Navigation – Plan du site

AccueilNuméros3La recherche par l’écritDossier thématiqueDocumenting and Digitizing with D...

La recherche par l’écrit
Dossier thématique

Documenting and Digitizing with Dignity: Ethical Considerations and the West African Frontier Force Personnel Records

Documenter et numériser avec dignité. Considérations éthiques à propos de la numérisation des dossiers du personnel de la West African Frontier Force
Documentar y digitalizar con dignidad. Consideraciones éticas sobre la digitalización de los expedientes del personal de la West African Frontier Force
Documentar e digitalizar com dignidade. Considerações éticas sobre a digitalização dos registos do pessoal da West African Frontier Force
Katrina Keefer, Eric Lehman, Michael McGill et Gabriela Mattia


Cet article examine certaines des conséquences de la numérisation des dossiers du personnel de la West African Frontier Force conservés dans les archives publiques de la Sierre Leone. Ces dossiers rendent possible l’accès à une image d’un passé encore vif, puisqu’ils contiennent des informations délicates et confidentielles, telles que des détails médicaux et personnels, qui ne sont pas normalement révélés au public. Il faut tenir compte des bonnes pratiques et de l’éthique pour protéger la vie privée des sujets, et cela doit s’appliquer à toutes les archives liées à la traite des esclaves transatlantique. Les bonnes pratiques numériques valorisent la dignité humaine des personnes qui risqueraient autrement d’être traitées comme des données. Dans cet article, nous commençons d’abord à explorer l’archive sous ses aspects fragmentaires ainsi que physiques : dans certains cas, les archives abritent des documents essentiels à nos recherches, mais cela dans des conditions très précaires. Nos réflexions sur la matérialité de l’archive nous ont menés à de bonnes pratiques qui respectent aussi bien la provenance des sources primaires que les personnes qui les conservent. Nous passons ensuite à des considérations méthodologiques et éthiques : comment organiser, anonymiser et standardiser les données et les métadonnées. Nous espérons que cette section servira de guide à d’autres projets de recherche semblables en proposant une méthode efficace pour organiser des systèmes complexes de documents, surtout pour ceux où il faudrait respecter le plus possible la structure d’origine des fichiers pour en respecter la signification. Enfin, nous abordons les problèmes que pose la localisation des origines des soldats sur les cartes topographiques ; cela nous a amenés à analyser les divers problèmes de la cartographie coloniale dans sa globalité, et ainsi à trouver des solutions adéquates. Notre article éclaire les défis de la recherche et de la numérisation de vies individuelles à travers des données d’archives complexes : dans chaque cas, nous avons décrit et analysé les obstacles auxquels nous avons été confrontés et les moyens trouvés pour les affronter ou les résoudre, avec la conviction qu’ils pouvaient constituer de bonnes pratiques qui pouvaient être appliquées à d’autres projets.

Haut de page

Texte intégral

  • 1 Fula’s name is spelled throughout his personnel file variously as Mussa Fula, Musa Fula, or Musa F (...)

1It was 1944, and Musa Fula1 was in the process of being sentenced to three years of hard labour for his part in a series of thefts which took place at the place where his battery was stationed. A year earlier, he had been punished for being insufficiently alert while on duty and fined for his offense. Twice that year, Fula had been away without leave, and each time he was fined for his offense. After his sentence was completed, he was discharged from service with dishonour for his crime. He had been in the Frontier Force for only three years.

  • 2 Fula’s paternity is of interest because his father was one of the main suspects in the 1913 Specia (...)

2Prior to his enlistment in 1941, Fula was an eighteen-year-old student, and was a practicing Muslim, like his father, the paramount chief Bunting Williams of Mattru in Jong Chiefdom.2 The crime he was accused of committing took place in India, far from his homeland of Sierra Leone, where British colonial forces like the Frontier Force were to fight against the Japanese. Both in India and in Burma, African soldiers fought valiantly, meriting medals and losing their lives as World War II raged. While broad-strokes narratives of these conflicts are known, and some details exist concerning the decorations which were earned, there have been few sources to understand what life was like for colonial soldiers like Musa Fula. In his file, which is held in a rice bag in the Sierra Leone Public Archives with approximately 6,000 others, the full transcripts may be found of his court martial and the evidence heard against him. In his attestation papers, his vaccination scars are described, along with his height of 5’7”, his skin tone, his religion, his home village, and his medical history during his brief time in the Frontier Force. Signatures as well as carefully inked and pressed thumbprints adorn the documents which present a snapshot of a young man finding himself in India, in the army, and before long, in trouble.

3For soldiers who spent decades enlisted, the files which exist represent lengthy narratives of their lives. They detail family visits, spouse’s, parents’, and children’s names, vaccination records, accomplishments, offenses, and more. They often include photographs of individual soldiers, hand- and fingerprints, and dental records as well as identification tags in some cases. Regimental Conduct sheets present detailed, year-by-year accounts, each one dated, signed, and documented. Attestation papers require each recruit to identify their home village, district, previous occupation, father’s name, and ethnonym. They describe and typically draw all identifying facial and body marks, differentiating between so-called “tribal” marks—including tattoos, secret society initiation marks, and ethno-linguistic scarification identifiers—and marks from medical interventions or past injuries. The value to modern scholars of this rich dataset is extraordinary, and applicable to a wide range of fields of study. Preserving, protecting, digitizing, and organizing these documents so that scholars can analyze and explore them is a clear need, and unquestionably, the information within this remarkable trove of material demands attention. In a broader context, this largely untouched dataset can be used as an exemplar for other, comparable datasets. Records of enslavement and manumission may present the same sets of challenges in terms of their important cultural heritage and need for preservation; the Frontier Force data poses a different challenge which pertains specifically to best practices. Because the information within these files is deeply personal, and in many cases comparatively recent, questions of anonymity, identity, and privacy arise which impose a reconsideration of best practices in these types of data. Scholars of slavery may wonder why this additional ramification is important particularly with respect to centuries-old data, but it is our contention that when handling any individual and biographical data which could be connected to a living person, consideration should always be given to applying ethical best practices. This approach rehumanizes the individuals and respects the fundamental dignity which belongs to every person, alive or dead.

4From another perspective, the Frontier Force material is also well positioned to be an example of a dataset that requires technology and digital analysis due to its rich content. The richness of detail in each soldier’s file necessitates the use of databases, and the condition of the documents demands that they be preserved. These data are clear, informative, and each file holds data which would be of interest to a variety of disciplines. There are significant challenges, however, which have caused our team to develop best practices around each of them as we have engaged with them. We have identified the central challenges as ethical, geographic, and organizational. In our haste to collect, preserve, and examine the West African Frontier Force collections held throughout the former British West African colonies, it would be easy to forget that these men whose lives are so clearly recorded may well be the fathers and grandfathers of people alive today. The archives in Freetown hold extraordinarily rich material, and in quantities which are staggering in comparison to the usual dearth of accounts concerning the rank and file Frontier Force troops. Our first challenge is that within these files are many personal details of these men; their medical records often record venereal diseases, and court martial records can reveal behavior that would be embarrassing were the men in question is alive today. There are ethical considerations which we as researchers must bring to bear as we begin the crucial process of preservation and digitization. Without question, these documents hold extraordinary potential—but its potential which must be tempered by conscious efforts to protect the individual privacy of the people whose lives we are studying. Further, these documents represent the cultural heritage of a nation—as we work to preserve it, the issues of ownership and agency arises. We argue that in complex circumstances such as we find in Sierra Leone, new approaches must be employed to respect the archive itself as well as the lives it holds. Our second challenge arises from the methodology we employed as we began to organize the digitized documents. In organizing the dataset, we have developed an approach which permits us to efficiently catalogue individual soldiers’ lives and documents. A rich set of documents like the Frontier Force records represents a situation where we have individual soldiers’ files within which are multiple official forms, each one comprising multiple pages (see Figure 1).

Figure 1. Court Martial of Musa Fula

Figure 1. Court Martial of Musa Fula

Source: West African Frontier Force Collection, Sierra Leone Public Archives, Fourah Bay College, Freetown, Sierra Leone, unclassified document. Photo of the author.

5As we digitize these, we generate a large number of files which must be arranged so that their contents can easily be sorted and analyzed. Our solutions are drawn from museological practices, and the theories of cataloguing which have been developed for artifact acquisition. From the outset of our analysis we realized that given the detail concerning individual soldier’s home villages, we could plot demographics and patterns of recruitment. But this rapidly became a clear challenge as well. With respect to geography, we must carefully consider any effort to map these individual soldiers, as mapping itself poses methodological considerations. A deliberate approach is required for the complex task of reconstructing trajectories, origins and narratives from within this rich dataset. Overall, our team has explored each challenge in turn and collaboratively developed solutions which we propose as best practices in approaching any such dataset. Our proposals must encompass the scholarly applications, archival necessities, and ethical considerations which we argue should govern all digital humanities projects concerning historical lives.

The Archive

6When considering the ethical elements of digitization and preservation of data sources such as this one, it is important to recognize the unique circumstances facing the archive itself. Established in 1895, the Sierra Leone Public Archive was formed to house British colonial government documents. During the Second World War, as the archival space was being used for the war effort, the documents from the archive were transferred to military buildings where they remained until after the war. In the 1950s, Christopher Fyfe was hired as Government Archivist to catalogue, restore and preserve the collection of archives which had been left in disarray and was deteriorating from neglect. In 1965, under the Sierra Leone Public Archives Act, the Public Archives Office was founded and, under the direction of the Minister of Education, was designated to act as custodian of the national archives. Many of the documents in the archive are of vital interest and importance in understanding the trans-Atlantic Slave Trade and are made more remarkable by their survival in the face of social unrest and conflict. Today, the SLPA is located on the fourth floor of the Kennedy Building at Fourah Bay College atop of Mount Aureol in Freetown, the first westernized university established in West Africa.

7Due to location changes during the war, previously well catalogued and indexed collections have become fragmented, leading to a chaotic situation within the archive which is aggravated by chronic understaffing and underfunding. When our team visited in February 2019, the Frontier Force records in particular were without an index, and were held in approximately 60 rice bags kept on the floor around the perimeter of the archive. Each bag held at least one hundred personnel files. Clearly, this is not an ideal method of physical preservation, given the exposure to temperature and humidity in the space where the archives are housed. While the situation seems dire, understanding the underlying causes is important in working toward solutions. The problems faced by the archive are not due to lack of initiative or effort on the part of the team working within it but rather are lack of resources and funds flowing to the SLPA. These make it difficult on an institutional level to properly catalogue and preserve this otherwise rich collection, and the Frontier Force records in particular. The SLPA archivists outlined five distinct challenges to the archives when interviewed, all of which derive from a lack of predictable capital to sustain the organization. The first issue is staffing deficiencies and fair wages for the university-trained archivists who earn very little each month. To put this into perspective, the monthly stipend for an archivist at the public archives is approximately the same as a one-room accommodation for one night at a hotel in Aberdeen in Freetown. The second issue is personal safety and lack of protective equipment to handle deteriorating documents which may have insect infestations or mold, and lack of access to equipment that would prevent dust and chemical exposure. Archivists noted that if they are exposed to illness-inducing levels of toxicity, they are responsible for their own medical bills. The third difficulty concerns education and the fact of being isolated and not part of an international community. There is no institutional budget to pay national or international membership fees, so archivists cannot regularly participate in seminars and training initiatives beyond those provided by generous international scholars on an individual basis. The fourth obstacle is at the level of technology. With ongoing advances in digital technology, cameras able to capture the highest resolution images of degrading documents are essential to preserve materials for future use. Regular and stable internet access is vital to allow cloud storage of digitized data, but this issue is connected with ongoing problems around power generation in the country. Here again, cost restrictions militate against the proper preservation of archival materials. The fifth restriction has to do with the physical archive reaching capacity and the inability to receive new acquisitions. A lack of physical space means that the archive could potentially be turning down new collections that come to them. In addition to the five problems outlined by the archivists, there is another. The condition of the building in which the archives are held does not lend itself to effective conservation. During our 2019 visit there was a broken window by the tables where the team was working to digitize documents, exposing them to dust and other environmental elements from outside, and aiding in the deterioration of the archive’s holdings. These issues are systemic, pervasive, and governed by limitations within the country which cannot be easily met by securing periodical external grants or through personal funds raised by researchers.

  • 3 British Library, “Endangered Archives Programme,” (last accessed, October 2020)

8In recent years, the British Library’s Endangered Archives Programme (EAP) has been actively involved in funding preservation efforts at the SLPA. Four grants totalling just under £78,000 have been awarded to research and digitize records at the SLPA.3 Complex research networks funded in part by these grants have been developed by leading international scholars including Paul E. Lovejoy and Suzanne Schwarz, and these scholars have actively worked to secure ongoing funding to support the archive and seek training for the archivists, including forthcoming grants both from the British Library and Canada’s Social Sciences and Humanities Research Council. Lovejoy and Schwarz’s projects in particular have systematically digitized materials beginning with the earliest documents and moving forward chronologically in an ongoing effort. Under the EAP funding initiative, many historical documents including the Registers of Liberated Africans, Registers of Birth, John Clarkson’s Letterbook and Journal, important Treaties and Agreements, census, military, school and police records, and Railway and Communication Documents have been digitized and shared online to assist research into the histories of Sierra Leone and the trans-Atlantic Slave Trade. While the EAP is a valuable granting approach and the work of international teams has been extraordinary, the Programme does not go nearly far enough and cannot be expected to fill a funding gap—these problems are deeply rooted and solutions must ultimately come from within the nation. No matter how monumental the ongoing work to digitally preserve these documents has been, the hurdles are greater by several orders of magnitude than can be met by any international team or granting agency. The underlying problems of chronic underfunding at the national level, an inventory in disarray and a building in disrepair, are compounded by the fundamental disruption of the civil war, and many documents were left in very bad condition, which has necessitated a triage approach for the teams working in the archive over the past ten years. This is a matter of developing capacity in-country, and working toward a self-sustaining system which is less reliant upon international granting agencies, and funds the archive adequately on a regular yearly basis from within the country, permitting the reorganization and inventory of holdings, and the preservation of the physical documents. The mandate of the EAP is not to conserve the actual physical documents but to digitally preserve the information contained in the archives for future generations and provide access (Case 2015: xliv). As Barry Supple observes:

it was not envisaged that the EAP would support focused work on the physical preservation of material… it was not always easy to maintain the distinction between preserving by sustaining and copying, and preserving by protecting archives through the acquisition of storage materials and other means of physical maintenance of archives (Supple 2015: xii).

9Beyond the issues posed by an exclusively digital approach including the difficulty of ensuring that digital records remain uncorrupt, there is a reflective consideration here: to call an archive “endangered” as the granting Programme does is problematic on a conceptual basis, as it suggests that it needs saving, much like an endangered species. This is at odds with the need to develop capacity to ensure the physical preservation of original records within these nations. And yet the goals of the EAP were established for the conservation of physical documents in archives themselves, but instead focused on the digital preservation and free access to these records. Ongoing efforts by international scholars such as Lovejoy and Schwarz with respect to the support and funding of the physical archives in Freetown are testament to their individual commitment to the incredible collection there rather than representing an aspect of the EAP’s overall purpose. In explaining the function of the EAP in 2008, the Chief Executive of the British Library, Lynne Brindley, stated that it is “to assist researchers and repositories to rescue archives under threat and make them available to wider communities of interest, while retaining them in the region of origin” (Brindley 2008: 67). The problem here is that while archival records may be removed from “threat” through digitization, the archives themselves remain at risk for all the reasons we have outlined. In some ways, the EAP is acting through what film festival scholar Sonia Tascón calls “the humanitarian gaze.” Drawing from Laura Mulvey’s theories on the “male gaze,” the “humanitarian gaze” is a way of looking, expecting to see, and reproducing what is seen filtered through representations of violence and deprivation evoking an emotional and alarmist response (Tascón 2015: 35 –7). The EAP acts out of the fear that these documents will be wiped from existence if something is not done to preserve them now. This is not an unreasonable fear, but digitization is a short-term solution to a systemic challenge and it does not address the physical preservation of materials which have survived for centuries and require care.

10To call the Sierra Leone Public Archives endangered, or even neglected, misrepresents the efforts of the archivists themselves as well as committed international collaborators. These archives are underfunded. The archivists who work there do so out of a pride and affection for the material not necessarily held by researchers who step through the doors. It is their history, or as one archivists, Aiah Yendeh, stated in 2019, it is their “beloved archives.” He elaborated in an interview with our team that “The motivation, I think, is just passion. For me, I see these records as my baby. You know, I love it so much. It’s not always about money. Preservation of our history, you know, to me, is key. That is why I am here, and I have no intention of leaving.” Similarly, archivist Joannes Caulker stated that

[T]hese documents have [great] importance, they have vital records […] The money wasn’t my focus, actually. I was thinking about how best I could promote the institution in my own little way and giving out what I can. So that is what has been keeping me and because of the nature, like I said, of the [archive] I grew a love for it and it’s because of the love that has kept me in the profession up to this time.

11What is also implied in Brindley’s statement concerning digital preservation is the colonial practice of mining for resources; a system where the goods are effectively plundered, giving little credit and support to the nations and institutions where these resources originated. This is the crux of the ethical considerations we have developed: although we respect the incredible importance of digitally preserving these materials through Programmes like the EAP, and the work done by teams of researchers to this point, our team also believes that the best practices that should be developed in this case and others in equally challenging circumstances must also be respect for the dignity of the archive itself, and the continued preservation of original documents. As such, our project is dedicated to developing a clear process and method as we engage with these records, as well as with the physical building and the people who work with them. While still abiding by the British Library standard procedure for digitization to ensure the highest possible quality of image, we believe that digital best practices must also empower the archivists—in this case at the SLPA—to govern their own archives, and that researchers must find ways to support the physical preservation of these materials as well as their digitization. Digital researchers must not only be engaged in the process of preserving the historical documents that are housed within the archive, but also in making relationships with the archivist teams who have worked to preserve them. Lovejoy and Schwarz’s efforts over the past decade are exemplars of this approach, and we believe that all teams should develop close connections and assist in building local capacity to their best of their abilities. For these archivists, the archives are like home. As Jacques Derrida noted, the word “archive” is derived from the Greek word arkheion meaning house, and the archons are the guardians of the documents which reside in the archive (Derrida 1996: 2). Archives are sites of power. What archivists do is “literally creat[e] archives, deciding what is remembered and what is forgotten, who in society is visible and who remains invisible, who has a voice and who does not” (Kenosi 2000: 67). These theoretical considerations and ideas are not abstract in our example. Archives can preserve colonial legacies and maintain histories of unequal access, privileging outside researchers over the people whose documents they are. This is not to say that this is inevitably the case, but the power imbalance is one which we believe must be considered as research teams approach archives which are already facing challenging circumstances. We argue here that the most effective approach in these post-colonial nations must explicitly be to return the power to African control, and to work in harmony with those who are preserving their own cultural heritage. Just as we are working to consider issues concerning individual privacy for the lives recorded within these archives, we believe that the guiding principle in digital humanities projects include explicit consideration and respect for the physical space of the archive and those who work within it.

Approaching the data: cataloguing concerns and access possibilities

12As we develop guidelines on a theoretical level engaging with the ethics of the archive, we also have been developing a meticulous approach to the ingestion and digitization of the personnel files themselves. The variety of types of information held within an individual soldier’s folders is remarkable, both in the sense of types of record generated by the military itself and in the variations we have documented of these records. This poses a series of unique challenges as we organize and transcribe the information within each document. In the course of three years of cataloguing and the digitization of raw data, we have identified over a hundred types of military record. This systematized intake process is necessary not only for this dataset but for other rich sources of data which offer complex problems in terms of their diversity or organization.

13The process of digitizing the original documents happens in situ, following the British Library guidelines given the unique environment of the archive itself. Not every building in which the documents are housed has electricity, and in particular, the Student Centre in which the Frontier Force materials are held is without power. The digitization is done with a black fabric laid out on the archive’s floor, with a color reference to ensure accuracy in reproduction, a digital SLR camera, remote tethering software and controls, and a tripod. Our research teams were trained in digitization at the Harriett Tubman Institute at York University, and all such digitization adheres to them as closely as is possible. Documents are photographed and checked for focus using a remote shooting application on a tablet, and are saved in both RAW and .jpg format to permit the highest resolution as well as quick reference. Over three years of trips collecting this data, we have accumulated over four thousand photographs using this process. The team is now developing ways of organizing and presenting the documents so that scholars and others interested in conducting primary research can access the information in digitized files. A central aspect of that organization hinges on our team’s theories of identity, ethics, and the meanings behind data. Not only do we want people to read the transcripts and the original documents, but we also want to organize what is collected in such a way that people will understand a fundamental theme of our work: before the image, before the paper, even before the moment something happened and that someone decided to write it down—there were people. People living their lives, within a society, with complex relations and organization. The document will never fully explain that particular life, but considered in the context of other documents, oral history, and visualization, scholars and the general public can potentially understand more of the connexion between past peoples and modern cultures.

14To achieve this goal, it is fundamental to know the content of the files and to catalogue the information. Only then is it possible to create and to provide tools to use the raw data and produce knowledge. The cataloguing process itself requires a strategy to match the users’ needs (Rodrigues 2006). Therefore our team works through meetings and multidisciplinary discussions to conceive what would be an ideal tool to analyze and visualize these documented lives. What possible cross- searching tools, maps locating the data, web exhibitions and user-friendly access can we implement to ensure a visually pleasing and intuitive system? Facing thousands of digitized files divided by the year they were made, we have identified the main hurdles: how to approach the files ethically? How can we keep the information interlaced and connected without dismembering the folders and files? How can we keep all the files together if we have medical records within other documents and need to apply an anonymizing filter?

15We decided to approach each person’s personnel file as one object for the purposes of organization. This means that all the individual documents of one person are seen as subdivisions of the object, each file receiving a sequenced number, the first one being “0001” and the others following. Each such entry is in effect a digital folder, identified by the army number and the person’s last name. Within it, we input the files replicating its original sequence and so respect any meaning that the original sequence may have had. A problem we haven’t solved yet and which we are still discussing is the number of the whole object and collection in our catalogue. We understand that for privacy reasons, the last name and army number can’t be the main information provided to the general public. We recognize that those who have personal, familial connections to a given soldier whose documents we have digitized would reasonably expect access to their family member’s biography. We are therefore working to implement procedures relating to concerns about access and privacy which respect family connections while preserving the dignity of the person behind the data. One possibility is to create an alphanumeric number for each such object. This would mean that by looking at the inventory of alphanumeric Frontier Force data, researchers would immediately ascertain which collection the file is from and its origin—this has the potential to allow further research into West African Frontier Force documents in different archives beyond the Sierra Leone dataset. It would also show us how many objects we have and how many subdivisions each one has. The alphanumeric number would start with the collection, “WAFF” for West African Frontier Forces, followed by its origin, “SLPA” for Sierra Leone Public Archives. The object receives a number (according to the order of appearance), and its files have the sequenced number (beginning with “0001”). So, as an example, the first file of the first object, the alphanumeric number would be “WAFF_SLPA_I_0001”. If the team finds some West African Frontier Force folders in the National Archives of Nigeria, for example, the first object with its first file would be “WAFF_NAN_I_0001”.

16Each object, meaning each personnel file, has a form created as it is digitized to provide metainformation and allow for data management. This can be accessed by the team manager of this project, with some sections available to the public. Metainformation around data management begins with a section of identification with a catalogue number, army number and last name, and original (physical) source. There would be also the year the data was recorded; the researcher responsible for the digitization; date of last edition in the form; how many subdivisions the form has. Then a table showing the file’s alphanumeric number in the first column, the title of the military record in the second column, and the third column showing which documents within the personnel file are confidential. The last column to be filled would be completed if a personnel file includes additional data pertaining to research projects—in our case, body-marking information. We are proposing that the section would end with a contextualization that would explain in a less formal way what the files contain about that person, and that would be available in the platform for general users.

17Following the identification section, we are developing an Object History element. Based on Peter Van Mensch’s dimensional matrices and his approach to the incorporation of artifacts into museums (1987), it is being organized with one designated space for physical properties (information on the material that supports the information), a second for meanings and roles (primary, second and metaphysical) and a third for the object’s history (where it was produced, who filled it; its first uses and how it ended in the archive (object path/track) (Mensch 1987). In addition, we are developing a Management of the Object section. Ideally, we would have automatically updated data on the use of the object and, specifically, its splits. As an example, if there were a video that appears in one file, it would appear in the object’s documentation form and would track access and views by the general public in the platform. This would permit real-time harvesting of meta-data around the uses of harvested and curated information by researchers and the general public. In this section there is also information on current study-projects that are using the object, references to studies published, uses in web exhibitions, blog posts, educational activities and any other type of activity.

18It is important to keep in mind that the catalogue and the information about each document will keep growing. Not only will the team continue collecting raw digitized files, but we intend to harvest metrics around how many people open the files, which files, and who the people are who are accessing the documents. One way of achieving this is for users to access our platform through a social media, such as Google or Facebook, sharing demographic information (gender, age, country). They could have a simple profile enabling them to have easy access to what they have found and “saved”, the reports they have created through the advanced searching tool.

19But how can we make the data more approachable and accessible? We aim to provide a relevant and useful platform for scholars and researchers which at the same time engages students of elementary and middle schools, the public, and all of those interested in non-formal ways of learning. We must also continue to respect the dignity of the people whose data this is, which means working to anonymize certain information to preserve privacy. We are still studying ways of achieving this, through Patrimonial Education, Museology approaches, Public History, and Web Design combined in an interdisciplinary approach (Saye 2015). Our team is also collaboratively developing protocols to manage the often sensitive information which these personnel files include. The medical records in particular would not be entirely accessible to researchers nor to the general public, but they would be catalogued. While we fully recognize the importance of studying medical history—particularly military medical history—the team is developing a different approach to the data. We intend to have all information transcribed and organized in metadata, which will allow researchers to access reports which would be anonymized and cleaned of identifying individual information. Those reports would be created based on the advanced search tool, with the term searched and period of time, resulting in quantified information. This might be exclusively applied to medical information or even expanded to other types of document. Our purpose here is to preserve the dignity of individual soldiers while ensuring access to importance data points for statistical analysis.

The Challenges of Mapping and a Colonial Source

20The West African Frontier Force material offers scholars and researchers an extraordinary glimpse into the careers and lives of enlisted Africans. Files also permit possible demographic analyses, as every attestation includes information which locates the recruit in time and space. It would seem simple enough to develop a map of these clearly indicated origins and homes, and determine distribution of recruits from particular provinces, districts, and villages. But in practice it proved to be considerably harder than we had anticipated.

  • 4 Recent scholarship such as Henry B. Lovejoy’s contributions to the methodology of mapping have bee (...)
  • 5 Beyond honest inaccuracies, there are considerable issues with maps which consistently under-repre (...)

21African geography drawn from colonial sources has been a consistent problem for cartographic scholars.4 Mapping even in the supposedly rational nineteenth century was ultimately propositional in nature (Bassett & Porter 1991: 370). Maps are at their core visualizations which serve to shape our collective idea of geography. As a medium, cartography is represented as authoritative and accurate, but until the age of satellite mapping, cartography was often not as objective nor as true as viewers might imagine.5 The debate over the ultimately imaginary “Kong Mountain range” is one example of a case where cartographers and geographers’ efforts to define and reify the land they saw was filtered through theories, inherent biases, and debate (Bassett & Porter 1991: 381). It was assumed that there had to be a mountain range to explain the transition between the Sahara desert and the forest belt, and European cartographers ultimately decided that the “Kong Mountains” stretched across the continent to explain this phenomenon; of course, they do not actually exist! They were first created in 1798 on a map of the continent which relied on accounts by Mungo Park, and were replicated on all subsequent maps. These fictional mountains existed purely in the realms of imagination from the early 19th century until eventually dispelled after the late 1880’s expedition of Louis-Gustave Binger (see Figure 2).

Figure 2. The Kong Mountain Range

Figure 2. The Kong Mountain Range

John Cary, A New Map of Africa, From the Latest Authorities, London, J. Cary, Engraver & Map-seller, 1805. Source: Wikimedia Commons, Licence CC-PD-Mark.

  • 6 As Stickler explains, however, often this pragmatic need to reduce the complexity of towns in sett (...)

22More often than not, in fact, explorers’ and geographers’ mapping relied heavily on local knowledge of the regions being mapped. African scholars have described many of the challenges which European explorers faced when attempting to grasp spatial relationships in Africa (Jones & Voigt 2012: 12). Mapping served to articulate a European-dominated perspective of the continent as a largely empty land in such narratives. Traveller accounts, which included African accounts of geography and regions, had to be compared carefully to European accounts for veracity, but all such accounts were relying upon accessibility of terrain, and the so-called reliability of African informants (Fritsch 2009). Maps are perceived as unbiased and neutral but, particularly during the colonial era, they served to reinforce biases. If a map published by a reputable press did not indicate that a place existed, then it did not exist. All maps distort reality to some degree, as they are ultimately abstractions which attempt to reduce a complex reality into analog clarity (Stickler 1990: 329).6 Few maps prior to the present day allowed enough space to accurately reproduce smaller settlements, which often led to the erroneous conclusion that regions of a given map were uninhabited.

23West Africa is one of the densest geographic regions on the continent with respect to the variation of ethnic groups, language, religion, and heritage. Attempts to map this region, especially historically, have proven extremely difficult, as the density and turbulence of West Africa makes tracing groups and states challenging. The sheer amount of variation in the area has historically led to immense changes over short periods of time, quickly making previously up-to-date maps obsolete, and as H.B. Lovejoy has noted, historical maps of the region depended on misguided ideas of the geography, inaccurate accounts, and incomplete data. Once Europeans began to map the region in pursuit of colonial goals, they did so with little sensitivity to accuracy. European records were sloppy, inaccurate, and often filled with misinformation. Moreover, European cartographers were reliant on not only African informants, but also on well-trodden African trade paths and networks (Fritsch 2009: 89). Colonial European explorers were stepping into a region with its own spatial understandings already at work, intent on making names for themselves back in Europe as discoverers, relying on African narratives and help. Some European travellers recorded the ways in which local knowledge and approaches to mapping differed from their own, anticipating further complications that we would encounter in our own effort to accurately place Frontier Force birthplaces reliably onto a map of Sierra Leone. Maps of the colonial era represent a shift in the way that Europeans approached Africa on a conceptual level. They reflect the establishment of administrations on the ground which demanded the identification of local populations but were otherwise often wildly inaccurate (Stone 1988: 59). This lack of accuracy becomes more serious when dealing with identities and origins during the slave period. Our team approached the colonial period with care, initially assuming that as more maps were being produced at that time for purposes of expansion and exploitation, we might have more material to work with. Instead, as we rapidly discovered, the issues cartographic scholars have raised with mapping and colonialism made achieving accuracy extremely difficult.

24One of the central requirements of our project is to trace and map the trajectories of the individuals who served in the Frontier Force. This is a fundamental prerequisite for the visualization of demography and analysis of enlisted Africans. The data within each soldier’s attestation is helpful information, but once again, the colonial officers who recorded this information were prone to mis-identification characterized by misspellings or arbitrary information. This has presented specific and unique problems as we attempted to trace the geographic origins and destinations of these individuals. It is of the utmost importance to outline and approach these challenges carefully, to be aware of cultural sensitivities and change, and to work towards finding the best possible outcomes. Some complexities in mapping have already been laid out, but it is important to explore them in more detail. Even using current best practices in QGIS, there are problems, as “within this scope of an immeasurable amount of missing data, the maps… must always be treated as approximations because of inexact and missing data, limitations in software functionality and any errors of interpretation of the sources” (Lovejoy 2019: 129). Accurate historical mapping even with the best possible tools is no small challenge. Regional conflicts during the precolonial period, as well as during the era of the slave trade, saw kingdoms and smaller states rise and fall on a regular basis. Of course, the invasion and formal colonization of much of West Africa in the 19th century saw borders redrawn, states abolished, and ethnic groups either obliterated or forced into common larger groupings. The cartographic face of West Africa was in regular flux and not merely in the context of uncertain maps and inadequate data.

25The source material itself for developing a clear and accurate map of Sierra Leone from the appropriate period is problematic. When the Europeans arrived, there was little care taken to preserve the history of the region, which was largely handed down by local traditions and oral communication. With the division, separation, and exploitation of many groups, the oral source material was lost, making rebuilding the exact geography difficult. There were certainly attempts at mapping the region, but many of the European efforts, without the contextualization of local traditions, were inaccurate. This, again, is because earlier Europeans largely remained along the coast to trade slaves and other goods, and took extreme liberties in mapping the interior. Moreover, later cartographers would work with these older maps and base their new renditions on that misinformation. In effect, errors were reified and often magnified as maps were drawn from previous maps rather than from personal experience. The Kong Mountains provide the best example of mistakes becoming normalized.

26In specific terms, as we tried to reconcile the Frontier Force data with existing maps of Sierra Leone, the largest immediate problem we faced was the mislabelling and misspelling of toponyms. If information is not accurate, it can very quickly make a map obsolete. The notes of the colonial officers and cartographers used to produce many presumably authoritative nineteenth century maps were regularly inconsistent, and were often based on phonetic representations of an original word. Moreover, colonial-era maps were often generated in comparative haste, necessitating Colonial Surveys which relied on District Officers, themselves amateur cartographers at best. At their best, these mapmakers required professional expertise they often lacked, the maps themselves necessitating cadastral surveys intended to define alienated land, townships and building plots, roads, and railways in order to administer and establish control (Stone 1988: 60–1). Such surveys took time and care, and as a result of the fragmented expertise available, the colonial period represents an often disjointed approach to mapping in Africa caused by the pressures of administration and the exertion of colonial authority. Colonial maps for Sierra Leone offer numerous examples of this disjunction, not the least of which is the irregularity of ethnonyms applied to particular regions and the inconsistent naming of settlements and villages. For instance, the Temne ethnic group would often be labelled as “Timmannee” or “Timene,” or the Vai as “Foy” or even “Gallina.” Given the complex tapestry of ethnic groups in the region, this often blurred the lines for colonial officials, who had very little prior knowledge of these groupings. These are some of the clearer examples of misunderstanding to be found in European notes and writings. Despite the descriptions of home villages in the Frontier Force attestations, when we tried to locate the villages in questions, some proved entirely untraceable.

Figure 3. “Map of Sierra Leone (Provisional Issue), Compiled in the Intelligence Division,” London, War Office, 1901 [1st ed. 1898].

Figure 3. “Map of Sierra Leone (Provisional Issue), Compiled in the Intelligence Division,” London, War Office, 1901 [1st ed. 1898].

© 2018 Barry Lawrence Ruderman Antique Maps.

27For our project, we began with a map (see Figure 3) that we selected primarily because of the relative time- period in which it had been drawn, and the considerable number of small villages that we found named on it. It is a typical example of a colonial-era map and includes a remarkable number of small settlements and villages, primarily situated along the waterways which colonial officials could readily navigate. This map also represents a combination of the multiple issues we have identified. The survey was prepared by E. de P. O’Kelly in 1897–1898 and was published in 1898, which makes it one of the earlier extant maps of Sierra Leone’s interior. Previous maps focused on the coast and largely neglected the kingdoms inland. This one was important because it was published only a short time after the Paris Treaty of 1895, which formally established the colony’s boundaries. Even more significantly, the map was created at the outset of the Hut Tax War of 1898, fought between Sierra Leoneans and the British over severe taxation of local populations. In the end, that conflict decimated the population, and quite literally wiped entire villages off the map (Abraham 1974). Because of the results of the 1898 war, as well as other colonial pressures that we have outlined, subsequent maps do not provide the same level of detail as the 1898 version, and many villages can no longer be found. This is only one part of the issue, however. The 1898 map, while much more useful in its detail, falls victim to the mislabelling and misspelling typical of similar colonial maps. Many of the village and town names are entirely misspelt or placed in the wrong location altogether. The shifting borders, which were being negotiated at that time and only settled into their modern configuration in 1912, make it nearly impossible to map smaller settlements in the present-day border regions with Guinea and Liberia. These are just some of the hurdles facing us when it comes to mapping the origins of the Frontier Force individuals. The intricate combination of density, colonialism, and cartographic biases in the region have greatly impeded scholars’ ability to accurately map this period of Sierra Leone’s history. It is difficult to retrace and follow the trajectories of individuals when there is little chance of locating their origin in the first place.

28Our project approaches these challenges in a variety of ways. First, we are fortunate that the Frontier Force files contain a large quantity of detailed information. This allows us to work with a number of factors when searching for an individual’s location. If their village or town name is misidentified, we are still able to deduce a more approximate location based on the other information. For example, in the event that a village name is incorrect, the individual’s province and chiefdom are still available to work from. The chiefdoms still largely reflect those established at the end of the 19th century, which is beneficial to us. Once we have located the areas in question, we are able to search more closely on present-day maps to try and determine an approximate location for a village. Best practices would therefore require us to analysis derived from many maps spanning many years, prioritizing the earliest and most detailed, and using more recent maps to cross-reference. This is in line with Lovejoy’s methodology and use of multiple primary and secondary source-maps (Lovejoy 2013: 444). While this does not always succeed, we are able at least to deduce what region or chiefdom of the country these individuals originated from, which is still entirely useful data. Modern mapping technology, particularly satellite-based examples like Google Maps, is useful in many cases when it comes to mapping smaller villages. However, it is still necessary to conduct research into possible alternate names and spelling, so that searches will prove fruitful. This project is about tracing origins with the tools we have, and thanks to newer technologies we are drawing steadily closer to our goals. For the time being, however, the methods we are developing require this careful cross-referencing system, and are time-consuming and inefficient.

29A major element to our approach to the mapping of individuals onto geography is the importance our project places on awareness of potentially neocolonial processes. Our efforts require ongoing collaborative discussion with those whose countries and histories we are mapping. Their experience within the regions, and their access to traditions which may clarify the ways that placenames have changed, are crucial parts of reconstructing Frontier Force soldiers’ lives. Working to share ideas with scholars and archivists on the continent also ensures that there is ongoing African input concerning the cultural histories which after all originate in and belong to Africans.

30The difficulties of mapping West African history, especially in regard to ethnicity, religion, and language, are well established. Descriptions of locations, groupings of people, or other ethnic signifiers were largely written and produced by Europeans, many of whom were colonial officials. The sources are often misleading, inaccurate, misspelt, or incorrectly labelled, with the result that reliable sources became ever more difficult to find. Colonialism and subsequent postcolonial conflict in West Africa has led to the loss of much useful source-material as well as to the production of problematic, frequently inaccurate, and sometimes biased maps.

31We have explored our specific project and how it relates to the visualization of West African geographic history, and in particular the complexiy of situating enlisted Africans in relation to their birthplaces. With a close reading of the Frontier Force records, we are beginning to piece together information and to map the locations of soldiers. But this is only a small contribution. We have taken steps to work with connections in Sierra Leone itself, which is a vital methodological approach to mapping history. Without the assistance of local West Africans, and their valuable input as members of the communities whose geography we are trying to rebuild, all efforts would be in vain. The people know their own societies and cultural traditions better than anyone; their involvement is therefore of paramount importance. It is not that we should work with West Africans in attempting to map the historical geography of the region, but that we must work with them to successfully and honestly retrace what has been lost. It has been written that “history is both a discourse of knowledge and a discourse of power” (Mudimbe 1988: 188). Rebuilding the past is no different. We must share that power equally and avoid straying once again down a colonial road.


32Even as maps have served to define colonial power, so too have archives served to preserve selective evidence. Just as the blank space on an inaccurate map can neglect or obscure African power in a framing of narrative, so too have archives held the potential to frame national narratives in how they preserve and what they preserve. Our team is therefore considering these elements as we develop a process which may be applied to similarly complex collections. We are working to develop a best practice which recognizes the two forces at play here. Without question, it is important to preserve important documents such as the Frontier Force collection for future generations, but at the archival level we must move beyond mere preservation and towards empowering the archivists and giving them agency to act as guardians of their own records, keepers of their own house. In this way, the project will continue to engage in discussion with the archival team to develop initiatives and partnerships which address the needs they have expressed to us. The eventual ideal outcome would be that even in the absence of privileged scholars arriving with significant grant monies, the archive can preserve and conserve the histories of their nation on an ongoing basis. More training and dialogue are planned for when the team returns, and projects are being developed which would fuel capacity-building and autonomy. While researchers may have received grants for their own research, there is often little to spare for conservation efforts. As regards raising salaries, hiring more archivists, and designating more space and adequate conditions for documents, this an evolving process. Helping the archivists in their mission to validate the importance of this national treasure is a step in the right direction. Individual project teams can only do so much, but as we collectively develop a method which not only emphasizes the importance of data, but prioritizes people, we believe that we can do much more than salvage history.

33This development extends to efforts to map history and origins; as with the archive, it is crucial for scholars to consider all sources of information including oral histories and local knowledge. This is integral to the practice of restoring control to the descendants of those whose lives these documents describe and integral to our methodological practice as we continue our own project and research. Our central contention with respect to best practices is this: that any and all work being done on the continent must begin with a theoretical consideration of ways of developing and facilitating agency in a way that is not just non-colonial but actively anti-colonial in its approach. The data which we digitize belongs to the nations within which it is found, and the stories we work to tell as historians demand that we dignify the people whose stories they are. Otherwise they vanish into obscurity, like Musa Fula, whose brief experience captured within data from the Frontier Force represents the only glimpse at a life. As the son of a chief who was wrongfully accused and convicted of cannibalism in 1912, but who clearly returned to father him, Musa Fula’s story deserves more remembrance than a trial for petty theft in 1944 and then dust-choked silence.

Haut de page


Abraham, Arthur, 1974. “Bai Bureh, The British, and the Hut Tax War,” The International Journal of African Historical Studies, No. 7/1, pp. 99–106.

Bassett, Thomas J. & Philip W. Porter, 1991. “‘From the Best Authorities’: The Mountains of Kong in the Cartography of West Africa,” The Journal of African History, No. 32/3, pp. 367–413.

Brindley, Lynne, 2008. “The International Dimensions of Digital Science and Scholarship: Aspirations of the British Library in Serving the International Scientific and Scholarly Communities,” in Rae Earnshaw & John Vince (eds.), Digital Convergence—Libraries of the Future, Springer, London, pp. 65–73.

Case, Anthea, 2015. “The Endangered Archives Programme after ten years,” in Maja Kominko (ed.), From Dust to Digital: Ten Years of the Endangered Archives Programme, Cambridge, Open Book Publishers, p. XLIII-XLVI.

Derrida, Jacques, 1996. Archive Fever: A Freudian Impression, Chicago / London, University of Chicago.

Fritsch, Kathrin, 2009. “‘You Have Everything Confused And Mixed Up...!’ Georg Schweinfurth, Knowledge and Cartography Of Africa in the 19th Century,” History in Africa, No. 36, pp. 87–101.

Jones, Adam & Isabel Voigt, 2012. “‘Just a First Sketchy Makeshift’: German Travellers and Their Cartographic Encounters in Africa, 1850–1914,” History in Africa, No. 39, pp. 9–39.

Kenosi, Lekoko, 2000. “Managing the Records of Restricted organizations,” Information development, No. 16/2, pp. 65–9.

Lovejoy, Henry B., 2013. “Redrawing historical maps of the Bight of Benin Hinterland, c. 1780,” Canadian Journal of African Studies/Revue canadienne des études africaines, No. 47/3, pp. 443–63.

Lovejoy, Henry B., 2019. “Mapping Uncertainty: The Collapse of Oyo and the Trans-Atlantic Slave Trade, 1816–1836,” Journal of Global Slavery, No. 4/2, pp. 127–61.

Mudimbe, Victor Y., 1988. The Invention of Africa: Gnosis, Philosophy, and the Order of Knowledge, Bloomington, Indiana University Press.

O’Kelly, E. de P., 1901 [1898]. “Map of Sierra Leone (Provisional Issue), Compiled in the Intelligence Division,” London, War Office.

Rodrigues, Ana Marcía L., 2006. “A teoria dos arquivos e a gestão de documentos,” Perspectivas em Ciência da Informação, No. 11/1, pp. 102–17.

Saye, Faye, 2015. Public History: A Practical Guide, Indiana, Bloomsbury.

Stickler, P. J., 1990. “Invisible Towns: A Case Study in the Cartography of South Africa,” GeoJournal, No. 22/3, pp. 329–33.

Stone, Jeffrey C., 1988. “Imperialism, Colonialism and Cartography,” Transactions of the Institute of British Geographers, No. 13/1, pp. 57–64.

Supple, Barry, 2015. “Preserving the past: creating the Endangered Archives Programme,” in Maja Kominko (ed.), From Dust to Digital: Ten years of the Endangered Archives Programme, Cambridge, Open Book Publishers, pp. XXXIX-XLI.

Tascón, Sonia M., 2015. Human rights film festivals: Activism in context, Springer, London.

Van Mensch, Peter, 1987. “A structured approach to museology,” in Peter Van Mensch (dir.), Object, Museum, Museology, an Eternal Triangle, Leiden, Reinwardt Academy.

Haut de page


1 Fula’s name is spelled throughout his personnel file variously as Mussa Fula, Musa Fula, or Musa Fulla.

2 Fula’s paternity is of interest because his father was one of the main suspects in the 1913 Special Commission Court cases which were heard in Gbangbama, Sierra Leone. They concerned claims of cannibalism and a society of men in the region called the Human Leopards, and as a group the cases display competing interests and ongoing tensions in the Protectorate there. For more, see Keefer, Katrina, Under the Leopard Skin, forthcoming.

3 British Library, “Endangered Archives Programme,” (last accessed, October 2020).

4 Recent scholarship such as Henry B. Lovejoy’s contributions to the methodology of mapping have been of considerable help for our own project. In the next phase of our project, we will be applying his QGIS methods to address the issues raised in this article. See Lovejoy 2019.

5 Beyond honest inaccuracies, there are considerable issues with maps which consistently under-represent African settlement sizes, reinforcing colonial biases and perceptions. See Stickler 1990.

6 As Stickler explains, however, often this pragmatic need to reduce the complexity of towns in settler societies such as South Africa led to the excision of African towns, while settler towns remained, producing a false and biased view of reality.

Haut de page

Table des illustrations

Titre Figure 1. Court Martial of Musa Fula
Crédits Source: West African Frontier Force Collection, Sierra Leone Public Archives, Fourah Bay College, Freetown, Sierra Leone, unclassified document. Photo of the author.
Fichier image/jpeg, 260k
Titre Figure 2. The Kong Mountain Range
Crédits John Cary, A New Map of Africa, From the Latest Authorities, London, J. Cary, Engraver & Map-seller, 1805. Source: Wikimedia Commons, Licence CC-PD-Mark.
Fichier image/jpeg, 1,9M
Titre Figure 3. “Map of Sierra Leone (Provisional Issue), Compiled in the Intelligence Division,” London, War Office, 1901 [1st ed. 1898].
Crédits © 2018 Barry Lawrence Ruderman Antique Maps.
Fichier image/jpeg, 1,6M
Haut de page

Pour citer cet article

Référence électronique

Katrina Keefer, Eric Lehman, Michael McGill et Gabriela Mattia, « Documenting and Digitizing with Dignity: Ethical Considerations and the West African Frontier Force Personnel Records »Esclavages & Post-esclavages [En ligne], 3 | 2020, mis en ligne le 27 novembre 2020, consulté le 29 mai 2023. URL :

Haut de page


Katrina Keefer

Adjunct professor, Department Of History, Trent University, Canada

Eric Lehman

Ph.D. Candidate, Department of Canadian Studies, Trent University, Canada

Michael McGill

Department of History, Trent University, Canada

Gabriela Mattia

Universidade Federal do Rio Grande do Sul, Brasil

Haut de page

Droits d’auteur


Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search