Le Play House and the Regional Survey Movement in British Sociology 1920 - 1955 | CONTENTS | Introduction | Theory | Origins | Early C 20th | Le Play House | IoS | WW2 | Postwar

Theory and Methodology

"We are so astonished that bards long dead should have modern ideas that we marvel if in what we believe to be an ancient Gaelic epic we come across one which we should have thought as most ingenious in a contemporary. A translator of talent has only to add to an ancient writer whom he is reconstructing more or less faithfully a few passages which, signed with a contemporary name and published separately, would seem to be merely agreeable; at once he imparts a moving grandeur to his poet, who is thus made to play upon the keyboards of several ages at once. The translator was capable only of a mediocre book, if that book had been published as his original work. Offered as a translation, it seems a masterpiece. The past is not fugitive, it stays put ....."
(Marcel Proust, "The Guermantes Way" Vol. 3 of À la Recherche du temps perdu, 1920)

The Problem of Methods

Any discipline which focuses excessive attention on itself can justifiably be regarded as somewhat self-indulgent. It is however a luxury readily available to only a few disciplines, notably in the areas of the arts and the social sciences, where the theories and methods used for dealing with the more obvious subjects of enquiry can equally well be turned upon the practitioners of the discipline itself and their activities and writings. Sociologists perhaps more than historians, geographers, economists or other social scientists tend to be more intensely interested in the nature and development of their own discipline, as an aspect of the sociology of knowledge. Sociologists are also interested in the nature and development of other disciplines, and the interest is reciprocated to a certain extent, for as Puttman observes, although routine workers in the physical sciences are on the whole not particularly interested in social and philosophical questions, those at the 'frontiers of knowledge' are more aware of the fragility and fallibility of their enterprise and are consequently more interested in epistemology. (Magee 1978).

However, despite the interest most sociologists show in the nature and development of their own discipline, and the fact that sociologists are often criticised for devoting too much attention to abstract theory rather than 'the real world', literature on 'how to do' the history or sociology of sociology remains sparse. A sociologist embarking on a 'conventional' piece of empirical work, using survey techniques, or participant observation is well provided for with technical manuals explaining what is to be done at every step of the way. This is not so for the kind of techniques necessary to investigate the history of sociology. Inevitably much of the work tends to be with documents rather than with people, and where it does involve contact with people, the standard techniques of social research such as the interview are not immediately appropriate (Platt 1981b). As Platt observes elsewhere, "'Documentary research' is not a clear-cut and well-recognised category, like survey research or participant observation", and she goes on to note that well under half of a selection of texts on sociological method make any mention of documentary research at all. (Platt 1981a) Even those texts which do devote themselves extensively to documentary research (Webb 1966: Plummer 1983) tend to be less than wholly relevant in that they still work within a model of the relationship between the researcher and the subjects of the research, which assumes that the former is in a higher social and institutional position to the latter, and is making use of materials which themselves were not constructed with sociology in mind. The situation is different for research into the lives and activities of those who were themselves part of the discipline. Documents relevant to the history of sociology are themselves part of sociology and have helped to construct the discipline which is in turn examining them.

Work on the institutional history of sociology is somewhat easier, especially in the USA where detailed course lists and other university records from the late 19th century onwards provide a ready made data set for subsequent research. The proliferation of institutions offering courses in sociology from an early stage, despite the conflicting notions of what the discipline was about provide much that can be used (Sica 1983; Bulmer 1981b; Martindale 1980). Such work is less fruitful in Britain, because of the much smaller number of institutions offering sociology in any shape or form and because of the less systematic way in which courses are described and recorded (The same problem exists for historical research in many other countries, such as Germany (Kasler 1980)). Very few people have attempted such research in Britain (Fincham (1975): Heath and Edmonson (1981)). The methodology most frequently used for institutional research in sociology, which relies on the identification of 'schools' and the personal and professional connections between individuals in different institutions which produce them (Mullins 1973; Tiryakian 1979) is also inappropriate for investigating sociology other than in university departments, or where a large enough number of such departments exists to form any kind of comparison.

Notwithstanding these limitations however, the problems faced by a historian of sociology are the same as those faced by a sociologist undertaking any research project; the identification of a perspective (and the corresponding evaluation and rejection of alternative perspectives), and the selection of a practical method (or methods) for carrying out the research. In practice the processes tend to be simultaneous, in that only the act of doing the research reveals the appropriateness or otherwise of a particular perspective or method. In any case, in dealing with the history of sociology, the choice is strictly limited.

Approaches to the History of Sociology

It has already been remarked how dependent sociology has become upon the consideration of its own history, when compared with other disciplines (Peel 1978:347). However, sociologists differ in the ways they interpret their history, and some consideration of the different theoretical approaches which sociologists and others interested in the history of the social or human sciences take to the area is necessary. Here I will be considering the theoretical or epistemological problems of studying the history of sociology; the following section will cover the empirical or practical problems.

The identification of any clear 'pattern' to sociology, or the recognition that sociology was an international discipline with a history that could be examined in a cross-cultural context can be said to date from Parsons' creation of the idea of 'classical sociology' (Parsons 1937), through his attempt to integrate the work of Weber and Durkheim among others ("before 'The Structure of Social Action', the sprawling academic activities called sociology had no clear and generally recognised genealogy" (Therborn 1980), although Giddens among others has criticised the way in which such 'myths' are created in the history of sociology. (Giddens 1976). (It might be noted in the context of a history of British Sociology that Hobhouse was a major influence on the young Parsons. His name appears in Parsons' early lectures as a primary influence, yet is entirely absent from "The Structure of Social Action" when it eventually appears in 1937).

However, books purporting to be histories of sociology had appeared earlier, the first probably being Bristol's "Social Adaption" (1915) (Hinkle 1978), Lichtenbergers' "The Development of Social Theory" (1923) and Sorokin's "Contemporary Sociological Theories" (1928). Timasheff (1955) identifies three different ways of dealing with the history of social theory in early works - through a consideration of different schools, through studying the development in historical (and by implication international) sequence, and in different geographical areas. There were a number of examples of the third type in existence by 1937, and many more to follow, although major compilations of different accounts in single volumes did not appear until after the Second World War (Gurvitch and Moore 1945; Barnes 1948). It is these which are often mistakenly reckoned to be the first accounts of the history of sociology.

Interest in the history of sociology as a theoretical problem rather than simply a matter of correctly charting an agreed sequence of events is comparatively recent, dating from Peel's seminal biography of Spencer (1971), influenced by the work of Quentin Skinner on the history of political theory (Skinner 1966; 1969). Several short, narrative histories of sociology also appeared during the 1960s (Maus 1962; Mitchell 1968), and Fletcher's much more substantial work (Fletcher 1971) shortly afterwards. The recognition of the history of sociology as a reputable academic specialism, rather than simply a hobby for retired sociologists can be marked by the establishment of the Journal of the History of the Behavioural Sciences (1965 - although this was more the province of psychologists initially), the International Sociological Association Research Committee on the History of Sociology (1971), the establishment of a History of Sociology Section at the American Sociological Association Annual Meeting and the 'Journal of the History of Sociology' (now simply 'History of Sociology') (1978) and the British Sociological Associations' study group on the Sociology of Sociology and Social Research (1981). Within the last few years there have been many books, articles and papers on the theory and content of the history of sociology and the area can be said to be well established on an international level, although as recently as 1977 Jones could write that, "self-conscious reflection on historical method has not been typical of historians of sociological thought". (Jones 1977:282).

Peel (1978:347) suggests that there are at least five different forms of history of sociology.

In the first place, there are "triumphalist Whig histories celebrating the general scientific achievements of the subject". The term "Whig" or "Whiggish" to describe history originates from the work of Sir Herbert Butterfield (1931) who described 'the Whig interpretation of history' as being "to emphasise certain principles of progress in the past and to produce a story which is the ratification if not the glorification of the present". He goes on to argue that "the danger in any survey of the past is lest we argue in a circle and impute lessons to history which history has never taught and historical research has never discovered - lessons which are really inferences from the particular organisation that we have given to our knowledge....Our assumptions do not matter if we are conscious that they are assumptions, but the most fallacious thing in the world is to organise our historical knowledge upon an assumption without realising what we are doing, and then to make inferences from that organisation and claim that these are the voice of history." (Butterfield|1931:24-5). Butterfield argues that historians should focus their interests on issues and ideas that mattered at the time to the individuals concerned, rather than working backwards from contemporary concerns. 'Whiggism', or 'presentism' as it is often called is the butt of much current criticism in the history of sociology (Peel 1971; Jones 1977; Collini 1979; Camic 1979; Kuklick 1980) although the critique itself has also come in for criticism (Seidman 1983), and there are serious limitations to Butterfields own approach ("it is highly unlikely that there will be substantial agreement as to 'what matters' among the writers of any historical period; and ... even if there were, it would be a tedious and unsatisfactory 'history' which merely endorsed such judgements" (Jones 1977:1137). Peel suggests that the work of Barnes (1948) and Fletcher (1971) falls into the category of 'whig history'.

Secondly he refers to "histories written to give support to a very particular contemporary position", and cites Parsons (1937), Nisbet (1967) and Harris (1971) as examples, although Nisbet's account of the origins of sociology as being a 'conservative' reaction to the radicalism of the enlightenment, and an 'art form' as much as it is a science (Nisbet 1976) is perhaps the most typical. Hawthorn's account of the development of sociology as a series of different approaches to the nature and the lessons of history (Hawthorn 1976) is another example.

Thirdly, Peel writes of "exposito-critical histories, dealing with the founding fathers whom it is judged important for the present generation to know and get right". Any number of introductory texts could be included here, but specifically (although not necessarily critically) cited are Giddens (1971) and the whole University of Chicago 'Heritage of Sociology' series. The notion of 'founding fathers', apart from the inaccuracy inherent in its sexism (Morgan 1980) is a 'whiggish' term, and one which Giddens (and others) in later work (Giddens 1976: Jones 1977) have argued against, on the grounds that it tends to construe the history of the discipline in terms of arguments between a limited circle of individuals in the past who the present has deemed to be the significant figures of their epoch, when in fact they were frequently either not aware of each others' existence, or in debate with other long-forgotten individuals considered much more important at the time. The celebrated 'mutual non-awareness' of Durkheim and Weber (Tiryakian 1966: Lukes 1973), and Durkheim's intense debates with Robertson-Smith (Jones 1977) are examples.

Fourthly, Peel refers to "intellectual history ... merely dealing with a topic of intense interest" citing Hughes (1959) and Burrows (1966).

Finally there is "the sociology of sociology" (Reynolds and Reynolds 1970), described unhesitatingly as "a farrago", "belabouring establishment sociologists in America for unjustified pretensions to value-freedom" which "has little to do with the history of the subject" (Peel1978:357). However Peel concedes that other works within this field, such as Oberschall (1972), Lazarsfeld (1962) and even Friedrichs (1970) are more worthwhile, although he remains critical of Gouldner (1971) describing it as the kind of "mythologised history" or "prophetic mode of sociology" which Giddens (1976) and (paradoxically) Friedrichs himself (1970) also condemn.

Since Peels article however, debate in the history of sociology has centred around the issue of "whiggism" and its counterpart, usually referred to as "historicism", and the two are conveniently dichotomised to 'presentism' and 'historicism' in some current papers (Seidman 1983). Jones (1977) on the interpretation of Durkheim's work on religion is the starting point for much of this debate, and this itself draws heavily on Skinner (Skinner 1966; 1969; 1972; 1974) and Dunn (1966) within philosophy, and of course Butterfield (1931) and to a certain extent Kuhn (1962). It has led to a wide ranging debate (Johnson 1978; Camic 1979; Kuklick 1980; Camic 1981; Seidman 1983) in the course of which the theoretical and methodological principles of both presentism and historicism have been subjected to scrutiny. Few historians of sociology would describe themselves as 'presentists', but there are permutations and complications inherent in the idea of 'historicism' which mean that in practice it is more complex and possibly no more satisfactory a form of explanation.

Seidman (1983) provides the most cogent summary of the 'presentist' and 'historicist' positions. 'Presentism' sees "a continuity in subject matter and problems between past and present social scientists" (Seidman 1983:80), and "subordinates past to present. The past is not intrinsically valuable, for it is only worth studying as a record of (people) striving to realise the present. History is the story of the triumph of progressive forces over reactionary ones; and the present is superior to all." (Peel 1971:259). It is more vilified (Jones 1977: Skinner 1969) than practiced, although the work of Abraham (1973) and Fletcher (1971) could be said to be recent exceptions. It assumes that the problems and doctrines of the present serve as criteria for organising, interpreting and judging past ideas, that 'the past' involves a simple division between 'false ideas' (refuted) and 'true ideas' (incorporated), and that the history of social science must disentangle 'valid scientific knowledge' from 'ideological residues'.

Its attitude is evaluative and ideological. History becomes "a pack of tricks we play upon the dead" (Skinner 1969:178) - "the history of sociology is merely quarried to provide spurious pedigrees for current claimants to sociological legitimacy" (Peel 1971:ix), leading to historically absurd interpretations - 'the present' seems always to be foreshadowed in 'the past', "a history of thoughts which no one ever actually succeeded in thinking, at a level of coherence which no one ever actually attained". (Skinner 1969:18)

In contrast, 'historicism' seeks "to understand the science of a given period in its own terms" (Stocking 1968:8). "Past, present and future are bound into an organic pattern which can only be discovered from an analysis of the past. Historicism ... leads to improved techniques for recovering the past" (Peel 1971:259). It assumes 'discontinuity' between past and present - "the conditions, and therefore the questions and answers of science vary historically" (Seidman 1983:81). It avoids the notion of 'influence' - "a story which reads like the first chapter of Chronicles, though without the genetic justification" (Skinner 1969:25), and claims that to specify the meaning of a text or set of ideas, one needs to recover the authors 'intentions'. (Jones 1981b; Skinner 1972; 1976)

In this, it assumes that author's statements of intention are adequate, whereas "in the light of psychoanalysis ... we cannot assume that an author has privileged access to his or her own intentions" (Seidman 1983:83). It tends to deny what Ricoeur and others call 'the autonomy of the text' - "What the text says matters more than what the author meant to say" (Ricoeur 1971:534). "The real meaning of a text, as it speaks to the interpreter does not depend on the contingency of the author and whom (s/he) originally wrote for". (Gadamer 1975:263). Ultimately, it leads to an illusion of 'objectivism', that a 'real' description of the past is possible.

This is not the place to indulge in a detailed epistemological examination of the nature of historical explanation. Much has been written in recent years that is critical of traditional historical methods and which is particularly relevant to the forgotten parts of the history of sociology. Suffice it to say that the present study will attempt to provide an account of the Institute of Sociology, Le Play House and the Regional Survey movement in British social research which is informed by historicism, although mindful both of the limitations of the approach and of the tendency to slip into presentism without always being aware of doing so.

The Practice of Historical Research in Sociology

The research was originally intended to take three forms

  1. A detailed investigation of the records of Le Play House and its related organisations, together with such other artefacts (letters, manuscripts and survey materials) as are deposited with the formal records in the Library of Keele University.
  2. An examination of other related records elsewhere. These would have included papers of individuals who were in any way involved with the work of Le Play House, records of organisations sharing common interests and a common membership with Le Play House, and the records of institutions (such as the London School of Economics) which were active in the development of the discipline of sociology in this country during the period in question.
  3. Interviews with people who were active in any way in the work of Le Play House, or who knew of its work in any other capacity.

In practice, only the first and the third aims were realisable. Some of the organisations whose records would have been worth the effort of consulting appeared to have no records which could be traced, and it is unlikely that a search of the papers of any of the individuals whose collections are open to public scrutiny would have revealed any information which would have added significantly to that available at Keele. In any case, as the research proceeded, the scope tended to become narrower as the amount of material available and the amount of work it required became apparent. In dealing with an area where no cataloguing of the available documents has been undertaken, where no secondary research of any kind has made use of the material, and where the detailed pattern of the history of the organisations in question has not been charted in any surviving form, a large part of the thesis has become of necessity an attempt to do these three things.

Research on the records of Le Play House was undertaken at Keele University Library during three visits of two weeks each in 1981, 1982 and 1983, and a number of shorter visits. Details of how and why the records ended up at Keele University itself forms part of the research (Section 8.1) and has been explained elsewhere (Farquharson D 1955; Mountford 1972). The records consist of, among other things

  1. Minute Books and other formal records of the Sociological Society, the Institute of Sociology and various other smaller organisations and committees which were associated with them between 1903 and the early 1950s. These are not complete - for example, no minutes of Institute of Sociology Council or Executive meetings exist for the period between 1936 and the early 1940s.
  2. Correspondence and other papers, both private and professional of, among others, Victor Branford, Patrick Geddes (although the majority of his extensive collection of papers are elsewhere), and Alexander and Dorothea Farquharson.
  3. Various manuscripts of books and articles by Victor Branford and others
  4. Survey materials (notebooks, scrapbooks, maps, lantern slides, charts and other artefacts) from surveys undertaken in Britain and Europe between 1903 and 1952. (Those undertaken under the auspices of Le Play House are listed in Appendix ?)

Most of these materials were deposited in the Archives at Keele University at two separate times. The first was after the dissolution of the Institute of Sociology (1955) when many of the formal papers were received through the efforts of W A Campbell-Stewart, then Professor of Education, and Stanley Stewart, the Librarian. The second was after the death of Dorothea Farquharson (1976), when the recipient was the University Archivist Ian Fraser; although it is likely that some things were received in the intervening period. The library of the Institute (some 10,000 volumes) was given to Keele in 1953, but Victor Branford's papers, and other related items were retained by Dorothea Farquharson after the dissolution of the Institute with a view to their being edited and published. This had been a condition of the Victor Branford Trust, set out in Branford's will, and was nominally an obligation imposed upon the Guild of St George to which the Trust was passed after dissolution of the Institute. Lewis Mumford was chosen to act as editor of the Branford manuscripts, and visited Britain in 1956 with this in view (among other things), but due to the death of Alexander Farquharson, the recent dissolution of the Institute, the sale of most of Le Play House in Ledbury and the inevitable lack of funds, the papers were not easily accessible, and most of Branford's manuscripts were eventually passed on to Keele unedited and unsorted. Some preliminary listing and sorting of his papers had been undertaken in the early 1940s (VB1), and it is likely that the papers are currently stored at Keele in approximately the same order. In any case, all those items received prior to 1976 were listed in the order in which they were unpacked, and are referred to henceforth as the 'VB Series'. It must be understood that the prefix VB does not mean that the papers concerned are exclusively those of Victor Branford, although the majority of his most important materials are included in this series. (See notes by Dorothea Farquharson in VB1)

Those items received after 1976 had not been catalogued or listed in any way at all by Keele University Library, and I therefore prepared my own listing, which again did not involve any but the most minimal sorting and reorganisation. This is referred to henceforth as the 'AF series', although like the VB series it contains a very varied mixture of papers and other artefacts.

In addition, there is a large amount of material, mainly survey records and printed papers, which is stored in the basement of Keele Hall elsewhere at the University under the charge of Stanley Stewart, the former University Librarian who arranged the transfer of the Institute's records to Keele in the early 1950s. I prepared a listing of all that seemed to be directly relevant to the history of Le Play House, concentrating especially on the survey materials. It is likely that these were sorted in the late 1940s (Hill 1983). A great many pamphlets, journals and other printed papers from organisations which Le Play House was in contact from all over the world remain uncatalogued.

There are a limited number of texts on the methodology appropriate for this kind of sociological research. Platt (1981a) suggests that there are five major problems with documentary research, based largely on her work on the history of the antique trade (Platt 1978). These are

  1. authenticity
  2. availability
  3. quantity (whether there is enough to permit sampling)
  4. reliability
  5. other inferences

The extent to which scrupulous attention need be paid to any one of these problems depends on the nature of the documents being studied and their origin. The authenticity and 'truth' of a document and its contents can normally be covered by the practice of assuming both unless other evidence suggests the contrary. (Platt quotes 'Craig's Rules of Historical Evidence' of 1600 that 'all (people) have an equal right to be believed unless the contrary has been established from elsewhere').

Sampling is only relevant for very large datasets, which require some selectivity on the part of the researcher. This is not necessary with the Archives at Keele, although the technique could have been used for a more detailed examination of the survey records, of which there are a large number within an identifiable category.

Webb et al (1966) suggest two major sources of bias in running records; selective deposit and selective survival. Both of these are more likely to affect the records of voluntary or informal organisations rather than those of governmental or official bodies. In the case of records which can be defined as 'political', any evidence of selective deposit or survival should be viewed critically, since selective editing of records may be a deliberate administrative practice - 'particularly in the political area, the holes that exist in data series are suspect'. On the other hand, records, even the most important are often destroyed or mislaid casually, particularly in the course of office moves, and in that the 'holes' in the data series of the Institute of Sociology are proximate to two office moves (1940 and 1946) the former of which was made necessary by wartime conditions, there is unlikely to be any ulterior motive behind the loss. In any case, problems created by the selective or erratic survival of documentary evidence are 'solvable', the authors suggest, by 'consensus test', which involves searching for agreement (or discrepancy) from different contemporary sources - the absence of any noticeable evidence which suggests otherwise enables the rational assumption that things continued 'as before'. Although there were periods of fairly intense disagreement and even personal animosity during the thirty five year history of Le Play House (particularly during the split which led to the creation of the Le Play Society in the early 1930s (See Section 6.3) and during the discussions about the dissolution of the Institute in the early 1950s (See Section 8.3), there is ample supporting evidence, largely in the form of correspondence, to the effect that the formal records of the Institute of Sociology during these periods are an accurate, if not entirely complete record of what happened.

A second type of record in documentary research is what Webb et al (1966) call the 'episodic or private record', and which for the most part consists of letters. Plummer (1983), basing his evidence on Thomas and Znanecki's monumental study of the correspondence between Poland and the USA in the early twentieth Century (1920) suggests that there are five types of letter - ceremonial, informing, sentimental, literary and business, and examples of each type occur in the Le Play House archives.

The significant difference between the way in which the current research has made use of letters, and the relevance of letters to research such as that of Thomas and Znanecki is the same as that referred to above - namely, that the content of the letters is useful as factual evidence relating to the substance of the research, rather than simply as 'data' for subsequent analysis. In all correspondence however, what Webb et al (1966) refer to as the 'dross rate' is very high !

Documentary research in sociology may take several forms, and even though there is relatively little discussion of the technicalities of such research, there are many similarities with other research methods. As Platt observes, "only the most limited technical problems of documentary research are wholly specific to it." (Platt 1981a) However, she goes on to note that "the combination of characteristics which it can possess is sufficiently different from those more typically written about by sociological methodologists for differences of emphasis to amount almost to qualitative distinctiveness." The specific problems of documentary research which she draws attention to are

  1. Documents are studied in contexts drawn entirely from other documents. This would not be so in all research using documents, but is almost inevitable in historical research - although the definition of 'document' can obviously be a broad one.
  2. One does not have all the documents that are relevant. One can never know either, particularly in dealing with 'informal' organisations exactly (or even approximately) how many documents are missing, in that letters, reports, and committee minutes are not always produced regularly and sometimes never appear at all.
  3. Not all those involved have produced documents. Many people who 'should' have written letters do not appear to have done so.
  4. One wants to make inferences about things to which documents do not refer.

One result of this is that documentary research will inevitably involve a great deal more speculation and inference than other contemporary forms of research, and be less amenable to the possibility of testing hypotheses, because of the general lack of reliable, consistent, independent data to do it with. This is a problem that historians have long been familiar with, and have accepted as an inevitable condition of their work - it is only sociologists delving into what is for many of them a new area who have felt the need to 'problematise' it.

The present work is therefore an attempt both to offer a crude map of a largely uncharted land, and to suggest some reasons why previous travellers have seen fit to bypass it altogether. One conclusion might well be that they did well to avoid it - although there is a sense in which anything that has been almost studiously ignored must at the very least, in Michelin's words, 'merite un detour'. At the risk of producing a somewhat mixed metaphor, one could quote Bernard Crick who suggested that Collini's seminal biography of Hobhouse used a "vast net with a fine mesh to catch a middling plump fish which he wishes neither to eat nor to preserve." (1979) Sometimes the mere existence of something is justification enough for simple curiosity.

Le Play House and the Regional Survey Movement in British Sociology 1920 - 1955 | CONTENTS | Introduction | Theory | Origins | Early C 20th | Le Play House | IoS | WW2 | Postwar