W. E. ARMSTRONG AND SOCIAL ANTHROPOLOGY AT CAMBRIDGE 1922-1926
James Urry
Victoria University of Wellington
W. E. Armstrong was the first person to hold an official post, if temporary, in social anthropology at Cambridge where he succeeded his teacher, W. H. R. Rivers, after the latter's death in 1922. The unpublished lecture notes of Armstrong are examined to reconstruct his vision of social anthropology, particularly his concern with psychology, social organisation and diffusionism. Armstrong's ideas are placed in their historical context and compared with those of his famous contemporaries, Bronislaw Malinowski and A. R. Radcliffe-Brown. Armstrong's influence on his students and the circumstances which led him to abandon a career in anthropology are briefly discussed.
Most British anthropologists who secured academic posts after the second world war recognised that the study of primitive society was their major concern and thus preferred to call themselves 'social anthropologists'.1 But A. C. Haddon, who belonged to an earlier generation of anthropologists, and who had been largely instrumental in the establishment of academic anthropology in Britain and the British Empire, preferred the term 'ethnology' to 'anthropology' (Quiggin 1942:10). The change in title reflects a fundamental shift in British anthropology between the two world wars, and the ultimate triumph of social anthropology over a more holistic view of anthropology. This shift has been discussed in many recent histories of the discipline. Kuper (1973; 1983) and Goody (1973) have both viewed Bronislaw Malinowski and A. R. Radcliffe-Brown as the founding-fathers of British social anthropology. More recently Langham (1981) has argued that the influence of W. H. R. Rivers, more specifically his study of kinship systems, should be seen as the central paradigm from which modern social anthropology developed (see also Slobodin 1978 on Rivers's importance).
The origins of British social anthropology, however, should not be sought either in the work of individual scholars, or in such a narrow and technical field as the study of kinship. The crucial period for the emergence of British social anthropology lies in the years 1900 to 1914; its consolidation with refinement in the inter-war period (see Stocking 1984). In the first period a new generation of anthropologists emerged, mostly trained in the natural sciences, who proposed new methods of inquiry (Urry 1972; 1984a; 19846) and new theoretical issues, including the study of primitive society.
Man (n.s.) 20, 412-433
JAMES URRY
413
During this first period Rivers was undoubtedly the major figure concerned with the study of primitive society. Through his writings and teaching a new generation of anthropologists took up his interests, the most famous being Radcliffe-Brown. Another student, less well-known than Radcliffe-Brown, was Wallace Edwin Armstrong. After Rivers's sudden and unexpected death in 1922, Armstrong succeeded Rivers in teaching social anthropology at Cambridge. Although Langham briefly discusses Armstrong, neither Kuper nor Goody mentions his name, perhaps because he taught anthropology for only a brief period and subsequently pursued a career in another subject. More surprisingly, however, Meyer Fortes (1953) in his Cambridge inaugural address on the development of social anthropology at Cambridge, fails to mention Armstrong. In fact, Armstrong was the first holder of a lectureship in social anthropology at Cambridge. In 1981 the National Library of Australia acquired Armstrong's anthropological papers, which include his Cambridge lecture notes (see Urry 1983). The aim of this article is to examine these lecture notes, more particularly his lectures on social anthropology, and relate them to the changes which occurred in British anthropology during the 1920's.
Armstrong never published a general account of his vision of social anthropology, though the fact that his lecture notes were typed-up in full may indicate that he intended to do so. Whatever his intentions, these texts present opinions he never expressed in print, and when placed in their historical context throw an interesting light on the period when social anthropology was being refined and consolidated after the first world war, and following the death of Rivers.
Armstrong's anthropological career
W. E. Armstrong was born in England in 1892 and won an exhibition to Sidney Sussex College, Cambridge, before the first world war.2 At the outbreak of war Armstrong joined the Royal Army Medical Corps, but was wounded in action in 1915 and subsequently lost a leg. He returned to Cambridge to complete his degree; now a confirmed pacifist he joined the Union of Democratic Control, the major peace movement of which Bertrand Russell was also a member (Vcllacott 1980:19).3 Armstrong completed his degree in the Moral Sciences Tripos in 1918.
In his final year at Cambridge, Armstrong concentrated on psychology and was introduced to Rivers who fired him with an interest in anthropology. After completing his degree Armstrong studied anthropology under Haddon (Armstrong in Langham 1981:205), and in 1919 was awarded the Anthony Wilkin Scholarship to carry out field research in Papua New Guinea. Armstrong first worked in the Suau district of what was then South-Eastern Papua, and early in 1921 was engaged by the Papuan government to collect further ethnographic material in this region (Strong 1922: 29; Armstrong 1922). He was later appointed Assistant Anthropologist to the government and spent two months on Rossel (Yela) Island in the far east of South-Eastern Papua.
Rossel Island was little known, but had acquired an infamous reputation after a French ship carrying over three hundred Chinese to Australia was wrecked on
414
JAMES URRY
its coasts in 1858. Nearly all the survivors were killed and eaten by the islanders (Armstrong 1928a: Appendix 1). Anthropologically, the island is of considerable interest as its people speak a non-Austronesian language in contrast to the mostly Austronesian-speaking inhabitants of the Massim area to the west.4 Armstrong was probably attracted to the island on account of its ethnological significance, but his initial intention had been to study the islanders' kinship system (Armstrong 1928a: vii). In the course of his research he discovered a unique 'monetary' system and concentrated on this aspect of the island's culture. The majority of Armstrong's ethnographic writings are concerned with Rossel and most deal with the monetary system (Armstrong 1923/24; 1924a; 19246). His Suau work appears never to have been finally written up.
The position of Assistant Anthropologist was only temporary and, though offered the post of Government Anthropologist, Armstrong declined and returned to Cambridge in 1922.5 After Rivers's sudden death in June 1922, Armstrong's name was put forward to give instruction in social anthropology, and he was offered and accepted the post.6 It was stressed that the new post was temporary, as Rivers had no official position in anthropology and had given his lectures gratuitously. The following year, however, the post was renewed and Armstrong was asked also to take over Haddon's lectures on the Religion of Backward Peoples.7 Armstrong's position remained precarious, though the post was renewed in 1924 and 1925. In 1926 Haddon retired as Reader in Ethnology and was replaced by an ex-Indian Civil Service Officer, T. C. Hodson, who had experience of ethnographic research in India, had published widely and had lectured on anthropology at British and American universities.8
As early as 1923 Haddon had made representations to the University to establish his Readership (Leach 1984: 5) and secure more posts in anthropology. In 1925 a Memorial was sent to the Appointments Committee of the Council of the Senate of the University to establish three posts as recognised lectureships in the University. The three posts were to be in cultural anthropology (Camilla Wedgwood), prehistory (M. C. Burkitt) and social anthropology (Armstrong). It was not an opportune time to make such a request as a Royal Commission was investigating the University and posts were being frozen or revised. In the end only Burkitt was appointed and the other posts turned down, but money was again made available for the teaching of social anthropology. At this point Hodson requested that the money be used for teaching only 'anthropology' as he intended himself to lecture on social anthropology.9
Armstrong gave instruction to Tripos students in the Ethnology of the Special Area (Melanesia) in 1926-27 and offered lectures and supervision until 1931 (Leach 1984: 11) but his career in anthropology had effectively ended. During the early 1920's Armstrong had become interested in economics and from 1926 to 1939 he acted as a supervisor and occasional lecturer in economics at Cambridge. In 1939 he accepted a post as Lecturer in Economics at the University College, Southampton, later the University of Southampton. Eventually, after steady promotion, he became Professor of Economic Theory in the University and retired in 1961. He died in March 1980.
JAMES URRY
415
The scope of social anthropology
British anthropology in 1900 still covered a broad range of topics, physical anthropology, ethnology, folklore, material culture and archaeology, with little specialisation and with most explanations couched in evolutionary terms. During the following decade various sub-areas of anthropology were to become differentiated, with anthropologists specialising in particular topics and utilising a wider range of explanation. This change in emphasis occurred in conjunction with an increasing trend towards the professionalisation of anthropology, the teaching of the subject in universities and the emergence of a new generation of trained anthropologists. The study of primitive society was one area to receive special attention. Such studies had a long history in British anthropology, but after 1900 they received a new lease of life as anthropologists carried out direct field research and contacts were made with the emerging discipline of sociology. The term 'social anthropology' became more common; Sir James George Frazer was given the title Professor of Social Anthropology at Liverpool University in 1908, and R. R. Marett was appointed to a University Readership in Social Anthropology at Oxford in 1910. At the London School ofEconomics a number of lecture courses were given before the first world war which dealt with primitive society, most notably by Radcliffe-Brown, Malinowski and Rivers (Firth 1963).
Anthropology had been taught in Cambridge since the early 1890's and after Haddon was appointed to a Lectureship in Ethnology in 1900 he developed the teaching of physical anthropology and ethnology, although both Rivers and Radcliffe-Brown gave lectures on social organisation before 1914 (Firth 1956: 288-9; Stocking 1984a). The Anthropology Tripos was first examined in 1915 and the Student's Handbook to Cambridge for that year included an entry on 'social institutions' (Dr E. S. Leedham-Green pers. comm.) and social anthropology appears to have been taught 'intermittently' by Rivers after this date (Haddon 1923a). In 1918 Haddon noted that social anthropology was included in the anthropology offered at Cambridge under the following schedule.
Social anthropology—social organization; marriage and kinship; property and rank, descent, inheritance, and succession; age grades; totemism; caste; secret societies; government; trade and currency; slavery; customs and ceremonial of birth, childhood, adolescence, marriage and death; religion and magic; animism; cult of the dead; animal and vegetable cults; gods; mythology (Haddon 1918:261).
This schedule undoubtedly reflects Rivcrs's interests although his final lectures, delivered in 1921 shortly before his death, did not cover all the topics.10 The schedule, however, was only a guide to courses, not a firm syllabus.
When constructing his own course of lectures, Armstrong had not only the schedule to guide him, but also the text of Rivers's last lectures. Preserved within Armstrong's papers is the typed top copy of Rivers's final lectures.11 An undated, handwritten list of lectures which closely follows the published schedule probably reflects the structure of Armstrong's early lectures (table 1). A selection of rough handwritten notes, and one of two handwritten complete lectures may be attributed to this early period. The complete set of typed
416
JAMES UHRY
Table i .   An early list of lectures given by Armstrong at Cambridge possibly 1922-1923 (APNLA 2(19)).
I	Nature of Culture.	IX	Exchange. Kula. Rossel.
II	Methods. Evolutional		Exchange. Potlach.
	and Diffusionist.	X	Puberty and Secret Societies.
	Sociological (and Group Mind).	XI	Invention. Diffusion.
	Individualism.		Megalithic Culture.
III	Continuity and Discontinuity in	XII	Magico Religion. Introduction.
	Social Origin. The Family.		Examples of Sympathetic Magic.
	Clan and Tribe.	XIII	Origin of Magic.
IV	Nature of the Classificatory System.	XIV	Animatism and Animism.
V	Marriage.		Manaism
VI	Totemism. A method.	XV	Evolution of gods. Ancestor
VII	Totemism. A method.		Worship.
VIII	Exchange. Evolution of	XVI	Hero Cults. Mythology. Taufo
	money etc.		Series.
lectures (see table 2) are probably from his last years of lecturing, and reflect Armstrong's 'mature' view of the course. The lectures deviate from the public schedule, devoting more time to psychology and totemism, and saying little about religion. Some of these changes can be explained by the fact that after 1923 Armstrong took over Haddon's religion course, allowing him room to expand the social anthropology lectures.12 The 'mature' lectures may date from this period.
Table 2.   The 'mature' list of lectures given by Armstrong sometime after 1923 and before 1926
I	Definitions. Psychological Basis.	XIII	(Marriage Classes and Section Systems)*
II	The Instincts of Man.	XIV	Totemism.
III	Suggestion, Sympathy,	XV	Totemism (2).
	Imitation.	XVI	Totemism (3).
IV	Association, Habit. The	XVII	Property.
	Sentiments, Group Sentiment.	XIX	Succession. Chiefs.
V	Earliest Groupings a priori.	XX	The Men's House. Age
VI	Original Groupings (cont.)		Grades.
	Evolutionary theory.	XXI	Puberty, Voluntary Groupings.
VII	Kinship and Family. Possible origin of clan.	XXII	Fraternities and Secret
VIII			Societies.
	Genealogical method.	XXIII	Origin of Secret Societies.
IX	Tribe. Rivers' Theory of	XXIV	Migrations in Melanesia.
	Group Marriage.	XXV	Contact. Interaction.
X	Marriage.		Heliolithic Culture.
XI	Marriage (cont.)	XXVI	Heliolithic Culture (cont.)
♦ This lecture lacks a first page; it is mainly concerned with Australian Aboriginal systems and discusses the work of A. R. Radcliffe-Brown.
The published schedules do not appear to have been altered until after Armstrong had left anthropology. The 1925 schedule is almost identical with that of 1918 except the topics 'descent' and 'inheritance and succession' have been dropped, and the word 'heroes' added to that of'gods' at the end.13 After the reorganisation of the Anthropology and Archaeology Tripos in 1926,
JAMES URRY
417
however, the schedule began with the words 'Social psychology and the conception of culture' and ended with 'Medicine and science. Ethics and aesthetics in relation to custom and religion. Games and play' (Cambridge. Students' Handbook 1927-8:351). Though the latter wording may reflect Hodson's influence, the opening phrase mirrors Armstrong's lectures.
Although he may have had guidance as to the structure of his lectures, Armstrong had little to draw upon to define the nature and scope ofhis subject. Frazer's inaugural address (1913), though purporting to define social anthropology, belonged to an older vision of anthropology. Most of the newer 'professional' anthropologists before 1920, including Rivers, tended to refer to the study of primitive society simply as 'sociology' (Rivers 19141916]; 1926 etc.).
In a draft from his early lectures Armstrong ventured his own definition of social anthropology:
Social anthropology may perhaps be most conveniently regarded as [an arbitrary division in the general science of sociology] that part of the more general science of sociology which deals with primitive society (APNLA 2(12) the phrase in parenthesis has been struck out).
In this early draft Armstrong wrestled with the problem of the word 'primitive' and whether it allowed him to separate the study of 'primitive' institutions from those of'civilized peoples'. In his mature lectures he appears to overcome this difficulty by talking instead of'simple' and 'complex' societies. The study of simple societies allowed anthropologists to make generalisations concerning the nature of all social groupings.
The mature lectures begin with the simple statement,
The object of social anthropology in its narrowest sense is to describe the various forms of social structure found throughout the world, and to submit them to a comparative study (APNLA i(2)Li:i).
This comparative study could take two forms: the historical and the functional. Social anthropology was concerned with 'the comparative study of social structures throughout the world and their history in as far as it can be constructed from records or traditions, or deduced from clues in the present social organizations', and also with 'the comparative study of social groupings with the social functions accompanying the groupings' (APNLA 1 (2)Li:i).
The study of social anthropology was to be differentiated from sociology because:
Sociology ... [is] a deductive science which utilizes the data of psychology and social anthropology including history, to explain the formation of actual or possible social groupings (APNLA i(2)Li:i).
Armstrong was particularly interested in the relationship of psychology and social anthropology to sociology. He illustrated his ideas with a schema on the blackboard (table 3).
4l8
JAMES URRV
Table 3.   Armstrong's schema of the relation of psychology and social anthropology to sociology and the nature of sociology.
Schema
Psychology
Social Anthropology (Sc Statistical Material)
part relevant to Sociology.)
(Analysis of consciousness and laws of mental process. Social Psychology, that
(Comparative study of forms of social organisation. Statistics, e.g. economic, relevant to sociological constructions.)
the data for Sociology
consisting of hypothetical general propositions
1 Development of societies, an internal causal process;
2 Effect of introductions of new factors by non-deducible invention from within or by introduction from without. Interaction between societies;
3 Behaviour of individuals of a group under stimulus deduced psychologically from the fact of membership of the group. Behaviour of group under stimulus of all members of a group.
Psychology and anthropology
It is clear from the lectures that the basis of social anthropology for Armstrong lay in psychology. This reflected not only Armstrong's earlier connexions with psychology, but also a particular trend in British anthropology which had developed after 1900. Armstrong had been introduced to Rivers by C. S. Myers (who, with Rivers, had been a member of Haddon's Cambridge Anthropological Expedition to the Torres Straits in 1898) and his assistant F. C. Bartlett (Langham 1981:205). Bartlett had also fallen under the spell of Rivers, had contemplated doing anthropological fieldwork but instead pursued a career in psychology, becoming a dominant figure in British psychology (Bartlett 1930: 40). During the 1920's Bartlett maintained close links with anthropology and wrote a number of works on folklore, anthropology and psychology (see Bartlett 1923). He remained a close friend of Armstrong at Cambridge (see Armstrong's (1923) review of Bartlett's book). Armstrong also maintained his links with psychology, and between 1925 and 1931 was a member of the British Psychological Society.14
Psychology in Britain during the 1920's was a diffuse subject (Hearnshaw 1964). It included not only experimental psychology maintaining links with physiology and medicine, but also had connexions with philosophy, the emerging subject of sociology and the newly established field of psychoanalysis. Rivers was involved in all these areas and he related his interests to his anthropology. In 1916 Rivers declared that for him 'the final aim of the study of society is the explanation of social behaviour in terms of psychology', psychology being defined as the study of 'mental phenomena, conscious and unconscious' (Rivers 1926: 5).
On methodological grounds, however, Rivers wished to differentiate sociology from psychology and treat them as if they were independent disciplines. He argued that the observation of social behaviour had to precede the investigation
JAMES URRY
419
of social psychology, but this was purely for heuristic purposes as ultimately social and psychological observations were to be synthesised in 'a science of collective psychology' (Rivers 1923:10-11).
Social psychology in Britain during the 1920's was dominated by the work of William McDougall who also had been a member of the Torres Straits Expedition. His influential Introduction to social psychology went through a number of editions after its initial publication in 1908 and McDougall's theory of instincts was widely discussed among anthropologists, sociologists and philosophers (see Fletcher 1957). The other influential work was Alexander Shand's book on sentiments (see Shand 1914). Armstrong used both works, particularly McDougall's writings on instinct, and in his lectures he followed closely McDougall's classification of instincts.
Armstrong assumed that there was little difference between 'savage' and 'civilised' man in the relative strength of innate 'psycho-physical' dispositions since all people possessed the same instincts. But the 'mature psychological dispositions' varied enormously in different societies and 'had to be explained in terms of difference of environment, and of this environment the only important element is the social'.
The more complex the society the greater will be the variety of the mature psychological dispositions of its members, and it will be part of our task later on to correlate such. . . variations of psychological dispositions with variations of society and social groupings within societies (APNLA i(2)Li:8).
After outlining the innate instincts and their basis in neurology, Armstrong discussed the 'pseudo-instincts' of suggestion, sympathy and imitation which:
correspond to the three aspects of the mind, the cognitive, affective, and conative, and are purely social in the sense that they depend entirely on relations with other persons. Suggestion is the induction of similar ideas in other persons, Sympathy the induction of similar emotions, and Imitation the induction of similar movements (APNLA 1(2)1^3:1-2).
These distinctions, being social, are learned and liable to further modification in accordance with psychological laws. In terms of cognition the law of association operates, imitation is influenced by habit acting on motor action and they are both developed in the process of socialisation. Sympathy and the emotions are developed in complexes of sentiments; this is by far the most important element in the formation of the individual in a social environment, and the area in which there is most cultural variation. This variation is the product of differences in social organisation.
Even the most primitive society . . . has within itself several modes of grouping of its members, and to every mode of grouping there corresponds a sentiment of its members, and this sentiment has the same object for every member of the group. The sentiment towards the group is just that which distinguishes a member of the group from members of other groups (APNLA i(2)L4:6).
The complexity of sentiments in any individual is related directly to that individual's membership of the different groups which exist in all societies; the more complex the society the greater the number of groups and correspondingly there was an increase in the complex of sentiments.
420
JAMES URRY
One would like to exhibit these social sentiments as effects of social groupings—to deduce from given social grouping the appropriate sentiments. If we could do this, we should have achieved much, but merely to correlate grouping and psychological disposition would be a great deal and take us further than the usual loose method of ascribing most differences of character of peoples (e.g. nations) to race, which is so unlikely to be the cause that one wonders how the explanation has managed to persist for so long (APNLA 1(2)14:7).
The social environment not only imposes modes of behaviour and sentiments, but also a body of knowledge which Armstrong called 'tradition'. The 'social organisation with its body of tradition. . . moulds the individual, giving him those social sentiments, which differentiate him as a member of that group from members of other groups' (APNLA 1(2)1.4:7-8). Here Armstrong differed both with McDougall and other writers who posited a 'group-mind', and with what he took to be the view of the French school of Dürkheim, Lévy-Bruhl and others who spoke of a 'group-consciousness'.
Deduction of the laws ofbehaviour of members of the group as a whole, involves deduction of the group-sentiment from the group organisation and group-tradition, and deduction of the group-behaviour from the group-sentiment (APNLA 1(2)14:8).
It is obvious that for Armstrong social organisation and tradition were the important factors to be considered by social anthropology. Although Armstrong recognised a 'self-regarding' sentiment, which was individual and involved the development of self-consciousness, this sentiment was only achieved in a social environment. Though born with innate human instincts, the individual was the product of social processes.15
Social organisation
The major aim of Armstrong's lectures on social organisation was to explain the origin and development of social forms. His method, however, was not that of the nineteenth-century evolutionists. At the outset he warned his students that existing primitive societies were not a good guide to the reconstruction of original forms or the various stages of development; contemporary primitive societies were complex institutions, the result of long processes of change. Evolutionary speculations, he argued, assumed a universal pattern of development and failed to take into account individual variation and the fact that the history of culture was as much a product of chance as 'biological and psychological necessities' (APNLA 1(3^5:2). Continuing Rivers's tradition, Armstrong declared: 'Our method must, therefore, now become largely historical', by which he meant the reconstruction of culture contact and patterns of diffusion (APNLA i(3)L6:ii).
In spite of this statement Armstrong did not examine actual historical examples in his discussion of social organisation. Instead, he chose to adopt a mode of reasoning based upon the construction of a priori hypotheses. Armstrong first posited the logical nature of early social forms and then the likely path of their development, taking various factors into account. It could be argued that the result of this approach was a set of speculations not unlike those
JAMES URRY
421
of the nineteenth-century evolutionists whose methods he had disparaged. It was also quite unlike the more functionalist arguments Rivers had employed in his last lectures on social organisation and different from the approaches of the hyperdiffusionists such as Perry and Elliot Smith with whom Armstrong shared many assumptions. In modern terms Armstrong's method would be called 'model-building'.
The basic pattern of development of social forms Armstrong outlined was relatively simple. The earliest 'genetical' group was the family which progressed in scale to the clan and eventually to the tribe. Armstrong set out to define each form and to account for its development. His definitions of the family, clan and tribe were later incorporated in a series of entries he wrote for the Encyclopaedia Britannica, although his ideas on their development are not made explicit in these articles (Armstrong 1929).16 Armstrong differed with Rivers in terms of definition and the process of change, though he agreed with him on the general pattern of development. His major criticism concerned Rivers's use of the term 'family' and while he misread Rivers on this point, he did draw a distinction between 'absolute' groups such as the clan and tribe, and 'relative' groups (or groupings' as he preferred to call them) such as the nuclear family.17 Armstrong developed this point profitably in his ethnographic monograph on Rossel (Armstrong 1928«; see also Armstrong 1925).
Armstrong also addressed himself to the problem of the development of the classificatory system of relationships, a central concern of Rivers's analyses of kinship systems. Armstrong suggested that such classificatory schemes could only have developed after the establishment of speech and the formation of the family; terms for brother and sister later were extended to at least parallel cousins, which assumed the earliest form of group organisation beyond the family consisted of matrilineal kin groups. Clans therefore developed in conjunction with the extension of the classificatory system.
My conclusion then is that classificatory relationship is an early result of genetic grouping, i.e. not an historical improbability but a psychological probability. And Classificatory relationship is one of the main conditions of the development of a clan organisation. I think Lowie and Rivers put the cart before the horse in so far as they imply that the clan-organisation has given rise to the classificatory system of relationships (APNLA 1(4)1,8:3-4).
With this position established, Armstrong moved on to speculate on how different clan systems were formed in association with different marriage systems. Particular types of classificatory systems were reflected in different forms of social organisation. In this way, Armstrong explained the origin of dual organisation and more complex marriage class systems without resorting to theories of the existence of systems of group marriage which Rivers had developed in his earlier writing. But Armstrong was unwilling to go as far as Radcliffe-Brown in this regard, and although he devoted an entire lecture to a consideration of Radcliffe-Brown's writings on marriage classes (moieties, sections and sub-sections) he did not accept all Radcliffe-Brown's arguments (APNLA 1(7)1,13).18 The subject of marriage classes obviously fascinated Armstrong, and though he devoted only a small part ofhis lectures to the subject he later included a detailed analysis in his Rossel book and wrote an entry for
JAMES URRY
423
was paralleled by similar concerns in America and in Germany (Goldbeck 1980; Heine-Geldern 1964). The geographical/historical explanations were not incompatible with evolutionism but, as ideas on the migration and diffusion of cultures developed, many anthropologists rejected the claims of nineteenth-century evolutionism. In 1911, Rivers announced his rejection of evolutionism for historical explanation (Rivers 1911, see also Rivers 1926), and his major study of the development of Melanesian society (1914ft) championed not only the historical approach, but also the concepts of culture contact, migration and diffusion. During the first world war, Rivers's close association with Grafton Elliot Smith and W. J. Perry strengthened his support for such ideas and brought him into close contact with their more extreme hypotheses on migration and diffusion which rejected the idea of independent invention.
Elliot Smith and Perry were convinced that most major influences on the development of cultures stemmed directly or indirectly from ancient Egypt, and that the mark of Egypt could be seen in all parts of the world where cultures had developed beyond a primitive state of abject savagery (Daniel 1973). Rivers toyed with these hypotheses and, influenced particularly by the work of Perry (1918) and the ethnographic reports of C. E. Fox (1924), he was willing to admit that in his own research area, Melanesia, there might be evidence to support such ideas (Rivers 1922).19
Armstrong would have imbibed such ideas not just from Rivers but also from the general anthropological atmosphere of the day. It is understandable that Armstrong should include a discussion of the hypotheses in a lecture series on social anthropology, as ideas on culture contact, migration and diffusion were an integral part of Rivers's approach to social organisation and fitted Armstrong's own developmental analysis. Armstrong began his final lectures by considering Rivers's account of migrations in Melanesia, and from this he set out to make some general statements on culture contact and the diffusion of cultures.
Armstrong differentiated three kinds of exchange between cultures: 'simple diffusion', whereby mainly artefacts and basic skills were diffused through the contact of neighbouring communities; 'cultural migration', where new groups of people settled in existing communities bringing more complex influences such as religious ideas, social organisation etc., but having little effect on the 'racial' make-up of the community, and 'actual migration', where newcomers altered both the 'racial' and cultural appearance of the community (APNLA 1(10)1.24:78). With ponderous logic, Armstrong devoted an entire lecture to how the process of contact may have occurred, utilising a priori hypotheses. These dealt with the nature of initial and subsequent contact between groups, depending on whether or not they were peaceful, which group was dominant, etc.
Armstrong reasoned that the diffusion of more complex cultural elements required more intensive forms of contact, and the right mix of types of contact. His three kinds of exchange represented this increasing intensity of exchange, and aspects of social organisation were only spread after cultural migration had occurred (APNLA i(io)L25).
Elliot Smith's ideas were then examined in detail and tested against
422
JAMES URRY
the Encyclopaedia Britannica on the subject, although it was never published (Armstrong 1928: Appendix III, APNLA 5(45) 'Marriage Classes').
Armstrong included three detailed lectures on totemism, with reference to current theories and the relationship between totemism, clan organisation and marriage systems (APNLA i(7)L 14-16). He obviously read widely in this subject, as extensive notes are preserved in his papers. The gist of his lectures was later brought together in an article he contributed on the subject for the Britannica (Armstrong 1929^).
Social organisation for Armstrong, however, was much more than the study of kinship systems. In spite of his increasing interest in economics and the analysis of the Rosscl Island monetary system, Armstrong devoted very little space to an examination ofprimitivc economics, providingjust a brieflecture on exchange (APNLA i(8)Li7). But like both Rivers (1924) and Lowie (1920) in their books on primitive society, Armstrong spent a number of lectures examining formal and informal groups based on age, sex and religious principles. In a sense, such groups were both a reflection of early 'genetic' groupings, and later more sophisticated assemblies which went beyond ties created by kinship and marriage. Discussing Men's Houses, Armstrong noted:
Gregariousness must have caused the aggregation of man in earliest times; and, furthermore, out on any primordial aggregate, the working of the other fundamental instincts must have produced real social groupings with potentialities to expansion and a form of social organisation . . . But gregariousness and economic necessity need not go together; moreover, the sexual instinct would flagrantly upset the associations which economic necessity would forge from gregarious aggregates. Mental conflict was the price that man had to pay for his organisation into groups. (APNLA i(8)L2o:i).
This passage clearly reflects Armstrong's return to a psychological explanation, as well as continuing his approach to formulating the historical development of social forms. In fact, throughout his discussion of social organisation the psychological ideas with which he had started his lectures remained implicit rather than explicit. The general idea was that as social forms expanded and became more complex, so instincts and sentiments also increased in complexity. But Armstrong never really tied the two approaches together. Instead his discussion of social organisation (APNLA i(a)L22-23), particularly of secret societies in which he criticised the work of Webster (1908), moved in another direction: towards a consideration of current theories of migration and diffusion with which Armstrong was to end his lectures.
Culture contact and diffusion
The increased involvement ofBritish anthropologists in field research after 1900 resulted in the formulation of new approaches to the analysis of ethnographic material. While evolutionary problems were still examined, the survey field research conducted by anthropologists, such as Rivers in Melanesia and C. G. Seligman in Melanesia and the Sudan stimulated an interest in the relationship between culture, geographical regions and historical development (Urry 1984a: 46-8). This continued a trend begun in Britain by Haddon (Urry 1982:70) and
424
JAMES URRY
Armstrong's own classification of contacts. Elliot Smith had proposed that at the core of heliolithic culture was a set of central elements which had been diffused throughout the world and could still be seen, even though they had suffered transformation and often degeneration since first contact. Armstrong examined this set of elements and noted that while some could have been spread through simple migration, others required cultural migration to have reached remote regions. Armstrong's consideration of these issues was not critical of Elliot Smith's basic assumption of the Egyptocentric nature of some features of culture, and his aim was not to question the hypothesis but merely to elaborate upon it. What was needed was more evidence to confirm the spread of culture and more sophisticated ideas on culture contact. Armstrong thus ended his lectures on a positive note in support of the Egyptocentric bias of Elliot Smith and Perry:
The alternative to direct cultural migrations from Egypt to America20 is the supposition that at certain points on the heliolithic track, e.g. in India and Indonesia, interaction produced new centres of Egyptian culture which, either by the inhabitants acting as middle-men for traders from Egypt, or for their own needs, led to trading expeditions of converted aborigines (or half-castes in the beginning) in search of gold, pearls, cowries, etc. If there were any such centres beyond which the traders from Egypt did not go, this should be revealed by further investigation, since all heliolithic elements of culture, which crop up to the East of any such centre, must once have occurred in no less elaborate a form in that centre (APNLA i(io)L26:6).
Armstrong's advocacy of diffusionism might at first appear strange. But in the early 1920's such ideas received wide support and the more extreme views of Elliot Smith were also examined with interest. Perry's book, The children of the sun (1923), was widely reviewed and Haddon summarised its main argument with apparent approval (19236; 1924). As the decade advanced, however, the reaction against Perry and particularly Elliot Smith increased, assisted in part by Elliot Smith's extravagant claims, often couched in extreme terms, which alienated those who had at first been sympathetic to his ideas. This counter-reaction reached a peak between 1926 and 1928, and by the early 1930's the ideas of the extreme diffusionists had been largely discredited (Myres 1925; Seligman 1926/27; Marett 1927; Elliot Smith et al. 1928). Armstrong's lectures were written before these events, and his final position on the ideas of the extreme diffusionists is unclear. The only indication in print of Armstrong's support of diffusionism occurs in his 1925 review of Rivers's Social organization. Discussing Perry's diffusionist explanation of the dual organisation, which Perry had included as an appendix to Rivers's book, Armstrong noted that there was 'an inherent improbability in this theory even for one who accepts the main contentions of the "Diffusionist" school' (Armstrong 1925:45). In his ethnographic writings, including his Rossel book, and in his Britannica articles, Armstrong fails to mention diffusionist ideas.21
Armstrong and his contemporaries
In 1923 Malinowski was appointed Reader in Social Anthropology at the University of London, but it was not until 1925 that he taught a course under the
JAMES URRY
425
heading Social Anthropology. The syllabus for the course is somewhat similar in outline to that printed at Cambridge:
The Biological Basis of the Social Order. Instincts. Habits and Ideas in relation to the forms of social grouping. Social Psychology and the morphology of Primitive Communities. The Fundamental Aspects of Primitive Organisation; Kinship, Local Groupings; The Clan; Economic Organisation and the Constitution of the Tribe (Calendar of the LSE igz6-2j, 1926:98).
The course was supposed to consist of twenty-five lectures, but Firth notes that Malinowski presented seminars instead (1963: 5-6).
Malinowski had chosen the title of his Readership himself (see Firth 1963: 5 for details), and while he certainly was eager to examine the social significance of culture he was not at first too concerned with the title 'social' anthropology. In an early article he referred to his work as 'ethnology' (Malinowski 1922), and later was quite willing to use the terms 'society' and 'culture' in the same context.
Social, called also Cultural Anthropology, studies questions of the culture and social organisation of primitive tribes and nations. As a rule, a somewhat vague line of demarcation is drawn between peoples of simpler culture and those more highly developed, such as the modern inhabitants of Europe and North America. The study ofhigher civilisations is then assigned to Sociology. The distinction, however, is unsatisfactory and it would be more correct to say that Social Anthropology is a branch of Sociology, as applied to primitive tribes (Malinowski 1929: 862).
Like Armstrong, Malinowski was extremely interested in the connexion between psychology and anthropology. McDougall's (1908) Introduction to social psychology was one of the three books suggested for his course and he recommended Shand's (1914) work on sentiments to his students (the other texts were Rivers 1924 and Lowie 1920, see also Firth 1957:7,1981:125). But Malinowski's attitude altered during the late 1920's and early 1930's. His initial enthusiasm for McDougall and Shand declined as his interest in psychoanalysis increased, but this too waned, to be replaced during the 1930's by a more general concern with behaviouralism (Fortes 1957:161).22 The study of social institutions was central to Malinowski's anthropology. He discussed the family, clan and tribe in much the same way as Armstrong and tended to see the clan as a consequence of the expansion of culture and social co-operation in more complex groups (Malinowski 1926:135).23 But the similarities with Armstrong's approach are superficial; Malinowski was little concerned with discussing the development of social units in detail, nor was he interested in examining social structure in abstract terms. Social organisation was something to be explained in terms of the context of everyday life in concrete situations; Malinowski's opposition to the formal analysis of kinship, and his rejection of historical and evolutionary ideas, is well known (Malinowski 1930; 1931&: xxiv-xv, see also Fortes 1957). By the late 1920's Malinowski's theory of functionalism, which had been implicit from his early writings, now became more explicit and central to his explanation of society and culture. It was an anthropology founded firmly on direct field research, not on abstract analysis of social structure, or the postu-lation of a priori hypotheses.
Radcliffc-Brown began to formulate his distinctive vision of anthropology
426
JAMES URRY
before the first world war. Influenced by his teachers, Rivers and Haddon, his early work was couched in evolutionary terms. Such ideas were, however, in a state of flux (Stocking 1983:94) and Radcliffe-Brown, influenced by the sociological approaches of Herbert Spencer and Durkhcim and the study of Australian social organisation, began to reformulate his ideas. An important factor in this reformulation was Radcliffe-Brown's reaction to Rivers's advocacy of historicism after 1911 and an extensive correspondence with Rivers (see Stocking 19846; 145-56) which helped focus his thinking on a number of subjects. In 1913, responding to Rivers's paper on 'survivals' (1913), Radcliffe-Brown argued that the task of anthropology was largely 'psychological' (1977 see also Stocking 19846: 154).24 Following McDougall, he described psychology as the science of human behaviour and this science included sociology which was merely a special branch of psychology. In terms of the analysis of society Radcliffe-Brown objected to Rivers's historical speculations and, following Spencer, he differentiated between static and dynamic sociology, stating that the analysis of social statics logically had to precede that of social dynamics. Although Rivers rejected both Radcliffe-Brown's definition of psychology and his stress on the importance of synchronic studies, he considered Radcliffe-Brown's arguments with some care as can be seen in his final lectures on social organisation.25 The first world war, with Rivers involved in war-work in England and Radcliffe-Brown living in Australia and Tonga, prevented the continuation of their dialogue.
The synchronic analysis of social systems was to remain a central feature of Radcliffe-Brown's vision of anthropology until his death, and he published a number of versions of what Lowie (1937: 222) called his 'pronunciamentos' on the proper classification of the fields of anthropological study. The earliest of these appeared in 1923 and others followed in later years, with minor alterations and the major features remained remarkably consistent (Radcliffe-Brown, 19236; see also Radcliffe-Brown, 1958 which reprints this paper and others on this theme). In the earliest version, physical anthropology concerned itself with the biological nature of man, psychology with 'individual behaviour in its relation to the individual' and social anthropology with 'the behaviour of groups or collective bodies of individuals in its relation to the group' (Radcliffe-Brown 1958:17). Evolutionary and historical explanations were not really scientific and therefore were not even called anthropology, but were lumped under the term 'ethnology'. The major change which had occurred in Radcliffe-Brown's thinking since 1913 was that the study of society now had nothing to do with psychology; social anthropology was an independent science, with its own subject matter and its own methods. The ultimate aim of social anthropology was not to relate social systems to psychological universals, nor to trace the development of social forms, but to establish universal laws of the structure and function of social systems.26
It is clear that Armstrong's social anthropology, with its psychological base, its a priori hypotheses and concern with historical patterns, diffusion and migration would have been anathema to Radcliffe-Brown. But Armstrong's consideration of aspects of social organisation would have found parallels in Radcliffe-Brown's thinking. They both derived their interest and much of their
JAMES URRY
427
methodology on this subject from Rivers. But there were few opportunities for them to discuss such common concerns during the 1920's, though they may have corresponded. Radciiffe-Brown was abroad during this period; from 1920 to 1921 he was the Ethnologist at the Pretoria Museum, from 1921 until 1926 he was Professor of Anthropology at the University of Cape Town, and from 1926 to 1931 Professor at Sydney University in Australia.
Armstrong's social anthropology, though sharing common concerns with those of his major contemporaries, was also distinctive. In a sense it was a continuation and elaboration of Rivers's concerns; Radcliffe-Brown had begun to separate himself from River's concerns before 1914, and Malinowski, never a student of Rivers, had experienced other influences. During the early years of the 1920's both Malinowski and Radcliffe-Brown were establishing themselves in their academic careers and developing their distinctive approaches. It is not surprising that Armstrong should have taken little note of them in his lectures, as the full impact of their thinking was not felt until later in the decade.
Armstrong's own students were few in number, and his influence was not sustained as he abandoned anthropology and most of his students came under the influence of Malinowski and especially Radcliffe-Brown. Armstrong's lectures were not delivered to undergraduates eager to seek a career in social anthropology as there were so few openings in the subject at this period. Many of his students were recruits for the Indian Civil Service, but anthropology was not seen as an essential part of their training or that of other colonial officials at this time, in spite of attempts by Haddon to get it so recognised (see Leach 1984:9-10). A few students, however, did take anthropology courses in the hope of doing fieldwork and seeking a career in the subject. Three students of Armstrong stand out in this respect, A. B. Deacon, Camilla Wedgwood and Gregory Bateson.
Deacon was awarded a First Class in the Anthropological Tripos in 1925. In a letter to Haddon he wrote of Armstrong, 'On the theoretical side of social anthropology, I found him very interesting and illuminating. He is extremely fair and unbiased and a confirmed logician' (in Haddon 1934: xiv). Deacon's first publication involved a comparison of the similarities between religious cults of Ccram in eastern Indonesia and those of parts of Papua New Guinea. These he -explained in terms of diffusion (Deacon 1925). This concern with culture patterns and diffusionism he derived from both Haddon and Armstrong and it played a part in his field research in the New Hebrides, which he began in 1925 (Deacon to Haddon in Haddon 1934: xv; Deacon 1934; ch. 24; Larcom 1983: 177). In the New Hebrides, Deacon concentrated on social organisation and he published a major paper on marriage classes in Ambry in, thus continuing the work of Rivers which he had learnt through the medium of Armstrong's instruction (Deacon 1927; Deacon to Haddon 1934: xxiii). Deacon, however, was to die of Blackwater fever near the end of his research, although his fieldnotes, edited by Wedgwood and published thanks to Haddon's efforts, appeared after his death.
Armstrong's influence on Wedgwood and Bateson is more difficult to assess. Though both received their initial training from Armstrong and Haddon, both were deeply influenced by the new ideas in British social anthropology in the late
428
JAMES URRY
io2o's and early 1930's, through direct contact with Malinowski and Radcliffe-Brown. Some of Wedgwood's early papers may reflect Armstrong's concerns, but the arguments are functional (Wedgwood 1927; 1930a; 19306). Bateson failed to recognise Armstrong as one of his teachers (see acknowledgements in Bateson 1936: ix) and in 1972 even claimed that no one at Cambridge was interested in social anthropology in his day (quoted in Lipsct 1980:121).
Conclusion
Armstrong's social anthropology had a distinctly Cambridge flavour. His concern with social organisation, culture contact and diffusionism was derived from Rivers's teaching; his interest in psychology, though derived in part from Rivers, also came from Bartlctt; his a priori reasoning may have been a product of Cambridge philosophy. Armstrong thus built upon the teachings he had received at Cambridge and developed them in a Cambridge atmosphere independently of his contemporary colleagues, Radcliffe-Brown and Malinowski, who were formulating their own positions elsewhere.
The continuation and elaboration of Rivers's concerns is of interest. Recently Langham has argued (1981: 159, 198, 311 etc.) that there were two Rivers, an 'earlier' Rivers, whose interests were in social organisation and the scientific analysis of kinship, and 'later' Rivers, dominated by historical and diffusionist ideas, including extreme diffusionism. In making this distinction Langham is following the rationalisations made of Rivers's scholarly life by Radcliffe-Brown and others in the 1920's and by later writers on the subject. This was just one of the popular distinctions concerning Rivers's vision of anthropology made during the 1920's in reaction to the claims of Elliot Smith and Perry that Rivers had given priority to their ideas in his final years (see Urry 19846). The conflict over Rivers's final position sealed the fate of Rivers's reputation leaving behind a sense of ambiguity concerning his vision of anthropology. While it is true that this sense of ambiguity remains, it is clear that Rivers did not necessarily consider that historical, including diffusion studies, nor social and psychological approaches were contradictory, but that in fact they could be complementary. Armstrong's lectures appear to confirm this.
Armstrong's social anthropology belonged to a transitionary period in British anthropology. His students, and those trained by Marett at Oxford during the 1920's, all fell under the spell of a different social anthropology propounded by Malinowski and Radcliffe-Brown. The study of social organisation became more focused, the study of kinship and its politico-jural aspects became increasingly important in the 1930's as the emphasis of ethnographic field research swung away from Oceania towards Africa (Kuper 1982). Interests in the relation between sociological and psychological analysis declined, historical interests were replaced by a concern with social change and the term 'culture contact' now applied to relations between European and other cultures. The vision of social anthropology, which was to be established in the new academic departments founded after 1945, was largely that of Radcliffe-Brown and it was his ideas which were to prevail in the triumph of the 'social' anthropologists.
JAMES URRY
429
notes
I should like to thank Mr J. G. Armstrong for permission to use and quote from his father's papers. Dr E. S. Leedham-Green kindly provided details from the Cambridge University Archives. I should also like to thank Dr Charlotte Carr-Gregg, Professor Sir Raymond Firth, Margaret Gardiner, Miss B.J. Kirkpatrick, Professor Sir Edmund Leach and Professor George Stocking for their assistance in writing the article.
1 This change is epitomised by the foundation in 1946 of the Association of Social Anthropologists to represent the interests of professional academic anthropologists in Britain and the British Commonwealth.
2 Unless otherwise indicated, details of Armstrong's career are derived from his entry in Who was Who, 1971-1980 and an anonymous obituary notice in The Times (London), 26 March, 1980:19.
3 When Armstrong's name appeared in a letter quoted in his autobiography Russell noted with economy of phrase: 'Armstrong was a man whom I came to know as an undergraduate at Cambridge. He enlisted at the beginning of the war, lost a leg and became a pacifist' (Russell 1968:6 in).
4 See Haddon 1928 where Haddon summarises the ethnological evidence on the Massim area and places Rossel in context. Malinowski, when working in the Trobriand islands in 1915—1917 was constantly encouraged by his anthropological supervisors to visit Rossel, but never did (Stocking 1983:97; Young 1984). Langham (1981:205) reports that Malinowski later suggested Rossel to Armstrong as a suitable place for fieldwork.
5 For a record of Armstrong's service see the letter of H. W. Champion dated 21 April 1922 in APNLA 5(44)-
6 Minutes of the Board of Archaeological and Anthropological Studies 12 June 1922, University Archives, University of Cambridge. The other names put forward were W. J. Perry, N. W. Thomas, B. Malinowski and Paul Radin.
7 Report of the Board for Archaeological and Anthropological Studies to the General Board of Studies, January 1923, University Archives, University of Cambridge.
8 See his entry in Who was Who igsi-ig6o and Morrison (in press). Armstrong apparently also applied for the Readership (Langham 1981:179).
9 Minutes of the Board of Archaeological and Anthropological Studies February 1926, 20 June 1926, 13 October 1926, University Archives, University of Cambridge. I am grateful to Dr E. S. Leedham-Green of the University Archives for pointing out the circumstances of the events of 1926. Hodson's decision obviously came as a shock to Haddon who later, in a reference he wrote for Armstrong in 1928, noted that 'to my dismay he [Hodson] did not retain Armstrong on the staff and that 'thereby Ethnology lost one of its most promising students', Haddon Letter 7, University Archives, Cambridge.
10 With some minor alterations and re-organisation these were published as Rivers 1924.
11 The lectures are scattered through various boxes in the collection and are not identified by name though they can easily be collated with the published version (see note 10). They are the top-copies. Rivers's Fitzpatrick Lectures (Rivers 1916) are also included in typescript form with Rivers's own corrections.
12 The full typescript texts of this lecture series is also in Armstrong's papers. It is difficult to reconstruct which courses Armstrong actually taught as entries in the Handbook, the archives and in the University Reporter are impossible to reconcile. I am grateful to Professor Edmund Leach for alerting me to these problems. Professor Leach has also provided me with A. B. Deacon's lecture notes of Armstrong's lectures which appear to belong to the period 1923-1924. Copies of these have been deposited with the Armstrong Papers in Canberra.
13 Offprint from the Student's Handbook, 1915 in APNLA 3(25).
14 See membership lists in the British Journal of Psychology for the period; Armstrong was involved with Bartlett, J. T. MacCurdy and Haddon in a discussion of 'social constructiveness' in the Society's journal; see Armstrong 19286.
15 In a later article Armstrong argued that one could talk in abstract terms of three kinds of group behaviour: the human response, which was the product of innate instincts; the cultural response, the product of social learning; and the individual's life-history and innate genetic make-up, but influenced by the human and cultural response; see Armstrong 19286:396,
43°
JAMES URRY
16 Armstrong 1929a, b, c. The last article is not attributed to Armstrong but the original typescript is in his papers, APNLA 5(45).
17 Rivers's own position altered on this point but in his last lectures he differentiated four types of family: the nuclear, the bilateral kindred and two forms of unilateral groups (=lineages) based on patrilineal and matrilineal descent; see Rivers (1924: 15). In his lectures and his review of Rivers's book (1925) Armstrong confused the first two. On the general misinterpretation of Rivers's views on this issue see Freeman (1961: 197-200).
18 Rivers had likewise discussed Radcliffe-Brown's ideas (Rivers 1924 Appendix II). Radcliffe-Brown's papers on Australian marriage systems (1918,1923a) were the major focus of attention. It is clear from Deacon's notes on Armstrong's courses that considerable attention was paid to Radcliffe-Brown's arguments, not so much in the lectures but perhaps in classes. This may well have been merely for Deacon's own interest.
19 Much of Fox's work first appeared in articles before Rivers's death and Rivers also consulted Fox's writing in manuscript.
20 Elliot Smith (1924) had claimed, amid some controversy, that Middle American civilisation had been derived from Egypt.
21 Armstrong (1929c/) wrote anthropological entries on New Guinea and Oceania for the Britannka and never mentioned diffusionist ideas. In his 'totemism' article (Armstrong 19290") he concluded by recognising that functionalist explanations should take preference over historical reconstructions. But in comments made to an anthropologist who utilised his material on the Rossel monetary system in the 1960's, Armstrong indicated he still shared many of Rivers's ideas as he believed the Rossel system reflected an earlier system which had degenerated (see Baric 1964: 39).
22 The word 'behaviour' becomes common in Malinowski's work from the early 1930's; see his statement that 'Social organization is the standardized manner in which groups behave' (Malinowski 19310:622).
23 Malinowski dropped this argument when he revised his article for inclusion in the 14th Edition under the title 'Social Anthropology'.
24 Radcliffe-Brown's monograph on the Andaman Islanders, written at this period but not published until 1922, reveals similar concerns and sympathy with the psychological ideas of McDougall and Shand (see Stocking 19846:157).
25 Although Radcliffe-Brown's article remained unpublished, the paper, and comments made in correspondence, encouraged Rivers to reply (see Rivers 1926: 3).
26 Radcliffe-Brown's attitudes to psychology remained antagonistic, although he later differentiated between types of psychology, some of which he approved of more than others.
27 Langham (1981:163-5, 179-80, 215) has difficulty explaining Deacon's interests in culture history because he fails to realise that this was a continuing concern of Cambridge anthropology during the 1920's.
28 Wedgwood taught under Radcliffe-Brown in Sydney during the late 1920's and early 1930's when most of these papers were written. She maintained contact with Armstrong, however. I am grateful to Charlotte Carr-Gregg for letting me see letters from Wedgwood to Armstrong from this period.
references
APNLA Armstrong Papers, National Library of Australia, Canberra, Australia (Ms. 6507) (Box
number/Folder Number/Lecture number/page). Armstrong, W. E. 1922. Report on Suau—Tawala. In Australia-Papua: Annual Report for the Year
1920-21. Melbourne: Government Printer (Reprinted as Anthropology Report No. 1 by the
Government of the Territory of Papua, 1922).
-1923. [Review of] Psychology and primitive culture by F. C. Bartlett. Camb. Rev. 45 (1102), 66.
-1923/4. Rossel Island religion. Anthropos 18/19, 1-11.
-1924a. Rossel Island money: a unique monetary system. Econ.J. 34, 423-29.
•-— 1924(1. Shell money from Rossel Island, Papua, Man 24, 161-2.
--■ 1925. [Review of] Social organization by W. H. R. Rivers. Man 25, 25-45.
-1928a. Rossel Island: an ethnological study. Cambridge: Univ. Press.
-19286. Social constructiveness III. Brit. J. Psychol. 18, 396-99,
JAMES URRY
43i
-1929a. Clan. Encyclopaedia Britannka (14th edn) 5, 752-3.
-1929/). Family. Encyclopaedia Britannka (14th edn) 9, 59-60.
——— 1929c. Tribe. Encyclopaedia Britannka (14th edn) 22, 465.
-19290". Totemism. Encyclopaedia Britannka (14th edn) 22, 315-18.
———■ 1929c New Guinea, Anthropology. Encyclopaedia Britannka (14th edn) 16, 198-300. --1929/. Oceania. Encyclopaedia Britannka (14th edn) 16, 693-4.
Baric, L. 1964. Some aspects of credit, saving and investment in a 'non-monetary' economy. In Capital, saving and credit in peasant societies (eds) R. Firth & B. S. Yamey. London: Allen and Unwin.
Bartlett, F. C. 1923. Psychology and primitive culture. Cambridge: Univ. Press.
-1930. [Autobiography] In A history of psychology in autobiography Vol. 3. Worcester (Mass.):
Clark Univ. Press.
Bateson, G. 1936. Naven: a survey of the problems suggested by a composite picture of the culture of a New
Guinea tribe drawn from three points of view. Cambridge: Univ. Press. Cambridge University Student's Handbook 192;, 1927-8. Cambridge: Univ. Press. Daniel, G. 1973. Elliot Smith, Egypt and diffusionism. In The concepts of human evolution (ed.) Lord
Zuckerman. London: Academic Press. Deacon, A. B. 1925. The Kakihan society of Ceram, and New Guinea initiation cults. Folk-Lore 36,
332-61.
-1927. The regulation of marriage in Ambrym. J. R. anthrop. Inst. 57, 377-97.
-1934. Malekula; a vanishing people in the New Hebrides. London: George Routledge.
Firth, R. 1956. Alfred Reginald Radcliffe-Brown 1881-1955. Proc. Brit. Acad. 42, 287-302. -1957. Introduction: Malinowski as scientist and as man. In Man and culture: an evaluation of the
work of Bronislaw Malinowski (ed.) R. Firth. London: Routledge'& Kegan Paul. -1963. A brief history of the Department (1913-1963). London: School of Economics & Political
Science.
-1981. Bronislaw Malinowski. In Totems and teachers: perspectives on the history of anthropology
(ed.) S. Silverman. New York: Columbia Univ. Press. Fletcher, R. 1957. Instinct in man in the light of recent work in comparative psychology. London: George
Allen & Unwin.
Fortes, M. 1953. Social anthropology at Cambridge since 1900. Cambridge: Univ. Press.
- 1957- Malinowski and the study of kinship. In Man and culture: an evaluation of the work of
Bronislaw Malinowski (ed.) R. Firth. London: Routledge & Kegan Paul. Fox, C. E. 1924. The threshold of the Pacific. London: Kegan Paul, Trench & Triibner. Frazer, J. G. 1913. The scope of social anthropology. In Psyche's task. London: Macmillan. Freeman, J. D. 1961. On the concept of the kindred. J. R. anthrop. Inst. 91, 192-220. Goldbeck, A. L. 1980. Quantification in ethnology and its appearance in regional culture trait
distribution studies (1899-1939)./. Hist, behav. Sci. 16, 288-40. Goody, J. 1973. British functionalism. In Main currents in cultural anthropology (eds) R. Naroll &
F. Naroll. Englewood Cliffs, New Jersey: Prentice Hall. Haddon, A. C. 1918. Anthropology as a University subject. N. Z.f. Sci. Technol. 1, 259-61.
-1923a. A brief history of the study of anthropology at Cambridge. Cambridge: Privately Printed.
-19236. Migration of peoples of the south-west Pacific. In Proceedings of the Pan-Pacific Science
Australia, 1923 (ed.) G. Lightfoot. Melbourne: Government Printer.
-1924. The cultural history of the Pacific. N. Z. J. Sci, Technol. 7, 101-6.
-1928. Introduction. In Rossel Island: an ethnological study by W. E. Armstrong. Cambridge:
Univ. Press.
-1934. Preface. In Malekula: a vanishing people in the New Hebrides by A. B. Deacon. London:
George Routledge.
Hearnshaw, L. S. 1964. A short history of British psychology 1840-1940. London: Methuen. Heine-Geldern, R. 1964. One hundred years of ethnological theory in the German-speaking
countries: some milestones. Curr, Anthrop. 5, 407-18. Kuper, A. 1973. Anthropologists and anthropology: the British school 1922-1972. London: Allen Lane.
-1982. Lineage theory: a critical retrospect. Ann. Rev. Anthrop. II, 71-95.
-1983. Anthropology and anthropologists: the modern British school. London: Routledge & Kegan
Paul.
432
JAMES URRY
Langhara, I. 1981. The building of British social anthropology: W. H. R. Rivers and his Cambridge
disciples in the development of kinship studies, 1898-1931. Dordrecht: D. Reidel. Larcom, J. 1983. Following Deacon: the problem of ethnographic reanalysis, 1926-1981. In
Observers observed; essays on ethnographic fieldwork (ed.) G. W. Stocking (Hist. Anthrop. 1).
Madison: Univ. of Wisconsin Press. Leach, E. R. 1984. Notes on the mythology of Cambridge anthropology. Camb. Anthrop. 9, 1-16. Lipset, D. 1980. Gregory Bateson: the legacy of a scientist. Englewood Cliffs, New Jersey: Prentice
Hall.
London School of Economics and Political Science 1926, Calendar 1926-1927. London: LSE. Lowie, R. H. 1920. Primitive society. New York: Boni & Liveright.
-1937- The history of ethnological theory. New York: Holt, Rinehart & Winston.
MacDougall, W. 1908. An introduction to social psychology. London:
Malinowski, B. C. 1922. Ethnology and the study of society. Economical, 208-19.
- 1926. Anthropology. Encyclopaedia Britannica (13th edn, Supplementary volume) 29,
131-4°-
-1929. Social anthropology. Encyclopaedia Britannica (14th edn) 21, 862-70.
-1930. Kinship. Man 30, 19-29.
•-• 193 la. Culture. Encyclopaedia of the Social Sciences 4, 621-46.
-19316. The sexual life of savages in north-western Melanesia (2nd edn). London: Regan Paul.
Marett, R. R. 1927. The diffusion of culture (The Frazer Lecture in Social Anthropology). Cambridge: Univ. Press.
Morrison, C. in press. Three styles of imperial ethnography: British officials as anthropologists in
India. In Knowledge and society: studies in the sociology of culture past and present 5. Myres, J. L. 1925. The geographical view of the historical method. Geogr. Teacher 13, 3-38. Perry, W.J. 1918. The megalithic culture of Indonesia. London: Longmans Green. -1923. The children of the sun.
Quiggin, A. H. 1942. Haddon the headhunted a short sketch of the life of A. C. Haddon. Cambridge: Cambridge Univ. Press.
Radcliffe-Brown, A. R. 1918. Notes on the social organization of Australian tribes. I.J. R. anthrop. Inst. 48, 222-53.
-1922. The Andaman Islanders: a study in social anthropology. Cambridge: Univ. Press.
-1923a. Notes on the social organization of Australian tribes II. J. R. anthrop. Inst. 53, 424-47.
■-— 19236. The methods of ethnology and social anthropology. S. Afr.J. Sci. 20, 124-47.
-1958. Method in social anthropology. Chicago: Univ. of Chicago Press.
-■ 1977. The study of social institutions. Camb. Anthrop. 3, 35-48.
Rivers, W. H. R. 1911. The ethnological analysis of culture. Nature, Land. 87, 356-60.
-1913. Survival in sociology. Sociol. Rev. 6, 293-305.
■-- 1914a. Kinship and social organization. London: Constable.
-19146. A history ofMelanesian society. Cambridge: Univ. Press.
Rivers, W. H. R. 1916. Medicine, magic, and religion. The Lancet Jan 8th, Jan 15th: 59-65, 117-23.
-1922. The unity of anthropology. J. R. anthrop. Inst. 52, 12-25.
-1923. Psychology and politics and other essays. London: Kegan Paul, Trench, Triibner.
-1924. Social organization. London: Kegan Paul, Trench, Triibner.
-1926. Psychology and ethnology. London: Kegan Paul, Trench, Triibner.
Russell, B. 1968. The autobiography of Bertrand Russell (vol. 2, 1914-1944). London: George Allen & Unwin.
Seligman, B. 1926/7. Critical notice of W. H. R. Rivers Psychology and culture. Brit. J. Psychol. 17, 370-6.
Shand, A. 1914. The foundations of character: being a study of the tendencies of the emotions and sentiments.
London: Macmillan. Slobodin, R. 1978. W. H. R. Rivers. New York: Columbia Univ. Press. Smith, G. E. 1924. Elephants and ethnologists. London: Kegan Paul, Trench, Trubner.
-et al. 1928. Culture: the diffusion controversy. London: Psyche Miniatures.
Stocking, G. W. 1983. The ethnographers magic; fieldwork in British anthropology from Tylor to
Malinowski. In Observers observed: essays on ethnographic fieldwork (ed.) G. W. Stocking (Hist.
Anthrop. 1). Madison: University of Wisconsin Press.
JAMES URRY
433
- 1984a. Dr Dürkheim and Mr Brown: comparative sociology of Cambridge in 1910. In
Functionalism historicized: essays in British social anthropology (ed.) G. W. Stocking (Hist.
Anthrop. 2). Madison: Univ. of Wisconsin Press. -19846, Radcliffe-Brown and British social anthropology. In Functionalism historicized: essays
on British social anthropology (ed.) G. W. Stocking (Hist. Anthrop. 2). Madison: Univ. of
Wisconsin Press.
Strong, W. M. 1922. First annual report by anthropologist. In Australia: Papua. Annual Reportforthe
Year 1920-21. Government Printer. Urry, J. 1972. Notes and Queries on Anthropology and the development of field methods in British
anthropology 1870-1920. Proc. R. anthrop. Inst. 1972, 45-57. -1982. From zoology to ethnology: A. C. Haddon's conversion to anthropology. Canberra
Anthrop. 5.
-1983. The Wallace Edwin Armstrong Papers in the National Library of Australia, Canberra,
Australia. Hist. Anthrop. Newsl. 10, 3-5. - 19840. A history of field methods. In Ethnographic Research (ed.) R. F. Ellen. London:
Academic Press.
- 19846. Facts, fantasies and falsehoods: Ian Langham and the history of British social
anthropology. Unpublished ms. Vellacott.J, 1980. Bertrand Russell and the pacifists in thefirst World War. New York: St Martins Press. Webster, H. 1908. Primitive secret societies: a study in early politics and religion. London: Macmillan, Wedgwood, C. H. 1927. Death and social status in Melanesia. J, R. anthrop. Inst. 57, 377-97.
-1930a. Some aspects of warfare in Melanesia. Oceania i, 1-33.
-19306. The nature and function of secret societies. Oceania 1, 129-45.
Young, M. 1983. The intensive study of a restricted area, or, why did Malinowski go to the
Trobriand Islands? (Unpublished ms.).