- Split View
-
Views
-
Cite
Cite
Robert (R M ) Duiveman, Institutionally embedded research; how academic and governmental practices enable and constrain collaborative research, Science and Public Policy, Volume 47, Issue 6, December 2020, Pages 855–864, https://doi.org/10.1093/scipol/scaa049
- Share Icon Share
Abstract
To elucidate how authoritative knowledge is established for better dealing with unstructured urban problems, this article describes how collaborations between researchers and officials become an instrument for conceptualizing and addressing policy problems. A case study is used to describe a research consortium evaluating the controversial practice of ‘Lifestyle’ based housing allocation in the Dutch domain of social-housing. Analyzing this case in key episodes, we see researchers and policymakers selectively draw on established institutional practices—their so called ‘home practices’—to jointly (re-)structure problems. In addition, we find that restructuring problems is not only intertwined with, but also deliberately aimed at (re-)structuring the relations within and between the governmental practices, the actors are embedded in. It is by selectively tinkering with knowledges, values, norms, and criteria that the actors can deliberately enable and constrain the ways a real-world problem is addressed.
1. Introduction
Official: ‘Are residents’ lifestyles stable over time?’ That was not the research question. The research design doesn’t allow you to address that subject, so remove it from the report.
Researcher:What we can say for certain with this research, is that the results of the policy are not what we would have expected based on the conceptual mechanism underlying your policy.
Official: This will get us into trouble. I want it removed from the report! I don’t want a report that states ‘University U finds that ‘lifestyle’ doesn’t work’
Researcher: The report doesn’t say that.
Official: No, but that’s how media will report on this and that’s what local policy makers will remember. I want it removed.
(paraphrased dialog—from fieldwork notes consortium meeting lifestyle 11/06/28 -translation by author)
This testy dialog between a city official and researcher about which research findings should be published, exemplifies many of the tensions and dilemmas around research that is co-produced with policy makers. At first glance, this argument on how to publish the findings of collaborative research seems to be about ‘getting the truth out’ or ‘speaking truth to power’ and for the researchers involved it very well might be. The official, however, is concerned with ‘what ought to be done?’ and the reputation management of his administration. Not able to find common ground, their discussion leads to a rather unproductive negotiation between the requirements of scientific and policymaking practices. This, clearly, is not what governments and universities aspire to when they set up, currently popular, collaborative arrangements such as research consortia, urban experiments, or living labs—in a desire to improve the usefulness and impact of research (Evans et al. 2016; Marvin et al. 2018). Such new knowledge practices introduce transdisciplinary methods to overcome, not reproduce, received ways of addressing policy problems. To succeed in this endeavor, however, the officials and the researchers involved must learn to not only deal with the policy problem but also with each other and—this is the central claim of this article—that success in the latter is a prerequisite to accomplishing the former. This article describes such a new collaborative knowledge practice and analyzes what factors enable and constrain such collaboration. It will allow us to critically review Lawrence Mead’s contested suggestion that to achieve ‘engaged research’, it should be organized from within government (2015). Indeed, I will argue that the major constraint is the actors’ unreflective adherence to, what I call their academic and policymaking ‘home practices’, the everyday organizational practices that define their professional and institutional identity. The case concerns lifestyle-based public housing allocation, and at the time controversial urban housing policy.
1.1 Knowledge for Strong Cities (KSC)
From 2006 onward, the Dutch central government allowed NICIS Institute to invest over 40 million euro in scientific research that was aimed at analyzing and finding solutions to urban issues.1 As one of the requirements for obtaining research funding from NICIS was that researchers and city officials co-produced the research in jointly funded research consortia (NICIS 2006), its major program called KSC provides an excellent setting for studying the collaboration between municipal officials and academic researchers. Between 2006 and 2014, NICIS co-funded over fifty such research consortia. NICIS guidelines for the design of consortia are remarkable in three ways. First, instead of a detailed research design, as is common when applying for scientific research funding, NICIS required partners to elaborate on how they organized their deliberation before and during their collaboration. This provision aimed at providing partners in a research consortium the opportunity to shape the research along the way based on joint learning. A second characteristic of KSC is that partners in the research consortia, including NICIS itself, share the costs equally. This important consideration was to level the playing field between policy makers and researchers. It in effect stated that researchers did not work under a contract, with its implicit principal–agent relationship, but as co-equals with officials and professionals. The third characteristic is that the actors in the consortium have a shared responsibility for the operationalization, interpretation, and dissemination of the research.
1.2 Embedded knowledge practices
Just like in an increasing number of living labs and urban experiments (Evans et al. 2016; Marvin et al. 2018), the NICIS research consortia were designed as a knowledge producing environment. They are sites where knowledge was developed, tested, debated, used, shared, advocated, framed, and much more, in order to better deal with complex urban problems. From previous studies into KSC (Duiveman 2020) , we know that through collaboratively collecting and interpreting data, the KSC consortia developed new knowledge practices. These stand apart from well-known intermediary arrangements between science and policy, which aim at transferring knowledge by bridging or brokering boundaries between academic and governmental communities (Meyer 2010; Stone 2012).
New knowledge practices emerge as researchers and officials collaborate in a new setting with the intention to more effectively address a real-world controversial problem while also remaining attached to different academic, governmental, organizational, and professional practices; their so called ‘home practices’ (Duiveman 2020) . This article aims to describe the ‘unfolding’ problem processing (Wagenaar and Cook 2011) in a new knowledge practice within the KSC program and analyses the consequences of its embeddedness in multiple, interdependent institutional home practices.
1.3 Preview of the case study
The case study concerns the KSC consortium ‘lifestyle’, which is short for ‘lifestyle-based housing Allocation’. In 2009, officials from four municipalities and researchers from University U received KSC funding to establish a consortium to explore the problems and opportunities associated with using ‘lifestyle’ research for allocating public housing. Housing corporations, who develop, own, maintain, and allocate complexes for low-income households in The Netherlands also participated in the consortium. A reason for establishing this consortium was actually that an increasing number of the, recently privatized, housing corporations wanted to use lifestyle research to group residents with similar lifestyles in designated housing complexes in order to better match residents with estates and with each other. However, from the start lifestyle allocation was considered controversial. Its proponents argued that grouping residents with similar or matching socio-cultural preferences in designated housing blocks would lead to residents being more tolerant towards each other, experiencing less nuisance or disturbance by each other or even experiencing better living conditions (Van Kempen and Pinkster 2002; Windel 2008; Nio 2010). Its opponents do not argue whether such lifestyle segmentation would work as predicted, theirs is an ethical objection. It should not even be attempted. To them differentiating between people’s personal preferences in order to match them to their neighbors in housing estates constitutes a type of social engineering bordering on, or even resembling, discrimination based on socio-cultural characteristics. As proponents and opponents stereotyped each other as either ‘believers’ or ‘skeptics’, policies based on ‘lifestyle’ led to divisions among researchers, policymakers, and politicians in the Dutch field of social housing (Van Kempen and Pinkster 2003; Heijs et al. 2005, 2009; Muskee 2010; Nio 2010). It is against this backdrop that a KSC consortium is established. It set itself the task to structure a policy issue the definition of which is—and remains—highly contested.
Let us first review some literature to obtain sensitizing concepts that may be of help when we analyze how a new knowledge practice worked out for addressing this controversial policy.
2. Problems, practices, and translations
2.1 Problems
The interaction between officials and researchers in a new knowledge practice—in case the ‘lifestyle’ consortium—is propelled by a shared interest in defining and resolving a pressing problem. Describing how exactly these actors structure the problem provides a productive beginning for analyzing the content of their interactions. I will therefore trace how problems are structured over time and place (Hisschemöller and Hoppe 1996). To do so, a problem is seen as a discrepancy between an actors’ representation of an actual (or anticipated) situation and a desired situation. We can differentiate how different actors structure a problem and, most importantly, we can map how a problem is restructured over time and space: if all parties involved agree on the facts of a current (or anticipated) situation and the values to apply to establish a desired one, we speak of a well-structured problem. Unstructured problems concern situations where facts are contested (ambiguity) and/or values are disputed (ambivalence). It stands to reason that as more actors are party to a problem it becomes less likely they apply the same facts and values. By describing over time which facts and values are included and excluded, we can analyze to which academic and institutional the research agenda of the embedded knowledge practice becomes more closely or more loosely related.
2.2 Practices
The concept of ‘practice’ has been used a number of times in this article to designate sites where knowledge is developed, and problems are structured. In line with Shove et al. (2012), practices are understood as recurring combinations of the elements ‘material’, ‘competence’, and ‘meaning’. People (as practitioners) doing everyday things, like driving, exercising, or cooking, (re-)make the links between these elements and consequently enact practices. By driving a car, for example, people link specific materials (motoring technology, rubber and steel) with competences (ability to steer, shift gear, and refuel) and meanings (sense of autonomy or the appearance of wealth). It is by doing everyday things like driving or having a meeting, that people carry the practices, which constitute the stable cultural and material reproduction of daily life.
Beside this conceptualization of stability, this approach to ‘practice’ is also helpful for describing change. Change occurs when the links between practices’ constituting elements are broken, (re-)made or replaced. When cars start driving themselves, for example, the material elements that make up the practice of driving are linked to other competences such as programming your destination on a touch screen. When environmental awareness grows the positive meanings previously linked to revving an engine can be substituted with meanings such as shame over pollution. As the result of links between elements being broken or (re-)made, over time and space, practices can change or disappear.
Shove et al. (2012: 15) state that the relation between practices-as-entity (which emphasizes structure and stability) and practices-as-performance (which emphasizes action and change) is recursive: as everyday acts are performed differently, people break and (re-) make the links that constitute the practices by which their actions are structured. The case study in this article will describe how people’s actions can both be structured by a practice—and how action can restructure practices.
Over the course of a day, people enact multiple practices. In the case study, we will differentiate between the new knowledge practice (the research consortium) and the academic, municipal, or professional home practice in which officials and researchers are embedded. Beyond the organizations, the consortium partners are employed by, these home practices can consist of the professional and social settings, which provide people with the discursive, institutional, and material structure that he or she is part of (Duiveman 2020) .
2.3 Translations
For officials and researchers embedded in different governmental, corporate, and academic practices, to develop a shared approach to a controversial policy problem such as ‘lifestyle’, they must work toward an approach that is meaningful in terms of their collaborative knowledge practice and in terms of their institutional home practices—what Grin and Van de Graaf (1996) label ‘congruency’. It is about finding an interpretation of a problem, which leads to knowledge that is simultaneously relevant to—for example—the official writing of a policy paper for the mayor, to the professor preparing a lecture, and to the professional dealing with an upset client. To describe how such congruency is achieved, we trace actors as they give meaning to their own as well as their collaborative actions in different sites under changing circumstances. The concept of ‘translation’ (Latour 1987; Stone 2012) is then used to characterize the work done by actors operating within and between multiple practices.
To describe ‘translations’ we follow actors as they displace knowledge from one practice to another and pay special attention to adaptations in the knowledge that actors apply when moving between practices. This tracing of ‘translations’ draws attention to actors establishing a transformation of meaning. As the perspectives of municipalities, housing corporations, and universities—on the policy of ‘lifestyle’ research for allocating homes—diverge or converge, their interests and relations change, increasing or decreasing the possibilities to pool material resources, meanings, or competencies.
3. Research design and methods
The case study presented in this article describes the unfolding of one NICIS consortium as an embedded practice. The research question is How do academic and governmental home practices enable and/or constrain a collaborative knowledge practice addressing a controversial real-world issue?
The case study is part of a larger longitudinal study into the interactions between researchers and officials within the KSC program. The author of this article (as external researcher) evaluated the collaborative process within various consortia based on document study, observations of consortium meetings, interviews, and a survey. The author was allowed unrestricted access to documents and all meetings pertaining to the design and management of the KSC program and its consortia over a three-year period (2011–4). The overarching study consists of four cases and a survey analyzing how researchers and policymakers establish a practice for knowledge development that is scientifically sound and societally relevant. The perspective employed here emphasizes the interactions within a consortium to reconstruct the actors’ efforts to align their knowledge practices with those of the other partners. For this case study, the research covered over ninety documents. These were reports, agendas, minutes, proposals, and articles produced by or for the knowledge practice. External documents, articles, papers, and publications were also included when relevant for reconstructing the ‘translations’. Three extensive meetings were observed, and multiple interviews were held with six of the central actors: the PIs, three officials and the NICIS program manager.
Two rounds of analysis were performed. First, a number of sensitizing concepts (Bowen 2006) were developed from the literature: problem structuring, translation, practices, and congruency. These concepts were used to trace actors as they structure and restructure the meaning of ‘lifestyle’ in the practices involved and analyze how their enactment of this concept makes, breaks, or reshapes the links within and between these practices. Through abduction (Blaikie 2010), the process of translations in three episodes was analyzed, described, and supplied to the interviewees for a second round of analysis. This consisted of iterative rounds of written and verbal checking of the ‘integrity’ of the case study by verifying it with the respondents (Blaikie 2010: 90). This was done by verifying and validating research findings by discussing them extensively with the people involved in and around the lifestyle consortium (Latour 1987). The case study, as it was eventually constructed, was made available (in Dutch) by NICIS. This serves as the basis for the account of three specific episodes, which for pragmatic reasons will refer to only a limited set of empirical sources.
The case study consists of three episodes. Each pinpoints and represents a period in which the consortium was compelled to re-structure the problem they are researching to be able to move their collaboration forward. By analyzing these, episodes we can highlight how exactly the relations between multiple practices develop as translations are being proposed to achieve closure.
4. Case study: lifestyle allocation in the domain of housing
The NICIS consortium on ‘lifestyle allocation in the domain of housing’ will be analyzed by describing three key episodes. Each episode describes how new events necessitate the consortium to restructure the research problem and how actors re-establish congruency to keep their project moving forward. The relevant policy context of the consortium is outlined in a pro- and epilogue.
4.1 Prologue
In the 1990s, housing corporations in the Netherlands were privatized. The new, not-for-profit enterprises that were created became the owners of hundreds of thousands of one-family houses and apartments with the obligation to make and keep them available to low-income households. Although the corporations were still subsidized and regulated by government, the new liberal policy also provided them with more freedom: on how to allocate living units, how to approach area development, and how to manage investments. A number of the larger corporations began to resemble commercial enterprises, acquiring a larger share of the market through mergers and acquisitions, investing in commercial real estate development abroad and paying hefty bonuses to their new CEO’s.
The corporations were expected to invest in the public domain and in area development. In some cases, this conflicted with their drive for growth and expansion. As Dutch housing corporations were just adapting to their recently acquired identity as (semi-) private enterprises, they were not all that eager to voluntarily invest their financial reserves to improve the urban public domain. On the other hand, the government could threaten them by enforcing the stipulated legal obligations and, if the corporations would still not invest in the public domain, the national government threatened to levy a so-called landlord tax (additional taxes on the rent collected on social housing). So, the corporations were required to cooperate at least to some extent.
Against this background, a number of corporations suggested using a new method for allocating houses: sociocultural segmentation. This method, so they claimed, in fact entailed an investment in the public domain because it would address livability issues—a catch all term for urban nuisance that discomforts or angers citizens. Based on a questionnaire, the corporations proposed to categorize households in terms of their ‘lifestyle’ and claimed that by allocating housing units to people with matching or compatible ‘lifestyles’ they could group tenants in a manner that improved the overall livability. In 2011, thirty Dutch housing corporations experimented with sociocultural segmentation through ’lifestyles’ and in nine out of the thirty-three largest municipalities ‘lifestyle’ criteria were used in housing allocation (VROM Inspectie 2011).
‘Lifestyle’-based allocation in the housing sector was an admittedly controversial policy. The controversy was a public problem as its pro- and opponents had to collaborate on a day-to-day basis to ensure comprehensive urban housing policies for lower income households. What raised the stakes was that the central government was preparing a new social housing act in which it would consider the institutionalization of ‘lifestyle-based’ allocation of houses. In the period that the consortium took shape a very stringent law was in place on how to allocate social housing. These regulations in effect prohibited ‘lifestyle’-based allocation of houses. The above-mentioned nine municipalities were turning a blind eye by licensing the ‘lifestyle’ based housing allocation under the heading ‘Local experiments in public housing’ (VROM Inspectie 2011). As the central government was preparing consultation on the matter of ‘lifestyle’, its proponents and opponents tried to determine the position their corporation or municipality was going to take when they would be consulted by the ministry on the new act.
4.2 Episode 1
In November 2008, fourteen professionals from housing corporations and a researcher from University U were deliberating on the extension of their joint research program. They were considering research into experiences with ‘lifestyle’ and other concepts for socio-cultural segmentation. The minutes of the meeting (24 November 2008) stated that a ‘lively discussion’ ensued. Some of the professionals voiced their moral objections to sociocultural segmentation. There was a discussion on whether ‘lifestyle’ policies constitute discrimination between tenants based on socio-cultural characteristics or whether it just enabled corporations to better know and address the personal and behavioral preferences of their tenants. University U proposed not to delve into the morality of this issue but to simply make an inventory of current use of the ‘lifestyle’ concept. By making such an overview, knowledge gaps might be uncovered, which could possibly, at a later stage, be formulated into research questions.
In this same period, the University U principal investigator (PI) participated in a NICIS Institute workshop on urban housing. There he suggested a research project into ‘lifestyle’-based housing policies. A conversation about this idea amongst some municipal officials led them to formulate questions for a possible consortium: ‘What is the meaning of the concept of “lifestyle” in urban practice? What effects have been achieved with experiments in this area, both in terms of housing, neighborhood management and area development?’ (Nicis 2009). NICIS and University U combined forces and together with four municipalities paired up with fourteen corporations to establish a research consortium within the NICIS KSC program. From the very start, rifts regarding ‘lifestyle’ segmentation began to appear between consortium partners.
Corporations: The fourteen participating corporations were interested in how ‘lifestyle’ might contribute to improving the satisfaction of their tenants and the quality of life in the neighborhood while at the same time reducing management and maintenance costs. This is made clear in the project proposal for the consortium. However, the ‘other’ corporations, the ones that expressed moral objections to this approach and who associated ‘lifestyle’ with enabling discrimination, were not interested in research that measured its effects in terms of livability. According to the PI and an official that was involved, these latter corporations excluded themselves from participating in the consortium.
Municipalities: There were four municipalities who supported the consortium. Two of those, ‘Seacity’ and ‘Rivercity’, participated actively in the consortium meetings concerning ‘lifestyle’-based housing allocation. The officials involved state that both of these cities participated because they expected that the research findings could provide closure for the ongoing normative debate within their organization. From interviews with the civil servant involved, we know that in ‘Seacity’ the civil servants were skeptics while the alderman was known to be a so-called believer. In ‘Rivercity’, the officials state that differences divided the several administrative divisions involved in local housing policies.
University U: The PI is a senior researcher who heads a three-person research team. These researchers all work at a nationally renowned institute with a long history of research and advice in the field of social housing. The researchers took no position in the feud over ‘lifestyle’ based housing allocation—so they inform us in interviews. The view on this concept in the academic field of social housing was summarized at the time by Heijs et al. (2005, 2009) who outlined a triple criticism: based on a literature study they concluded that the concept of ‘lifestyle’ is insufficiently defined, insufficiently reliable and valid, and finally unnecessary because there are better alternatives.
In the research design for the consortium, the researchers problematized the issue of ‘lifestyle’ by recognizing that although it is frequently applied, it remained controversial, both in the research community and in society. This type of situation would normally benefit from a research design that allows for multiple interests and stakes to be explored and synthesized through carefully orchestrating puzzling, powering, and participation (Hoppe 2011; Bammer 2013). The PI, however, explained that he considers factual and value-free research as the best way forward, to circumvent the moral issues that ‘lifestyle’ segmentation raise and, to establish a consortium of partners willing to collaboratively research whether the ‘lifestyle’ instrument worked (interview PI lifestyle consortium). Thus, we see that the PI, in his own words, chooses to redefine the issues surrounding lifestyle more objectively, thereby positioning the research as a factual arbiter between proponents and opponents. The consortium partners indeed agree to this and decide to collaborate in an objective empirical study.
It should be noted at this point that the normative question, which divided the policy field into skeptics and believers—and which the corporations extensively debated in the first meeting—do not have a place in the inventory that the researchers made (University U memo—March 6 2009). It was in fact by ‘organizing out’ contested values that the consortium structured the controversial issue of ‘lifestyle’ housing allocation into a researchable problem and acquired the legitimacy provided by ‘science’. The researchers drew on their scientific home practices to structure the problem; leaving out the values that led to conflicts of opinion. Stating that leaving out certain values is necessary for an objective, scientifically sound approach, added legitimacy to the new knowledge practice.
Now, instead of wrestling with a policy controversy, the consortium could engage in a technical puzzle that is solvable through more or less straightforward scientific measurement. By collaborating with this academic practice, the municipalities and corporations involved in the consortium could use scientific competencies (research methods and knowledge) to engage with ‘lifestyle’ policies in a manner that circumvented the well-known moral issues. Drawing on the scientific home practice of University U, the officials in the consortium were enabled to delve into instrumental questions (how does ‘lifestyle’ work, when does it work, etc)—allowing them to provide new meanings to the concept of ‘lifestyle’ (Consortium lifestyle project proposal 21 August 2009; Nicis Lifessytle and Branding progress report 2009–10).
By restructuring the problem—making it ‘an objective puzzle that can be solved by research’ instead of ‘a controversy to be dealt with by politics’—the consortium reshaped relations between the practices involved. Thereby the civil servants and professionals became dependent on scientific method to solve the question of does ‘lifestyle’ work? It was by proposing research aimed at ‘getting the facts straight’ that the researchers acquired a central position in the policy controversy. Officials with a stake in ‘lifestyle’ policy acceded to this, as it allowed them to research the effects of the policy while the moral discussions were put ‘on hold’. Kicking the can down the road so to speak.
In the effort to provide clarity by ‘getting the facts straight’, the consortium also restructured the issue in a way that conforms to the interests of some officials, but clearly deviated from that of others. It led to the formation of a coalition of actors within the policy field that concurred with a distinct approach to ‘lifestyle’ allocation. Quite understandably, corporations and municipalities with ethical reservations were not inclined to provide the means for measuring the effects of an instrument they rejected on moral grounds. So, beyond linking specific scientific and policy making practices, setting up the new knowledge practice also led to (self-)exclusion of certain practices.
Civil servants from municipalities that were still undecided about their position regarding ‘lifestyle’ (such as in ‘Seacity’ and ‘Rivertown’) decided to participate in the consortium. According to the officials involved they took part because the project proposed overcoming this issue by performing research that had been designated as ‘objective’ and could therefore act as a neutral arbiter in the disputes within their municipality.
This episode entailed a ‘translation’ that consisted of the redefinition of a controversial policy problem, making and breaking links between competencies and meanings involved and the inclusion and exclusion of practices that have a stake in addressing ‘lifestyle’-based housing allocation. Furthermore, we described a reassembling of the relations between the practices in the policy field, providing University U with a prominent role as the controversy is now understood in terms of a puzzle that the researchers were best qualified to solve.
4.3 Episode 2
The ‘morally dissenting’ actors had already excluded themselves. Still, the consortium consisted of actors with widely divergent interests; researchers who wanted to empirically study a scientifically contested concept, officials from corporations who aimed at gaining leverage in an administrative struggle with civil servants and politicians over ‘lifestyle’ as a new policy instrument, and municipalities that craved objective knowledge to get closure on the controversial issue. As the consortium went to work on the research, we will see that the restructuring of problems and practices, which allowed these actors to collaborate, did not create closure but was rather just the beginning of a new round of controversy.
The collaborative research consisted of three consecutive studies into ‘lifestyle’; first a study reviewing the history of the concept, then an exploration of the concept’s use in urban area-development, and finally a couple of case studies researching its application in housing allocation (Consortium lifestyle project proposal 21 August 2009). As the researchers were designing the research for the case studies that made up the third part of the research, they encountered a paradoxical question: How to evaluate policy without introducing the contested values they previously designated as subjective? The consortium needed to resolve this conundrum to prevent the normative issue—that was excluded in the first episodes—from resurfacing and threatening their collaboration. Additionally, the researchers needed to find corporations willing to open up their properties to a survey with which the researchers could establish the results of ‘lifestyle’ housing allocation. Obviously, getting this access would be easier if the research—once again—circumvented the moral sensibilities.
Soho, one of The Netherlands largest housing corporations owned an estate where the consortium wanted to do their case study research. The project leader of this estate was recently promoted to be the manager responsible for ‘lifestyle’ allocation policy at Soho—tasked with advocating the further use of the instrument with divisional corporation directors and local aldermen. When this Soho official was approached by the researchers, he responded positively. In an interview, he admitted that he was keen for this renowned university to prove the instrument’s success. After debating the pros and cons of ‘lifestyle’ on a daily basis with his colleagues, civil servants, and politicians, the housing official was familiar—and admittedly annoyed—with what he labeled as the unfounded moral objections. The need to overcome this resistance was key for him to participate in the consortium research. The Soho official, in other words, collaborated in the new knowledge practice to bring closure to the normative debate in the policy field as well as within his own organization by having scholars from the renowned University U objectively establish whether lifestyle policy achieved its goals in ‘his’ properties.
The civil servants from the municipalities that made up the consortium also supported the case study approach. Officials from ‘Seacity’ as well as ‘Rivercity’ considered the normative issue a nuisance; an obstacle to effective policymaking and problem solving. They wanted to get the facts straight in an effort to move beyond the current debate between ‘believers’ and ‘skeptics’. In the course of multiple interviews, civil servants from Seacity and Rivercity explain that they knew their alderman as well as their colleagues put trust in University U, which meant that by collaborating with these researchers on case study research they were working toward closure, enabling them to move on.
Evaluation without values is of course an oxymoron. However, to study the results of ‘lifestyle’ allocation the researchers needed a tool for measuring the effects of the policy. Yet deciding on which effects to monitor would require the specification of relevant norms, and bring the divisive normative issue back on the table. To proceed, the researchers and officials needed to decide what facts to measure when studying the issue of ‘lifestyle’ allocation within their collaborative knowledge practice without breaking the relations with, and between, the home practices they drew upon—for example, by involving the contested values that opponents of lifestyle adhere to. So, here once again, the consortium needed to make a successful ‘translation’ to arrive at a new structuring of the problem that allows them to continue their practices.
A way forward was found by measuring the effects of ‘lifestyle’ allocation per case study in terms of the local policy in which the program was instrumental. This meant housing corporations were approached with a survey that would measure to what extent ‘lifestyle’-based housing allocation had contributed to the policy goals it was intended to achieve. In other words, this was not an evaluation in terms of the generic goals associated with the concept (like livability or tenant satisfaction). Nor would the researchers inquire into other, more questionable effects associated with ‘lifestyle’, most notably discrimination. Actually, the preclusion of generic and questionable norms and criteria made the survey resemble a local policy evaluation: ‘Does the instrument realize the goals it (locally) intends to achieve?’ It was by applying this limited perspective that the effects of ‘lifestyle’ allocation could be measured, while at the same time keeping a safe distance from the contested normative issue. Consequently, the consortium partners agreed to this approach and the participating corporations—so the Soho spokespersons explains in an interview—were willing to open up their properties for measuring the effects of ‘lifestyle’ allocation.
This second episode again describes a ‘translation’ that consists of restructuring the problem in a manner that allows the knowledge practice to move on, to restructure relations between different home practices (now including properties with ‘lifestyle’ allocation) and to produce knowledge that could be displaced as each actor expected the results to be meaningful in terms of his or her own home practice. The official from Soho for example was motivated to collaborate in the new knowledge practice as he wanted to draw on the competencies of a scientific practice. If University U confirms what he already senses from experience to be true—‘lifestyle’ works—the policy attains a new, positive meaning, legitimizing its further use in his home practice, increasing his competence to win over other colleagues, civil servants, and politicians.
The researchers stated in interviews that they are acting in a manner fitting to the scientific practices they represent; they clarify a societal issue by developing knowledge, getting the facts straight on an issue that had been the subject of discussions but not of empirical studies. That the corporations were allowed to provide the tools with which to measure whether ‘lifestyle’ worked in their properties does not bother them, as it left them sufficient freedom to address the questions relevant to their home practice: an empirical verification of the academic view that ‘lifestyle’ is ill defined, both unreliable and invalid, and unnecessary (Heijs et al. 2005, 2009). They were confident that in the end they would have the competencies to provide meaning to the instrument that was relevant to their home practice.
The municipalities did not interfere with how the effects of ‘lifestyle’ were measured, as long as it was done in an objective and competent manner (interviews with civil servants Sea City). The value of the knowledge practice to them lies in its capacity to provide closure. Objective research by University U was considered sufficient to provide meaning to the policy in a manner that carries enough weight to either establish the use of ‘lifestyle’ within their municipality or to reject it (interviews civil servants River City).
This second episode described a further structuring of the problem. By tracing actors back-stage so to speak, we could see that they acted out of differing motivations and interests. The research into ‘lifestyle’ had a different meaning to each actor’s home practice and it had to cater to them all in order for the collaborative practice to succeed. The choices made with regard to the collaborative research were therefore as much aimed at studying and analyzing the substantive issue of allocating housing through ‘lifestyle’, as they were aimed at (re-) structuring the problem in a manner that could keep all the related parties interested and involved, and allowed the officials to participate in the knowledge practice in a manner that is meaningful to their home practice.
So far ‘translations’, ensuring that practices and problems are all kept in line, have been successful. However, they come at a cost. They paper over differences. In the next episode, the partners will have to collaboratively deal with the startling results of their survey.
4.4 Episode 3
Corporation Soho applied ‘lifestyle’-based allocation and management in three adjacent complexes to increase tenants’ satisfaction within their housing and neighborhoods (Windel 2008). Surveying these properties, the researchers found an anomaly in the distribution of ‘lifestyles’ within the three complexes. After having allocated x-number of houses to tenants with a certain ‘lifestyle’ over a certain period of time, you would expect to find that number of tenants living in the complexes. This was not so. The number of respondents with a certain lifestyle did not correspond to the number of tenants with that ‘lifestyle’ allocated to each complex. The researchers stated that the ‘placement’ of tenants with a certain ‘lifestyle’ did, over time, not lead to a corresponding increase of residents with that ‘lifestyle’. Reporting on this to the consortium in a meeting on 7 April 2011, the researchers wrote that ‘ .. we must place serious question marks with regard to the stability of the ‘lifestyle’ instrument, a conclusion which also matches earlier comments regarding the concept .. ’. They referred to publications in scientific journals in which they, as well as other researchers, voiced their concerns about the stability of the ‘lifestyle’ concept.
Between April 2011 and June 2011, the consortium partners discussed the case study report drafted by the researchers. Much of the discussion was centered on the phrase ‘The instrument is less stable than assumed and this [..] limits the usefulness of the instrument..’ This phrasing of the results triggered a divergence of problem definitions within the consortium. According to Corporation Soho all that the results of the survey made clear is that reductions in nuisance or increases in satisfaction could not be assigned to the current segmentation of ‘lifestyles’ in the complex. No matter how surprising this may be, to the corporation’s spokesperson stated at the meeting on 28 June that this finding was of no consequence when evaluating the instrument; after all, even if the ‘lifestyle’ of tenants was not stable over time the allocation instrument could still contribute to an increase in residential satisfaction. According to the Corporation Soho spokesperson, it was irrelevant whether ‘lifestyle’ is stable, as long as it produced the intended outcome. From his perspective, the incapacity of the researchers to understand or explain the current distribution was a purely scientific matter with no relevance for policymaking. Instead, the spokesperson suggested a focus on other findings in the concluding chapter. Based on a different selection of statistics in the report they could conclude that those who were allocated correctly in terms of ‘lifestyle’ were satisfied with their residential surroundings.
In the June meeting, the researchers disputed the structuring of the problem suggested by the Soho official, as well as the claim that the actual distribution is not meaningful to policy making. Even if there were sufficient grounds for a positive conclusion about housing satisfaction in the complex, which they contested, the current distribution would make it impossible to formally explain this as the result of ‘lifestyle’ allocation. Moreover, the fact that ‘evidence-based’ interventions such as renovation and intensive management had recently been applied in the surveyed complexes made it plausible that any increase in tenant satisfaction should at least in part be attributed to these. In fact, the researchers were considering mentioning the role of these other interventions in the report. To the spokesperson of Corporation Soho such a comparison between interventions was irrelevant as well as misguided and would constitute a breach of the agreed upon collaboration. It was irrelevant because a lower-end, run-down housing complex like the one in the case study would always require multiple interventions. It was misguided because the real-world policy choice was not either-or. The question was whether ‘lifestyle’ allocation was a positive addition to the known instruments, not how it compares to them.
While the civil servants in the consortium refused to get involved in the discussion on the stability of the instrument (a purely technical matter) they did take an interest when the discussion turned to whether to include a comparison between interventions. They argued that because of the high costs, aldermen preferred not to apply the expensive evidence-based interventions. The officials laid out their problem structure by arguing that to municipalities that it is important to know whether ‘lifestyle’ allocation did better or worse in terms of satisfaction, against lower costs, than other interventions.
The discussion on stability and comparison continued over multiple sessions. In them, the Corporation Soho spokesperson argued—to no avail—that the research was methodologically flawed (selective nonresponse); the research was useless and harmful (journalists will capture the question mark on stability as ‘lifestyle’ does not work); and finally the spokesperson threatened to pull Soho’s support and forbid publication (meeting June 24 and interview with Soho spokesperson). Between meetings, while scrutinizing their data, the researchers uncovered an anomaly in the data analysis. The publication of the report was delayed. The final draft was discussed by the researchers and corporations’ spokespersons, and subsequently published. The contested paragraph questioning the stability of the instrument was preserved in the final report and made public in 2012.
Previous episodes showed that a problem was structured to accommodate a coalition of practices willing to invest to keep a collaborative knowledge practice moving on. Now we see that structuring out the value dissent that enabled this collaboration in the first place did not lead to solving the policy controversy (the differences that were previously papered over resurface). Additionally, in this third episode, a further round of restructuring problem and practices was required, due to the ‘unexpected’ and due to some ‘undesirable’ results of the survey; the situation ‘talks back’, as. Schön (1983) famously put it. The data found in the survey had a different meaning to each actor, their reactions depending on the home practice to which the new insights have to make sense. There was no longer a congruent problem that unified interest in the common knowledge practice, and the relations between actors became strained. The ensuing interactions—propelled by the ‘event’ of unexpected data—illustrated most explicitly how actors drew on home practices when re-structuring of the (research)problem. They proposed to include and to exclude the normative issue to support and to suppress the report on stability and argued for and against comparing ‘lifestyle’ with intensive management. Each of these options for restructuring the problem drew on different home practices and, more importantly, the actors empathized how choices will affect their capacity to displace insights from the joint knowledge practice to these home practices (will the results be published, will they be accepted and debated, publicized and disseminated or will they be silently filed in a drawer?) strengthening or weakening the bonds between them.
It was eventually the interpretation of the researchers that prevailed and that was published in name of the consortium. This makes this episode stand apart from previous ones. We did not see the consortium drawing strategically on the home practices involved to arrive at congruency. Instead, it was the researchers, who at the outset restructured the policy controversy into a puzzle only they could solve, that decided how the new knowledge practice finally structured the problem.
4.5 Epilogue
In the autumn of 2011, it became clear that the new national Social Housing Act would not allow housing allocation based on ‘lifestyle’. The Minister stated moral objections to sociocultural segmentation: ‘…the fact that some representatives of a group cause nuisance does not mean that other people with similar social characteristics behave in the same way. The application of such a provision [of ‘lifestyle’, rd] will, in practice, lead to discrimination against groups of (potential) tenants which is not permitted and is considered undesirable’ (translation by author).
The Minister did not value the ‘lifestyle’ instrument in terms of its local policy effects but in terms of the risk, it poses for sociocultural exclusion, thereby giving center stage to the values that were central to the normative question previously excluded by the consortium. This outcome demonstrates that the strategies the consortium used to avoid crippling conflict in the end were of limited value because ‘lifestyle’ allocation was prohibited by the central government on moral grounds—precisely the value aspect of the problem that the consortium partners unsuccessfully tried to suppress. It was exactly through exclusion of the values by which a relevant research problem structure was assembled that the knowledge practice became less relevant to the policy field.
5. Analysis
The central question of this article, How academic and governmental home practices enable and constrain a collaborative knowledge practice addressing a controversial real-world issue, was shown in the description of the three episodes. The case study described the knowledge practice as an on-going process of ‘translation’. In each episode, there were events that required the actors to re-assemble their initially meticulously articulated group as well as the problem structure. By analyzing these, ‘translations’ we see officials and researchers collaborating in knowledge practices breaking and (re-)making links between the elements that constitute their home practices. While (re-)structuring problems, they are actually mediating the distances between the institutions they are embedded in and the practices that are developing within the research consortium, increasing or decreasing them depending on the requirements of their agendas.
The ways actors dealt with this struggle was described as an on-going process of ‘translation’. In each episode, there were events that required the actors to re-assemble their initially meticulously articulated group as well as the problem structure. By analyzing these ‘translations’, we see officials and researchers collaborating in knowledge practices breaking and (re-)making links between the elements that constitute their home practices. While (re-)structuring problems, they are actually mediating the distances between the institutions they are embedded in and the practices that are developing within the research consortium, increasing or decreasing them depending on the requirements of their agendas.
The case study described how the norms actors use for evaluating knowledge that is developed in a practice is not only derived from established home practices but can emerge from the process of problem structuring itself. Unexpected research findings may change how actors appreciate a certain structuring of the problem and make them willing to deliberate and accept an alternative. The Soho Corporation actually aimed to capitalize on this phenomenon. If the university research structured the controversial policy by taking out the value conflict—which they initially seemed to do with their objective, value free measurements—the knowledge practice would contribute to restructuring the problem in line with the views of Soho. Capitalizing on the university’s prominent place in the policy field, the Soho Corporation thus aimed to use knowledge development to restructure how municipalities and other corporations address housing allocation through ‘lifestyle’. It is in this light that we come to appreciate the fierceness of the debate over the phrasing of research findings; it is not so much about what formulation is more or less truthful in relation to the surveyed housing complexes, it is about the manner in which the policy problem will be structured in the policy field and, thereby, the ways in which corporations and municipalities will relate to the policy problem and each other. In some cases, this seems chiefly motivated by political expediency (Sea City and Corporation SoHo) and in other cases (University) the drive to arrive at publishable insights.
This knowledge practice, in other words, does not only contribute to the development of knowledge but it also influences the institutional environment in which it is embedded. For example, by highlighting the facts relevant to one policy adversary but not the other, and thereby framing the debate in a network to conform to some interests while ignoring others. Relevant knowledge in such a setting is not attained by tailoring the content of research at the outset to the needs of a certain home practice or group of practices. We see an ongoing performance by actors trying to construe and mend a relation between problems and practices in which neither is fixed, and both can change and adapt under influence of the other.
Initially, this seems to take place by actors explaining to others what is more or less relevant to their home practices. Over time, we see that what actors consider to be of interest to their home practice is not stable but changes; it unfolds over time. This is most clearly illustrated in the third episode where actors discuss whether to include a comparison between ‘lifestyle’ allocation and other interventions that can improve livability. The arguments the actors provide here do not comply with the criteria they previously offered in name of their home practices; up to this point, the civil servants stated that for municipalities to appreciate the knowledge practice outcome, its decisions must be based exclusively on scientific method and arguments. Moreover, it is exactly because anything other than scientific values and judgments were formally excluded from the research that the municipalities were interested in participating. In this third episode, however, the civil servants agree on the inclusion of a comparison of different interventions because it is relevant from a political perspective, namely to allow local aldermen to cut costs. Taking up such a comparison in the research means the civil servants now do support a proposition to involve new values (economic) and judgments (political) into the project.
This shift in position is instructive because it is an instance where officials deviated from the criteria they previously claimed where central to their home practice in order to come to a more relevant outcome in terms of that home practice. In other words, in abandoning the criteria about ‘scientific method and arguments’, a possible way of addressing a problem can be transcended so the underlying issue might be addressed; re-framing the situation so it complies with the alderman’s agenda of cutting costs.
With this episode a powerful phenomenon becomes manifest; restructuring problems is not only intertwined with, but deliberately aimed at (re-)structuring relations within and between the institutional home practices that they are embedded in. Actors both draw on the proceedings and characteristics of home and of knowledge practices to influence each other, thereby enabling and constraining what meanings and competencies are linked to a substantive problem such as housing allocation. By strategically drawing on practices, tinkering with knowledges, values, norms, and criteria, actors can deliberately enable and constrain the ways a real-world problem is addressed. As a consequence, knowledge intended to be instrumental to municipal practices governing a problem can actually come to govern those practices.
6. Conclusion
6.1 Key findings
This article set out to analyze collaborative knowledge development in currently popular institutionally embedded knowledge practices such as urban living labs, urban experiments, and research consortia. A case study shows how the characteristics of the academic and governmental practices in which these practices are embedded, enable, and constrain the ways a policy problem is addressed.
Portraying embedded practices as an ongoing process of problem structuring illustrates how actors selectively draw on their home practices to influence the proceedings and characteristics of the knowledge practice. It is through strategically drawing on both home and knowledge practices that, depending on circumstance and agenda, the actors can increase or decrease the distance between governing practices in the policy field. By highlighting facts and values, actors, moving between practices, use the characteristics and proceedings in one practice to influence relations in the other, thereby enabling or constraining the ways an actor, a practice or a policy field address a real-world problem.
Based on this approach and insights, an interesting contribution can be made to the ongoing debate on the relevance of research for policymaking (Daviter 2015; Mead 2015; Newman and Head 2015). Mead (2015) lists the reasons for the limited influence of policy analysis on policy making, and urges scholars to improve their capacity to collaborate in policy making—more specifically to improve their ability to ‘translate’ research findings to be of use in the everyday policymaking practice. Newman and Head claim that such limited influence results from adhering to Caplan (1979) who considers research relevant to policy making when it can be made instrumental to it. Appreciating that research has a broader impact, Newman and Head advise to invest in cross-boundary collaborations; improving knowledge transfer and knowledge brokering to intensify interactions and convey knowledge from research in formats digestible in the course of regular policy making (Newman and Head 2015).
By analyzing ‘translations’ as applied in this article, we can re-conceptualize the two-community approach and thereby contribute to this debate. By following researchers and policymakers as they struggle and manipulate multiple knowledge claims, values, and interests, we have come to see how knowledge and groups shape each other. From this, I derive two objections against making research relevant by organizing it ‘from the inside’ of government (as suggested by Mead 2015).
First of all, because (urban) policy making is increasingly developed in between and ‘outside’ government. Second, because the value of research for policy does not result from the integration of these two practices as such, but from organizing them in such a way that they can complement, enlighten, conflict with, and inspire each other. To the officials in the case study, for example, the relevance of lifestyle research was contingent on how new findings and insights (meaning) enabled and constrained municipalities and housing corporations to allocate (competency) social housing complexes (material). In other words, it is the capacity to break or (re-)make links between the specific ‘meanings’, ‘competencies’, and ‘materials’ that constitute their home practices (Shove et al. 2012), which determines whether research is relevant to the policymakers.
Better connections between research and policymaking (Mead 2015; Newman and Head 2015) can provide important but not sufficient conditions to achieve relevance. To capture how relevance is achieved in practice, I suggest we focus on the performance of Translations. Translation with a capital T to allude to the concept coined by Latour (1987) and as an enrichment and a synthesis of the ‘translations’ advocated by Mead (2015) and the linkages proposed by Newman and Head (2015).
6.2 Further research into knowledge governance
This case-study shows actors in a knowledge practice (re)structuring an urban problem by selectively drawing on their home practices and thereby (re)structuring relations within and between those practices. This approach highlights how a knowledge practices can operate as a mode of governance (compare Grin and Loeber 2007), an idea that is central to scholars writing on knowledge governance (Van Buuren and Eshuis 2010; Gerritsen et al. 2013). Studying knowledge governance by analyzing ‘Translations’ could contribute to uncovering how learning leads to changing relations and regimes that influence power, and how such changes may provide impetus and conditions for collective learning (Grin 2010).
Acknowledgements
The author is grateful to John Grin and Hendrik Wagenaar for reviewing previous versions of this article and their suggestions for its improvement. He would also like to thank the interviewees for their cooperation.
Funding
This work was financed by the University of Amsterdam with support of NICIS Institute as part of a PhD study.
Conflict of interest statement
No potential conflict of interest was reported by the author.
Note
1. NICIS Institute was a temporary not-for-profit knowledge center founded for stimulating state-of-the-art scientific research on urban issues in close collaboration with Dutch municipalities. After a merger, NICIS is part of a larger knowledge center called Platform 31.
References
NICIS (
Nicis (
VROM Inspectie (