An ethnographic study of multilingual language policy localization with a focus on the resolution of communication problems in international Non-governmental Organizations (INGOs)

The paper addresses how multilingual language policy is localized within an INGO office in China to meet communication needs in a globalized era where multilingualism is pervasive. INGOs are characterized by decentralized decision-making, creating space for local adaptation of language policy. The study focuses on ICOM China (International Council of Museums, Beijing) to examine how communication problems are solved and how individual agents’ language practices interact with institutional and supra-national policies. The research aims to: (1) identify roles of individuals in internal and external communication; (2) describe institutionalized language policy developed to meet communication needs; and (3) explain how ICOM International’s multilingual policy is localized in China.

The review distinguishes two key constructs: language brokers, who mediate across linguistic/cultural boundaries and can influence policy (originally studied in immigrant families, later in workplaces), and language nodes, intermediaries in MNCs bridging HQ and subsidiaries, often powerful but not decision-makers. Studies highlight benefits and drawbacks of relying on such individuals, including enhanced authority and potential information bottlenecks. Limited research on IGOs/INGOs shows internal staff networks handling language tasks (e.g., UN), local peer educators serving as nodes (WHO in Tanzania), and staff expanding repertoires in ICRC. In China’s large events (e.g., Beijing 2008 Olympics), top-down planning emphasized Mandarin and English. Gaps include few studies of how individuals’ practices appropriate policy, and limited work in INGOs within traditionally monolingual contexts. The present study targets these gaps in an INGO in China.

The study uses an Ethnography of Language Policy approach to produce a thick, contextualized description of how ICOM China resolves communication problems and formulates/implements language policy. Setting: ICOM China office in Beijing, part of ICOM (HQ Paris), whose official/working languages are English, French, Spanish; China’s broader context is largely monolingual with Mandarin prioritized. Fieldwork: 3.5 months onsite prior to the 8th MPT-EXPO plus 2 weeks during the event. The event involved 98 participants from 39 countries; 14 staff organized it (7 ICOM China members, 2 volunteers including Author 1, and 5 expatriates from local museums). Author 1 worked as a volunteer-observer and later joined the news reporting group, enabling extensive access. Data: policy and working documents (office and event brochures/manuals; ICOM International and ICOM China websites), ~240 minutes of office meeting recordings, ~200 minutes of event recordings, interviews (6 organizational actors plus the ICOM Chair), and detailed field notes. Triangulation included biweekly debriefings with an independent expert, cross-checking data sources, and iterative assessment of data sufficiency. Analysis: thematic analysis with open coding of roles in solving communication problems, followed by focused coding on ICOM China’s language policy and its formation/implementation, considering sociolinguistic factors across layers (international, national, institutional). Translations were performed and verified by the authors for Chinese materials selected for inclusion.

- Roles in communication: Two staff emerged as language brokers—Eyre (Vice-Secretary) and Clara (Foreign Affairs Specialist). Eyre held greater administrative authority, brokered in official settings (e.g., hosted panels; substituted for an absent Chinese official by delivering an English speech), and made decisions to resolve crises. Clara interfaced with ICOM International, intermittently interpreted for the ICOM Chair, coordinated problem-solving, and made micro-language planning decisions in less formal settings (e.g., directing interpreters to assist diners and mandating bilingual Chinese-English food labeling for dietary needs). A volunteer, Misha, acted as a language node in unplanned scenarios (e.g., assisting guests during breaks, highlighting labeling issues), exercising initiative but without decision-making authority. Other interviewees (Chair, Martin, Chen, Heng) did not function as brokers/nodes due to roles or language proficiency. - Institutionalized language policy: Overt policy via recruitment required English proficiency, effectively making English the common language and entry threshold (e.g., job postings specifying strong written Chinese/English and CET 4+, English-related degrees for foreign affairs roles). Covert policy converted international life experience into symbolic capital, shaping position allocation and conferring ‘legitimate domination’ in communication-related decisions (e.g., Eyre’s UK study experience; Clara’s English background and Australian master’s degree). Together these policies formed a talent pool of potential brokers/nodes, structuring who mediates and who decides. - Localization of ICOM multilingual policy: In practice, Chinese dominated internal staff communication; English was used for external communication and formal events. Event materials (manuals, brochures, posters) were bilingual (Chinese-English); French and Spanish were not used. Applications for training/events required fluency in English and English documentation, evidencing the de facto prioritization of English over de jure multilingual policy. Cost and convenience were primary reasons for using English as the only translation language; other language services could be arranged at participants’ expense. Resource availability and China’s English-focused foreign language education further supported this choice. Quantitatively, the event included 98 participants from 39 countries; the organizing group comprised 14 staff.

The findings illustrate how formal top-down multilingual policies intersect with informal, emergent practices, producing localized, bilingual outcomes in a traditionally monolingual context. The emergence of language brokers and nodes reflects institutional agency oriented to solving communication problems under linguistic instrumentalism and resource constraints. English’s de facto status as lingua franca drives ‘Englishization,’ aligning with supranational norms while interacting with national policies (Mandarin promotion, English-dominant foreign language education) and institutional practices (recruitment, role allocation). The coordinated overt (English proficiency requirements) and covert (valuing international experience) policies create a talent pool from which brokers emerge, occupying decision-making roles in policy implementation and micro-planning. This multiscalar dynamic underscores the top-down–bottom-up co-construction of language policy in INGOs and highlights potential implications for power, access, and linguistic diversity.

The study identifies distinct roles—language brokers and a language node—in ICOM China’s resolution of communication problems and shows how multilingual policy from ICOM International is localized into a functional Chinese–English bilingual regime. It demonstrates that top-down policies are not implemented straightforwardly; instead, institutional and individual agency translate, adapt, and at times narrow multilingual commitments in response to local sociolinguistic realities, costs, and available resources. Contributions include a conceptualization of localization processes centered on communication problem-solving, the interplay of policy layers (international–national–institutional), and the mechanisms by which overt and covert policies produce a broker/node talent pool. The study suggests that accumulating comparable ethnographic cases across sectors and contexts can further illuminate evolving language policy practices and the ongoing dynamics of Englishization and multilingualism in international workplaces.

Single-site, explorative ethnographic case focused on an INGO office in China, which may limit transferability to other contexts. Fieldwork covered 3.5 months prior to, and 2 weeks during, a specific global event. Due to data volume, only selected materials were transcribed and translated into English for presentation.