Swahili Language - Orthography

Orthography

Swahili is currently written in a slightly defective alphabet using the Latin script; the defectiveness comes in not distinguishing aspirated consonants, though those are not distinguished in all dialects. (These were, however, distinguished as kh etc. in the old German colonial Latin alphabet.) There are two digraphs for native sounds, ch and sh; c is not used apart from unassimilated English loans and occasionally as a substitute for k in advertisements. There are in addition several digraphs for Arabic sounds which are not distinguished in pronunciation outside of traditional Swahili areas.

The language had previously been written in the Arabic script. Unlike adaptations of the Arabic script for other languages, relatively little accommodation was made for Swahili. There were also differences in orthographic conventions between cities, authors, and over the centuries, some quite precise, but others defective enough to cause difficulties with intelligibility.

Vowel diacritics were generally written, effectively making the Swahili-Arabic script an abugida. /e/ and /i/, /o/ and /u/ were often conflated, but in some orthographies /e/ was distinguished from /i/ by rotating the kasra 90°, and /o/ from /u/ by writing the damma backwards.

Several Swahili consonants do not have equivalents in Arabic, and for these often no special letters were created, as they were for example in Persian and Urdu. Instead, the closest Arabic sound is substituted. Not only does this mean that one letter often stands for more than one sound, but writers made different choices as to which consonant to substitute. Some of the equivalents between Arabic Swahili and Roman Swahili are,

This was the general situation, but conventions from Urdu were adopted by some authors; for example, to distinguish aspiration and /p/ from /b/: پھا /pʰaa/ 'gazelle', پا /paa/ 'roof'. Although not found in Standard Swahili today, there is a distinction between dental and alveolar consonants in some dialects, and this is reflected in some orthographies, for example in كُٹَ -kuta 'to meet' vs. كُتَ -kut̠a 'to be satisfied'. A k with the dots of y, ڱ, was used for ch in some conventions; this ky is historically and even contemporaneously a more accurate transcription than Roman ch. In Mombasa, it was common to use the Arabic emphatics for Cw, for example in صِصِ swiswi (standard sisi) 'we' and كِطَ kit̠wa (standard kichwa) 'head'.

Word division differs from Roman norms. Particles such as ya, na, si, kwa, ni are joined to the following noun, and possessives such as yangu and yako are joined to the preceding noun, but verbs are written as two words, with the subject and tense–aspect–mood morphemes separated from the object and root, as in aliye niambia "he who asked me".