibility. British Journal of Criminology 50(5): 914-934 Graham G (2013) Exploring imaginative futures writing through the fictional prototype ‘crime-

sourcing.’ Futures 50: 94-100 Homel R, Mackay PM, and Henstridge J (1997) The long-term effects of random breath testing in four

Australian states: a time series analysis. Federal Office of Road Safety Howe, J., (2006), The Rise of Crowdsourcing. Wired, 1 Jun. Available at:

http://www.wired.com/2006/06/crowds/ Lokot T (2015) Ukrainian crowdmapping of the ’12 Elections. Civic Media Project: The MIT Press,

Mar.9. Available at: http://civicmediaproject.org/works/civic-media-project/ukranianelections.12

Mann S, Nolan J and Wellman B (2003) Sousveillance: Inventing and Using Wearable

Computing Devices for Data Collection. Surveillance & Society 1(3): 331-355 Marwick AE (2012) The Public Domain: Surveillance in Everyday Life. Surveillance & Society 9(4):

378-393 Mathiesen T (1997) The Viewer Society: Michel Foucault’s Panopticon Revisited. Theoretical

Criminology 1(2): 215-234 Monahan T. (2006) Counter-surveillance as Political Intervention. Social Semiotics 16(4): 515-534 Nhan J, Huey L and Broll R (in press) Digilantism: An analysis of crowdsourcing and the Boston

Marathon Bombings. British Journal of Criminology, 1-21 O’Reilly T (2007) What is Web 2.0: Design patterns and business models for the next generation

of software. Communications & Strategies 1: 17-37 Poblet M (2013) Visualizing the law: Crisis mapping as an open tool for legal practice. Journal of

Open Access to Law 1(1): 1-20 van Dijck J (2014) Datafication, dataism and dataveillance: Big Data between scientific paradigm

and ideology, Surveillance & Society 12(2): 197-208

9

Trottier D (2012) Social Media as Surveillance: Rethinking Visibility in a Converging World. Surrey, UK: Ashgate

Trottier D (2014) Crowdsourcing CCTV surveillance on the Internet. Information, Communication

& Society 17(5): 609-626 Trottier D. (2015) Coming to terms with social media monitoring: Uptake and early assessment.

Crime Media Culture 11(3): 317-333 Wall T and Linnemann T (2014) Staring Down the State: Police Power, Visual Economies, and

the “War on Cameras.” Crime Media Culture 10(2): 133-149 Werbin KC (2011) Spookipedia: intelligence, social media and biopolitics. Media, Culture & Society

33(8): 1254-1265 Wilson D and Serisier T (2010) Video Activism and the Ambiguities of Counter-Surveillance.

Surveillance and Society 8(2): 166-180