Myths of the Chicago School

3. Chicagoans as ethnographers

3.5 Participant observation at Chicago

3.5.1 The Hobo as a participant observation study

3.5.2 The Taxi-Dance Hall as a participant observation study

3.5.3 The Gold Coast and the Slum as a participant observation study

There is a widespread assumption that 'Chicago sociology' was not only predominantly ethnographic and hostile to quantification but that its predominant methodic practice was participant observation. For example, Cavan (1983) suggested that besides case study and statistics there was another method that was the result of Park's interest and influence and that was 'observation, participant and otherwise'. This method, Cavan claimed, was neither formalised nor named at the time but was the basis for a number of studies including Anderson (1923), Thrasher (1927), Zorbaugh (1929), Cressey (1932), Young, P., (1932) and Reckless (1933). However, it was left to Becker, she suggested, to give participant observation more formal shape (in the 1950s). Zorbaugh's work, in particular, has been taken as an exemplar of the participant observation approach of the 'Chicago School' (Madge, 1963; Hunter, 1983).
 
However, an examination of textbooks, methodological writings and substantive works suggests that the view of the 'Chicago School' as participant observers is misleading (Harvey, 1981; Platt, 1982). The early use of participant observation was not always conscious, was more related to case study and does not fit current conceptualisations of the method.
 Nonetheless, there are some grounds for arguing that participant observation in the loosest sense is evident in the work of the Chicagoans from the beginnings of their empirical endeavours. MacLean (1910), for example, engaged in a series of short-term participant observer experiences, as an employee in various industries, in her survey of women in the labour force and this approach was representative of research carried out in Chicago in its very early days (Fish, 1981).
 
In his review of the early period of the 'school', Park (1939, p. 3) suggested that around 1910 there was a boom in graduate students going into the social sciences but that apart from 'applied social science' courses with their atheoretical orientations, there was

no special provision for students who wanted to study a living society and no opportunity ... to study social problems in the field, or so to speak, "on the hoof". Chicago University provided that possibility, following Thomas' innovative development of the tradition he 'inherited' from Small, Henderson and Abbott. (Park, 1939, p. 3)

However this did not mean that the second decade of the century saw a blossoming of a relatively well-established participant observation study tradition. The study of 'social problems on the hoof' is indicative of the growing concern to establish an empirically based social science. At its most radical this implied getting out and seeing what is going on. But such observation rarely involved any degree of active participation. Of the forty two theses surveyed in detail, only two (5%) utilised complete participant observation (both after 1940), six (14%) used some kind of partial participant observation [3] while another seven (17%) employed casual observation, relied on past personal involvement or used the report backs of other observers. Nearly two thirds (64%) made no use of observation as a technique at all (see Appendix 6).
 
Even the 'golden era' of the 1920s was not a period in which participant observation blossomed. Cavan's (1983) suggestion, for example, that Thrasher's (1926) The Gang was a participant observation study is dubious. On the face of it, a thesis entitled The Gang would seem perfect for such a method, but the subtitle A Study of 1,313 Gangs In Chicago belies this assumption. Nobody could observe so many gangs as a participant. The study was not so much an insider account of gang life as a sociology of 'the gang as a type of human group' (Thrasher, 1927). As Park (1927) put it in the editor's preface:

The title of this book does not quite describe it. It is a study of the gang, to be sure, but it is at the same time a study of "gangland"; that is to say, a study of the gang and its habitat, and in this case the habitat is a city slum.

The book deals with the natural history of the gang, including: an attempt at typification; life in the gang; organisation and control in the gang; and the gang problem. Despite being hailed as 'an advancement in the general-survey and case study method' and as 'superior to earlier studies of the gang in that its conclusions grow out of concrete material' (Rice, 1931), Thrasher obtained most of his information through interviews. He interviewed about one hundred and thirty people consisting of sixty-one gang members, a large number of social workers who made available case material, a dozen policemen, half a dozen politicians and a number of others such as lawyers and club owners. The material gathered in this way was used qualitatively rather than quantitatively and was augmented by the written life histories of twenty-one of the gang boys, newspaper reports and a mere ten of Thrasher's own observations (Madge, 1963). However, Thrasher was very concerned about the accuracy of his work and its objective scientific status.

Such formulations as are presented, however, must be regarded as tentative hypotheses rather than as scientific generalizations. Certain of the suggestions made here may prove fruitful in dealing with the practical problems which the gang forments [sic], but the investigation has probably raised more questions than it has answered. Too great precision, furthermore, must not be claimed for the materials collected, although every effort has been made to render them accurate. The study is primarily an exploratory survey designed to reveal behavior-trends and to present a general picture of life in an area little understood by the average citizen. (Thrasher, 1927, author's preface)

Reckless' study of vice, also cited by Cavan as a participant observation study, also lacked participation to any degree. His dissertation, presented in 1925, had to rely primarily on official statistical sources. In the preparation of his book, published in 1933, he was able to augment this original study with case material made available from social workers. The study was directed to providing a picture of the location and degree of concentration of prostitution in Chicago and to assessing whether there was any correlation between vice and demographic features of Chicago as revealed by census material.
 
The difference between the observational approach adopted by the Chicagoans in the 1920s and that of more recent ethnographers is further illustrated by examining, in depth, three works supposedly indicative of Chicago participant observation studies and with much stronger claims to the method than Thrasher's or Reckless' research.
 
The core features of participant observation, as discussed above, are direct observation through a participating role, which enables the organisational and symbolic processes of a group under study to be scrutinised in order to assess the meanings in use that define the subjects’ perspective of their social milieu.

3.5.1 The Hobo as a participant observation study

The Hobo, by Nels Anderson (1923) is often cited as an early example of Chicago ethnographic work. Usually it is seen as representing the beginning of the published participant observation studies and being the forerunner of the kind of work undertaken by Becker and others in the 1950s and 1960s. However, there are considerable differences between the later participant observation studies and Anderson's work, to the extent that, using the current meaning of the term, The Hobo was not a participant observation study at all.
 
Anderson did not live as a hobo but rather stayed in a hobo hotel in hobohemia, which is counter to the taken-for-granted view even of people who were at Chicago at the same time as Anderson. In recalling the hobo research, Johnson, for example, implied that Anderson collected information as a 'complete participant observer'.

I shared a room when I first got to Chicago with Nels Anderson. He was the hobo. He had hoboed all over the country. And he hoboed from Utah to Chicago. And he spent the previous night, before I first met him, sleeping under a concrete ledge around the smoke stack of the university power plant... He said he was used to that kind of thing because he was always sleeping outdoors or under a railway bridge with a bunch of hobos, warming up food in a tin can, and all that. Anderson did his dissertation on the hobo, he wanted to supplement his information with some surveys of hobos down in what you could call hobohemia, down at South Halstead Street. So several times on weekends he would go down there and I would go with him. We got a room in one of those cheap rooming places, almost like a flop house. We'd spend a couple of days just going around talking to people. And he established rapport very quickly with them because he was a very folksy kind of man. Maybe, Friday or Saturday night we would visit, so-called, 'Hobo College'. On one occasion he was asked to say something and he got up and gave a very nice little talk. (Johnson, 1972)

Anderson had, apparently, 'jumped a freight train' from Utah in order to get to Chicago and had encountered hobos. He may also have adopted this form of travel on other occasions but in no way, as he pointed out himself, could he be said to have taken on the role of a hobo for research purposes (Anderson, 1983). Before starting his study he had worked in a 'Home for Incurables' in Chicago, which had put him in contact with hobos and it was through this contact that he developed his research. His approach was not participant observation of hobo life; rather it was observation of hobos in an institutional setting with a heavy reliance on informal, in-depth conversations with residents. Like many of Park's later students, Anderson researched an area to which he already had access.
 
Neither Anderson, nor any commentators at the time referred to his work as involving participant observation. Mention of his work in the Bulletin of the Society for Social Research made no mention of participant observation. Nor did his style of work reflect the concerns of participant observation practitioners of more recent decades except that, somewhat against the tenor of the times, he was sympathetic to his subject. This made the 'insider' report that he wrote, based primarily on his discussions with hobos encountered at work and in the hotel, appear to be a dispassionate, 'scientific' document. This appearance was amplified by the study being clearly at variance with what the University, at the time, saw as appropriate fields of study for its graduate students. Anderson (1983) suggested that his lack of moral stance appealed to Park and Burgess who made the decision to publish the work. They saw his study as scientific despite Anderson's lack of any substantial sociological background or use of sociological concepts. Indeed, Anderson suggested that the book was scientific without him having to work on it, simply because his background and approach was unlike that of the predominant 'clergy' at Chicago [4]. Its candid abandonment of conventional ameliorative wisdom seems to have been its main appeal for Park and Burgess.
 
The book simply stated what hobos did, what types of hobos there were, how they lived in both cities and rural areas, how they were seen by other non-hobo sectors of the community and concluded with an assessment of why hobos were disappearing, which concentrated on the lack of demand for migrant labour. In short The Hobo was a detailed descriptive account that analysed the usefulness of the hobo and their increasingly rapid disappearance. It lacked sociological 'pretentiousness' and was primarily a report of the state of affairs that challenged some of the taken-for-granted notions about hobos.
 
Working in the home had been an important factor in Anderson attracting the research funds but Anderson's decision to research the hobo was not one developed through any direct affiliation to the University of Chicago. It arose as the consequence of discussions with a Dr. Reitman who was interested in the subject of the hobo and raised the money for a study from his friend Dr. W.A. Evans and placed the money with the United Charities of Chicago. The director of the United Charities was Joel Hunter who became the treasurer of the hobo study committee to whom Anderson was responsible. The other two members of the committee were Reitman, the 'authority on the area and its inhabitants' and E.W. Burgess who, as 'scientific advisor' and chairman provided the link with the Sociology department at the University. Notably (and rarely, see also Blumethal, 1932a) Anderson's study was published as a book by the University of Chicago Press before he was awarded the M.A. for the work (1925).
 
As the study was based on his work in the home and the contacts this generated rather than as the result of any direct participation as a travelling and working hobo, there is very little grounds for according The Hobo the status of an early representative of a 'Chicago School' participant observation studies tradition.

3.5.2 The Taxi-Dance Hall as a participant observation study

The Taxi-Dance Hall (Cressey, 1929) is sometimes seen as a clear early attempt to develop participant observation at Chicago. It is seen, in particular, as an early example of participant observation of a deviant situation. The study has a strong claim in this respect because it involved a decision to engage in unobtrusive participation. Nonetheless, it is a relatively isolated study of this kind. Furthermore, the term participant observation is never used by Cressey nor did the Bulletin of the Society for Social Research refer to it as a participant observation study. Moreover, the extent to which the research represented even an embryonic participant observation study, in the sense explored above, is at least debatable.
 
The sensitive nature of Cressey's enquiry required subterfuge; but Cressey's comments show that the data came from many more sources than secret participant observation.

Most of the data upon which this study is based was secured from the case records of social agencies, notably the Juvenile Protective Association, and from the reports of observers and investigators. Published material upon such a new phenomenon as the taxi-dance hall was found to be scanty and of little value; and formal interviews were abandoned as unsatisfactory. (Cressey, 1932, preface)

Cressey points out that co-operation was not forthcoming and so the decision was made to carry out the study without the co-operation of proprietors and in spite of the deliberate opposition of some of them. This lack of co-operation made it logistically impossible to secure what otherwise would have been desirable statistical data. The research was forced to adopt other approaches. Nonetheless, Cressey assures the reader that the 'considerable amount of case material which has been amassed' over five years (of which only a small amount is included in the text) 'afford a reasonable basis for the validity of the generalizations made'.
 
The observational method finally adopted was outlined by Cressey as follows:

Observers were sent into the taxi-dance halls. They were instructed to mingle with the others and to become as much a part of this social world as ethically possible. They were asked to observe and keep as accurate a record as possible of the behavior and conversations of those met in the establishments. Each observer was selected because of his past experience, his training and his special abilities. These investigators made it possible to gather significant case material from a much more varied group of patrons and taxi-dancers than could have been secured by any one person. The investigators functioned as anonymous strangers and casual acquaintances. They were thus able to obtain this material without encountering the inhibitions and resistance usually met in formal interviews. Further, the independent reports from different observers upon their contacts with the same individual made possible a check upon the consistency of the documents obtained. Moreover, this information concerning patrons and taxi-dancers made it feasible to secure much ancillary data from the records of social agencies. (Cressey, 1932, preface)

This research clearly does involve the collection of ethno-graphic material through secret participant observation type approaches and, in the same mould as Anderson's study of The Hobo, the point of view of the participants appears to be taken into account. Indeed, Cressey discussed the observational element of his research, in an unpublished paper (Cressey, 1983), exploring the role of the stranger in interactive situations (Bulmer, 1983a).
 
However, this methodic orientation does not transform the research into a participant observation study in the current sense because Cressey's observers were more in accord with Lindeman's 'objective observers' than participant observers (Madge, 1963, p. 119). Despite appearances, the study differs from later participant observation studies on four important counts. First, the participant observation material was support data for case records. Second, it was regarded as somewhat suspect as research material and needed a great deal of cross-verification. Third, ironically, a preconceived moral position that regarded the halls as 'unwholesome' (if temporarily unavoidable) underpinned the research. Fourth, while the requirements of patrons were considered, no attempt was made to engage the perspective of the female taxi-dancers. They were talked to but only in order to provide classificatory schemes, to assess why they adopted the profession and so on; but there was no attempt to explore the dancers' points of view.

3.5.3 The Gold Coast and the Slum as a participant observation study

A similar conclusion emerges from the study of Zorbaugh (1929) The Gold Coast and the Slum, the third 'exemplar' of early 'Chicago School' participant observation studies, (Madge, 1963; Hunter, 1983). This study was of the extremely diverse area of Chicago known as the Near North Side, which lies just north of the central business district (the Loop). The area, one-and-a-half miles long by one mile wide, is on the shore of Lake Michigan and, in the 1920s, its economic prosperity and consequent social standing declined rapidly as one moved inland from the shore. Lake Shore Drive, known as the Gold Coast, was a highly desirable residential area. Backing on to this was an area around Clark Street that had become a rooming house district and beyond that was 'Little Italy', a slum area that had gone through various transitions but was, at the time of Zorbaugh's study, primarily a Sicilian enclave. Incongruously, the 'bohemian' area, known as Towertown, lay in the middle of the slum area. This was, according to Zorbaugh, a rather second-rate community of artists. The Near North Side was thus an area of extremes.
 
The study was primarily concerned to provide a detailed description of the complexities of the Near North Side and, to that end, was broken down into an examination of each of the four sections separately. A considerable amount of the empirical evidence consisted of demographic data and ecological analysis. The types of shops on North Clark Street, for example, were used as indicative of the area. The high concentration of cheap lunch rooms and restaurants was related to the rooming house district where residents had little or no opportunity to prepare meals for themselves.
 
The latter part of the study included an historical case study of the Lower North Community Council, which ultimately failed, and which Zorbaugh used to illustrate his conclusions about the inadequacies in local community institutions throughout the area.
 
Although Zorbaugh made himself familiar with the area under investigation, he did not undertake the kind of participant observation of a community that was to be attempted later by Blumenthal, Warner and the Lynds. The source data for his study came principally from documents provided by residents, from life histories collected, presumably, by the author, and from case histories, particularly those in the files of the United Charities. For example, when discussing the Gold Coast, Zorbaugh relied heavily on fourteen anonymous written contributions. The life histories included one provided by a pawnbroker and another by a 'charity girl'. However, it is not clear whether they were written by the contributors or compiled by the researcher on the basis of extensive interviewing. Further evidence came from a large number of essays written by school children; from a school census probably conducted on behalf of the board of education; from comments to census workers; from key informants; from personal documents such as letters; from existing records, such as the 'Illinois Lodging House Register'; and from the records of the Juvenile Protective Association. A survey of the rooming houses was also undertaken, but the information gathered from this appears to be little used, and there is no clear indication of the kinds of questions asked, the sampling procedure or the number of respondents involved. Very little of the evidence presented in the study seems to have been direct observation by the author.

While a descriptive ethnographic account, it was preoccupied with an ecological analysis rather than the perspective of the subjects. Throughout, there is a taken-for-granted view that the area was 'disorganized' and that it represented the antithesis of what a community ought to be. However, Zorbaugh was at pains to point out that he was not comparing the area with an idealised community; rather, taking on Park's approach, he argued that it was necessary to accept the cultural traditions of the community as they are and attempt to understand them. To that end, it was necessary to discover the nature of the community, how it operated and the impact of industrialisation upon it. Nonetheless, Zorbaugh was far more concerned with a generalised descriptive account than any insider attempt to unravel the perceptions of the subject groups. There is nothing in the methodology of the study to suggest that Zorbaugh was a part of the area. Thus, in current terms, it does not constitute a participant observation study.

Notes

[ 3] Complete participant observation refers to those instances where the researcher takes on the role of the group under observation and joins in on a more-or-less full-time basis. Partial participant observation refers to those situations where the researcher merely engages as a participant observer on a part-time or convenience basis. Return

[ 4] Anderson was not an atheist or agnostic and he suggested that he had a two-year struggle relating his social scientific and notably Darwinian evolutionary views to his Mormon fundamentalist background. This reflects similar views of graduates of the 1920s, as seen in chapter two. Return

Next 3.6 Participant observation and community studies