Roger Clarke's Web-Site
© Xamax Consultancy Pty Ltd, 1995-2018
|Identity Matters||Other Topics||Waltzing Matilda||What's New|
Review Version of 12 August 2018
Journal of Information Technology 34,1 (Mar 2019)
This brief paper is an invited Response to four Commentatories on my paper on 'Risks Inherent in the Digital Surveillance Economy: A Research Agenda'
Roger Clarke **
© Xamax Consultancy Pty Ltd, 2018
Available under an AEShareNet licence or a Creative Commons licence.
This document is at http://www.rogerclarke.com/EC/DSER.html
The four Commentaries in part confirm, articulate and elucidate the argument in my article. However, there are also some material differences of opinion or valuation between author and commentators. This Response provides a few clarifications, and grasps a couple of opportunities to further develop the thesis.
I first address three substantive aspects - the significance of GDPR as a regulatory intervention (an element of Scenario 2), the prospects for consumer resistance (Scenario 3), and the application of the techniques and data of the digital surveillance economy (DSE) to the polity as well. I then turn to two methodological issues - the nature of a research agenda, and the need for futurist techniques to be applied to IS research. The final section takes a bolder step. It discusses competing ideologies within the IS discipline and proposes that we should recognise and declare the values that underlie our work.
In Scenario (2), the last 3 of the 7 themes are concerned with the extent to which formal laws may be effective in safeguarding consumer interests. Two commentators considered that the EU's General Data Protection Regulation (GDPR) needed to be specifically addressed.
It's true that many people hold enormously high hopes for the GDPR's impact, both within and beyond Europe. I've conducted research and presented in multiple European venues, specifically in relation to the Data Protection Impact Assessment (DPIA) element (Clarke 2014a, 2017a, 2017b). That research suggests that considerable obstacles will have to be overcome if the GDPR is to result in the reining in of consumer marketers' excesses and the achievement of substantial changes in DSE practices.
European countries have taken a long time to move against exponents of the new business model, and there is no guarantee, even post-GDPR, that the necessary actions will be taken by Data Protection (DP) Commissioners, nor that any such actions will be effective. Almost all of the major players at the heart of the DSE are domiciled in the USA, and there are practical limits to the extra-territorial reach of European domestic laws. In addition, the European Commission (EC) is strongly driven by economic motives. To the extent that DP Commissioners become activist, the EC may take steps to limit the harm that they do to European economic interests - particularly if the Trump Administration's trade warfare is prolonged or extended.
Prof. Cecez-Kecmanovic noted the lack of serious consumer resistance to date. The muted response reflects a point made in s.5.1 of my paper: "Of the 12 activities in the model of the digital surveillance economy ..., at least 9 are conducted out of consumers' sight [and] many people have very little understanding of the nature of the digital surveillance economy". There could be real value in the proposal to apply the ideas of Latour and Zuboff to refer to 'civilisation-scale' 'anti-programs'. Three areas are addressed in my article (Privacy Enhancing technologies, consumer activism, and new laws and regulations). Prof. Cecez-Kezmanovic draws attention to the possibility that 'fee for service' business models might yet emerge, particularly when some future new wave of market players seeks to re-define the marketspace.
Prof. Leidner proposed that the attributes of the digital persona deserve more serious research attention. She also exhorted individuals to not resign themselves to passive roles, but instead empower themselves. Self-empowerment is challenging, however, because, in the DSE, digital personae are imposed on consumers (Clarke 2014b). As a result, consumers have limited ability to influence what those personae contain, and how they are applied. A further concern is whether the IS discipline's gatekeepers will even be tolerant, let alone supportive, of papers on digital activism generally, and on means for consumers to effectively collectivise in order to achieve countervailing power against marketing corporations.
Prof. March, on the other hand, was emphatic that "Consumers certainly have the freedom to decide ... consumers can take defensive actions ... consumers have a choice in determining how they wish to participate or not participate in this economy". Such contentions fly in the face of the evidence - even that within his own Commentary. Substantial asymmetries exist in relation to information and power, as discussed in my article in the segments on ad targeting, behaviour manipulation, micro-pricing, and the conduct of 9 of the DSE's 12 activities out of consumers' sight. As I was drafting this segment of text, yet another media article on the theme appeared, with the tag-line "Australians can't tell if they're making a fair trade with the data they're giving up" (Biggs & Hatch 2018).
The significance of social media in this context is commonly over-estimated. It supports brief spurts of outrage. On the other hand, it's of limited use to activists who need to convey complex messages and to mobilise supporters over an extended period of time, in order to force changes in organisational practices, technical infrastructure and even societal values.
The main thrust of my article is concerned with economic aspects of the phenomenon, extending out to some of its social implications. The article does, however, touch on the possible emergence of a "high-tech corporatised-government State" and "corpocracy". Towards the end of the article's development period, data arising from the DSE was applied to political aims, and the article accordingly noted public concerns about Cambridge Analytica's behavioural profiling based on dubiously legal access to Facebook data. Those concerns have since broadened, via the 'Russian interference' meme, into a national sovereignty issue. It may therefore be helpful to postulate 'the digital surveillance polity' as a bridge between the topic of this paper and the broader aspects of Zuboff's 'surveillance capitalism'.
In addition to substantive matters, the commentators identified a number of issues of a methodological nature. Firstly, Prof. March suggested that "It is not clear that [what the article proposes] is a research objective or a research agenda. It seems more a political, social or personal agenda".
Although the term 'research agenda' is widely-used in articles within the IS discipline, the concept has an unsatisfactory theoretical and methodological base. Even articles published very recently leap directly to exposition of the authors' proposals, without clarification of a research agenda's desirable characteristics. My paper cited the most useful sources that I had been able to locate, resulting in a working definition of a research agenda as the combination of:
all carried forward into the instrumentalist portion of the work, comprising:
Further, a 'research program' was characterised as "a framework within which large numbers of disparate projects adopt disparate approaches to disparate research questions, utilising bodies of theory and techniques from disparate disciplines, and achieving a degree of cohesion through inter- and multi-disciplinary elements of the program".
It would have been valuable if Prof. March's criticism had been supported by detailed engagement with the analysis that I presented. It is only through such processes that we will accumulate an adequate IS literature on what constitutes a research agenda.
Two of the commentators were dissatisfied with the form of the questions that accompanied the three Scenarios. The article describes these as "broad research questions that reflect alternative paths that society might trace", and "speculative questions that appear likely to require investigation [as a result of themes that will emerge from well-developed scenarios]". They are, in short, intended to be precisely as Prof. Böhme described them: "very broad, partly ill-defined, partly ambiguous". Their function is not to serve as testable hypotheses, and there is no need for them to be framed as falsifiable propositions.
On the other hand, Professor Leidner identified the advantages of framing even 'speculative questions' in a more open-ended manner than those in my article. This is more likely to lead to understanding, and toward measures that may be instrumental in achieving alternative futures.
Prof. Böhme also suggested that I was unaware of, or disregarded, existing work that tackles and partially answers more precisely posed forms of some of my general questions. The final para. of s.6.3 of my article talks of "[needing] to establish what existing research already tells us", and concludes that "The [research] program can utilise the theories, insights and research techniques of many different disciplines". Indicative examples of relevant research were provided. However, I accept that my article could indeed be seen as under-playing the extent to which relevant prior research exists, and in some areas may even have established a firm foundation for progress to be achieved.
Prof. March considered that many questions can be answered by means of a multitude of research methods, most obviously those used in marketing research. The article acknowledges that to be the case: "The [research] program can utilise the theories, insights and research techniques of many different disciplines". Ten indicative examples are provided, including "marketing research".
Further, Prof. March queried what phenomena must be studied using "future-oriented, instrumentalist research", given that "the activities and data flows depicted in [my Figure 1] are extant". My presentation of this aspect was less clear than I had imagined.
In s.2, I describe the digital surveillance economy as featuring "emergent phenomena, such as those that arise from the application of multiple new digital technologies", and argue that the need is for "future-oriented, instrumentalist research". The research agenda is concerned not merely with current instantiations of technology, architecture, infrastructure and practice, but with all relevant present and future activities. All aspects are of course likely to change over time. In order to provide a frame, however, a proxy model was proposed, represented graphically in Figure 1. It drew on and generalised observable activities and data flows: "The broad conceptual framework was articulated by means of a model of the key processes within the digital surveillance economy. This was developed from observations of consumer marketing activities, primarily in contemporary Web-based eCommerce and mobile commerce". Put another way, the particular model discussed in the article is not the final word, but is provisional: "The process model provided a basis for identifying the impacts of the new ways of doing business. It is also valuable as a basis for tests of the conceptual framework against continually changing real-world instances".
Prof. Böhme questioned what he perceived to be a claim that Òempirical research is uselessÓ. The relevant sentence, which is in the final paragraph of my article reads in full: "Conventional, rigorous, empirical research is useless in the face of such a challenge" (emphases added). The challenge that I had in mind is explained in s.2, which I acknowledge is far too distant from the sentence in question. I've adjusted the expression of the first element in order to reflect the clarification I provided in the immediately preceding paragraph. The challenge derives from the combination of the following characteristics of the digital surveillance economy:
The IS discipline has to date shied away from future-oriented research. My article suggests both that this reticence is unnecessary, and that it results in missed opportunities and undue constraints on the discipline's scope. Future-oriented research can be undertaken in a sufficiently careful manner that it can deliver value to decision-makers in business contexts, and also to policy-makers and civil society.
My article could be reasonably criticised for being too timid. It fails to directly engage 'an elephant in the room', and 'a debate that we ought to have'.
By substituting the perspective of consumers for that of consumer marketing corporations, I'm arguably issuing a challenge to established norms in the discipline. Perhaps I should enlarge the debate, and argue that the IS discipline needs to confront the issue of competing ideologies. I'm using the word 'ideology' in the non-pejorative sense conveyed by meaning 4 of the OED entry: "A systematic scheme of ideas, usually relating to politics, economics, or society and forming the basis of action or policy; a set of beliefs governing conduct".
One way in which discussion of ideologies has made its way into the IS discipline is through the notion of 'unconscious hegemonic participation' (Wall et al. 2015). The authors argue that each IS researcher, like any human being, has deeply-rooted assumptions that influence their behaviour (and, further, that some assumptions tend to dominate). The authors wrote cautiously, referring only to the unconscious form. It's important to recognise that both unconscious and conscious forms of hegemonic participation exist: many researchers are well aware of (some of) their deep-rooted assumptions, and some actively espouse them as values. Importantly, however, the value-sets subscribed to by researchers are diverse. To provide an indication of where awareness of competing ideologies might lead us, I'll outline four examples.
Firstly, a researcher working within the positivist tradition gains advantages, but self-imposes limitations. For example, Prof. March's Commentary included the statement that "research is intended to reveal, discover or produce truth" (emphasis added). The assumptions implicit in ideologies will lead some to read this statement as implying a singular truth. For others, there may be multiple perceived truths, and there may be conflicts among the interests of different stakeholders. Making explicit an ideological stance of this nature enables subsequent discussions to avoid diversions into superficial or derivative issues and to instead focus on underlying differences in approach.
A second example arises where devotees of free market thinking deprecate the notions of information and power asymmetry, and ascribe capabilities to consumers that they patently do not have. Similarly, a committed, say, socialist perspective also carries intellectual baggage. It's difficult for debate to be usefully engaged between competing viewpoints if either side treats their own views as 'going without saying', or the other side's views as being 'beyond the pale'. Worldviews that lack tolerance and are antithetical to alternative value-sets stifle academic debate. On the other hand, if the protagonists' ideologies are acknowledged, it becomes easier to iterate towards what it is that the two sides 'agree to disagree' about.
As a third example, I need to acknowledge the somewhat different ideological background that underlies my paper. As Profs. March and Leidner noted, my paper uses 'surveillance economy' in an overwhelmingly negative way, valuing individualism very highly and investigating risks to individual values, while deprecating corporate values and downplaying benefits to marketers and risks to society. I see my single-mindedness as being justified by the dominance of corporate values within the IS discipline, and the marginalisation and even exclusion of other value-sets. But, like every other IS researcher, I should recognise the blinkers that I've worn while conducting the research, and declare them to my readers.
My fourth example is the tendency among many business school academics to conceive their research in ways consistent with strong commitment to their clients' interests and a dismissive attitude to the interests of other stakeholders. This can be apparent not only when undertaking consultancies and commissioned projects, but even when performing academic research. This issue needs to be recognised and confronted.
In a series of recent papers, I've argued that a pair of assumptions has become engrained in the IS discipline (Clarke 2015, Clarke 2016, Clarke & Davison 2018). The first assumption is that it's appropriate to adopt a singular 'researcher perspective'. (By that term, I mean the viewpoint of a particular stakeholder in the relevant domain, which is adopted by a researcher as the viewpoint from which to observe phenomena during the conduct of a research project). The choice of researcher perspective strongly influences the conception, design, conduct and findings of the research, with that party's interests essentially being perceived as objectives, and the interests of other stakeholders treated as constraints on the chosen party's freedom of action. The second assumption is that the singular perspective that's generally appropriate to adopt is that of the system sponsor.
Single-perspective research is clearly problematical for all stakeholders other than the one chosen by the researcher. Moreover, the dominance of single-perspective research denies the IS discipline the scope to undertake research relevant to public policy questions. By public policy, I mean activities that "deal with matters of national or regional significance and involve enunciating objectives and principles; formulating steps to achieve them through legal and other regulatory measures; allocating resources; and conceiving, articulating, and implementing programs of action. Objectives are not just economic but social and environmental. Fulfilling policy objectives depends on cooperation among disparate organizations that often come from all sectors (public, private and voluntary)" (Niederman et al. 2017). In such contexts, conflict among stakeholder value-sets and objectives are the norm, not the exception. Research can only support public policy activities if it internalises, appreciates, juxtaposes and actively seeks balances among, diverse values and objectives.
Even within a narrow business school frame of reference, however, adoption solely of the system sponsor's perspective may inevitably result in poor client service. Such research fails to glean much information about the capabilities and likely behaviour of other stakeholders - not even of those that have sufficient power in the relevant context to act as adversaries to the system sponsor's interests. Too tight a focus on the system sponsor therefore fails to support the discovery of strategies that can deliver 'win-win' and 'win-win-win' outcomes. To justify a claim to be an academic discipline, IS research needs to be more ideologically pluralist, and to internalise, and manage, the resulting tensions and conflicts.
An implication of this line of argument is that considerable advantages would accrue to all parties if researchers, rather than adopting a singular perspective, reflected the interests of more than one party. This would enable a more balanced framework of benefits, costs and risks to be adopted. In the context addressed by my paper, the interests of both consumer marketing corporations and consumers can be regarded as objectives, and solutions can be sought that accommodate both. My article falls well short of the lofty ambitions articulated in this section. On the other hand, it may serve to balance the scales a little, and hence to lay a foundation for dual-perspective IS research.
The word 'ideology', when used to refer to academics' weltanschauung, often raises hackles rather than stimulating useful debate. The IS discipline's meta-discussions need to mature beyond the stage of each of us thinking that 'they' have an ideology, whereas 'we' have a coherent frame of reference. That will enable our debates to focus on the underlying points of difference, and our research to contribute more effectively to the understanding and management of real-world applications and implications of data and of information technology.
Biggs T. & Hatch P. (2018) 'Over a barrel: We are clueless about how much data we're giving up' Fairfax Media, 16 July 2018, at https://www.smh.com.au/technology/over-a-barrel-we-are-clueless-about-how-much-data-we-re-giving-up-20180713-p4zrd2.html
Clarke R. (2014a) 'Approaches to Impact Assessment' Xamax Consultancy Pty Ltd, For a Panel at CPDP'14, Brussels, 22 January 2014, at http://www.rogerclarke.com/SOS/IA-1401.html
Clarke R. (2014b) 'Promise Unfulfilled: The Digital Persona Concept, Two Decades Later' Information Technology & People 27, 2 (Jun 2014) 182-207, PrePrint at http://www.rogerclarke.com/ID/DP12.html
Clarke, R. (2015) 'Not Only Horses Wear Blinkers: The Missing Perspectives in IS Research' Keynote Presentation at Austral. Conf. Infor. Syst., December 2015, at https://arxiv.org/pdf/1611.04059, PrePrint at http://www.rogerclarke.com/SOS/ACIS15.html
Clarke, R. (2016) 'An Empirical Assessment of Researcher Perspectives' Proc. 29th Bled eConf., Slovenia, June 2016, PrePrint at http://www.rogerclarke.com/SOS/BledP.html
Clarke R. (2017a) 'The Distinction between a PIA and a Data Protection Impact Assessment (DPIA) under the EU GDPR' For a Panel at CPDP'17, Brussels, 27 January 2017, at http://www.rogerclarke.com/DV/PIAvsDPIA.html
Clarke R. (2017b) 'Big Data Prophylactics' Chapter 1 in Lehmann A., Whitehouse D., Fischer-Huebner S., Fritsch L. & Raab C. (eds.) 'Privacy and Identity Management. Facing up to Next Steps' Springer, 2017, pp. 3-14, PrePrint at http://www.rogerclarke.com/DV/BDP.html
Clarke R. & Davison R.M. (2018) 'Through Whose Eyes Are You Observing the Phenomena? The Critical Yet Latent Concept of Researcher Perspective' Working Paper, Xamax Consultancy Pty Ltd, August 2018, at http://www.rogerclarke.com/SOS/RP8.html
Niederman F., Clarke R., Applegate L., King J.L., Beck R. & Majchrzak A. (2017) 'IS Research and Policy: Notes From the 2015 ICIS Senior Scholar's Forum' Commun. AIS 40 (2017) Article 5
Wall J.D., Stahl B.C. & Salam A.F. (2015) 'Critical Discourse Analysis as a Review Methodology: An Empirical Example' Commun. AIS 37, 11 (2015) 257-285
Roger Clarke is Principal of Xamax Consultancy Pty Ltd, Canberra. He is also a Visiting Professor in Cyberspace Law & Policy at the University of N.S.W., and a Visiting Professor in the Research School of Computer Science at the Australian National University.
The content and infrastructure for these community service pages are provided by Roger Clarke through his consultancy company, Xamax.
From the site's beginnings in August 1994 until February 2009, the infrastructure was provided by the Australian National University. During that time, the site accumulated close to 30 million hits. It passed 50 million in early 2015.
Sponsored by Bunhybee Grasslands, the extended Clarke Family, Knights of the Spatchcock and their drummer
Xamax Consultancy Pty Ltd
ACN: 002 360 456
78 Sidaway St, Chapman ACT 2611 AUSTRALIA
Tel: +61 2 6288 6916
Created: 12 August 2018 - Last Amended: 12 August 2018 by Roger Clarke - Site Last Verified: 15 February 2009
This document is at www.rogerclarke.com/EC/DSER.html