Until the fifth/eleventh century, Islamic inscriptions were composed in Arabic alone. There are a few exceptions, such as the continuing use of Greek and Pahlavi on coinage in early Umayyad times, but this represents simply a temporary adherence to the previous administrative practice and was dispensed with after the reforms of 'Abd al-Malik (r. 65/685–86/705) that promoted Arabic as the prime language of the bureaucracy. It was not until the fourth/tenth to the fifth/eleventh centuries that Arabic’s monopoly as a literary and administrative language was challenged by the emergence of New Persian and Turkish. However, these new literary languages did not immediately affect epigraphic practices. The Samanid dynasty of Transoxiana, for instance, is generally credited with the promotion of New Persian, with the first use of Persian as a language of bureaucracy, and the Samanid amirs’ commissioning of Persian versions of Arabic classics, the History and Qur'an commentary of al-Ṭabarī. Yet the overwhelming majority of surviving inscribed ceramics and coins that were produced under Samanid rule continued to use Arabic; few Samanid monumental inscriptions have come down to us, but there is little evidence that the picture was substantially different there. While a handful of Persian inscribed objects may date to the fourth/tenth or even third/ninth century, and the odd Persian word may be found in Arabic inscriptions, there is a striking contrast between vernacular literary and epigraphic practices. Similarly, while Turkish literary texts, Qur'an translations and administrative documents have come down to us from the Turkish Qarakhanid dynasty, who succeeded the Samanids in many of their territories, there is little trace of any use of Turkish in their inscriptions. Instead, it is on Qarakhanid monuments that we have some of the earliest attested uses of Persian in epigraphy, while other early examples are found among the Qarakhanids’ Turkish contemporaries, the Ghaznavids. Aside from a small number of exceptions, it was not until the late seventh/thirteenth century that Turkish at all regularly appears on inscriptions in the Islamic world, and even then its employment was restricted to certain specific regional and cultural environments.

The emergence of new epigraphic languages cannot thus be simply explained with reference to broader processes of vernacularisation. Language choice in epigraphy differed from that of literary and administrative practice, and inscriptions in the vernacular only appear long after they are well established as literary languages.