YouTube Itak: a description of Ainu-related videos

The study addresses how Ainu-related multimedia on YouTube reflects Indigenous communication strategies and participation. Set within growing internet adoption and contra-flows in global media, the paper situates Indigenous media as tools for knowledge preservation, self-expression, sovereignty, and political influence. It argues that Ainu Mosir—understood as living, diverse knowledge and practices—extends to online spaces and merits systematic examination of openly available multimedia. Guided by frameworks including Weave Communication (threads, knots, holes), Indigenous media functions (inwards, outwards, border), and Ainu-specific Mosir model (Pokna/Ainu/Kamuy Mosir), the research asks: (1) What types of Ainu-related multimedia content exist on YouTube? (2) Who produces them and to what degree do Ainu participate? (3) What communication strategies are present on SNS? The objectives align to identify content types, participation by ethnicity and gender of collaborators, and communication strategies across platforms.

The review spans non-web-based and online Indigenous media. Early work explored adaptation of Indigenous communication to mass media and examined film, radio, and print across regions (Africa, Australia, Latin America, North America, Europe). Studies also investigated media framing of Indigenous issues, health, and political mobilisation. With social media, research since circa 2013 has focused on Oceania and the Americas, covering blogs/webpages, digital storytelling, hashtag activism, performativity on YouTube, racism, trauma, and tourism. Frameworks identified include media appropriation, policy/regulation, Indigenous media, and Buen Vivir communication; ACIN’s Weave Communication conceptualises media tools (threads), actors (knots), and topics (holes). Ainu-specific literature has focused on early ethnographic films, Kayano’s documentation efforts, politicised documentaries, NHK archives, and pop culture representations (e.g., Golden Kamuy, games). Hayashi-Simpliciano’s Ainu Neno An Ainu and Mosir model provide an Ainu-centred, transformative, spiritually grounded approach, framing identity via ancestral, present, and future planes (Pokna, Ainu, Kamuy Mosir). Gaps identified include limited analysis of Ainu communication strategies in contemporary online media and few large-scale YouTube studies.

Design: Mixed-methods combining quantitative YouTube metadata analysis with qualitative content verification and Indigenous frameworks (Weave Communication; Mosir model; Indigenous media functions). Data source and sampling: 428 Ainu-related YouTube videos were curated via an account-guided recommendation process and YouTube Data API extraction. Inclusion emphasised videos showing Ainu people or voices; most live videos and items without Ainu presence were excluded. Where multiple versions existed, the more interactive or English-subtitled version was selected. Variables via API: published date, title, description, tags, category, duration, definition (SD/HD), captions, view count. Human-coded variables: release date (original public release vs YouTube upload), conceptualisers, producers, presenters, and place of recording. Collaborator ethnicity categories: Ainu, Japanese, foreign, unclear; perceived gender: man, woman, unclear. Coding: Four researchers independently coded and reconciled classifications; videos grouped as produced by Ainu only, produced with Ainu, or produced without Ainu. Analysis: Non-parametric statistics (Kruskal–Wallis tests with pairwise comparisons) assessed differences across groups for time, category, duration, quality, tags/captions, views, places, collaborator ethnicity and perceived gender. Text analyses: Word frequency analyses on tags and descriptions using ConText and Japanese Text Analysis Tool; semantic networks graphed in Gephi to map cross-group differences. Theoretical lenses guided interpretation of communication functions (inwards, outwards, border) and Mosir planes.

Sample and grouping: 428 videos categorized into produced by Ainu (N≈97), produced with Ainu (N≈172), and produced by non-Ainu (N≈159). Temporal patterns: Peaks in releases and YouTube publications corresponded with policy milestones (1997 Ainu Culture Law, 2007 Indigenous recognition, 2013 Irankarapte campaign, 2018 Upopoy naming, 2019 new Ainu Law, 2020 Olympics-related promotion). Categories: Significant group differences (Kruskal–Wallis H(2)=28.52, p<0.001). Ainu-produced videos were most distinct and were primarily Education (69). With-Ainu and non-Ainu groups concentrated in Non-profits & Activism and People & Blogs. Duration: Moderate differences (H(2)=6.97, p=0.031); mean durations: by Ainu 439.97 s; with Ainu 403.51 s; by non-Ainu 399.76 s. Definition (quality): Significant differences (H(2)=29.56, p<0.001); non-Ainu group had more SD videos. Tags, captions, views: No significant differences in counts/usage (p>0.10). Most-viewed examples crossed genres (tourism, games, anime, documentaries), including two Upopoy promotions; top view counts exceeded 1 million. Place: Significant differences (H(2)=80.50, p<0.001). Most videos located in Japan, particularly Hokkaido/Yaunmosir; Ainu-produced works also featured Kanto. Ten videos were from New Zealand/Aotearoa; some from Taiwan, Singapore, North America, and Europe. Collaborators—ethnicity: Strong differences across conceptualisers, producers, presenters (p≤0.003). Non-Ainu productions had more foreign collaborators; Ainu conceptualisers were sparse in non-Ainu group; Japanese collaborators were prominent across groups. Collaborators—perceived gender: Significant differences for most roles (p≤0.001), except unclear presenters. High participation of women across categories; non-Ainu productions had more collaborators with unclear perceived gender. Tag networks: 63 frequent tags showed language (Ainu), traditional culture, music, tourism themes. Shared tag across all groups: “Ainu” (katakana). Ainu-produced videos used Ainu-language tags and references (e.g., tonkori, Upopoy, Kamuy/Imeruat). With- and non-Ainu groups used place and tourism-related tags (Akan, Kushiro, Hokkaido), ecology, and cooking terms; non-Ainu commonly used “eco” and “nature.” Description networks: Ainu-produced videos emphasised linguistics (Ainu phrases) and cited scholars (Kayano, Nakagawa, Tamura); cross-platform hashtags (Twitter, Instagram) appeared. With-Ainu videos centred on “story.” Non-Ainu used “culture,” “Japan,” “people”; locations like Shiraoi featured. Shared terms across groups included “Ainu” (katakana), “Hokkaido,” and “person” (Japanese). Sensory and food terms (salmon, fish) appeared more than venison and beef, reflecting traditional cuisine. Overall patterns: Ainu-produced content prioritised educational and inward functions (language and daily life); with-Ainu content promoted tourism (outwards/border); non-Ainu content framed Ainu within ecology and vlogging practices.

Findings demonstrate differentiated content and collaborator patterns aligned with Indigenous media functions and the Mosir model. Ainu-produced videos centre inward functions—education, language revitalisation, and everyday life—often led by women, reinforcing identity and cultural continuity (Ainu Mosir). With-Ainu productions combine inward, outward, and border functions, particularly in tourism linked to places like Shiraoi and Akan, bridging ancestral narratives (Pokna Mosir) with contemporary creative processes and future-oriented advocacy (Kamuy Mosir). Non-Ainu productions emphasise ecological framings and vlog formats, expanding outward visibility but often with fewer Ainu conceptualisers. Semantic analyses reveal distinct lexicons: Ainu-language and scholarly references in Ainu-led works vs. place/eco/tourism tags and broad descriptors in others. The temporal alignment with policy milestones underscores how institutional contexts shape online Indigenous media production and dissemination. Limited presence of Japanese broadcasters (e.g., NHK) on YouTube suggests platform/copyright constraints that may hamper broader diffusion of Ainu-related content. Transnational interactions (e.g., with Māori) and high-performing videos point to border functions and Indigenous cosmopolitanism in digital spaces, where Ainu creators leverage SNS to assert identity and counter platformed discrimination.

- Ainu-produced YouTube content is distinct and predominantly educational, foregrounding language and daily life to articulate contemporary Ainu modernities and serving mainly inward communication functions. - Productions with Ainu heavily feature tourism promotion (e.g., Upopoy, Akan), often using ambassadors and mainstream figures, serving outward and border functions by projecting Ainu culture to broader audiences and markets. - Non-Ainu productions frequently adopt vlogging practices and ecological framings. - Across groups, communication strategies prominently promote traditional culture and music; Ainu creators increasingly utilise cross-platform support (hashtags/SNS) and scholarly sources for language teaching. - Women are notably active as conceptualisers, producers, and presenters, underscoring gendered dynamics in cultural preservation and media production. - Large-scale, mixed-method analysis integrating Indigenous frameworks effectively identified Ainu-endorsed/controlled media and their characteristics in online contexts. Future research should expand to live recordings, consider overlapping identity categories, map collaboration networks, compare versions/formats, and incorporate Russian-language keywords to capture broader geographies.

- Exclusion of live recordings, despite their importance to Ainu oral traditions and performance. - Coding did not accommodate simultaneous categories (e.g., overlapping ethnicity/nationality), potentially oversimplifying identities. - Keyword strategy did not include Russian terms; minimal content linked to Russia was captured. - Limited availability of broadcaster content (e.g., NHK) on YouTube due to copyright/streaming constraints may bias the sample. - Lack of precise data on Ainu internet access and online engagement patterns limits generalisability. Future work should include network mapping, cross-format comparisons, and multilingual queries.