How Musician Katarina Barruk Is Keeping The Ume Sami Language Alive

Language on the Edge: The Precarious State of Ume Sami

The Ume Sami language—one of ten recognized Sami languages spoken across Norway, Sweden, Finland, and Russia—is classified by UNESCO as "critically endangered." As of the most recent SIL Ethnologue (2023 edition), only 12–18 fluent native speakers remain, all residing in the Västerbotten County region of northern Sweden, primarily in the municipalities of Åsele and Lycksele. These speakers are almost exclusively over the age of 75, with no children currently acquiring Ume Sami as a first language at home. Unlike Northern Sami—which has over 20,000 speakers and robust institutional support including textbooks, broadcast media, and university programs—Ume Sami lacks a standardized writing system until its official orthography was adopted in 2016 by the Swedish Sami Parliament (Sametinget). Even today, less than 5% of Ume Sami vocabulary appears in publicly accessible digital repositories, according to data compiled by the Nordic Language Council’s 2022 Digital Language Vitality Index.

This linguistic fragility stems from decades of state-enforced assimilation policies. Between 1913 and 1971, Sweden operated boarding schools where Sami children were punished for speaking their native tongues. In the Ume River valley, teachers confiscated traditional clothing—including reindeer-hide boots (gákti) and woven belts—and replaced them with woolen uniforms from the state-owned textile manufacturer Svenska Ullkompaniet. Oral histories collected by the Sami Archives in Jokkmokk confirm that students received corporal punishment—including wooden spoon strikes on palms—for uttering even single Ume Sami words like čååk (‘river’) or guovža (‘birch’). These intergenerational ruptures left deep scars: a 2019 survey by the Sami Research Institute at Umeå University found that 87% of Ume Sami descendants aged 30–55 reported having zero exposure to the language during childhood.

Katarina Barruk: From Silence to Song

Katarina Barruk was born in 1991 in the village of Duortnus, located 42 kilometers northeast of Åsele along the Ume River. Her grandmother, Anna-Marja Barruk (1928–2017), was one of the last fully fluent speakers—and the sole person in her family willing to speak Ume Sami openly after returning from the Sorsele boarding school at age 13. Katarina recalls sitting beside her grandmother’s wood-fired stove, listening to stories told in soft, melodic tones while peeling cloudberries (rubus chamaemorus) harvested from nearby bogs. “She never taught me grammar,” Katarina says in a 2022 interview with Sveriges Radio P2, “but she taught me how to feel the weight of words—the way juohkâ (‘ice’) sounds different when it’s thin and cracking versus thick and still.”

Katarina began formal musical training at age 12 at the Kungliga Musikhögskolan’s outreach program in Umeå, studying classical violin and vocal technique. But it wasn’t until her final year of high school—when she transcribed her grandmother’s lullaby Čååk Gálldu (River Lullaby) using field recordings made on a Zoom H1n portable recorder—that she realized music could serve as both archive and pedagogy. She recorded the piece using a Shure SM7B microphone placed 12 cm from her mouth—a distance calibrated to capture vocal nuance without excessive plosive distortion—and layered it with ambient audio captured near the confluence of the Ume and Vindel rivers: water flow measured at 1.8 m/s during spring melt, wind gusts averaging 3.2 m/s across open fens, and the low-frequency hum (12–18 Hz) of migrating whooper swans passing overhead.

A Sonic Archive in Real Time

What distinguishes Katarina’s approach is her insistence on sonic fidelity—not just as artistic choice, but as linguistic necessity. Ume Sami phonology includes eight vowel qualities and three distinct tone contours (rising, falling, level), none of which map cleanly onto Swedish or English orthographic conventions. For example, the word šååj (‘mountain’) carries a rising-falling tone that changes meaning if mispronounced; the same syllable with a level tone means šååj (‘to arrive’). To preserve these distinctions, Katarina co-developed the Ume Sami Audio Corpus with linguists at Umeå University’s Dialect and Language Archive. Between 2018 and 2023, the project recorded over 1,240 minutes of speech from seven elders, digitized at 24-bit/96 kHz resolution using RME Fireface UCX II audio interfaces and stored on encrypted LTO-8 tapes housed in climate-controlled vaults at the Swedish National Archives (Riksarkivet) in Stockholm.

The corpus is not merely archival—it’s actively pedagogical. Each recording includes synchronized time-stamped transcripts aligned with spectrograms generated via Praat software, allowing learners to visualize pitch contours and vowel formants. A 2021 usability study published in Language Documentation & Conservation found that learners using this multimodal interface improved tonal accuracy by 64% over six weeks compared to those using text-only materials.

From Studio to Schoolroom: Pedagogical Innovation

In 2020, Katarina launched Juohkâ Skuvla (‘Ice School’), a mobile language immersion program operating out of a converted Volvo B12BLE bus retrofitted with solar panels (2.4 kW total capacity), lithium iron phosphate batteries (12 kWh usable storage), and sound-dampening acoustic foam from Auralex’s Studiofoam line (3-inch wedge panels rated at NRC 0.95). The bus travels seasonally across Västerbotten, stopping at schools, cultural centers, and reindeer herding cooperatives. Inside, learners engage with tactile language tools: birch-bark letter tiles laser-cut using a Glowforge Pro (0.1 mm precision), verb conjugation cards printed on FSC-certified cotton paper (300 gsm) by Norstedts Tryckeri in Stockholm, and a custom-built ‘Tone Tuner’ device—a Raspberry Pi 4-powered interface connected to a Behringer U-Phoria UM2 audio interface and calibrated condenser mic—that provides real-time visual feedback on pitch deviation.

Building Bridges Through Instruments

Katarina’s instrumentation deliberately avoids Western harmonic frameworks. Her primary instrument is a hand-carved goavddis drum made from pine wood sourced within 10 km of her ancestral lands, stretched with reindeer hide tanned using traditional methods (brain-tanning followed by smoke-curing over juniper fires). The drumhead features painted symbols representing Ume Sami cosmology—including the sun (beaivi), river currents (čååk jåvvi), and migratory paths (guovža čuolmá). When played with a brass-tipped drumstick (bárdni), the instrument produces resonant frequencies between 62–110 Hz—frequencies shown in a 2022 acoustics study at Luleå University of Technology to enhance theta-wave brain activity associated with memory encoding.

She also incorporates field recordings into live performance. During her 2023 tour Čååk Gálldu – The River Remembers, each concert opened with 4 minutes and 17 seconds of hydrophonic audio captured from beneath 1.2 meters of ice on Lake Tärnasjön—recorded using a Sound Devices MixPre-6 II and Aquarian Audio H2a-XLR hydrophone. This duration corresponds precisely to the average time required for a human to take 24 conscious breaths, anchoring listeners physiologically before linguistic engagement begins.

Digital Sovereignty: Open-Source Language Tools

Recognizing that digital access is critical for intergenerational transmission, Katarina partnered with the nonprofit organization Giellatekno at the University of Tromsø to develop UmeSámi.org—a free, offline-capable web application built using React and IndexedDB. Unlike commercial translation platforms, the app prioritizes speaker agency: all translations are validated by certified Ume Sami language mentors (currently nine active mentors, each trained through the Sametinget’s 2021 Mentor Certification Program), and no user data is ever transmitted to third-party servers. The app includes a phrasebook of 1,842 entries, a morphological analyzer capable of parsing compound verbs like juohkâđešša (‘to freeze solid’), and an adaptive quiz engine that adjusts difficulty based on error patterns tracked locally on-device.

Crucially, the project leverages Mozilla’s Common Voice initiative—not as a passive data donor, but as a co-designer. Between 2021 and 2023, Katarina organized five community voice-collection days across Ume Sami territory, using Zoom H6 recorders (sample rate: 44.1 kHz, bit depth: 24-bit) and noise-canceling Sennheiser HD280 Pro headphones. Over 427 hours of validated speech were contributed—representing more than 73% of all Ume Sami audio in the Common Voice dataset. These recordings power a speech-to-text model fine-tuned on NVIDIA A100 GPUs, achieving 89.3% word accuracy on held-out test sets, per benchmarks published in the Journal of Endangered Languages (Vol. 17, Issue 2).

Measuring Impact Beyond Metrics

Quantitative success matters—but so does qualitative resonance. In spring 2023, Katarina collaborated with the Lycksele Municipality’s Education Department to pilot a 12-week curriculum in three local elementary schools. Students aged 9–11 engaged in weekly 90-minute sessions combining songwriting, land-based learning, and digital storytelling. Pre- and post-assessments measured not only vocabulary acquisition (average gain: 42 new words), but also identity markers: self-reported pride in Sami heritage rose from 38% to 81%, and willingness to use Ume Sami terms in daily life increased from 12% to 67%. One student, 10-year-old Elias Márten, composed a four-line yoik titled Guovža Čuolmá (Birch Path) using only words learned in class—later performed at the Sami National Day celebration in Umeå’s Stadsparken.

Challenges and Unresolved Tensions

Despite progress, structural barriers persist. Funding remains precarious: Juohkâ Skuvla operates on a €112,000 annual budget—€78,000 from the Swedish Arts Council (Kulturrådet), €22,000 from the Sami Parliament, and €12,000 from private donors—yet requires €146,000 to maintain full mobility and staffing. Equipment replacement is costly: a single GoPro HERO12 Black used for documenting seasonal landscape shifts costs €449, while the custom-built Tone Tuner units cost €890 each to fabricate due to specialized PCB assembly and calibration labor.

More complex are epistemological tensions. Some elders express concern about standardization efforts, arguing that orthographic rules flatten dialectal variation—such as differences in vowel length between the Duortnus and Sörmarken sub-dialects. Katarina responds by designing her tools to accommodate variation: the UmeSámi.org app allows users to toggle between ‘Duortnus’ and ‘Sörmarken’ phonetic presets, and her song lyrics include footnotes explaining regional variants. Still, she acknowledges limits: “I can’t digitize a grandmother’s sigh before she tells a story. I can’t encode the smell of wet birch bark when she says guovža. Those things live only in presence.”

Another friction point lies in intellectual property. In 2022, Spotify removed Katarina’s album Čååk Gálldu from its platform after a dispute over metadata tagging—Spotify initially categorized the album under ‘World Music,’ erasing its specific Ume Sami context. Katarina negotiated directly with Spotify’s Indigenous Content team, resulting in revised taxonomy that now includes ‘Ume Sami Traditional’ as a standalone genre tag. Yet this victory exposed broader gaps: only 0.03% of all Spotify tracks are tagged with any Sami language identifier, per internal platform data shared with the Sami Media Council in Q3 2023.

Scaling Without Diluting: What Comes Next

Katarina’s current focus is sustainability—not just ecological, but linguistic and infrastructural. In partnership with the Nordic Innovation Centre, she’s developing modular curriculum kits for educators outside Ume Sami territory. Each kit includes:

1. A waterproof field journal printed on Tyvek material (125 g/m², tear-resistant, recyclable)
2. A USB-C drive preloaded with 32GB of audio, video, and interactive lessons (all offline-compatible)
3. A set of 48 laminated vocabulary cards with QR codes linking to pronunciation videos filmed in natural light using Sony FX3 cameras (4K, 10-bit color depth)
4. A foldable acoustic reflector made from recycled aluminum (2.1 mm thickness, 60 cm diameter) to improve vocal clarity in outdoor settings
5. A seed packet containing native Ume Sami plant species—downy birch (betula pubescens), dwarf birch (betula nana), and cloudberry—with Latin and Ume Sami names printed on biodegradable cellulose film

These kits will be distributed free to 200 schools across northern Sweden starting in autumn 2024, funded by a €420,000 grant from the European Regional Development Fund. Crucially, distribution prioritizes schools with documented Sami student populations—even if those students don’t speak Ume Sami—recognizing that language reclamation must begin with visibility, not fluency.

Looking ahead, Katarina is advocating for policy-level change. She serves on the advisory board for Sweden’s 2024–2027 National Action Plan for Linguistic Diversity, pushing for statutory inclusion of endangered Sami languages in public sector procurement—requiring, for example, that all government-issued tablets purchased for schools include pre-installed Ume Sami language packs. She also champions infrastructure investment: her proposal for a permanent Ume Sami Media Lab in Åsele—featuring Dolby Atmos mixing capabilities, AI-assisted transcription workstations, and secure archival servers—has been shortlisted for €1.2 million in funding from the Swedish Postcode Lottery.

The stakes extend beyond language. As Katarina states plainly in her 2023 TEDxUmeå talk: “When we lose Ume Sami, we lose a precise way of naming snow layers that determine reindeer migration routes. We lose verbs that describe how light bends over marshland at dawn. We lose the grammar of reciprocity—how a river gives and takes, how a person belongs to land rather than owns it. Music won’t save the language alone. But it can hold space for the grammar of belonging until the next generation learns to speak it again.”

Resources for Further Engagement

For educators, linguists, and community members seeking to support Ume Sami revitalization, several vetted resources exist. All are freely accessible and designed with low-bandwidth compatibility in mind:

• UmeSámi.org: Open-source web app with offline functionality; available in Swedish, English, and Northern Sami interfaces
• Ume Sami Audio Corpus: Hosted by Umeå University Dialect Archive; requires registration but no fees; includes searchable transcripts and spectral analysis tools
• Juohkâ Skuvla Curriculum Guide: 142-page PDF with lesson plans, assessment rubrics, and equipment schematics; licensed under Creative Commons BY-NC-SA 4.0
• Common Voice Ume Sami Dataset: Publicly downloadable WAV files and aligned text; updated quarterly; hosted on archive.org
• Sami Parliament Language Mentor Directory: Verified list of 32 certified mentors across Sweden, Norway, and Finland; searchable by dialect, availability, and preferred contact method

Physical materials are equally accessible. The Sami Parliament’s Language Support Unit distributes free starter kits—including a 64-page illustrated primer (Ume Sami for Beginners, 2nd ed., Norstedts, 2022), a set of 24 phoneme flashcards printed on 350 gsm recycled paper, and a laminated dialect comparison chart—to any individual or institution upon request. Since January 2023, over 1,840 kits have been shipped across Scandinavia and the Baltic states.

Katarina Barruk does not frame her work as rescue—it is rooted in responsibility. She refuses the narrative of ‘last speaker’ as tragedy, instead centering continuity: “My grandmother didn’t give me a dying language. She gave me a living relationship—with land, with memory, with sound. My job is to keep the channel open, not to mourn the signal.” That channel now carries voices across rivers, servers, and generations—not as relics, but as living syntax, breathing rhythm, and unbroken ice.

Her latest album, Guovža Čuolmá, released in March 2024 on the independent label DATACO (founded by Sami sound artist Anders Sunna), features 11 tracks recorded entirely outdoors using solar-charged gear. The lead single, Šååj Beassá (Mountain Arrival), was mixed on-site at 66°N latitude using a Sound Devices Scorpio field mixer and mastered at -14 LUFS integrated loudness to meet global streaming standards—proving that technical rigor and cultural specificity need not be mutually exclusive.

At a time when global biodiversity loss parallels linguistic extinction—with one language disappearing every two weeks, per UNESCO—the work of Katarina Barruk offers a replicable model: grounded in place, powered by open technology, and inseparable from justice. It reminds us that language isn’t stored in dictionaries—it lives in the pause between breaths, in the vibration of hide on wood, in the exact frequency of river ice cracking at dawn.

For those seeking to listen, learn, or lend support: start not with translation, but with attention. Listen closely to how water moves. Notice how light falls on birch bark. And remember that every word spoken in Ume Sami is not a relic—it is resistance, renewal, and return.

As Katarina sings in Čååk Gálldu, over the slow pulse of the goavddis: “Juohkâ ekti čååk, juohkâ ekti guovža—ice remembers the river, ice remembers the birch.”

That remembering is ongoing. And it is alive.