Collaboration and Research Practice in Intelligence

Close, intensive research collaboration between universities, companies, and the public sector can open up new and different opportunities for qualitative research, and provide analytic and empirical insights that otherwise might be difficult to obtain. The aim of this paper is to explore collaboration as a means of doing research with the intelligence community. Experiences from a research project concerning dilemmas the practitioners face in their organization within the Swedish Armed Forces, serve as a starting point for this reflective discussion. It is argued here that collaboration is suitable when change is required. The mutual learning between the actors feeds into change processes. However, such collaboration raises fundamental ethical issues that are complex and highlight various academic, institutional, and personal perspectives. Collaborations should not be a set of “howto” recipes, but rather a research activity that can have substantial rewards for researchers and practitioners alike.

academia, but also attempt to stimulate collaborations between companies and universities to make research results accessible, and to make sure they reach various areas of society in which they can have impact. The different forms of collaborations are sometimes even a prerequisite for obtaining research grants. Universities' research and education strategies often see collaboration as something that can contribute to the use of research results in society, but also as something that gives rise to new research.

Close, intensive research collaboration with an organization can open up
new and different opportunities for qualitative research, and provide analytic and empirical insights that otherwise might be difficult to obtain.
Collaboration can also increase knowledge transfer that would not be possible for individual actors working in isolation. Still, such close collaboration is not entirely straightforward. The word "collaboration" indicates that the activities are, at least to some extent, carried out together.
It also promises that the participants gain something out of this work.
Collaboration, I believe, also raises a number of methodological issues, especially for qualitative research practices. These issues concern how hands-on research practices can be carried out together, but also concern issues on control of the research agenda and intellectual property rights, to name a few. What the issues are depends on the research context and, for example, the organizational setting. The aim of this article is to explore collaboration as a means of doing research with the intelligence community (IC). An IC is a particular type of organization in which practitioners carry out certain work tasks in a certain way to accomplish certain goals.
I explore how close collaboration can contribute to intelligence research, but also what the challenges are for such research. By doing so, I hope to contribute to the literature on research collaborations with the IC. My experiences from a research project -for my part, concerning dilemmas the practitioners face in their organization, the Military Intelligence and Security Service (the Swedish acronym MUST) within the Swedish Armed Forcesserve as a starting point for this discussion.

Background
There are different perceptions of collaboration within academic disciplines. Social science research has always relied on the active contribution of actors from outside of the university. In anthropology, for example, fieldwork as a research practice already indicates some sort of relationship between the researcher and the members of the community she or he is interested in. Even if cooperation is needed for successful fieldwork, this is not the form of collaboration I refer to here. Rather, the type of collaboration I am concerned with can be found within applied anthropology and similar approaches, for example, within the emancipatory approaches, in which -at least to some extent -the aim is to support a certain group of people in one way or another for various reasons. Within human-computer interaction research, as another example, collaboration between academia and various organizations and so-called user groups is common.
Collaborative research is also found within intelligence studies, in particular those with an applied approach.
Today, the outcomes and research practices of social science research are recognized among many agencies outside of universities (e.g., Savage and Borrows 2007). Various actors use the same or similar methods to understand the social practices they are interested in; this paves the way for understanding the research practices and therefore lays down a common foundation for collaboration. Collaboration, as described in this article, is thus not reserved for research within intelligence studies alone. However, collaboration is always a challenge and is not taken-for-granted, and it depends on the particular situation and its research arrangement.
For some years, attempts have been made to provide the Swedish Armed Forces intelligence function with a clearer organizational identity and structure. This has meant attempting to bring order, for example, to the policy documents that coordinate intelligence studies and development, as well as to agree upon a single model for production management. This approach, together with addressing a number of other workplace issues and organizational concerns, has led to changes in work practices. The organizational and work redesign, including taking into account practitioners' perceived needs, has been in development for a number of years.
The research presented here is based on the three-year study (2008)(2009)(2010) of the redesign of the military intelligence function. Although this effort can be studied from several perspectives, our research team studied past and present experiences of intelligence function, principles, models, and methods that support intelligence architecture, but also the dilemmas that intelligence practitioners face in their organization. The research focused on the method and technology development related to two intelligence functions: strategic analysis of the outside world and support for intelligence work. Although the others involved in the project already worked at the Swedish Defence University, I was contracted for the part of the study that examined the dilemmas that intelligence practitioners face in their organization. For security reasons, my background was checked against the authorities' various records, and I signed a privacy and confidentiality agreement before joining the team.
Initially, I was excited about the possibility to carry out research in an intelligence community, which was an unknown to me at the time. I was curious about the community and the work done there but knew very little of it. The IC is, after all, a somewhat exciting and mysterious field that almost always brings about enthusiastic looks and raised eyebrows when

Research Practices
The study used an approach borrowed from multidisciplinary and applied research, which have much in common with issue-driven interdisciplinarity (Robinson 2008) and action research (e.g., Checkland and 6 Secrecy and Society, Vol. 2, No. 1 [2018], Art. 6 https://scholarworks.sjsu.edu/secrecyandsociety/vol2/iss1/6 Holwell 2004). They both are, in overall terms, change-oriented approaches in which the researcher is involved in addressing practical issues, or in helping to solve problems in collaboration with the participating organization's representatives. The research team worked with Swedish Armed Forces representatives to define the study's objectives -objectives researchers and organization staff believed would have practical and scientific benefits. Such an issue-driven approach can draw attention to certain everyday practices, and can lead to their revision "on the fly." Thus, attempts were made here to reduce the gap between theoretical knowledge and practical application.
The terms "issue-driven" and "action research" were used during the project as ways of describing the overall approach, especially for academic audiences. However, the researchers themselves did not agree on any single way of describing the approach. Some of us preferred the terms and ideas behind an "issue-driven" approach, while others were more at ease with "action research." Whichever term we use is not important for the purposes of this article. What I wish to highlight here is that an "issue" urges participants to define and hopefully to agree on what that issue is -in other words, what is at stake here that is of importance both for the IC and the university? A shared issue and goal, or at least working towards this, strengthens the willingness to participate and contribute to the study. It indicates interest and therefore helps to prioritize these issues over other work issues. It also, I argue, sets the basis for collaboration in this project.
As a social scientist, I am interested in the everyday practices, as well as negotiating the aims of the study, which also offers bits of information that add to the research. A challenge, obviously, is how to negotiate the objectives of the study, and how to communicate them among the participants. A negotiation is a communicative situation and device. It refers to a situation with and about people, their conditions, and their everyday lives. In collaboration, the joint activity must be of benefit to both sides of the partnership for it to succeed. It is therefore important to find the core point between interests and the division of labour. Still, the practitioners should keep doing what they are good at, and the researchers should do what they are good at -challenging both the practical and the theoretical.
The research in this project focused on the activities and phenomena in the everyday situations in which they normally occur, rather than on measuring them quantitatively. Qualitative methods and techniques were used to collect information. As researchers, we carried out what may be called "polymorphous engagement" (Gusterson 1997, 116), which means interacting with informants in diverse ways, locations, and occasions, as well as collecting information eclectically from different sources, using a mix of research techniques. We conducted interviews, field observations, and document reviews. There were also several meetings, experiments, workshops, field exercises, and courses that we observed and collected data from.
In general, my research practice is based on the anthropological tradition. One way to characterize such an approach is as follows: Ethnographers listen, observe, participate, converse, lurk, collaborate, count, classify, learn, help, read, reflect and-with luck-appreciate and understand what goes on (and maybe why) in the social worlds they have penetrated. It is an unspoken methodological paradigm that is generally effective in not scaring away the phenomenon of interest […]. Preserving the apparent naturalness and everyday character of what is being studied is the stock and trade of ethnographic work on the ground (and in writing). (John Van Maanen 2001, 240) Traditional fieldwork offers possibilities to listen, observe, and lurk, but also requires access to the field more or less on a daily basis. In this research project, none of the researchers had unlimited access to the premises of MUST. I visited the premises, but only occasionally, and as a visitor, to meet with intelligence practitioners for interviews and other activities. I did not meet the intelligence practitioners every day, nor did I share the everyday activities of the IC.
Yet, although ethnographic immersion was not possible, other forms of data gathering methods were used and were important. The meetings, courses, and other social gatherings worked as a window into the community's everyday life. Collaboration between parties opened up these possibilities for participation. While several of the activities we participated in were regular happenings within the IC, others were arranged for the purposes of the research project. Some of the activities gathered only intelligence practitioners within MUST, while others gathered even military personnel from Swedish Armed Forces and from abroad, as well as researchers from academia with shared interest in intelligence practices.
Participation in such everyday activities is crucial and leads to substantial advantages for researchers and practitioners, because it usually leads to discussion about work activities as such, and therefore also gives everyone the opportunity to challenge these activities as everyday practices. That is, mutual learning occurs whenever both parties are present. Constantly creating, selecting, and managing new empirical data is a methodological challenge in research (as well as in intelligence work), because the "information retrieval" does not have a beginning and end in the traditional sense. In the project, we considered this process an ongoing delivery of research insights.
As part of the study, we conducted a number of semi-structured interviews with intelligence practitioners within MUST. For these interviews, respondents were selected by the intelligence organization. There was a risk, then, that only certain individuals would get a voice, and only certain viewpoints would be articulated and become explicit. After the interviews, the transcribed interview notes were corrected, completed, and approved by the interviewees. Even though this procedure was mainly done to make sure that the interview data did not include any confidential information, it also worked as a form of quality assurance. Document and image review was also a way to collect information.
Various documents and terminologies are produced as part of the everyday practices within an organization and are therefore important elements to study. These reviews were not carried out independently, but rather in connection to interviews and observations. Writing together represents close interactions between the co-authors. I suggest it may, in fact, be the ultimate form of collaboration between the participants. However, thus far I have not written any reports or articles together with the intelligence practitioners. Yet the analytical work, leading to reports and articles, was to some extent a shared activity-partly through the meetings in which findings were discussed, and partly while writing up the findings. Drafts of the research reports were circulated among a number of intelligence practitioners, and we received valuable comments and critiques. This practice added to knowledge transfer and knowledge production. Ultimately, however, it is the researcher -not the informants/practitioners -who develops and takes responsibility for the scientific representations and results.
Writing, as communication, raises important questions about academic freedom, control of the research agenda, and intellectual property rights, all of which must be addressed if this form of collaboration is to be successful.
One question that needs to be answered each time, regardless of the research project: What can I publish and how? During this project, one report was considered confidential, but other publications were unproblematic in the open literature.

In Retrospect
Collaboration changes participation, which in turn also affects the research practice and the information that is gained. Researchers and practitioners brought different forms of knowledge to the project, and both were actively involved in its research activities. This linkage between practice and research was achieved largely through project activities. When practitioners participate in the process, they expect to gain something in the long run. Given the prerequisites to carry out intelligence work, the time and effort expended on ordinary work tasks and the priorities thereof are challenges to any additional projects in that setting. Therefore, it was an advantage that the project's focus on the dilemmas practitioners face in their organization overlapped strongly with what was already engaging the practitioners. As Robinson (2008) suggests -and as I have experienced in collaborations with other organizations -striving for overlapping activities for both research and practice in the organization is essential for fruitful collaboration. When practitioners share the objectives of the study and feel responsible for it, they also tend to participate in the project differently. They tend to provide different kinds of, and perhaps more layered, information, and thus open doors for further enquiry. This gives a methodological advantage to the overall research project.
The mutual learning through the knowledge production that occurred between researchers and intelligence practitioners was substantial. It emerged from this collaboration either by the efforts of practitioners or by those of researchers. The collaboration meant that research results were critiqued as the study went on. By doing so, the collaboration contributed to knowledge production and helped to provide empirical insights that otherwise might have been difficult to glean. Yet one result of the collaboration was that practitioners could use these insights to spur change directly in their ongoing, everyday work. This also meant that the object of study was changing as we studied it. Although this may be a challenge for research practices and knowledge production, it may also be considered an advantage, as it made visible the pace of everyday activities the intelligence practitioners carry out in their organization. However, the reader should keep in mind that it is not always possible to establish whether or not learning and change have occurred in tandem. In this case, sometimes the activities led to a list of concrete changes to put forward within the organization; other times, more general reflection on practices occurred. As we all know, changes are sometimes easy and quick to make, and sometimes they encounter resistance and take time.
The collaboration described here is an activity based on the circumstances of project activities in research carried out with the intelligence community of MUST. The contributions and challenges of the collaboration presented here reflect this situation. The collaboration was possible despite the secrecy and security issues that are associated with the IC. Hence, I believe that the viewpoints may be transferred and applied to similar settings and research collaborations that take place between academia and other intelligence communities, as well as other organizations, with careful judgment as to which insights might be important in those particular circumstances.

Concluding Remarks
Overall, the experiences of collaboration in this research project correspond with those concerning issue-driven interdisciplinarity (Robinson 2008). Close collaboration in a research project can be time consuming and risky. It takes effort to learn about each other's organization and language to arrive at useful and mutually rewarding findings and results. It takes effort to build good-enough relationships between participants to make collaboration successful. Collaboration with the IC was rewarding and perhaps necessary in order to get more layered information of the everyday dilemmas that practitioners face in their organization. Without a mutual interest in the issues at hand, and without working toward a shared goal and having respect for one another, this research, I believe, would have taken a different turn. However, the challenge is to balance the emphasis on implementing organizational change with the kinds of theoretical and empirical advances typically associated with science.
Collaboration is suitable when change is required, such as in this research project. The mutual learning between the actors feeds into change processes as if it is a seamless, invisible element of everyday life. The potential of this engagement lies in how collaboration contributes here to the framing of change of intelligence practices. Engagement through collaboration aims to enable change to be understood not only as complex, contradictory, and uncertain, but also as a routine and ongoing everyday activity in an intelligence organization. Collaboration offers the researcher not only special access to the studied phenomenon, but also the possibility to change the phenomenon. However, such collaboration raises fundamental ethical issues that must be resolved before entering the field. These issues are complex and highlight various academic, institutional, and personal perspectives. Which organizations should I, as a researcher, collaborate with, and for what purposes? What kind of change will I, as a researcher, contribute to -especially if the possible change lies beyond the research task at hand? Collaboration may not be possible, applicable, or suitable in all fields or to all research questions. In some cases, it may even be strongly advised against.
We can consider collaboration as a vehicle to the field that situates the research practice. Collaboration may be considered an umbrella under which research methods are defined. Furthermore, such collaborations require us to question and reflect on scientific practices. Collaboration, I suggest, is both constitutive and generative -and potentially also transformative.
Consequently, collaboration as practice requires collaborating parties to reflect critically upon the value and significance of both knowledge production and the knowledge that comes of it.
However, collaboration is not a device with certain inherent boundaries or its own inner logics. What we can learn from collaboration, I suggest, is how to figure out things together so as to satisfy various objectives.
Collaboration can be understood as a device for articulating the relation between practitioners and researchers. It also suggests a changing division of labour in research: Practitioners learn about the research process as researchers learn about the practitioners' work practices. Rather than just passing over information to the researcher, the practitioner works with them in knowledge generation. This circulation of social science research practices and techniques across social life can be productive not only for intelligence studies and social science, but also for the various communities involved in research, as this research with the IC suggests.
Yet collaboration requires different types of incentives and institutional support systems in order to succeed. Robinson (2008) draws attention to institutional challenges for issue-driven interdisciplinarity, which are applicable even here. The challenges concern issues of academic freedom, control of the research agenda, and intellectual property rights -but also standards, quality, and evaluation of research, all of which are rooted in overall academic practices for employment and reward systems. So far, quality measures of academic research work against this type of research.
For example, a common way to measure academic success is by scientific publications. There is pressure on academics to publish scientific work in order to sustain and further one's career. Collaboration with an organization requires, as it did in our case, different types of reports and reporting to reach audiences other than merely academics. Obviously, finding a publication channel that satisfies both practitioners and scholars would be appreciated by both parties.
A further question remains: Does a research collaboration imply that the scientist has to "sell" herself/himself as something resembling a free-market entity, or can the parties work together, yet on their own terms? Such collaborations, whatever they might be, should not be a set of "how-to" recipes, but rather a research activity that can have substantial rewards for researchers and practitioners alike, as the collaboration with this intelligence community would suggest.