Journal of Museum Ethnography, 15, pp. 109-114

Alison Petch, Pitt Rivers Museum, University of Oxford

We are very grateful to the editor of JME and the chair of the Museum Ethnographers Group for agreeing to allow us to re-publish this article on this website.

Documentation in the Pitt Rivers Museum: the contribution of Sir Francis Knowles (1886-1953)

I was particularly struck, when reading Len Pole’s evaluation of our recent DCF cataloguing project, which was completed in 2002, by this sentence: 

'The assumption has been made by the present [retrospective computerisation cataloguing] team that previous curatorial input has been sufficient to validate the entry details.'

This statement is true—it has been assumed in all the recent documentation projects that the majority of information compiled by most members of staff at the museum since its founding in 1884 is accurate for that point in time. Not only is this a belief upon which projects like the retrospective computerisation of catalogues was based but it has been tested by other members of staff when examining the same objects more recently and in general it has been found to be true; the vast majority of information in our documentation system is accurate, although some of it, of course, is now historically outmoded.

I have often wondered why it is that Pitt Rivers Museum historical documentation appears to be so much better than a lot of other museums in the world, ‘it has long been recognised that the quality of the ethnographic records at the PRM are second to none, world-wide ..’ [Pole, [evaluation] 2002]. In the end I believe (though I cannot fully substantiate this claim) that this is due to two factors: first, that the museum has always been lucky enough to attract a very high calibre of staff who work at the museum for a long time; and second, that these staff have always had a very keen interest in the processes and techniques of documenting objects.

The Pitt Rivers Museum is lucky in that ever since 1884 staff have completed quite full descriptions of each object in the accession registers and other forms of documentation. Two card index catalogues of all objects, divided by provenance and typology, were later prepared using the accession registers as a base. This initial information has been augmented by members of staff as an object was researched or examined. The systems for accessioning and adding information were established early and continued to be carried out meticulously by all documentation staff so that even today more or less the same procedures are carried out by documentation staff as was done in 1884 (taking into account change due to the advances of technology and museum registration standards). In the very first annual report of the Pitt Rivers Museum, in 1888, it was reported:

'About 1500 of the specimens upon the screens have been fully catalogued: that is, a label upon each specimen refers by a number to a separate card, upon which is written an exact description of the specimen with measurements, where necessary, location and all data as well as references to literature.' [PRM Annual Report, 1888: 30]

There is proof of one long-term member of staff’s interest in the process of documenting an object: Beatrice Blackwood published The Classification of Artefacts in the Pitt Rivers Museum Oxford in 1970, a seminal publication which is still a basis for the Pitt Rivers Museum and other museums' ethnographic thesauri. It is possible to see this as a unique publication as few other members of staff, until recent years, have published their thoughts on the documentation process but I do not believe that it was. I think that it was the public part of an on-going internal debate about the documentation of objects.

In April 2002 I was augmenting the database entries for tobacco pipes by adding the information compiled in one of a series of detailed card catalogues. These card catalogues were to a bit of a mystery to staff working at the museum now. They are not part of the 2 main card catalogue systems and do not contain quite the same type of information. There are twelve of these catalogue boxes in total and they concentrate on a small number of types of things—amulets, pipes, footgear, lamps and medicine.

I had previously added the detailed card catalogue for footgear to the computerised database and had not found any information about why the catalogue had been compiled, or who had compiled it. However the catalogue did contain a great deal of new and detailed information about each object including very precise physical descriptions and construction and wonderful drawings. I thought that it was well worth continuing to add the information on these cards to the database as time allowed.

About half way through the card catalogue for pipes I came across two cards which suddenly made at least some information available about who had drawn the illustrations and compiled the information. The cards were headed ‘Key to Catalogue of Smoking collection in Pitt Rivers Museum (F.H.S Knowles 1939)’. This meant that the detailed card catalogues were a precursor to the two comprehensive card indexes, divided by type and provenance (and principally worked on by Penniman and Blackwood starting in 1939). It also explains why no object has been included that was received later than the 1930s.

F.H.S. Knowles was Sir Francis Howe Seymour Knowles (1886-1953). As Blackwood and Penniman related in his obituary he was one of the first two students to be awarded the Diploma in Anthropology and in 1909 was appointed Assistant to Professor Arthur Thomson, specifically to carry out teaching and research in physical anthropology, the first post of its kind in the University. In 1912 Knowles began fieldwork on the Iroquois Reserve, Ontario, and from 1914 to 1919 he held the post of physical anthropologist to the Canadian Government.  Typhoid forced him to give up his career as a physical anthropologist so he took up the study of the methods used to make stone tools and weapons, working through the collections in the Pitt Rivers Museum. He arranged a number of exhibition cases for the Pitt Rivers Museum, showing the techniques employed in stone-working from the earliest times to the gun-flint-makers of Brandon. It is obvious, however, that in addition to his principal interests in stone tools and techniques and weaponry he also contributed to museum documentation in other ways and it is these contributions that interest me at the present time.

Knowles is a bit of a lost figure in Pitt Rivers Museum mythology (like Ernest Seymour Thomas about whom little was known, despite his massive contribution to museum documentation, until Linda Mowat published a short article about him and his life in the Friends newsletter)*. Most people think of Balfour and Blackwood when they think of early documentation at the museum but in fact a very large number of people have contributed to this area either in funded jobs as assistants to the Curator, Assistant Curators or as volunteers. More recently the museum has been lucky enough to attract both research funding and Designated Challenge Fund monies to supplement the small number of permanent documentation and curatorial staff.

A detailed perusal of the Museum’s annual reports for the relevant period do not entirely remove the mysteries concerning the detailed card catalogues but they do explain some things:

'… Sir Francis Knowles very kindly volunteered to make a card-index of a collection of more than a thousand specimens illustrating primitive methods of illumination, which I have presented to the Museum. This work occupied him during many months.' [PRM Annual Report, 1932: 2]

'… The card-catalogue of the collection of Lighting-appliances, presented by me in 1932, was practically completed by Sir Francis Knowles, who also commenced a card-catalogue of the series of primitive Surgical and Medical appliances. I am greatly indebted to him for his valuable voluntary assistance. The cataloguing of the extensive comparative series of Footgear was kindly undertaken by Miss A. Powell-Cotton and is still in progress. …' [PRM Annual Report, 1933-4: 2]

'…. Sir Francis Knowles and Miss Powell-Cotton have kindly continued to serve as volunteer workers in the making of card-catalogues.' [PRM Annual Report, 1935: 1]

It seems from the Annual Reports that Knowles continued to work on the card catalogues at least until 1940. In addition it may have been that the Amulet detailed card catalogues boxes (5 in all) may have been prepared a good deal before the other catalogues, though the entry in the Annual Report for 1908 is not clear enough to be certain:

'A card catalogue of the very extensive collection of Amulets, Talismans and Magic-appliances, has been commenced by Miss B. Freire Marreco, a former Diploma student who has kindly devoted a considerable amount of time to this work.' [PRM Annual Report, 1908: 59]

In the 1909 Annual Report it was stated that Barbara Freire Marreco was continuing with the Magic card catalogue which was nearly completed, helped by 3 Japanese students [PRM Annual Report, 1909: 68]. In 1941 it was reported that ‘Mr Gibbs has greatly improved and developed the catalogue of our many amulets…’ [PRM Annual Report, 1941: 6] although of course it is not clear whether this was the same catalogue that Freire Marreco had worked on so many years earlier.

Knowles’ two cards of background information about the compilation of the detailed card catalogue of pipes makes interesting reading. The full text reads:

‘Key to Catalogue of Smoking collection in Pitt Rivers Museum(F.H.S Knowles 1939)Aim.(1) Ready identification of the specimen from the catalogue card(2) Concise description of specimen(3) Salient measurements(4) All information available and all references to literature concerning the specimen Method.(1) A drawing of the specimen to aid in speedy identification(2) Concise description of shape, material, colour and ornamentation.(3) Measurements of length of pipe, height of bowl, and outer width of rim (on the reverse of this card there is a diagram of a pipe showing exactly how and where these measurements were taken)(4) Identifying marks, chips, breakages, parts missing etc(5) Tribe, locality, native name, collector and how acquired noted(6) References to literature and notes and extracts referring to the particular specimen written on card, together with any other information available

On the reverse of this card is a ‘Diagram of composite tobacco pipe to illustrate descriptive terms and measurements used in making the catalogue of the Pitt Rivers smoking collection F.H.S. Knowles Feby 27th 1939’

The second card gives the numbers of cards compiled by Knowles (a total of 627).

The two card introduction by Knowles seems to me to be an exemplary setting up of a documentation and cataloguing project for the late 1930s. So far as I can ascertain the pipe catalogue was comprehensive and obviously carried out with the objects in front of the cataloguer. There are a few objects contained in these catalogues that had evaded the more customary accessioning processes and there are many objects for whom the only detailed information is given on these cards.

It is unlikely that Knowles worked on the other detailed card catalogues for lamps, amulets and footgear and the handwriting on these is very different, some of the handwriting could be Balfour’s or Thomas’ but other handwriting sources are unknown although one suspects that they were either more junior paid staff or volunteers working under Balfour, who died in 1939.

Each of the cards in the 12 boxes is set out in the same way. The cards are standard card catalogue size (152 x 101 mm) and printed with the same subdivisions:

Group: which refers to the type of object e.g. Footgear,Division: which refers to the sub-division with the group e.g. Sandals,Class: which refers to the sub-division of the Division e.g. Plaited,Number: which refers to any number associated with the object, for example a collector’s number,Description: which is a large section giving a physical description of the object and the measurements etc,People: ‘Tribe’ or cultural group name,Locality: Geographical provenance,Native name: (self-explanatory),Collected by: (self-explanatory),How Acquired: by the Pitt Rivers Museum,References: Publications etc.

I cannot tell whether these cards were printed in this format specifically for the Museum and this exercise (they were not used for the comprehensive card catalogues which are written on plain card catalogues) or whether they were commercially available for museums’ use. At the bottom of the cards  is printed ‘VS 186 The Globe Werniche Co Ltd London’, from whom the cards were obviously obtained, or the printers. I do not recall seeing these cards elsewhere (but I would be pleased to hear from any other museum which has similar ones).

I do not think it a coincidence that the types of objects covered by the detailed card catalogues are the artefacts of particular interest to Henry Balfour who published several detailed articles about them in learned journals. However Henry Balfour does not appear to have compiled more than a few of these card catalogues himself. I believe that it is most likely that he instigated the system, chose the topics to be considered (he may have intended this system to eventually cover all the types of objects—and therefore all the objects—in the Museum, and started with those objects closest to his own research interests). Other staff then (possibly under his instruction) worked on separate types of objects. If I am correct in this assumption then the likely date for this work is during the 1930s culminating and ending with Balfour’s death on 9 February 1939 (and the start of the second World War a little later). However there are earlier, vague, references to card cataloguing in the annual reports and it may have been that sporadic attempts at card cataloguing commenced much earlier (with the amulets work in 1908) and continued until 1939.

It may be that the detailed card catalogues are a definite precursor to the comprehensive typological and geographical cards compiled during World War II. The detailed card catalogues information could not have been assembled without the object being in front of the cataloguer because a great deal of the information is concerned with the artefact’s physical description, construction methods, materials and measurements. One of the reasons why museum staff worked on preparing the comprehensive card catalogues during the war was because they could not easily access the actual artefacts that had been left in the Museum because of the problems of lighting and air-raids. Thus these card catalogues were prepared from paper records and not the objects and the information that could be provided was necessarily different.

I have written this short article for several reasons: to celebrate the work of one of the unsung heroes of Pitt Rivers Museum documentation, Knowles; to shed some light on documentation practise in the 1930s, at least in one ethnographic museum, and in order that current documentation practises, processes, theories and successes can be seen in the light of previous generations’ work and achievements.

Alison Petch

Registar, Pitt Rivers Museum


Blackwood. B. 1970. The Classification of Artefacts in the Pitt Rivers Museum Oxford. Occasional Papers on Technology, no. 11.

Blackwood, B.  and T.K. Penniman,  ‘Obituary : Sir Francis Knowles: 1886-1953’, Man, June 1953, n. 127, pp. 88-89.

Mowat, L. 1995. ‘The Assistant Curator who walked through fire: A tribute to E.S. Thomas’. PRM Friends Newsletter no.14 (October 1995)

Pole, L. 2002 [not published] Evaluation of computerised data base at Pitt Rivers Museum Oxford

* Note original punctuation changed here as the original was inaccurate and misleading.

[Transcribed by Alison Petch as part of the Rethinking Pitt-Rivers project June 2010]

prm logo