Laboratizing and de-laboratizing the world

Abstract

How has sociology framed places of knowledge production and what is the specific power of the laboratory for this history? This article looks in three steps at how sociology and Science and Technology Studies (STS) have historically framed the world as laboratory. First, in early sociology, the laboratory was an important metaphor to conceive of sociology as a scientific enterprise. In the 1950s, the trend reversed and with the emergence of a ‘qualitative sociology’, sociology was seen in opposition to laboratory work. With the ascent of laboratory studies, the laboratory perspective was again applied to many fields, including sociology itself. Based on a definition of a laboratory as aiming at placeless knowledge and being inconsequential this article argues that the two waves of laboratorization were metaphorical and did not really turn the world into a laboratory. Instead, two alternative concepts, those of the unilatory and the locatory, are proposed to gain a more precise understanding of some of these metaphorical uses of the term ‘laboratory’.

Keywords

epistemology field history of sociology laboratory place STS

Introduction

In recent years, science studies and actor-network theory (ANT) have expanded the notion of the laboratory. For example, Bruno Latour writes: ‘having extended laboratory life to all of our collective existence, it seems that, as the project of modernism gradually exhausts itself, there is now no fact that is not also a cause or a claim’ (Latour, 2004: 24). Similarly, Miller and O’Leary in an article that analyses ‘the factory as laboratory par excellence’ claim that ‘Science and Technology Studies need to take a much wider view of what counts as a laboratory. [They] need to clear away the lingering demarcationism that characterizes the discipline and address those practices that seek to act upon and transform the world in specific and relatively bounded locales, even if this takes place outside the laboratories populated by physicists, chemists and the like’ (Miller and O’Leary, 1994: 470).

In this article, I would like to examine this extension of the laboratory historically. I follow the use of the term ‘laboratory’ in sociology and science studies and how it is employed as a tool to differentiate specific places of knowledge production. However, unlike Miller and O’Leary, I do not interpret the challenge of what might be termed the ‘laboratization’ of the world as one of overcoming demarcationism. Rather, I see contemporary practices of laboratization as a return to an earlier metaphorical use of the term laboratory that seeks to transfer the epistemic authority of science to other fields. Instead of participating in this transfer of authority, I want to understand historically the laboratization (and de-laboratization) of the world by sociology as well as introducing a number of alternative concepts which allow for the more precise analysis of different forms of research and how they relate to their location.

At the same time, this is not a history of laboratory experimentation in sociology. Also, it is well understood that the supposed difference between lab- and non-lab-produced knowledge is often not empirically sustainable. I am merely interested in the varying uses of the term laboratory and its varying metaphorical levels. The goal of this exercise is thus analytic. It is a case study of when, why and how sociology uses descriptions of one kind of place to describe other places.

André Kieserling has noted that ‘modern society has produced a rule for swear words in which its own structures are mirrored’ (Kieserling, 2004: 128). The rule consists in attributing words from one sphere to another sphere, such as when describing religion with the words of the economy (priests as accumulating capital) or the law with the words of religion (judges as priests), etc. The same rule works with specific places: one needs simply to describe a university as a factory (for producing graduates) or a court as a market (for bargaining for sentences). The extended use of the term laboratory is another instance of this rule, but with far more complex outcomes. As I show, the laboratory (within sociology) can be used in a metaphorical way both to justify and to critique sociology’s practices, and the metaphorical extension of the term varies heavily over time.

My objection to the metaphorical use of the laboratory is not an exercise in controlling language. I do not want to limit the sociological imagination by banning metaphors. Rather, I want to highlight a specific career of a specific metaphor that has in recent times resurfaced, and that obscures rather than clarifies sociological theorizing of the relationship between place and epistemology. I propose new terms myself to describe some of what the authors discussed here call a laboratory, but I consider these terms to be more precise and differentiating.

My article develops the argument in two parts. First, I begin by giving a definition of the laboratory as a result of a process to differentiate a controlled inside from an uncontrolled outside, thereby producing both a notion of placeless knowledge and the possibility of inconsequential action. Second, I analyse the history of the use of the term laboratory in sociology historically in three steps. In the first phase, until the Second World War, several notions of laboratory existed next to each other. Apart from the standard definition as the counterpart to the field, several other notions of laboratory existed that conflated the field and the lab, thus leaving the laboratory without its other, placeness and consequential action. My thesis is that the term laboratory in these instances was used in a metaphorical way. These early sociologists used the term to indicate that sociology was a science too, with an object that could be studied empirically. They did not attempt to separate an inside from an outside in order to allow for placelessness and inconsequential actions.

Only after the Second World War, in a second step, the lab–field dichotomy was established in the common sense of the term. Real laboratory experiments were done, but also critically examined. It was understood that sociology would in most cases not profit from becoming a laboratory science and the term ‘field’ became much more important. Because the laboratory was now real, but appeared not very suitable to many sociologists, the metaphorical invocation of the laboratory disappeared.

Third, since the 1990s, following the success of laboratory studies, the notion of lab returns to a more metaphorical use. As I will show, this happens primarily to extend the scope and methods of Science and Technology Studies (STS) to other places in society. I discuss four different kinds of expansions and contend that they return to a metaphorical use. In order to clarify these uses, I propose for two of these uses non-metaphorical and more precise concepts, the locatory and the unilatory. My discussion is restricted mostly to American sociology and does not purport to be a real history of the lab–field distinction in sociology, but a conceptual clarification.

The definition of the lab

Here is my definition of a laboratory. The laboratory is the result of a procedure that separates between an outside, an environment that is considered negligible for some epistemic claim or technological invention, and an inside, a (partly) controlled environment that is considered relevant for this claim or invention. The lab is not so much a closed space, but a procedure that often results in a space with the properties to separate controlled inside from uncontrolled outside. Control means not necessarily physical control but a procedure whereby data and objects are managed to behave in a way the scientist wishes. The separation between inside and outside allows for the two central features of the lab, placelessness and consequence-free research.

Before I elaborate this definition, let me compare it with other definitions of the laboratory. My definition of the laboratory is based on central tenets of laboratory studies, most notably, it is procedural and praxeological and moves away from questions of validity (Knorr Cetina, 2001: 8232). This definition also shares with laboratory studies a stress on the interventionist character of the laboratory and the idea that the laboratory is an assemblage of technologies and practice. But my definition is analytical and it does not try, as historians do, to capture every historical notion of the term laboratory (Gooday, 2008). More specifically, it allows one to differentiate laboratories from rhetorical invocations of the term laboratory. My goal is to differentiate the laboratory from what Karin Knorr Cetina has called ‘the laboratory perspective’, i.e. seeing the world as lab (Knorr Cetina, 2001: 8237). To do so, the only, but crucial, difference to laboratory studies is that this definition aims at differentiating the laboratory from other places of knowledge production exactly because it is a place of a highly specific form of socio-material interventions. Most obviously, the other side of the distinction is the field. The field is an environment for epistemic claims without a distinction between a controlled inside and a non-controlled outside. Because of the lack of this distinction the borders of a field are blurry and have to be defined by the researcher.

Let me look more precisely at the two defining main features of the laboratory. First, because of the distinction between inside and outside, the laboratory is a mechanism for generalization, since epistemic claims or objects derived from labs can be extended to other non-controlled environments. This is why Robert Kohler calls a laboratory placeless (Kohler, 2008: 766). More precisely, the laboratory is a mechanism for generalization, because it consists of two parts. One part, the lab, is stable, and the other part, the knowledge object – for example, a rat – is unstable. The controlled environment is stabilized and known beforehand. It is a proper technology in the sense that its stable qualities always produce the same output with the same input. If a rat runs with different speeds in a treadmill, the speed difference cannot be attributed to the treadmill but the rat.

Second, a laboratory allows what Krohn and Weyer called ‘consequence free research’ (Krohn and Weyer, 1994: 181). They observe that science works with the promise of ‘containment’. The practical operations in the lab, as well as epistemic operations of science, are both supposed to be without real-world consequences and reversible. As operations in the lab, an experiment, if it goes wrong, poses no danger to the outside world (obviously, if it works, it may be implemented, and then change the world, but the experiment itself does not). The lab as containment for inconsequential actions is also a source of unease and is constantly put in doubt. The unease is the source for an endless stream of fictional accounts about what happens if the objects cannot be controlled, escape from the lab and become consequential.

The lab as a container for inconsequential actions exists only because scientists created boundaries that separate the lab from the world. This does not imply that a lab is necessarily a room or a building, but it refers to the fact that a lab has to be created by scientists and that containment needs work to render actions inconsequential.

Both of these features are not mere rhetorical tricks, they cannot be achieved by writing. Mere generalization or mere claims that actions are inconsequential do not indicate a lab. These two features depend on scientific operations that first have to establish a boundary between an uncontrolled outside and a controlled inside and then operate on an object inside the laboratory. After these steps have been accomplished the laboratory has to be reduced slowly in order to test whether the facts and objects remain stable without it. This includes the attribution of anomalies that occur outside the lab to the particular circumstances and not to the lab. This step is only possible if before, in the lab, a proper distinction between lab and object has taken place.

Only after these operations have been completed does it become possible to generalize knowledge-claims and render them placeless. Only then is it possible to claim that the actions are inconsequential. Since a laboratory is a technology, it is also possible to search for its failure if either generalization or containment fails. Latour calls these steps ‘chains of translations’ or ‘networks’ and it is these chains of translations that stabilize facts outside the lab (Latour, 1987). Mere generalizations are non-stabilized claims.

This definition is not tied to any specific notion of science. A lab is neither needed for science, nor are laboratories always used for science in the strong sense. Labs are mechanisms to control knowledge objects and allow for inconsequential actions, which can even take place outside science, as, for example, in industry. Thus my use of the term is compatible with the uses of the term before the 19th century as described in Klein (Klein, 2008). Neither placelessness nor inconsequential action is an absolute term, but obviously they form a continuum. Different labs may be able to construe different levels of placelessness and inconsequential action. The term laboratory is a relative term in a space defined by a continuum (Kohler, 2002). Let me now look at how the historical use of the term laboratory relates to this definition.

Extending the notion of the term laboratory in early sociology

In early sociology the term laboratory was used regularly. The standard notion was used by several authors (Chapin, 1917a, 1917b; Kellor, 1901). They were influenced by neighbouring disciplines such as criminology, and tried to turn sociology into an experimental science. But beside the standard notion, varying other notions of laboratories flourished. Most notably in Chicago, the city, and the field in general, were seen as laboratories.

For the case of Chicago sociology, a range of texts on the use of the term laboratory already exist (Gieryn, 2006; Egloff, 2007; Gross, 2009; Gross and Krohn, 2005). From this literature one might get the impression that the extended use of the term laboratory was specific for the Chicago School and for the city as a place of research in particular. However, the extension of the laboratory in Chicago was but one example of a then common extension of the term laboratory in sociology. My discussion also notably differs from the one of Thomas Gieryn (Gieryn, 2006). In his analysis of the Chicago School he accepts the actors’ definitions of laboratories and reads them as ways to turn Chicago into a real placeless laboratory, as when he summarizes his discussion with: ‘the “wild, natural’ city has been made in the laboratory and brought under experimental (and maybe political) control’ (ibid.: 16). At least at some moments in his analysis – at others less so – he seems to assume that the rhetorical moves of the Chicago School had the effect of really turning the city into a lab. Contrary to this interpretation, I contend that this never happened. The Chicago School and the other sociologists who used the term lab, never really attempted to turn their places of fieldwork into placeless spaces where inconsequential action was possible.

Most often, if an extended notion of laboratory was used, it was used to describe any place for knowledge production, a notion that included placeness as well as consequential action. Since society could be in principle everywhere, it was not clear at all in early sociology where to look. The term laboratory did a twofold job: first, it secured that sociology was a science, by invoking the laboratory as a term that referred to the natural sciences; second, it allowed the framing of a particular place (mostly understood in the topographical sense) to be a legitimate object of inquiry. Let me start with those uses of laboratory that invoked science and then move on to those that framed a particular place as a legitimate site of research.

For example, in Chicago, George E. Vincent entitled an article ‘A Laboratory Experiment in Journalism’ (Vincent, 1905), but the paper simply describes a course in the history, theory and practice of journalism without even mentioning laboratory in the text, let alone discussing it. This generalized notion of laboratory was often used in the context of education, where the term did not invoke knowledge production, but a space where students would be trained not with books but with real-world objects, which included ‘fieldwork’ (Melvin, 1925; Aldrich, 1940).

Another use was to call the sociology institutes themselves laboratories. For example, W. E. B. Du Bois called the institute in Atlanta, which was founded in 1895 and which he had headed since 1897, a ‘sociological laboratory’ (Du Bois, 1903; Wright, 2002). In his article, the laboratory is not really defined, but it becomes clear that the laboratory was not so much the institute, but the location of the laboratory in the city as a concrete place to study African-Americans and, moreover, a place where they could be studied as a category since they were living more or less segregated.¹ This notion he later clarified in his autobiography Dusk of Dawn: ‘Social scientists were … still thinking in terms of theory and vast and eternal laws, but I had a concrete group of living beings … capable of almost laboratory experiment’ (Du Bois, 1968: 64).

Similarly, Jacob L. Moreno wrote about ‘the participant observer of the social laboratory, counterpart of the scientific observer in the physical or biological laboratory’ as the main role of the sociologist in his project of ‘sociometry’ (Moreno, 1937: 210). The sociological laboratory for Moreno was society itself and the role of the sociologist, whom he called a ‘field worker’ (ibid.: 218), was to animate people to participate in sociometry. This meant that they had to observe and interact with others to produce data for the sociologist in such a way until they became ‘participants in and observers of the problems of others as well as their own’ (ibid.: 211). For Moreno, a laboratory was thus simply a place in which to conduct empirical work, and referred neither to placelessness, nor to inconsequential action.

This confusion was also possible, because the search for a laboratory was only in part a search for a laboratory in the narrow sense defined here, but rather a search for empirical sites in general. For example, Frederick Gruenberg explicitly compared the tendency to conduct fieldwork with the ‘disposition to introduce more and more laboratory work in the biological and physical sciences’ (Gruenberg, 1923: 109). The use of the term laboratory for the field simply indicates that this is a place where empirical research is conducted rather than armchair speculation.

The second use related to specific research sites. The term laboratory referred then to the relative clear-cut boundaries of a specific field, though without trying to control the field. For example, Weatherly in ‘The West Indies as a Sociological Laboratory’ understood remote islands as laboratories because they were ‘isolated’, thus the ‘play of social forces is least disturbed by alien elements’ and one could observe ‘behaviour as nearly spontaneous as is possible under modern conditions’ (Weatherly, 1923: 290). Though he does not give a clear definition of a lab, it is obvious that he considered the West Indies to be labs not because they could be controlled, but because ‘social forces’ occur as found in nature, rather than produced and disturbed in modern cities.²

None of these laboratories allows for placelessness. As Gieryn himself writes: ‘It mattered that observations were done here in Chicago and not just anywhere, because Chicago and its constituent neighbourhoods and social patterns are at least distinctive, and possibly unique’ (Gieryn, 2006: 18). Even though the Chicago sociologists, and others who used the term laboratory, sought to generalize and thereby de-localize their findings derived from their particular research site, this was a rhetorical move (ibid.: 19–20), and not substantiated by control of the supposed lab itself, the city. The rhetorical move of the Chicago School sociologists consisted in claiming that what they found out about Chicago was typical for big cities, but they did not attempt to stabilize the claims with a lab.

Apart from simply using the term laboratory, some authors attempted direct comparisons with the natural sciences. Typical for these comparisons is that they do not seek to emulate the practices of the natural sciences but parallel different kinds of practices in the natural and the social sciences with the term laboratory.

In Chicago, Robert Park claimed that ‘social science has achieved something that approaches in character a laboratory experiment. For the purpose of these experiments the city … becomes … a device for controlling our observations of social conditions in their relation to human behavior’ (quoted from Gieryn, 2006: 15). The laboratory of Park, which only ‘approached’ the natural science lab ‘in character’, is surprisingly not a tool to control the object of research, but the researchers themselves.

Another use related to the lab as a closed – but not controlled – space to do research. Wilber Newstetter from Western Reserve University wrote in an article on measuring group adjustment: ‘There are many laboratories for the experimental study of the common cold; there are few laboratories for the experimental study of social behavior set up and directed from the point of view of sociology’, hoping for a sociology whose aim would be to have an ‘objective basis for the prediction and control of social behavior’ and airing his disappointment with ‘past sociology (Newstetter, 1937: 230). His laboratory, a ‘summer camp’, was defined by being a closed space, not so much one in which an experimenter could control the inmates, but one that allowed for overview. Newstetter also went on to conflate the lab and the field, when he wrote: ‘The study began in 1924 and the field investigation ended in 1933’ (ibid.: 230).

Similarly, Edwin Sutherland from the University of Chicago understood the prison as a laboratory, where he could have easy access to prisoners, and he already understood that the prison as laboratory was an ‘artificial’ environment. Thus he compared the advantages of the lab with the field: in the lab, the prisoner is ‘not in his “natural habitat’’. Which raised the criticisms that ‘a criminal can no more be understood in prison than a lion can be understood in a cage’ and prisoners should be ‘studied “in the open’’ (Sutherland, 1931: 132). But his understanding was based on the difference between the prison and ‘the open’ as more or less natural ‘habitats’ and not as places with more or less control. In each of these cases it is apparent that the authors do not try to turn these spaces into controlled spaces, but that they need the term laboratory to explain something about their concept of social research for which they are missing other words.

The metaphorical use of the laboratory also backfired, and from early on, specifically in Chicago sociology, the term laboratory was criticized too: William Henderson, already in 1899, opposed the notion because it implied ‘inquisitive investigators [who] may pursue methods of vivisection and torture, in order to illustrate or test sociological theories’ (quoted in Gross, 2009: 85). Jane Addams opposed it, because she believed that the settlements were ‘something much more human and spontaneous than such a phrase [laboratory] connotes’ (quoted in ibid.). The word ‘connote’ makes clear that she understood that settlements were in fact no laboratories and the use of the term was metaphorical. Both of these criticisms indicate that the Chicago sociologists had a clear understanding that their sociology was based on fieldwork that could not undertake and should not attempt inconsequential action. As Gross points out, Jane Adams did not use the term laboratory but rather the term experiment. Experiment for her was precisely not the contained inconsequential action, but exploratory tests of new social situations outside a laboratory, in the real world (ibid.: 86–8).

In general, in these early days of sociology in Chicago and elsewhere, laboratory and field were not necessarily understood to be opposing terms, and ‘fieldwork’ was not understood as it is today as ‘qualitative’ and opposed to quantitative or lab work. Fieldwork was not even a strong category; in Chicago the term ‘case study method’ was used instead. The term laboratory did not imply an attempt to create a proper laboratory. Rather, it was a sign of a nascent discipline trying to reflect on and define its methods.

The stabilization of the field–laboratory dichotomy in sociology

After the Second World War, when the differentiation between the social and natural sciences became more stable, the terms lab and field became differentiated and were used in their standard form as controlled environment vs. non-controlled environment (Sherif, 1954; O’Rourke, 1963). This development was probably mostly an effect of the success of philosophy of science and its application to the methodology of social science and the actual attempts to do proper laboratory experiments in sociology (for an overview see Bonacich and Light, 1978). After the Second World War, philosophy of science increasingly turned towards social science, and social scientists themselves showed an interest in clarifying and thoroughly comparing their own research techniques with those of other sciences, which resulted in fairly sophisticated methodological discussions (Lazarsfeld and Rosenberg, 1955; Braybrooke, 1965). The methodological texts of the 1950s and the 1960s did not create a rift between lab and field. Rather, they were driven by the urge to refine the empirical tools of sociology and were thus directed against theoretical synthesis in the wake of Talcott Parsons.

The methodologists in the social sciences accepted the methodology of the natural sciences and the meaning of specific terms, such as laboratory, experiment, testing of hypothesis, causation, induction and deduction. These discussions made the metaphorical use of laboratory impossible. An early flavour of this was already sounded by Howard Becker in 1940:

In spite of high-sounding phrases in graduate school bulletins about Harlem or rural Iowa or gangland Chicago as ‘a sociological laboratory’, most sociologists know full well that they cannot experiment, that they are not laboratory scientists, and that in the opinion of many competent judges they never will be. (H. Becker, 1940: 44)³

This led to the question what a laboratory in sociology could be, if it existed at all (see, for example, Ackoff, 1953: 2 ff.; Kaplan, 1964: 126–70). Specifically, with the increasing use of quantitative and statistical techniques, the laboratory as an ideal became problematic. Control over the situation and the reduction of variables appeared unnecessary:

The development of methods of ‘multivariate analysis’ has removed the necessity of manipulation and the laboratory, and it permits the scientist to go out into the world and tackle increasingly complex problems in their natural habitat. (Ackoff, 1953: 4)

The laboratory became clearly defined, but exactly this definition questioned its use for sociology: ‘Whether or not the work of sociology a decade hence will take the form of laboratory experimentation, it seems clear that few of our present techniques operate at their greatest effectiveness in that situation’ (Goode and Hatt, 1952: 94).

Indeed, in the following two decades, laboratory experimentation had been applied to many sociological problems, pioneered by Robert Bales in the 1950s and continued by scholars like Joseph Berger and Morris Zelditch. These studies ranged from role differentiation to the prisoner’s dilemma. The sociological laboratory had ceased to be a metaphorical concept to legitimate sociology. But exactly because the laboratory became real, it also became more problematic: the laboratory was only partly suited for sociological work; even the most avid proponents of laboratory experiments understood that ‘the variables in the laboratory seem to bear little relation to the names the experimenter gives to them’ (Bonacich and Light, 1978: 166). For sociology, the very difference between the inside and the outside of the real laboratory was considered to be too big to handle: ‘it is time … to leave the lab’ (ibid.: 167).

Conversely, ‘field work’ started to get scrutinized as well and could no more be confused with a laboratory (H. S. Becker, 1958; Zelditch, 1962). Howard S. Becker, for example, wishes that ‘qualitative research may become more a “scientific’ and less an “artistic’ kind of endeavor’ (note that the link is science, rather than a lab, and against ‘art’; H. S. Becker, 1958: 660). Thus, since the 1950s, a laboratory in sociology was strictly defined: it was neither an exemplar for those interested in statistics nor one for those who became qualitative sociologists and it could not be confused with a city, an island, or a summer camp as research site any more.

Only the thorough acceptance of philosophy of science and the emergence of real laboratories allowed developing alternatives. Qualitative social science could not call its field a lab any more. The field slowly became the romanticized counterpart to the lab and quantitative data. For a brief time, in the early 1960s, laboratory and experiment could still be thought to be compatible with an interest in verstehen (Schutz, 1965). At that time in qualitative sociology, typical themes of the Chicago School, such as the role of the stranger, were tested in the laboratory (Nash and Wolfe, 1957). As Garfinkel (Garfinkel, 1952, 2001) and Cicourel (Cicourel, 1964) intended to show, laboratory experiments could be used to prove the invariant foundations of social order based on ‘verstehen’. For Garfinkel and Cicourel, laboratory experiments were problematic only if they took common assumptions between experimenter and subject for granted, rather than using experiments to test the foundations of this reciprocal process.

However, soon ethnomethodology and what now became ‘qualitative’ sociology dropped this interest in experiments and the idea of ‘verstehen’ became narrowly identified with uncontrolled field situations.⁴ In this view, the procedure of ‘verstehen’ required a completely different epistemology, which could not be framed with the existing tools of philosophy of natural science based on the exemplar of the laboratory. A typical representative for this shift is Herbert Blumer, who was trained in Chicago and started his career by positively discussing laboratory experiments in his unpublished dissertation from 1929 ‘Method in Social Psychology’ (Hammersley, 1989: 136–7). But later, in 1969, in his central book on symbolic interactionism, he was at pains to define symbolic interactionism as a kind of ‘naturalistic investigation … that is directed to a given empirical world in its natural, ongoing character instead of to a simulation of such a world, or to an abstraction from it (as in the case of laboratory experimentation)’ (Blumer, 1969: 46). Indeed, the very idea of emulating natural science was mistaken: ‘Symbolic interactionism is not misled by the mythical belief that to be scientific it is necessary to shape one’s study to fit a pre-established protocol of empirical inquiry, such as adopting the working procedure of advanced physical science’ (ibid.: 48).⁵ Laboratory experiments were excluded from qualitative sociology and were identified with the new sister discipline social psychology (Good, 2000: 392).⁶ The laboratory was now the other of qualitative sociology. It was associated with useless aspirations of sociology to become a natural science and it was taken to represent a mistaken attempt to control, what in essence could not be controlled, namely interactions.

In short, no later than in the 1960s, the notion of laboratory in sociology had been in sync with the standard use of the term. A laboratory was now a controlled environment. Laboratory experiments led a shadowy existence, because qualitative sociology thought they would make ‘verstehen’ impossible and quantitative sociology had no use for them either.

With the distinction between lab and field firmly in place, new nuances between laboratory and field were found too. For example, in the 1960s Donald T. Campbell developed the notion of ‘quasi-experiments’, a sociological analysis of public reforms understood as a kind of experiment with and in society (Campbell, 1971). His aim was to improve policy-making by scientifically evaluating social reform programmes for their efficiency rather than for their political uses. However, Campbell very consciously calls them ‘quasi-experiments’, because they do not take place in a laboratory, but in society itself, and only allow for partial control (also see Gross, Hoffmann-Riehm and Krohn, 2005; Gross and Krohn, 2005).

The renewed expansion of the lab after its definition

The current use of the term laboratory again blurs this distinction, sparked by a renewed interest in laboratories as places of doing science. In the early 1980s Karin Knorr Cetina, Bruno Latour, Steve Woolgar and Michael Lynch published the first ‘laboratory studies’ (Latour and Woolgar, 1979; Lynch, 1985; Knorr Cetina, 1981). They provided for the first time an empirical analysis of what scientific laboratories do. In this and several subsequent studies, they established a detailed understanding of the features of laboratories, from which the definition in the previous section is drawn. Thus, today’s extension of the ‘laboratory’ takes place after two decades of specification of the term.

In early sociology, STS did not exist, and the notion of laboratory was a metaphorical aid to understand what sociology is doing. To call the city a lab, was a means to specify the status of sociology as a science at a moment when this status was contested and unclear. The renewed use of the term laboratory to designate all kinds of research takes place in a situation where the scientific status of STS and sociology is not contested. Rather scholars trained in STS are expanding their objects from science to other fields, such as the built environment, architecture and infrastructure. This renewed extension of the laboratory returns to the earlier ones under the umbrella of STS. My aim is to clarify these extensions in comparison with the earlier ones with the very tools of STS.

I would like to look at the following recent uses of the term laboratory: (1) lab as collaboration; (2) empirical extension of lab space (society as laboratory/real-world experiments); (3) lab as a generalized notion for spaces for knowledge production: the locatory; (4) lab as a container to test objects: the unilatory. Other than the historical uses of the term laboratory, some of the recent instances use the term laboratory not to describe the sociologists’ own practices, but as second order categories to describe other sciences. However, this extension to other sciences conceptually follows the earlier extensions of the laboratory of sociology itself.

1 The laboratory as collaboration: the local control of disciplinary norms

Probably the most unusual expansion of the laboratory comes from Paul Rabinow, Stephen Collier and Andrew Lakoff (Collier, Lakoff and Rabinow, 2006; Rabinow, 2007). They use the term laboratory to designate their way of collaboration in anthropology⁷ specified as ‘joint production of papers’ and ‘concept development, collective reflection, and shared standards of evaluation’ (Collier, Lakoff and Rabinow, 2006: 1). A laboratory for them is a way of doing collaborative work on similar topics across different institutions. Thus their notion of a laboratory has nothing to do with my definition or any other usual definition of a laboratory and could rather be called a network or, more prosaically, a working group.

However, what is remarkable about their ‘laboratory’ is that the authors have a background in STS and are thus familiar with precise definitions of laboratories. They are at pains to deny that they imitate natural science and that they do not aspire to ‘positivistic scientific rigor’ (Collier, Lakoff and Rabinow, 2006: 5). They spend some energy in legitimating their differing use and they use the term to criticize the practice of anthropology. They seek to redistribute the connection between reputation, knowledge-claims and validation with the help of their lab. For them, a laboratory is a counterbalance to the ‘individual project’ that ‘rests on a myth of sui generis intellectual production’, exemplified by ‘Clifford Geertz in Princeton’ (ibid.: 1; Rabinow, 2007: 8). With the individual project, knowledge-claims are validated based on a model, which distributes reputation not along intersubjective criteria but along non-transparent hierarchies.

The laboratory, in their view, allows for, among other things, ‘Experimentation as a way to put concepts to the test’, ‘established agreed-upon demonstrations of adequacy’, ‘search for impersonal methodological norms’, ‘recognition of legitimate authority based on knowledge rather than status’ (Collier, Lakoff and Rabinow, 2006: 4). These goals read as if taken from an introduction to the Mertonian norms of science (Merton, 1973). Their laboratory compensates for the fact that, in their view, current anthropology is not a Mertonian science, but defined by ‘a crisis in thinking about what constitutes a valid claim’ and a non-scientific distribution of reputation (Collier, Lakoff and Rabinow, 2006: 4). Their laboratory is a private, small-scale attempt to turn back anthropology into a Mertonian science. What is remarkable is not so much the implied sorry state of anthropology, but the idea that the problem shall be solved not by changing the discipline but by creating a small network called a laboratory. Whereas in science, the lab is a place for controlling the subjects of research, for Collier, Lakoff and Rabinow it is a place for controlling disciplinary norms.⁸ By doing so, they reverse one of the other scarce notions of the lab in recent anthropology: Bonté and Izard write in their French dictionary in the entry ‘méthode ethnographique’ under the subheading ‘le terrain’: ‘Taking a widely used formula, the field is the “laboratory’ of the anthropologist: it is his vocation to gain field experience until the point where the “first field’ has become the experimentum crucis that decides on a career’ (Bonte and Izard, 1991: 473). Whereas for Bonté and Izard the field-as-lab-as-experimentum-crucis is a public space under the control of the whole anthropological community, for Rabinow et al. the field-as-lab has become defective and is replaced by a private and thus controlled network-as-lab.

2 ‘Society as laboratory’: real-world experiments as de-laboratization of science

A second use of the term laboratory occurs in the thesis of ‘society as laboratory’ (Krohn and Weyer, 1988). The notion of society as laboratory describes an empirical extension of the laboratory space to the whole of society. This observation is driven by recent developments, such as atomic power plants, climate change, or the release of genetically modified organisms that cannot be tested within the controlled spaces of a laboratory. Any test of these technologies by definition takes place in society itself turning it into an ‘experimental society’ (ibid.: 349).

But the notion of laboratory does not really fit the insightful diagnosis. The difference between society as a laboratory and the real laboratory is that the laboratory introduces a difference between an inside and an outside that allows both ‘placelessness’ and ‘inconsequential action’. ‘Real-world experiments’ as the authors call them, do not allow for any of these. Climate change is located – even if on a very large scale. It cannot be generalized because it is already everywhere – except on other planets. Climate change cannot be controlled, tested and then released. It is always taking place in a real world. Krohn and Weyer themselves stress that these experiments cannot be contained. As they observe, with ‘society as laboratory’ or rather, the de-laboratization of science, science begins to resemble more closely the economy, the law, or politics, which do not know a test-mode either (Krohn and Wreyer, 1988: 350).⁹ Consequently, in later texts by the same authors, they expand the concept of real-world experiment and correctly drop the notion of ‘laboratory’ (Gross and Krohn, 2005).

3 Specific places as laboratories; or rats without labs: the locatory

Another recent use of the term laboratory is similar to the earlier uses as a generic notion for places of research. In this sense, the city or a country is a laboratory for economists, sociologists and planners. A typical case for such a use is Bockman and Eyal (Bockman and Eyal, 2002), who look at eastern Europe as a laboratory for economists. They do so by successfully introducing the language of actor-network theory to explain why neo-liberalism as an economic explanation came to be successful after 1989. In their view, eastern Europe is a laboratory, because economists could test knowledge-claims in a specific historic situation that was not available in the West.¹⁰ Similarly, Beth Greenhough uses the case of deCODE’s plans for a genomic database to discuss Iceland as a laboratory for science studies scholars (Greenhough, 2006). Another example is Lapp who uses the term laboratory to describe how American social scientists in the 1950s and 1960s conducted research in Puerto Rico, but without any mention of ANT or science studies (Lapp, 1995).

However, these cases simply follow their precursors such as Newstetter, Sutherland, or Weatherly and describe relatively clearly delineated field sites that do not allow a distinction between a controlled environment and a knowledge object.¹¹ The cases are like rats without a lab. There are knowledge objects to observe, but they cannot be controlled. Rather the opposite, knowledge-claims become located and tied to these places. Also, neither eastern Europe nor Iceland, nor Puerto Rico, allows for inconsequential action. Rather the opposite, everything that happens is consequential for all the inhabitants, which is exactly what makes these case studies so compelling.

In the case of Greenhough and Bockman and Eyal the authors are probably misled by the possibility of applying ANT language (translation, black-boxing, network) to their cases. Because ANT language has been developed with reference to laboratories, the fact that it can be applied to other cases leads the authors to deduce that these cases are laboratories too. But ANT is about to become a general sociological description language, which allows us to see the world as an assemblage of networks and translations (Latour, 2005). Hence, the application of ANT does not render automatically every place into a lab; it only renders the world according to ANT.

I propose to understand the above examples as distinct spaces of doing science and call them locatories.¹² Locatories are places where specific knowledge-claims can be made, which are not possible in other places. Locatories are thus not mere fields, where objects can be observed ‘in the wild’, or according to their ‘natural’ behaviour. Locatories tie the observation to a specific location. This is what makes locatories so valuable. The economists in Bockman’s and Eyal’s story could not build a place to test their theories. Also, they could not just go outside of their offices and observe people. They needed a specific place with specific properties to observe what they needed. A laboratory, wherever it would have been, and however large it would have been, would exactly not allow finding out what they wanted to find out. They needed a locatory, where their theories could be tested in a real-world setting, where real actors would do real actions with real consequences. Unlike a real-world experiment that is global and cannot be controlled, a locatory has defined boundaries. These boundaries may be geographical (an island) but they also may be defined by social definitions (a political system, an economic system, etc.). Actions in locatories are not inconsequential but consequential and typified. That is, locatories allow the understanding of actions not just as actions per se, but as functions of the locatory.

4 Selective control, or: the unilatory as lab without rats

The last extension of the laboratory is the laboratory as a container to test unique things or, in a more general sense, the laboratory as the workspace of designers. This expansion has been used for factory planning (Miller and O’Leary, 1994), the architectural office (Potthast, 1998), policy think-tanks, or creative incubators. A laboratory in this sense would be a space, where professionals manipulate models, drawings, signs and texts that refer to the outside world. The analogy to the laboratory is first and foremost that social scientists conduct a similar kind of empirical observation in those places as in laboratories. As in laboratory studies, sociologists and anthropologists can observe in situ how these professions manipulate their objects. This parallel has indeed been productive, since it made the ethnographic studies of architectural laboratories or think-tanks possible after all (the author is himself guilty of such a ‘laboratory study’ of consultancies).

However, neither consultancies nor architectural offices are laboratories according to my definition. In the case of consultancies and other think-tanks and creative incubators, the work is similar to the economists of Bockman and Eyal. In the case of architectural offices the problem is more complex. One could say that the architectural office is like a laboratory. The office creates a distinction between an inside and an outside, where the object, the building, is under full control of the architect. The architect can preview and manipulate the size, form and structural properties of the building.

But I would argue that this misses an important point about buildings in modernity, namely the fact that they are built to change society, and thus act as technologies on users. But these users cannot be brought into the architectural office and manipulated. What can be manipulated are the physical properties of the buildings. As long as the architect’s task is restricted to the construction of buildings, then the architectural office is indeed a lab. But as soon as the architect’s task is understood to be about creating buildings to produce specific effects in society, then the architect’s office is no lab, since it fails to manipulate and control what it claims to do. Such places I propose to call unilatories. They are defined by the fact that, like laboratories, they do create a difference between an inside and an outside, but the object they purport to manipulate cannot be controlled because it is not in the laboratory. In the case of the architectural office, this is so, because the properties of users are quite different from those of buildings. A plan or a model can precisely represent a building, but not a user. A user can be brought into a laboratory and can be researched in various ways, but architects hardly try to do so. And if they do so, they do not do it in the same space where they draw and model. In an architectural office as unilatory, architects manipulate drawings and models in order to change society, but society itself remains outside. A unilatory is not simply a laboratory that fails to control some aspects of the object. This is obviously true of every laboratory. Laboratories are designed to ignore certain aspects of the object that is manipulated. A unilatory fails to control exactly those aspects of the object that it purports to control. I do not claim that buildings that result from operations in unilatories do not indeed change society. They do, as has been shown over and over. But these changes stand in no direct relation to the operations of the unilatory, such as the formal properties of a building do. This is why I call a unilatory the laboratory without rats. In a unilatory the most elaborate treadmill can be set up, but the rats never fit in.

De-laboratizing the world

In this article I have discussed the history of the uses of the term laboratory in sociology and science studies. Based on my definition of a laboratory as a distinction between an uncontrolled outside and a controlled inside I traced the history of the laboratization and de-laboratization of the world through sociology. The historical view has shown that an initial phase of calling any research place a lab was followed by a de-laboratization of sociology after the Second World War. In this phase, (qualitative) sociology defined itself as the other of the lab. The recent rise of laboratory studies has led to a renewed laboratization of the world. However, rather than following the initial impulse to clarify the relationship between place and science, STS has repeated the universalization of the laboratory. I have critically discussed some elements of the laboratization of the world and proposed two alternatives that may help to further theorize the relationship between place and knowledge without turning the whole world into a laboratory. The first is the locatory, an instance of rats without a lab. The locatory is a specific kind of place that allows observing very specific objects that only exist there and are dependent of the locatory. Second, I defined the unilatory, a lab without rats, a specific kind of laboratory that can control only one part of its object whereas another, equally constitutive part cannot be controlled.

It may seem that this article is driven by an anti-metaphorical stance. I seem to purge sociology of using spatial metaphors to describe research places. But my goal is far more limited and precise: I do not object to the use of metaphors if they help to frame an object, and in early sociology the metaphor of the laboratory was indeed helpful to frame sociology as a legitimate science. But metaphors are only instructive when the metaphorical use of a term is apparent. When the metaphor is taken for the thing itself, as is often the case with the term laboratory, it tends to obscure rather than help to think. The laboratory metaphor in recent sociology has obscured the relationship between knowledge production and place, because it has reduced a diversity of forms to one historical blueprint. The problem here is not that the laboratory is used in a metaphorical way, but that the laboratory as metaphor does not suit any of the objects the authors want to understand and rather leads us to believe that these places have qualities of real laboratories, when in fact they do not. By suggesting the terms unilatory and a locatory I propose to reframe two of these cases and thereby add some precision to our vocabulary.

Footnotes

Notes

References

Ackoff

(1953) The Design of Social Research. Chicago, IL: University of Chicago Press.

Aldrich

J. C.

(1940) ‘The Community as a Laboratory for Social Studies’, Journal of Educational Sociology 14(1) (September): 33–9.

Barnard

Spencer

eds (1996) Encyclopedia of Social and Cultural Anthropology. London: Routledge.

Becker

(1940) ‘Constructive Typology in the Social Sciences’, American Sociological Review 5(1): 40–55.

Becker

H. S.

(1958) ‘Problems of Inference and Proof in Participant Observation’, American Sociological Review 23(6) (December): 652–60.

Blumer

(1969) Symbolic Interactionism: Perspective and Method. Englewood Cliffs, NJ: Prentice-Hall.

Bockman

Eyal

(2002) ‘Eastern Europe as a Laboratory for Economic Knowledge: the Transnational Roots of Neoliberalism’, American Journal of Sociology 108(2): 310–52.

Bonacich

Light

(1978) ‘Laboratory Experimentation in Sociology’, Annual Review of Sociology 4: 145–70.

Bonté

Izard

eds (1991) Dictionnaire de l’ethnologie et de l’anthropologie [Dictionary of Ethnology and Anthropology]. Paris: Presses Universitaires de France.

10.

Braybrooke

ed. (1965) Philosophical Problems of the Social Sciences. New York: Macmillan.

11.

Campbell

D. T.

(1971) ‘Reforms as Experiments’, Urban Affairs Review 7(2): 133–71.

12.

Campbell

D. T.

Stanley

J. C.

(1966) Experimental and Quasi-Experimental Designs for Research. Chicago, IL: Rand McNally.

13.

Chapin

F. S.

(1917a) ‘The Experimental Method and Sociology’, The Scientific Monthly 4(2) (February): 133–44.

14.

Chapin

F. S.

(1917b) ‘The Experimental Method and Sociology II’, The Scientific Monthly 4(3) (March): 238–47.

15.

Cicourel

(1964) Method and Measurement in Sociology. New York: Free Press of Glencoe.

16.

Collier

S. J.

Lakoff

Rabinow

(2006) ‘What is a Laboratory in the Human Sciences?’,ARC working paper, available at: http://anthropos-lab.net/documents/wps/.

17.

Couch

(1987) Researching Social Processes in the Laboratory. Greenwich, CT: JAI Press.

18.

Du Bois

W. E. B.

(1903) ‘The Laboratory in Sociology at Atlanta University’, Annals of the American Academy of Political and Social Science 21: 502–5.

19.

Du Bois

W. E. B.

(1968[1940]) Dusk of Dawn: An Essay toward an Autobiography of a Race Concept. New York: Schocken Books.

20.

Egloff

(2007) ‘Krise und Disziplin. Zur Entstehung der Fachsoziologie an der University of Chicago’ [Crisis and Discipline: the Development of Sociology at the University of Chicago],PhD-Dissertation, Universität Zürich, Forschungsstelle für schweizerische Sozial- und Wirtschaftsgeschichte.

21.

Filstead

(1970a) ‘Introduction’,in Filstead

(ed.) Qualitative Methodology: Firsthand Involvement with the Social World. Chicago, IL: Rand McNally College Publishing, pp. 1–14.

22.

Filstead

ed. (1970b) Qualitative Methodology: Firsthand Involvement with the Social World. Chicago, IL: Rand McNally College Publishing.

23.

Garfinkel

(1952) ‘The Perception of the Other: a Study in Social Order’. PhD dissertation, Harvard University.

24.

Garfinkel

(2001) ‘A Conception of and Experiments with ‘Trust’ as a Condition of Concerted Stable Actions’,in O’Brien

(ed.) The Production of Reality: Essays and Readings on Social Interaction. Thousand Oaks, CA: Pine Forge, pp. 370–80.

25.

Gieryn

T. F.

(2006) ‘City as Truth-Spot: Laboratories and Field-Sites in Urban Studies’, Social Studies of Science 36(1): 5–38.

26.

Giesen

Leggewie

(1991) Experiment Vereinigung [The Experiment of (German) Unification]. Berlin: Rotbuch Verlag.

27.

Good

J. M. M.

(2000) ‘Disciplining Social Psychology: a Case Study of Boundary Relations in the History of the Human Sciences’, Journal of the History of the Behavioral Sciences 36(4): 383–403.

28.

Gooday

(2008) ‘Placing or Replacing the Laboratory in the History of Science?’, Isis 99(4) (December): 783–95.

29.

Goode

Hatt

P. K.

(1952) Methods in Social Research, 1st edn. New York: McGraw-Hill.

30.

Greenhough

(2006) ‘Tales of an Island-Laboratory: Defining the Field in Geography and Science Studies’, Transactions of the Institute of British Geographers 31(2): 224–37.

31.

Gross

(2009) ‘Collaborative Experiments: Jane Addams, Hull House and Experimental Social Work’, Social Science Information 48(1) (March): 81–95.

32.

Gross

Hoffmann-Riehm

Krohn

(2005) Realexperimente: ökologische Gestaltungsprozesse in der Wissensgesellschaft [Experiments with/in Reality: Changing Ecology in the Knowledge Society]. Bielefeld: Transcript.

33.

Gross

Krohn

(2005) ‘Society as Experiment: Sociological Foundations for a Self-Experimental Society’, History of the Human Sciences 18(2): 63–86.

34.

Gruenberg

F. P.

(1923) ‘Some Experiences with Students’ ‘Field Work’, Journal of Social Forces 1(2) (January): 109–11.

35.

Hammersley

(1989) The Dilemma of Qualitative Method: Herbert Blumer and the Chicago Tradition. London: Routledge.

36.

Henderson

Aginsky

B. W.

(1941) ‘A Social Science Field Laboratory’, American Sociological Review 6(1) (February): 41–4.

37.

Herskovits

M. J.

(1948) Man and His Works: The Science of Cultural Anthropology. New York: A. A. Knopf.

38.

Kaplan

(1964) The Conduct of Inquiry: Methodology for Behavioral Science. San Francisco, CA: Chandler.

39.

Kellor

(1901) Experimental Sociology: Descriptive and Analytical: Delinquents. New York: Macmillan.

40.

Kieserling

(2004) Selbstbeschreibung und Fremdbeschreibung. Beiträge zur Soziologie soziologischen Wissens [Self-description and Description by Others: Contributions to a Sociology of Sociological Knowledge]. Frankfurt am Main: Suhrkamp.

41.

Klein

(2008) ‘The Laboratory Challenge: Some Revisions of the Standard View of Early Modern Experimentation’, Isis 99(4) (December): 769–82.

42.

Knorr Cetina

(2001) ‘Laboratory Studies: Historical Perspectives’, International Encyclopedia of the Social & Behavioral Sciences. Amsterdam: Elsevier.

43.

Knorr Cetina

(1981) The Manufacture of Knowledge: An Essay on the Constructivist and Contextual Nature of Science. Oxford: Pergamon Press.

44.

Kohler

R. E.

(2002) Landscapes & Labscapes: Exploring the Lab–Field Border in Biology. Chicago, IL: University of Chicago Press.

45.

Kohler

R. E.

(2008) ‘Lab History: Reflections’, Isis 99(4): 761–8.

46.

Krohn

Weyer

(1988) ‘Gesellschaft als Labor. Die Erzeugung sozialer Risiken durch experimentelle Forschung’ [Society as Laboratory: the Production of Social Risks through Experimental Research], Soziale Welt: 349–73.

47.

Krohn

Weyer

(1994) ‘Society as a Laboratory: the Social Risks of Experimental Research’, Science and Public Policy 21(3): 173–83.

48.

Lapp

(1995) ‘The Rise and Fall of Puerto Rico as a Social Laboratory, 1945–1965’, Social Science History 19(2): 169–99.

49.

Latour

(1987) Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard University Press.

50.

Latour

(2004) ‘Scientific Objects and Legal Objectivity’,in Pottage

Mundy

(eds) Law, Anthropology, and the Constitution of the Social: Making Persons and Things. Cambridge: Cambridge University Press, pp. 73–113.

51.

Latour

(2005) Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford University Press.

52.

Latour

Woolgar

(1979) Laboratory Life: The Social Construction of Scientific Facts. London, Thousand Oaks, CA and New Delhi: Sage.

53.

Lazarsfeld

Rosenberg

eds (1955) The Language of Social Research: A Reader in the Methodology of Social Research. Glencoe, IL: Free Press.

54.

Lynch

(1985) Art and Artifact in Laboratory Science: A Study of Shop Work and Shop Talk in a Research Laboratory. London: Routledge & Kegan Paul.

55.

Mead

(1928) Coming of Age in Samoa: A Psychological Study of Primitive Youth for Western Civilisation. New York: W. Morrow.

56.

Melvin

B. L.

(1925) ‘Laboratory Work in Rural Social Problems’, Journal of Social Forces 3(2) (January): 261–3.

57.

Merton

R. K.

(1973) ‘The Normative Structure of Science’,in Merton

R. K.

(ed.) The Sociology of Science: Theoretical and Empirical Investigations. Chicago, IL: University of Chicago Press, pp. 267–78.

58.

Miller

O’Leary

(1994) ‘The Factory as Laboratory’, Science in Context 7(3): 469–96.

59.

Molseed

M. J.

(1994) ‘Naturalistic Observation in the Laboratory’, Symbolic Interaction 17(3): 239–51.

60.

Moreno

J. L.

(1937) ‘Sociometry in Relation to Other Social Sciences’, Sociometry 1(1/2) (October): 206–19.

61.

Nash

D. J.

Wolfe

A. W.

(1957) ‘The Stranger in Laboratory Culture’, American Sociological Review 22(4) (August): 400–5.

62.

Newstetter

W. I.

(1937) ‘An Experiment in the Defining and Measuring of Group Adjustment’, American Sociological Review 2(2) (April): 230–6.

63.

Nichols

(1998) ‘Social Relations undone: Disciplinary Divergence and Departmental Politics at Harvard, 1946–1970’, The American Sociologist 29(2) (June): 83–107.

64.

O’Rourke

J. F.

(1963) ‘Field and Laboratory: the Decision-Making Behavior of Family Groups in Two Experimental Conditions’, Sociometry 26(4) (December): 422–35.

65.

Potthast

(1998) “Sollen wir mal ein Hochhaus bauen?’ Das Architekturbüro als Labor der Stadt [“Shall We build a Skyscraper?’ The Architectural Office as Laboratory of the City]. Berlin: Wissenschaftszentrum Berlin; available at: bibliothek.wzb.eu/pdf/1998/ii98-502.pdf.

66.

Rabinow

(2007) ‘Steps towards an Anthropological Laboratory’,available at: http://anthropos-lab.net/documents/conceptual_exchanges/.

67.

Schutz

(1965) ‘The Social World and the Theory of Social Action’,in Braybrooke

(ed.) Philosophical Problems of the Social Sciences. New York: Macmillan, pp. 53–67.

68.

Seymour-Smith

ed. (1986) Macmillan Dictionary of Anthropology. New York: Macmillan.

69.

Sherif

(1954) ‘Integrating Field Work and Laboratory in Small Group Research’, American Sociological Review 19(6) (December): 759–71.

70.

Stocking

G. W.

Jr (1982) ‘The Santa Fe Style in American Anthropology: Regional Interest, Academic Initiative, and Philanthropic Policy in the First Two Decades of the Laboratory of Anthropology, Inc.’, Journal of the History of the Behavioral Sciences 18(1): 3–19.

71.

Sutherland

E. H.

(1931) ‘The Prison as a Criminological Laboratory’, Annals of the American Academy of Political and Social Science 157 (September): 131–6.

72.

Vincent

G. E.

(1905) ‘A Laboratory Experiment in Journalism’, American Journal of Sociology 11(3): 297–311.

73.

Weatherly

U. G.

(1923) ‘The West Indies as a Sociological Laboratory’, The American Journal of Sociology 29(3) (November): 290–304.

74.

Wright

(2002) ‘The Atlanta Sociological Laboratory 1896–1924: a Historical Account of the First American School of Sociology’, Western Journal of Black Studies 26(3): 165–74.

75.

Zelditch

(1962) ‘Some Methodological Problems of Field Studies’, The American Journal of Sociology 67(5) (March): 566–76.