Writing systems
Script Reforms: When a Language Changes Its Alphabet
Why states switch scripts, what it costs a generation of readers, and the cases still under way.
The alphabet is the foundation stone of a nation's written language, but it can change. And for many countries, the decision to enact a script reform comes with costs as well as benefits. One of the most notable examples of alphabet reform was when Turkey in 1928, under the leadership of Mustafa Kemal Atatürk, eliminated the Arabic-script Ottoman alphabet in favor of today's Latin-script Turkish alphabet.
Alphabet reform is typically sold as a modernization project, or a step towards nation-building. Turkey's move to change its script, as outlined in the Law on the Adoption and Application of Turkish Letters, passed on 1 November 1928, was no exception. But there is also a heavy administrative aspect: reform can be the tool of rapid administrative change the state.
The Turkish alphabet law made the use of Latin letters compulsory in public communications and the education system. A specific language commission was set up in Turkey, too, to adjust the Latin script to the phonetic needs of Turkish, proving that reform is about standardizing more than just the script itself. Turkey's current alphabet still reflects this move, as it was established in the 1928 reform period and contains 29 letters.
While alphabet reform has a political driver, the burden falls to the population, who can lose access to texts written in the old script. It can take generations for the transition to complete. Uzbekistan is a live example. Even though Uzbekistan has been switching from Cyrillic to Latin since 1993, it has not yet completed the transition.
On July 7, 2026, Uzbekistan's Legislative Chamber approved a bill to further tweak the Latin-based Uzbek alphabet. The bill introduces new conditions, adding single letters instead of the existing four digraphs: Ş for Sh, Ç for Ch, Ö for Oʻ, and Ğ for Gʻ. The draft law itself was not shared for public discussion, with only its 2021 edition available online.
The Uzbekistan case underlines how a script change is more than a single act. The alphabet reform in Uzbekistan, and the continuous tweaks it went through, show that the alphabet isn't always constant or straightforward. Uzbekistan's experience shows that alphabet reform is open to change, with changes in the reform itself.
Alphabet reform is a technological change with both educational and administrative aspects, but it has a significant impact on the responsibility of archives and libraries. Alphabet reform affects what can be stored in archives, how information is cataloged and accessed in libraries, and the opportunities for archivists and historians.
Longer processes to implement a reform are increasingly common. Uzbekistan, for example, unveiled its plans to transition its alphabet in 1993, and initially intended to complete the process by 2000. But more than 10 years on, it hasn't yet completed the transition. It shows post-Soviet republics' alphabet reforms can drag on for a long time, with proceedings delayed or efforts frustrated by each other.
Alphabet reform rarely completes on one go: one thing leads to another, affecting all aspects of historical documentation and storage. Even decades after the reform, hearing the language in its original script becomes harder.
Why states keep revisiting this challenge is harder to pin down. Turkey's reforms were preceded by big nation-building efforts, while Uzbekistan pursued gradual, top-down change. Reform may be inevitable for some governments -- it can support and symbolize identity shifts, or provide a new identity for a transitioning state.
But this shows that script itself can't be entirely detached from political views or historical context. Script reform is as much an expression of a state's identity, as it is a way to govern.
From the cost of getting a population to learn to read a new script, to the economic cost of preserving and transitioning archives, script reform is not a simple, or quick decision. In Uzbekistan, it remains unfinished.
Where this sits in the Register
Written by the Linguasphere desk. Classification data referred to here comes from the Linguasphere Register — see sources and attribution.