Making ‘slave ownership’ visible in the archival catalogue: findings from a pilot project

ABSTRACT

This article outlines a pilot project aimed at making ‘slave ownership’ more visible in archival catalogues. The project began with the premise that it is incumbent upon academic communities and record-keepers to make known Britain’s slaving past and the ongoing legacies of that past as part of a drive to dismantle systemic (and often invisible) racism across the sector. Specifically, it explored different ways of cross-referencing the Legacies of British Slave-ownership database (https://www.ucl.ac.uk/lbs//) with the Special Collections catalogue at the University of St Andrews with a view to updating the information provided in the latter. Six methods of identifying matches were trialled, each of which is presented and reflected upon here. Although some methods produced more matches than others, the collective results point towards the need for a multifaceted approach. Our findings also raise important questions about types of involvement in enslavement (direct/indirect), how different levels of certainty regarding the identity of certain individuals might be indicated in the record, and how collection-level and item-level descriptions might be updated. The project also highlights how our own assumptions about who is — and is not — likely to have ‘owned’ enslaved people can influence our very methods for uncovering those people.

KEYWORDS:

• Slavery Compensation Act
• ‘slave ownership’
• enslavers
• descriptive practice
• cataloguing
• Legacies of British Slave-ownership database

Introduction

It is incumbent upon academic and recordkeeping communities to acknowledge and make known the uncomfortable history — and legacy — of enslaved labour. Academic researchers across different fields are striving in their publications to un-silence and make visible the stories and experiences of enslaved people. Among them are Sophie White and Trevor Burnard,¹ Gloria García Rodrígues,² Marisa J. Fuentes,³ Dominique Rogers,⁴ and Alice Bellagamba, Sandra E. Greene and Martin A. Klein.⁵ Others have put together important digital resources that allow materials to be read afresh by researchers seeking to understand better the phenomenon and experience of enslavement.⁶ Alongside this welcome interest in marginalized groups has come more ready acknowledgement of the nature and extent of European involvement in the trans-Atlantic slave trade in particular. In the case of Scotland, significant moments came with the publication of It Wisnae Us: The Truth about Glasgow and Slavery⁷ and an edited collection entitled Recovering Scotland’s Slavery Past: the Caribbean Connection.⁸ This movement was complemented by the University of Glasgow’s recent initiative to investigate the extent to which its own wealth was generated by the slave trade, and what this might mean today.⁹ Fife Council has recently set up an Enslavement Education Group, which is exploring Fife’s historical links with enslavement, looking particularly at street names and monuments.

Further impetus came, notably, from University College London, home to the ESRC-funded Legacies of British Slave-ownership project (2009–12) and the AHRC-funded Structure and significance of British Caribbean slave-ownership 1763–1833 project (2013–15). These resulted in the creation of an extraordinary set of resources, including a database documenting those who received financial ‘compensation’ under the Slave Compensation Act of 1837 following the abolition of slavery in the British Caribbean, Mauritius and the Cape in 1833.¹⁰ This has shone a spotlight on ‘slave ownership’¹¹ and its legacies not just by major plantation owners, but also by less prominent individuals. The fact that individuals of British ancestry can today enter their family name or address into the search facility makes it difficult to ignore the possibility that our ancestors were involved in what could otherwise have been dismissed as the remote and irrelevant past. The experience of entering one’s family name into the database is described in a blog post published by the Runnymede Trust (the UK’s race equality think tank) in 2018.¹² It was a search for ‘slave owners’ in the Scottish town of St Andrews for the same blog post that led to the project that is outlined in detail below. One of the three hits for the town was John Whyte-Melville (or Whyte Melville), whose portrait, by Sir Francis Grant, is owned by the famous Royal and Ancient Golf Club. A quick search of the University of St Andrews’ Special Collections catalogue brought up the Bennochy Estate Papers (part of the Pagan, Osborne and Grace Papers), which features in its group-level catalogue entry a brief biography of Whyte-Melville. Until 2018, the biography did not mention the fact that Whyte-Melville had been a ‘slave owner’ or, to use a term that further queries the legitimacy of such ‘ownership,’ an enslaver. Although none of the materials held in the Bennochy Estates Papers relates directly to enslavement or enslaved people, it was agreed that it was our responsibility to include this information in the biography as part of a wider movement within academia and archives to make enslavement more visible. Our update, based on the information provided by the Legacies of British Slave-ownership (LBS) database, was carefully worded in order to convey clearly the known facts of Whyte-Melville’s association with enslaved labour: ‘Following the abolition of slavery, John Whyte-Melville received “compensation” for 131 enslaved people on his estate in the Caribbean island of Dominica in 1835.’

While it was a relatively straightforward matter to update our own catalogue entry, we are still in the process of contacting other online systems, such as Archives Hub, to which we have contributed entries in order that these may be updated as well. More broadly, this single update raised a number of much larger questions about how we might usefully and responsibly make enslaved labour and enslavement more visible in our Special Collections catalogue more generally. Museums and Museum Studies are used to dealing with such questions — indeed, a whole section of Politics of Memory: Making Slavery Visible in the Public Space is dedicated to museum practice¹³ — and the issue of how best to acknowledge the stories of enslavers and especially enslaved persons is part of a growing conversation among archivists.¹⁴ However, what has not yet been addressed in sufficient depth is the question of archival catalogues and the inclusion of data on enslavers within authority records, collection-level and item-level records. The opportunity to begin to explore the matter arose when the University of St Andrews launched its Gender, Diversity and Inclusion fund in 2019. Thanks to a small grant, a team featuring a senior academic, senior archivist, manuscript archivist and rare books librarian, with the active support of the director of Special Collections, acted as advisors to a postdoctoral research assistant (PDRA) working towards a qualification in Archive and Records management. The PDRA began to explore methods for using the Legacies of British Slave-ownership (LBS) database to update our Special Collections catalogues in ways that could be applicable to other archive catalogues across the UK and beyond. Although the project did not produce the results that we had originally anticipated, its findings (and lack of findings), as well as the issues that arose along the way, are certainly of interest to the wider academic and archival community.

‘Slave owners,’ the catalogue, and descriptive practice

The debate surrounding the ethical imperatives of archivists to respond to the historical hierarchies of value that remain embedded in archival collections and practices is wide-ranging.¹⁵ Acknowledging that recordkeepers participate in the creation of memory, the profession has been challenged to reconsider how archival agency shapes the omissions, elisions and exclusions of the record.¹⁶ The centrality of questions of race, histories and memory to the archival profession are reflected in the Archive and Record Association’s recent participation in the heritage sector’s joint statement of intent, which underlines that ‘our nation’s history and heritage is an invaluable tool in the fight against racism and discrimination’ (3 June 2020).

Within the archival record, the historic narratives of enslavers run alongside, though are different from, those of enslaved persons. Perspectives of enslaved and not enslaved are unequally captured and unequally preserved in these archives. This dynamic of inclusion and exclusion of enslaved and dominant voices in the record is complex and multifaceted and perpetuated through systems of recordkeeping and their sociocultural contexts.¹⁷ Archival work to engage with the records of enslavement — and such dynamics of oppression and omission — has several aspects. Reparative work, including, in Scotland, Glasgow’s Runaway Slaves in Eighteenth-Century Britain database of notices seeking the return of enslaved persons, has sought to re-discover narratives of the enslaved.¹⁸ There is no small irony in the fact that one of the richest sources of information about enslaved individuals that remains is the runaway advertisement.¹⁹ A more nuanced analysis of the record of enslavement, seen in academic research, such as the Legacies of British Slave-ownership research conducted at UCL, and investigations undertaken at a regional level, for instance in the University of Aberdeen’s 2007 exhibition ‘A North East Story,’ have increased understanding and awareness of this history.²⁰ Uncovering and re-framing these archives runs parallel to the re-evaluation of the role of archival practices, including description, in marginalizing communities and narratives.²¹

The St Andrews project examines how to discover and illuminate records with links to enslavers within the archival catalogue. Often, where enslavement-linked records are preserved, this aspect of the record is not explicit in the record’s description, whether at collection level, item level, or in relevant authority records. Explicitly re-identifying these individuals as ‘slave owners’ within the catalogue can offer an archival reframing of the narrative of such individuals’ records, to re-inscribing enslavement into their stories in order to reveal such connections to archive users. The following discussion first considers practical approaches to using UCL’s LBS database to discover enslavers within archival collections, before considering the potential impacts of including slavery-compensation details within the archival catalogue based on the results of our project.

The LBS database includes estates identified in the British Caribbean in the period 1763–1833, and all known enslavers, attorneys, mortgagees and legatees for the estates between 1763 and 1833.²² The information is based on the records of the Slave Compensation Commission, supplemented by further sources.²³ Individuals named within the database are not limited to those who successfully claimed for ‘compensation’; anyone named within the ‘compensation’ records is included. It should be noted that development of the database is still ongoing (information from Jamaican inventories, for example, was added in January 2020).²⁴ The primary search tool is the database of individuals associated with slave ‘compensation’ claims, where it is possible to search by individual details, address or claim. The entry for each individual provides, where known, dates, biography, and associated estates, claims, addresses, and names. In the entry for each estate, further information regarding possession and sale of the estate and associated enslaved people is provided. Another tab provides access to eight ‘legacy’ strands collated by topic, including commercial legacies and imperial legacies.

Development and design

The Making ‘Slave Ownership’ Visible project was first envisaged as a partial investigation of a single search approach, aiming to find compensated individuals within St Andrews’ Special Collections and update collection- and item-level catalogue descriptions accordingly. This approach was informed by the discovery and amendment of details pertaining to John Whyte-Melville outlined above. An interdepartmental team, led by Julia Prest (Professor of French and Caribbean Studies) included Gabriel Sewell (Head of Special Collections), Elizabeth Henderson (Head of Rare Books), Rachel Hart (Muniments Archivist), Maia Sheridan (Manuscripts Archivist) and Miriam Buncombe (Post-Doctoral Research Assistant). Early project-planning concluded that a single-approach trial would provide data with a limited longer-term application. Although this approach would uncover ‘slave ownership’ information for St Andrews’ current holdings, it would not help to define effective methods for interrogating LBS in other recordkeeping contexts, such as future accessions. It was agreed that we would adopt a more experimental approach in order to gain insight into practical approaches by which the information available in the LBS database might be used by an archives or Special Collections team to improve the description of links to ‘slave owners’ within collections under their stewardship. In this way, the project could support the improved integration of data on enslavers and enslavement-derived profit into everyday descriptive practices within a wide range of archives.

The scope of the project was restricted by the size of the grant, which funded a total of 182 research hours for the PDRA. Based on the core dates of the LBS database, the search was thus centred around the period between 1760–1833. It was decided not to consult archival items directly during the trial, with the recognition that in relying on existing item- and collection-level descriptions we would be limited by the pre-existing descriptive choices reflected in our catalogue. With the aim of exploration, time was allotted to investigating a variety of approaches and both focussed approaches, for instance collating records by searches for colony name, and wide-net approaches, such as investigating LBS individuals in alphabetical order, were included.

Ultimately, six approaches were explored. Three approaches took the Special Collections’ catalogue as a starting point: 1. Searching by colony name; 2. Searching by collection type (Estate papers); and 3. Searching by the date of the archived record. A further three approaches used the LBS database as a starting point: 4. Searching by surname of listed individuals associated with enslavement in alphabetical order; 5. Searching by region linked to LBS entries showing a geographic overlap with our archive collection; 6. Searching by firms associated with enslaved-labour listed within LBS.

Results and reflections

Figure 1 provides a statistical summary of our findings. This shows each search criterion (column 1) and how many individual and/or company names for cross-examination were generated (column 2). The given raw total for each approach includes only names discovered within catalogue entries for records falling within a time period closely relating to individuals active 1760–1833. The third column indicates how many names from the initial pool showed any like-for-like name correlation between the Special Collections catalogue description (collection-, series- or item-level entries) and LBS database, prior to confirmation or exclusion. The certainty of correlation between LBS name and name appearing in catalogue description was assessed by examining contextual information within LBS and catalogue description, including the broader administrative history provided at collection level in the Special Collections catalogue, where available (but without undertaking any significant further research using external or secondary sources). Contextual details considered in this assessment included known dates of activity for the individual, known address, colonial region of activity and further family relationships. This correlation was ranked: 1 — correlation excluded; 2 — correlation not excluded, no further details; 3 — further correlating detail in addition to name but insufficient to confirm or exclude; 4 — two correlating details in addition to name (likely); 5 — clear individual match based on contextual details (certain). Totals for individuals identified as ‘Likely’ (level 4) or ‘Certain’ (level 5) for each approach are indicated in column four. Individuals or companies linked to a compensated party, such as inheritors, discovered in the catalogue were also noted and the certainty of this correlation assessed (1b — 5b). The totals for both matches and linked matches considered likely or certain are shown in the fifth column (4 + 4b + 5 + 5b).

Figure 1. Name correlations between LBS and Special Collections catalogue

Challenges arose across all methods trialled. The breadth of information within the LBS database makes it a rich resource for uncovering links to enslavement; its scope and configuration, however, mean that multiple search strategies must be employed in order to extract as much relevant data as possible. Use of the ‘Commercial Firm’ legacies tab, for example, led to the discovery of archival records for 14 different parties compared with the 14 individuals with links to the archival collection discovered through use of LBS’s ‘address’ search function. LBS does not have an option to search all fields across the ‘individual’ and thematic ‘legacies’ databases. In the context of a regional collection, such an option might have provided a valuable shortcut.

LBS is divided into multiple sections, and information is not always repeated across all relevant segments. For example, when investigating the name ‘James Graham,’ although the estate records pertaining to Graham’s claim, ‘Jamaica Friendship 535,’ indicated that the owner was ‘James Graham of colour,’ this was not mirrored in the individual’s biography.²⁵ In a different case, Patrick Playfair’s business partnership Playfair, Crichton and Gilbert, the ‘Crichton’ of this firm is only described under Playfair’s partner John Gilbert, where he is named as James Crichton.²⁶ This illustrates how LBS may offer evidence of an individual’s participation in enslavement even where this person does not appear under their own entry. Crichton, in particular, appears to be evidenced within LBS again as executor to the will of Charles William Este, a possibility that is corroborated through a Playfair collection letter, held by Special Collections at St Andrews, which points to Antiguan society links between Crichton and Este.²⁷ In this case, LBS contains two details of Crichton’s links to Antiguan enslavers, but neither is discoverable through a surname search. Extensive use of primary material within LBS, and the resultant variation in spelling of personal and estate names, as, for instance, the Newell estate shown variously as Wellekens, Willekens, Willikens, and Wilkins, similarly requires the use of multiple-route searches.²⁸

By colony — method 1

The approach that generated the most catalogue records with explicit associations with the ‘ownership’ of enslaved persons (20) was searching by colony. Here, the names of West Indian islands (Anguilla, Antigua, Bahamas, Barbados, Dominica, Jamaica, St Kitts, Tobago, West Indies) were used as search terms for descriptive text within St Andrews Special Collections CALM catalogue.²⁹ Any personal names or estate names appearing in these catalogue descriptions — at collection, series or item level — were then cross-checked in LBS, both within the ‘individual details’ and ‘Notes’ search fields, and under the ‘Estates’ tab as appropriate. This approach found 121 records, with 117 discrete personal names of which 50 found such-named individuals in LBS. Based on the correlation of identifying factors listed within LBS and the current Special Collections catalogue description, 20 of these were considered certain or likely to refer to the same individual.

Capturing catalogue description explicitly referencing the West Indies, it is unsurprising that links to enslaved labour appeared less ‘hidden’ through this method than in records revealed through other approaches. Investigation of ‘slave ownership’ in these records nonetheless put the nature of Scottish interests in the colonial West Indies into stark relief. It underlined the simple but important fact that where Scottish individuals have explicit connections to the West Indies during the time-period examined, c. 1763–1833, these are almost exclusively connections to enslaved labour whether through personal ‘ownership’ of enslaved persons or links to a colonial society centred around enslaved labour.³⁰ Indeed, the enslaved labour context implicit within such addresses is an important argument in favour of investigating any link to the West Indies when listing material of this period. Examining the names discovered through this method highlighted the tendency in this period of profit from enslavement to cohere within social networks and the continuation of West Indian economic and social networks in the Scottish context. A further valuable aspect of this approach was in demonstrating that individuals named within archives showing explicit connections with enslaved labour did not necessarily appear within LBS, as was the case for Captain Arthur Law, husband to Penelope Newell Law.³¹ This sounds an important note of caution regarding the limits of LBS as a resource in this context. The variety of links to slave-owning society illuminated by this comparison between the catalogue and LBS raises complex questions regarding the merits of delimiting the inclusion of connections with enslaved labour within archival description, at collection or item level, based on the form of an individual’s ‘compensation.’ Should catalogue entries that refer to ‘slave ownership’ be restricted to individuals who actually purchased (or who had purchased on their behalf) enslaved people, or should the catalogue’s definition of ‘slave ownership’ extend to all ‘owners’ of enslaved people, including inheritors of enslavers? Perhaps different forms of involvement should be suggested by distinguishing between different types of ‘slave ownership’ (e.g. direct or indirect) in the catalogue? The answers to these questions might depend on the level of cataloguing and which method is being used: a prose update may be able to move more smoothly between different verbs (e.g. ‘own,’ ‘purchase’ or ‘inherit’) than use of a keyword tagging system. Keywords may introduce their own issues as types of slave ‘ownership’ link, as shown above, are wide-ranging and hard to define precisely. But different tags for different types of ‘ownership’ or different contexts for ‘compensation’ may offer one way of making connections with slave ‘ownership’ clear within the archival catalogue and searchable.

By collection type (estate papers) — method 2

A reference list of 23 collections of Fife estate papers held within the archive was used as a starting point. The names of key individuals associated with these collections were identified using the current catalogue descriptions, both at collection and item level as available, of these collections, and any individual names and family surnames discovered in the catalogue were cross-checked in LBS, both in the ‘individual details’ and ‘notes’ search fields.

Unfortunately, the highly detailed contextual knowledge provided by estate collections proved a hindrance to the easy discovery of ‘slave owning’ individuals and only five individuals considered likely or certain to match a compensated individual named by LBS were identified. The Hay of Leys papers, for instance, include 222 catalogue records with reference to at least 10 estates in Scotland, as well as the Hay, Balfour and Paterson families. This provides a disparate body of potential search terms resulting in too many variables to establish clear links with LBS. Searching LBS by house or estate names within the address field, or individual notes field was partially effective. The family name associated with an estate, such as Hay of Leys, did not offer a productive route of inquiry. Using a broad search term, such as a common surname, generated many results, whose interrogation was complicated by the minimal biography for some LBS individuals, as well as the lack of overlap of LBS’ data with that of the archives. With a concentration on Scottish property, our estate papers, for instance, rarely name the West Indies explicitly, which might have offered a key point of correlation with the focus of record evidence underpinning LBS. Using the ‘notes’ or ‘address’ fields in LBS, searches by family name (e.g. ‘Hay’) could be narrowed to individuals demonstrating contextual alignment with the respective estate (for example, a connection to Scotland). This did not overcome the genealogical complexity of the divergence of cadet relationships that further complicated investigations to establish which branch was compensated for enslaved persons and if financial associations between branches remained at the point of involvement in enslavement. Estate papers, moreover, capture the wider social circles of the associated family. Exploration of estate papers thus quickly becomes an examination of a broader network within a social stratum, with enslavers identified via this route often emerging outside the estate papers generating the search term. This route into tracing connections with ‘slave ownership’ connections within social circles, although laborious, could nonetheless be valuable in investigating social networks where we have knowledge of a connection to the West Indies or enslaved labour. This again underlines how the archival record mirrors a context in which individuals benefitted from enslavement and ‘slave ownership’ without (necessarily) having been directly involved in it.

By date — method 3

By far the least effective approach (which produced no certain or likely matches to ‘compensated’ individuals) was selecting archival holdings based solely on date. Consecutive catalogue records for archival items listed as dating to within sample years 1763 and 1783 (from the beginning and middle of the project’s chronological limits) were examined, and any personal names found in the first 30 catalogue descriptions for each year were cross-checked in LBS. This process produced the worst search ratio: searching 112 names resulted in no positive matches. A particular issue, exacerbated by the number of local administrative records that this approach covered, was the limited context provided for most of the names discovered in such records. Two key factors contributed to this difficulty. Firstly, as in the case of the Burgh records, many individuals named within the archival document were not central to the purpose of the record. One example was ‘Thomas Scott,’ appearing in St Andrews’ Guildry accounts (1783–1 October 1785), a record offering few details through which to narrow the list of 16 ‘Thomas Scott’ individuals within LBS.³² These limitations resulted in insufficient information to determine any correlation between individuals. A potential advantage of such decontextualisation, however, was that this circumvented researcher preconceptions regarding whether an individual was ‘likely’ to have connections to enslaved labour, for example because of their known social class or business associations.

By surname — method 4

Using the search term ‘%%%’ within the ‘Surname’ field, records of all individuals associated with enslavement with a named entry within LBS were brought up in alphabetical order.³³ The first 170 of these people (up to Adamson) were then cross-checked with St Andrews’ archival catalogue by searching for each name in the descriptions of currently catalogued archival items. This LBS search was repeated, narrowing the complete list of named individuals by simultaneously applying the search term ‘Scotland’ under the LBS search field ‘location’ in addition to the search term ‘%%%’ within the ‘Surname’ field; 170 of the resulting names (Adair — Cuthill) were cross-checked for appearances in the Special Collections catalogue descriptions.

Although examining named individuals in the LBS database in alphabetical order was also slow-going and resulted in a limited number of confirmed matches, it revealed correlations between ‘slave owners’ and records held by St Andrews discovered through no other approach. Instances of a confirmed positive correlation increased slightly (two versus six) when the alphabetic list of names was further narrowed through use of region (Scotland). However, narrowing the search in this way excludes individuals of Scottish origin for whom no Scottish address is listed, and may overlook those records held by St Andrews without explicit Scottish links. While cross-checking the full alphabetic list of individuals named by LBS (without any further delimitation) is an extremely time-consuming process, it would be the most complete way to cross-check all individuals included within LBS with an archival collection. A further benefit of this approach is that it avoids being influenced by preconceptions regarding which archives, individuals, or even date-ranges are likely to have associations with enslavement. This has potential for an ongoing project to make enslavers visible within St Andrews’ holdings, though it would of course require substantial resources.³⁴

By region — method 5

Employing the ‘Address’ field within LBS, the terms ‘St Andrews’ and ‘Fife’ were used to identify individuals in the database with potential links to regions closest to St Andrews. These searches were supplemented by using the terms ‘St Andrews’ and ‘Fife’ in the notes field. All names identified in LBS with a St Andrews or Fife connection were then cross-checked for presence within the archival catalogue.

As the St Andrews archive’s collecting policy is influenced by regional criteria, the use of address-based terms to interrogate the LBS database appeared a potentially good route through which to pinpoint enslavers for whom St Andrews may hold records. However, this approach was not as fruitful as anticipated; although over 400 catalogue entries containing the names found in LBS within the project’s time frame were available, this process uncovered only five probable matches between a compensated person and an archival item. An influencing factor may be the London-centred context of the Compensation Commission. Initial advertisement of the scheme was only through English sources and payment, which required claimants or their appointed agents to appear in person, was administered via London.³⁵ Despite there being a disproportionately large number of Scottish individuals associated with enslaved labour compared with population size captured in the LBS database, this may still be an underrepresentation of the full scope of Scottish ‘slave ownership.’ Although in this instance ‘compensation’ links to Fife identified through LBS did not translate into individuals with direct matches with archival holdings, the results did offer a glimpse into the influence of economic impacts of enslaved labour in Fife. While this knowledge may not be directly applicable to current holdings, the increased understanding of local ‘slave ownership’ patterns provides a foundational awareness of the wider context for records of this period and may inform the treatment of future accessions.

By firm — method 6

Beginning with entries on ‘commercial firms’ provided under the LBS Legacies tab, which provides drop-down lists for ‘Firm Name’ and ‘City,’ all firms for Edinburgh and Glasgow were called up.³⁶ The titular company name for the LBS entry, variations of the firm name and key individuals for the firm, as described in the LBS entry, were cross-checked with record descriptions in the archival catalogue. This was repeated for firms listed for ‘London,’ where many Scots were active in businesses linked to enslavement.

This approach succeeded in generating a limited pool of contextual information, which provided multiple points of reference from which to identify individual names and confirm matches. Finding correlations between LBS and the archival collections proved most effective when the list of firms was narrowed to Edinburgh and Glasgow. Using the firm name and listed shareholders as search terms we were able to pinpoint previously unidentified links between individuals and firms and Special Collections’ holdings. Search terms discovered through LBS entries on commercial legacies (such as firm or stakeholder names) appeared most frequently in records created outside the company with a named entry in LBS. This included, for instance, archives relating to a firm shareholder’s other personal or commercial capacities, as is the case for William Frederick Burnley, of the LBS-named firm Eccles, Burnley and Co, who appears as correspondent with William Forbes Skene.³⁷

The results

The ease and clarity of the discovery of Robert Whyte-Melville within the LBS database and collections held by St Andrews belied the significantly murkier and time-consuming process of systematically applying the LBS data to archival collections. Cases of proven, direct matches between individuals named by LBS and named individuals associated with our archival records were rare for all search methods tested. Indeed, no potentially matching individuals (like-for-like names within the archival catalogue and LBS) were found at all for 60% of names checked. Moreover, out of the 842 names revealed and cross-checked in total, across all six approaches trialled, only 17 resulted in fully confirmable ‘certain’ matches (2%) between a named individual in LBS and a person named within or associated with an archival record. This figure rises to just under 5% if likely matches, classified as ‘likely’ if three or more points of correlating contextual information could be established, are also included. Our distinction between certain and likely matches raises another important question for the cataloguer: how to include levels of certainty in the catalogue. This is a distinction that would best lend itself to a tagging system of some kind. If, in addition to likely and certain matches, likely or certain instances of persons linked to ‘compensated’ individuals, such as their children, are also taken into account, the total proportion of matches is still just 8.4%. The results showed that of the 342 names (out of the 842 cross-checked) producing any match at all between the archival record and LBS database, 55% would require further research to be able to positively confirm or exclude a possible correlation between enslaver and archival individual.

A particularly important finding is that even where certain and likely links between enslavers and archival holding were found, the content of these archival records (as described in the catalogue) rarely showed any relation to enslavement. The overwhelming majority (67%) of archival items related to the individuals identified made no explicit mention of the West Indies, trade linked to the West Indies or enslavement. A further 9% had only brief content demonstrating a West Indian connection, such as an address. These results affirmed the foundational suspicion of the project team, and part of the motivation for the trial: although many individuals were ‘compensated’ for ‘slave ownership,’ and aspects of their histories are captured in archival records, the link between an individual who received ‘compensation’ and an individual as archival subject is neither an overt subject in surviving manuscripts nor explicit in the existing archival catalogue. It is our conviction that the absence of archival materials appearing (at first glance) to have a direct bearing on enslavement increases, rather than decreases, the need to make such a connection more apparent in our catalogue.

Updating the Catalogue

Across all approaches, this investigation has revealed 40 certain or likely matches between individuals named within LBS and subjects named in St Andrews’ archival catalogue.* However, it has also reflected the ways in which the economic impacts of enslaved labour spread widely beyond personal ‘ownership’ of enslaved people, thus echoing the findings of the LBS research. The results for St Andrews’ holdings showed only half of those individuals as having been ‘compensated’ for personal ‘ownership’ of enslaved persons. Individuals such as Penelope Newell Law participated in family-owned plantations, while other individuals such as Baron George Gavin Browne Mill inherited the ‘right’ to claim for ‘compensation,’ or, like William Pulteney Alison, gained a ‘right’ to enslaved persons through trusteeship.³⁸ Although such claimants do not fall within the primary remit of this initial investigation, descriptive clarity on these forms of profit based on enslaved persons would be valuable in illuminating the threads of links to individuals associated with enslavement and enslavement-derived profit running through St Andrews’ archival collections. These examples invite a further nuancing of our earlier distinction between direct and indirect enslavers. Any individual who bought (or had bought on their behalf) enslaved people is clearly an active participant in the trade in enslaved persons. An individual who inherits the ‘right’ to claim ‘compensation’ becomes active if and when they decide to make a claim.¹ But what if there is evidence of a reluctance to claim ‘compensation’ or even of coercion? Keeping prose updates or tags strictly factual is part of the solution to this problem, but tagging systems in particular run the risk of simplifying matters too much and it is crucial that users understand what any enslavement-related tag is indicating. The results of this trial reflect the varied connections between firms, families and individuals and slave ‘compensation,’ and likewise the complex ways in which parties associated with enslavement-derived profit are linked to the preserved record. Critical questions of how these different links might be reflected in the archival catalogue, whether any limits to descriptive catalogue inclusion are desirable or how these might be determined (for instance based on degree of association with enslavement; enslaver degree of association with the archival record?), remain open.

Consideration of the interrelationship of such factors also influences decisions on amending authority files and adding ‘compensation’ details to collection-level and/or item-level descriptions. Such descriptive choices must be weighed in light of the primary consideration of this investigation, namely how the inclusion of slave ‘compensation’ can increase the visibility of enslavers’ records for archive users.

Comparison of current description of archives linked to the colonial West Indies with the information found in LBS highlights the potential differences in visibility of ‘slave ownership’ and perspective on the archival record of this history achieved by engaging with ‘slave ownership’ in cataloguing practice at all levels of description. The will of Henry Barham, for example, is currently described at item level as ‘Copy Last Will and Testament Henry Barham late of Jamaica now of London. Executors: Elizabeth Barham his wife, Messrs Roger Drake and Beaston Long, Colonel George Rickets and Dr James Paterson. 22 May 1746.’³⁹ Using LBS, it is possible to demonstrate that Barham owned 287 enslaved persons at probate;⁴⁰ Roger Drake and Beeston Long were partners in Long, Drake and Co., which ran, in various iterations, as a West India Merchant between c. 1730–1780s;⁴¹ Beeston Long was Mortgagee in possession of Saltspring Pen estate on Jamaica, and left £39,000 in his will;⁴² Col. George Ricketts may have been the same man who is shown as a Major-General of the Jamaican militia, owning 288 enslaved persons at probate.⁴³ Finally, Dr James Paterson may be (or be related to), James Paterson of Carpow, in whose archives Barham’s will is found, and whose records demonstrate links to colonial Jamaica and Jamaican planters such as Julines Beckford.⁴⁴ Using LBS’ data, Beckford is reframed as the owner of 662 enslaved persons at probate.⁴⁵ While this information is not new to researchers of enslavement, it is not explicit in the archival object or the item level catalogue entry; descriptive changes using ‘compensation’ data should help to bridge the gap between the work of specialists and a wider audience of archive users. Inclusion of ‘compensation’ information at all levels of description would make this link to enslavement visible to users independently of their approach to the catalogue. For Barham’s will, ‘slave ownership’ data transforms the descriptive portrait of the record offering a glimpse of colonial connections to a depiction of a record (and record subject) tied into economic and social connections based around enslaved labour. This creates the potential for seeing ‘slave ownership’ despite the fact that Barham appears in only two brief records held by St Andrews’ Special Collections; this is a valuable insight as 37 of the 40 certain or likely claimants uncovered appear in only a small number of archival records held, with their names appearing only in isolated item-level descriptions.

As Barham’s will indicates, one outstanding issue raised by our trial is the difficulty of confirming an enslaver’s identity within records, and what degree of certainty is necessary for ‘slave ownership’ to be included within description.⁴⁶ This process of decision-making is itself marked by a context influenced by a current understanding of, and socio-cultural discomfort surrounding, this history.⁴⁷ A cataloguer’s choices regarding which research on a party’s ‘slave ownership’ to prioritise intertwine with current (mis)understandings about participation in ‘slave ownership.’⁴⁸ Retrospectively, it may be contended that preconceptions of a higher probability of correlation between land-owning classes and enslavement may have influenced our own choice to use estate papers as one route of inquiry. Certainly, there was surprise when the project identified potential ‘compensated’ individuals within the local fishing communities of St Andrews (Helen Braid) and Anstruther (Charles Wightmann).⁴⁹

Restrictions unwittingly imposed by underlying assumptions are further compounded by the limitations on available archival evidence, which disproportionately affect marginalized groups including less affluent individuals or women. Within the LBS database, entries for individuals making smaller claims are less likely to include biographical details. The absence of information in LBS is mirrored in the archival perspective where working-class individuals are less likely to be captured in the record.⁵⁰ Therefore, the issue of match certainty appears linked to socio-cultural power, with the possibility of a confirmed match being more probable for individuals of historically powerful groups. In this instance, the priorities of the archival records of the Compensation Commission, amongst others, reflected in the LBS database, may be perpetuated through using certainty of an individual’s identification as a criterion in embedding ‘slave ownership’ information in the archival catalogue. This highlights the importance of continued collaboration between academic researchers and archivists.⁵¹ Considering that current understanding and preconceptions of ‘slave ownership’ are based, in part, on the visibility of enslavement in records, the issue of proof raised by this trial strongly points to the need for greater access to the body of archives evidencing enslavers. This further demonstrates the potential value of this trial in investigating approaches for enhancing descriptive practice to enable a more nuanced understanding of ‘slave ownership.’

Conclusion

This exploratory project has investigated ways for archivists to use the Legacies of British Slave-ownership database to make ‘slave ownership’ visible at all levels of description in the archival catalogue. Our pilot has resulted in two key conclusions: first, that using LBS as a resource in archival descriptive practice would increase the visibility of enslavers and strongly support improved access to the archival record of this history (and, for this reason, it is now the case that all new acquisitions at St Andrews are systematically checked against LBS upon receipt, and any descriptions relating to individuals who received ‘compensation’ now include this information as a matter of course). The resounding sense of the value of integrating this approach must, however, be tempered by the second finding that, even using LBS, the process of retrospective correlation of collections with ‘compensation’ data is neither swift nor clear cut. Archival collections will need to allocate sufficient resources for such work.

Our trial suggests that using targeted search terms, such as colony names, was most effective at discovering archival matches with compensated individuals. Yet no single approach found a pool of compensated parties that entirely overlapped with the group from other approaches. While cross-checking compensated individuals within LBS through alphabetic listing would provide a complete investigation of all enslavers listed, this is impractical for most archives. Examining the archival catalogue based solely on date was ineffective for St Andrews’ collections. Beginning with a sufficient but delimited body of information, such as a section of related archival records, gave the greatest chance of discovering links with enslavement. Cross-matching data in LBS and the catalogue was most effective when multiple searches were used to provide alternative angles on collections through which connections to different enslavers could emerge. Two practical factors were underlined by this examination: the importance of interrogating LBS via multiple search fields, including the ‘notes’ field, and the value of checking all names mentioned within a single record (where one person within an archival record was identified as an enslaver, other individuals named by the records were often also linked with enslavement).

This trial has also stressed that a pre-existing understanding of collections may influence which areas are investigated for ‘slave ownership.’ Such knowledge is often coloured by assumptions, not always borne out in practice, regarding likely enslavers, and may exclude unexpected areas of correlation between collections and enslaved labour. A multi-search route approach partially counteracts researcher bias by widening the scope of investigation. Current understanding of ‘slave ownership’ for Fife, and in Britain, is compounded by the interplay between social groups, record presence and the ability to confirm ‘ownership’ of enslaved people — something that would benefit from further collaborative investigation between archival professionals and academic researchers. A number of legacies strands, such as ‘Imperial Legacies’ or ‘Literary Legacies’ were not explored at all, and these may be highly informative for collections with strengths in corresponding areas.

Finally, it is important to underline that the approach explored in this project should be considered only one step in a broader, continual process. The results of this project, by necessity limited in scope, could be complemented, for instance, by searches for commodities produced using enslaved labour (sugar and coffee), or products manufactured for use by enslavers (linen or herring). The time period investigated, 1763–1833, does not mark the beginning or end of Scottish profit derived through enslaved labour, leaving a large segment of our archive unexplored. Similar work with other databases of historic enslavement, such as Slave Voyages,⁵² would also complement the work of this project, engaging a different spotlight through which to highlight records’ links to enslavers. It should also be acknowledged that adding information about ‘compensation’ to our various catalogue entries only makes ‘slave ownership’ visible to those who read the additional text. If this information is embedded in a long description (as is the case for Whyte-Melville), it may well be overlooked. Even a tagging system can be limited in its capacity to make ‘slave ownership’ visible to any users who do not actively search for the appropriate tag. We recognize that including details of ‘slave ownership’ does not automatically make it as visible in the catalogue as we would like. This is a matter for ongoing discussion within the archival and academic communities. Above all, we must together consider how descriptive inclusion of ‘compensation,’ which increases the presence of enslavers’ narratives in the catalogue, can be embedded within wider practices that make visible the stories of enslaved people themselves.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Miriam Buncombe

Miriam Buncombe is a recent Archives and Records Management graduate (University of Dundee, 2020). Following undergraduate studies at the University of St Andrews and the Friedrich Wilhelm University of Bonn, she completed postgraduate research into the role of ritual in mediaeval mystical narratives. She is currently project archivist working to catalogue the Archive of European Intellectual Life held at the University of St Andrews Special Collections.

Julia Prest

Julia Prest is Professor of French and Caribbean Studies at the University of St Andrews. She has published widely on early-modern French and Creole theatre, and her current research is on theatre and enslaved people in the former French colony of Saint-Domingue (now Haiti). Julia is founder of the Colonial-Era Caribbean Theatre and Opera Network (CECTON) and creator of the trilingual (English-French-Creole) website and performance database, Theatre in Saint-Domingue, 1764-1791: Plays, Ballets and Operas, https://www.theatreinsaintdomingue.org.

Notes

^* This number excludes a duplicate name that appears in Figure 1

1. White, Voices of the Enslaved. White and Burnard, Hearing Enslaved Voices. Burnard, Hearing Slaves Speak.

2. Rodríguez, Voices of the Enslaved in Nineteenth-Century Cuba.

3. Fuentes, Dispossessed Lives: Enslaved Women, Violence, and the Archive.

4. Rogers, Voix d’Esclaves.

5. Bellagamba et al., African Voices on Slavery and the Slave Trade, vol 2.

6. See especially Marronage, Accessed February 25, 2020, http://www.marronnage.info.

7. Mullen, It Wisnae Us.

8. Devine, ed., Discovering Scotland’s Slavery Past.

9. Mullen and Newman, ‘Slavery, Abolition and the University of Glasgow’. University of Glasgow. September 2018. https://www.gla.ac.uk/media/Media_607547_smxx.pdf. Other universities, including Bristol and Cambridge, have undertaken similar inquiries. In October 2019, Bristol appointed its first Professor of the History of Slavery, Olivette Otele.

10. LBS, Accessed February 25, 2020, https://www.ucl.ac.uk/lbs. See also Hall et al, Legacies of British Slave-ownership.

11. When writing in our own voice, we have chosen to put ‘slave ownership’ in inverted commas in order to query the notion that one human could ‘own’ another and consider them a ‘slave’. We also put ‘compensation’ in inverted commas to contest the notion that enslavers were owed ‘compensation’ following abolition.

12. Prest, ‘Slavery is a local — as well as a global — issue’, Race Matters (Runnymede Trust’s blog) (November 12, 2018).

13. Araujo, ed., Politics of Memory: Making Slavery Visible.

14. See for instance anti-racist descriptive resources by Archives for Black Lives in Philadelphia; on cataloguing records of enslavement see for example National Records of Scotland, “James Montgomery Slavery Case”.

15. Early calls for engagement with archival inequalities of representation include Zinn, “Secrecy, Archives, and the Public Interest”, and more recently in the work of Michelle Caswell, Terry Cook, and Verne Harris.

16. On archival agency see for instance Harris, “The Archival Sliver”; Caswell et al., “To be able to imagine otherwise”; Kaplan, “We Are What We Collect”; Cook, “Evidence, Memory, Identity, and Community”.

17. On whose voice is captured, see Wood et al., “Mobilizing records”; Bastian, “Reading Colonial Records”; Aljoe et al., “Obeah and the Early Caribbean Digital Archive”; on post-abolition cultural shifts, see for example Oldfield, Chords of Freedom.

18. Regarding approaches to colonial records, see Haberstock, “Participatory description”; Bastian, “Reading Colonial Records through an Archival Lens”.

19. For an example of how these can be used productively, see Waldstreicher, “Reading the Runaways”; and Fuentes, Dispossessed Lives, Chapter One.

20. The University of Aberdeen, “A North East Story”; and Hackney Museums and Archives, “Local Roots/Global Routes.”

21. On changes to practice, see for example First Archivist Circle, “Protocols for Native American Archive Materials,” Accessed February 25, 2021, https://www2.nau.edu/libnap-p/protocols.html.

22. LBS, Project details, Accessed February 1, 2020, https://www.ucl.ac.uk/lbs/project/details/.

23. LBS, The Database, Accessed July 6, 2020, https://www.ucl.ac.uk/lbs/project/details.

24. LBS, Database Updates — January, 2020, Accessed July 6, 2020, https://www.ucl.ac.uk/lbs/project/new2020.

25. LBS, Friendship Settlement, Jamaica, Westmoreland, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/estate/view/3205; LBS, James Graham, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/24142.

26. LBS, John Gilbert of Antigua, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146652843.

27. LBS, Charles William Este, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146645189; Letter 5, Diaries and Letters of Mrs Patrick Playfair of Kilmarnock, msdep14/2/48/3/5.

28. LBS, Researching Slave Owners, Accessed July 6, 2020, https://www.ucl.ac.uk/lbs/project/research; LBS, Penelope Newell Law (née Hepburn), Accessed July 7, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146658943; LBS, William Hepburn, Accessed July 7, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146651741; LBS, Willekins, Jamaica, St Dorothy, Accessed July 7, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/estate/view/14721.

29. Total list of colonies in LBS pick-list September 2019; this has since increased.

30. Scottish slavery connections are not restricted to this period, see Devine, “Did Slavery make Scotia great?”; and Duffill, “The Africa Trade”.

31. Letter from Penelope Newell Law to William Berry, 30 July 1807, msDA817.B4 ms3881.

32. Guildry Accounts, St Andrews: Robert Niven, 1783–1 October, 1785, B65/22/11/9/1; LBS, search term ‘Thomas Scott’, 16 hits, Accessed July 7, 2020, https://www.ucl.ac.uk/lbs/search/.

33. After putting all names in alphabetical order, it is only possible to move through the results one page at a time, LBS, Search %%% under Surname field, Accessed July 8, 2020, https://www.ucl.ac.uk/lbs/search/.

34. LBS, search term %%% under Surname field, Accessed July 7, 2020, https://www.ucl.ac.uk/lbs/search/.

35. Draper, “Scotland and Colonial Slave Ownership,” In Rediscovering Scotland’s Slavery Past, edited by, Devine, 177.

36. The only Scottish cities available from LBS pick-list, September 30, 2019.

37. Letter from William Forbes Skene to William F Burnley, October 30, 1869, msdep19/4/35 (see also msdep19/4/36), Papers of George Hay Forbes, University of St Andrews Special Collections, St Andrews, Fife.

38. LBS, Penelope Newell Law (née Hepburn), Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146658943; LBS, Baron George Gavin Browne Mill, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/16873; and LBS, William Pulteney Alison, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/44168.

39. Copy Last Will and Testament Henry Barham, 1746, ms36220/1888. On Barham, see Dunn, “A Tale of Two Plantations.”

40. LBS, Dr. Henry Barham, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146635190.

41. LBS, Long, Drake & Co, Accessed July 8, 2020, https://www.ucl.ac.uk/lbs/firm/view/2144928789.

42. LBS, Beeston Long senior, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146644619.

43. LBS, George Ricketts, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146644275.

44. Papers of the Hay Family of Leys, 1312–1924, ms 36,220; see for instance ms36220/1359 and 1360. The Beckford family’s associations with slavery are well known, on Julines’ brother see Gauci, William Beckford.

45. LBS, Julines Beckford, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146650155; Paterson may also be the man named in the will of William Grahame of Mossknowe, transcribed by LBS, see LBS, William Graham of Mossknowe, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146634698.

46. On the ethics of description, see for example, Duff and Harris, “Stories and Names”; MacNeil, “Picking our text”; and Light, Hyry, “Colophons and annotations”.

47. On coloniality in archival practice, see Agostinho, “Archival encounters”; Ghaddar and Caswell, “To go beyond”.

48. On socio-cultural memory of British participation in slavery see Donnington et al., eds., Britain’s History and Memory of Transatlantic Slavery.

49. Helen Braid appears in Claim of Service by James Braid as agnate to John Braid, 14 January 1786, B65/22/1/6/1; see also B65/22/1/8/1; LBS, Helen Braid, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/-641637211; Wightman may be the man named in Charter of Resignation and confirmation by Sir John Anstruther in favour of Charles Wightman, 3 October 1749, msdep121/1/1/7/2 (identification of this individual remains inconclusive); LBS, Charles Wightman, Accessed July 8, 2020, http://wwwdepts-live.ucl.ac.uk/lbs/person/view/2146632323.

50. See Flinn, “Other ways of thinking,” In What are Archives? 109–128.

51. See Stoler, Along the Archival Grain; Jeurgens and Karabinos, “Paradoxes of curating colonial memory”.

52. https://www.slavevoyages.org/.