Decolonising Description: Addressing Discriminatory Language in Scottish Public Heritage and Beyond

Decolonisation and UK Heritage

Put simply, decolonisation is a restorative justice movement that seeks to dismantle the systems, institutions, and ideologies of colonialism that still serve to perpetuate inequalities today. As the cross-disciplinary body of scholarship known as critical race theory has illuminated, the ‘saturating presence of discrimination within social institutions and their practices and behaviours’ preserves the unequal existence of ‘dominant and disenfranchised cultures, communities, or social groups’ in the UK.⁶ Although the issue of race inequality sits at the core of the decolonialisation project, racism and its systemic structures cannot be disentangled from other prejudices of the British Empire, including but not limited to patriarchy, heteronormativity, classism, ableism, anti-Semitism, and xenophobia.

The need to decolonise the UK heritage sector stems from the fact that its structures and methods are rooted in colonial and imperial histories, ultimately serving the interests of those who held power and privilege in the past.⁷ What this means is that the behaviours and practices of the British heritage sector, including descriptive and cataloguing practices, traditionally functioned to prioritise the perspectives and experiences of the white, the male, the upper/middle class, the heterosexual, the cis-gender, and the able-bodied. A wealth of literature on archival silences has shown that the voices of underprivileged groups have been excluded or erased as a result of this colonial bias.⁸ In the past two decades, the absence of these narratives from mainstream archival and heritage records has led to a rise of independent community archives in the UK, through which marginalised peoples have sought to actively document their own heritage.⁹ The UK public heritage sector, therefore, must decolonise its protocols and procedures in order to repair its relationships with groups who have not only been excluded from heritage spaces in the past, but whose needs have been almost entirely disregarded by the sector.

The ‘overwhelmingly whiteness’ of Scotland’s cultural institutions has attracted especial critique in recent years, with the absence of diverse ethnic representation across the nation’s galleries, libraries, archives, and museums being explicitly visible.¹⁰ This is not a uniquely Scottish problem, but one that plagues the UK heritage sector. A study of the UK information workforce in 2015 recorded that 96.7 per cent of employees identified as ‘white’.¹¹ Whilst my work seeks to address all areas of inequality in heritage, the sector’s race problem warrants particular attention. During the nine months of my internship in which I was engaged in remote conversations with various heritage organisations across the UK sector, I was overwhelmed by the experience of talking about issues of diversity and inclusion in meeting after meeting, at conference after conference, where I was the only person of colour present. Like many others from marginalised ethnic backgrounds in the UK, I experienced the sector as a hostile and isolating work environment.¹² What was so alienating was that I regularly experienced microaggressions in situations where multiple other people were present, yet nobody else expressed discomfort or offence. These incidents happened below the level of awareness of even my well-intentioned white colleagues because they were not trained to identify, through first-hand experience or education, the ‘commonplace daily verbal, behavioral or environmental slights, whether intentional or unintentional, that communicate hostile, derogatory, or negative attitudes toward stigmatized or culturally marginalized groups’.¹³

The decolonisation of language is, therefore, just one aspect of the broader effort that needs to take place to dismantle the sector’s inherited colonial structures and legacies. Inclusive description work will be superficial unless it is accompanied by a cultural shift in attitudes across the sector and greater representation of people from marginalised backgrounds in the workforce, particularly in higher pay grades and positions of management. From reviewing recruitment policies to increasing the accessibility of building spaces, abolishing unpaid internships, improving job security, offering professional development relating to issues of equality and equity, and creating more effective support systems for employees engaged in emotional labour, the sector needs to be more accommodating towards people from disadvantaged backgrounds.¹⁴ As part of this multi-faceted project, we need to see institutions acknowledge the provenance of objects and materials that were acquired through colonial aggression or the profits of Europe’s transatlantic trade in Africans along with the increased restitution and repatriation of stolen cultural artefacts and human remains currently held in the UK.¹⁵ Only through committing to a culture of transparency and accountability over a prolonged period of time can heritage institutions hope to regain the trust of their under-served communities. To remain relevant to contemporary society, memory institutions will need to build reciprocal relationships with marginalised groups to understand their needs and better serve their interests, which must include remuneration for their time and labour when appropriate.¹⁶ In his Black Lives Matter Charter for the UK Heritage Sector, Errol Francis also adds that ‘Arts and heritage organisations should devise programmes that appeal to everyone in our society by commissioning contemporary diverse artists and curators to address the history and present-day issues around racism, prejudice and social exclusion’.¹⁷

Despite a valiant transnational effort to debunk the myth that heritage institutions uphold some kind of ‘suprapolitical objectivity’, many in the profession continue to believe in the ‘neutrality’ of the sector.¹⁸ As public institutions, galleries, libraries, archives, and museums are inherently political. Mike Murawksi and La Tanya Autry, founders of the Museums Are Not Neutral campaign, have commented on how ‘the claim of neutrality fosters unequal power relations’. As they have explained, ‘what museums take for granted as “neutral,” “objective,” “normal,” “professional,” and “high quality” is all part of a status quo system … that perpetuates oppression, racism, injustice, and colonialism’.¹⁹ Ultimately, curators and cataloguers make decisions about ‘which records, will get full, partial, or no archival attention’.²⁰ In this way, public heritage plays an active role in constructing a nation’s collective memory and it opens up spaces where power struggles play out.²¹ South African archivist Verne Harris articulates that heritage is ‘a battleground for meaning and significance … A place and a space of complex and ever-shifting power plays.’²² Not only is the action of collecting, curating, and preserving ‘historically significant’ records or artefacts riddled with selective biases, but the act of conferring significance and authority on those materials through description and cataloguing is innately subjective.²³ In the words of Randall C. Jimerson, ‘Far from being a neutral repository for recorded memory, archives (and archivists) actively mediate and shape the archival record’.²⁴

The Power of Description

On top of selecting which materials are and are not worthy of acquisition and preservation, the main way through which curators and cataloguers shape the public’s understandings of history is through writing description. In the literature on archival science produced in the 1970s and 1980s, ‘description’ is defined as the process of ‘establishing intellectual control over holdings’; it is the archivist’s role to capture, collate, analyse, and organise any information that will help ‘interpret the holdings’ and ‘explain the context’.²⁵ The function of catalogue metadata is to bring to the surface the most essential information about an item or artefact to provide users with the context that they need to discover relevant resources.²⁶ When creating information about historical records and artefacts, curators and cataloguers must make subjective decisions about what information is deemed culturally and politically ‘important’ and what terminology best reflects the communities that are represented. Wendy M. Duff and Verne Harris note that ‘When describing records archivists will remember certain aspects and hide or forget others. They will highlight some relationships and ignore others’.²⁷ Evidently, description – whether created for finding aids, website resources, or exhibition text – is a tool for controlling the historical narrative, and it is biased in its nature. Put concisely, ‘Every representation, every mode of description, is biased because it reflects a particular world-view and is constructed to meet specific purposes’.²⁸

Since the 1970s, a vast body of postcolonial scholarship has highlighted the imbrication of power and knowledge production, demonstrating the violent role that language has played within the context of the British Empire. There is power in textual representation. When a text paints a derogatory and dehumanising image of an entire group of people, it can serve to legitimise hatred towards them. In his scholarship on ‘Mau Maus of the Mind’, for example, John Lonsdale explores how Kikuyu resistance fighters were constructed in the British imaginary as an epitomic emblem of African barbarism and savagery.²⁹ British government-printed reports that described the Mau Mau as ‘obscene’ and ‘bestial’ in nature, and which purported that Kikuyu men and women could not be purged of ‘the filthy thing Mau Mau’, were used to justify the creation of British torture camps in Kenya in the 1960s and to legitimise the brutal suppression of the anti-colonial movement in the public eye.³⁰

Throughout the history of British imperialism, this kind of literary mechanism has served to rationalise the dispossession of people and the stripping of their autonomy, often through violent means.³¹ Exemplified by Edward Said in his ground-breaking text Orientalism (1978), the Western imperial project produced a large body of knowledge about non-Western peoples, denying them the right to represent themselves.³² The Enlightenment gave birth to the idea that only Western empirical knowledge was ‘true’, and within this context the written word was imbued with especial power.³³ The ‘unequal control over historical productions’ led to inaccurate and biased portrayals of non-Western societies and cultures as ‘Other’, resulting in false representations and erroneous stereotypes that still pervade Western thought today. The colonial project of ‘Othering’ not only relied on explicitly racist depictions of non-white savagery, primitivity, and sub-humanity, but also more subtle portrayals of the Other’s innate differences. Exoticised and romanticised representations of non-Western peoples as ‘relics of the past’ who have submissive, child-like, and innately ‘feminine’ characteristics are equally stigmatising.

With their historic involvement in Europe’s transatlantic trade in Africans and the racial sciences being well-documented, it is no secret that older British heritage institutions were themselves ‘an instrument of colonialism’.³⁴ Consciously or not, descriptions and subject headings written and used by cataloguers and curators often reproduce the semantics of difference embedded in colonial discourses of power. When heritage institutions replicate these colonialist fallacies today, they not only contribute to the circulation of misinformation, but they perpetuate the violence of misrepresentation. To quote Shiraz Durrani and Elizabeth Smallwood, ‘[librarians] have a social responsibility to ensure that people get correct information. It is a matter of ethics that they challenge misinformation’.³⁵ When we apply ‘essentialized, ahistorical categories and labels’ to entire groups of people, we risk the creation of distorted information and the reproduction of harmful stereotypes.³⁶ When our language is homogenising, we cause additional harm by erasing diversity and dishonouring people’s self-identities. As summarised in Colleen McGloin and Bronwyn L. Carlson’s study of the politics of language in Indigenous Studies at the University of Wollongong, Australia, written text has the ‘capacity … to construct reality; produce ideas, beliefs, and stereotypical representations, represent, misrepresent, and imbue readers and viewers with particular views about the world.’³⁷ From a human rights perspective, moreover, Stacy Wood et al. noted that

archival description and recordkeeping more broadly have been identified by both archival scholars and government inquiries as key agents in the oppression, marginalization, silencing, alienation and traumatization of individuals and communities that have been involved in social justice and human rights movements, for example, through how acts and victims are classified, euphemized, or submerged.³⁸

The discipline of ‘archival science’ was born out of Enlightenment traditions, thus representing a distinctly Western mode of thought.³⁹ Seeking to deny cultural pluralism, the principles of archival science taught heritage institutions to standardise catalogue and collection data according to a Western value system, whilst systematically ignoring and obscuring non-Western forms of knowledge in the process.⁴⁰ Thus, Western cataloguing and classification systems have functioned to erase Indigenous concepts and names from the historical record. This is better known as ‘epistemic violence’, a term coined by Gayatri Spivak to refer to the way that ‘Western forms of knowing preclude or destroy local forms of knowledge’.⁴¹ If we turn our attention to controlled vocabularies, for instance, we are quickly reminded just who is in control. The most widely used subject indexing and classification systems, namely Library of Congress Subject Headings (LCSH) and the Dewey Decimal Classification (DDC), are rooted in colonial biases and have functioned to exclude and misrepresent marginalised peoples, including women, those with LGBTQIA+ identities, and disabled people.⁴²

In the words of Hope Olson, ‘Individual libraries, as well as the institutions that govern our standard, must be held accountable for poor and biased access to information’.⁴³ From a pragmatic standpoint, the crux of the non-inclusive language problem is that biased and inaccurate description reduces the discoverability of historical materials and the relevance of collection information. Materials relating to marginalised communities are often described using vague and homogenising language; they are often mislabelled; and they often fail to mention the presence of marginalised peoples altogether, thus rendering them ‘invisible’ or ‘silent’. Untrained researchers, moreover, are unlikely to use archaic, outdated language in their search terms. Essentially, providing access to historical records for the purpose of research and education is the British heritage sector’s raison d’être today.⁴⁴ By refusing to meet the needs of today’s society, the British public heritage sector risks becoming inaccessible and, ultimately, irrelevant.⁴⁵

As McGloin and Carlson neatly summarise, ‘language is a primary element in the continuity of colonial discourse, but that mindfulness about language use can and does effect discursive change’.⁴⁶ We must recognise that representation is eternally selective and that it will always be incomplete.⁴⁷ Description, nevertheless, plays a significant role in the ‘construction of meanings and the exercise of power’.⁴⁸ Memory institutions must therefore take responsibility for the language that they choose to use and the stories they choose to tell. Heritage professionals have equally a responsibility ‘to mitigate harm caused to record subject communities and users’.⁴⁹ Language choices can enact forms of non-physical violence – such as psychological, bureaucratic, and symbolic violence.⁵⁰ Words can also encourage and provide legitimation for hate crimes and physical acts of violence. There is power in description; hence, language must be understood as a tool through which heritage professionals can create and curate socially conscious catalogues, collections, displays, and learning resources.⁵¹ To pose a question that builds off the work of Duff and Harris: what if we re-imagined archival description standards as an instrument through which to challenge ‘the instinct merely to replicate existing power relations’?⁵²

The National Library of Scotland’s Equalities, Diversity, and Inclusion Internship, 2020–21

As part of their ongoing review and development of non-discriminatory cataloguing practices, the National Library of Scotland funded an Equalities, Diversity and Inclusion (EDI) internship from September 2020 to June 2021. The purpose of this internship was to develop a formal policy to govern the way that the National Library of Scotland manages its descriptive and interpretive practices over time in order to meet modern standards of description for protected characteristics and other areas of cultural sensitivity. The scope of the work included not only the review of new and existing catalogue descriptions, but also of the language used across the Library’s collections, finding aids, learning resources, and exhibitions. Although much of the work was internal in its focus, the nature of the project was directly shaped by conversations with other UK-based heritage organisations. As these conversations progressed, the need for cross-institutional collaboration on this area of work became increasingly apparent.

With its exclusive focus on the question of decolonising description, this internship appears to have been the first of its kind. Internships are themselves an established part of the industry that facilitate the valuable exchange of knowledge between heritage professionals and those at the beginning of their careers. It is worth noting for anybody who is considering emulating the National Library of Scotland’s approach, however, that using sponsored internship as a strategy to address reparative description had both pros and cons. Whilst it was an invaluable career opportunity for me as a recent graduate, the efficiency of the work was certainly hindered by my lack of cataloguing experience, the fact I had no prior education in Heritage Studies, as well as my newness within the organisation. The brevity of my employment also posed challenges in terms of the transfer of knowledge and the question of how to sustain the project following my departure. On the plus side, however, the internship route enabled the National Library of Scotland to make significant headway in an area of work that their permanent employees, given their existing workloads, would have otherwise struggled to explore in such depth. Additionally, I think that my personal background as a mixed-race woman of colour combined with my formal training in the discipline of History, wherein I specialised in understandings of race and empire, enabled me to bring a fresh and critical perspective to this area of work.

It would be difficult to say whether hiring an intern or external contractor for a short-term project is more or less effective than having an existing team of cataloguers and curators complete the work themselves, albeit at a slower pace, in collaboration with scholars and relevant communities. I do not write this article because I think that sponsored internship is the best or only way through which other public heritage institutions should seek to approach reparative description, therefore, but simply because I wish to share the fruits of my labour. As the National Library of Scotland’s former EDI intern, I held a unique job role within the UK heritage sector which enabled me to give the topic of inclusive description my full, undivided attention. Over the course of nine months, I conducted a substantial amount of research on the subject and was able to produce a range of inclusive description resources for the sector. Not only was I positioned to co-ordinate the descriptive efforts of different departments within the National Library of Scotland, but I acted as a linchpin for discussions about descriptive standards that were taking place across the sector, which culminated in the creation of the Cultural Heritage Terminology Network (CHTNUK). The problem of outdated and discriminatory language appearing within catalogues and collections is not unique to the National Library of Scotland; it is a problem that universally impacts public heritage institutions in the UK. Given the challenges of widespread under-staffing and under-funding in the sector, the resources shared in this section are designed to help heritage professionals across and beyond Scotland to make progress in this area of work despite time and budget constraints.

The Need for New Terminology Guidance

In order to determine the direction that the National Library of Scotland would take, it was essential to first review the existing guidance that had been published on the topic of inclusive description. I therefore began the EDI internship project by surveying the descriptive strategies of a range of cultural heritage institutions in North America, Europe, and Australia. In the USA, this included guidance that had been produced in collaboration with Indigenous and African American populations respectively, most notably, the Protocols for Native American Archival Materials (2007) published by First Archivist Circle and the Anti-Racist Description Resources (2020) produced by Archivists for Black Lives in Philadelphia (A4BLiP).⁵³ Studies conducted by Michelle Caswell, Associate Professor of Archival Studies at UCLA, have also provided a foundational basis for inclusive description efforts in the United States.⁵⁴ Similarly, the Presbyterian Historical Society (PHS) shared useful practical guidance in an online blog post about how their archivists ‘address the legacy of outdated and offensive language in [their] collection descriptions’.⁵⁵ Along with this guidance, PHS also released their ‘Terminology Crosswalk’, essentially a list of the outdated and preferred terms that they had used to update the language in their online catalogue beginning in 2020.⁵⁶ Another useful guide about harmful terminology relating to disability comes from the Harvard Center for the History of Medicine.⁵⁷ Moreover, in 2020 the Digital Public Library of America hosted and recorded a particularly insightful three-part conference titled ‘Introduction to Conscious Editing’ that is available to view online.⁵⁸ It must also be noted that amongst other large heritage organisations in the United States, a significant number of university libraries have published statements on their websites addressing the presence of harmful language within their online catalogues and collections.⁵⁹

In Australia, a notable amount of work has taken place to improve the relationship between the heritage sector and the Aboriginal and Torres Strait Islander people. Beginning in 1995, the first version of the Aboriginal and Torres Strait Islander Protocols for Libraries, Archives and Information Services was published by the Aboriginal and Torres Strait Islander Library and Information Resource Network (ATSILIRN).⁶⁰ Updated in 2012, the Protocols state that heritage institutions have ‘a responsibility to … respond appropriately to the existence of offensive materials’.⁶¹ There is no elaboration, however, on what an ‘appropriate’ response might look like. Both the State Records New South Wales (NSW) Protocols for Staff Working with Indigenous People and the Aboriginal and Torres Strait Islander Data Archive (ATSIDA) Protocols (2013) recommend the use of content warnings.⁶² We see content advice implemented by the National Library of Australia, for example, which firstly warns its users about ‘culturally sensitive’ language and materials that might appear in its online catalogue.⁶³ Secondly, on its Trove website users are required to opt whether they would like to see advisory warnings or not.⁶⁴ In terms of inclusive indexing, the Australian Institute of Aboriginal and Torres Strait Islander Studies (AIATSIS) thesaurus was produced in close collaboration with Aboriginal Australian and Torres Strait Islander communities to provide inclusive subject headings relating to their languages, people, and places.⁶⁵ In both the United States and Australia, the process of developing inclusive descriptive standards has focused on involving Indigenous and other marginalised communities in the archival process.

The most detailed guidance on harmful terminology in Europe was produced by the Netherlands’ National Museum of World Cultures, whose 2018 publication Words Matter offers critical and comprehensive reflection on the discriminatory language of the Dutch Empire.⁶⁶ In terms of inclusive indexing, IHLIA LGBTI Heritage Amsterdam also created the Homosaurus LGBTQ+ linked data vocabulary in 1997, which is now used internationally.⁶⁷

If we turn our attention to the United Kingdom, historian Norena Shopland’s book A Practical Guide to Searching LGBTQIA Historical Records offers additional guidance for both users and cataloguers working with LGBTQIA+ historic materials.⁶⁸ Another useful terminology guide is Historic England’s ‘Disability History Glossary’, which provides information on language surrounding disability and how it has developed throughout English history.⁶⁹ Generally speaking, inclusive description work in the UK has moved at a slower pace compared to the USA, Australia, and the Netherlands. That is not to say, however, that no efforts have been made by heritage professionals. On her online blog for Intersectional GLAM, a business which offers bespoke anti-discrimination training for heritage professionals, Jass Thethi shared her own set of guidelines on ‘respectful descriptions for marginalised groups’ in 2018.⁷⁰ Alicia Chilcott also published a practical set of recommendations on anti-racist description for UK public archives in 2019.⁷¹ Furthermore, that same year, Tim Jerrome collated a set of suggestions about how to deal with ‘offensive language in archival description’ from an email list survey for British-based archivists.⁷²

Whilst UK memory institutions need to deal with the specific language legacies of the British empire, many continue to rely on foreign terminology guidance and internationally controlled vocabularies. In particular, there is a dependency on Library of Congress Subject Headings (LCSH) which have been widely criticised for their biases and are, rather problematically, regulated by the United States Congress.⁷³ Subject headings produced in the USA do not reflect the specificities of language relating to Britain and its former colonies. Whilst it is possible to request changes within the LCSH system, it is surprising that the British heritage sector does not produce its own vocabularies at a national level.

From the conversations that I have had across the sector, it appears that Words Matter and A4BLiP’s Anti-Racist Description Resources are the two most popular inclusive language guides currently being used by heritage professionals engaged in inclusive description work within the United Kingdom. Whilst the information contained within these guides is certainly helpful, they are both greatly limited in their scope. In this case, the former deals with the language legacies of the Dutch Empire, with an emphasis on racially harmful terminology, whilst the latter specifically addresses the description of materials relating to African American communities in the United States. It is therefore necessary to supplement this information with other terminology guidance produced not only by heritage institutions, but also by NGOs, government institutions, and journalistic organisations around the world. The problem with using non-heritage guides, however, is that they usually only provide information about contemporary language use.

The Inclusive Terminology Glossary

The primary concern that UK heritage professionals have shared with me is that they do not feel confident in their knowledge of harmful language and the preferred contemporary terminology of groups with protected characteristics. This comes as no surprise, since heritage professionals in the United Kingdom are rarely provided with adequate training and education opportunities when it comes to dealing with issues related to equality and equity. To make the task of detecting and supplementing discriminatory language easier for myself and others, therefore, I began to collate all the existing guidance about preferred terminologies and the language that has been used to describe groups with protected characteristics in the past. After consulting a wide range of source materials, including where possible any texts that had been produced by the communities being represented, I contacted community members and academics with relevant expertise to review the information contained within the Glossary. Although labour-intensive, there was an indisputable rationale for creating this new Inclusive Terminology Glossary from scratch.⁷⁴ As already mentioned, there is little terminology guidance that has been written with a UK-focus. This has created a major gap in knowledge about the unique language legacies of the British Empire, which are of course global in scope. UK heritage professionals require an understanding of the language of race, diaspora, disability, and LGBTQIA+ issues in Britain and its former colonies, as well as specific local and regional terminologies that are used across the UK, including terminologies in all languages and dialects of the UK.

When conducting research into the preferred terminology of different marginalised groups, I encountered a lot of discrepancies across the available sources of information. Existing guidelines often present only a single view, therefore failing to acknowledge the presence of any inconsistencies or offer an explanation as to why they occur. For example, although identity-first vocabularies on disability are standard in Britain (i.e. ‘disabled person’) people-first vocabularies are preferred in the USA (i.e. ‘person with a disability’). Cultural heritage professionals require an understanding of these nuances so that they can determine when a term may or may not be deemed appropriate in a specific context. Although created with the UK heritage sector in mind, therefore, the Glossary remains international in its scope and seeks to explain, not erase, alternative perspectives. After all, British galleries, libraries, archives, and museums possess materials relating to the whole world. Moreover, there is no existing terminology guidance that considers all areas of protected characteristics. Projects to create inclusive vocabularies for materials relating to race, LGBTQIA+, and disability issues have all developed independently from one another. There is no reason, however, why heritage professionals could not address all these areas of discrimination simultaneously.

My Inclusive Terminology Glossary is specifically designed for use by cultural heritage professionals, who require understanding of both the modern and historical usage of terms as well as knowledge of how and why meanings may have shifted over time and space. Nevertheless, the information that the Glossary contains might prove useful to other sectors, such as journalism and publishing. Existing glossaries that have been produced by cultural heritage organisations tend to lack any sense of a timeline. Without an idea of the time periods in which certain terms were used, it can be incredibly difficult for both staff and users to locate materials relating to marginalised groups within a particular era. Therefore, my Glossary contains a column to denote the specific time period that a term was in use, if known. In this sense, the Glossary has a dual function as not only a guide for cataloguers, but as guide for researchers who are trying to locate archival documents related to people from marginalised communities.

Screenshots of Inclusive Terminology Glossary, Section 1.1. African American History and the Atlantic Slave Trade, accessed 8 July 2023. https://docs.google.com/document/d/1JaJ8VchUCbtg7jPmhwiZOQYsabBqKLxZ7n69urQS8VM/edit?usp=sharing.

Furthermore, few anglophone institutions have sought to provide guidance on harmful terminologies that may appear in non-English language material. It would be a mammoth task to incorporate discriminatory terminology that appears in all languages, but it is certainly worth including as many relevant examples as possible to avoid reinforcing an anglophone bias. In the Scottish context, this includes paying especial attention to the indigenous languages of Gaelic and Scots. Above all, I wanted to create a resource that is truly collaborative. Language debates are complex and contested, and it is crucial that the document reflects multiple perspectives. The Glossary would not be truly ‘inclusive’ if it did not invite the communities who are represented in its pages to assert their own self-identities. Created as a Google Document, anybody can suggest edits or add entries to the Glossary, meaning it is ‘live’ in its nature. This allows both heritage professionals and users to provide feedback and actively contribute to the work, meaning that the quality of information will continually be improved over time. It is the intention that the Glossary will, in the future, be developed into a responsive and sophisticated open-access database by the National Library of Scotland with the help of other parties.

Collections, Content Advice, and Catalogue Metadata

Within the scope of my internship project, my Glossary had many practical applications. To start with, I needed a list of terms that I could use to conduct key word searches across the National Library of Scotland’s online catalogues, collections, and websites. Using the Glossary, I was able to manually audit the presence of potentially discriminatory materials so that I could better understand the scale and nature of the issue. I found that not all ‘hits’ were from catalogue descriptions and websites, but also from published titles and the searchable transcriptions of surrogates of original historic materials. Key word searches have their limitations, however, particularly when it comes to identifying harmful material across visual collections like photographs and moving images. In terms of my methodology, I therefore also had one-on-one conversations with curators, who helped me to identify further problematic areas within their respective collecting areas. At present, I think it would be ineffective to conduct automated key word searches using artificial intelligence because the task requires manually filtering through the search results to determine whether a word or phrase is being used in a way that is discriminatory or not. For example, the word ‘coloured’ (or ‘colored’ in US English) is discriminatory when it is applied to people to describe them in racial terms, yet it is also an everyday term that appears in hundreds of item descriptions, usually to denote maps, images, or films that have colour. Human judgement is therefore required to gauge the nature of the context in which words are being used. Nevertheless, manual searches can be time-consuming, especially since a catalogue’s search algorithm can impact the discoverability of materials; some digital catalogues may return only exact matches, whereas others will return anything containing the search term in addition to related terms. As a reflection of the tedious nature of the work, the Glossary therefore includes both plural and singular forms, multiple spelling variations, archaic forms of spelling (such as replacing ‘s’ with ‘f’), as well as common misspellings that might return relevant materials.

Locating and identifying harmful materials would be the first step for the implementation of content advice and filters. Advisory notices are important for a number of reasons. Many British collections contain materials that have been used in the past to justify the oppression, segregation, torture, and sometimes genocide of groups with protected characteristics. It is vital to recognise this material as harmful and refuse to normalise the narratives that they contain, whilst safeguarding the emotional and mental welfare of users by warning them when they might encounter distressing materials. In terms of content advice, there are various different approaches that an institution might take, the least labour intensive of which is an institutional statement or catalogue-level warning that explains to users that they might encounter discriminatory materials across online catalogues and collections.

Examples of catalogue-level warnings can be seen on Bristol Archives’ British Empire & Commonwealth Collection Catalogue, the University of Virginia Archives catalogue, and at the bottom every item page on the Horniman Museum’s online collection catalogue.⁷⁵ Adding individual item level warnings is certainly laborious, but the National Archives UK has added a note to a handful of individual files to advise users that they contain ‘language that is by its nature offensive to Black and other minority ethnic groups’.⁷⁶ At an item-level, some institutions have also opted to put content warnings in the item title as an effective means of making sure that the user sees the warning if they access the page directly from an external search engine rather than through the catalogue. The Digital Public Library of America (DPLA) and the New York Public Library (NYPL) have taken this approach to warn users about the graphic content contained within lynching photographs, for example.⁷⁷ Institutions such as BBC Rewind and the Victoria and Albert Museum (V&A) have implemented more sophisticated content filters. On BBC Rewind, users must create an account and opt whether they wish to view certain types of sensitive material or not. On the V&A online catalogue search, potentially discriminatory images are blurred out until the user chooses to view them. Along with this, the V&A provide item-level content warnings; many of their descriptions contain a sufficient amount of historical context about their items; and they have a ‘Suggest Feedback’ button that users can use to report discriminatory material.

The other purpose of my Glossary is for editing and improving metadata, as well as using it to inform the creation of new learning resources, website features, and exhibition text. As I already mentioned, preferred terminology is sometimes debated by communities today, but heritage professionals nevertheless need to make informed decisions about what language they should use to describe materials relating to groups with protected characteristics. Often this will require breaking away from the confines of archival science, in which the classification of people and concepts is neatly standardised according to Western preferences, not allowing for nuance. Instead of adhering to such restrictive practices and viewing description as something that must be both concise and definitive, heritage professionals will often need to take the necessary space to add language disclaimers and provide users with relevant historical context for understanding the terminology choices that have been made. When the software does not allow for additional descriptive fields, institutions can work with the software developers (such as Ex-Libris) and petition for the creation of new features. Sometimes it will be appropriate to acknowledge multiple terminologies (whether they are competing or complimentary) and other times it will be necessary to admit that the institution does not have the knowledge to provide adequate labels. Supplementing records with the most appropriate terminology not only communicates to under-served groups that their self-identities are being acknowledged and respected, but it also enhances the quality of collections information and improves the discoverability of materials. Inclusive description does not dictate that historic language be altogether removed from records but that it is no longer normalised, which can be achieved through simple formatting choices such as the addition of quotation marks or inverted commas.

During my time at the National Library of Scotland, I used the Glossary to make a small sample of suggested metadata changes in the Archives & Manuscripts Catalogue which included, for example, replacing ‘chattel slaves’ and ‘negroes’ with ‘enslaved Africans’; capitalising the word ‘Indigenous’ in the North American context; changing ‘lunacy’ to ‘mental health’; and supplementing a quoted reference to ‘Hottentot girls’ to include ‘Khoikhoi’ in square brackets. In every instance, these metadata changes demanded careful consideration of the context of their use. Another recurring issue within the Archives & Manuscripts Catalogue that myself and my manager addressed was the incorrect or unclear use of terms like ‘African’ and ‘Indian’ in our agent epithets. Generally, these terms had been used in the past to describe the activities of white British men, built on the assumption that African men and women could not hold titles such as ‘explorer’, ‘historian’, or ‘missionary’. For instance, we changed ‘African explorer’ to ‘explorer in Africa’; ‘African traveller’ to ‘traveller in Africa’; ‘African notebook’ to ‘notebook … on the subject of Africa’; and ‘Indian historian’ to ‘administrator in India and historian’. Another term that required greater clarity was the label ‘planter’, which has loosely been used in the past both to describe Scottish tea-planters in India and, euphemistically, to refer to Scottish enslavers in the Caribbean. Additionally, we reviewed the use of the epithet ‘Orientalist’, an outdated term that was historically used to refer to scholars who took interest in any aspect of Asian or ‘Middle Eastern’ history, languages, geography, and cultures. To avoid ambiguity, the term was replaced with more specific labels – such as ‘Sinologist’, ‘Semitic Studies Professor’, and ‘civil servant in India’ – which more accurately describe the person’s career, travels, scholarly interest, or linguistic expertise, thus improving the specificity of these records so that users interested in any part of the vast geographical regions of Asia and the ‘Middle East’ can locate them more easily. In most instances, these examples of outdated language that appeared in our catalogue’s epithets could be traced back to the Oxford Dictionary of National Biography.⁷⁸

Policy Work and Principles of Inclusive Description

Alongside the review of harmful terminology across the National Library of Scotland’s catalogues, collections, and websites, I also worked on drafting a formal descriptive and interpretative practices policy that will govern the way the National Library of Scotland manages its linguistic choices over time. For the sake of accountability, the published policy will clearly outline, for both its staff and users, the professional ethical standards that the National Library of Scotland aims to meet in its descriptive and interpretative work. Applicable to all areas of the library, the document will guide the long-term development of the various strands of inclusive description work – from editing catalogue metadata to implementing content advice and developing the Inclusive Terminology Glossary in the future.

Policy development represents a top-down approach that can help an institution to internally implement and co-ordinate its inclusive description efforts, especially across different areas of work such as cataloguing, exhibitions, and social media. When published, a policy can also provide users with vital insight into an institution’s methodologies, ethical standards, and intentions. With that in mind, however, it is worth noting that corporate documents are generally inaccessible to users in terms of their writing style and the use of heritage-specific jargon. To improve transparency and accessibility, policy publication would be best accompanied by a user-friendly public statement or media release that is written in clear and concise prose. The other danger of policy work is that it can appear as a superficial kind of ‘empty gesture’, thus exacerbating the culture of distrust between heritage institutions and marginalised communities. When an institution releases a statement about their commitment to decolonisation, it is vital that they follow through by making meaningful and observable changes to their behaviours and practices, and marginalised communities should be involved in the policy-creation process itself.

Whether policy reform should be an institution’s ‘first step’ when it comes to inclusive description work remains undecided, but it is certainly useful for staff to have access to clear set of shared aims that will guide the reform of descriptive standards across the sector. Numerous codes of ethics for cataloguers as well as more generic ethical frameworks for librarians and archivists have already been developed by a range of organisations, some of which are international in their scope.⁷⁹ These ethical guidelines provide the foundation for the principles of inclusive description; however, the practice of inclusive description must extend beyond just catalogues. Building off the policy work that I completed with the National Library of Scotland, therefore, I have developed a set of ten foundational principles for inclusive description work that I wish to share. These generic principles are applicable to all heritage institutions in the UK and can be adapted to meet individual needs. The principles are as follows:

1. Acquisition or providing access to an item does not imply endorsement of any statements or opinions contained therein.

2. It is the duty of cultural heritage institutions to accurately maintain the historical record. It is not appropriate for cultural heritage professionals to alter or censor the original historical record, including language contained within collections, published titles, and official names.

3. Cultural heritage institutions have a responsibility to meet modern descriptive standards and describe and interpret materials in a manner that is accurate, respectful, and responsive to the communities who create, use, and are represented in their collections. When appropriate, this work will involve the remunerated involvement of the communities in question.

4. It is the duty of cultural heritage institutions to improve the quality of educational information for the purpose of enhancing and maintaining the authenticity, integrity, and reliability of records and resources and to promote the ethical use of the information contained within their collections and resources.

5. Where the original language in collections is harmful or discriminatory, cultural heritage institutions should strive to provide additional historical context and appropriate advisory content.

6. Description and interpretation should be harnessed as a tool to improve the discoverability of resources relating to marginalised communities.

7. Descriptive and interpretive practice should reflect professional values and ethics and contribute to a culture of accountability, trust, and transparency.

8. Cultural heritage institutions have a responsibility to support their staff with the task of inclusive description work. This includes emotional support as well as professional development and training to ensure staff are equipped to complete inclusive descriptive work judiciously and thoughtfully.

9. Cultural heritage institutions should share their expertise and use their voices to reform national and international frameworks of agreed standards and vocabularies.

10. Descriptive and interpretive standards will need to continue to be updated to reflect changes in knowledge, practice, and values in the future.

Within a broader effort to decolonise the UK heritage sector, these ten principles provide the essential framework in which heritage professionals should use the Inclusive Terminology Glossary. Policy aims will be meaningless if institutions do not commit time, money, and resources to achieving these principles, however.

Networking and Next Steps

To take a meaningful step towards decolonisation, moreover, UK heritage institutions will need to put their individualistic ambitions aside and start working together to address the language legacies that are specific to Britain and its former empire. If the goal is to achieve reparative justice, then it is unethical to capitalise on or gatekeep knowledge and resources that will help to mitigate the discrimination experienced by marginalised groups today. Not all institutions have the funding, resources, or expertise to launch their own inclusive description initiatives from scratch, and hence the sharing of resources and good praxis is essential to facilitating positive change across the public heritage sector. The creation of a more diverse and inclusive heritage sector requires the collation of knowledge into open-access guides like the Inclusive Terminology Glossary, and readers are urged to contribute to this project, which is designed to help heritage professionals create accurate and inclusive metadata, exhibition text, advisory notices, and other resources. Anybody can make suggestions and provide feedback to help improve the quality of the information it contains, and whilst I am hopeful that it will be developed into a funded database project in the future, its usefulness and future relevance currently depends on crowdsourcing and wider input.

Individuals who are interested in reparative description efforts are also encouraged to engage with the Cultural Heritage Terminology Network (CHTNUK), an online platform designed to facilitate cross-institutional collaboration in this area.⁸⁰ The multi-dimensional website and social media pages are intended to spark conversations between different heritage organisations, whilst providing a space where those who have the task of making difficult and sometimes burdensome language decisions can support and advise one another. The task of inclusive description can seem daunting, especially when one seeks to address all aspects of social inequality, but it becomes more manageable and takes the pressure off individual curators if we put our heads together. Through improving cross-institutional communication, it is hoped that heritage institutions will avoid repeating research initiatives that have already been done elsewhere and will instead focus their energy and resources on implementing actions and practical changes that support and build on the foundational work that heritage professionals, especially those from marginalised backgrounds, have already begun. The website is a place to access updated inclusive description resources and it hosts a blog and discussion platform where heritage professionals can come together with academic researchers and community members to learn from, listen to, and collaborate with one another on a range of inclusive description issues. For descriptive practices to be truly “inclusive”, after all, our efforts must reflect and respond to a range of different perspectives.