What Literacy Instruction in Multilingual Settings Actually Means

Literacy instruction in multilingual settings is the deliberate design of reading and writing acquisition across two or more languages of schooling, where the language of instruction, the script, the curriculum sequence, and the assessment regime are all policy variables rather than neutral background. In the post-Soviet space, this is not an abstract pedagogical question. It is a budget line, a textbook print run, a teacher’s stavka load, and a gradebook entry. The core dilemma is simple to state and hard to solve: a child who enters grade one speaking a home language different from the language of instruction must learn to decode, comprehend, and produce academic text in a language they may not yet control orally, while the state simultaneously measures them against national assessment benchmarks.

This article traces that dilemma instrument by instrument — from ministry order to classroom — across Kazakhstan, Uzbekistan, Kyrgyzstan, Tajikistan, Georgia, Armenia, Azerbaijan, and Moldova. It treats teachers as professionals negotiating real constraints, not as obstacles. And it refuses both the “education is broken” lede and the utopian fix.

Why the Language of Instruction Is a Fiscal Instrument, Not Just a Pedagogical One

When a ministry designates a language of instruction, it simultaneously commits to a textbook pipeline, a teacher-training pipeline, and an assessment pipeline in that language. In Kazakhstan, the 2019 On the Status of the Teacher law and subsequent amendments to the Law on Education formalized a transition toward trilingual instruction (Kazakh, Russian, English) in secondary grades. The rollout was phased by region and subject, not by a single national switch. The practical consequence: a physics teacher in a Kazakh-medium school may be required to teach in Russian by a specified grade band, which means retraining hours measured in contact hours, not in workshops attended.

Uzbekistan’s 2017 Strategy of Actions and the 2020 Law on Education reoriented the system toward Uzbek-medium instruction with expanded Russian and English provision. The Ministry of Preschool and School Education has since published per-pupil textbook norms; in 2023, the stated target was one textbook per pupil per subject, with reprints budgeted in sums. Whether that target is met in Karakalpakstan or Surkhandarya is an empirical question, not a rhetorical one.

In Kyrgyzstan, the 2012 Law on Education and the 2014 Concept of Education Development left language-of-instruction choice largely to school founders, which means a village school may operate in Kyrgyz, Russian, or Uzbek depending on local demography and parental petition. The fiscal consequence is that small schools carry disproportionate per-pupil costs for textbook and teacher provision in minority languages.

Script Transitions as Literacy Policy

Script is not cosmetic. In Azerbaijan, the 2001 transition from Cyrillic to Latin script in schools was completed over a decade, but the literacy consequences persist in adult cohorts and in the availability of graded readers. Tajikistan retained Cyrillic; Uzbekistan completed its Latin-script transition in 1993 with a phased implementation that left many adults literate in Cyrillic but not Latin. For a grade-one teacher, this means that a child’s home literacy environment may be in a script the school does not use, which changes the decoding sequence and the time allocated to letter-sound correspondence.

Georgia’s 2011 Law on Education and subsequent amendments established Georgian as the primary language of instruction, with minority-language schools in Kvemo Kartli and Samtskhe-Javakheti operating in Azerbaijani and Armenian respectively. The 2015 Zurab Zhvania School of Public Administration and the Teacher Professional Development Center run Georgian-language courses for minority-school teachers, measured in contact hours. The literacy question is not whether Georgian is taught, but whether the transition from home language to Georgian-medium academic text happens early enough to prevent comprehension gaps by grade four.

Assessment Regimes and the Multilingual Learner

National assessments are the point where language-of-instruction policy becomes a number in a gradebook. In Kazakhstan, the External Assessment of Educational Achievements (VOUD) and the Unified National Testing (UNT) are administered in Kazakh or Russian, with the choice made by the student. A student from a Uzbek-medium family in Turkistan region may sit the UNT in Kazakh, which is their second or third language. The assessment does not measure literacy in the home language; it measures academic literacy in the language of the test.

Uzbekistan’s State Testing Center administers university entrance exams in Uzbek and Russian. The 2021 Presidential Decree on Education introduced new assessment formats, but the language-of-test question remains. In Kyrgyzstan, the National Testing Center (ORT) is administered in Kyrgyz and Russian; a student from an Uzbek-medium school in Osh may sit the ORT in Kyrgyz, which is not their first language.

This is not an argument against assessment. It is an argument for disaggregating assessment data by language of instruction and home language, which most post-Soviet systems do not routinely publish. Without that disaggregation, a literacy gap looks like a school-quality gap, and the policy response targets the wrong variable.

Teacher Labor Rules and the Literacy Load

A teacher’s stavka is a workload unit, typically 18 to 20 contact hours per week in most post-Soviet systems. In a multilingual classroom, the same stavka may include students at different oral proficiency levels in the language of instruction. The teacher is expected to differentiate, but the differentiation time is not budgeted. In Armenia, the 2010 Law on Education and the 2014 Teacher Certification procedures tie salary increments to certification, not to multilingual classroom load. In Moldova, the 2014 Education Code and the 2018 Teacher Salary Reform introduced per-pupil funding formulas, but the formula does not weight for language-of-instruction diversity.

The result is a predictable pattern: teachers in multilingual schools absorb the literacy load through unpaid time, or they teach to the middle and the students at the margins — often the ones who need the most support — fall further behind. This is not a teacher failure. It is a workload-design failure.

What Works, What Doesn’t, and What Is Untested

There is no single literacy instruction model that has been rigorously evaluated across all eight systems. What exists is a set of partial interventions with different evidence bases.

Mother-tongue-based multilingual education (MTB-MLE) has the strongest evidence base globally, but in the post-Soviet space it is politically and fiscally difficult. Tajikistan’s 2010 Law on Education allows instruction in Tajik and, in principle, in minority languages, but the textbook pipeline for minority languages is thin. Uzbekistan’s 2020 law permits instruction in Karakalpak and other minority languages in areas of compact settlement, but the assessment regime remains Uzbek- or Russian-medium.

Structured phonics in the language of instruction is the default in most early-grade classrooms, but it assumes oral proficiency. In a classroom where half the students do not speak the language of instruction at home, phonics without oral-language support produces decoding without comprehension.

Bilingual and transitional programs exist in Georgia’s minority schools and in some Kazakh schools, but they are not systematically evaluated. The 2018 PISA results for Georgia show a persistent gap between Georgian-medium and minority-language schools, but the data do not isolate literacy instruction from other variables.

Textbook Runs and Per-Pupil Budgets

Textbook provision is a material precondition for literacy instruction. In Kazakhstan, the 2020 Law on Education amendments set per-pupil textbook funding norms, but the actual per-pupil allocation varies by region and by language of instruction. In Uzbekistan, the 2023 budget allocated sums for textbook reprints, but the distribution to schools in remote districts is a logistical question. In Kyrgyzstan, the 2019 Education Development Strategy set targets for textbook provision, but the per-pupil budget in som is not publicly disaggregated by language.

Without textbook runs measured in copies per pupil per subject per language, literacy instruction is a plan, not a practice.

FAQ: Literacy Instruction in Multilingual Post-Soviet Classrooms

What is the biggest literacy instruction challenge in multilingual post-Soviet schools?

The mismatch between the language of instruction and the language of assessment. A student may learn to read in one language and be tested in another, which measures language proficiency rather than literacy skill.

How do teacher workload rules affect literacy instruction?

Teacher stavka hours are typically 18–20 contact hours per week, but multilingual classrooms require additional preparation time that is not budgeted. This means teachers either absorb the load unpaid or teach to the middle, leaving struggling readers behind.

Are mother-tongue-based programs feasible in this region?

They are legally permitted in several countries, including Uzbekistan and Tajikistan, but the textbook pipeline and assessment regime are not aligned with them. Feasibility depends on whether the state is willing to fund textbooks and assessments in minority languages, which is a political and fiscal decision, not a pedagogical one.

What data would help improve literacy instruction?

Disaggregated assessment data by language of instruction and home language, per-pupil textbook provision by language, and teacher workload data that includes multilingual classroom load. Most post-Soviet systems do not routinely publish these.

Practitioner-Facing Provocation

If you are a teacher in a multilingual classroom, the next time you are handed a textbook in the language of instruction and a class list that includes students who do not speak that language at home, ask your administration for the per-pupil textbook allocation in your language and the contact-hour budget for differentiation. If those numbers do not exist, you have found the policy gap. The question is not whether you can close it alone — you cannot — but whether you can name it precisely enough that someone with a budget line can act on it.