𐱁𐰑

There seems to have been no problem to introduce onset /š/ with borrowings, in view of the appearance of the title šad already in the runiform inscriptions and the widespread use of šïmnu / šamnu to refer to the ‚devil‘ (also among Buddhists; there = māra). Cf. also the term šïk for a measure of capacity, borrowed from Chinese. Note, though, that šad appears as čad in Taryat N4 (twice) and Tes W6, two runiform inscriptions of the Uygur steppe empire. šato ‚ladder‘ is also a borrowing and was, in any case, adapted to Turkic in the form of č(a)to in an early text (M I nr. 1 III v 8); Wordlist has it with this meaning and o in the second syllable. If we do find šato itself well attested in classical and later texts (including Qarakhanid), the reason may be that the Turks subsequently got used to having š in onset position. The replacement of an onset foreign sibilant by /č/ appears to have been an areal phenomenon: čïxšaput / čaxšapat (etc.) ‘commandment’ is not a direct loan from Skt. śikṣāpada but comes over Sogdian čγš’pδ. The history of šï: ‘moist(ure)’ (found e.g. in BuddhKat 4) may have been a bit more complicated.^185
185 The modern Chinese word of this shape and meaning originally had a final consonant which should have been borrowed into Old Turkic; what we have appears to have resulted from contamination between that word and Turkic čï: ‘dew’ etc., attested (together with verbal derivates) in the DLT and in many modern languages.