Empirical Legal Research in Action

<cn>7. <ct>Cases under construction <au>Irene van Oorschot and Peter Mascini <a>INTRODUCTION In this chapter, we seek to demonstrate the uses of the case study for sociolegal research on criminal law. The case study, first and foremost, is less a single research method than a research approach. Unlike experimental setups or survey research, case studies are often conducted using a variety of methods and a variety of source materials. And while survey and experimental research tend to want to cover many cases, the case study, in contrast, tends to concentrate on only one, or a highly limited number of cases. The case study is “an in-depth, multifaceted investigation, using qualitative research methods, of a single social phenomenon. The study is conducted in great detail and often relies on the use of various data sources.” Its object “can be an organization; it can be a role or role-occupants; it can be a city; it can even be an entire group of people.”1 Case studies may consist of single or multiple cases (for example, a school innovation at different locations) and may contain single (for example, a work practice) or multiple units of analysis (for example, individual employees in an organization).2 It is difficult, then, to generalize about case studies: The questions of precisely what constitutes a relevant and important case and precisely how one should go about using which method for data collection need to be addressed in the context of a research question. As such, the case study requires a certain prudence—or, in the Aristotelian phrasing, phronesis—that contrasts with a more technical attitude that is often drawn on in survey research or experimental setups. With the case study there are no foolproof recipes or catchall solutions, for it lacks the safety afforded by standards of large-N data collection or the controls included in experimental setups. It does not isolate the phenomenon in question but seeks to identify interconnections, mechanisms, and change. In that capacity, the “case” of the case study is not a passive repository of information, but rather an unfolding and constructed thing that speaks back to the researcher—sometimes in ways that rearticulate one’s theoretical purposes and one’s methodological choices. Important decisions need to be made throughout the whole process, taking into account one’s own theoretical expectations, the various uses of different methods, and the limitations of one’s access to the phenomenon under study. 1 Joe R Feagin, Anthony M Orum and Gideon Sjoberg (eds), A Case for Case Study (University of North Carolina Press 1991) 2. 2 Robert K Yin, Case Study Research: Designs and Methods (Sage Publishers 1989) 46–59. 1 Precisely because it is unclear how one should decide on studying which “case,” and precisely because one faces a choice between very different research methods, this prudence consists first and foremost of the willingness and capacity to critically and reflexively interrogate one’s own choices, position in the field, methods, theories, expectations, and claims. Working through the experience of doing a case study of a Dutch criminal court, we will demonstrate the intricacies of making such choices and their consequences for the “case” we were eventually able to make. In so doing, this chapter is not to be read as a set of hard and fast rules for doing a case study, but rather as a realistic and practical demonstration of one of its possible applications. In order to “make our case” we have used as our guiding research question how the judges of a magistrates’ court in the Netherlands arrive at their decisions as a matter of everyday work practice. How do they come to interpret evidence and “see” cases clearly from the moment they open a case file, and what do they do to finally arrive at a verdict? How is this unfolding work practice distributed over different actors and materially mediated by the case file? By opening the black box of judicial decision making, we aimed to complement survey research and experimental research design, which tends to problematically isolate the impact of legal and extralegal factors on sentence outcomes and to ignore the processes and mechanisms that take place in between. Moreover, we also sought to complicate juristic understandings of adjudication and sentencing that understand these activities as purely cognitive or intuitive leaps in the rule-governed dark.3 Instead, we were interested in the activity of decision making as a matter of everyday work. Therefore, we have stuck closely to the daily work practice of judges: their holistic approach to cases; their careful weighing of evidence and the personal circumstances of the defendant; and their local methods of arriving at a verdict. <a>LEGAL CASE MAKING We start our discussion of a case study in the making with legal practices of case making. There is much to learn from practicing jurists at the level of practice when it comes to doing a case study. Like those wanting to carry out a case study, practicing jurists tend to be concerned with the specifics of a case, the delineation of “the case,” the limitations of various sources of information about “the case,” and the question whether this or that is a case of 3 Paul Scholten, Dorsten naar gerechtigheid (Thirsting for Justice) (Wolters Kluwer 2010); HLA Hart, The Concept of Law (Oxford University Press 1994). 2 something else. First and foremost, however, legal actors tend to treat cases as entities “under construction”: as entities that involve active work of selecting, weighing, evaluating information; of testing out possible scenarios; of grappling with surprising elements. It is for that reason that the following vignette, taken from the first author’s fieldwork, is such a helpful guide in our demonstration of the case study: A case study requires the researcher to be attentive to its constructed character. The first author of this chapter invites you, then, to join her in her “field”: the offices of lower criminal court judges in a court in the Netherlands. This research field is embedded in the moderately inquisitorial Dutch criminal law system, in which judges function both as fact checkers of the evidence and charges in the criminal file and as sentencers. The court receives a steady stream of “cases” from the public prosecutor in the form of case files. Clerks and judges prepare these cases prior to the court hearing. The practices studied here are those of lower magistrate judges (politierechters), who adjudicate and sentence only the relatively mild cases (cases with offenses punishable up to one year in prison). On a morning in early 2014, the first author of this chapter finds herself in the office of one of these judges. <quotation>The judge I am observing today, Judge Cohen, is slightly weary. I asked her if I could observe her preparation for tomorrow’s court session, but she worries that “everything I do is just not interesting enough.” I try to reassure her, but she goes on to confess that she finds it a little strange to have me sitting there, peeking over her shoulder to watch her taking notes and click through the digital files. “But as long as you think it’s relevant for you I don’t have a problem with it,” she concludes uneasily. I tell her everything she does is interesting to me, and try to draw her attention to the digital files on her computer screen. The first case scheduled for tomorrow, she tells me, is a case of theft involving two defendants, defendant Peter and defendant Shelley. Peter and Shelley are accused of stealing a mobile phone from a woman’s purse. Somewhat irritated, Judge Cohen clicks through the digital file looking for the evidentiary, written materials. “What is so strange about this case is that these two are both scheduled for the same time to appear in court, but their charges differ somewhat. That’s something to check up on, I think.” Her eyes wander over the digitized file but come to a halt when they encounter highlighted text on the screen. The clerk assisting her has digitally marked important phrases in the file for her. Taking down notes to the police on a copy of the official summons, she works through several of these statements: those of the defendant, of the victim, and of the police officials 3 themselves. “What’s is unclear to me because of the way the file is structured now is how they ended up with these defendants. What is the link between the offense and the defendants? Both of them deny any involvement, I see. . .” Again taking down notes on the copy of the summons, she tells me: “I just want to have something on paper. Even when it isn’t marked yellow [by the clerks] I write some things down, like the defendants’ personal circumstances. Just so I know for tomorrow, if I am to find them guilty and have to decide on a sentence.” Eventually, she encounters a witness statement. “God, this is so unclear. Now he [the witness] is talking about a very tall man, then he is talking about a guy with a small beard, and then again someone with a red shirt . . . Ah, I see. The guy with the red shirt is the tall one. I get it now. Still, I don’t really know how the police ended up with the first defendant to start with. . .” After a while, Judge Cohen closes the file. She is not entirely satisfied with its contents. First, it is unclear how the police ended up interrogating defendant Peter—but did they have reasonable cause to suspect any involvement on his part? Second, the witness testimony is rather incoherent, and the police have failed to include in the file footage from a security camera in the bar in question. Could there be definite proof on that tape? If so, why have they not included it in the file? Last, it is unclear whether both defendants in fact collaborated to steal the phone from this woman’s purse, as suggested by the official charges. All in all, the case puzzles Judge Cohen: neither the facts nor their legal qualification are quite straightforward at this point. Tomorrow’s court session, including as it does an interrogation of the defendants, will have to help her make a decision.</quotation> Weaving together reading, writing, and commenting on the file, Judge Cohen’s work practice illustrates some of the complexities of the legal construction of a case as well as some of the conundrums involved with doing a case study in the social scientific sense. While social scientists do not typically seek to adjudicate, let alone sentence their objects of study, there are affinities between the two practices of “case construction.” First, based on the case file, Judge Cohen has to establish what the case is about: She has to distinguish signal from noise, relevant information from irrelevant background. She also has to interpret and weight the evidence presented in the file and to link it to the charges and the person of the defendant. Although she is highly dependent on materials provided to her by the public prosecution, it is only as a result of her own actions, reflections, and interpretations that she starts seeing what the case is about. Sometimes these materials suffice, but not so much in this case: Judge 4 Cohen is rather unhappy with the fact that the case file does not seem to include a potentially incriminating piece of evidence, that is, the security camera footage. The case file both tells a story and silences other stories; it seems unfinished. Therefore, Judge Cohen intends to use the trial to interrogate the defendants in order to elicit the additional information she believes is needed to make sense of this particular case. Second, throughout her work practice, she must be aware not only of “the facts,” but also of their legal qualification. Even if that phone was indeed stolen, which of the two did it? And then: If the phone was stolen by one and given to the other, does that mean the defendants actually collaborated? Can she make that case? This is the zone of tension she has entered: Yes, something has happened—but is its existence to be proven beyond doubt? And can that event be qualified legally? In her engagement with “the case,” Judge Cohen approaches commonsensical understandings of it. Merriam-Webster tells us that “case” is derived from the Latin root cadere (to fall) and refers to something that “has befallen” or “has happened.” In that sense it is a thing in and of reality, out there. On the other hand, “the case” is also a constructed entity. “Making a good case” is a phrase evocative of the activity of assembling and recombining information into a persuasive whole. “The case,” in common parlance, tends also to be evocative of a broader category of things: a patient can present a “case” of pneumonia, or an event can represent a “case” of a (legally qualified) theft. Judge Cohen’s work practices similarly suggest that a case is, first and foremost, “under construction”; she has to select, interpret, and weight the information that is available in the case file and to determine whether or not it is sufficient to arrive at a verdict. Moreover, it is up to her to decide whether or not the particular case exemplifies or refers to something else, that is, a broader category or kind. These components of “the case” are also crucial to the “case” in the case study. On the one hand, those doing case studies wish to study “cases” as events, organizations, individuals, or communities in the world. We may choose courts, professional communities, or interest groups as the object of our study. On the other, “the case” we make is always the product of decisions and questions. But how do we end up with this case, rather than another? How do we investigate? What materials, methods and sources do we use? How do we draw a boundary around “our case”? How do we construct claims about these cases? <a>HOW TO CHOOSE A CASE? 5 Let us start at the beginning, with our choice of a “case.” While survey research and experimental designs often take for granted the existence of their cases “out there,” the case study destabilizes the idea that “cases” are simple things “out there” that we can measure or tap into. Often, survey research and experimental setups take individuals as their unit of analysis, but the case study may concentrate on phenomena as diverse as organizations, professional groups, communities, specific sites, or practices. With these phenomena, it is not always clear where they begin or end and what their boundaries are. Where does a “court” begin and end? Where are the boundaries between different professional groups? Where do judicial work practices take place? <b>Opening the Black Box of Judicial Decision Making: A Theoretically Informed Research Question A research question is crucial to begin delineating “the case” in question. In our research, we were interested in judicial work practices, raising the question of how, as a matter of everyday work practices, judges construct cases and “get things done.” We had our theoretically inspired reasons to concentrate on concrete judicial work practices. While survey research concentrates on measuring “input” and “output” to judicial decision making and helpfully distinguishes between offender and offense characteristics in order to measure their relative effects on sentence type and duration, we felt such research “black boxed” the processes and mechanisms that take place in between input and output. We wondered, instead, what happens in between. We were sensitized by admonitions to take seriously the ways in which judicial actors made sense of individual cases.4 Judges, in other words, are not computers processing offense- and offender-related bits and pieces of information, but do active work in the production and construction of actual cases—practical, unfolding, and to some extent uncertain work of which Judge Cohen’s work practices are evocative. The fragment above also illustrates that Judge Cohen aims to “see” the case in its entirety. She puts some information from the file at the forefront while completely ignoring other information, intends to collect additional information during the court session, and interprets and weights bits and pieces of information in order to be able to see the case clearly. The contrast between this holistic way of working and the rationale of experimental research, the goal of which is to isolate the influence of one particular factor on decision making—for example, the influence of an anchor, frame, bias, or other manipulated variable—is large indeed. Our research, then, Cyrus Tata, “Sentencing as Craftwork and the Binary Epistemologies of the Discretionary Decision Process (2007) 16(3) Social & Legal Studies 425; Cyrus Tata, “Conceptions and Representations of the Sentencing Decision Process” (1997) 24(3) Journal of Law and Society 395. 4 6 sought to shed light on the processes of interpretation and everyday work practices that both survey research and experimental research tend to neglect. For this, we sought help from the case study. Indeed, one of its advantages over survey and experimental setups is that the case study helps approach these practices as they actually and practically occur. <b>Hanging on to Sensitizing Concepts At this stage, Tata’s criticism of statistical methods in the study of law in action and his conception of judicial work practice as an unfolding work practice functioned very much as a sensitizing concept: a line of reasoning, a term, a phrase that helps to ask a question and open a line of inquiry.5 It allows us to ask questions such as: Where and how do various actors work on and with cases? What happens to the case in such work practices? How are these work practices distributed? In so doing, the notion of unfolding work practices does not unproblematically point to a set of instances or cases; however, it needs to be made operative in the succeeding stages of the research. Blumer understands sensitizing concepts not as simple descriptions of things in the world, but rather as instruments that “suggest directions along which to look”: 6 <quotation>A definitive concept refers precisely to what is common to a class of objects, by the aid of a clear definition in terms of attributes or fixed bench marks. This definition, or the bench marks, serve as a means of clearly identifying the individual instance of the class and the make-up of that instance that is covered by the concept. A sensitizing concept lacks such specification of attributes or bench marks and consequently it does not enable the user to move directly to the instance and its relevant content. Instead, it gives the user a general sense of reference and guidance in approaching empirical instances. Whereas definitive concepts provide prescriptions of what to see, sensitizing concepts merely suggest directions along which to look.</quotation> There are basically two ways, Blumer argues, to make a sensitizing concept operational. 7 One way is to carefully operationalize them, as is customary in surveys and experiments. That is, we may try to turn a concept such as “institutions,” or “culture,” or “professionalism” into a measurable entity by devising measurement scales. Survey and experimental research tends to work with such operationalizations: On the basis of prior research, this type of research may adopt preexisting operationalizations or devise new ones. For Blumer, this is a way of turning a sensitizing, “vague” concept into a definitive one, or, in other words, translating a 5 Ibid. Herbert Blumer, “What Is Wrong with Social Theory” (1954) 18 American Sociological Review 3. 7 Ibid. 6 7 theoretical concept into a measurable variable. With such an operationalized concept, one can cast aside everything that is not measured by it and zoom in on one facet of the object under study. Yet for Blumer this is an undesirable procedure, precisely because it isolates facets and factors and fails to address these concepts as empirical instances. The crucial value of the openended character of the case study is that sensitizing concepts can retain their undefined character a bit longer:8 <quotation>One moves out from the concept to the concrete distinctiveness of the instance instead of embracing the instance in the abstract framework of the concept. This is a matter of filling out a new situation or of picking one’s way in an unknown terrain. The concept sensitizes one to this task, providing clues and suggestions.</quotation> Instead of devising a set of questions or measurements that would make the notion of “unfolding work practice” measurable, we wanted to be able to see how such unfolding takes place, and to take our cues from the practices we were to encounter in the field. <b>What Is a Field? Of course, this led to yet another question. We may now have a theoretically informed sensitizing concept, but, in the absence of definite measurement techniques (a survey, an experiment) or definite units of analyses (individuals), we do need to find practical and concrete ways to bring our sensitizing concept into touch with actual practices. In our case, that means we need a specific and concrete site to investigate these. In other words: We need a “field.” Much has been written about the problems posed by “the field” to those aspiring to do a case study. While it may have been plausible to map “the field” and “the object” onto each other in the past, for instance, by going to a different country to study its culture, 9 the relationship between the object of study and a site of research may raise important challenges. 8 Ibid p. 9, p. 8. And precisely this mapping of object of study and physical location is, even in the case of early twentiethcentury ethnography, very much disputed: because to what extent is “the field” an isolated thing in and of itself? And do “cultures” map onto physical sites, as is suggested in anthropological work on, for example, “the Nuer” (see Edward Evan Evans-Pritchard, The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People (Clarendon Press 1940)? The gesture that maps one onto the other not only glosses over difference within physical locations or “cultures” but also neglects concrete histories of exchange, and—in the case of nineteenth and early twentieth-century anthropology—especially colonialism itself, of course (see Claude Lévi-Strauss, Tristes tropiques (Criterion Books 1961), and the pieces gathered together in Writing Culture by James Clifford and George E Marcus, Writing Culture: The Poetics and Politics of Ethnography (University of California Press 1986). 9 8 Many practices and objects of study are in and of themselves multisited,10 dispersed over a variety of physical locations and practices. To name an obvious example: Think of the legal case itself. The legal case is a multisited entity: It begins somewhere, with an event, and then goes through a series of bureaucratic translations. Statements are gathered, witnesses heard, further evidence sought; the case file grows and starts to travel between legal–bureaucratic actors. Or think of “the legal profession”: also a multisited professional group, working in a variety of settings on a variety of tasks. Fields may also be “fuzzy,”11 their boundaries quite unclear. Take “the court” itself: Where does the court start? Do its administrative workers count as part and parcel of the court, or are we more interested in the “proper” legal actors? To make matters are a little more complicated: Do legally trained and assisting clerks count as being part and parcel of that field? Where is the field that allows us to study our case? These are intricate questions indeed, and make clear that it is not always easy to move on from a choice of an object of study to a physical site or “field” that will allow us to study this object. Again, we must mobilize a theoretically informed research question to justify our choice. In our case, we recognized the necessity of trying to get as close as possible to these judicial practices of case making. Precisely because we were theoretically interested not in the input and output but in the process of decision making, we looked to the court for help. After all, the courthouse is the physical space that contains the file room where administrative personnel file cases, the offices where clerks prepare cases for the judges, the back offices where judges read their files, and the courtroom where judges interact with their defendants. As such, the courthouse would give us the space to track and trace the process of making sense of cases on a daily basis. There, we might converse with administrative personnel, clerks, and judges, and read files. There, we might observe just how judges manage the proceedings in court. We also decided to consider the practices of lower magistrate judges who preside over less severe cases on their own. In contrast with the full-bench panels consisting of three judges, the lower magistrate judge adjudicates and sentences relatively low-level cases. As we were less interested in legally complex “hard cases,” but precisely in judicial work as a matter of everyday, to some extent routinized work practices, our choice to consider lower magistrate judges resonates with our theoretical aims. <b>On Access George E Marcus, “Ethnography in/of the World System: The Emergence of Multi-Sited Ethnography” (1995) 24 Annual Review of Anthropology 95. 11 See e.g. Eva Nadai and Cristoph Maeder, “Fuzzy Fields: Multi-Sited Ethnography in Sociological Research” (2005) 6(3) Forum: Qualitative Social Research, Article 28. 10 9 Gaining access to this court was by no means easy. While it is tempting to think of access as a single event, in actual fact it is often a multisited series of events. First, researchers in institutional settings often face institutional gatekeepers (professional interest organizations, team and department heads who may have to agree), and in the study of legal practices, such negotiations were rendered more difficult by two concerns specific to the judiciary. The first was the confidentiality of these judicial practices: Could we guarantee no identifying information about individual defendants would come out? The second was the court’s fear of unwanted media attention: “What if it’s us on the front pages of the tabloids next year?”, one team head pointedly asked in the negotiations leading up to our eventual court access. Assuaging these concerns was not always easy, but our commitment to assuring confidentiality for both defendants and individual judges (and also other court workers), as well as our commitment to providing nuance to overtly politicized understandings of judicial work practices, helped our case. This was when we noticed that access is, in institutional settings, a multifaceted thing. Access to the court required that the first author of this chapter undergo a brief ceremony promising confidentiality and discretion in her treatment of her findings. It was also an infrastructural event: Not only did this author receive a digital access pass to the court’s “backstage” offices, she was also equipped with a user name and password so that she could log on to the court’s computers and access the digital files. In the first stages of the research, she was to be seated among the court’s administrative personnel, close to the file room. There, she would be able to read (physical and digital) files and mobilize individual judges to cooperate in her research. There, she would be able to read (physical and digital) files and mobilize individual judges to cooperate in her research. Indeed, the team head who provided her with access had warned her that she would have to negotiate “access” to individual judges each time anew, as he could not force “his” judges to cooperate (this, incidentally, is an issue an increasing number of researchers of the judiciary or other professions may face, as professionals tend to emphasize and value their autonomy with regards to management). Fortunately, another senior judge was herself interested in the research, and introduced the researcher to her team. The introduction included a brief presentation about the research and its theoretical background. This introduction, and the court’s faith in the researcher and the research communicated by it, significantly helped her in her later requests for individual permission. Exactly what constitutes access, therefore, and where and how it has to be negotiated, is dependent on the specifics of the site of study. 10 It seemed, then, that we were all set up. We had our theoretically informed research question: “How do judges arrive at their decisions as a matter of everyday work practice?”; a sensitizing concept: “Judicial work as an unfolding work practice”; and a theoretically informed selection of a field: the court house. But of course, that is where the real work started. <a>“MAKING THE CASE TALK”: ON SITUATEDNESS, METHODS, AND THEIR CONSEQUENCES The real work, in this case, existed in experimentation with different methods of generating data—for data is not simply “gathered,” as if these are simply bits and pieces we happen to stumble upon, but rather produced in concrete interactions with both people and things. In pragmatist terms: Knowledge about the world is always a consequence of situated collaboration, experimentation, and interaction.12 This holds true regardless of the specific methods used. Indeed, experiments or surveys are similarly ways to collaborate with realities, even though the numerical data they yield have the persuasive gloss of “mechanical objectivity”:13 a kind of objectivity that is rooted not in the specific expertise, knowledge, or authority of the individual researcher, but rather in the calibration and sophistication of the measurement instrument itself. With surveys and experiments, the individual’s expertise, authority, background knowledge, and indeed social identity may matter somewhat less to the validity of the conclusions reached than would be the case in the more situational and embodied modes of data generation typical of case-study matters. For this reason, techniques have been developed that can enhance the reliability and validity of these modes of data generation.14 However, these techniques do not free us from the necessity to address the questions of, first, just how we are situated vis-à-vis the case under study, and second, precisely what kind of knowledges specific methods of data generation yield. In order to address these questions, a relatively brief period of exploratory research in the field is a helpful first step. In our case, it helped us to address both questions. William James, The Meaning of Truth: A Sequel to “Pragmatism” (Prometheus Books 1997 [1909]); Donna Haraway, “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective” (1988) 14(3) Feminist Studies 575; Larry A Hickman and Thomas M Alexander, The Essential Dewey (Indiana University Press 1998). 13 Lorraine Daston and Peter Galison, “The Image of Objectivity” (1992) 40 Representations 81. 14 Anthony J Onwuegbuzie and Nancy L Leech, “Validity and Qualitative Research: An Oxymoron?” (2007) 41(2) Quality & Quantity 233. 12 11 <b>The Researcher’s Situatedness: Limitations and Affordances While academic writing may privilege the passive and impersonal voice—in which “data are found”—the experience of doing a case study is inevitably one of running up against one’s own “situatedness.”15 We understand situatedness very broadly, in both the highly physical sense and the less tangible sense. Physically, the first author’s “situatedness” was one of being, in the first stages of the research, stationed in the administrative offices of the court. This situatedness provided her with easy access to case files and to the court’s digital infrastructure, but also to the more informally shared stories (some would call it gossip) told about the judges. In that capacity, the administrative personnel of the court, so often invisible in studies of knowledge practices (for example, the medical profession, the judiciary, studies of academic work), proved an invaluable source of information about the court’s organization and the reputations of individual judges. Collegial links between the administrative offices and the groups of assisting clerks also provided her with connections among these clerks, to whose shared offices she would move in a later phase of the research. Being stationed at the administrative offices also had its drawbacks, however. In later phases of the research, the first author of this piece was often mistaken for an administrative worker herself. Her social identity as relatively young and female plausibly played a role in these appraisals as well—we can never leave our social identities at home when doing fieldwork. In such moments, she had to negotiate and metaphorically prove her academic credentials to individual judges. At the same time, being perceived as “lower” in the court’s hierarchy, or as a “mere Ph.D. student,” may also have helped her, in that it made her a harmless, nonthreatening presence to the judges. Depending on judges’ willingness to explain and elaborate on their work practices, the first author of this piece quickly recognized that the perception of her as an uninformed outsider could work to her advantage. Situatedness, then, is both tangible and less tangible, physical and social. It is not necessarily a handicap, but a fact of doing case studies that has to be reflected upon and, where possible, turned into an advantage. Gender, ethnicity/race, and age are particularly likely to shape interactions with people in the field, and one’s situatedness within a specific site within the institutional field matters as well. Indeed, later in this chapter we will concentrate on the theoretical possibilities of this situatedness among the administrative and clerical assistants. 15 Haraway (n 12). 12 <b>Selective In/Visibilities: Methods and Their Performativities As we will describe in this section, the research also consisted of experimenting with, and reflecting on, the uses of different methods (interviews, observations, document analyses). Our methodological approach reflects a more recent development in the study of knowledge practices. Methods are increasingly understood not as neutral “probes” that tap into a preexisting phenomenon, but rather as active components of knowledge production. In other words: Methods do not simply help to “represent” a reality; they are also active in the performative sense16—they shape the realities about which we seek to speak. That does not mean they do not yield valid results—there are no methods without their performativity. Surveys, for instance, are performatively active in that they treat individuals primarily as bundles of social factors (age, class, and ethnicity) that display preferences or behaviors, while their reliance on a “sample” always evokes a broader, often nationally or professionally circumscribed “population.”17 In this way, this method is ill suited to attend to the ways people live their lives in and through meanings and narratives, for instance. Experimental setups may helpfully allow researchers to “control” for background variables and isolate the causal mechanism under study. At the same time, they work with a similar conception of individuals as automata who mechanically “respond” to different experimental conditions. However, interviews, observations, and document analyses do not automatically privilege the researcher: They, too, have their specific performativities, that is, their affordances and limitations. In the following, we will highlight the performativities of these methods in our research. Our research question, you will remember, focused on how judges make sense of individual cases. We were interested in how they create order in the information that they have at their disposal, how they perceive of individual defendants, and how they tailor their sentences to the personal circumstances and the specifics of the offense. What better way to answer these questions than to simply ask judges? <b>Interviews Judith Butler, Gender Trouble: Feminism and the Subversion of Identity (Routledge 1990); John Law, “Seeing Like a Survey” (2009) 3(2) Cultural Sociology 239. 17 See, for example, Alain Desrosières, The Politics of Large Numbers: A History of Statistical Reasoning (Harvard University Press 1998); Law (n 16). 16 13 Trying to get to judicial sense making by way of either structured interviews or conversation is, however, not an unproblematic operation. First and foremost, it depends on a specific idea of judicial work, that is, it conceives of judicial work as a purely cognitive thought process. We will say more about this issue later on; for now it suffices to say that already embedded in the interview or conversation format is that people have ideas “in their heads,” which we can elicit and probe. However—and we will return to this issue much later on—actual work practices may be neither purely “cognitive” nor directly accessible to the respondent. Interview formats, as well as informal questions about work practices, depend on people’s self-reflexive capacities, which are their abilities to relate, account, explain, and render intelligible what they do to others. As such, these verbally mediated methods perform a specific kind of human being, that is, a respondent capable of reflection and narrativization of his or her (past) actions. Of course, many people are in actual fact just that: Many of us reflect upon, try to account for and understand, our own actions in terms acceptable to us. But in this, such narratives and accounts also slide into the realm of justification or rationalization. This realm enables respondents to present a favorable self-image or to tell stories they believe the researcher is expecting to hear from them.18 In our first encounters with judges, this performative effect of the interview format was especially salient. The first author of this chapter quickly noticed that her openended questions about “how you typically deal with cases” fell flat, as judges were not providing her with a description of their “typical” work process but with rather philosophical reflections on the role of different punishment goals, on the role of the judiciary vis-à-vis government and civil society, or on the role of the judge vis-à-vis that of the prosecution and defense. Asked for a description of their work practices, they provided justifications and philosophical reflection instead. Not that this is a bad thing: We do not mean to suggest that such justifications and rationalizations have the status of “lies,” nor that these cannot be important research objects themselves. Indeed, discovering how judges justify and account for their practice is a worthwhile empirical endeavor—but our purpose was to gain insight into judges’ actual work practices. There was a second performativity of verbally mediated modes of data generation at play here. Note that the researcher tended to cast her question in general terms, for example, “how do you usually/typically/generally deal with cases?” Early on during the fieldwork, it became clear that judges tend to have difficulties answering such questions. “It all depends on the 18 Jerome Kirk and Marc L Miller, Reliability and Validity in Qualitative Research, vol. 1 (Sage 1986). 14 specific case,” judges would object; or, going along with the question, they would pepper their responses with reference to past cases. A rather more sociological, general approach to their practices of case making resonated badly with their own casuistic orientation.19 This was particularly salient at the start of the research project, which was marked by controversy triggered by the publication of a study concluding that defendants who looked “foreign,” particularly if they did not speak Dutch, were more likely to receive a prison term than defendants who both look and speak Dutch.20 Judges whom the first author was to meet later, during fieldwork, asserted in no uncertain terms that they were not happy with the study’s implication that they would disadvantage people with certain legally irrelevant characteristics.21 They raised questions about the “partiality” of the research and about the qualifications of the researchers themselves—“do they have any idea how a judge arrives at a sentencing decision?”22 They were particularly piqued by the suggestion that they were focusing on extralegal factors such as foreign looks and the ability to speak Dutch when sentencing. Instead, judges argued that they try to see cases holistically and synergistically, and to arrive at a sentence after a careful process of weighting and evaluating charges and evidence and taking into account the defendant’s person. Through this controversy, we realized it would be far more productive to elicit judges’ commentary on actual, individual cases. Given the performativities of the interview format and its tendency to stick to generalities rather than specifics, we elected to “shadow” individual judges in their “doings with cases.” This method, relying on study of the files and observation of and informal talk about these files with judges during their pretrial file work, as well as on observations in court, was far better suited to studying how judges make sense of cases. After all, for judges the case itself was a highly specific and unfolding thing. To be able to do justice to their case making, then, we had to follow the case wherever it went: from file, to backstage office, to the court, all the way up to the eventual verdict. <b>Observations of Work Practices and In-Court Interactions: Cases and Narratives 19 Peter Mascini and others, “Understanding Judges’ Choices of Sentence Types as Interpretative Work: An Explorative Study in a Dutch Police Court” (2016) 37(1) Recht der Werkelijkheid 32. 20 Hilde Wermink, Jan de Keijser en Pauline Schuyt, “Verschillen in straftoemeting in soortgelijke zaken: een kwantitatief onderzoek naar de rol van specifieke kenmerken van de dader” (Differences in Sentencing Outcomes in Comparable Cases: A Quantitative Study into the Role of Specific Offender Characteristics) (2012) 87(11) Nederlands Juristenblad 647. 21 Irene van Oorschot, “Ways of Case-Making” (Erasmus University Rotterdam 2018). 22 Ibid. 15 As a result of our choice to observe judges’ file work and the proceedings in court, we were not only able to elicit judges’ responses to individual cases and observe their practices of reading and summarizing case files, but also encountered in effect a salient ingredient of their practices: the whole-case narrative. We arrived at these whole-case narratives by paying attention to recurring tropes in judges’ comments on individual cases. Some cases, we learned, were more typical than others. Of course, there were always cases that did not fit any pattern, cases which actually surprised judges. But some of the cases evoked, to them, broader categories or kinds. One such typical narrative was that of the drug-addicted defendant. Petty theft, a long criminal record containing similar offenses, and histories of drug abuse were often salient ingredients of such cases. Another “typical kind” of case was that of the “angry young man”: a young male accused of criminal assault, often in public places and often at night. Judges would point to inebriation or perceived slights to such defendants’ masculinity as typical of these whole-case narratives: “guy goes out, has a few drinks, gets violent,” as one judge summarized this narrative. Some domestic violence cases were similarly typified. Especially when the couple in question was caught up in divorce proceedings, some judges expressed the belief that “the wife isn’t an angel herself.” Importantly, we started to see that these typified whole-case narratives tended to be accompanied with specific appraisals of remorse. Drug-addicted defendants, judges felt, may be very remorseful about the things they have done, but this tended not to be seen as reason to mitigate the prosecution’s proposed sentence: After all, their long criminal records clearly showed, as far as the judges were concerned, these drug-addicted defendants’ lack of capacity to “take responsibility for oneself.” Angry young men, in contrast, tended to be chastised for their actions in court, and were typically expected to express great remorse – something that was quite difficult for defendants, especially if they felt the victim was partially to blame for the escalation of the violence. Defendants charged with domestic violence at times similarly pointed, in court, to their partner’s role in the escalation of violence (often pointing to his or her “nagging” character or psychological issues). Such finger pointing—or, rather, such externalization of responsibility—was not always judged as harshly as in cases involving “angry young men”: after all, the partner may not “be an angel herself” or may have filed a criminal complaint as part of a legal disagreement between the couple in a custodial case.23 Had we stuck to interviews only, we might have drawn the conclusions that judges conceive of their work practices in legal–philosophical terms only and that there was nothing See for these arguments I van Oorschot, P Mascini and D Weenink, “Remorse in Context(s): A Qualitative Exploration of Remorse and its Consequences” (2017) Social & Legal Studies 359. 23 16 general to say about cases, as “every case is different.” However, through case-specific observation and informal conversation we started to arrive at quite a different picture. Yes, judges tend to have a casuistic attitude. But their commentary on, and decision-making in, actual cases nevertheless suggested that they also typify individual cases—in other words, that they work with certain cognitive and moral templates that assist them in their sense making, and these templates often involve a kind of narrative and plot. The interview format and casespecific questions and observations each produce their own selective visibilities and invisibilities. <b>“The Silence of the Social” and the Abundance of Writing Moving away from structured interviews and general questions about judicial work practices, the data discussed so far were still, to a large extent, found through verbally mediated forms of data generation. While the focus on individual cases was more in tune with judicial work practices, we nevertheless relied to a large extent on judges’ commentaries, explanations, and accounts of individual cases and the verbal interactions taking place in court. In that sense, our focus was rather “phonocentric”:24 It privileged the spoken word both as an object of study—in our emphasis on narratives—and as a source of data—in our emphasis on judges’ verbalizations. In so doing, however, it glossed over two things: first, the fact that judicial practices themselves are, to a large extent, mediated precisely by written materials in the case file; second, the fact that many of these judicial work practices are, simply speaking, “silent”—judges can make sense of files without narrating their thought processes to us. Let us tackle the presence of written materials first. Many organizational field sites are rife with documentary materials: charts, graphs, memos, briefs, reports, and indeed files. In the court studied here, they were truly abundant, raising the question of just how we might include them in our data generation practices. This is on the one hand a practical challenge: It is not easy to make clear summaries of individual case files. They are, to many academics, an unfamiliar genre of institutional writing, and experience has taught us that it takes practice to start to know “how to read a file.” Yet we nevertheless needed a sense of their content so that we might ask good questions about these individual cases. These summaries must also allow retrieval of information in the future, making it possible to go back to them in a few years’ time. Dealing with files is as much a data generation as an archival practice. Depending on one’s theoretical interests, files can also serve different theoretical goals. One can read files “representationally”: as more or less transparent windows onto the reality of the offense and 24 Jacques Derrida, Of Grammatology (John Hopkins University Press 1967). 17 offender. The problem with such a reading, however, is that many organizational documents may not have as their principal purpose this kind of “truth telling,” but may additionally serve other organizational purposes.25 After all, case files, for instance, do not only present evidence about the event in question, but also enfold materials that render visible its procedural history. In other words, their use is both “epistemic” and procedural, and the researcher looking only for a veracious story about the event in question may miss out on the opportunity to study the file’s embeddedness within legal–procedural networks and practices.26 One may also read files, or documents in files, “hermeneutically,” as reflections of authorial intent: What did the author of this or that presentencing report mean when he wrote that the defendant is not ready to be rehabilitated? (However, we were primarily interested in what these documents meant to judges and how they interpreted them.) Or one can read them as embodying of broader cultural or social discourses,27 and raise questions such as: Why do presentencing reports incorporate fewer and fewer stories about the defendant and more and more risk-assessment scales? Is this a reflection of shifting punitive logics within the criminal justice field? This is typical for studies that conceive the law as a representation of something else, for example, historical developments or social relations. Instead of this, we were interested in the practice of judging proper. We read case files “forensically,”28 as clues to different organizational, institutional, and legal realities, as when one searches a file for its own history of production in police offices and its history of circulation along chains in the criminal justice network. This reading implies that the documents themselves are interrogated—a process of interrogation that crucially includes the histories, routines, and decisions these documents render invisible. Taking seriously how facts and truths are made in legal settings alerted us to the role of procedure and the materiality of time in legal proceedings. Since the file can serve many different purposes for the researcher, one needs to make a decision about what to incorporate in one’s reports on these files. These decisions need to be made in dialogue both with theory and with the field itself. A similar attention to theory and the field itself is present in a second component of these practices: the fact that many of these practices were nonverbal, that is, “silent.” Hirschauer suggests that many methods of inquiry rely on people’s capacity to verbalize thoughts, 25 See Harold Garfinkel, Studies in Ethnomethodology (Blackwell Publishing 1967). Irene van Oorschot, “Seeing the Case Clearly: File-Work, Material Mediation, and Visualizing Practices in a Dutch Criminal Court“ (2014) 37(4) Symbolic Interaction 439. 27 Teun A van Dijk, “Principles of Critical Discourse Analysis” (1993) 4(2) Discourse & Society 249. 28 van Oorschot (n 21). 26 18 reasons, actions, and motives.29 Yet many components of social life are “silent”: Think, for instance, of rituals, of taken-for-granted cultural or social ideas, or indeed of routinized work practices. For Hirschauer, these instances of social life are an acute challenge to description, for they raise the question of what the researcher should try to record and how—given that complete description is an impossibility—as well as asking how the researcher can impute valid conclusions from these “silent” episodes of social life. In our case, one such “silent” episode was judicial “file work.” On the one hand, judges commented on the content of these cases. Yet on the other, many of these judges were “silent” about the technicalities of case making: about how they read a file or how they order and structure information. Hirschauer suggests that in such instances, researchers reflect theoretically on what it is they seek to discern within these practices—in other words, that they do not try to record everything but, in dialogue with the field, come to arrive at theoretically informed choices as to what to record, and how to record. In sum, it was through interviewing judges, in the early stages of the fieldwork, about how they deal with cases in general that it became apparent that, in order to understand the unfolding practice of sense making, it was necessary to focus on how judges handled specific cases. Moreover, observations of judges’ file work made it clear that judicial decision making is largely a “silent” process; document analysis of the file for its history of production alerted us to the role of procedure and the materiality of time in legal proceedings, and observations of the court proceedings and informal conversations showed that judges use cognitive and moral templates as aids in their sense making. In other words, shadowing judges in their “doings with cases” meant confronting both quiet episodes of file work—and of course the case file itself—and a setting saturated with talk and discourse: the courtroom. Precisely by experimenting with different methods, it became apparent that each method generates its own type of data and its own specific visibilities and invisibilities. For example, we learned that although interviews can yield cognitive justifications and philosophical reflections on how work is done, they do not help much to gain an understanding of unfolding work practices, and that although case-specific observations in court and informal conversations can render visible judges’ conceptions of specific cases in broader categories, only the analysis of case files, and especially observations of their implication in actual work practices, proved methods capable of attending to the “silent” dimension of judicial practice. This conclusion about the performativities of different methods ought not to be read as a plea for mixed Stefan Hirschauer, “Putting Things into Words: Ethnographic Description and the Silence of the Social” (2006) 29(4) Human Studies 413. 29 19 methods per se. The desirability of using different methods depends on the research question. Our goal was to gain understanding of the entire process of how judges arrive at their verdicts. In order to achieve this goal it was necessary to use methods that were tailored to the silent pretrial work as well as to the commentary on and verbal exchanges during trial. Had our study been limited to nonverbal behavior or conversations in court, for example, then observations of court proceedings might have sufficed. The following section is an account of our choices in the face of the two ingredients we realized were crucial to understand judicial practices: the abundance of written materials, and the “silence” of pretrial preparation practices. As we will show, it was in taking these two facets of judicial practices seriously—by listening to the field speaking back to us—that we made significant theoretical advances in the study of law in action. <a>THE FIELD SPEAKING BACK: ENCOUNTERS WITH THE UNEXPECTED Remember that we were interested in the way judges make sense of cases. But there was a real way in which we had unwittingly narrowed down our focus already. That is, we had located the activity of sense making exclusively in the work practices of judges, whose reasoning, motivations, and interpretations we tried to capture using interviews. However, this focus on judges’ narrativization of cases glossed over two salient ingredients of these practices identified in the above: case files and the “silence” of judges’—and clerks’— preparatory work practices. Faced with these two ingredients of these work practices, we tried to allow these to “make us think” and not merely “recognize,”30 that is, reproduce commonsensical or theoretical categories.31 By taking written materials and file work seriously, we learned, locating the activity of judging in the conscious or intuitive leaps of individual judges is already a theoretical move. However, what happens to our theoretical conceptions of judging when we aim to incorporate these case files and “silent” work practices in our understanding of judicial work? How can we conceptualize judging in a way that is attentive to both the spoken and the written word? In the following, we show how “the field is a powerful disciplinary force: assertive, demanding, even coercive.”32 The unique Isabelle Stengers, “Introductory Notes on an Ecology of Practices” (2005) 11(1) Cultural Studies Review 183. Clifford Geertz, After the Fact: Two Countries, Four Decades, One Anthropologist (Harvard University Press 1995) 119. 32 Ibid. 30 31 20 affordance of the case study, in our view, is that it seeks not to silence the field, but rather to listen carefully. <b>Case Making as a Distributed Work Practice Our prior idea that sense making is the province of the individual judge was simply a presupposition derived partially from common sense—after all, the judge is the one accountable for the verdict—and partially from the wider field of study surrounding these judicial practices, which may for instance try to measure “influences” on judicial decision making (such as the influence of extralegal characteristics on the judicial decision). It is reflected, too, in legal theory emphasizing the central role of the judge in the application, and sometimes the making, of law.33 However, our experiences and situatedness in the field significantly rearticulated this conception. Stationed among the court’s administrative staff and later among a group of its assisting clerks, the first author of this piece became intimately acquainted with the crucial role played by the preparation practices of administrative staff, and especially clerks.34 Judges, she learned, rarely read and make sense of cases without the case file having already undergone a distributed process of ordering, coding, highlighting, and summarizing. For these court workers, case files are recalcitrant things: They may be incomplete; they may be messy or “overcomplete” (involving too many copies of the same documents), their internal order and structure obscure to those trying to read them. No wonder, then, that the court also draws on assisting clerks to prepare case files. Interestingly, it is at this point that the first author of this chapter also started to notice exactly how much this preparatory work slides into actual engagement with the case file’s contents. Assisting clerks do not only order the case files; they also physically mark sheets of paper with color-coded sticky notes, highlight salient phrases in the case file, and summarize the case file in more or less extensive “extracts.” Are these practices merely peripheral to judicial decision making, just part of their “context”? Or is it perhaps worthwhile to think about these practices as part and parcel of distributed case making? The first answer to that question—these practices are only a “context,” not an ingredient— is perhaps the more common one. After all, the judge decides, so sense making must take 33 Hart (n 3); B. Latour The Making of Law: An Ethnography of the Conseil D’Etat (Polity Press 2010). See also Nina L Holvast, “Considering the Consequences of Increased Reliance on Judicial Assistants: A Study on Dutch Courts” (2014) 21(1) International Journal of the Legal Profession 39. 34 21 place at the level of the judge. Yet we were not wholly satisfied with that solution. Sure, the judge is accountable for the verdict, and only the judge can make a decision. But at the same time, the judge is part of a chain of work practices revolving around the case file, and draws explicitly on these work practices in his or her own case preparation and sense making. Judge Cohen in the earlier description, for instance, used both the file and the clerks’ summary to come to an understanding of the case. This led us to conclude that it is worthwhile to try to understand judicial case making as a distributed work practice rather than a phenomenon isolated at the level of the judge. This insight into the distribution of sense making not only made us aware of the contribution clerks’ practices make to judicial decision making, but also helped us to grasp the crucial role played by nonhuman objects in various settings. Actor network–theoretical approaches,35 for instance, gave the researcher conceptual tools to deal with the material side of judicial decision practices. The case file, we realized, was not simply a carrier of information but rather a physically and informationally recalcitrant object: an object that exercises agency and shapes work practices around it. Interestingly, and serendipitously, the fieldwork moreover coincided with a project aimed at the digitization of all case files—a project that significantly reshaped the, to some extent, medium-specific and distributed work practices of clerks and judges.36 Tracing how clerical and judicial summarizing practices shifted in response to this new “tool of the trade,” we demonstrated just how much of this work does not take place inside the head of the judge, but is in fact a materially mediated and socially distributed practice.37 In so doing, we were able to contribute to the social study of legal work practices, arguing that the theoretical assumption that sense making takes place only at the level of the judge is a poor approximation of actually existing legal practices. We were also able to caution against the assumption that knowledge work is just immaterial, cognitive work—because it is precisely that assumption that is at work in the (sometimes too unproblematic) adoption of digital methods of work. Because existing work practices were tailored to the specificities of paper, the digital file disrupted some of the clerks’ and judges’ routines and tasks and significantly hampered the smooth flow of work in the early stages of the digitization project. 35 See for example Annemarie Mol, The Body Multiple: Ontology in Medical Practice (Duke University Press 2002); Bruno Latour, Reassembling the Social: An Introduction to Actor-Network Theory (Oxford University Press 2005). 36 Giovan Francesco Lanzara, “Reshaping Practice Across Media: Material Mediation, Medium Specificity and Practical Knowledge in Judicial Work” (2009) 30(12) Organization Studies 1369. 37 van Oorschot (n 26). 22 <a>HART’S INTERNAL POINT OF VIEW VS. PRACTICES OF VISUALIZATION There was yet another implicit starting point which we were forced to reconsider as our case study developed. This was based on the distinction between “external” and “internal” points of view, especially as developed by Hart.38 This distinction is first and foremost a distinction between orientations toward the legal rule. Only the internalized and practical acceptance of the legal rule is a true instance of the internal point of view; other conceptions of the rule are external in that they take the rule as an object of contemplation, or sideline the rule entirely in efforts to study it empirically.39 As sociologists ourselves, we could not hope to assume this internal viewpoint: After all, we were in the business of description, not judging. Nor could we hope to study it empirically, we thought: After all, the internal point of view is more of an intuitive, practical leap of rule acceptance than a tangible, empirical entity. Again, the field spoke back to our theoretical presuppositions. Confronted, for instance, with the work practices of clerks and judges, we started to notice something interesting about these case summaries. Both clerks and judges would speak of these practices as attempts to “see the case clearly” or as ways to “arrive at a picture of the case.” Like Hart, clerks and judges used a similarly “perspectivist” vocabulary to speak about their engagement with cases. Drawing on paper copies of the formally specified charges, clerks and judges would use these to produce their summaries. Procedural elements of the case would be written in the top-right corner. Information regarding the event in question would surround the text of the charges in the middle of the paper. At the bottom, clerks and judges would jot down notes with regards to the defendants’ personal circumstances. In these practices of visualization,40 we soon started to recognize the careful building of an internal point of view: of summarizing the case in such a way that the leap from the facts to the legal rule became a possibility. The internal point of view, then, was not one point but rather a distributed practice of visualization; and it was not entirely locked up in the “head” of the actor at all. In substituting Hart’s emphasis on the internal point of view with the concept of “practices of visualization,” we were also able to contribute to the ethnomethodological study of legal work practices. Ethnomethodology is a rather specialized subspecies of sociology. Instead of adopting many of sociology’s explanatory concepts—the state, culture, class, capitalism, and so on—it seeks 38 Hart (n 3). Ibid; Scott J Shapiro, “What Is the Internal Point of View?” (2006) 75(3) Fordham Law Review 1157. 40 van Oorschot (n 26). 39 23 instead to treat these as explananda: as abstractions that need to be explained in reference to practical, ongoing activity.41 In the study of legal practices, these studies have often drawn attention to the problem of rule-following behavior. How is it that legal actors come to recognize and “follow” rules as a matter of practical activity?42 We pointed to the practical activity of file work as a salient and empirically investigable site of such practical rulefollowing activities. Thus, contrary to common conceptions regarding judicial work practices, the internal viewpoint toward the legal rule is also accessible to the sociologist; by assessing case summarization practices, the practical leap of rule acceptance can be rendered visible. In sum, it was by listening carefully to the field that we gained a more in-depth understanding of judges’ unfolding work practices, one that challenged commonsensical theoretical conceptions about these practices. Rather than taking place “in the head of the judge,” sense making occurred to be materially mediated—as shown in the disturbance of the summarizing practice by the digitization of case files—and socially distributed among clerks and judges (as well as other officials active in the preceding stages in the judicial decision chain, such as police officers, prosecutors, and rehabilitation workers). Moreover, rather than an entirely intuitive or cognitive process taking place “inside the head of the actor,” the process of linking facts to legal rules is empirically accessible by studying the summaries clerks and judges add to the prints of the charges formally specified by the prosecutor in order to visualize what the case is about. <a>A CASE OF WHAT? Remember Judge Cohen’s attempts to see if a series of events could be legally qualified as an instance of coordinated theft? Jurists will recognize the legal operation of subsumption in Judge Cohen’s concerns. Judge Cohen does not seek a fundamental pattern or law applicable to all defendants, offenses, or cases, but rather what story can be told about the case, and how that story is to be related to legal rules. But this is not that far removed from what social scientists do on the basis of their case studies. They, too, are concerned with the question of what this case is an instance, exemplar, or kind of. Indeed, this is what researchers doing case studies consider when they want to generalize on the basis of their fieldwork. In contrast to 41 Garfinkel (n 25). Baudouin Dupret, Adjudication in Action: An Ethnomethodology of Law, Morality, and Justice (Ashgate 2011); Baudouin Dupret, Michael Lynch and Tim Berard (eds), Law at Work: Studies in Legal Ethnomethods (Oxford University Press 2015). 42 24 surveys or experimental research, the case study does not always lend itself perfectly to generalization to a wider population of similar cases. But that does not mean the case study is devoid of generalizing value. While random sampling (in survey research) or random assignment (in experimental research) are procedures that ensure at least the possibility of gauging representativeness, the researcher who has conducted a case study has usually not selected the case at random, nor can the single case in its individuality tell us something about a broader class of similar cases. At the same time, cases are usually selected purposively. Case selection procedures focus on typical, diverse, extreme, deviant, influential, most similar, or most dissimilar cases.43 This purposive sampling automatically implies that the ambitions of the case study are never only particularistic: “the ‘case’ implies a family; it alleges that the particular is a case of something else . . . Cases claim to represent general categories of the social world.”44 The question with our case study was, then: Of what is this case a case? It was helpful at this point to think not in terms of general, social laws (and deviations thereof), but rather in terms of family resemblances. The notion of family resemblance is derived from Wittgenstein.45 Phenomena, to Wittgenstein, can be tied together not because they share a common identity—for instance, because they spring from the same social norm or are exemplifications of the same social law—but rather because they share resemblances to each other. They are recognizable as members of a “family” or category not because they are identical, but because they share features with each other. With this conception of family resemblances, we could start to address the question of generalization in a more fruitful manner. Our case study is not representative of all judicial practices in that it does not seek out regularities and patterns that recur everywhere in such practices: After all, judicial practices differ immensely between legal fields, national jurisdictions, and through history. However, there are concrete ways in which these judicial practices bear affinity with and resemble other practices, legal or otherwise. Our claims that judicial practices are not purely “cognitive” but take place in relation to the physical properties of legal case files and in socially textured communities of practice may for instance sensitize other researchers of other work practices to pay attention to the medium-dependency of knowledge work. Medicine, academia, or the “creative industries” all tend to be conceived of as sites of practice relying on Jason Seawright and John Gerring, “Case Selection Techniques in Case Study Research: A Menu of Qualitative and Quantitative Options” (2008) 61(2) Political Research Quarterly 294. 44 J Walton, “Making the Theoretical Case” in CH Ragin and H Becker (eds), What Is a Case? Exploring the Foundations of Social Inquiry (Cambridge University Press 1992) 121. 45 L Wittgenstein, Philosophical Investigations (Blackwell 1953). 43 25 their members’ cognitive capacities. Yet how do the material dimensions of these practices shape these forms of laboring? How do infrastructures and their possible “inversion”46 due to digitization affect these practices? Another affinity: Think of bureaucratic practices in relation to our insistence on the crucial role of the case file. The work practices of not only judges, but also other street-level bureaucrats, are mediated in important ways by the “documentary reality” of case files.47 By drawing connections between this judicial practice and other practices similarly reliant on case files, we may again exploit resemblances rather than identities, and suggest fruitful avenues for future research. On a yet more general level, our case is also suggestive of the crucial role played by nonhuman objects in work practices more generally—an insight that has the capacity to rearticulate not only conceptions of “judgecraft,”48 but also the character of the social itself. Indeed, where many researchers tend to privilege human actors and their stories, we have instead concentrated on the entanglements of humans and nonhumans and their coproduction of social life. Generalization on the basis of our case study, then, was not a language game in which we tried to connect a case to a population of cases and make claims about the social laws governing their behavior, but rather a matter of seeking resemblances and affinities. <a>CONCLUDING NOTES: ON OBJECTIVITY AND THE CASE STUDY In the preceding pages, we have aimed to provide some concrete pointers regarding the choices researchers must make in relation to the case study. First, the question is whether their research question indeed merits a case study. In our view, the case study, specifically when coupled with a variety of methods, lends itself particularly well to attending to not only the meaningful and subjective dimensions of social life, but also its practical, situated, and “messy” character. Not satisfied with the statistical–formal modeling of adjudication and sentencing practices, we instead wanted to emphasize its ongoing, practical character as an everyday work practice. We have also demonstrated that individual methods have their limitations and affordances. In other words: Methods are performative, and experimentation with different methods is an important ingredient of (the early stages of) a case study. In our case, the use of interviews to arrive at a sense of the unfolding and constructed character of 46 Geoffrey C Bowker and Susan Leigh Star, Sorting Things Out: Classification and Its Consequences (MIT Press 1999). 47 Dorothy E Smith, “The Social Construction of Documentary Reality” (1974) 44(4) Sociological Inquiry 257. 48 Tata (n 4). 26 “cases” was less valuable than observation of work practice and case-specific informal conversations. Documentary analysis of the case files, particularly in terms of their structure, implication in work practices, and content, proved relevant both for methodological and theoretical reasons: methodological, as it allowed case-specific lines of inquiry; theoretical, as it alerted us to the recalcitrance of these objects in judicial work practices. We have also concentrated on the capacity of “the case”—judicial work practices—to “speak back” and to rearticulate theoretical presuppositions. The case spoke back by showing, first, that the practice of sense making of cases is distributed among clerks and judges rather than isolated at the level of the judge; second, that sense making is materially mediated by the case file rather than a purely cognitive process; and third, that case-summarizing practices render empirically accessible how judges connect the evidence to the charges, rather than viewing the making of this connection as an entirely intuitive process that is only comprehensible to professional judges. Given this, we suggest that criticism of the case study method as tending toward the subjective and a focus on the verification of prior knowledge is misplaced: In fact, it is very well suited to disrupt prior conceptions.49 This argument is reiterated by Julien Etienne in his chapter on the case study in administrative law. He encountered observations on self-reporting behavior about “bad news” that conflicted with the prevailing hypotheses derived from rational choice theory, which forced him to generate and test alternative explanations for self-reporting behaviors. When applied with care for the object studied, the case study has the power to surprise, and precisely therein lays its crucial value to social scientific work. With regard to the issue of generalization, our chapter has paid attention to the way “the case” may not unproblematically represent a wider population of cases, but can nevertheless be strategically mobilized along different vectors of resemblance. Indeed, many “cases” are multiple:50 they cannot easily be contained by one explanatory framework, nor is their value restricted to one theoretical problem only. Their strength resides precisely in the different connections they allow us to make between empirical reality and theory, but also between different objects of research. Hence, our case study not only allowed us to place interventions within sociological debates about law in action, but also gave us the opportunity to position ourselves in relation to debates about bureaucratic practices, the role of documents in work practices, and methodological debates on the merits of including in one’s accounts the agency See also Donald T Campbell, “Degrees of Freedom and the Case Study” (1975) 8(1) Comparative Political Studies 178; Bent Flyvbjerg, “Five Misunderstandings about Case-Study Research” (2006) 12(2) Qualitative Inquiry 219. 50 Mol (n 35). 49 27 and recalcitrance of nonhuman actors. Ultimately, our case is evocative of how “the social” and “the legal” are both human and nonhuman constructions. We also want to reiterate that the specific details of these pointers depend on the interaction between the researcher and the particularities of the research setting and the content of the research question. For example, it is conceivable that addressing the question of how judges construct cases and get things done on a daily basis might have brought to light different limitations and affordances of methods had the research been executed in an adversarial legal system. In such a system, the role of judges is restricted to ensuring the legality of the procedures followed by opposing lawyers in establishing and challenging the facts of a case in the courtroom to a greater degree than is the case in the Dutch inquisitorial system. This may imply that in an adversarial setting phonocentricity may be less of a problem, and reading the file may have less added value in understanding judges’ methods of case making. Had the research question focused, to name another example, on the question of how the implementation of key performance indicators interferes with the daily practice of judicial decision making, it may have been more useful to tap into the conscious deliberations and reflections of judges by interviewing them than to shadow judges in their work on cases. This shows that the pointers we have derived from our case study ought to be read not as hard and fast rules for case studies, but rather as broad checkpoints for choosing methods in case studies when producing data in interactions with both people and things. Running through this chapter has been the affinity between legal practices of case making and sociological practices of doing case studies. We are comfortable with that affinity; in fact, we think there is much to learn from jurists. Think of the specific kind of objectivity that is valued in legal practices. This objectivity is of a different kind than that of the disinterested scientist, who is not supposed to judge the phenomenon under study. Well, there is a real sense in which the case study explicitly asks for the exercise of situated and reflexive judgment: not the question of what, in general, is the best course of action, but rather that of what is to be done in the here and now. This we understand to be the objectivity demanded of those doing case studies: an objectivity predicated not only on technical knowhow, but also on the more humble and self-aware capacity to make judgments in the here and now of inquiry.51 Those doing case studies cannot put their faith in sophisticated measurement instruments, nor can they seek technical solutions; instead, together with their notepads, computers, and prying eyes, they are the measurement instrument. This is not to say that the case study and its 51 Flyvbjerg (n 49). 28 methods are therefore subjective. The opposition between the objectivity of the survey or the experiment and the subjectivity of the case study is a bad opposition. It is precisely by reflecting on, and remaining aware of, the performativities of one’s situatedness and one’s method that one arrives at a more robust conception of what, objectively, is going on. Such a prudential appraisal of “what is going on here” is impossible without losing sight of the limitations of one’s approach. Like judges, we must always keep in our minds that our “file” is never a full and accurate picture of what really happened. Our methods help us, but also render things inaccessible or invisible. Our social identities can be a boon or bane in the field, but in any case are consequential in the way we encounter others. We cannot leave our selves at home, nor can we disengage from the field to search for a detached “view from nowhere.” We are part of the knowledge we make. The task we face, then, is one of remaining accountable for what we have learned,52 and of making sure we never claim to have had the last word. For here is a real contrast between jurists and scientists: While the first write arrêts—stops, halts to the discussion by rendering verdicts—scientists write “please-goons.”53 On that note, we urge you, please go on. <a>BIBLIOGRAPHY Bade JJ and Van der Nat MM “Discrimineert de rechter?” (Does the Judge Discriminate?) (2012) Nederlands Juristenblad 1189. Blumer H, “What Is Wrong with Social Theory” (1954) 18 American Sociological Review 3. Bowker G and Star SL, Sorting Things Out: Classification and Its Consequences (MIT Press 1999). Butler J, Gender Trouble (Routledge 1990). Campbell DT, “Degrees of Freedom and the Case Study” (1975) 8 Comparative Political Studies 178. Clifford J and Marcus GE, Writing Culture: The Poetics and Politics of Ethnography (University of California Press 1986). Daston L and Galison P, “The Image of Objectivity” (1992) 40 Representations 81. Derrida J, Of Grammatology (Johns Hopkins University Press 1967). 52 53 Haraway (n 12). Latour (n 32). 29 Desrosieres A, The Politics of Large Numbers: A History of Statistical Reasoning (Harvard University Press 1998). Dewey J, The Essential Dewey (Indiana University Press 1999). Van Dijk TA, “Principles of Critical Discourse Analysis” (1993) 4 Discourse & Society 249. Dupret B, Adjudication in Action: An Ethnomethodology of Law, Morality, and Justice (Ashgate 2011). Dupret B, Lynch M and Berard T, Law at Work: Studies in Legal Ethnomethods (Oxford University Press 2015). Evans-Pritchard EE, The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People (Clarendon Press 1940). Feagin J, Orum A and Sjoberg G, A Case for Case Study (University of North Carolina Press 1991). Flyvbjerg B, Making Social Science Matter: Why Social Inquiry Fails and How It Can Succeed Again (Cambridge University Press 2001). Flyvbjerg B, “Five Misunderstandings about Case-Study Research” (2006) 12 Qualitative Inquiry 219. Garfinkel H, “Studies in Ethnomethodology” in Studies in Ethnomethodology (Blackwell Publishing 1967). Geertz C, After the Fact: Two Countries, Four Decades, One Anthropologist (Harvard University Press 1995). Haraway D, “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective” (1988) 14 Feminist Studies 575. Hart HLA, The Concept of Law (Oxford University Press (1961 [1994]). Hirschauer S, “Putting Things into Words: Ethnographic Description and the Silence of the Social” (2006) 29 Human Affairs 413. Holvast NL, “Considering the Consequences of Increased Reliance on Judicial Assistants: A Study on Dutch Courts” (2014) 21 International Journal of the Legal Profession 39. James W, The Meaning of Truth: A Sequel to “Pragmatism” (Prometheus Books 1997 [1909]). Kirk J and Miller ML, Reliability and Validity in Qualitative Research (Sage 1986). Lanzara GF, “Reshaping Practice across Media: Material Mediation, Medium Specificity and Practical Knowledge in Judicial Work” (2009) 30 Organization Studies 1369. Latour B, Reassembling the Social: An Introduction to Actor-Network Theory (Oxford University Press 2005). 30 Latour B, The Making of Law: An Ethnography of the Conseil D”Etat (Cambridge University Press 2010). Law J, “Seeing Like a Survey” (2009) 3 Cultural Sociology 239. Lévi-Strauss C, Tristes tropiques (Criterion Books 1961). Marcus G, “Ethnography in/of the World System: The Emergence of Multi-Sited Ethnography” in Marcus G (ed), Ethnography through Thick and Thin (Princeton University Press, Princeton 1995) 79–104. Mascini P and others, “Understanding Judges“ Choices of Sentence Types as Interpretative Work: An Explorative Study in a Dutch Police Court” (2016) 37 Recht der werkelijkheid 32. Mol A, The Body Multiple: Ontology in Medical Practice (Duke University Press 2002). Nadai E and Maeder C, “Fuzzy Fields: Multi-Sited Ethnography in Sociological Research” (2005) 6 Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, Article 28. Onwuegbuzie AJ and Leech NL, “Validity and Qualitative Research: An Oxymoron?” (2007) 41 Quality & Quantity 233. Van Oorschot I, “Ways of Case-Making” (Erasmus University Rotterdam 2018). Van Oorschot I, “Seeing the Case Clearly: File-Work, Material Mediation, and Visualizing Practices in a Dutch Criminal Court“ (2014) 37 Symbolic Interaction 439. Van Oorschot I, Mascini P and Weenink D, “Remorse in Context(s): A Qualitative Exploration of Remorse and its Consequences” (2017) 26 Social & Legal Studies 359. Scholten P, Dorsten naar gerechtigheid (Thirsting for Justice) (Wolters Kluwer 2010). Seawright J and Gerring J, “Case Selection Techniques in Case Study Research: A Menu of Qualitative and Quantitative Options” (2008) 61 Political Research Quarterly 294. Shapiro SJ, “What Is the Internal Point of View?” (2006) 75 Fordham Law Review 1157. Smith DE, “The Social Construction of Documentary Reality” (1974) 44 Sociological Inquiry 257. Stengers I, “Introductory Notes on an Ecology of Practices” (2005) 11 Cultural Studies Review 183. Tata C, “Conceptions and Representations of the Sentencing Decision Process” (1997) 24 Journal of Law and Society 395. Tata C, “Sentencing as Craftwork and the Binary Epistemologies of the Discretionary Decision Process” (2007) 16 Social & Legal Studies 425. 31 Walton J, “Making the Theoretical Case” in Ragin CH and Becker H (eds), What Is a Case? Exploring the Foundations of Social Inquiry (Cambridge University Press 1992) 121–37. Wermink H, De Keijser J and Schuyt P, “Verschillen in straftoemeting in soortgelijke zaken: Een kwantitatief onderzoek naar de rol van specifieke kenmerken van de dader” (Differences in Sentencing Outcomes of Comparable Cases: A Quantitative Study into the Role of Specific Offender Characteristics) (2012) 87 Nederlands Juristenblad 647. Wittgenstein L, Philosophical Investigations (Blackwell 1953). Yin RK, Case Study Research: Designs and Methods (Sage Publishers 1989). 32 33