Securing the right to assimilate: How the drafting of the genocide convention helped undermine language rights

Abstract

This article focuses broadly on the problem of global language loss, and the underlying political dynamics that drive it. Noting that the United Nations currently promotes a rights-based approach to addressing this problem, I explore why language rights mechanisms are currently weak and fragmented. To do this, I focus on the tensions that exist between language rights and the sovereign rights claimed by states, based on a genealogical analysis of the drafting of the United Nations Convention on the Prevention and Punishment of the Crime of Genocide. Specifically, I examine why many drafters wanted to exclude protections for languages and linguistic minorities from the convention. My detailed narrative reconstruction shows that drafters removed consideration of languages from the convention in order to assert states’ rights to promote national languages, assimilate minorities, and prevent outside influence in domestic affairs. The article concludes by making suggestions for further research to deepen our understanding of the connections between sovereignty and language rights, while also arguing that effective solutions to the global crisis of language loss require the creation of a more robust language rights regime that acknowledges and confronts the implicit rights that states claim for themselves.

Introduction: Global language loss, language rights, and state sovereignty

Conservative estimates currently suggest that at least half the languages in the world will no longer be used at the end of the 21^st century (Campbell and Belew, 2018). The language loss we are now observing is both quantitatively and qualitatively different from what has come before (Hale et al., 1992; Nettle and Romaine, 2002; Roche, 2022a; Wurm, 1991), and research suggests that the rate of language loss is continuing to accelerate (Bromham et al., 2022). Recognizing that Indigenous peoples are disproportionately impacted by this ongoing crisis, the United Nations has launched an International Decade for Indigenous Languages that will run from 2022 to 2032, with the aim of preserving, revitalizing, and promoting Indigenous languages (UNESCO, 2022). Central to this endeavor is a rights-based approach to protecting linguistic diversity.

The United Nations’ adoption of a rights-based approach to language builds on work to theorize, advocate for, and legally encode language rights, which stretches back over 40 years, but has intensified significantly in the twenty-first century (Jiménez-Salcado and Mowbray, 2023; Kymlicka and Patten, 2003; May, 2011, 2023; Mowbray, 2012). Enshrined in a number of international and domestic laws, as well as regional agreements and non-binding declarations, language rights aim to protect the freedom of individuals and groups to use and maintain their languages (Skutnabb-Kangas and Phillipson, 2023). One of the most detailed and robust examples of language rights in international law is found in the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP, 2007), which enshrines rights to revitalize, use, develop and transmit Indigenous languages (Article 13), rights to establish control, and access education systems in Indigenous languages (Article 14), and the right to establish media in Indigenous languages (Article 16). It furthermore safeguards Indigenous languages by ensuring freedom from any kind of discrimination (Article 2) and from forced assimilation or destruction of Indigenous culture (Article 8).

At the same time, the United Nations Declaration on the Rights of Indigenous Peoples also demonstrates one of several significant shortcomings of language rights mechanisms in international law, insofar as it contains an ‘escape clause’ that empowers states to reject any initiative that can be construed as “authorizing or encouraging any action which would dismember or impair, totally or in part, the territorial integrity or political unity of sovereign and independent States” (UNDRIP, 2007). A second limitation of language rights mechanisms in international law is that various rights are scattered across a patchwork of different mechanisms; instead of a single convention or treaty on language rights, we have a fragmented scattering of language-related articles and clauses in mechanisms that focus on other issues (de Varennes and Kuzborska, 2016; Dunbar, 2023). ¹ In addition to the rights scattered among these laws, other non-enforceable conventions and statutes exist that have overlapping jurisdiction. ² Furthermore, this weak, fragmented web of language rights exists in a world where most states are officially monolingual (Jiménez-Salcedo and Mowbray, 2023), and continue to promote linguistic homogenization through policies based on ideologies and practices that draw on monoglot nationalism (Bauman and Briggs, 2003; May, 2012; Roche, 2024), colonial epistemologies and hierarchies (Heller and McElhinny, 2017; Meighan, 2023), and neoliberal market rationality (Piller and Cho, 2013). Meanwhile, intensifying assaults on human rights defenders and civil society organizations around the world are currently undermining the ability of individuals and organizations to advocate for more robust protections of their language rights (Roche et al., 2023).

Within this context, this essay is motivated by the following question: why do we lack strong mechanisms to enshrine and protect language rights despite recognizing an urgent and widespread need for them?

To answer this question, I draw on theories of the modern state that see its sovereignty as a bundle of rights claimed over and against those of citizens and other residents of its territory. I take inspiration from the literature on biopolitics, originating in the work of Michel Foucault (2003; 2004, 2007, 2008), that sees sovereignty as fundamentally rooted in the state’s right to kill: to secure life and promote flourishing for the state’s ‘normal’ population by killing those who are considered threatening to the state’s normal functioning (Mbembe, 2003). ³ Drawing on this approach, Jasbir Puar, for example, has theorized that modern states exercise a ‘right to maim’: to debilitate through a combination of bodily injury and social exclusion (2017). I connect this approach with studies that examine the state’s prerogative to homogenize and rationalize its populations through violence (Acheson, 2023; Nagengast, 1994; Rae, 2002; Jun, 2019; Scott, 1998), and to do so with impunity (Cronin-Furman, 2022), including what legal theorist Leo Kuper describes as “the right to commit genocide” (Kuper, 1981: 161). Finally, I also draw on critical human rights studies that problematizes the liberal, progressive idea of rights as inherently liberatory, and instead explore how rights can be used as ‘weapons’ for projects that defend inequalities (Bob 2019) and entrench and justify domination (Perugini and Gordon, 2015). Broadly, therefore, my approach sees individual and group rights, including language rights, as always in tension with the de jure and de facto rights claimed by states as expression of sovereignty.

In combining a biopolitical theory of sovereignty with research on state violence and a critical approach to rights, I can reframe the motivating question above, to ask instead: what rights do modern states claim that effectively undermine language rights in domestic and international law? To answer this question, I turn to the UN Convention on the Prevention and Punishment of the Crime of Genocide, and explore the history of its drafting for clues as to how states limit language rights as part of their efforts to assert sovereignty.

Genocide and languages

The UN Convention on the Prevention and Punishment of the Crime of Genocide provides an important case study through which to examine the relationship between sovereignty and language rights. A brief historical excursus explaining the origin of the term will explain why.

The concept of genocide originated with the Polish jurist Raphael Lemkin, who introduced the concept in his 1944 book Axis Rule in Occupied Europe, in which he defined genocide as “the destruction of a nation or of an ethnic group.” ⁴ He clarified that genocide did not necessarily refer to the “immediate destruction” of such a group, but rather, to “a coordinated plan of different actions aiming at the destruction of the essential foundations of the life of national groups, with the aim of annihilating the groups themselves” (Lemkin, 2008 [1944]: 79). Lemkin then went on to classify eight categories of ‘different actions’ that might make up the ‘coordinated plan’ of genocide: political, social, cultural, economic, biological, physical, religious, and moral. Language is mentioned in several categories. For example, in ‘political’ techniques of genocide, Lemkin discusses the changing of placenames and public signs from one language to another, and in ‘social’ techniques he discusses changes to the language used in the courts. However, language is discussed most extensively in regard to cultural techniques of genocide, where he describes a situation in which the local population of a conquered territory “is forbidden to use its language in schools and in printing” (Lemkin, 2008[1944]: 84). He then goes on to talk about control over publishing and other artistic production, and the destruction of libraries and archives. For Lemkin, then, the state suppression of language, and efforts to limit the languages that could be used in the public realm, was an integral part of genocide.

When genocide was finally codified in international law by the United Nations, however, its excluded all mention of language. When approved by the UN General Assembly on 9 December 1948, the UN Convention on the Prevention and Punishment of the Crime of Genocide (UN Resolution 260[III]) defined genocide with reference to the following acts (when “committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such”):

a. Killing members of the group;

b. Causing serious bodily or mental harm to members of the group;

c. Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part;

d. Imposing measures intended to prevent births within the group;

e. Forcibly transferring children of the group to another group.

Emerging at a pivotal moment in the creation of international law, the Genocide Convention had a powerful potential to establish norms and expectations about the appropriate regulation of state behavior and to define the rights that could be exercised, in law or fact, as part of sovereignty. Had the convention clung more closely to Lemkin’s definition of genocide, it would have created legal mechanisms and precedents for regulating state behavior in relation to languages, in ways that could have supported a more robust regime of language rights in international law than the weak and fragmented one we currently have. Examining the drafting process of the convention offers clues as to why the drafters decided to exclude languages from the final text and what bearing, if any, attempts to define and defend sovereign rights had on their arguments and decisions. Although others have traced more generally the historiography of genocide (e.g., Akhavan, 2012; Churchill, 1997; Kuper, 1981; Meiches, 2019; Moses, 2021; Short, 2016; Starblanket, 2018; Stone, 2010), and the fate of cultural genocide therein (Bachman, 2019; Moses, 2010; Novic, 2016; van Krieken, 2010), this study is the first to specifically focus on the issue of language in the convention, and consider its relation to state sovereignty. ⁵

To explore these issues, I use Foucault’s genealogical methodology (Bleiker, 2000; Foucault, 1977, 2004; McWhorter, 2009; Meiches, 2019; Scott, 2014). Genealogy is a historical methodology that highlights the role of power in forming our commonsense understandings of what is true, normal, and routine, and how these are expressed and experienced through things that seemingly lack history, such as our “sentiments, love, conscience, [and] instincts” (Foucault, 1977: 139). It seeks to reveal the struggles that led to particular forms of knowledge being “buried and disqualified” (Foucault, 2004: 8), and in doing so, to expose the violence that operates in our current common-sense understandings, showing how the past “actively exists in [and] continues to actively animate” the present (Foucault, 1977: 147). Such reconstruction is not carried out for its own sake, but rather with a liberatory aim; genealogy aims to destabilize assumptions, deprive concepts of their naturalness, and develop forms of counter-knowledge that “can be effective in disrupting stable patterns of life that dominate, suppress, and otherwise injure people” (Scott, 2014: 167). This article, then, examines how language was removed from genocide in order to uncover the historical agents who created this rupture, and to reconstruct their motives, in order to better understand how language rights might be better defended in the present.

To do this, I offer a close reading of the travaux préparatoires, or preparatory documents, surrounding the drafting of the Genocide Convention, as collated by Abtahi and Webb (2008). As the collators explain, they aimed to “gather together in a single publication the records of the multitude of meetings which led to the adoption of the Genocide Convention” (xxiv), thus bringing out “the voices of the negotiators and drafters of the Convention by providing readers with the actual texts to interpret for themselves” (xxviii). Documents for the collection were assembled by a variety of methods: searching electronic databases, cross-referencing physical documents, and trawling through records at The Hague and the UN’s Dag Hammarskjöld Library in New York City. Through this process, the editors claim to have collated “every single relevant document” (xviii) produced during the drafting process between 1946 and 1948, consisting of 337 unique documents and including drafts of the convention, commentaries by various parties, meeting minutes, working documents, and corrigenda (Abtahi and Webb, 2018; Webb, 2023).

In order to reconstitute how language was removed from the genocide convention, I read through the documents collated by Abtahi and Webb (2008) and collected every mention of language onto a series of index cards, which I then used to reconstruct the narrative presented here. My primary aim was narrative reconstruction: to establish how the text of the convention changed over time and who said what, in which context. My hope is that the rich, detailed narrative that I present here will not only support my arguments about language rights and sovereignty, but that it will also enable others to advance different claims about the relationship between language and genocide.

The narrative that follows is divided into three main stages, each corresponding to a different draft of the convention: the initial draft by the UN Secretariat, a subsequent draft created by a specially-appointed Ad Hoc Committee, and a final draft created by the UN Legal Committee. After presenting this reconstructed narrative, my conclusion summarizes the bundle of rights that I believe the drafters of the Genocide Convention were seeking to defend when they removed language from the convention. I explore how this finding relates to current efforts to protect threatened languages through language rights mechanisms, and make some suggestions for further research to help expand our understanding of this issue.

The issue of languages in the drafting of the UN Genocide Convention

Resolving to draft a Genocide Convention

The concept of genocide as originally introduced into UN forums for discussion was broad and vague, and although that definition did not explicitly refer to language, some delegates were already drawing this connection.

On the 9th of November, 1946, during the first session of the UN General Assembly, Cuba, India, and Panama introduced a draft resolution on genocide, which they defined as the destruction or the denial of the right to exist of “racial, ethnical or religious groups,” while describing it as a crime that “shook the conscience of mankind” (Abtahi and Webb, 2008: 3, A/BUR/50). ⁶ During the ensuing discussions of this proposal, the Lebanese delegate, addressing the Legal Committee, tied the proposed convention to the issue of language for the first time, referring to precedents in the Pan-American Union, which included resolutions “condemning persecutions for reasons of language, race or religion” (emphasis mine, Abtahi and Webb, 2008: 13, A/C.6/91).

Then, on the 11th of December, 1946, UN Resolution 96(I): The Crime of Genocide, was passed by the General Assembly. The resolution confirmed that genocide was a crime, and requested the Economic and Social Council to undertake research in order to draft a convention. The resolution defined genocide broadly as “a denial of the right of existence of human groups,” referring to “racial, religious, political and other groups,” but making no mention of language (Abtahi and Webb, 2008: 34). Resolution 96(I) was then forwarded to the Economic and Social Council, where it was decided that the Secretariat should prepare a draft of the convention that could then be sent to the Committee on the Development and Codification of International Law for discussion.

Stage 1: The secretariat draft

The first draft of the Genocide Convention to be introduced for discussion was prepared by three legal experts—Raphael Lemkin, Henri Donnedieu de Vabres, and Vespasian V. Pella—and is known as the Secretariat Draft (E/447). Language was explicitly included within the text of this draft, but immediately emerged as a point of contention between the three drafters, and between the state delegates who discussed the draft. Many of the arguments that we will meet later in the drafting process appear here in embryonic form.

The Secretariat Draft included language in its definition of genocide in several ways. To begin with, ‘linguistic groups’ were included in the list of protected groups in the convention’s opening section, which stated, “The purpose of this Convention is to prevent the destruction of racial, national, linguistic, religious, or political groups of human beings” (emphasis mine, Abtahi and Webb, 2008: 115, A/AC.10/42). Language was also included in the description of genocidal acts, which were broadly divided into three types: (1) physical genocide (‘causing deaths’ to members of the group); (2) biological genocide (‘restricting births’ of members of the group); and (3) cultural genocide (‘destroying the characteristics of the group’) (Abtahi and Webb, 2008: 115, A/AC.10/42). Language is the explicit focus of two of the five acts listed under the category of cultural genocide, namely (Abtahi and Webb, 2008: 215, E/447):

c. prohibition of the use of the national language even in private intercourse;

d. systematic destruction of books printed in the national language or of religious works or prohibition of new publications.

The UN Secretariat, in its instructions to Lemkin, de Vabres, and Pella, raised specifically the question of protected groups, noting that humans exist in “racial, national, linguistic, religious and political groups,” and asked, “Should the Convention on genocide protect all or only some of these?” (emphasis mine, Abtahi and Webb, 2008: 224, E/447). Later, in their comments on the draft, the Secretariat’s commentary pointed out that linguistic groups had been added to the list of protected groups provided in UN Resolution 96(I) (Abtahi and Webb, 2008: 230, E/447).

The inclusion of linguistic groups and language in the Secretariat Draft was largely due to Lemkin’s advocacy: the other two drafters opposed their inclusion. De Vabres, for example, expressed concern regarding, “how to define against what kinds of groups the crime of genocide might be committed: e.g., racial groups only (for instance, the Jews), or would religious, linguistic and political groups be covered too?” (emphasis mine, Abtahi and Webb, 2008: 163, A/AC.10/SR.28). He also expressed opposition to the idea of cultural genocide, stating that, “the French delegation was unable to recognize any but physical genocide” (Abtahi and Webb, 2008: 164, A/AC.10/SR.28). He therefore opposed both the protection of linguistic groups, and including acts of cultural genocide in the convention. Pella’s views, meanwhile, are less clear, but the documents record that both he and de Vabres, “held that cultural genocide represented an undue extension of the notion of genocide and amounted to reconstituting the former protection of minorities (which was based on other conceptions) under cover of the term genocide” (Abtahi and Webb, 2008: 234, E/447).

In refuting de Vabres and Pella’s arguments against cultural genocide, Lemkin centered language, stating that the practice of cultural genocide went beyond a policy of “moderately coercive assimilation” to include, “prohibition of the opening of schools for teaching the language of the group concerned, of the publication of newspapers printed in that language, of the use of that language in official documents and courts, and so on” (Abtahi and Webb, 2008: 234-5, E/447). In justifying the inclusion of another aspect of cultural genocide—“Forced and systematic exile of individuals representing the culture of a group”—Lemkin again centered language, claiming that in the process of such exile, a group’s language “…is degraded to the rank of a dialect, and hence those using it are reduced to a position of inferiority and the language has no longer any political or social significance” (Abtahi and Webb, 2008: 234, E/447). Language was clearly important to Lemkin’s understanding of genocide.

In the ensuing discussions of the Secretariat Draft, there was little support for cultural genocide, and little mention of language. In one statement supporting cultural genocide that mentioned language, the Polish delegate stated that “...the Germans [under Bismarck] did unquestionably commit cultural genocide by forbidding not only the teaching of Polish, but even the use of that language in schools,” (Abtahi and Webb, 2008: 165, A/AC.10/SR.28). Siam (Thailand), meanwhile, expressed tacit support for the inclusion of language and cultural genocide when they sought clarification about which kinds of language might be suppressed, suggesting that “language of the group” should be used instead of “national language” (Abtahi and Webb, 2008: 639, E/623/Add.4). Others, such as the Lebanese delegate, expressed the hope that “a prominent place would be given to [cultural genocide] in the draft convention,” but did not specifically mention language (Abtahi and Webb, 2008: 589, E/SR.139).

Positions against cultural genocide and concerns about including language were much more common. The USA spoke against the inclusion of ‘linguistic groups’ within the list of protected groups, stating, “The inclusion of ‘linguistic’ groups is believed to be unnecessary, since it is not believed that genocide would be practiced upon them because of their linguistic, as distinguished from their racial, national or religious characters. Racial, national, and religious groups are covered, and that should be sufficient.” (Abtahi and Webb, 2008: 373, A/401). Elsewhere they described the creation of the genocide convention as an act of “extreme gravity,” and argued that the “prohibition of the use of language” was insufficiently “barbarous” to warrant inclusion (Abtahi and Webb, 2008: 374, A/401). France also drafted their own version of the convention, which excluded linguistic groups and limited protection to groups defined by “nationality, race, religion or opinions” (Abtahi and Webb, 2008: 577, E/623/Add.1); their draft excluded cultural genocide.

Most critics focused broadly on cultural genocide rather than language specifically. China, for example, spoke against the inclusion of cultural genocide, claiming that it was “a difficult matter and the Committee should not perhaps deal with it” (Abtahi and Webb, 2008: 607, E/AC.7/SR.37). The Netherlands opposed the inclusion of cultural genocide, because it “touches on the rights of man” and was thus best dealt with in a human rights mechanism (Abtahi and Webb, 2008: 635, E/623/Add.3). The USA stated that they accepted “...the idea of ‘physical genocide’ and ‘biological genocide’, but generally reject[ed] the idea of ‘cultural genocide’…” (Abtahi and Webb, 2008: 526, E/621). The French delegate stated that genocide should be limited to physical and biological genocide (Abtahi and Webb, 2008: 383, A/401) and that they opposed “the idea of cultural genocide, since this invites the risk of political interference in the domestic affairs of states and in respect of questions which, in fact, are connected with the protection of minorities” (Abtahi and Webb, 2008: 527, E/621 23). Finally, the UK representative “would have been ready to support a convention had it dealt only with physical genocide…” (Abtahi and Webb, 2008: 403, A/C.6/SR.42).

Following these lengthy debates, on 12 Feb 1948, the USA delegate proposed to the Economic and Social Council that an ad hoc committee be set up, consisting of France, Venezuela, Denmark, and the USA, in order to draft a new version of the convention (Abtahi and Webb, 2008: 583, E/662). This request was approved (Abtahi and Webb, 2008: 584, E/662), with the final committee including China, France, Lebanon, Poland, USSR, USA, and Venezuela.

Stage 2: Ad Hoc draft

Discussions on creating a new draft of the Genocide Convention lasted from the 1st to the 30th of April 1948, taking up a total of 26 half-day sessions. Halfway through these discussions, on the 16th of April, the Chinese delegate presented a complete draft convention to the Committee (E/AC.25/9), which became the basis for discussion. Throughout these debates, cultural genocide, and whether it ‘broadened’ or ‘weakened’ the concept of genocide, was a central concern. The role of language in cultural genocide was a focus of these discussions. Although language was retained in this draft of the convention, the debate around its inclusion intensified, and more significantly, it was eventually isolated into a distinct article on cultural genocide, separating it from the core concept of genocide. Seemingly a mere point of clarification at the time, it was to have profound significance in the next stage of the drafting process, once the Ad Hoc Committee had finalized their text.

The terms of reference from the UN General-Secretary that established the Ad Hoc Committee instructed its members to pay special attention to the issue of protected groups: “What human groups should be protected? Should all racial, national, linguistic, religious, political, or other human groups be protected, or only some of them?” (emphasis mine, Abtahi and Webb, 2008:464, E/AC.25/2). Secondly, the terms of reference also requested the committee to consider whether cultural genocide should be excluded, and “consideration given only to physical and biological genocide” (Abtahi and Webb, 2008:646, E/AC.25/2). On this issue, the Ad Hoc Committee also received a note from the Secretariat that claimed that genocide “in the most restricted sense” consisted only of physical destruction: “Thus defined, genocide is mass murder” (Abtahi and Webb, 2008: 661, E/AC.25/3). Meanwhile, in their preliminary comments on the committee’s task, the USSR explicitly confirmed the important place of language within genocide: “The concept of genocide must also cover measures and actions aimed against the use of the national language… e.g., (a) the prohibition or restriction of the use of the national tongue in both public schools and private life; the prohibition of teaching in schools given in the national tongue; (b) the destruction or prohibition of the printing and circulation of books and other printed matter in the national tongue…” (Abtahi and Webb, 2008: 697, E/AC.23/7).

Following these preliminaries, the discussion turned to the issue of protected groups. From the outset of these discussions, the ‘linguistic groups’ that had been protected from genocide in the Secretariat draft were omitted without debate: only racial, national, and religious groups were included. The justification for this more limited set of protected groups was most clearly given by the Polish delegate, who stated that it was necessary to limit protection to these three groups, or “otherwise there would be no reason for not having an endless list protecting every conceivable kind of group” (Abtahi and Webb, 2008: 718, E/AC.25/SR.4). Once the Chinese draft was accepted (by unanimous vote), the committee voted to restrict the protected groups to “national, racial, religious [and] political groups” (E/AC.25/SR.13), once again choosing to exclude linguistic groups.

Language was discussed in more detail during debates around the concept of cultural genocide. Again, the USSR advocated to include cultural genocide (which they referred to as ‘national-cultural genocide’) and in discussing the issue made specific reference to the ‘prohibition’ and ‘restriction’ of languages (E/AC.25/SR.5). The Chinese delegate affirmed this relationship between language prohibition and cultural genocide, stating that the “...suppression of [a group’s] language” was one means of destroying the group (Abtahi and Webb, 2008: 736, E/AC.25/SR.5). In another session, the Chinese delegate also referred to a hypothetical case in which British authorities, “...might have forbidden the use of a dialect in a territory… Now that the concept of cultural genocide had been accepted, the case could be reported to the United Nations by the parties concerned” (Abtahi and Webb, 2008: 809, E/AC.25/SR.8). Despite this, the Chinese delegate was also careful to point out that “...when a national Government promoted its national language that was not suppressing other languages” (Abtahi and Webb, 2008: 853, E/AC.25/SR.11). The Venezuelan delegate also took a similarly ambivalent position, stating that a ‘comprehensive’ definition of genocide should include “prohibition of a language” (Abtahi and Webb, 2008: 849, E/AC.25/SR.11), but went on to clarify that, “Foreign groups settling in a country might be a danger to the culture of the country itself if their languages were not suppressed in order to defend the national language of the country in which they lived” (p853, E/AC.25/S.11).

During these discussions, the chair of the Committee, who was also the US delegate, asked the members, “Whether they thought cultural genocide should be covered in a separate article,” with the intention that this would enable members states to notify the UN of their objections to this specific aspect of the genocide convention (Abtahi and Webb, 2008: 837, E/AC.25/SR.10). The French delegate spoke in favor of this idea, stating that it, “would be easier for them to vote on the question of cultural genocide if it formed the subject of a separate article” (Abtahi and Webb, 2008: 839, E/AC.25/SR.10). The proposal to separate cultural genocide in a distinct article was approved three votes to one with two abstentions, a decision which became highly significant in the next round of discussions.

Having thus separated cultural genocide into a separate clause, a special session was devoted to this article on the 21st of April 1948. To formulate a definition of cultural genocide, the committee took as their model a draft submitted by the USSR delegate on the 19th of April (E/AC.25/SR.12). After some discussion of the wording of the article, the following text was put forward by the Chinese delegate (Abtahi and Webb, 2008: 888, E/AC.25/SR.14):

In this convention, genocide also means any of the following deliberate acts committed with the intention of destroying the language and culture of a nation, racial or religious group on grounds of national or racial origin or religious belief:

1. prohibiting the use of the language of the group in private intercourse, in schools and in publications;

2. destroying, or preventing the use of, libraries, museums, schools, historical monuments, places of worship or other cultural institutions and objects of the group.

Language thus appeared in the definition of cultural genocide both in the opening of the article, and in the specified acts.

The French delegate immediately opposed the article, particularly in regard to the ‘delicate’ issue “...of the rights of the State with respect to minority groups and the rights of minority groups with respect to the State.” He was concerned that “some of the acts which it was proposed to include in the concept of cultural genocide might have a lawful basis.” To exemplify what he meant, the French delegate went on to state that, “...current legislation acknowledged the right of States to impose certain restrictions on the use of the national language of minority groups living in their territory.” He thus cautioned the committee to “avoid stating the problem of genocide in such a way as to incriminate States exercising their powers in a normal way” (Abtahi and Webb, 2008: 889, E/AC.25/SR.14).

Following the French delegate’s discussion, the Polish delegate concurred, noting that the convention aimed only to prevent the “violence, persecutions and excesses which aroused the conscience of mankind,” not “the inevitable absorption of certain minority groups into the national whole” (Abtahi and Webb, 2008: 890, E/AC.25/SR.14). The US delegate added their support to the position of the French and Polish delegates and recommended that the issue of cultural genocide was best dealt with elsewhere. The Lebanese delegate, meanwhile, stated that he believed that, “...the articles on cultural genocide… listed acts which were not as serious or as shocking to the conscience of the world” as physical and biological genocide. However, he conceded that, “those which could cause the complete and rapid disappearance of the culture, language or religion of a group,” should nonetheless be targeted by the convention (Abtahi and Webb, 2008: 891). The suggestion to include this perspective in the draft was, however, rejected, due to the difficulty of defining ‘rapid’ disappearance.

When a revised version of the draft article was re-introduced to the committee following this discussion, the Lebanese delegate again pushed for revisions, suggesting the addition of a third technique of cultural genocide: “Subjecting members of a group to such conditions as would cause them to renounce their language, religion or culture” (Abtahi and Webb, 2008: 892). In defending this suggestion, the Lebanese delegate stated that, “...there had been examples in history of minority groups that had been subjected to conditions that forced them to give up their language, religion or culture although no formal measures such as those enumerated in the draft had been taken against them” (Abtahi and Webb, 2008: 892–893). However, this proposal was rejected by the USSR delegate, who claimed this additional clause was unnecessary because, “It would be hard to imagine… that members of a group would give up their language for any reason other than that they were forbidden to use it” (Abtahi and Webb, 2008: 893).

After voting on the proposals put forward by the Lebanese delegate, the following text—a very slightly modified version of the previous version—was adopted as the article on cultural genocide (changes are noted in square brackets):

In this convention, genocide also means any of the deliberate acts committed with the intention of destroying the language, religion [inserted] or culture of a national, racial or religious group on grounds of national or racial origin or religious belief, such as [replaced ‘following’ in the previous draft’]:

1. prohibiting the use of the language of the group in daily intercourse or in schools, or prohibiting the printing and circulation of publications in the language of the group;

2. destroying, or preventing the use of, the libraries, museums, schools, historical monuments, places of worship or other cultural institutions and objects of the group.

This version of the article on cultural genocide, including its focus on the prohibition of languages, was part of the final draft convention that was approved by the Ad Hoc Committee on 30 April 1948, with five votes in favor, one against, and one abstention (E/AC.25/SR.26).

Following, several delegates expressed their opposition to the inclusion of this article, with specific reference to language. Venezuela expressed concern that sub-paragraph one of the article, pertaining to the prohibition of language, would “not protect the parties against accusations when they take measures with a view to protecting their own language” (Abtahi and Webb, 2008: 1046, E/AC.25/SR.26). Following, the Canadian delegate raised a number of objections, starting with the claim that cultural genocide was incomparable to physical genocide: “It was a far cry from the unspeakable crimes which had been perpetrated at the Nazi crematoria, and which had so fundamentally shocked mankind, to the prohibition of the use of a museum cherished by some particular cultural group” (Abtahi and Webb, 2008: 1225, E/SR.218). He also claimed that “... if the term [cultural genocide] were broadened to include the suppression of a minority-language newspaper or the closing of a school, confusion would inevitably ensue” (Abtahi and Webb, 2008: 1225). And finally, he returned to what he considered the incomparable nature of physical and cultural genocide, with specific reference to language: “...no one would suggest that to prevent an individual from speaking his native language could possibly constitute homicide” (Abtahi and Webb, 2008: 1225). France, the USA, and the UK all added their voices to the opposition against cultural genocide, but without specific mention of language (E/SR.219).

Despite these objections, cultural genocide, with its focus on prohibiting languages, was retained in the draft that was then sent for discussion to the UN’s Sixth Committee, which dealt with legal questions. This was to prove the decisive step in removing language from genocide.

Stage 3: Sixth committee draft

The Sixth Committee’s examination of the Ad Hoc Draft was the final stage before the convention could be submitted to the General Assembly for a vote. The debates in the Sixth Committee took place from the 21st of September to the 2nd of December 1948, consisting of a total of 73 sessions. It was during this stage that language was finally removed from the text of the Genocide Convention, by deleting the article on cultural genocide in its entirety.

From the onset, the discussion immediately turned to the question of cultural genocide. In some of the earliest sessions, ⁷ several states expressed their strong opposition to cultural genocide on the basis that it could prevent the assimilation or integration of minorities. The only delegate to do this with explicit reference to language was the Philippines, who raised concerns that including cultural genocide “...could be interpreted as depriving nations of the right to integrate the different elements of which they were composed into a homogenous whole as, for instance, in the case of language” (Abtahi and Webb, 2008: 1315, A/C.6/SR.65). Other countries made similar arguments but without explicit mention of language. The USA, for example, argued that in the case of “new countries… care should be taken... not to favour minority movements which would tend to oppose the legitimate efforts made to assimilate the minorities by the countries in which they were living” (Abtahi and Webb, 2008: 1291, A/C.6/SR.63). The Egyptian delegate expressed concern that, “the concept of cultural genocide might hamper a reasonable policy of assimilation which no State aiming at national unity could be expected to renounce” (Abtahi and Webb, 2008: 1293, A/C.6/SR.63). South Africa, meanwhile, worried that the inclusion of cultural genocide, “...went too far in respect to the protection of minorities” (Abtahi and Webb, 2008: 1301, A/C.6/SR.64).

France remained a vocal opponent to including cultural genocide, speaking on the issue over two consecutive days (A/C.6/SR.63 & A/C.6/65). On the 30th of September, the French delegate repeated their claims that recognizing cultural genocide “…would soon lead to intervention in the domestic affairs of states” (Abtahi and Webb, 2008: 1295, A/C.6/SR.63) They also argued that the definition of cultural genocide was insufficiently precise, and that the issue was, anyway, related to the issue of human rights, and thus “logically came within the province of the Third Committee” (i.e., the Committee for Social, Humanitarian & Cultural Issues) (Abtahi and Webb, 2008: 1295, A/C.6/SR.63). On the 1st of October, they reiterated that cultural genocide was “not sufficiently well defined” and “came within the sphere of the protection of human rights” (Abtahi and Webb, 2008: 1321, A/C.6/SR.65).

A similar definitional claim was made by Iran, who argued that the concept of cultural genocide “...went far beyond the aims of the convention” (Abtahi and Webb, 2008: 1324, A.C6/SR.66). And finally, Venezuela also spoke against it, arguing that cultural genocide should be used only “...with reference only to violent and brutal acts which were repugnant to the human conscience” (Abtahi and Webb, 2008: 1312, A/C.6/SR.65).

Few countries defended cultural genocide, and among these, only Pakistan specifically mentioned language, in relation to the status of Urdu in India, where, although Urdu “had not been prohibited by law, it was under heavy attack” (Abtahi and Webb, 2008: 1298, A/C.6/SR.63). The Ukrainian SSR delegate stated simply that, “...no country which was genuinely anxious to combat the crime of genocide and to prevent it as well as punish it, could oppose the inclusion of cultural genocide” (Abtahi and Webb, 2008: 1319, A/C.6/SR.65). Finally, Lebanon spoke in support of including cultural genocide, saying that cultural and physical genocide “...were but two facets of one and the same act having the same origin and the same purpose, namely, the destruction of the group…” (Abtahi and Webb, 2008: 1326), and thus these two “aspects” of genocide were “indivisible” (Abtahi and Webb, 2008: 1327, A/C.6/SR.66).

The issue of cultural genocide was then not discussed again (except briefly on the 23rd of October) until the 25th of October 1948, ⁸ during the 83rd session of the Sixth Committee, when special attention was given to whether ‘cultural genocide’ should be included in the convention or not. The discussion was brief, not even lasting one session of less than 4 hours. ⁹ By comparison, Article I, which defined the basic concept of genocide, was discussed across two entire meetings, and Article II, on physical genocide, was debated for a full 14 meetings. The debate on cultural genocide was as definitive as it was brief: the Sixth Committee voted, with a clear majority, to delete the article on cultural genocide.

Although the prohibition of language had played an important role in defining cultural genocide in the Ad Hoc Committee debates, and during the opening debates of the Sixth Committee, it was hardly mentioned during the 83rd session. Only two delegates mentioned language in their defense of cultural genocide: the Byelorussian SSR and China. Byelorussia argued that cultural genocide should be retained in the convention in order to target, “…actions aiming at the destruction of the language, religion or culture of a group for reasons of national, racial or religious hatred” (emphasis mine, Abtahi and Webb, 2008: 1512, A/C.6.SR.83), and concluded by stating that the deletion of the article “...would be tantamount to stating in advance that crimes against the culture, religion or language of a group would remain unpunished” (emphasis mine, p1513). The Chinese delegate, meanwhile, declared that they were in favor of retaining the clause on cultural genocide because, “Although it seemed less brutal, that aspect of the crime against the human group might be even more harmful than physical or biological genocide, since it worked below the surface and attacked a whole population, attempting to deprive it of its ancestral culture and to destroy its very language” (emphasis mine, Abtahi and Webb, 2008: 1507).

Meanwhile, three countries argued in favor of retaining the article on cultural genocide, but suggested removing mention of the prohibition of languages. This campaign was led by Pakistan, despite their earlier championing of cultural genocide with specific reference to the fate of the Urdu language in India. Changing tactics, on the 13th of October, the Pakistan delegate submitted a proposed revision to Article III, which removed language and focused primarily on religion (Abtahi and Webb, 2008: 1983, A/C.6/229). Their commentary in the 83rd session then focused on attesting to the importance of including cultural genocide and systematically refuting previous arguments that advocated for deleting it. Then, after assuring other delegates that they “…understood perfectly that new countries desired to assimilate immigrants in order to create a powerful national unit,” (Abtahi and Webb, 2008: 1503, A/C.6/SR.83) the Pakistan delegate advocated for their revised version of the article which deleted the subparagraph dealing with the prohibition of language. This proposal was supported by Egypt with oblique reference to the Nakba, then underway in Palestine. Brazil, meanwhile, claimed to be generally opposed to retaining cultural genocide because, “…a State might be justified in its endeavor to achieve by legal means a certain degree of homogeneity and culture within its boundaries” (Abtahi and Webb, 2008: 1507, A/C.6/SR.83); the delegate then stated that if the article were retained, they favored Pakistan’s draft, which excluded mention of language.

Despite these three states specifically targeting language within cultural genocide, most were opposed to cultural genocide in its entirety. They provided a wide range of justifications, many echoing points they or other delegates had made earlier. Several countries claimed that the inclusion of cultural genocide exceeded the mandate given by draft resolution 96(I), either because cultural genocide was not mentioned (Iran, India), or because it proposed a definition of genocide that was “too broad” (Peru; Abtahi and Webb, 2008: 1513) or represented an “indefinite extension” (Belgium; Abtahi and Webb, 2008: 1515) of the concept of genocide. Numerous states similarly argued that the question of genocide more properly belonged in the jurisdiction of human rights or other mechanisms: Sweden, ¹⁰ India, South Africa, Netherlands, USA, and Canada. The Netherlands (Abtahi and Webb, 2008: 1514), also felt that the concept was “too vague,” while Denmark found it “not clearly defined” (Abtahi and Webb, 2008: 1508). Several states also highlighted what they felt was the incomparable nature of physical and cultural genocide: the Netherlands referred to the two concepts’ “existential difference” (Abtahi and Webb, 2008: 1514); Iran claimed it would be “artificial” to place the two concepts “on the same level” (Abtahi and Webb, 2008: 1511); Denmark said that it would “show a lack of logic and a sense of proportion to include in the same convention both mass murders in gas chambers and the closing of libraries” (Abtahi and Webb, 2008: 1508). The New Zealand delegate expressed a pro-assimilationist stance by arguing that certain “cultural traits” could impede the “political and social advancement” of colonized peoples (Abtahi and Webb, 2008: 1511), while the South African delegate indulged in hyperbole to express a similarly pro-assimilationist stance: “No one could, for example, approve the inclusion in the convention of provisions for the protection of such customs as cannibalism” (Abtahi and Webb, 2008: 1513). Finally, Iran included a tactical justification for their opposition: “The adoption of article III might, on account of its political implications, prevent some countries from ratifying the convention” (Abtahi and Webb, 2008: 1511).

Following these discussions, and despite a proposal from Egypt to adjourn the meeting due the large number of absent delegates, a vote was taken regarding the fate of the article on ‘cultural genocide’. The Sixth Committee voted to remove the article on cultural genocide, with 25 votes in favor, ¹¹ 16 against, ¹² 4 abstentions, ¹³ and 13 absences, thus removing any explicit mention of language from the text of the convention.

The UN General Assembly vote

The final step left was to conduct a vote on the genocide convention in the General Assembly, which was holding its third regular session at the time the convention was approved by the Sixth Committee, scheduled to finish on the 12th of December. The convention was to be discussed on the 9th of December. During this stage, a final, unsuccessful effort was made to reinsert the article on cultural genocide, along with its mention of language, into the text of the convention.

This move was initiated by the USSR delegate, who submitted six proposed amendments, one of which restored article III (Abtahi and Webb, 2008: 2038, A/766). Several delegates then spoke against the amendment. The USA reiterated its position that, “…acts such as the prohibition of the use of a certain language in schools or in daily intercourse, as also in publications” was best treated as a human rights issue (Abtahi and Webb, 2008: 2052, A/PV.178). India added their support to this position, stating that “…any attempt to destroy the language …of a group within a State was wholly reprehensible” but concurred that this was a human rights issue. The most relevant comment regarding the place of language in the genocide convention, however, came from the Venezuelan delegate, who said they had “…great hesitation in accepting the part of article III which had prohibited any ban on a group using its own language. Countries whose population was composed of immigrants and for which the defense of their national language was a vital necessity, had felt similar doubts” (Abtahi and Webb, 2008: 2047, A/PV.178).

Following these discussions, the USSR amendment on article III was rejected by 31 votes to 14 with 10 abstentions, and cultural genocide, and language along with it, were thus definitively excluded from the genocide convention. Voting could then proceed on the convention itself, which was adopted unanimously by the 56 members sitting in the General Assembly, thus passing the United Nation’s first major convention—the Universal Declaration of Human Rights was passed the next day—and permanently separating language from genocide both in the popular imagination and international law.

Discussion and conclusion: securing the right to assimilate

The rich details presented in the preceding narrative demonstrate, in line with the genealogical methodology employed, the contested, contingent, and power-laden nature of the processes that give rise to our common-sense, taken for granted concepts. Not until the final vote was completed in the General Assembly could it be assumed that genocide bore no relation to language. This ‘fact’ only emerged after a long process of debate, during which a variety of positions emerged, and were advanced with a range of rhetorical strategies and political maneuvering, such as when the Sixth Committee’s vote on cultural genocide was rushed through despite the large number of absentees. This realization opens up space to reconsider the relationship between genocide and language, to imagine how our present reality might have been otherwise, and to ask a range of new and important questions. Here, I return to the question I raised in the introduction about language rights and sovereignty, to ask: when the drafters of the Genocide Convention removed language from the text, what sovereign rights were they asserting, and how might these have contributed to the current language rights regime we have? ¹⁴

The drafters of the Genocide Convention aimed to protect a range of state rights that they saw as being threatened by the inclusion of languages and linguistic groups in the convention. The first of these to emerge in the debate was the state’s right to prevent outside interference in domestic affairs. This was raised by the French delegate during discussions of the Secretariat Draft, and again during the Sixth Committee discussions: cultural genocide was seen to invite “the risk of political interference in the domestic affairs of states” (Abtahi and Webb, 2008: 383) or potentially “lead to intervention in the domestic affairs of states” (Abtahi and Webb, 2008: 1295). The next sovereign right to appear was the state’s prerogative to limit the number of groups it was obliged to recognize and protect. This is seen in the Polish delegate’s concern that the convention would contain an “endless list protecting every conceivable kind of group” (Abtahi and Webb, 2008: 718).

During the creation of the Ad Hoc draft, new sovereignty-based justifications emerged. China emphasized the state’s right to promote a national language (“...when a national Government promoted its national language that was not suppressing other languages” Abtahi and Webb, 2008: 853), which the Venezuelan delegate also expressed concern about, worrying that the convention would “not protect the parties against accusations when they take measures with a view to protecting their own language” (Abtahi and Webb, 2008: 1046). Elsewhere, Venezuela also raised the possibility that migrants “might be a danger to the culture of the country itself if their languages were not suppressed in order to defend the national language of the country in which they lived” (Abtahi and Webb, 2008: 853). The French delegate framed their opposition to cultural genocide with explicit reference to “…the rights of the State with respect to minority groups,” including the right to “impose certain restrictions on the use of the national language of minority groups living in their territory.” Including cultural genocide in the convention would thus, according to the French delegate, “incriminate States exercising their powers in a normal way” (Abtahi and Webb, 2008: 889).

During the Sixth Committee discussions, sovereignty-based rights related to minorities and language were returned to and clarified by a range of delegates. The Philippines argued that cultural genocide “...could be interpreted as depriving nations of the right to integrate the different elements of which they were composed into a homogenous whole as, for instance, in the case of language” (Abtahi and Webb, 2008). The USA sought to defend “legitimate efforts made to assimilate …minorities” (Abtahi and Webb, 2008: 1291) and Egypt aimed to protect “…a reasonable policy of assimilation which no State aiming at national unity could be expected to renounce” (Abtahi and Webb, 2008: 1293). In attempting to extract language from the definition of cultural genocide, the Pakistani delegate assured other drafters that he “understood perfectly that new countries desired to assimilate immigrants in order to create a powerful national unit,” (Abtahi and Webb, 2008: 1503). Brazil defended states’ right to “…achieve by legal means a certain degree of homogeneity and culture within its boundaries” (Abtahi and Webb, 2008: 1507). And finally, the Venezuelan delegate reiterated their opposition to legal restraint on banning languages, claiming that, “Countries whose population was composed of immigrants and for which the defense of their national language was a vital necessity, had felt similar doubts” (Abtahi and Webb, 2008).

Delegates therefore explicitly expressed their desire to defend a range of sovereign rights when they advocated to remove language from the genocide convention. In ensuring that the convention could not regulate state behavior in relation to languages, delegates were aiming to establish and protect what they perceived as legitimate state rights: the right to delimit obligations to minorities, the right to maintain exclusive control over internal affairs, the right to promote national languages, the right to restrict or ban the use of minority and migrant languages, and the right to linguistically homogenize the state’s population, whether through integration (the Philippines) or assimilation (USA, Brazil, Pakistan, and Egypt). These rights that states secured during the drafting of the Genocide Convention collectively provided them with impunity to act against linguistic minorities; in the words of the Byelorussian delegate, these rights ensured that “…crimes against the …language of a group would remain unpunished” (Abtahi and Webb, 2008: 1513, A/C.6.SR.83). Importantly, this impunity was granted to all UN member states, regardless of the position they took during the drafting debates.

Through the drafting of the Genocide Convention, sates therefore defined and claimed a series of de facto rights, and in doing so, rejected measures they saw as incompatible with these rights, including those that would have modified state practice in ways that would have benefitted linguistic minorities, including Indigenous peoples and others. Such de facto rights are likely to have played an important role in shaping the fragmented and ineffective language rights regime we have today. However, further research is needed to better understand how the drafting of a range of international laws has promoted the rights of states over those of linguistic minorities. A fuller picture of the range of sovereign rights obtained by states through the drafting of international conventions could be obtained through genealogical studies on the Universal Declaration of Human Rights, the International Covenant on Economic, Social, and Cultural Rights, the United Nations Convention on the Rights of the Child, and other mechanisms. Such studies will be supported by embedding them within broader histories of human rights (Moses et al., 2020; Whyte, 2019). Concepts from Third World Approaches to International Law (TWAIL), which critically interrogates how international law perpetuates colonial patterns of domination, will also likely be useful in elucidating the underlying dynamics (Chimni, 2003; Gathii, 2011; Mutua and Anghie, 2000).

At a practical level, what the present study makes clear is that we need to understand language rights in relation to a countarvailing set of de facto and de jure sovereign rights claimed by all states. Efforts to promote language rights, sustain global linguistic diversity, and to preserve, revitalize, and promote Indigenous languages will need to acknowledge and confront the impunity that these rights have secured for all modern states. So long as that impunity remains in place, current efforts will continue to fall short of their stated goals, and the unfolding crisis of global linguistic diversity will continue unabated.