Abstract

The work aims to delve into the multifaceted and broadening realm of Intangible Cultural Heritage (ICH), with a focus on the preservation of linguistic diversity in the age of digitalisation. Article 2(2) of the UNESCO Convention (2003) defines “oral traditions and expressions, including languages, as a vehicle of the intangible cultural heritage”. Nonetheless, drawing on interdisciplinary discourses surrounding ICH and its relationship to language preservation, practice, and revitalisation, the work suggests that language be considered a form and practice of heritage in and of itself, not merely a vehicle of ICH. Indeed, considering the crucial role of languages as one of the main domains in which cultural heritage is manifested, it is paramount to provide a comprehensive understanding of the theoretical foundations underpinning sociolinguistics studies on the hegemony of English in the digital world. The article presents a state-of-the art overview of the concept of language as heritage following an interdisciplinary approach ranging from individual aspects of heritage language knowledge to broader societal, educational, and policy concerns in local and international contexts. In particular, the contribution of digital technologies will be analysed as a means for prioritising and at the same maintaining linguistic diversity. Inclusive and sustainable strategies for the documentation, revitalisation, and preservation of heritage languages will be explored with the aim of guaranteeing wider accessibility to culture via digital platforms and to understand better the implications of linguistic accessibility for ICH in the digital era.

Keywords: intangible cultural heritage, linguistic diversity, digitalisation, linguistic sustainability, international legal policy

1. Introduction

The digital revolution has drastically transformed the global landscape, especially in terms of communication, cultural dissemination, and knowledge production. The rapid spread of digital platforms, together with the globalization process, has led to the dominance of a few major languages - most notably English - in the digital sphere. This phenomenon, together with socio-economic factors and perceived linguistic prestige, has had a profound impact on linguistic diversity, is one of the major causes of the marginalisation of all the other languages, and poses a threat to the maintenance and preservation of minority languages.

It is important to clarify that, in this context, the broader, more inclusive definition of minority language is adopted. That is, one that does not merely consider the relationship between the number of speakers in a given geographic area since this label would not take into account other crucial factors, including economic, social and political prestige as well as legal recognition. Indeed, one should bear in mind that a minority language is not necessarily, or not only, a language spoken by a limited number of speakers. To better understand how controversial the concept is, we can consider the example of Spanish. It is a majority language in a number of countries but a minority language in the United States overall. At the same time, in US counties or regions with large Latino populations, it is much more prevalent and is indeed spoken by a majority of the population in some counties in Texas and New Mexico. The European Charter for Regional or Minority Languages (ETS No. 148), adopted in 1992 by the Council of Europe, allows us to resolve these controversial socio-linguistic situations. Indeed, it defines minority languages based on two criteria: a numerically smaller speaker population and a lack of official status. Hence, languages (such as Irish) which enjoy an official status but are spoken by smaller percentages of a given population do not count as minority languages. However, the charter excludes dialects and migrant languages, even though the classification of a language variety as a dialect is as much a sociopolitical judgment as a linguistic one (Edwards 2010).

Although an estimated 7,000 languages are spoken worldwide, the predictions of a study by Bromham et al. (2022) indicate that, by the end of this century, half of them may no longer exist. More specifically, to shed light on the key factors responsible for this threat to language diversity, the work examines 6511 spoken languages with 51 predictor variables including aspects of population, documentation, legal recognition, education policy, socioeconomic indicators and environmental features. Interestingly, the results reveal that, among all the other factors mentioned, two in particular are responsible for language endangerment: that is greater road networks and schooling. Indeed, the first encourages population movement, while the second is consistent with a growing number of studies showing a negative impact of formal schooling on minority language vitality, particularly where bilingual education is not supported or, in some cases, is actively discouraged (Hornerberg 2002; Skutnabb-Kangas 2013).

Another recent work (Kandler and Unger 2023) reveals how alarming future projections are regarding language loss with the consequent erasure of such rich cultural heritage and unique world perspectives. The study indicates that around 3000 languages could disappear by the end of this century, at a rate of one every two weeks. Notably, previous research on the threats to linguistic diversity (Aalberse et al. 2019) showed that the disappearing rate was one every forty days, that is, nine languages per year. The accelerated rate of language extinction together with its consequences is also confirmed by other studies. For example, to demonstrate this trend, Skirgard et al. (2023) employed Grambank, the most important database on language structures available, to inquire about the threat posed to linguistic diversity. Grambank is designed to be used to investigate the global distribution of features, language universals, functional dependencies, language prehistory and interactions between language, cognition, culture and environment. The research used a metric called “functional richness”, one of the main components of cultural diversity, to specifically estimate the potential effects of language loss on both a global and regional level. Finally, Bromham et al. (2022) emphasised the need for investment in language documentation, bilingual education programmes, and other sustainable initiatives to avoid the potential loss of 1500 languages by the end of the century.

Thus, the disappearance of minority languages would represent a significant reduction in the world’s intangible cultural heritage, as language is a primary medium for transmitting cultural knowledge, traditions, and identity from generation to generation.  Indeed, according to The Language Conservancy, the foremost organisation working with endangered languages in North America, after global warming, language loss is emerging as the Earth’s most acute crisis (Collette and Kennedy 2023). This linguistic scenario has profound socio-cultural implications as it intersects with issues of identity and belonging. In fact, language is a critical marker of cultural identity, and for many immigrant and indigenous communities, maintaining their language is essential for preserving their heritage and fostering a sense of community. The use of a particular language not only serves as a means of functional communication but also expresses the speaker’s cultural identity, as well as celebrating the cultural heritage developed by all previous users of that language. From this perspective, one can say that measures that allow for the public use of a particular language, or that impose the use of that language in certain contexts, also contribute to the preservation of the cultural heritage of a human community (and of the world).

The Intangible Cultural Heritage (ICH) of human communities thus comprises an oral heritage expressed through language. The importance of orality was clearly expressed by the UNESCO programme that preceded the adoption of the Intangible Heritage Convention, namely the Programme for the Proclamation of Masterpieces of the Oral and Intangible Heritage of Humanity (2006).

The intangible cultural heritage means the practices, representations, expressions, knowledge, skills – as well as the instruments, objects, artefacts and cultural spaces associated therewith – that communities, groups and, in some cases, individuals recognize as part of their cultural heritage. (UNESCO 2003: art. 2.1).

Hence, the UNESCO Convention for the Safeguarding of Intangible Cultural Heritage (2003) marks a significant achievement acknowledging the importance of language, particularly oral traditions, in the preservation of cultural heritage. However, in the digital era, new challenges have emerged regarding language maintenance and loss, which require a thorough re-evaluation of how linguistic diversity can be safeguarded and promoted. As will be discussed in the following sections, while digitalization offers new instruments and opportunities for documenting and revitalizing endangered languages, it also facilitates the further marginalization of minority languages.

This paper seeks to explore the complex interplay between language as a fundamental component of intangible cultural heritage and the ways in which digital technologies are affecting its preservation, in terms of both challenges and opportunities. The article suggests that heritage studies would benefit from greater attention to language as a cultural practice, unique to human beings, and as both embodiment and manifestation of individual and collective belonging. To that end, engaging with the existing literature on heritage and language, this study aims to contribute to the ongoing academic debates on language as a form and a vehicle of intangible cultural heritage and provide some critical considerations on the nature of the relationship between heritage and language.

First, the work addresses how language is perceived and theorised within the socio-political landscape, as well as in the academic scenario, within different types of heritage discourses. Starting from analysis of international legal frameworks, it addresses the relationship between national law and international conventions to attempt to shed light on their impact in terms of safeguarding linguistic diversity. Second, the work discusses the most significant factors that put linguistic diversity at risk, with particular attention to the digitalisation process and AI-based linguistic technologies. Finally, digital tools, initiatives and case studies of language revitalisation efforts will be reviewed to provide insights into the future of linguistic diversity in the digital age.

2. Linguistic Heritage as Intangible Cultural Heritage (ICH)

Although communicative functionality represents one of the most important goals of languages, all the other fundamental aspects represented and carried by each language cannot be overlooked. Indeed, linguistic heritage conveys a sense of cultural identity and belonging, it is a repository of knowledge, traditions, culture and, therefore, it is an essential component of human diversity (Sarma 2015). Indeed, language represents one of the most important aspects of intangible heritage since it stands for a unique worldview and is the vehicle of collective memories and values (Eaton and Turin 2022).

However, as the reviewed studies suggest, half of the 7,000 documented languages currently spoken across the world could be endangered. As such, the preservation of linguistic heritage is integral to the broader effort to safeguard ICH and every action aimed at preserving ICH can be considered as sustainable. Within this framework, the role of language in preserving cultural identity cannot be underestimated Through language, communities express their values, beliefs, and worldviews. Language is also a living, evolving medium that reflects the history and experiences of its speakers. In many cases, the loss of a language leads to the loss of unique cultural practices and knowledge systems that are embedded within the linguistic structure. However, an important concern that emerges from this definition is the role of language as a mere “vehicle” of ICH rather than as an essential part of it.

To better understand the complex interplay between language, thought and culture, it is worth recalling the academic debate, which draws upon various disciplinary approaches, from linguistics to anthropology, and cultural studies, opposing two contrasting views: that is language conceived as ICH per se and language as a vehicle of ICH. Central to this discourse are the ways in which language both embodies cultural practices and serves as a medium through which these practices are transmitted through generations. While the Sapir-Whorf Hypothesis (Sapir 1921; Whorf 1956) stresses the importance of language in shaping worldview and cultural practices, Canagarajah (2006, 2013) offers a more nuanced view that acknowledges the evolving, adaptive nature of language in a globalized context. Both perspectives contribute to our understanding of how language not only reflects but actively participates in the transmission of intangible cultural heritage.

Different scholars maintain that language is inseparable from all those intangible aspects of culture including beliefs, customs, and worldviews. Hence, language embodies social practices, rituals, and historical narratives that define a community's sense of belonging. This means that when a language is endangered, the loss is not only linguistic but also cultural, as the language is the means for expressing and preserving community heritage.  Nonetheless, as already discussed, language also serves as a vehicle for transmitting intangible cultural heritage. In this view, language is understood primarily as a tool through which cultural knowledge is preserved, performed, and transmitted. It is the medium through which cultural expressions such as music, dance, oral traditions, rituals, and folklore are shared. From this perspective, scholars point out that to understand cultural practices fully, one must also understand the linguistic expressions that encode them. This conception aligns with the notion that language does not simply reflect culture but actively shapes and constructs cultural meanings. In communities with diverse linguistic traditions, a single language can serve as the conduit for a complex network of cultural practices, allowing them to be embodied in the words, expressions, and narratives of its speakers.

The influential Sapir-Whorf Hypothesis (Sapir 1921; Whorf 1956), also known as “linguistic relativity”, plays a significant role in this discussion. It states that people experience the world based on the structure of their language, and that linguistic categories shape and limit cognitive processes and how speakers of a given language perceive time, space, and social relationships. Since differences in language affect thought, perception, and behaviour, according to this view, speakers of different languages think and act differently. In the context of ICH, this suggests that language is not merely a passive reflection of culture but an active force in shaping cultural norms and practices. Accordingly, the loss of a language could mean the loss of unique ways of understanding the world, as cultural knowledge embedded in linguistic structures would also be eroded. The relationship between language and thought thus implies that ICH, expressed and understood through language, could be deeply impacted by language extinction.

Indeed, as the hypothesis contends, if linguistic structures influence not only how individuals perceive the world but also how they conceptualize and interact with their cultural environment, then language must be seen as a fundamental component of heritage itself. The hypothesis implies that cultural knowledge is not merely conveyed through language but is in fact shaped by it: the grammatical categories, vocabulary, and metaphors embedded in a language encode culturally specific ways of thinking and being. For example, the presence or absence of certain tense structures, classificatory systems, or spatial orientations in a language can reflect and perpetuate particular social norms, cosmologies, and epistemologies. In this light, language does not simply act as a neutral carrier of cultural content but constitutes the very framework through which that content is interpreted and experienced. Thus, preserving a language also means preserving the distinctive cognitive and cultural patterns it fosters, which may not be fully translatable or replicable in other linguistic systems. This perspective reinforces the idea that the safeguarding of linguistic diversity is essential not only for communication but for the continued vitality of the world’s intangible cultural heritage.

On the other hand, Suresh Canagarajah’s works (2006, 2013) provide an important contemporary perspective on the role of language in ICH, particularly in the context of globalization. Canagarajah challenges the idea of linguistic purity and emphasizes the dynamic, hybrid nature of language practices in a globalized world. In particular, in his influential work Translingual Practice: Global Englishes and Cosmopolitan Relations (2013), the author maintains that if modern globalisation was associated with the values of territoriality and homogeneity, post-modern globalisation celebrates mobility and diversity. Hence, these changes have led to social conditions which significantly differ from those promoted by the Herderian Triad. The latter posited an intrinsic and almost organic link between a people (Volk), their language, and their land. In Herder’s own view, language is not merely a communicative tool but represents the embodiment of a community’s soul, shaping and shaped by the experience, culture, and worldview of that people. Remarkably, Bauman and Briggs (2003), commenting on Herder’s philosophy, describe the concept in the following terms: “The desired goal of unification rests upon discursive unity, provided by the authority of tradition and a unified adherence to the national spirit. And here too, linguistic homogeneity is a necessary condition: “One people, one fatherland, one language” (p. 193). Accordingly, each language was considered unique and inextricably tied to a specific nation and geographic territory. This formulation gave rise to the notion that not only are linguistic purity and homogeneity natural, but they are also culturally and morally desirable, contributing to the preservation of the integrity of national identity.

Nonetheless, Canagarajah challenged the Herderian paradigm, arguing that it is not suitable to account for the fluid, hybrid, and translingual practices that characterise contemporary communication in an increasingly interconnected and mobile world. More specifically, he maintains that the conception of language as “the innermost spirit, thought, and value of the community” contributes to strengthen the sense of nationalism, linguistic homogeneity and purity that justified the imposition of the language of the dominant community on all the other minority languages, which lost their claim to that specific place and got suppressed. In other words, this perspective historically justified the marginalization and suppression of minority languages in favour of a monolingual national standard. Hence, heritage languages were often deemed illegitimate or inferior, losing not only their communicative functionality but also, most importantly, their cultural acknowledgment.

Another major point of the Herderian Triad critique advanced in Canagarajah’swork, concerns the legitimacy attributed to the so-called native speaker, which significantly contributes to the current discussion on the hegemony of English. Indeed, if a person is defined as a native of a single language, there is only one language belonging to one’s speech community and a place. That is to say, others who use our language are interlopers into ours. According to Canagarajah, the risks of such a notion of a pure, place-bound language speaker gives rise to the myth of the monolingual native speaker as the only legitimate user of a language. By privileging those who are defined as ‘native’, who are often from historically dominant geopolitical regions, the Herderian triad overlooks and does not give voice to the multilingual and translingual practices of individuals in postcolonial and diasporic contexts.

In contrast, Canagarajah highlights the fluidity of language in the transmission of ICH. Indeed, according to this perspective, language is not static but constantly evolving, absorbing influences from global forces while maintaining cultural relevance. The multilateral flow of people, things, and ideas across borders has made more visible mixed forms of community and language in highly diversified geographical spaces. Accordingly, language practices in these multilingual and multicultural communities can be seen as forms of resistance and adaptation, where ICH is renegotiated through everyday language use. Thus, differently from earlier perspectives that emphasised linguistic "purity," Canagarajah's point of view suggests that language as a vehicle for ICH may involve a more complex, hybridized process where cultural practices are constantly reshaped through intercultural exchange.

An interesting perspective that emerged from a UNESCO study (2003) highlights the normative reasons why languages should be considered part of the world’s cultural heritage:

The extinction of each language results in the irrecoverable loss of unique cultural, historical and ecological knowledge. Each language is a unique expression of the human experience of the world. Every time a language dies, we have less evidence for understanding patterns in the structure and function of human language, human prehistory, and the maintenance of the world’s diverse ecosystems. Above all, speakers of these languages may experience the loss of their language as a loss of their original ethnic and cultural identity (UNESCO 2003: 2).

Thus, the restrictive definition of intangible heritage, in the Convention’s text, does not deny that language is an element of cultural heritage. Instead, it is the expression of what has been defined the ‘long-standing resistance of States to any external interference in national cultural policies and the codification of protection for intangible heritage, especially languages, at the international level’ (Vrdoljak 2014; de Witte 2020: 373). More specifically, laying down legal obligations for states regarding their linguistic heritage might spill over into those States’ policies on the official use of languages, which is traditionally a sensitive matter linked to State-building and national identity.

Nonetheless, despite the de facto acknowledgement of language as a critical component of ICH, challenges remain in terms of translating this recognition into de iure effective policies and actions. While the UNESCO Conventions and recommendations provide a framework for protecting linguistic heritage, the implementation of these measures often depends on national and regional governments, which may prioritize other aspects of cultural heritage over linguistic diversity. Moreover, the global dominance of a few major languages, particularly in digital spaces, has created additional pressures on smaller languages, further complicating efforts to preserve linguistic diversity.

Indeed, the primary reason why several threatened or almost moribund languages have been successfully revived in recent decades (such as, for example, Hawaiian, Māori, and Welsh, to name just a few) is because legislation was adopted that grants legal rights to use those languages and imposes corresponding duties on public authorities to use them and ensure their transmission through education and the media. Conversely, the lack of official recognition of many other smaller languages is the main reason for their threatened extinction. The following sections will focus on some examples of successful language policies and strategies adopted to preserve minority languages in order to shed light on the complex interplay between language law, intangible cultural heritage, and linguistic sustainability.

3. Legal and Policy Frameworks for Protecting Linguistic Heritage

The protection of linguistic heritage is shaped by a complex relationship between national laws and international conventions. At the national level, language laws often focus on the regulation of language use in public life, including education, government, and media. In multilingual societies, language policies have the power to either promote linguistic diversity by recognizing minority languages or reinforce the dominance of a single national language. At the international level, legal frameworks such as the European Charter for Regional or Minority Languages and the UNESCO Convention on the Protection and Promotion of the Diversity of Cultural Expressions play a crucial role in safeguarding linguistic diversity.

At the global level, it is worth recalling that the UNESCO Convention on the Protection and Promotion of the Diversity of Cultural Expressions (2005) further reinforces the importance of linguistic diversity. Although the Convention does not explicitly focus on language, it emphasizes the need to promote cultural diversity in all its forms, including linguistic diversity. The Convention encourages member states to adopt policies that support the creation, production, and distribution of cultural goods and services in a wide range of languages. This is particularly important in the context of globalisation, where the cultural products of smaller language communities are often overshadowed by those produced in dominant languages.

Notably, the UNESCO Conventions do not formulate strict legal obligations for the member states, although their provisions serve as a basis for enforceable rights for the members of a linguistic group. Other international legal instruments that aim at the protection of minority languages and the rights of the speakers of these languages are formulated in somewhat stronger terms. At the global level, article 27 of the 18 International Covenant on Civil and Political Rights (UN 1967) is particularly relevant for this discussion. The wording of this provision seems to imply that states merely have a negative duty not to interfere with the private choice of people to speak a minority language: 'In those States in which ethnic, religious or linguistic minorities exist, persons belonging to such minorities shall not be denied the right […] to use their own language’.

However, the Human Rights Committee insisted, in one of its general comments, that article 27 is also a source of positive duties: ‘Although the rights protected under article 27 are individual rights, they depend in turn on the ability of the minority group to maintain its culture, language or religion’ and, therefore, ‘positive measures by States may be necessary to protect the identity of a minority and the rights of its members to enjoy and develop their culture and language in community with the other 19 members of the group’ (UN 1967). Despite this ambitious interpretation, article 27 plays only a limited role in the individual applications submitted to the Committee and it has not served, so far, as an effective source for the protection of language heritage.

On the other hand, the most explicit international instrument for the protection of linguistic heritage is the aforementioned Council of Europe’s European Charter of Regional or Minority Languages (CoE 1992), which has been ratified, at the latest count, by twenty-five European countries. Its preamble contains references both to the protection of languages as heritage and to the protection of linguistic rights. The substantive provisions of the Charter reflect to some extent the hybrid nature of its underlying aims. Whereas many of the provisions seek to promote the guarantee of language rights in education, public administration, and the media (the traditional areas of minority language rights legislation), some other provisions are more directly concerned with the protection of language as cultural heritage.

The latter concern is particularly visible in article 12, entitled ‘Cultural activities and facilities’, which contains some rather specific duties for the member states. The provision requires them ‘to encourage and/or facilitate the creation of a body or bodies responsible for collecting, keeping a copy of and presenting or publishing works produced in the regional or minority languages’, and ‘to create and/or promote and finance translation and terminological research services, particularly with a view to maintaining and developing appropriate administrative, commercial, economic, social, technical or legal terminology in each regional or minority language’ (CoE 1992: art. 12g). Here, the concern is clearly not with granting individual rights to use or learn the minority language but, rather, with the adoption of schemes of governance that will help the sustainable development of those languages.

Ever since the adoption of the Convention for the Safeguarding of the Intangible Cultural Heritage (CSICH) in 2003, indigenous and minority cultural rights have enjoyed increasing recognition. At the same time, they have been exposed to public discourses and homogenising language that might detrimentally affect their right to access and participation in the creation of ICH and ultimately cultural life (see Eichler 2021).

Different national examples of effective language revitalisation efforts provide valuable insights into the role that legal frameworks can play in preserving linguistic heritage. In Wales, for instance, the Welsh language has undergone a remarkable revival over the past few decades, thanks in large part to legislative measures that grant Welsh official status alongside English. The Welsh Language Act of 1993 and the Government of Wales Act of 1998 both recognize the right of Welsh speakers to use their language in public life, including in education, government, and media. These laws have been instrumental in increasing the visibility and use of Welsh, both in everyday life and in digital spaces. From an educational point of view, it is worth mentioning that translanguaging, the influential “practical theory of languages” (Li Wei 2018), was first experimented in Wales with the aim of reinforcing Welsh, the minority language, through constant use and alternation with English. First used in the 1980s by Cen Williams (1994, 2000), the term ‘translanguaging’ started to draw attention to the systematic alternation of input and output languages in bilingual education. Hence, translanguaging can be used as an umbrella term including a wide variety of examples of both theories and practices of fluid and more inclusive use of languages, breaking the traditional conventions and the strict purist ideologies to get closer to the way people communicate in their everyday life.

A similar successful example of language revitalisation can be found in New Zealand, where the Māori language has been revitalised through a combination of legal recognition, education initiatives, and media support. The Māori Language Act of 1987 granted Māori official status in New Zealand, ensuring that it could be used in courts, government, and other public institutions. This legal recognition was accompanied by efforts to promote the use of Māori in schools and the media, including the establishment of Māori-language television and radio stations. As a result of these initiatives, the number of Māori speakers has increased, and the language has gained greater visibility and prestige within New Zealand society.

As regards national policies aiming at the preservation and maintenance of minority languages, Canada represents an emblematic and complex case. The country is widely recognized for its linguistic and cultural diversity, officially reflected through bilingualism in English and French, as mandated by the Official Languages Act (1969) and reinforced by the Canadian Charter of Rights and Freedoms (1982). In addition, the Multiculturalism Act (1988) acknowledges the preservation of non-official languages, highlighting the country's commitment to its multicultural and multilingual heritage. However, the linguistic reality extends beyond official bilingualism. According to the 2021 census (Statistics Canada), while English (54.6%) and French (22.8%) are the predominant mother tongues, 23% of Canadians speak a language other than English or French at home, illustrating a broader linguistic complexity. This includes both immigrant languages—such as Mandarin, Punjabi, and Spanish—and Indigenous languages, with over 250,000 Canadians reporting the use of languages like Cree, Inuktitut, and Ojibwe. The Indigenous Languages Act (2019) and various community initiatives aim to protect and revitalize Indigenous languages, emphasizing their cultural importance.

Nonetheless, not all countries have been as successful in protecting their linguistic heritage. In Italy, for example, regional laws provide financial support for cultural activities in local languages, but there is no national policy that grants official status to these languages. Indeed, despite being home to 31 endangered minority languages (UNESCO 2018), Italy is still partially neglected in the relevant academic literature. Many regional languages in Italy remain at risk of extinction as they lack the legal protections necessary to ensure their continued use in public life. An interesting study by Lonardi (2020) analyses the role of tourism in pursuing the maintenance of minority languages in two small Italian towns, Luserna and Giazza (both in the north of Italy), where a population with Bavarian origins still speak the dialect of their ancestors.

Through qualitative interviews, the study not only revealed important data on the value of minority languages for tourism but, most importantly, it raised awareness on the role of tourism for ICH and especially for minority languages. Another important conclusion drawn from this research is that “economic sustainability is intrinsically related to social/cultural sustainability because it guarantees language transmission and, therefore, language preservation” (Lonardi 2020: 60). This highlights the need for strong frameworks, including tourism, education, and the law, for safeguarding linguistic diversity and the need for comprehensive policies that go beyond cultural activities to address language use in all fields such as education, government, and media.

As discussed, at the international level, the protection of linguistic heritage is often complicated by the reluctance of states to adopt binding legal obligations. Many countries are wary of external interference in their national language policies, which are closely tied to issues of national identity and sovereignty. Hence, international legal instruments like the UNESCO Conventions and the European Charter often rely on voluntary commitments, rather than enforceable obligations, making it difficult to ensure that states take meaningful action to protect linguistic diversity. Despite these challenges, international legal frameworks provide an important foundation for the protection of linguistic heritage. By establishing norms and guidelines for the promotion of linguistic diversity, these frameworks help to raise awareness of the importance of language preservation and encourage states to act. However, more work is needed to ensure that these legal frameworks are effectively implemented, particularly in countries where minority languages are most at risk.

In addition, concerning the peculiar contexts of safeguarding languages in the digital context, broader policies are needed to promote multilingualism and linguistic rights in cyberspace. It is worth recalling UNESCO’s 2003 Recommendation Concerning the Promotion and Use of Multilingualism and Access to Cyberspace, which provides a framework for promoting linguistic diversity online. The Recommendation calls for the development of multilingual content, the promotion of language education online, and the creation of tools that enable speakers of minority languages to access digital resources in their native tongues. However, to implement these policies, governments and international organizations must work together to address the economic and technical barriers to multilingualism in cyberspace. This includes providing funding for the development of digital content in minority languages, supporting language education initiatives, and promoting the use of minority languages in public and private sectors. By taking a comprehensive approach to multilingualism, it is possible to ensure that all languages have a place in the digital world.

4. Impact of Digitalization on Linguistic Diversity

As regards the impact that digitalisation may have on this panorama of linguistic diversity and any attempts to maintain and consolidate minority languages, it can be argued that the digital age presents a paradox for language preservation. Indeed, while digital technologies offer new tools for documenting and revitalizing endangered languages, they also contribute to the marginalization of these languages. The challenge for language preservation in the digital age is to harness the potential of digital technologies in a way that promotes linguistic diversity while mitigating the risks of linguistic homogenization. The digital revolution, with the consequent hegemonic role of English, has been reshaping linguistic diversity and has had a profound impact on several levels. On the one hand, as will be further analysed in the following sections, digital platforms offer new opportunities for documenting, revitalising, and promoting endangered languages. On the other, the dominance of a few major languages in digital spaces – particularly English – represents a threat to minority languages, accelerating their loss.

The hegemony of English in global communication together with the implication it has in terms of linguistic sustainability is a pivotal issue, with longstanding roots, which requires careful examination. The dominant role of English as a global lingua franca has been addressed by several influential scholars in the past, who focused on specific issues. For instance, Crystal (2003) has highlighted the impact of English in terms of language loss as well as of marginalisation of minority languages. Phillipson (1992) argues that the linguistic imperialism of English contributes to further emphasise social and cultural inequalities, hindering the preservation of linguistic heritage. Finally, Skutnabb-Kangas (2013) analyses the issue from the perspective of menacing linguistic rights as human rights.

This complex scenario is further impacted by the fastest technological adoption in our history. Indeed, the digital revolution, especially in the United States, where all the major IT companies are located, has determined a significant shift towards English as the dominant lingua franca. Moreover, the relatively recent development of generative AI by companies such as Microsoft, Google, and Amazon has exacerbated the phenomenon by using, predominantly, English for the creation of Large Language Models (LLMs) (Lai et al. 2023). If one considers the potential that these new technologies have to control and permit communication worldwide, it can be argued that such an English-centric approach significantly contributes to the debated language loss. Indeed, it has been noted that these models, with the nearly exclusive use of English, overlook and devalue all the other languages, especially the endangered ones at risk of extinction (Salomone 2022).

The English dominance in LLMs can be explained in terms of both data availability and structural inequalities in online content, reflecting broader linguistic hierarchies on the internet and in digital communication. According to W3Techs (2023), over 50% of websites use English, despite English being the first language of only about 5% of the global population (W3Techs, 2023; Ethnologue, 2023). This data imbalance translates into training corpora that are skewed toward English, which in turn affects model performance across languages. As Bender et al. (2021) claim in their influential paper “On the Dangers of Stochastic Parrots”, language models "reflect the biases and distribution of their training data," meaning that linguistic overrepresentation leads to uneven capabilities and potential marginalization of low-resource languages.

The Internet, considered as the primary medium for communication, education, and entertainment nowadays, has become a critical battleground for the preservation of linguistic diversity. According to Internet World Stats, just 10 languages dominate 77% of the world’s digital content, with English accounting for more than half of that content. This digital imbalance mirrors broader global inequalities, as speakers of minority languages are often excluded from digital spaces due to a lack of resources and infrastructure (UNESCO 2023). An additional concern is represented by the gradual recorded trend towards orality over literacy. As Hutson et al. (2024) discuss, with the improvement of natural language communication with computers, a shift can be observed from literacy to orality as language models are facilitated by the adaptability and immediacy of spoken interactions. For the current discussion on linguistic sustainability and the maintenance of endangered languages, this could lead to both advantages and drawbacks. If on the one hand this evolution highlights and strengthens the interface of technology and languages, on the other hand, it contributes to make the hegemonic role of English more explicit in global communication and culture.

Hence, for minority languages to survive in the digital age, it is paramount to make them visible in the digital sphere. That is, they must be represented online, where an increasing amount of communication and cultural exchange takes place. The absence of minority languages from digital spaces means that younger generations, who are increasingly reliant on digital technologies for learning and social interaction, are less likely to use their ancestral languages. This has significant implications for the transmission of linguistic and cultural knowledge, as the younger generation’s disengagement from their heritage languages often leads to language loss.

Yet, even if one is aware of the digital paradox for the preservation of linguistic diversity, it is important to focus also on the notable opportunities that the digital world may offer for language revitalisation. For instance, language learning apps, digital dictionaries, and online archives have made it easier than ever to document and preserve endangered languages. These tools allow speakers of minority languages to connect with one another, share resources, and access language learning materials, even when they are geographically dispersed. Indeed, if on the one hand, it has been stressed (Vieira et al. 2020) that the inappropriate use of Machine Translation (MT) may exacerbate linguistic inequalities, on the other, there are also valuable benefits in democratising online communication.

Indeed, more and more under-resourced languages are being catered for free on online MT systems. Google Translate, for instance, currently covers 109 languages including some examples of underrepresented languages: Scots-Gaelic, Chichewa, and Tatar (Moorkens 2022). Another example is No Language Left Behind (NLLB) (NLLB Team, 2022), a MT system representing a significant advancement in promoting linguistic diversity and inclusivity in the digital sphere. By supporting a broader range of languages, particularly low-resource and minority languages, NLLB contributes to both the survival and online visibility of these languages. Its architecture and training approach are specifically designed to overcome the data scarcity challenges that hinder the development of accurate MT models for underrepresented languages. In doing so, NLLB not only facilitates cross-linguistic communication but also strengthens the digital presence of endangered languages, offering speakers new opportunities for cultural expression, access to information, and participation in the global digital ecosystem.

Among the initiatives worth mentioning that involve digital technologies that can support linguistic diversity there is the Rosetta Project[1], a global initiative aimed at creating a digital archive of the world’s languages. The project has collected linguistic data on over 2,500 languages, many of which are endangered, and has made this information available to researchers, educators, and language communities. By providing a comprehensive digital record of these languages, the Rosetta Project helps to ensure that they will not be lost to future generations, even if they are no longer spoken. Another promising initiative is the Ma! Iwaidja app,[2] which was developed to support the documentation and revitalization of the Iwaidja language, spoken by fewer than 200 people in Northern Australia. The app allows users to record words, phrases, and translations, and includes a dictionary and grammar guide to help learners. By providing a practical tool for language learning and documentation, the app helps to ensure that the Iwaidja language will be preserved for future generations.

Digital preservation resources such as the Rosetta Project and the Ma! Iwaidja app demonstrate the potential of technology to support endangered languages. By collecting and archiving linguistic data, these initiatives help to ensure that even languages with few remaining speakers will be available for future study and revitalisation efforts. In addition, digital platforms like YouTube and Instagram have become important spaces for linguistic activism, where speakers of minority languages can share content in their languages and raise awareness about the importance of language preservation.

However, in this context, it is necessary to bear in mind some ethical considerations in terms of digital preservation. First, the process of digitizing linguistic and cultural heritage must be done in a way that respects the rights and traditions of the communities involved (Wagner and Clippele 2023). Indeed, as digital technologies become central to language preservation, it is essential to ensure that these efforts are guided by principles    of inclusivity, community engagement, and respect for cultural sovereignty. In many cases, digital preservation efforts have been led by external organizations, with little input from the communities whose languages are being preserved. This can lead to a fracture between the goals of the preservation project and the needs and priorities of the community. To address this issue, it is essential to involve communities in the decision-making process and to ensure that they have control over how their linguistic heritage is represented in digital spaces.

Second, a further significant ethical issue is the risk of cultural appropriation and exploitation. In some cases, digital platforms may commodify cultural knowledge, turning it into a product that can be consumed by a global audience without the consent of the communities from which it originates. This can lead to misrepresentation of cultural practices and the erosion of community control over cultural heritage.

To better understand these concepts, it is worth analysing Eichler’s (2021) opposition between homogenisation and differentiation related to the safeguarding of ICH and the consequent risks for cultural minority groups. Both are considered as important threats for cultural heritage as well as for all the related practices to maintain it. Specifically, the risks of homogenisation are due to the pressure on cultural bearers to adapt ICH to popular demands, affecting the essence of some cultural practices. Differentiation, instead, tends to portray a very limited image of what particularity means, with a negative impact on communities in deciding which aspects of ICH they want to make more visible based on what they consider more meaningful and significant. In other words, the main threat of differentiation is limiting the communities’ cultural self-determination.

To address these concerns, digital preservation efforts must be guided by principles of cultural sovereignty, ensuring that communities retain ownership over their linguistic and cultural heritage. Indeed, it is crucial to consider that while, on the one hand,  the human orientation of the ICH approach stresses asymmetries and social inequalities in terms of participation, self-determination, recognition, and safeguarding of ICH, on the other, it could be a valuable tool to protect communities against misappropriation (and/or undermining) of their cultural heritage. Indeed, as Eichler points out, “cultural bearers’ equal access to participation in ICH traditions, practices and so on may well need to be understood in human rights terms, beyond cultural heritage regimes” (Eichler 2021: 799). In other words, ICH has hitherto been considered under the umbrella of cultural diversity and heritage, rather than within the framework of international human rights law as such. The re-positioning and acknowledgement of ICH as a human right could provide multiple opportunities to mitigate the impact of inequality, enhancing diversity in terms of identities, expressions and decisions about what ICH constitutes for minority groups.

Another challenge is ensuring that digital preservation efforts are made sustainable. Many digital projects rely on external funding, which can be short-term and subject to changes in political or economic priorities. To ensure that digital preservation efforts are successful in the long term, it is essential to develop sustainable funding models that provide ongoing support for language preservation initiatives. This could include partnerships between governments, private sector organizations, and non-profits, as well as the development of community-driven funding models that allow language communities to take control of their own preservation efforts.

Overall, despite the risks that have been discussed here, the opportunities for minority languages in terms of digital preservation are noteworthy. Digital technologies have the potential to democratise language preservation by making linguistic resources accessible to a global audience. For example, online language learning platforms like Duolingo have introduced courses in endangered languages, such as Irish, Hawaiian, and Navajo, providing speakers of these languages with new opportunities to learn and practice their ancestral languages. Similarly, digital archives and databases provide valuable resources for researchers and educators, who can use these tools to document and revitalise endangered languages.

Digital platforms can also offer new opportunities for community-driven language preservation. Crowdsourcing initiatives, where community members contribute to the documentation and preservation of their languages, have become increasingly popular in recent years. These initiatives allow communities to take an active role in preserving their linguistic heritage, ensuring that the knowledge and expertise of local speakers are incorporated into digital preservation efforts.

A successful example of community driven actions is represented by the Masakhane project[3] for African languages. In response to linguistic marginalisation reinforced by digitalisation, researchers across Africa have launched an open source, pan-African artificial intelligence initiative aimed at developing machine translation systems for African languages. Named after the isiZulu phrase meaning “We Build Together,” the project seeks to empower Africa’s rich linguistic diversity (including over 2,000 languages) by fostering technological inclusion and connectivity. Central to Masakhane's philosophy is the conviction that Africa’s engagement in the Fourth Industrial Revolution cannot be mediated solely through English. Despite the widespread presence of colonial languages such as English, French, and Portuguese, the majority of African languages remain digitally invisible. This exclusion reinforces existing inequalities such as access to information and services, and participation in the global digital ecosystem is predominantly reserved for speakers of dominant Western languages. Masakhane challenges this trend by striving to ensure that African linguistic heritage is not left behind in the digital age.

4.1 Bias in AI-Based Language Technology

Artificial Intelligence (AI) and Machine Translation (MT) technologies may offer further potential solutions for expanding the linguistic reach of the Internet. AI-powered translation services, such as Google Translate, have made significant advances in recent years, with support for an increasing number of languages. While these services are far from perfect, one cannot deny that they provide a valuable resource for enabling communication between speakers of different languages and for making digital content more accessible to speakers of minority languages.

Nonetheless, it is worth analysing important concerns about the increasing reliance on MT and the consequent marginalisation of minority languages. More specifically, the language modelling bias could be seen as one of the major causes of the recently coined expression “digital language divide”. Indeed, it has been reported (Kornai 2013) that about 10 years ago, less than 5% of the world’s languages were represented on the Internet. However, no recent improvement of the linguistic gap between major and minority languages can be noticed nowadays (Joshi et al. 2020). Moreover, machine translation systems are typically trained on large datasets of digital content which are overwhelmingly dominated by major languages, with English in the foreground. As a result, translations of minority languages may be less accurate due to the limited availability of datasets for these languages. The issue of translation accuracy when dealing with under-resourced languages has been addressed by recent research (e.g. Yier et al. 2024). Interestingly, it has been observed that LLMs often underperform compared to NMT systems, primarily due to insufficient parallel data and the complexities introduced by diverse training data. In particular, a study by Song et al. (2025) has systematically evaluated LLM performance across 200 languages using benchmarks like FLORES-200. The findings confirm that while LLMs have advanced, significant performance gaps remain when translating into low-resource languages, especially when compared to specialized NMT systems.

In addition, translations into low-resource languages often exhibit significant interferences from the source language, a phenomenon that compromises the naturalness and fluency of the target output. As Toral (2019) notes, the lack of sufficient parallel corpora and linguistic resources leads to translations that closely mimic the syntactic and lexical patterns of high-resource source languages, rather than adhering to the norms of the target language. Similarly, Vanmassenhove et al. (2021) demonstrate that neural machine translation systems tend to produce output in low-resource languages that is overly influenced by the structure and style of dominant source languages, resulting in degraded grammaticality and reduced idiomaticity. These interferences underscore the importance of developing translation models that are better attuned to the specific linguistic features of under-resourced languages. Hence, the use of machine translation may encourage the homogenisation of language, as speakers of minority languages may rely on translations into dominant languages rather than developing digital content in their own languages.

Interestingly, recent work (Helm et al. 2024) has addressed the issue of the digital language divide and linguistic bias, advancing an analysis of the main causes underlying the problem. First, the authors maintain that language modelling bias can result in systems that, while being precise regarding languages and cultures of dominant powers, are limited in the expression of socio-culturally relevant notions of other communities. Second, at the root of this issue, they identify a systematic tendency of AI-powered tools to apply a simplistic understanding of the concept of diversity itself, that is, one which neglects the essential differences that languages, as well as the cultural communities speaking them, represent and embody.

For a deeper understanding of the concept, it is worth recalling the distinction, borrowed by the authors from Tsing (2012), between meaningful and scalable diversity. The former refers to diversity that changes things while the latter refers to the type of diversity that can only be integrated into pre-existing standards. This would explain the criticism of current attempts of AI-powered systems to increase linguistic diversity (that is, by simply multiplying the number of linguistic options available). Indeed, as the authors stress, “these attempts fail to account for the more profound cultural and epistemological differences, which are incorporated into different languages and which, as we claim, should be at the heart of the diversification effort” (Helm et al. 2024: 3). Thus, drawing on the concept of epistemic injustice, broader ethico-political implications are addressed, showing how AI can lead not only to a disregard for valuable aspects of diversity but also to an under- representation of the needs of marginalised language communities.

To consider these challenges and avoid the undermining of essential aspects of linguistic and cultural diversity, a more comprehensive approach to languages in the digital age is needed. This includes not only the development of digital tools and content in minority languages but also broader policies that promote multilingualism and linguistic inclusion. Governments, international organizations, and the private sector must work together to ensure that all languages have a place in the digital world and that speakers of minority languages have access to the resources they need to participate fully in digital spaces.

5. Concluding remarks

Considering the crucial role of digital platforms nowadays in terms of communication, education, and cultural exchange, it becomes essential to ensure that all languages have a place in the digital world for safeguarding the world’s intangible cultural heritage. If on one hand the dominance of a few major languages, most notably English, in the digital spaces poses significant challenges for minority languages, on the other, digital technologies may also offer new tools and opportunities for language preservation.

As has been argued, one of the key obstacles in promoting multilingualism online is the economic and technical barriers faced by minority language communities. Developing digital content in minority languages requires significant investment in infrastructure, technology, and education. In many cases, minority languages lack the standardisation and resources needed for creating digital content, such as dictionaries, grammar guides, and language learning materials. As a result, speakers of these languages are often excluded from participating fully in the digital economy and global culture.

Beyond these economic barriers, there are also technical challenges to promoting multilingualism online. Many digital platforms are not designed to accommodate a wide range of languages, particularly those that use non-Latin scripts or have complex grammatical structures. This can make it difficult for speakers of minority languages to create and share content in their languages, further marginalising them in digital spaces.

On the other hand, as remarked, the digital space also offers valuable opportunities for promoting multilingualism online. The development of machine translation technologies has the potential to increase linguistic inclusivity by providing automatic translation between languages. However, machine translation alone is not sufficient to address the broader issue of linguistic marginalisation. More careful ethical considerations on the concept of linguistic diversity are needed to develop a truly inclusive approach that considers the undermined aspects of languages as cultural heritage. While translation services can help bridge the gap between dominant and minority languages, they often fail to capture the cultural nuances and complexities of minority languages. To promote linguistic sustainability, it is essential to develop digital tools and contents that are specifically designed for minority languages, rather than relying solely on translations from dominant languages.

Thus, to conclude, the protection of linguistic heritage requires a comprehensive approach that includes sociolinguistic acknowledgement of the concept of language as an essential part of heritage, legal frameworks, digital tools, and community-driven initiatives. International legal instruments such as the European Charter for Regional or Minority Languages and UNESCO’s Convention for the Safeguarding of Intangible Cultural Heritage certainly provide important protections for linguistic diversity, but more work is needed to ensure that these guidelines are effectively implemented. If the digital tools reviewed, such as language learning apps and online archives, may offer valuable resources for documenting and revitalising endangered languages, these tools should be developed in collaboration with the communities whose languages are being preserved.

As the world continues to move into the digital age, it is essential that governments, international organisations, and communities work together to create sustainable, inclusive strategies for preserving linguistic diversity. By adopting responsible digital preservation practices, promoting multilingualism in cyberspace, and ensuring that communities retain control over their linguistic and cultural heritage, we can ensure that the voices of all languages continue to be heard, enriching human culture and knowledge for generations to come.

Acknowledgments

I would like to thank Professor John Patrick Leech for his valuable suggestions and constructive feedback, and the two anonymous referees for their insightful comments, which greatly contributed to the improvement of this paper.

About the author(s)

Francesca D’Angelo is Junior assistant professor (RTD-a) in English Language, Translation and Linguistics (ANGL-01/C) at the Alma Mater Studiorum University of Bologna, Department of Translation and Interpreting. She conducted her Ph.D. study on Third Language Acquisition between the University of Salerno and the University of Edinburgh. She is specialised in bilingualism, ESPs (legal English), translanguaging, translation, and gender studies from a sociolinguistic perspective. Among her most recent works: Teaching and Learning Third Languages (2023); Assessing Translation Quality: a Survey of Research into Human Translation, Post-editing and Machine Translation (2023); Integrating Multilingual Practices in Content and Language(s) Integrating Learning (2024).

Email: [please login or register to view author's email address]

• Next article