Abstract
Scientific and technological demonstrations are usually used to create credibility for scientific claims or to demonstrate the utility of technical devices. However, they can also function as dramatic instances of boundary work. Drawing on an ethnographic study of the development of an automated video surveillance system, I show how a government-funded, transdisciplinary group of researchers used theatrical practices when communicating to the funding institution, to stage their work as applicable. Their ‘technoscientific drama’ did not primarily produce credibility for their surveillance system’s utility, but more powerfully established the researchers’ credibility as ‘scientist-entrepreneurs’.
Introduction
Since the 1980s, there has been a shift in government commitments towards emphasizing applicability in federal funding arrangements for science. Germany is no exception, although changes in federal research policy arrived about one decade later there than in the United States. Since the mid-1990s, the German government has increasingly directed national spending for research towards economic development by pressing for its applicability to national problems and the commercialization of results. Such directed research funding is exemplified by the government’s creation of specialized funding schemes dedicated to security, health, energy and climate, transportation and communication. These government efforts to mobilize university research for economic growth have increased pressure on university researchers to turn their research into commodities.
Strategic funding schemes have been discussed in the literature as a symptom of ‘radical’ changes in the organization and control of academic research, as well as the very nature of knowledge production (cf. Gibbons et al., 1994; Nowotny et al., 2001; Slaughter and Rhoades, 2004; Ziman, 2000). Theories of radical changes in knowledge production have received their fair share of criticism, especially for their sweeping generalizations and erroneous claims about the history of science (e.g. Elzinga, 2004; Hessels and Van Lente, 2008; Weingart, 1997). Nonetheless, the theories do seem to point to important developments in science policy. In Germany, for example, demands for applicable research are now widely formalized in research programmes funded from the federal budget. However, it is less evident that scientists abide by such political demands for useful technologies. Academic institutions, such as disciplines, tend to be quite stable (Krücken, 2003; Shinn, 1999) and are not always geared towards solving national problems or producing useful commodities. Although scientists working in directed research projects have to anticipate demands for applicability if they want to obtain funding, it seems unlikely that they will give up their commitment to their academic disciplines. Thus, these researchers may have to navigate disciplinary, political and economic expectations, which often might be at odds with each other (cf. Owen-Smith and Powell, 2002; Tuunainen, 2005b).
Drawing on an ethnography of the development of an automated video surveillance system, this article analyses how scientists working in strategic funding schemes navigate conflicting commitments to their careers, their departments and the government commission. Adopting a dramaturgical framework (Goffman, 1959), I show how shifting back and forth between social worlds was one major strategy to performatively and temporarily align what they experienced as competing demands. While the researchers’ everyday work at times diverged significantly from what funders commissioned them to do and did not yield a marketable surveillance system, their demonstrations and presentations served to present themselves as entrepreneurial researchers and stage their work as applicable. Their ‘technoscientific drama’ then produced neither the credibility of scientific fact nor that of their surveillance system’s utility, but instead established the researchers’ credibility as ‘scientist-entrepreneurs’ (Daston and Galison, 2007: 414). A focus on the performative aspects of their strategies highlights the mundane work necessary to constitute these scientific subjectivities, as well as the latter’s relative fluidity and fragility.
Staging alignment through technoscientific drama
Drawing on social worlds/arenas theory (cf. Clarke, 1991; Clarke and Star, 2003, 2008; Strauss, 1991), we can understand scientists facing commercialization pressures as simultaneously negotiating multiple commitments in competing or misaligned social worlds. From an interactionist perspective, academic disciplines and specialties can be viewed as social worlds, as groups that share commitments to common activities, as well as resources and ideologies stipulating how to go about their work (cf. Clarke, 1991: 131; Strauss, 1991). Social worlds are amorphous and lack clear boundaries, but can more or less coincide with formal organizational structures such as university departments. If changing political discourses enter this worksite via directed research funding, they can be viewed as posing another set of constraints on university researchers (cf. Clarke, 2005: 72). This is because university researchers in Germany are more amenable to demands from governments and industry as the number of tenured faculty has declined significantly over the past decade in relation to student numbers (Kreckel, 2008; also Lave et al., 2010). In this arena rife with potential conflict, scientists then have to balance commitments to their research, their academic careers and political demands for marketable technologies.
Misalignment between social worlds is not an unusual feature of academic research. Things do not always go as planned, and scientists routinely have to fit together different levels of work organization through a mundane process of continuous reorganization and tinkering (Fujimura, 1987). Levels of work organization include scientists’ work practices (e.g. experimenting), the organizations involved (e.g. their departments) and their disciplinary cultures or fields of research. This means that in addition to their intellectual labour, scientists have to ‘articulate alignment’ – ‘pulling together everything that is needed to carry out production tasks: planning, organizing, monitoring, evaluating, adjusting, coordinating and integrating activities’ (Fujimura, 1987: 258). Articulating alignment serves to create scientific problems that are feasible given available skills and resources, connect to concerns in wider fields of research or disciplines and are interesting for funders.
However, in some cases, there are strong constraints on articulation work. This is the case if alignment between scientific work and diverse social worlds is a central funding requirement (as is the case in transdisciplinary projects), but scientists experience misaligned social worlds as competing demands and typically lack the resources to split up their work and refer undesirable tasks to other organizational units or subcontractors (for examples of such divisions of labour, see Baumeler, 2009; Fujimura, 1987; Tuunainen, 2005a, 2005b). In such cases, rather than articulating alignment, scientists may temporarily stage alignment through what I call ‘technoscientific dramas’.
Drawing on Goffman’s (1959, 1967) dramaturgical framework, I understand ‘technoscientific dramas’ as short-term events set up by scientists to persuade funders of the seamless alignment between their work and their heterogeneous work environment. These events may be relatively disconnected from scientists’ everyday work, in the sense that the scientific problems on which they work may be to only a small extent, or even not at all, portrayed. Nevertheless, they are consequential for scientists because current and/or potential funding with the same institution is at stake (and with that, the resources to carry out research, as well as the livelihood of the researchers funded through the projects). 1
Goffman’s dramaturgical metaphor is useful to understand how the scientists in my study tried to control the funders’ perception of them. According to Goffman, a stable (visible, audible) separation between ‘front stage’ and ‘backstage’ is central to the successful conveyance of a desirable situation. ‘Front stage’ is the bundle of practices, personas and artefacts that comprise the performance (Goffman, 1959: 107). In the context of my study, the front stage came in the shape of presentations and demonstrations in which scientists tried to persuade funders that the diverse social worlds in their project were aligned. They attempted to provide an impression consistent with the funding institutions’ demands for commercialization by temporarily shifting into the personas of entrepreneurial scientists. The central prop in their demonstrations was, due to the nature of their commission, a prototype of their surveillance system. As Suchman et al. (2002) note, prototypes can be powerful in aligning ‘multiple, discontinuous social worlds’ by working within a ‘dynamic assemblage of interests, fantasies and practical actions’ (p. 175).
However, performances are only effective if they are kept separate from the ‘backstage’. The backstage is the bundle of practices, personas and artefacts needed to prepare the performance. In my case, the scientists crafted the narratives of their presentations and demonstrations, assembled the prototype, arranged items of their entrepreneurial front and disciplined members of their group. The separation between front stage and backstage cannot always be maintained neatly: there is a risk of slippages, that the technology does not work or that actors cannot be disciplined properly and fall out of character. In Goffman’s (1967) terms, these events were ‘fateful’ situations. Situations are fateful when they present both great risks and opportunities because short-term action in the present might have long-term consequences for the actors’ futures. On the one hand, these events are risky, because funders might discover the misalignment or might interpret these moments of breakdown as evidence of misalignment, issues on which the scientists have little control. On the other hand, they present opportunities, because if successful, scientists will be able to keep their funding and work on problems divergent from funders’ demands, and if they are very successful, performances may increase chances for future funding with the same institution.
The German security research arena and the development of a ‘smart’ closed-circuit television system
This article is based on an ethnographic study of a research project funded by the German government within the German Security Research Program (SRP). Over the course of two years, I accompanied a group of researchers who were commissioned to develop the software for an automated closed-circuit television system (CCTV). This system, in contrast to non-automated surveillance systems, was supposed to automatically detect ‘dangerous’ situations and behaviours.
The SRP, initiated in 2007 and ongoing at the time of writing, mainly funds the development of security and surveillance technologies and solutions. By investing in university and corporate research and development, the programme’s goal is to increase citizens’ security. To ensure that the research meets these goals, the government has formalized its demands in the programme’s review criteria. In terms of content, research proposals have to clearly outline their projects’ contribution to the solution of national security problems and the developing technologies’ projected marketability. In terms of organization, research projects are required to work in a transdisciplinary fashion, collaborating not only across disciplines but also with end users and small- and medium-sized enterprises. By incorporating both end users and industry, the government hopes to ensure the development of useful technologies.
These criteria are policed through a programme-specific review and selection process. The SRP does not recruit reviewers from within academia. Instead, it outsources the review and management of projects to a spin-off organization of the Association of German Engineers (Verein Deutscher Ingenieure [VDI]). Proposal review and project supervision are often carried out by the same people. Although some of them have a doctoral degree in the natural or engineering sciences, they have left their academic careers to be employed full-time by this organization. Once these reviewers have made their selection, they forward the project proposals to the Federal Ministry of Education and Research for final approval. Shifting discretion in the review process from academic review panels to bureaucratic entities, this altered review process can be understood as the German government’s expansion of social control in order to protect its investments.
The government’s expectations of the researchers are fairly straightforward. Mobilizing imageries of crime and terrorism, and referring to the limited capabilities of human security staff (cf. Bundesregierung, 2011), the government expects the researchers to develop technical fixes to social problems of crime and terrorism, as well as to increase the efficiency of surveillance processes by mechanizing them. These demands express shifting political notions of university researchers’ professional ‘jurisdictions’ (Abbott, 1988): expectations to contribute to the solution of security problems declare academic researchers experts on crime and terrorism, and expectations to render surveillance processes more efficient and effective position researchers as experts in security work.
The government’s goal to change the ecology of university–industry relationships is clearly expressed in the ambitious hope for the SRP to exploit what they see as a growing demand in security technologies: Through research and innovation, [the SRP] offers the possibility of promoting the competitiveness of the companies involved, as well as their security technologies’ marketability, to establish security as a national locational and economic factor and to open up possibilities on a European level. (Bundesministerium für Bildung und Forschung, 2007: 7; author’s translation)
2
By assembling teams of academic and industrial researchers, consulting agencies and potential users, it aims to stimulate national growth and increase international competitiveness. Thus, the SRP expects the researchers not only to assume responsibility for the solution of security problems, but frames their work explicitly as an economic activity.
The researchers in my study had applied to the programme by proposing to develop the software for an automated CCTV system. Their goal as outlined in the grant proposal was to mechanize surveillance processes in order for the system to identify ‘dangerous’ situations automatically and in real-time, and to alert the human security staff in such cases. The idea was to eliminate the need for operators to watch the screens at all times, instead having the system alert operators to an event of interest. They argued that their surveillance system, in contrast with non-automated CCTV systems, would facilitate intervention before the fact and would also reduce personnel cost through automation.
The university researchers included computer scientists, geoscientists, electrical engineers and legal scholars. Furthermore, the project included members of two private research institutes who are mainly computer scientists by training. On the corporate side, the project included a consulting agency that carried out cost–benefit analyses and an information technology (IT) company that was supposed to integrate the system for technology transfer. Finally, the project included two officers from regional police crime units, who were expected to share their expertise in detecting criminal behaviour. The project was relatively large and during its course involved at different times between 25 and 30 members, about half of whom were university researchers. In my analysis, I have focused on the university researchers involved in the project. Thus, when in the remainder of this article I refer to researchers or scientists, I mean the project’s senior researchers on the faculty level, as well as their doctoral candidates, all based in various universities across Germany. I have substituted all names, places and unique technical terms with pseudonyms.
The government’s demands put considerable pressure on some of the university researchers. This is not surprising, considering that working at a university is not usually framed in terms of solving national problems or trying to jumpstart national economies. The researchers experienced government demands as conflicting with more academic demands: on one hand, the government articulated an economic vision of their tasks and activities to promote economic development through technological innovation, while, on the other hand, their institutional environment placed much more importance on finding interesting problems and writing peer-refereed papers for career advancement. A consequence was that the majority of academic researchers in my study showed no ambition to pursue technology transfer, although this was a clearly articulated expectation in the funding programme.
The junior researchers in particular, all of whom were in the middle of their doctoral research, problematized straddling these different social worlds, which at times they saw as unfeasible. They saw themselves caught between conflicting expectations and accountabilities from science, politics and the wider public. For example, the government’s expectation that they solve practical problems relating to surveillance technology did not always seem to provide legitimate topics for the junior researchers’ doctoral dissertations and hence did not align with disciplinary expectations within the academic system. Furthermore, most researchers struggled with criticism from the public that framed their work as the violation of privacy rights, and some even with their own reluctance to contribute to growing public and private surveillance. These struggles, in some cases intentionally and in others not, resulted in the development of a surveillance system that was unsuited for commercialization. The resistance to both political demands for applicability and surveillance indicates that researchers’ struggles were rooted in their commitments to their disciplines, in formal organizational constraints at their universities, as well as in their political stances.
However, the researchers in my study did not try to protect their work from government control by quarrelling with the funding institution over the boundaries between science, politics and industry, unlike scientists in other research on multiple commitments in directed research (cf. Gieryn, 1999; Jasanoff, 1990; Tuunainen, 2005a; Wehrens et al., 2014). Instead, they used a different kind of boundary work, which Calvert (2006) describes as the ‘tailoring’ of research problems: making one’s ‘work appear more applied in order to gain funding and resources’ (pp. 208–209). 3 The tailoring of research can be understood as a mutual translation between researchers’ scientific interests and government expectations, thereby creating research problems that are recognizable as both scientific problems to the researchers’ peers and practical solutions to the government.
This is exemplified in the researchers’ splitting up of their promise of one single piece of software into a loose federation of projects,
4
which they aligned with their departments’ lines of research. Although for funding purposes the detection of dangerous situations was the project’s main goal, developing behavioural analysis algorithms was only one part of the larger project. For instance, the participating computer science department worked on decentralized system architectures, arguing that this would make the system more stable against potential damage (e.g. through vandalism). One of the participating geosciences departments worked on developing stereoscopic techniques to create three-dimensional visualizations, arguing that this would facilitate crime scene investigations. Yet another part of the project, run by a department focused on network modelling, promised to supplement the surveillance system with additional movement detection and acoustic sensors. All of these technologies fell neatly in line with current lines of research in the departments. As one researcher explained, From our perspective, video surveillance isn’t actually our original interest; we deal with the deduction of coordinates in some geometric form from photos and visual material. With that we’d also be able to process videos as sequences of pictures, and primarily our interest is precisely in the geometric analysis of such pictures. They don’t necessarily have to be surveillance footage, but that’s just one field of application where we can work on new subjects for us. (Doctoral researcher M.T., January 2011)
More importantly, as junior researchers from these departments reported, they were not functionalities suited for a marketable surveillance system, because they were ‘too unstable’ and ‘too costly’ (doctoral researchers M.T., R.L., K.F., April 2012). But framing their research as practical solutions allowed them to ‘experiment’ and follow their interests, something they ‘would never be able to do in the industry’ (K.F., April 2012). Although these tailored research problems were in fact not suited to developing a functioning and marketable surveillance system, they served to keep up the appearance that the different groups of researchers were conducting applicable research and development.
However, in order for the researchers to maintain the appearance of producing useful technology, tailoring their research problems was not enough. Because they knew that their work was unlikely to yield a marketable surveillance system, considerable tensions remained between what the researchers defined as their work and the work the government expected them to carry out. These tensions had to be continuously hidden when communicating with the funding institution, even more so because government control was formalized in regular presentation duties. Thus, the researchers had to actively ‘sell’ their research to the funding institution, and in order to do so they presented themselves as entrepreneurial researchers.
Keeping up appearances
The majority of university researchers in my study used PowerPoint presentations and technological demonstrations to hide the tensions between their commitment to academic work and the obligations they had to fulfil for the funding institution. Hiding these tensions required considerable backstage work: constructing narratives of technological use, mobilizing allies for the show and disciplining group members to perform in unison. Work on the front stage, on the other hand, used a range of dramatic techniques, which included creating and controlling an audience, public demonstrations and economic rhetoric. While hiding perceived vulnerabilities is a common strategy to maintain the appearance of professionalism (cf. Abbott, 1988), what distinguished this ‘technoscientific drama’ from both academic and corporate practices were its purpose and its audience. In contrast to scientific demonstrations, the purpose was not to produce the credibility of scientific facts and intellectual capital with an academic audience. In contrast to technological demonstrations in corporate settings, it was not set up to sell to organizations an existing or projected technology for profit. Targeting mainly the funding institution’s representatives who controlled the project’s progress, the purpose of their theatrical practice was to prove that the government’s money was well spent, in order to secure their present funding and future funding possibilities.
PowerPoint and performance: The researchers and the funding institution
The researchers had to report to the funding institution on the state of their research at intervals of six months. These meetings usually took two days, where all project members met with two representatives of the funding institution. The first day was usually dedicated to fulfilling their reporting duties, whereas the second day was dedicated to internal discussions without the funding institution’s representatives present. The distinction between ‘internal’ and ‘external’ meetings corresponds to Goffman’s (1959; cf. Hilgartner, 2000) distinction between front stage and backstage. In this case, the meetings’ front stage and backstage were not necessarily separated spatially and physically, but switched their gestalt depending on the physical presence of the funding representatives. Rather than indicating deceit, the separation of front stage and backstage can be understood as shifting from one social world to the other by mobilizing different cultural registers.
Although the meetings always took place in a different location, the structure of their ‘official’ parts remained the same. The principal investigator welcomed the group and presented the agenda for the day, and this was followed by presentations by each of the nine group units. After each unit had presented their current state of research, the group allotted some time for feedback from the funding institution’s representatives, as well as for administrative matters. Feedback, if there was any, was usually short and signalled approval to continue in the way suggested by the scientists. The days always ended with everyone dining together in an upscale restaurant, an occasion used by the senior researchers to informally inquire with the funding institution’s representatives about funding opportunities for future projects.
The group used the presentations to prove that the government’s money was well spent. Proving the quality of their work was difficult because the researchers did not think that observing them coding software or troubleshooting bugs was quite persuasive enough, and there simply was no materialized surveillance system to look at – only pieces of software dispersed across Germany. They thus had to construct indicators of their competence that tied in with the government’s expectations of developing a marketable surveillance system. These indicators were to compensate for the lack of visible evidence and make their work both communicable and recognizable to the funding institution. Such indicators encompassed much more than just rhetoric and included the slides’ design, researchers’ clothing and shoes, refreshments and coordinating intricate stage plays. When the researchers presented to the funding institution, they were importantly performing their competence (cf. Knoblauch, 2008), and they did so by substituting for proving the quality of their work proving that they were the right people for the job.
To make a persuasive case for a novel and marketable surveillance system, the scientists hid problems and portrayed their research as rationalized work processes. This served to pre-empt scrutiny, stress the feasibility of future technological applications, and communicate an idea of linear progress. The content usually excluded process detail and instead focused on product, that is, the benefits of their work for surveillance processes. The structure of individual presentations was in most cases a reiteration of their grant applications, where they had divided their work into ‘work packages’ along a timeline. Their presentations then restated the goals of the work packages in relation to how they would improve surveillance systems and added two to four bullet points on how they successfully went about it. However, the researchers rarely explained details about their methodology, let alone expounded on the problems they encountered in their everyday work. For example, the work packages (unsurprisingly) did not reflect the temporal order of their work, as they were often working on several packages at the same time. Furthermore, some researchers routinely complained to me how ill-constructed some of the problems in the grant proposal were. Tinkering, incompleteness, abandoned ideas or dead ends are a normal feature of scientific work. However, unless it was absolutely unavoidable, they discussed such realities of scientific work only when the funding institution’s representatives left the room.
The project members also had a shared understanding that they had to speak in a single, unified voice. It was clear to all group members that under no circumstances would any presenter be criticized by another group member with the funding institution’s representatives in the room. Disputes were placed in the open either on the second day of the meetings or in separate meetings among the senior researchers. The following event from the very last presentation to funders shows a particularly vivid instance of putting up a unified front to hide what could be perceived as a major problem by the funding institution: Most project members have already taken their seats. I find a place next to Prof. Dreyer, who is busy typing on his laptop. It is the day after the demonstration, and everyone looks a bit tired. Mr. Gehrke [one of the supervising project officers] sets out to say a few words about the project and to introduce the final presentations. Somewhat approvingly, he comments on the demonstration: ‘So, in comparison with the first demonstration, this one was very good. However, what we’re really interested in is what’s going to happen with it [the surveillance system] from here on out’. The senior scholars suddenly seem much more awake than a second ago, and blurt out a series of statements, putting the emphasis somewhere between reassuring the officers and distracting from the question. Prof. Schmidt exclaims: ‘If only Prof. Vogel were here! Then we’d at least know how the project would continue scientifically!’ Prof. Dreyer interjects ‘He’s outside negotiating about an extension’, glancing across at Mr. Gehrke (without any success of getting a hint whether Prof. Vogel is going to be successful). Mr. Feyl addresses Mr. Gehrke and nods while he says: ‘Well I have addressed this question in my slides’. But Mr. Gehrke insists. ‘It would be nice to know what Mr. Meyer has to say about this’. (Field notes, February 2013)
Mr Meyer was the representative of the IT company that was supposed to integrate the software into one platform and transfer the technology to the market. Having a corporate member in the project was not only a funding requirement but was also used by the scientists as evidence that the group was aiming for commercialization. However, what the project officer did not know was that Mr Meyer had not shown up for the previous year. While he still was an ‘official’ project member (in terms of paperwork and funds), in terms of everyday project work, he had left. During the last presentation, the group hid this instance of extreme misalignment from the officers and told them that Mr Meyer was sick that day and could not make it, but had sent the slides. This led to the rather awkward situation that one of the doctoral researchers presented those slides in place of the IT company’s representative, although he was utterly unfamiliar with the slides’ content. Perhaps fortunately for them, he could re-use a somewhat familiar practice they called ‘PowerPoint karaoke’, a farewell ritual for graduating PhDs at one of the participating computer science departments.
My inquiries to find out why the most important corporate part of the project had dropped out yielded no significant results, as my informants did not know details. Striking was that the researchers seemed only mildly bothered. The fact that one project partner could just disappear without major repercussions for their work shows how effectively they had divided the work among themselves, allowing them to operate relatively independently of each other. Although their response to the project officers’ insistence on pursuing technology transfer was not particularly smooth, they still successfully constructed the corporate partner discursively and performatively in order to stage alignment.
Other techniques leaned more explicitly towards corporate practices. Performing as a unity also meant unifying the visible delivery. On one hand, this meant that some of the senior researchers urged the younger members to use the project’s design template for their slides, with its standardized layout, fonts, colours and logos. On the other hand, they also meticulously stipulated the dress code. Whereas the researchers wore formal clothing on the first, ‘official’ day of the meetings (suits or at least collared shirts), more casual clothes were the norm on the second day and in their everyday work (e.g. jeans and running shoes). The following field note shows how senior researchers sanctioned informal clothing and rewarded professional business attire at formal occasions: In the hall with Ben, whom I ask about the graphical user interface. Ben has already changed from jeans and long-sleeve into a suit. He looks slightly awkward in his turquoise-coloured suit, which is a size or two too big for him. While Ben explains to me how the interface restricts access for different kinds of users, Mr. Feyl [senior researcher at one of the private R&D institutes] approaches. ‘You’re looking dapper!’ he comments on Ben’s outfit and pats him on the shoulder. Ben blushes and chuckles a bit, but looks happy that his superior approves. Jonas [a postdoctoral researcher], who had overheard the conversation, in turn comments on the situation: ‘perhaps I should’ve put on a suit as well … But I’m an engineer!!’ Mr. Feyl raises his eyebrows and addresses Jonas, nodding in Ben’s direction: ‘Much on the contrary, this is exactly right’. (Field notes, February 2013)
Maintaining the appearance of developing a marketable surveillance system thus not only included rhetoric but also required the researchers to embody a persona that Daston and Galison (2007), in their history of scientific subjectivities, describe as the ‘scientist-entrepreneur’ (p. 398). However, the field note above also shows that embodying the persona of entrepreneurial researchers for the junior researchers meant a process of enculturation, and hence that this embodiment remained always somewhat fragile.
In situations in which the researchers perceived the boundary between front stage and backstage as particularly precarious, they tied the above elements together and created persuasive narratives. The dramaturge-in-chief for these performances was Dennis, one of the junior researchers in the coordinating unit. He was in charge of distributing and coordinating the play’s characters and making sure that the technical infrastructure worked in support of their performances. The following event from my fieldwork illustrates just how much invisible work went into crafting the entrepreneurial subjects the government demanded: The meeting takes place in a sober room with a large oval table with a screen at one end. Everyone seems a little tired. After we get our coffee and snacks, we sit down at the table, and all group members open their laptops. Martin [the principal investigator] welcomes Mr. Gehrke and Mr. Hirsch [the funding institution’s representatives] and introduces the schedule of presentations. … Next, Cecilia takes the stage and talks about the legal possibilities of integrating acoustic sensors into the system. She uses a lot of jargon, which makes her presentation quite hard to follow. She discusses the relevant legal regulations that need to be considered [displaying excerpts from legal code on her slides]: secrecy in telecommunications, confidentiality of the spoken word, and that explicit consent of the monitored person isn’t needed for data gathering, but for data analysis. She concludes that while integrating acoustic sensors for purposes of research could be possible because consent could be solicited from members and volunteers in an experimental set-up, this wouldn’t be possible in a real-life situation. When Cecilia ends her presentation, I see that Dennis sends a thankful nod in her direction. Cecilia’s was the last presentation, and I leave the room with Dennis, Kim, and Kai. Outside, Kim turns to Dennis: ‘Do you think it worked?’ Dennis shrugs and responds ‘I hope so. Otherwise we’ll have to think of something else’. They both look a little distressed, and so I ask what that was all about. (Field notes, May 2012)
Dennis explained that Kim’s project unit had promised to integrate acoustic surveillance sensors into the system. When it became clear that work towards the promise could not be made to fit meaningfully into Kim’s dissertation and that he did not have the necessary time to fulfil the promise, they came up with a plan to circumvent the problem. They checked with Cecilia, a post-doc in the legal unit, whether they could justify excluding work on the acoustic sensors because of its legal inadmissibility, instead of justifying their failure to achieve the goal because of lack of resources (doctoral researcher D.P., May 2012). Cecilia agreed to examine the case, and, when it turned out that it was possible to argue that operating the system with acoustic sensors would not be legal, prepared a report. At the meeting, Cecilia then debated at length whether, given privacy legislation in telecommunications, acoustic data could be gathered and analysed within the surveillance system. As we have seen above, she argued that the system would be legally inadmissible in cases in which it would be employed in practice. Thus, the researchers exploited the project’s transdisciplinarity and the relative flexibility of legal opinions to their advantage. As Ulrike, a doctoral researcher in the legal unit, had explained earlier to me, many cases can be argued both ways, particularly with technologies that are not yet employed in practice, because there has not yet been a court ruling (doctoral researcher U.M., May 2011).
This narrative can be seen as contributing to the project in multiple ways. First, the researchers externalized the reasons for the lack of practical applicability to legal regulations they could legitimately claim lay beyond their control. By referring to external factors, they pre-empted suspicions that failing on their promise had anything to do with their work. Second, they justified excluding the acoustic sensors with reference to a lack of practical applicability, thereby mimicking the SRP’s rhetoric of advertising useful and privacy-friendly security solutions (cf. Bundesministerium für Bildung und Forschung, 2012). By giving priority to practical applicability over a bounded research-only setting, the researchers presented themselves as developers concerned with applicability and privacy. And third, appealing to the cultural register of legal language gave their case more authority because trustworthy expertise in Germany is largely tied to institutional knowledge (cf. Jasanoff, 2005: 262). It was perhaps by coincidence, but Cecilia’s presentation, which made heavy use of legal jargon and slides filled with quotes from German legal code, helped to theatrically underpin an impression of authority.
Staying rhetorically and performatively within the parameters of the SRP was a common strategy to deal with tensions between their work and the government’s demands. As Robin had explained to me at an earlier meeting in the planning of this presentation, sometimes the promises made in the original proposals turned out to be uninteresting, unfeasible or both: ‘But saying “we’re not doing it because we can’t” is not an option. What we need then are professional reasons’ (doctoral researcher R.L., April 2012). Robin’s statement underlines that the group was well aware of the government’s demands and that both the failure and success of their work had to be framed in terms of applicability. Staying in persona was crucial to achieving this.
‘Theatre of use’: 5 The researchers and the wider public
In addition to their regular presentation duties, the researchers promised to organize two public demonstrations of their surveillance system. On these occasions, the researchers presented not only to the funding institution’s representatives but also to potential users, industry representatives and the wider public. The first demonstration was scheduled for the project’s mid-term mark, after 18 months, and the second one for the end of the project, after 36 months. The SRP put considerable pressure on the researchers, reserving the right to terminate the project and withdraw funding if it evaluated the project as a failure after the first demonstration.
The researchers took these events very seriously. Their staging script planned the same order for both demonstrations: a welcome with coffee and refreshments, a PowerPoint presentation by the principal investigator, a live demonstration of the CCTV system and its functionalities and a poster session with time for questions. After the demonstration, the project members and the funding institution representatives would go out for dinner to an upscale restaurant. These events, even more than the regular meetings, took on the character of intricate stage plays.
In contrast to their presentations, it was not enough for the researchers to present themselves as the right people for the job in order to prove that the government’s money was well spent. As with Boyle’s vacuum (Shapin, 1988; Shapin and Schaffer, 1985) or Pasteur’s anthrax vaccine (Latour, 1988), the researchers had to create a complex scenography to make their work visible. In order to prove that the government’s money was well spent, the researchers had to make them ‘see the software’, and this in turn was only possible once all these elements had been assembled and orchestrated. Creating the setting included crafting a front stage that showcased the surveillance system, assembling a material system that would appear as a discrete entity, mobilizing an audience that would bear credible testimony to the surveillance system’s utility, and creating narratives of use, Through all of this, the researchers had to present as scientist-entrepreneurs.
Again, these elements were not orchestrated arbitrarily but appealed to a cultural register within which the scenography became recognizable as what I will call the ‘tech demo’. Like the PowerPoint presentation, the tech demo can be understood as a genre of business communication. IT companies usually use them to convince potential buyers or investors of their product’s utility or the feasibility of a technological approach, and as such they are recognizable as corporate practice (Coopmans, 2011; Rosental, 2005; Simakova, 2010; Smith, 2009; Suchman et al., 2002). Drawing on the knowledge of those researchers who had experience with such corporate practices, the researchers enacted this particular genre to make their work appear to tie in with the government’s expectation of developing a marketable surveillance system.
The ‘tech demo’ pulled together an interesting mixture of familiar techniques of scientific demonstrations and ‘technologies of stage magic’ (Smith, 2015). The demonstration produced its effectiveness by merging the world and the laboratory with one another (cf. Latour, 1983) and by hiding the backstage work that went into the production of the demo’s effects (cf. Shapin, 1989). However, while the effects produced in scientific and engineering demonstrations usually are taken by the audience to be evidence of reality (Collins, 1988), it was clear to the audience that they were not witnessing a real ‘dangerous situation’ that the system then would detect, but a realistic simulation of one (Smith, 2015: 325). In this sense, there were differences in assumed realness with reference to the situation, the system and the researchers’ work. As in stage magic, the audience knew that it was a simulation, but could be expected to act as if the situation was real. However, with reference to the system and the group’s work, the demonstration had to maintain the appearance that the demonstration was giving the spectators natural and unmediated access to their work. This was tricky, because at the same time they had to control the demonstration’s outcome as much as possible. How they managed layers of simulation and naturalness shares interesting similarities with what Smith (2015) in his analysis of technologies of stage magic calls ‘dissimulation’: the production of an effect by not only concealing the secret methods and mechanisms behind this effect but by rendering them absent from the situation.
Importantly, producing this kind of ‘naturalness’ required a tremendous amount of ‘invisible work’ (Star and Strauss, 1999). First, the researchers had to create a front stage in which to showcase the surveillance system. In contrast to the group’s presentations, front stage and backstage were separated physically and spatially in their demonstrations and looked radically different. The researchers chose one of the universities’ atriums to be the front stage for the demonstration and the Dean’s hall for the preceding welcome and introduction presentation. The atrium was a huge rectangular space flooded with light, lined with supporting columns on two sides and featuring staircases winding up to a gallery. The cameras were mounted to the columns in the first demonstration and to tripods on the gallery in the second one. The space between the columns was supposed to represent the monitored space. On one side, just beyond the column line, the group had set up ‘the system’, as well as their posters and a table with coffee and refreshments. Part of a mid-19th-century palace, both rooms looked impressive and grand, and underlined the seriousness with which the group presented itself.
Second, the group had to assemble a material system that the funding institution’s representatives could see and with which they could interact. Because there was no system as such, but only pieces of software dispersed all over Germany, the researchers had to make one from scratch. It is telling that what the researchers referred to as ‘the system’ during the demonstration did not embody all the parts of their work, but only the graphical user interface. The researchers presented the interface in the form of a huge touch screen, set up as a table. The other parts of the CCTV system – the computers and servers that ran the software they developed – were kept backstage, hidden from sight. The backstage was located two stories above the atrium in two small and stuffy rooms. In various respects, these rooms were the very opposite of the front stage. They were dominated by neon lights, chipped wall paint, numerous PCs, tangles of cables, whiteboards scribbled all over, used coffee cups and leftovers from the preceding days, when the group had worked until the last minute to hard code the software to the specific space of the atrium. Hiding the messy infrastructure of their work and revealing the glamorous parts were central to creating a discrete object that would foreground imaginations of future usage.
Third, they had to mobilize an audience that would bear credible testimony to their technology’s utility. To that end, it would not have sufficed to show a working surveillance system to the funding institution’s representatives: considering that there was a clear expectation of producing a commodifiable piece of technology, witnesses who would testify to the technology’s utility had to come from industrial and user contexts. Consequently, they invited representatives from the security technology industry and private security services, and the police officers who were already part of the project. Creating this kind of audience was not only directed towards creating credible testimony. It also signalled that the researchers were building relationships with the industry and users, and thus with their aspirations to turn their work into a marketable product.
Fourth, the researchers had to weave all these elements together into narratives of use that would stage their technology in action as the dramatic culmination of their previous work. These ‘storylines’, a term they themselves used, had to be close enough to their original commission – detecting ‘dangerous behaviour’ – and at the same time simple enough to demonstrate the technology without risking the system’s breakdown. This implied rigorously selecting parts of their software, hiding the unstable and ‘unimpressive’ ones and showcasing the stable and dramatic ones.
The course of the demonstration then had to be carefully scripted and practised. This included fleshing out the storylines, distributing roles among the researchers, determining what they would do at any given moment in the demonstration, writing a running commentary that would relate the technical features to the benefits in practice, and anticipating questions, comments and criticisms from the audience. Carefully orchestrating all these elements was crucial to achieving the perceived naturalness of their demonstration.
Despite their hard preparation, at the first demonstration their script did not work quite as they imagined, because the technology failed to produce the desired visible effects. The failure to produce a desirable situation highlights how risky their technoscientific drama was, as it opened pathways for the officers to scrutinize their work: Everyone is gathered around the big touch screen in the hall. Other than the researchers, there are industry representatives, police officers, security services companies, and the funding institution representatives present. Dennis introduces the ‘use case’: he explains that someone will steal the painting they put in the middle of the hall, and that the system will send an alarm to the screen when that happens. This would be based on the analysis of the video material. ‘So who wants to steal the painting? Jakob?’ Before Jakob can answer, a very excited looking Professor Bode jumps up, raises her hand, and walks towards the painting. Everyone’s heads turn back and forth between her and the screen, looks of anticipation on their faces. The professor finally reaches out to the painting and takes it from its easel. Everyone’s attention shifts to the screen. Anticipation turns into extreme awkwardness when nothing happens. The researchers look at the screen with horror. Dennis tries to salvage the situation with some self-deprecating humour: ‘I forgot to say that this is also an opportunity to see work in progress’. (Field notes, November 2011)
Their ex-post-facto ‘investigation’ into the failure turned up a number of mundane factors: they had not standardized the format for messages between the system’s components, they had forgotten to turn on some of the sensors, there were restrictions in how they could mount the cameras and so on. While the computer logs they analysed afterwards showed that the software did in fact register ‘the theft’, the mundane errors led to a failure to produce visible effects. Because the alerts were not produced on the screen, it seemed as though nothing had happened, which undermined the whole purpose of the demonstration.
After the demonstration’s undesirable outcome, Dennis then was quick to announce that the audience would have the opportunity to ask questions about the surveillance system, trying to regain at least some control over the situation. One of the project officers jumped at the opportunity. Interestingly, instead of inquiring why the system did not work, he started asking questions about the touch screen – although the hardware was not a product of their work, as they were building software. Technology buyers from private security service companies joined them, and together they talked at length about their fantasies of having such a piece of high-tech equipment at their disposal (and complained about their budget restrictions). In Goffman’s (1967) terms, their response was one of ‘studied nonobservance’: they ignored the breakdown and played along so as to not let the situation derail completely. The fact that the officers contributed to restoring the situation suggests that there might have been some degree of ambivalence on the side of the funders about aims for commercialization. 6 At the next day’s meeting with the officers, however, the researchers were hard-pressed to argue away the dysfunction as mere technicalities and ‘glitches’, insisting that their original concept remained valid and making more promises for the future.
At the second demonstration, the group knew better how to anticipate demands of the funding institution. This time, the researchers decided to show how the system would detect a person entering a dangerous zone and set off an alert. Although again this scenario showcased only a minor fraction of their work, it was feasible and the researchers could relate it to its application in a train station. Relating the demonstration to a practical application was what Dennis’ running commentary would accomplish, by pointing out how this functionality could be used to prevent people from running onto train tracks. After gathering everyone around the user interface, everyone in their business attire, Jakob – responsible for the graphical user interface – had to mark a ‘forbidden zone’ on the grounds map on the screen. This forbidden zone was supposed to represent an area where people were forbidden to loiter. Then, Dennis would ask the audience for a ‘volunteer’ to test whether the detection worked, whereupon Robin, geosciences doctoral candidate, would step forward to walk through the monitored space. The grand finale would be the alert – visible and audible for all guests. Of course, Robin and Jakob had agreed upon a small area of the monitored space beforehand, so Robin walked unerringly through the 600-m2 space without having even looked at the screen to identify the corresponding area in the atrium.
What is important to note is that both demonstrations were relatively disconnected from the researchers’ everyday work. The ‘use-cases’ were either unaddressed problems, set up solely for the purpose of the demonstration (the case of theft), or were minor parts of their work (the case of forbidden zones). The slightly comic effect of showcasing stable scenarios rather than the scientific problems about which they cared was not lost on them. According to Jakob, the ‘forbidden zone’ train station scenario was somewhat like shooting at sparrows with cannons: ‘Honestly, it would be so much cheaper and effective to install a laser beam along the tracks which registers if the beam is interrupted’ (doctoral researcher J.K., February 2013).
Finally, based on their experience with the first demonstration, they built additional safeguards against breakdown. In addition to running numerous tests to troubleshoot bugs and practising the script, they created an emergency strategy to guard against such unwanted breakdown: Before the guests arrive, I notice a small, inconspicuous table half-hidden behind a column. A PC and monitor are set up on the table, and Max [a geosciences doctoral candidate] sits in front of the computer and stares at the monitor. I sit down next to Max and ask him about this suspiciously inconspicuous looking computer. Max explains: ‘That’s only for us to control the situation, in case something happens, we’re able to reboot the system manually. We’ve tested it and it only takes a couple of seconds, so that no-one should notice it during the demonstration’. I am curious: ‘So it’s like your own personal control room?’ Max chuckles and confirms. He explains that during the days preceding the demonstration, they had written a small piece of software that would reboot the surveillance system unnoticed by the audience. (Field notes, February 2013)
The problem was that the system cache that stored the spatial and temporal data of the analysed video footage was full after only a few minutes of analysis. This would cause the system to freeze, and the only way to get the system running again was to reboot it. Their emergency script guarded against a breakdown of the demonstration even when the system broke down, by ensuring that the system performance could temporally re-align with the researchers’ performance. In this respect, this ‘emergency code’, just like the other theatrical elements, assumed the role of dramaturgically repairing existing instabilities that could endanger a successful demonstration.
Technoscientific drama: The credibility of what?
As in previous accounts of demonstration and presentation practices, the researchers’ technoscientific drama worked so well because it produced credibility in virtue of ‘seeing and believing’ (Shapin and Schaffer, 1985). But exactly what credibility was at stake here? Clearly, the demonstration did not foreground the researchers’ credibility as scientists, as their performances excluded large parts of their work and were for a primarily non-academic audience. Nor did the demonstration produce the credibility of their technology’s utility for potential investors because a discrete technology simply did not exist, and some of the researchers even emphatically refused to take part in proliferating private and public surveillance. Because the researchers’ theatre of persuasion largely rested on presenting themselves as the right people for the job, I argue that the main objects – or rather subjects – of credibility to be produced were themselves: as scientist-entrepreneurs developing a functioning and marketable surveillance system.
The co-constitution of theatrical practice and scientific personae resonates with Shapin and Schaffer’s (1985) account of how 17th-century experimental practice created a new kind of scientific subjectivity. The new set of literary and social technologies that produced scientific credibility also evacuated the scientific subject from accounts of experimental practice, thereby creating the ‘scientist-as-modest-witness’ who would let the ‘evidence speak for itself’. By analogy, the researchers in my account used a set of material, rhetorical and social technologies to prove that the government’s money was well spent. They thereby also constituted themselves as entrepreneurial subjects who quite immodestly advertised their technology – much in the sense of Foucault’s (1982, 1988) ‘technologies of the self’. Their theatrical practices form the link between how they are ‘summoned’ into the subject position of scientist-entrepreneurs and their ongoing struggle to (temporarily and incompletely) constitute themselves within this discourse (cf. Hall, 2000: 27). Such practice is deeply embodied practice: just as scientists had to turn their bodies into truth-tellers through specific exercises that would suppress their subjectivity (Daston and Galison, 2007), the researchers in my studies had to transform their bodies into those of businessmen by slipping into their approximation of business attire, shifting into genres of business communication and emphasizing their subjectivity by stressing the unique benefits of their technology.
I purposefully use the term ‘businessmen’, as the gendered aspects of their performances are hard to ignore. This is not because most of the group members were men, but because ‘the business executive’ is what Connell (1998) claims to be the current hegemonic form of masculinity – understood as a set of hybrid but authoritative practices (Demetriou, 2001) – within a system of masculinities in the Global North.
The co-constitution of theatrical practice and specific kinds of masculinity resonates with Haraway’s (1997) and Potter’s (2001) insistence that 17th-century experimental practice not merely reproduced forms of masculinity, but that the scientists’ masculinity was at stake. Because modesty was at odds with the prevailing form of masculinity, the ‘hero-warrior’, scientists were vulnerable to being labelled as effeminate. They thus re-interpreted the dominant cultural ideal of masculinity from ‘conquering territory’ to ‘conquering nature’ to retain both their social and epistemic authority. Drawing an analogy with Haraway’s and Potter’s observation, we might then speculate that presenting themselves as ‘service providers’ for the government would have meant a significant loss in status and authority for the researchers. By contrast, appropriating and recombining elements from the persona of ‘the business executive’ positioned them as leaders within a capitalist economy largely invested in technological innovation. Situating their gendered practice within the continuity of scientific masculinities, we might then think of their self-presentation as amounting to the reframing of ‘conquering nature’ as ‘conquering transnational technology markets’ – a reinterpretation that ties in neatly with the government’s aspirations to strengthen Germany’s position within European security technology markets.
Enacting this specific masculinity solved a problem that resulted from their tailoring and staging practices. Keeping up the appearance that they were submitting to the government’s expansion of control put the researchers’ social authority as scientists somewhat in jeopardy because the authority of science significantly depends on the ability of scientists to draw clear boundaries between science and non-science (Gieryn, 1999; Jasanoff, 1990). Open disputes in the staging situation, however, were clearly impossible. Using cultural registers of business masculinity, and thereby shifting into a different position of cultural authority, helped the researchers to retain their social authority in the face of the government’s expansion of control into their work.
While shifting into the persona of the entrepreneurial researcher certainly helped the researchers keep politics near, but still out, it is hard to say how successful they were in convincing the funding institution. On one hand, the funding institution’s representatives expressed their overall satisfaction with the group’s work in their regular feedback after demonstrations and presentations, the one exception being the first demonstration. Although they certainly were not happy by the group’s failure to produce desired results at the first demonstration, they accepted the group’s explanations and did not discontinue funding. Despite occasional ruptures in the group’s self-presentation, the way in which they anticipated funders’ demands thus seems to have been convincing enough. On the other hand, the SRP rejected the group’s application for a successive project. When pressed for reasons for this rejection, the funding institution’s representatives explained that their proposal was positively evaluated, but referred to the massive number of applications they received, declining to give any further explanations. Only the legal scholars and private research institutes were able to obtain further funding for their individual research through the SRP, while the remaining university departments continued to obtain prestigious grants from the German Research Foundation (Deutsche Forschungsgemeinschaft [DFG], comparable to the National Science Foundation [NSF]) to fund their doctoral candidates.
Conclusion: Shifting gears
The researchers did not present themselves in an arbitrary manner, but drew on already existing cultural registers that they recombined and fitted to their needs. However, this was equally true for other contexts. For the funding institution, they staged their work as having applications; when addressing an academic audience (e.g. in their academic papers), they framed it as scientific; when meeting for week-long ‘hackathons’ to integrate their work, they appropriated elements from hacker cultures (Coleman, 2013; Irani, 2015). And in tune with the researchers shifting in and out of different contexts, the surveillance system, too, changed its appearance from a more or less loosely coupled ‘prototype’ to a disconnected array of software pieces that worked within the wider research of the respective university departments. The relationships between subjects, objects and practices were thus continuously reconfigured as the group moved through different kinds of situations.
These shifting configurations highlight the relative fluidity of the researchers’ subjectivities. It is seductive to conclude that what they ‘really wanted’ was to conduct ‘basic research’ and that their theatrical practice was merely a convenient deceit. However, it would be asymmetrical and static to assume that the scientists’ academic selves were somehow more authentic than their entrepreneurial selves. After all, academic personas, too, are created, often in long processes of enculturation. Rather than treating the distinction of front stage and backstage as one of authenticity and deceit, it should be understood as shifting from one social world to the other by mobilizing different cultural registers, without preferential treatment of one or the other. 7 This is also clear in the ambivalence with which the university researchers in my study described themselves. They identified at the same time as both ‘scientists’ and ‘developers’ (doctoral researcher K.F., April 2012) and wanted their research to be useful ‘someday’, but just ‘not right now’ and ‘not in this context’ (doctoral researcher M.T., May 2012). Expressed in such statements is the multiplicity of social worlds that unfold with increasing pressures for commercialization. Attending to the ways in which scientists experience these as conflicting demands and navigate them might elucidate some of the more interesting responses to such pressures.
I want to caution against interpreting the technoscientific dramas described here as simple cases of deceit or even sabotage, because doing so might too easily elide their social conditions. The tremendous amount of invisible work the demonstrations and presentations entailed indicates just how intensely the researchers struggled with being pushed and pulled in different directions by politics, academia and the wider public. Their theatrical practices were at least partly a product of powerful competing or misaligned social worlds.
The researchers’ technoscientific dramas reveal problems at the intersection of science and politics that are easily overlooked if viewed through simple dichotomies. Rather, we gain a richer understanding of current interactions between science, governments and industry by examining the variety of ways in which academic researchers respond to pressures for commercialization. I have highlighted one such mechanism here: by ‘shifting’ in and out of the persona of the scientist-entrepreneur, the university researchers compensated for misalignment between their work and the diverse social worlds involved and managed to protect their research without jeopardizing their present funding. Thus, asking whether or not knowledge production is changing under current commercialization pressures is perhaps putting it the wrong way around. Instead, my research suggests that it might be productive to analyse under what conditions researchers ‘shift gears’ and what kinds of work and objects make shifting gears more or less feasible.
Footnotes
Acknowledgements
I am grateful to Geof Bowker, Francisca Grommé, Alberto Morales, Kavita Philip, Winnie Poster and Nima Lamu Yolmo for donating their time to discuss earlier versions of this paper. I also thank Sergio Sismondo and the anonymous reviewers for their very valuable comments and suggestions.
Funding
The research for and writing of this paper were partially supported by grants from the German Ministry of Education and Research and the German Academic Exchange Service.