Abstract
In this paper, I will examine diacritics in order to show that the graphemics of a language is in principle dependent on its phonological structure. I distinguish three types of diacritics: (1) Those that represent distinctive features, i.e., that have a consistent function in the specific writing system. Examples include the dieresis in German indicating vowel fronting, and the “caron” or háček in Czech indicating palatalisation; (2) those that are used to distinguish phonemes, usually consonants, but in an inconsistent manner; this is a very flexible type of diacritic used in an inconsistent manner such as in the adoption of a “foreign” writing system and its adaptation in order to represent the phonemes of the adopting language, e.g., not only dots added to Arabic rasms to distinguish consonants in Arabic, but also those added to represent non-Arabic consonants that occur in languages such as Persian, Urdu, and Sindhi; (3) Those that indicate vowels in abjad writing systems such as Arabic and Hebrew.
Notes
1 My thanks to an anonymous reviewer for pointing this out to me.
2 This is avoided in the Perso-Arabic scripts for religious reasons (Wellisch, Citation1978, p.113). The Qur'an – in Arabic and in the Arabic script—is believed to be God-given, brought down from heaven to Muhammed via Gabriel (Ibrahim, Citation1983).
3 See, for example, the grapheme for /ɖh/ in Sindhi (below in “The Perso-Arabic script in Indian languages: Urdu and Sindhi”).
4 Here, I am referring to the dots that are placed above or below the rasm. The lack of diacritics representing vowels, on the other hand, may lead not only to phonological ambiguity, but also to morpho-syntactic ambiguity as in “the recurring ambiguity of active vs. passive vs. imperative forms” (Attia, Citation2008, p. 31).
5 That is the rasm.
6 As has been suggested by one of the reviewers.
7 This is a synchronic analysis. Historically, the plosives preceded the fricatives, and then spirantisation took place (Tamar Zewi, personal communication).