Extract from a paper given at the 'Friday Seminar' at the Pitt Rivers Museum on 23 January 2009 as part of the Revisiting Victorian Anthropology series run by the Other Within project. References to slides have been omitted. The thoughts of the author in January 2009 about Pitt-Rivers would not necessarily be shared by the author at the end of March 2011 (when this paper was added to the website), knowing what she knows now.

Total immersion or paddling?: different models of fieldwork in Victorian anthropology, 1874-1914

Abstract: Fieldwork has been called 'the central ritual of the tribe' of anthropologists. [Stocking 1983: 70] This paper asks why Malinowski's model of fieldwork, based upon his work in Papua New Guinea during the First World War, became the defining form, and whether alternative models of fieldwork carried out by Victorian anthropologists between 1874 and 1914 might also be suitable.

Fieldwork is seen by many to be at the core of what it is to be an anthropologist. George Stocking in his 1983 paper 'The Ethnographer's Magic' calls it 'the central ritual of the tribe'. [1983: 70] Malinowski is often portrayed as the founding father of modern anthropological fieldwork; he described it as follows:

I have lived in that archipelago for about two years ... during which time I naturally acquired a thorough knowledge of the language. I did my work entirely alone, living for the greater part of the time right in the villages. I therefore had constantly the daily life of the natives before my eyes, while accidental, dramatic occurrences, death, quarrels, village brawls, public and ceremonial events, could not escape my notice.' [Malinowski 1922: xvi-xvii]

As Henrika Kuklick as pointed out though, 'reportorial conventions of anthropological accounts have permitted readers to imagine fieldworkers experiencing exotic ways of life through total immersion, but field practice has often departed from this ideal'. [Kuklick, 1998: 164] Malinowski's diary certainly suggests that the reality was a little more complicated that his quotation implies. Following his supposed lead, the ideal twentieth-century anthropologist was expected to select one culture to study, identify the best location in which to base themselves for a prolonged period (usually taken to be a year), to learn to speak the local language as fluently as possible, totally immerse themselves in local culture and act, at all times, as a 'participant-observer'. Uptake of fieldwork à la Malinowski diminished towards the end of the century as restrictions in funding and concerns over safety etc made it increasingly impractical for many anthropological postgraduates.

Actually it is recognized that there never has been a single model for fieldwork. Some anthropologists today are eagerly embracing other models, like multi-sited ethnography. [Marcus, 1995] The big question is, why was Malinowski's model apparently chosen as THE 'distinctive apprenticeship', that Adam Kuper for one identifies, in the first place? [Kuper, 1996: 1] There is insufficient time to answer this large question here so I will concentrate on the aspects of it that affect one of the branches of mainstream anthropology — museum ethnography or museum anthropology, call it what you will — and look at the kind of fieldwork which resulted in the acquisition of artefacts, specifically objects which ended up here at the Pitt Rivers Museum. I will ask whether some of the alternative models used in Victorian times might work just as well for acquiring 'good' collections for museums today.

This paper considers four alternative models of fieldwork used by late Victorian anthropologists which I have labelled:

1.  Letting others do the hard work

2.  Surveying anthropology

3.  The field as laboratory, the anthropologist as team

4.  Field site as home

Naturally there is some slippage between these categories, even in the few examples I discuss at length, which I hope you will forgive.

Quite a few anthropologists could be used to exemplify these models but I have selected five men who are all linked to the Pitt Rivers Museum in some way. Four of these men were almost exactly the same age, Pitt Rivers was older but they were all Victorians. [1] There are other connections between these men (apart from their gender). Three of them were trained natural scientists (Spencer, Balfour and Haddon) at a time when the second edition of Notes and Queries, published in 1892, defined anthropology as 'the natural history of man'. [2] [Garson and Read 1892: B] Pitt Rivers' first collection founded the eponymous museum at Oxford in 1884, which Balfour curated from 1891-1939. Gillen, through his partner, Spencer, was acquainted with Balfour and contributed to the museum's collections. Balfour was a close colleague of Haddon and Spencer. Spencer was advised by Haddon to take a phonograph and ciné camera on his second expedition with Gillen. Interestingly, at least one of them was known well by Malinowski: Spencer sponsored him whilst he was in Australia before his PNG fieldwork and was known by him as 'Old Baldie'. Malinowski was also linked to the Pitt Rivers Museum — we have five canoe prows loaned by him in 1916 but not formally accessioned until 1971; another part of his collection was sold to the Museum, after his death, by his wife but later returned to her. [3] Four of the men undertook fieldwork at the same time, using the same key texts and answering some of the same questions.

Because Malinowski's fieldwork is well-known, and time is short, I will not discuss the details of his fieldwork model. Haddon's work has also been extensively examined and published recently (see Herle 1998) and I will deal with it cursorily. Balfour and Pitt Rivers' contributions, though generally less well-known, are of course known to us here; the most time therefore will be spent detailing the fieldwork of our fifth man, Frank Gillen, whose work (particularly when seen in isolation from his more famous partner) is probably the least familiar of the five.

So to our first fieldwork method: controversially perhaps, it can be seen as the choice of the lazy researcher.

First alternative: Paddling -  or Letting others do the hard work

Pitt Rivers chose not to visit the field himself to obtain data, which might have seemed the obvious thing to do; he believed that it was possible to base anthropological theory on information, and artefacts, obtained by other people. [Petch 2007: 29] He, like Tylor, believed that artefacts were cultural facts, and that 'cultural forms rendered into material forms [were] amenable to empirical study', 'myths and kinship were considered along with weapons and weaving as elements of culture of equivalent type'. [Brown et al, 2000: 259, 263] The artefacts Pitt Rivers acquired were in essence the 'facts' he was interested in, he was indifferent to the other types of data sent back by travellers: vocabularies, measurements, pictures, maps, and stories. Objects could be collected by others, just as well as by himself, and by concentrating on a UK-based hunt for artefacts, he ensured the widest possible range of sources and geographical spread. By trawling the market and dealers throughout Europe, and pursuing every traveller with a collection to sell, he managed to acquire one of the foremost ethnographic collections of the Victorian (or any other) age.

In order to ensure that he based his work upon the best possible sources of both forms of data, he helped to establish Notes and Queries in 1874 which directed travellers and other amateur anthropologists in the way to obtain data, and the questions to ask.[3] Pitt Rivers financially supported the first two editions of this publication and he also wrote several sections. [Petch 2007: 27] Some of these survived in the ensuing editions of the Guide until the fifth edition, published in 1929. Although facts and artefacts came to him in a haphazard manner he then tried to impose an order upon them in his publications, often in the process disguising the essentially random, opportunistic method of acquisition. [Petch 1998, 2006]

Pitt Rivers was, therefore, one of the group of what Stocking called 'evolutionary titans, seated in their armchairs, [who] culled ethnographic data from travel accounts to document their vision of the stage of creation of human cultural forms'. [1983: 71] As Stocking also points out, by making 'empirical data collected by gentleman amateurs abroad' feed the 'systematic inquiries of metropolitan scholarship', anthropologists were only following the lead of other mid-Victorian scientists. [Stocking 1983: 72] Pitt Rivers' laissez-faire attitude to data and 'facts', and what might be seen as his preference for sweeping generalization based on little hard evidence, can be interpreted as part of his armchair, home-based methodology. Separating data-collection and theory was seen as a positive thing, not a necessary evil - because data collectors in the field would not be biased by theoretical ideas, and theorists saw as much data as possible before developing their theses. Greater anthropological theorists than himself, like Tylor and Frazer, employed exactly the same methodology to greater theoretical effect. Indeed this method may have facilitated theorizing, freeing the scientist to compare and contrast between different cultures and ways of life that he himself was not intimately acquainted with.

None of our other, later, examples separated data collection from theorizing in this way: for each of them the collection of facts and the consequent theorizing was part of the same package, carried out by a single individual (or in the case of two of the examples, as part of a team).

Second alternative: Anthropology as survey

Pitt Rivers was self-taught, indeed he started working in anthropology long before it was possible to formally study it at a British university. The next generation of scientists were more fortunate. Both Henry Balfour and his contemporary and friend, Walter Baldwin Spencer, studied the natural sciences at the University of Oxford and attended lectures by Edward Burnett Tylor. Spencer came up to Exeter College in 1881 and graduated with a first class degree in 1884. He was then appointed Demonstrator with Henry Nottidge Moseley (who knew Pitt Rivers well, and also taught Balfour) and was asked to help transfer the Pitt Rivers Museum's founding collection from London to Oxford. Balfour came up to Trinity College, Oxford in 1882 (a year after Spencer) and graduated with a second class degree in 1885 before taking up the job working at the Pitt Rivers Museum. In 1891 he was appointed Curator of the Museum, a post he held until his death in 1939. Balfour and Spencer's correspondence, held in the Museum's manuscript collections proves the longevity of their connection.

For over forty years Balfour was the linchpin in the Museum —designing displays, acquired artefacts, battled with the University authorities to obtain funding, teaching the Diploma students, and carrying out his own research. Working full-time as a museum curator left Balfour very little time for fieldwork in the Malinowskian manner. In any case the kinds of questions that Balfour was interested in, including the development of particular technologies, lent themselves to being investigated by the survey method — spending short periods in the field, travelling over large tracts of country and researching a relatively limited number of questions. Balfour preferred 'facts' to theorizing, he thought that anthropological endeavour was divided between generalists and specialists. In his 1937 Frazer lecture at Oxford entitled 'Spinners and Weavers in Anthropological Research', he made it clear that his own sympathy was with the 'weavers' or generalists. It is clear that, for Balfour, theorizing took second place to collecting and communicating data. Both specialists and generalists, he argued, should know something of the other, but the latter kind of anthropologist would always have a grasp of the wider picture.

Balfour had to restrict his longer journeys to outside University terms, but he tried to take advantage of most Long Vacations to travel abroad. Up to 1914 he travelled to Finmarken and Russian Lapland, South Africa (three times), Mozambique, Zimbabwe, Australia, Indonesia, Singapore, Malaysia, Sri Lanka and India. He continued to travel in this vein until the late 1930s. He kept journals of his travels (they will hopefully be on-line at some point). They are full of direct observations, sketches, photographs, exact times, names and locations, but they often lacked anecdotal or personal commentary regarding the conversations he had or his thoughts and opinions on daily events and interactions.

However, Balfour's travels were not fieldwork as anthropologists after Malinowski would come to accept it: he did not spend very long in one place, and did not inquire in depth on any one topic (apart from, possibly, stone tools). Most of his travels were opportunities to obtain additional artefacts for the Museum rather than opportunities to obtain raw data (though, of course, for Balfour objects were raw data). His travels, although extensive, were largely busy tours involving day trips and excursions from the main towns and cities, accompanied by residents or in larger groups. Shopping in towns and buying artefacts from local stores was an important way for him to acquire material quickly and efficiently as he travelled around. There simply was not sufficient time to establish long-term relationships with the people who made the objects he bought. Instead, he had to rely on middle-men, and contributed to the thriving trade in local collectables at the same time. That said, there were occasions when he was able to acquire things more directly, from their original owners and users.

In addition to his own fieldwork, Balfour collected data (artefactual and factual) from other people. The main thrust of his research work was cumulative: he gathered ethnographic data and artefacts from hundreds of sources during his long life in an effort to build up as complete a picture as possible of the manufacture and use of particular types of object throughout the world. Basically Balfour was a mixture of fieldworker and armchair anthropologist, using whatever method best suited the job to hand. Survey work was common during the period in question, another Oxford anthropologist of the age, Robert Ranulph Marett, confirmed 'Touring, indeed, proves the ideal method of anthropological research' so Balfour was not alone. [Stocking, 1983: 110] Survey fieldwork was ideally suited to the limited time he had available and also his need to acquire a large number of artefacts fairly quickly. It also suited the exploration of the topics he was interested in: his interest in stone tool technology, for example, required large number of stone tools from different locations for comparative purposes. It can be argued that there are perhaps a limited number of research topics where survey work is the right or only methodology that would fulfill requirements, but Balfour certainly thought it had its place.

Third alternative: The field as laboratory, the anthropologist as team

Most listeners are probably familiar with the Torres Strait Expedition so I will not linger on the details. In March 1898 seven men set out for the Torres Strait islands off Australia. Their expedition had been sponsored by the University of Cambridge and was led by Alfred Cort Haddon, a distinguished natural scientist and ethnologist. The expedition also included scholars from the fields of psychology, physiology, medicine and linguistics. They spent approximately seven months on the islands. [Herle and Philp 1998: 8] Stocking has suggested that Haddon used the great eighteenth century maritime expeditions as a model for his team. [1983: 76]

Four years before the expedition set out, Haddon had described the essentials of fieldwork methodology:

Information must be patiently collected, critically examined and carefully compared and checked. In no case should the investigator theorise, it is the native's explanation that is required. [Haddon, 1894: 271 quoted in Herle and Philp 1998: 26]

Six years earlier he had doubted that he had the skill-set to be a 'proper anthropologist':

one requires wider knowledge and more versatile talents than I can lay claim to. He should be a linguist, artist, musician, and have an extensive knowledge of natural and mechanical science etc. [Haddon 1888 journal:108 quoted in Herle and Philp 1998: 25]

The expedition members used the scientific methodology of their particular disciplines to record as much measurable data as possible about the islanders, their lives and customs. Haddon was a great believer in accurate and detailed recording and the Expedition therefore amassed great quantities of data.[4] The team were expected to reach some sort of consensus about their findings, and used methodology which conformed more to contemporary laboratory practice than that of the field. Kuklick argues that, for Haddon, the 1898 expedition was:

a demonstration project, representing the optimal synthesis of laboratory and field practice. It would inspire anthropology to follow the lead of other natural history sciences, which had earlier abandoned the division of labour between collectors who toiled in the field and theorists who worked in libraries and laboratories. [Kuklick 1998: 150]

She defines the anthropological work undertaken by the Torres Strait expedition as 'primarily 'survey' work'. However, she also concludes that this was ideal for 'a team of specialists who divided investigative tasks along lines consistent with their individual expertise'. She sees this as the expected reflection of anthropology's development from the natural sciences where such collaborative research was common. [Kuklick 1998: 159] James Urry, by contrast, believed the expedition mixed 'detailed 'intensive' research with surveys of neighbouring areas'. In this way he sees the expedition as forming a middle ground between one model and another. [Urry, 1998] Stocking remarks that by the time the final volume of work arising from the 1898 expedition was published, in 1935:

the embracive evolutionary conception of anthropology that justified such a diffuse and multidisciplinary enterprise had been called seriously into question. Far from being accepted as a general methodological paradigm, the Cambridge Expedition became something of a methodological albatross for several early academic anthropologists who, alone in the field, attempted to carry on a similar range of inquiries. [Stocking 1995: 114]

Haddon's collaborative model of fieldwork was not one that British anthropology followed. This may have been because of the drift of social anthropology away from the natural sciences and towards the humanities or social sciences. Later anthropologists did not have the scientific training that anthropologists like Baldwin Spencer, Henry Balfour and Alfred Haddon had had. If anthropology was considered a branch of science, dealing with matters of hard evidence, it could be argued that it did not matter so much how the evidence was collected, so long that it was before it was lost forever - later anthropologists were seen as social scientists, dealing with interpretations, personalities and perspectives bringing the emphasis much more on the how. [Larson, pers. comm.]

Fourth and final alternative: Field site as home, total immersion

Few can ever have done anthropological fieldwork in quite the way that Frank Gillen did. [6] He was of Irish Catholic descent, born and brought up in South Australia. His formal education ended very early and he joined the postal service in 1867 at the age of twelve, as a messenger. He started work in the outback in 1875 as a telegraph operator and by 1892 he had been appointed the Post and Telegraph Station Master at Alice Springs and the local magistrate; one of the most, if not the most, senior authority figure then on permanent station in Central Australia. However, his background, love of gambling, investment in (worthless) gold mines and boisterous sense of humour meant that he was always a bit of an outsider in stuffy Australian bureaucracy. Another factor that contributed to his difference, was his sympathy and liking for members of the local Aboriginal community. He had some professional responsibility for them, being in charge of the mechanism by which rations were distributed in times of need, but the relationship between them was much deeper and was confirmed when he prosecuted a policeman who had murdered an Aboriginal man—although the police trooper was acquitted, Gillen's reputation with the Arrernte was sealed.

Gillen's anthropological fieldwork pre-dating 1894 has not yet been fully researched. There is little remaining evidence of it, apart from photographs held in the South Australia Museum, vocabulary lists written in 1875 and published in 1886 as part of Curr's survey of languages, fieldnote books now held in the University of Adelaide and Gillen's contribution to the Horn Expedition volume. [Gillen 1896] [Gillen, 1995: 4, 5] He is also known to have been a keen (if somewhat haphazard) photographer of the local population at the time. [Jones 2005] Because his home was in central Australia, and he had interested himself in Aboriginal affairs, Gillen was eventually very familiar with both general and restricted areas of Arrernte life.

In 1894 Gillen's life changed for ever when he met Walter Baldwin Spencer. Spencer was then the Professor of Biology at the University of Melbourne and had travelled to Alice Springs with the Horn Scientific Exploring Expedition as its photographer and zoologist. Gillen later contributed some twenty five pages of notes he had prepared on Arrernte life and customs to the Expedition volume, this must have been based on work he mostly carried out before he met Spencer. [Gillen 1896] The two men began an extensive correspondence in 1894 and later agreed to collaborate on an anthropological endeavour, Gillen in the field and Spencer at home in Melbourne. Their correspondence, or rather what is left of it, is now held in the manuscript collections of the Pitt Rivers Museum.

In 1899 Spencer and Gillen wrote:

It has been the lot of one of us to spend the greater part of the past twenty years in the centre of the continent, and as sub-protector of the Aborigines he has had exceptional opportunities of coming into contact with, and of gaining the confidence of, the members of the large and important Arunta tribe, amongst whom he has lived, and of which tribe both of us, it may be added, are regarded as fully initiated members. [Spencer and Gillen 1899: vii]

The final statement is controversial. They were trying to infer that they were privy to all the secrets of local culture, including that normally restricted from European knowledge. It is clear that, if this was true, it was so because of Gillen. He was allowed access to information and ceremonies that would normally have been forbidden to outsiders, women and uninitiated males because of the special relationship he shared with the local community. His reputation for fair dealing spread the length and breadth of the Northern Territory giving both men privileged access to other groups as well.

Gillen carried out fieldwork whilst continuing to work for the Telegraph Service. He worked closely with a number of Aboriginal men, mostly elders but also some younger men whom he paid in kind with tobacco, flour and clothes. Spencer and Gillen also carried out two further intensive periods of fieldwork together. The first was in the summer of 1896-7 when Gillen had 'arranged' that a cycle of ceremonies take place at the area they called the 'Engwura ground' [Angkwerre] next to the Telegraph Station. The second lasted for an entire year in 1901-2 when they travelled slowly through the whole of Australia from the Great Australian Bight to the Gulf of Carpentaria. These two fieldwork periods and Gillen's own daily fieldwork formed the basis for their first two books, Native Tribes ... (1899) and Northern Tribes of Central Australia (1904).

In 1896-7 Spencer and Gillen were as much 'participant observers' as nineteenth century anthropologists could be: during the two months of the cycle they lived in a wurley, close to the main ceremonial ground, and joined in some of the ritual action associated with the revelation of sacred things. Of this experience, they wrote, in Malinowskian mode:

... after carefully watching the natives during the performance of their ceremonies and endeavouring as best we could to enter into their feelings, to think as they did, and to become for the time being one of themselves ... [Spencer and Gillen 1899: 12]

In 1901-2 they pioneered sound recording on wax cylinders and shot ciné film under conditions of some hardship. The field journals they both kept during this journey (which are now in Australian archives) demonstrate their field methodology in detail but there is insufficient time to examine them in any depth here.

There has always been some debate about who contributed what in their partnership. It seems clear that Gillen provided the raw data, commentary and criticism whilst Spencer turned Gillen's work into polished prose suitable for an international audience. When both had worked in the field, Spencer's role was to synthesise their individual sets of fieldwork data into one seamless whole. Spencer was certainly more familiar with anthropological key texts (he often suggested that Gillen read them), and also with contemporary theories and practices; Gillen was the man with the inside, in-depth knowledge of Arrernte culture (the people they mostly worked with). He was always very deferential in his letters to what he perceived to be Spencer's finer mind and greater education but in fact he was a fine, if informal, writer and his great skill in obtaining and processing data from every informant made them the great anthropologists they became. There has also been debate about just how much Arrernte Gillen actually spoke (it is generally agreed that Spencer spoke very little). The greatest aid to being able to do empathic fieldwork is to speak the local language well. For a long time it was believed that Gillen was not fluent, but detailed analysis of his letters and, more particularly, his budgets of information, show that, in fact, he had a very good grasp of it.

Their final piece of joint fieldwork took place in 1903 when they travelled briefly to Lake Eyre. Spencer and Gillen intended to continue their anthropological partnership but this was prevented by Gillen's increasing ill-health, he died in 1912. Spencer continued to work intermittently on anthropological projects after this date, carrying out further fieldwork around Darwin in the Northern Territory. He finally died in 1929 during his last fieldwork trip, to Tierra del Fuego.

With hindsight their fieldwork can appear quite modern. They were very careful to cross-check the data they obtained with several sources. They also 'recorded information which went against their original hypotheses and modified their interpretation to take the new material into account'. [Mulvaney et al, 1997: 44] Their full documentation of their thought processes while in the field, or preparing their publications, through their letters and journals make it unusually clear what these processes were. In addition to collecting anthropological facts, they also amassed large collections of material culture, most of which is now in museums throughout the world, some in this very one.

In summary, Spencer and Gillen operated as a team, bringing different skills to the field but working in harmony and agreeing the results of their theorizing. Gillen believed strongly in the need for fieldwork, and for that fieldwork to be long term:

... without living amongst them and possessing their entire confidence [the anthropology] will be valueless — You know from our own experience how extremely difficult it is to make any headway [even] when you live amongst them and possess their Confidence ... [Letter 31, Gillen's letters to Spencer, Spencer papers PRM ms collections]

Frank Gillen knew his chosen anthropological subjects very well, he lived and worked alongside them from 1875 to 1899, and in 1901-2. His day-to-day interactions with the Aboriginal communities around Alice Springs were part of his reality. He was able to use this data to further human knowledge, to help one community (the Europeans in Australia, and Europe) understand another, apparently alien, culture. With his partner, Baldwin Spencer, he produced two publications which had a seminal effect on British and European anthropology and on the way that anthropology developed in the early twentieth century. Both Frazer, in the UK, and (to more lasting effect) Durkheim, in France, used their data as a source of much revolutionary theorizing. Malinowski believed that half of anthropology theory written between 1904 and 1913 was based on their writing, and all but a tenth heavily influenced by it. [Stocking 1983: 79] It is Stocking's view that Spencer and Gillen's work marked 'the beginning of fieldwork in the modern anthropological sense'. [Stocking 1995: 111]

Conclusions

This paper has given details of four different models for carrying out fieldwork employed by many anthropologists at the turn of the nineteenth and twentieth centuries. Each of these models was successful so far as their proponents were concerned in that they produced the results the anthropologists were looking for. One of the products of their fieldwork, in each case, was the acquisition of objects that eventually arrived at the Pitt Rivers Museum.

It was a surprise to me, though I am not sure why, when I first started work at the Museum, that so many of the classic social anthropologists of the twentieth century had collected material culture, and given it to the museum. Evans-Pritchard is the obvious example but also Marilyn Strathern and, of course, other representatives from [the Institute of Social and Cultural Anthropology] including Peter [Rivière] and Marcus [Banks] ... Thus the dominant twentieth century Malinowskian model of fieldwork was successful in ensuring material culture collections in the museum, although these collections were generally not discussed in the publications of their donors.

Each of the other fieldwork models appears to have been successful with regard to acquiring large numbers of artefacts. The difference, so far as the Museum is concerned, between the different methods probably lies in the depth of documentation provided for each object. As might be expected, acquiring artefacts at second hand is the least effective method of ensuring abundant information is also available and one can safely say that, by and large, the founding collection of the Museum is also one of the most poorly documented. Interestingly, though, another relatively poorly documented collection is that acquired and donated by Henry Balfour, who actually worked in the Museum. This is presumably because he often acquired objects away from their site of production and use, but also because his method of survey fieldwork left him little time to acquire detailed information. To be fair to him, it also reflects the fact that he may have felt he did not need to record all the details of artefacts as he himself could remember all about them. He worked at the very beginnings of professional museum practice and was not necessarily very consistent in his approach to documentation. Frances Larson has suggested that both Pitt Rivers' and Balfour's ways of anthropologising (going round a large region, or using secondary sources, to gather up lots of objects to work on at a later date) is only possible when objects are seen as a primary form of data, just like we see 'words' today (for example, in an informant's testimony). The object could be interrogated as though it could speak, it did not necessarily need background information as well. [pers comm] Thus the potential deficiences of their methodology were immaterial to them.

Survey fieldwork was carried out to different degrees by Balfour, Spencer and Gillen in 1901-2, and Haddon (as well as many other anthropologists like the Seligmans). It has long fallen into disuse as a model and is now disregarded as producing 'shallow' results. If comparison of different cultures is thought of as important, then survey work is an almost ideal method. It allows the researcher to identify the key questions, ask them systematically in a series of different locations and then work with the results. As regards acquiring collections for a museum, this method would probably ensure artefacts from the widest geographical area; depending upon the knowledge base of the collector, it may or may not produce well documented artefacts.

The collections of Haddon, Spencer and Gillen are well documented, indeed they are among the best documented artefacts of the period. This is because the authors not only knew the material culture of their chosen societies very well, but also because they took the time to publish information about the material culture in detail. They were helped by the relatively large number of man-hours devoted to field investigation in both cases (though the distribution of the man hours was different, with the Torres Strait having many men working simultaneously over a short period and Spencer and Gillen relying more upon one man working many hours, days and years in one location).

How would fieldwork documentation supplied by a Malinowskian anthropologist compare? The documentation for the Malinowski collection itself survives even if most of the artefacts are no longer in the Museum. The artefacts are very well physically described, but less well documented in terms of use, methods of production etc. This, however, may not reflect the level of documentation obtained by Malinowski but rather the fact that many of the artefacts were sold after his death by his wife via a third party (Audrey Richards). From experience of other Malinowskian fieldworkers, however, it can be safely concluded that the level of documentation associated with material culture entirely relies upon the anthropologists' interest in it. Where an anthropologist has collected as a side-line, little information will be obtained and passed on; where the anthropologist is interested, abundant information may be accrued. Malinowskian fieldwork is a method that can certainly be used to acquire well-documented artefacts from a single location.

The model of fieldworker in the twentieth century was very much one of 'anthropologist as [solitary] hero', fighting alone against his peers, seeking to convince his seniors that he is deserving of academic honours and funding. [Stocking 1983: 109] Collaboration and teamwork are surely options that should be explored more often. Working as part of a larger anthropological team is a much less common practice today though it is still the established method for other scientific disciplines. Spencer, Gillen and Haddon's work suggest that it can be a very fruitful method when applied to anthropology and ethnography. My current experience on this research project would confirm that. It allows several different research avenues to be explored simultaneously, brings several peoples' talents together and does not rely on a single polymath.

It must be clear by now that Victorian and early twentieth century anthropology was happy to employ a number of different models for fieldwork, and these all proved to have advantages and disadvantages of different kinds. Now comes the 'She would say that, wouldn't she' moment. It seems to me that one of the most effective methods is that employed specifically by one of our five examples: Frank Gillen. Making the site of your fieldwork your home, not for a limited duration but for many years, or for life, has many advantages. Not only do you have the time to really understand another culture in depth, and see as many facets of it as possible, but you are able to visit and revisit the same issues, gaining more knowledge each time. As a method for acquiring well-documented collections I do not see how it can be bettered. Over time you could acquire many more artefacts than would be possible on a single visit, to see changes in technology and usage over time, to buy more artefacts from individual makers or get a range of products from many different makers, to put right small deficiences in previous collections. I am sure that my audience can think of many disadvantages to this model, so I will stop now and hand over to them ...

Notes

[1] I will use the name Pitt Rivers throughout, for clarity, though he did not take this name until 1880.

[2] probably the edition of Notes and Queries used by them all, if it was not the 1899

[3] According to the accession records the 'specimens were acquired under a misunderstanding and were returned .. for shipment to ... Mexico in July 1947'.

[4] In his archaeological work Pitt Rivers was more active, leading many excavations in England, Wales and Ireland. He had spent short periods abroad; whilst serving in the Army (Canada, Malta and the Crimea) and had visited Egypt and several European countries on holiday or when visiting spas for the sake of his health but he never experienced a period of ethnographic enquiry in the field.

[5] Six large volumes of Reports were published from 1901 to 1935 of the Torres Strait Expedition findings. Unusually copies were also sent back to the islands for comment. Spencer and Gillen also showed the photographs and films they made in Central Australia to their informants, but there is no record of them distributing copies of their books outside the European community in Central Australia.

[6] Except, ironically, TGH Strehlow who worked in the same area, and professed to despise Spencer and Gillen's professional work.

Bibliography

http://www.adb.online.anu.edu.au/biogs/A120043b.htm?hilite=walter%3Bbaldwin%3Bspencer

British Association for the Advancement of Science (BAAS) Annual Report 1874.

British Association for the Advancement of Science (BAAS). Notes and Queries on Anthropology. London: Edward Stanford.

Brown, Alison, Jeremy Coote and Chris Gosden. 2000. 'Tylor's tongue: Material culture, evidence and social networks' Journal of the  Anthropological Society of Oxford vol XXXI no. 3: 257-276.

Gillen, F.J. 1896. 'Notes on some manners and customs of the Aborigines of the McDonnell Ranges belonging to the Arunta tribe'. In Report of the work of the Horn scientific expedition to central Australia. (ed.) W.B. Spencer vol. 4, pp. 162-186 London.

Gillen, F.J. 1968. Gillen’s Diary: The Camp Jottings of F.J. Gillen on the Spencer and Gillen Expedition across Australia 1901–2. Adelaide: Libraries Board of South Australia.

Gillen, R.S. (ed.) 1995. Gillen’s First Diary: Adelaide to Alice Springs March to June 1875. Adelaide: Wakefield Press.

Herle, Anita and Jude Philp. 1998. Torres Straits Islanders: An exhibition marking the centenary of the 1898 Cambridge Anthropological Expedition University of Cambridge Museum of Archaeology and Anthropology

Herle, Anita and Sandra Rouse (eds.) 1998. Cambridge and the Torres Strait: Centenary Essays on the 1898 Anthropological Expedition Cambridge University Press.

Jones, Philip. 2005. 'Indispensable to each other: Spencer and Gillen or Gillen and Spencer?' Collaboration and Language Strehlow Research Centre, Alice Springs (Occasional Paper 4) pp. 6-25

Kuklick, H. 1991 The Savage Within: The social history of British anthropology, 1885-1945 Cambridge University Press

Kuklick, H. 1998 'Fieldworker and physiologists' in Herle and Rouse op.cit

Malinowski, B. 1922. 'Argonauts of the Western Pacific' London: Routledge

Mulvaney, D.J.  and H. Morphy, A. Petch. (eds.) 1997 My Dear Spencer.  Melbourne: Hyland House.

Kuper, Adam. 1996. Anthropology and anthropologists: the modern British School' Third edition London: Routledge

Larson, Frances. 2006 'Anthropology as Comparative Anatomy? Reflecting on material culture studies during the late 1800s and the late 1900s' in Journal of Material Culture vol 12 (1): 89-112

Marcus. George. 1995. 'Ethnography in/ of the World System: The emergence of multi-sited ethnography' Annual Review of Anthropology vol. 24 (1995) pp. 95-117

Mason, Timothy. 'From Culture to Cultures : or Anthropology in the Desert' http://www.timothyjpmason.com/WebPages/Publications/Cultures.htm#sdfootnote64sym

Mason, Timothy. 'The Anthropologist's bagmen: Frazer, Spencer and Gillen, and the primitive in Australia. http://www.timothyjpmason.com/WebPages/Publications/Spencer_Gillen.htm

Mason, Timothy. 'Anchors in the desert sand: Sounding the cultural waters in Australasia' http://www.timothyjpmason.com/WebPages/Publications/Anchorage.htm

Mason, Timothy. 'Swallowing stones: The anthropologist's magician http://www.timothyjpmason.com/WebPages/Publications/Magician.html

Petch, Alison. 2008 'Walter Baldwin Spencer and the Pitt Rivers Museum' Journal of Museum Ethnography 20

Petch, Alison. 2007 'Notes and Queries and the Pitt Rivers Museum' Museum Anthropology vol. 30 no. 1 Spring 2007 pp. 21-39

Petch, Alison. 2007 'Opening the Pitt Rivers Museum' Journal of Museum Ethnography 19 pp. 101-112

Petch, Alison. 2007 [with Gosden and Larson] Knowing Things: Exploring the collections at the Pitt Rivers Museum 1884-1945 Oxford University Press

Petch, Alison. 2006.'Chance and Certitude: Pitt Rivers and his first collection' Journal of the History of Collecting 2006 18: 249-256

Petch, Alison. 2003. ‘Spencer and Gillen’s work in Australia - The interpretation of power and collecting in the past’ Journal of Museum Ethnography, No. 15 pp. 82-93

Petch, Alison. 2000. ‘Spencer and Gillen’s collaborative fieldwork in Central Australia and its legacy’ Journal of the Anthropological Society of Oxford 31/3 (2000) 309-328 [author][finally published in 2005]

Petch, 1998. '‘Man as he was and Man as he is’: General Pitt Rivers’ Collections.' Journal of the History of Collections 10(1):75–85.

Petch, Alison. 1997My Dear Spencer. Melbourne: Hyland House Publishing, Melbourne, Australia. [co-editor with H. Morphy and D. J. Mulvaney]

Peterson, Nicholas. 2006. 'Visual knowledge: Spencer and Gillen's use of photography in The Native tribes of Central Australia' Australian Aboriginal Studies vol 1 pp. 12-22

Spencer, W.B. 1928. Wanderings in Wild Australia. London: Macmillan

Stocking, George W. 1983 'The Ethnographer's Magic: Fieldwork in British Anthropology from Tylor to Malinowski' in Stocking (ed.) 'Observers Observed: Essays in Ethnographic Fieldwork' History of Anthropology volume 1 The University of Wisconsin Press pp. 70-120

Stocking, George W. 1987. Victorian Anthropology London: Macmillan

Stocking, George W. 1995. After Tylor: British Social Anthropology 1888-1951 Madison The University of Wisconsin Press

Urry, 1998 'Making sense of diversity and complexity: the ethnological context and consequences of the Torres Strait Expedition and the Oceanic phase in British anthropology, 1890-1935' in Herle, Anita and Sandra Rouse (eds.) op cit. pp. 201-33

AP, January 2009