Collecting the Past, Facing the Future: Revealing Collections’ Connections Using the Archives of the Wellcome Historical Medical Museum and Library Open Access

Precisely in order to facilitate these kinds of research, the Wellcome Historical Medical Museum and Library archives were comprehensively scanned and made openly available online in late 2017.⁷ Image digitization dramatically improved the archive's availability but did little to address its accessibility: No longer was it necessary to make an extended research trip to London and work through many boxes of paper records. But navigating the archive, digitized or otherwise, is not straightforward. Identifying references to any one object meant searching amongst hundreds of thousands of index cards and register entries, often heavily annotated, with multiple overlapping and—frankly— downright confusing different numbering and recording systems used by the museum and library over the course of time; documentation remained challenging and time consuming to negotiate, even for those with a good knowledge of Wellcome's institutional history. The archive has also been successively rearranged as a result of the dispersals processes, during which large quantities of material were gifted, loaned, or occasionally sold on to other museums.

All of which, combined with a well-worn metanarrative about the disorder of the museum collections at Henry Wellcome's death in 1936, gave rise to what seems to be quite a common misconception among museum peers and researchers that the Wellcome archive is either impenetrable or hopelessly limited, or perhaps both. Such narratives may have arisen as a means to translate the complexities of Wellcome's collections to museum-visiting public. They gained a higher profile through the 2003 British Museum temporary exhibition Medicine Man: The Forgotten Museum of Henry Wellcome, which used wonder and “forgotten”-ness as key themes alongside deep dives into the collection (Hawkins and Olsen 2003; Arnold and Olsen 2003). Similarly, in 2007 when Wellcome Collection opened as a cultural venue in London, its early marketing narratives also publicly mythologized the “mysterious” collections as largely undocumented, with little provenance information, mostly unpacked or unknown (for example, Kohn 2011, 2012). All of this was perhaps a convenient marketing tool for a young institution trying to sell a complex story.⁸ However, we follow Simon Chaplin (2005: 244) in gently challenging this narrative, observing that “our inability to respond to the Wellcome collections with any critical faculty other than wonder is as much the product of neglect as obliviscence.”

It is, of course, important not to downplay entirely the limitations of working with the Wellcome archives. Certainly there are limitations: backlogs of documentation exacerbated by documentation gaps, not least since acquisition frequently outpaced the capacity of the museum and library staff to keep up with accessioning. WHMM staff registered 16,000 objects in a year in 1936 in the Euston Road premises (Fig. 1), and as such they did not, or could not, go into a lot of detail. Often records simply copied whatever information was readily at hand, verbatim and without critique: typically a terse (sometimes highly inaccurate) one-liner from an auction house sales catalogue. The archive is limited, too, by who among the staff had the specialist knowledge needed to understand the huge range of material acquired—for material originating outside of Europe, these interpretations almost always reflected British or European anthropological perspectives.

There are some significant archival holes too, including one elusive sequential run of around 100,000 museum accession cards pertaining to material from ancient Egypt. And so the impression of a thinly described, poorly managed collection is both—in some senses, and at certain periods— true, and yet at the same time does a real disservice to the records’ creators and Wellcome curatorial staff who made successive attempts over many years to improve the comprehensiveness and detail of collections’ documentation, as evidenced by the size of the archive left behind from their endeavors. The scale of the inventory achieved across such a wide ranging and quantitatively enormous collections is still striking, while acknowledging the lack of provenancial detail and cultural specificity in many of the descriptions.

The archive can also be offensive; the language used is outdated at best, racist and inaccurate at worst. Entries for human remains, or for items which are considered secret or sacred by some communities, appear alongside records for mundane domestic objects. Yet contending with this raw—sometimes very raw—archive data is perhaps a vital discomfort on a journey toward a clearer understanding of the power dynamics at play behind objects’ acquisition, and the assumptions and cultural appropriations made in documenting, categorizing, researching, and managing museum and library collections at Wellcome over the course of the long twentieth century. Through the archive we can also show how Wellcome curators and staff drew on, contributed toward, and very occasionally challenged the dominant colonial narratives of the time, and demonstrate how these understandings constructed at the height of British Imperialism sometimes continue to influence how these objects are presented in museums today (Turner 2020; Sera-Shriar 2023). More hopefully, the archive can also present opportunities to engage in dialogue with originating communities on how objects might be better documented, displayed, and cared for in future.

This complexity, discomfort, and challenge brings a need to develop better resources to make the archive easier to navigate. Transcribe Wellcome began during COVID lockdown in 2020 as a pilot project coordinated by members of the Collections Information team at Wellcome Collection in collaboration with the Egypt Centre in Swansea. Transcribers included members of Wellcome staff, together with staff and volunteers associated with the Egypt Centre and other recipients of former Wellcome material. The project aims to transcribe and make the text locked within the hundreds of thousands of accession cards and museum and library register entries fully searchable and to open up new research entry points into the full richness of the archive.⁹

As a prototype, the Transcribe Wellcome database is available digitally on request but has not been publicly advertised; however, we very much welcome contact from researchers and museums working with former Wellcome material. Further information about the transcription project is available from the counterpart documentation website, as well as a bibliography and growing repository of research notes into the practices and people associated with Wellcome's museum and library.¹⁰

One role for transcription data is in making object numbers searchable. While transcription is still very much in progress,¹¹ the Transcribe Wellcome database enables users to search for a known object number to see where it appears among the batches of digitized index cards and museum register volumes, even where those records have not yet been fully transcribed. This hugely reduces the barrier for researchers, who would otherwise be required to navigate manually the different numbering systems, duplicate record series,¹² and extensive renumbering to find the complete registration history of an item. Renumbering and recataloguing material is part of any museum's practice history, though perhaps not always discussed with visiting publics, and museum documentation practices are deeply situated in particular knowledge-making contexts (Turner 2020).¹³ WHMM's formal registering started in 1899 for library collections and in 1914 for museum material, when the institution was operating through the staffing structures of Wellcome's pharmaceuticals business. As it transformed over decades into a private museum and library with professional staffing, its processes aimed to emulate established museum norms.¹⁴ Three different museum numbering series were successively implemented: original registration (R) number, accession (A) number, and new registration (R-year) number (Table 1). A single item may have a combination of all three types. Sometimes previous numbers are cross-referenced, but there are many examples where acquisition information is recorded under only one of an object's given identifiers. Therefore, it is necessary to research every number extant.

Doing this manually was complex and laborious (see flowchart, Fig. 2) and crucially, the Transcribe Wellcome database now allows searching for a museum number without needing to know which type it is—often a source of confusion, now and historically, because A-number and original R-number series were commonly used without prefixes (see Table 1). The database allows quick checking of any potential matches, plus any other numbers connected to that object. As more record cards and accession registers are transcribed, eventually it will become possible to search using descriptive terms as well, obviating the need to know an object's WHMM identifier in order to gain entry into specific material types, places, or names documented in the archive.

Material from the former WHMM is held today in over 120 museums and institutions. Tracing a particular object, or group of objects, held by a recipient museum involves negotiating this legacy of fragmentation through dispersal as well as fractures resulting from collecting. Our understanding of the Wellcome collections dispersals is largely a result of retrospective research by the museum workers involved in the final wave of transfers in 1978-83. Georgina Russell's articles remain the key source recording when material left WHMM and where it was redistributed (Russell 1986, 1987). Based on Russell's data, we have created Google Maps showing institutions that received collections from Wellcome (Fig. 3).¹⁵ Red pins indicate dispersals between 1936-78, and orange pins show the final wave from 1978-83. Dispersals data has also now been imported into Transcribe Wellcome for searchability.

The reasons for separating the collection were varied over the twentieth century, tied to shifting concepts about museums and academic study and specific local changes in WHMM's governance and direction. Henry Wellcome's museum collections were at their largest point in 1936, the year in which he died—said to comprise around one million objects and purportedly around five times the size of the Louvre in Paris (Skinner 1986). Due to its enormous size and the difficulty of managing it, the Trustees appointed to oversee Henry Wellcome's estate began a process of reshaping the collections. Parts started to be separated off and dispersed over the twentieth century through auction sales and some deaccessions, but mainly through transfers, gifts, and loans to other museums. The following paragraphs outline the overall pattern of collections movements, and African collections were dispersed during all of these phases (see Tables 2-3), reflecting the wide range of material and how it was perceived by decision-makers.

From 1936 until 1945, dispersals occurred mainly through auction sales. Auctioneers Alsopp & Co. held twenty-seven sales in a specially rented mansion, including items from Henry Wellcome's private home. Sales occurred for financial reasons, as the Trustees were liable for taxes on the value of Sir Henry's estate, which included the collections; the proceeds went toward paying these duties.

There were further sales during World War II, and some scrap disposals of wood and metal to aid the war effort. The Trustees had also decided not to pursue the broader-scope anthropology museum Henry Wellcome increasingly desired—a so-called museum of mankind. Such diversity was practically unworkable, and the academic landscape had shifted away from the encyclopaedic museums of 1900 toward greater specialization. Instead, the Trustees pursued a tighter focus on the history of medicine.

Further research is needed to explore who was making dispersal decisions regarding material perceived to hold medical connection. Museum staff reports from the 1940s, however, indicate that staff were tasked to separate out anthropological collections deemed medically related—and therefore suited for eventual display—and to crate material considered nonmedical and therefore less appropriate for retention.¹⁶ Crated items quickly exceeded available space in the museum's Euston Road basement, and when a storage facility at Willesden closed down in April 1945, agreement was made to temporarily store around 1,300 cases of material at the British Museum (Hall and Bembridge 1986: 46).¹⁷ The British Museum had agreed to store this material and assist in distributing it to other museums, in return for being offered first choice. They were then followed by other UK museums, including Liverpool, which sought to rebuild its collections after World War II. Through this route, the majority of anthropological (then termed “ethnographic”) collections, originating from across the Global South, were dispersed in ten rounds between 1949 and 1960. Some items were also offered to institutions situated in countries of origin. Recent analysis by Helen Mears indicates how these transfers, framed as a form of cultural return in Russell's dispersals summary (Russell 1986: S9), acted to reinforce colonial institutions and relationships (Mears 2023).

In 1960 Wellcome's Trustees officially purchased the collections from the commercial company, gaining direct control of their management, and a new wave of movements followed. Archaeological material was dispersed in the 1960s and early 1970s, predominantly ancient Egyptian and Sudanese collections—with this separation from other African collections reflecting a long trajectory in European academia. Ancient Egyptian material was rehoused with University College London's Petrie Museum in 1964. After selecting items for their collection, the Petrie acted as a redistribution center for material to other museums. Around 1971, representatives from other UK museums were invited to choose material, mostly sight unseen. These Egyptology dispersals were explored in a 2021 conference organized by Ken Griffin of Swansea University's Egypt Centre (see Griffin, this issue).¹⁸ From 1965-68, the University of California Los Angeles received large gift of material to what became the Fowler Museum at UCLA, which included extensive African collections (Hill 2006; see Jones and Forbes, this issue).

A final wave of rationalizations happened in the 1970s, as the Wellcome Trust refocused on its library, on research, and on funding academic history of medicine. British history of science and medicine had shifted toward textual and visual sources, with objects considered to have little research value.¹⁹ As the Wellcome Trust's then director, Peter Williams, later recalled, “museum exhibition was quite out of gear with the rest of our activities, historical and scientific. The collection did not generate academic history of medicine research. To exhibit it was expensive and the visitors were relatively few” (Williams 1994: 77). The Trust placed the remaining historic medical collections on permanent loan to Science Museum, London, to be displayed for public benefit (Bracegirdle 1986). This collection of around 117,000 objects has been cared for and managed by the Science Museum Group since that time, and Wellcome Collection works closely in partnership still.²⁰ Around 68,000 further “nonmedical” items were distributed to around eighty other museums by 1983, mainly through gifts, and the Wellcome Historical Medical Museum was then effectively wound up as an institution. Henry Wellcome's collections of paintings, historic books, and manuscripts were retained at the Wellcome Trust in the institute's Library. These have grown into the rich collections held today by Wellcome Collection, representing wide-ranging historical and contemporary perspectives on health.²¹

From the perspective of the museums that received objects from the former WHMM, independent researchers, and those representing the objects’ originating communities—often with a view to submitting requests for restitution, repatriation, or return—the already heavy challenges of tracing an item in the Wellcome archive are magnified further by museum practices during and following the dispersals process (Russell 1986; Sera-Shriar 2023). Much of this is due to the extended timeframes, the variability of record-keeping, the necessity of laboriously recreating descriptive information about thousands of objects in each recipient museum's documentation system,²² and the practice of using intermediary organizations to help with the distribution. As already discussed, ancient Egyptian material passed via the Petrie Museum, and much material at the Fowler Museum at UCLA apparently passed through the British Museum before being shipped to the United States.

Even simple practical considerations can impede the chances of successfully matching any one object to information in the archive: for example, items were commonly labelled only by means of a parcel tag tied on with string; these either fell off, or in some instances may even have been deliberately removed by recipient museums during sorting (Cooke 2018). Even where a label stayed attached, the details recorded are heavily abbreviated, almost cryptic, in the way they are set out. Also, objects were commonly sent on with little or no accompanying documentation, often at best a 6” x 4” index card sent separately from the objects themselves, sometimes decades afterward. In any case, there were multiple series of these index cards in existence, often with critical details recorded in only one series, which may or may not have been the version sent on.

The Transcribe Wellcome database has been used to prototype a virtual reconstruction of the dispersed Wellcome collections by linking object entries in the WHMM archive to contemporary online catalogue records at recipient museums and libraries (see Table 4). Where recipient organizations have recorded former Wellcome object identifiers in their own online catalogue records, or they are captured in digitized object images, links can be retrospectively created to corresponding entries in the digitized Wellcome archive.

Unfortunately the critical significance of these previous numbers to recreating the chain of provenance through the archive has not previously been well understood (even internally at Wellcome Collection) and this, coupled with the many different ways in which Wellcome identifiers could be expressed on objects and labels, means that this data, even where it is already available online, often requires considerable clean up and human reinterpretation before these retrospective matches can be made. The authors welcome contact from recipient museums about numbers of unknown significance and will gladly offer assistance.

We are already starting to use the transcribed information in Transcribe Wellcome as a tool for our own research and to answer external enquiries. The following sections share some samples of this work, and other contributors to this volume have shared additional examples.

In autumn 2019, a journalist from the Guardian newspaper contacted Wellcome Collection requesting further information about a bronze head from the Kingdom of Benin pictured in a black and white photograph available on the Wellcome Collection website (Fig. 4). The photograph dates from 1952, when the head and tusk formed the centerpiece of the problematically titled WHMM exhibition Medicine of Aboriginal Peoples in the British Commonwealth (Ashworth Underwood 1952). It was unusual for WHMM photographers to record museum registration numbers on their images, but here the photographer had fortunately noted the object number—R9165/1936.²³ The corresponding museum register entry documents that the head was purchased for WHMM from Sotheby's in London in July 1934, but the description does not mention the attached intricately carved ivory tusk, although the entry is stamped “cancelled” with “for UCLA” and “with tusk” added opposite in pencil but crossed through (Fig. 5). Curators at the Fowler Museum at UCLA, to whom both head and tusk had been gifted in the 1960s, were able to confirm through research during their Mellon African Art initiative that the tusk had been purchased separately from Foster's, another London auction house, in 1931, and only later attached to the bronze head, probably for the 1952 exhibition.²⁴

Alongside the entry for this particular bronze in the Wellcome register were another four heads from Benin: two of wood, painted gold, purchased at the British Empire Exhibition held at Wembley in 1924; another of brass from the same Sotheby's sale in 1934; and a wooden head inlaid with metal strips, bought from a William Hewlett, who gave an address in north London. It was becoming clear that the item(s) that had interested the Guardian might prove only the tip of an iceberg when it came to Wellcome's acquisition of objects from Benin.

Since lots from one auction sale were commonly also accessioned consecutively, a targeted transcription approach seemed likely to reveal more Benin material among the former Wellcome collections. No further Benin purchases from the 1834 Sotheby's sale were identified, but Wellcome's agents had successfully bid for no fewer than seventeen Benin lots in the Foster's sale in 1831. Whereas the Foster's sale catalogue in Wellcome's possession merely notes the sale of “property of a gentleman” for these lots, the copy at the British Museum is annotated with the name of the “gentleman” concerned: Captain Herbert Sutherland Walker, a known member of the “punitive expedition.” Correspondence files in the Wellcome archive also indicated that a considerable quantity of material from Benin had been sold to Wellcome by a former District Officer for Benin, Hugh Nevin Nevins (1888-1868), including many stone axe heads (Basu 2020), a large number of ceremonial items, and significant amounts of ivory. Further Benin material still was identified through reverse matching of known museum recipients’ catalogue records to entries in the WHMM archive, followed by transcription of sequentially filed accession cards. Our research has now established a total of 515 Benin objects known to have been acquired and subsequently dispersed by the WHMM, meaning that Wellcome formerly possessed one of the largest accumulations of Benin material in Europe.²⁵ Of these 515 items, 32 objects have clearly documented connections to officers and others present during the attack on Benin City by British forces in 1897. Current museum locations have been traced for 232 (45%) of these objects. While our knowledge of WHMM's acquisition of material from Benin remains partial and provisional, an open knowledge exchange among institutions has been critical to the success of this provenance tracing project, since crucial documented links to the looting in Benin in 1897 evident from the Wellcome archive have often been lost through the processes of dispersal and reregistration at recipient museums. All of the transcribed Wellcome provenance information has been shared with Digital Benin,²⁶ and we continue to work with recipient museums to reconstruct these institutional chains of provenance.

In the past, assumptions have sometimes been made that much or most material culture from the African continent in Wellcome's collections came directly from anthropological field collecting or colonial officials. An example of this type of collecting follows, but in actuality a significant proportion was purchased from London auction houses. Frances Larson's (2009) study and prior PhD thesis (Knight 2004), both conducted before image digitization and transcription of records, indicated that the auction market was a significant collecting route. But the true extent of auction purchasing was challenging to demonstrate effectively using predigital research methods.

The data in Transcribe Wellcome gives provisional figures on the proportion of material acquired from different routes. We’ve currently transcribed just under 8% of the estimated total number of accession and registration entries for the WHMM, comprising 25,690 individual entries.²⁷ Of entries so far, 38% of these (9,769) come from private sources or the source is unknown, and 62% (15,921) are auction purchases (Fig. 6). Looking specifically at the auction purchases, a few London auction houses dominate: 29% (7,462) from Stevens’, 19% (4,947) from Sotheby's, Glendining 5%, Foster 2%, then a long tail of other auction houses (see Fig. 7 for further detail of this breakdown). The dominance of Stevens’ (47% of the material purchased at auction) and Sotheby's (31%) has been consistent throughout our four years of transcription—although the exact proportions have varied, and the dataset comes with some heavy caveats regarding source data duplication bias and could also be skewed by the time period of the accession cards transcribed to date.

As we have seen in the Benin material, auction house catalogue information can sometimes be a method to trace back through named sales and collectors to understand acquisition routes. Yet often auction catalogue information offers nothing but a date of sale, a brief description from a European gaze, and a dead end, obscuring further details about how material was dislocated from its place of origin and ended up in London. In the future, it should be possible to study, at the broad overview scale, when and where material from a particular part of the world was sold to Wellcome from which auction houses. This question can be explored in more concrete terms for particular geographies and time periods, revealing the layers within the whole collecting enterprise and how exactly Wellcome's museum and library participated in the market for circulating artifacts in the first decades of the twentieth century.²⁸ Such quantitative views ask what may be revealed through studying breadth of collecting as well as depth—complementing single object biographical case studies which, by their nature, follow deep stories. Wellcome's scale of collecting was so large that it reflected what was appearing on the market at a given time: a global market for artifacts shaped by the sociopolitical contexts of colonial and imperial activity.

The Transcribe Wellcome database also provides customizable ways to search for, tabulate, and visualize connected groups of material, where material from the same source may have been separated to different recipient museums. One example relates to museum objects collected for WHMM by Mervyn D.W. Jeffreys (1890-1975), a South African anthropologist and colonial administrator in Nigeria and Cameroon. According to research by historian Diana Marsh (2015), Jeffreys became associated with the Museum from 1926 through friendship with WHMM's curator at that time, Louis W.G. Malcolm (1885-1946). Both men were anthropologists by background and had associations with Cameroon's colonial administration (Marsh 2015; Clark 2016). Malcolm's curatorial appointment reflected Henry Wellcome's intent in the 1920s to focus on an anthropological understanding of medicine. Jeffreys was working as a government officer in Southeastern Nigeria and studying for a PhD through the University of London and was a collector of artifacts from the regions where he worked. Malcolm began funding Jeffreys to collect material for the WHMM in southern Nigeria and Cameroon, and large shipments were sent to London from around 1827 to 1838. The material originates from Aro, Ibibio, and Efik peoples in southern Nigeria and Cameroon and include ceremonial garments and items of clothing, carved figures and ritual objects, and indications that some human remains were also acquired. WHMM archives contain over 600 pages of correspondence on this collecting, including handlists of connected groups of material. Many items didn't stay at WHMM for long: some were offered as gifts to other institutions as early as 1842, four years after Jeffreys stopped collecting. WHMM's reasons for separating the material likely included intent to retain only medical items, and attempts to give “representative” selections to different institutions. Objects collected by Jeffreys were gifted to the Pitt Rivers Museum in Oxford in 1842 (potentially over 3,370 objects); the British Museum in London in the 1850s (another 3,380 objects); UCLA in 186668 (now the Fowler Museum at UCLA); some were placed on longterm loan to the Science Museum, London in 1878; and the dispersal location of the others is still to be determined.

Archival records can help to restore information on cultural and geographical origins that may not have been passed on at the time of dispersal.²⁸ One illustrative example is an Igbo kola nut bowl (ọkwa ọjị) used within the Igbo rituals of hospitality, now at Pitt Rivers Museum (Fig. 8).³⁰ This decorative carved wood bowl holds kola nuts and condiments offered in ceremony to welcome guests. Acquired by Jeffreys in Nigeria for WHMM, it was subsequently gifted to Pitt Rivers Museum in 1842. Their catalogue does record the Jeffreys connection and also notes an “other number 5122,” which Transcribe Wellcome established as WHMM accession number A1522.³¹ The A1522 inventory card at Wellcome—written based on Jeffreys’ own handlists and descriptions—recorded that the bowl is Ututu in cultural origin, and that Jeffreys acquired it in Arochukwu in 1830. This information in Wellcome's records wasn't passed to Pitt Rivers at the time of the gift, and these small amounts of new research emerged through recent dialogue around loaning the bowl.

Material associated with Jeffreys that is traceable using his name in the Transcribe Wellcome database so far represents only a small proportion of the overall assemblages, but already the complexity of the dispersals are becoming evident, with single Wellcome accession groups of common place and time of origin being split among multiple recipients. Identifying material associated with a “collector” can digitally remake connections between physically separated objects—particularly where recipient museum systems have incompletely documented the connections. But going forward it will also be possible (and essential) to move beyond this privileging of collectors within records themselves and instead locate material by cultural origin, geography, and theme. We can also better scrutinize, through the detail of transcription, how collections were framed and contextualized within the twentieth century museum and how these understandings shifted over time, and use these understandings to reflect on how we might improve our contemporary collections information practices: for example, in more sensitively contextualizing Wellcome's collecting legacies or in seeking to make material discoverable and searchable using terminology chosen by originating communities.

Through our research we are also getting a more nuanced picture about the practices of staff within the museum and library, in displaying and interpreting material as well as making decisions about dispersals. Most written histories of the Wellcome Historical Medical Museum and Library usually name Henry Wellcome as founder, financier, and figurehead, and the main curators who worked for him. Less is understood about the junior curatorial staff, casually employed specialists, local agents, or practical staff who carried out the work. Larson (2008) notably does feature some junior staff and agents involved in assembling the collection, and John Symons (1883) characterizes the staff of Wellcome's library; both have been key sources for our own work. While Larson focuses on the museum story, our work is also indicating the porosity between museum and library collections that challenge expectations of what constitutes a “museum” collection. For example, there are Coptic manuscripts originally accessioned to the library (now at Petrie Museum, UCL)³² and an ancient Egyptian funerary portrait from a woman's mummified body (now at The Egypt Centre, Swansea)³³ which was accessioned and registered as a “painting” in a distinct and separate knowledge system from the other museum collections.

Digitization also makes it possible to research a wider range of WHMM staff more quickly and build up a more holistic picture of their activity across museum and library structures.³⁴ Junior staff were mainly White, educated-class men, but employees at times included White women, Jewish émigrés and, over time, a handful of men of color of South Asian and East Asian heritage whose language skills and cultural identities were employed for the museum's collecting aims.³⁵ These staff often act as “invisible technicians” in the museum (to use Steven Shapin's 1888 sociological term), their work mainly subsumed under the authorship of main figures and harder to reconstruct without deeper archival engagement.

It is important to recognize when examining the constitution of the WHMM staff that it never directly employed, to the best of our knowledge, anyone who was of Black African or African diasporic heritage to work on any of its collections. Looking specifically at the staff who catalogued and interpreted African objects in the 1830s under curator Louis Malcolm—describing, naming, making decisions about what may be displayed or dispersed—these were White women graduates at the start of their careers, educated into the turn-of-twentieth-century colonial model of anthropology and ethnography. S. Rosa Burstein (1887-1871) arrived at the museum in 1828 after studying folklore and anthropology in Wales and Oxford. She oversaw the Ethnographic section and the other women working on these collections.³⁶ In 1852 she curated the exhibition Medicine of Aboriginal Peoples in the British Commonwealth; its highly colonial presentation remains underhistoricized but appears important in ascribing European-gaze medical interpretations onto material.³⁷ Margaret Rowbottom (1808-88) became an assistant in the Ethnographic section in 1933 after a bachelor's degree in Physics. She sorted, categorized, and catalogued “ethnographic” objects, later specializing in European history of science, and remained at WHMM for her whole career.³⁸ Mary (Molly) Borer (1906-95) was a UCL anthropology graduate employed as a scientific assistant in the Ethnographic department from 1928-35. Borer viewed objects at auction sales, marked up catalogues with items of interest, and identified and registered items bought from sales. She worked on material from the Democratic Republic of Congo acquired from the Henry Pareyn collection auction sale.³⁹ Joan Lillico (1911-2001) graduated in Archaeology and Anthropology at Cambridge UK in 1935 and joined Wellcome's museum the same year, aged 24, cataloguing “ethnographic” collections. Our recent research, matching handwriting in staff monthly work reports with inventory cards, has established that Lillico catalogued much of the material collected by Mervyn Jeffreys in southern Nigeria, based upon Jeffreys's own descriptions and handlists.⁴⁰

In another brief example from in the museum's history, we have identified a three-letter classification system for anthropological collections as being introduced in 1958 by Klaus Wachsmann (1907-84) during his tenure as “Keeper of Ethnology”; this has allowed more specific dating of inventory cards and the information contained within them.⁴¹ Wachsmann, an ethnomusicologist, was previously curator at the Uganda Museum, Uganda and left WHMM for a professorial appointment at UCLA (Nannyonga-Tamusuza n.d.), highlighting the continuity of the intellectual traditions behind these museum establishments and running through the interpretation of their collections.

The intent here is not to venerate or rehabilitate particular staff involved, nor to downplay any of the work undertaken by them. Rather, it is to nuance the picture of who was actually undertaking the documentation and interpretation work and better situate the knowledge-making practices: understanding who was making knowledge, about whom, and for whom. As Turner (2020: 161) argues, “attributing names to documentation potentially allows others to trace the intellectual roots of those thought processes,” and consequently are as important to establish historically as well as in contemporary museum and archive practices (Light and Hyry 2002).

Image digitization, transcription, the Transcribe Wellcome database tools, and the building of research networks with peer museums across the dispersed collections is enabling a new level of access and possibilities for alternative modes of analysis around how Wellcome's collections were collected, understood, displayed, interpreted, mobilized, and dispersed. The collections overseen by the Wellcome Historical Medical Museum and Library resulted from, and were shaped by, the complex federation of business and research organizations bearing Wellcome's name, which themselves shifted in form, mission, and audience over the course of the twentieth century. Collections were gathered together through different routes—the auction market, dealers and agents, academic contacts and field collecting, colonial administrative networks—and were processed (accessioned, registered, often then registered again, researched and exhibited, dispersed) through many different hands within the institution and for many different purposes. Just as there is no single “collector,” there is no unifying rationale for the whole collection or single stable meaning across time (although this has been imposed upon it at times for particular ends). Similarly, a reductive view of “African collections” has been imposed, but this disguises a huge heterogeneity of material, acquired from different regions and with different intentions for how the collections might then be deployed. The WHMM's attempts to impose knowledge and ordering changed over time, shifting in terms of value assigned, but always inherently colonial and Western-centric in its stance.

Tracing particular objects and their histories requires the researcher to be able to navigate these multiple and parallel tracks of relationships, context, and time, locating and following the archival layers that hold museum-generated knowledge of this material. As custodians of the archives but no longer the objects, we recognize our institutional and professional responsibility to explore the archival systems and to understand these “heterogeneous assemblages of people and material technologies that manufacture knowledge in museums” (Turner 2020: 185). This knowledge is acquired through detailed readings, transcription, questioning, discussing, corresponding, and comparing between museums the labels, dog-eared cards, and scruffy carbon-copied paperwork. It is all in the detail, moving carefully and systematically along those rough archival surfaces, sharing knowledge generously, making explicit the “archival aporia: gaps and uncertainties that open possibility even as they hurt” (Nowviskie 2019), and “posit[ing] the record [and their connected artifacts] as always in the process of being made, the record opening out of the future…[an] open-ended making and re-making” (Duff and Harris 2002: 284).

Recognizing the heterogeneities of African collections and their meanings, and implementing new digitized collections infrastructures and practices which challenge inherited collections legacies is foundational, large-scale work that will never be finished. Transcribe Wellcome and our work on the institutional archives is one strand of activity within Wellcome Collection's larger commitment to develop more inclusive practices, including through exhibitions, new collecting, and audience-facing work.⁴² Although Transcribe Wellcome is early stage, it is already starting to make the archives a more effective research tool for researching objects displaced and disrupted by structures of colonial violence, museum categorization and subsequent scattering. Bethany Nowviskie, drawing on the concepts and philosophies of Afrofuturism, describes “an orientation towards past culture as future-oriented technology: codes to crack, tools to use, and collections to transform.” This aligns with our ambition for Transcribe Wellcome's future, “not as content to be received but as technology to be used… not as statements about what was, but as toolsets and resources for what could be” (Nowviskie 2019, italics in original): for the Wellcome Historical Medical Museum and Library archive to become a site of active endeavor and agency, debate and challenge, hope, healing, and respect.