Last Updated: 2020-06-29

COMPROMISED: Conflation between /æ/ and /ɛ/; slight ambiguity between vowel nasalization and underlying /n/; unreliable marking of attested phonemes brought in through loanwords

Background

Language Family: Indo-European / Indo-Iranian / Indo-Aryan / Central zone / Western Hindi / Hindustani

Phonology

Consonants

  • All consonants can appear as geminates word-medially (Ohala 1999, 101).
  • Although /f/, /ʃ/, and /z/ only occur in loanwords, they are well established phonemes within the language (Ohala 1999, 101).
  • The phonemic inventory below is largely based off Ohala (1999) (p. 100).
    • However, possibly due to dialectal variation, the analysis done by Koul (2008) is somewhat different (pp. 12, 19). He argues:
      • for an alveolar trill rather than a flap
      • that /s/ and /z/ are the only alveolar consonants
      • for /ɳ/
      • for /x/
      • that velar and palatal nasals are allophones of /n/, assimilating in place to following consonants
    • Given that Ohala (1983) treats nasals, excluding /m/ and /n/, as allophones, I have opted to do the same (p. 2).
    • Everything else is kept consistent with Ohala (1999) (p. 100).
Place of Articulation
Manner of Articulation Labial Dental Alveolar Postalveolar Retroflex Palatal Velar Glottal
Stops (plain) p b t̪ d̪ ʈ ɖ k ɡ
Stops (aspirated) pʰ bʱ t̪ʰ d̪ʱ ʈʰ ɖʱ kʰ ɡʱ
Affricates (plain) tʃ dʒ
Affricates (aspirated) tʃʰ dʒʱ
Fricatives f s z ʃ h
Nasals m n
Flaps (plain) ɾ ɽ
Flaps (aspirated) ɽʱ
Approximants ʋ l j
Note: For phonemes that share a cell, those on the left are voiceless and those on the right are voiced.

Vowels

  • The vowel /æ/ only appears in English loanwords and is not orthographically distinct (i.e. /æ/ and /ɛ/ are represented by the same character); I only account for /ɛ/ in the rules (Ohala 1999, 101). However, given that English loanwords are fairly prominent in Hindi, such that some sources include the phoneme in the inventory (e.g. Ohala 1999), I’ve marked this language as compromised.
  • Vowel nasalization is contrastive (Ohala 1999, 102).
    • Nasal /ɐ/ and /ɛ/ aren’t reflected in the Crúbadán corpus.
  • Vowel length is contrastive for /i/, /a/, and /u/ (Koul 2008, 26).
    • /i/, /u/, and /ɑ/ are considered long, whereas /ɪ/, /ʊ/, and /ɐ/ are short.
  • Adjacent vowels are not realized as complex nuclei (Ohala 1999, 101). This applies to /ɐi/ and /ɐu/, commonly interpreted as diphthongs in related languages.
  • Although several sources represent the inherent vowel as /ə/ (e.g. Kachru 2006, xix), I have chosen to use /ɐ/, as other sources (e.g. Ohala 1983, 1) infer that it is actually lower than the mid central position. Even Ohala (1999), who uses /ə/, states that /ɐ/ would be more appropriate (p. 102).
Front Central Back
High i u
Near-High ɪ ʊ
High-Mid e o
Low-Mid ɛ ɔ
Near-Low ɐ
Low æ ɑ

Alphabet

Consonant Grapheme Phoneme Comment
ब /bɐ/
भ /bʱɐ/
द /dɐ/
ध /dʱɐ/
ज /dʒɐ/
ज़ /zɐ/ indicated orthographically with an adjoining nutka diacritic (see Diacritic section below)
झ /dʒʱɐ/
ड /ɖɐ/
ढ /ɖʱɐ/
ड़ /ɽɐ/ indicated orthographically with an adjoining nutka diacritic (see Diacritic section below)
ढ़ /ɽʱɐ/ indicated orthographically with an adjoining nutka diacritic (see Diacritic section below)
ग /ɡɐ/
घ /ɡʱɐ/
ह /hɐ/
य /jɐ/
क /kɐ/
ख /kʰɐ/
ल /lɐ/
म /mɐ/
न; ण; ञ; ङ /nɐ/
प /pɐ/
फ /pʰɐ/
फ़ /fɐ/ indicated orthographically with an adjoining nutka diacritic (see Diacritic section below)
र /ɾɐ/
स /sɐ/
ष; श /ʃɐ/ ⟨ष⟩ originally referred to /ʂ/, but the contrast has merged (Mishra and Bali 2011, 1393)
त /t̪ɐ/
थ /t̪ʰɐ/
च /tʃɐ/
छ /tʃʰɐ/
ट /ʈɐ/
ठ /ʈʰɐ/
व /ʋɐ/
Independent Character
आ /ɑ/
ए /e/
ऐ /ɛ/
अ /ɐ/
ई /i/
इ /ɪ/
ओ /o/
औ /ɔ/
ऊ /u/
उ /ʊ/
ऋ /ɾɪ/
Diacritic
ा /ɑ/
े /e/
ै /ɛ/
ी /i/
ि /ɪ/
ो /o/
ौ /ɔ/
ू /u/
ु /ʊ/
ृ /ɾɪ/
ँ; ं called an anunasik and an anusvara, respectively, these diacritics indicate nasalization of the preceding vowel (Kellogg 2008, 6). The anusvara is generally used more so than the anunasik (Language 2007, 4).
् called a virama, this diacritic deletes the inherent vowel (Shapiro 2014, 24)
॰; ॱ; ।; ॥ punctuation - no phonemic realization (Shapiro 2014, 26)
़ called a nukta, this diacritic adjoins with consonants, indicating attested phonemes of loanword origin (e.g. /f/) (Lehal and Saini 2010, 3). This diacritic is not always written, compromising the language.

Syllable Structure

Misc. Rules

References

Agnihotri, Rama Kant. 2007. Hindi: An Essential Grammar. Taylor & Francis Ltd.

Kachru, Yamuna. 2006. Hindi. John Benjamins Publishing Company.

Kellogg, Samuel Henry. 2008. A Grammar of the Hindi Language: In Which Are Treated the Standard Hindi, Braj, and the Eastern Hindi of the Ramayan of Tulsi Das (1876). Kessinger Publishing, LLC.

Lehal, Gurpreet Singh, and Tejinder Singh Saini. 2010. “A Hindi to Urdu Transliteration System.” In Proceedings of Icon-2010: 8th International Conference on Natural Language Processing.

Masica, Colin P. 1991. The Indo-Aryan Languages. Edited by R. Lass B. Comrie C. J. Fillmore. Cambridge University Press.

Mishra, Diwakar, and Kalika Bali. 2011. “A Comparative Phonological Study of the Dialects of Hindi.” In ICPhS.

———. 1999. “Hindi.” Illustrations of the IPA.