Native toKenya, Ethiopia
EthnicityTurkana people
Native speakers
990,000 (2009 census)[1]
  • Northern Turkana
  • Southern Turkana
Language codes
ISO 639-3
tuv – Turkana

Turkana /tɜːrˈkɑːnə/[2] is the language of the Turkana people of Kenya and Ethiopia. It is spoken in northwestern Kenya, primarily in Turkana County, which lies west of Lake Turkana. It is one of the Eastern Nilotic languages, and is closely related to Karamojong, Jie and Teso of Uganda, to Toposa spoken in the extreme southeast of South Sudan, and to Nyangatom in the South Sudan/Ethiopia Omo valley borderland; these languages together form the cluster of Ateker Languages.

The collective group name for these related peoples is Ateker.



Labial Alveolar Palatal Velar
Nasal m n ɲ ŋ
voiceless p t k
voiced b d ɡ
Fricative s
Lateral l
Trill r
Approximant j w


There is a phonemic distinction with voiceless vowels, which only occur word-finally, and which are only realized as voiceless before a pause:

Front Central Back
voiced voiceless voiced voiceless voiced voiceless
Close i u
Near-close ɪ ɪ̥ ʊ ʊ̥
Close-mid e o
Open-mid ɛ ɛ̥ ɔ ɔ̥
Open a

Turkana features advanced tongue root vowel harmony.[3] The vowels /i e o u/ and their voiceless counterparts are produced with an advanced tongue root, while ɛ a ɔ ʊ/ and their voiceless counterparts are produced with a retracted tongue root. The advanced tongue root vowels are usually somewhat breathy in terms of voice. In most circumstances, vowels in any given word must either be all advanced or all retracted in their tongue root position. An exception is made for vowels that come after /a/, which can be either advanced or retracted (while vowels coming before /a/ must be retracted). In parallel to this, vowels following the semivowels /j/ and /w/ can be either advanced or retracted, but vowels preceding them must be advanced. However, the semivowels and /a/ do not affect each other: either may occur before the other, despite conflicting in their tongue root position.

Vowel harmony is usually controlled by the root of a word, so that the vowels of other morphemes assimilate to the root vowels' tongue root position. However, some suffixes are "strong" and instead assimilate the root along with any preceding suffixes. Prefixes are always weak and do not control other vowels. The vowels paired in such assimilations are /i/ vs. /ɪ/, /e/ vs. /ɛ/, /o/ (from earlier /ə/) vs. /a/, /o/ vs. /ɔ/, and /u/ vs. /ʊ/; either element of each pair will turn into the other to match the tongue root position of the controlling morpheme. Vowel harmony does not cross word boundaries, and a phonological word can be defined as a unit across which harmony operates. Vowel harmony is also blocked at the boundary between roots in a compound.

Long vowels occur phonetically, but are best analyzed as sequences of short vowels rather than phonemes in their own right. In roots, /ɔ/ and /ɛ/ may be realized as [wa] and [ja], respectively. High front vowels are deleted between a palatal consonant and another vowel. Voiceless vowels before a pause are lost after glides and nasals (after de-voicing them). Nonhigh voiceless vowels before a pause are furthermore often lost in general.


Turkana has two tonemes, high and low, and one or the other is carried by every vowel. Most syllables carry a high tone, so that low tone is more marked. Some tones are "floating" and not carried by a syllable at all. Minimal pairs for tone are rare, and so it is not important in distinguishing words, but it is used to distinguish verb tenses and noun cases, so that it is important in terms of grammar.

Morphology and syntax

Turkana is a verb-initial language with both verb–subject–object (VSO) and verb–object–subject (VOS) as basic, unmarked word orders.[3] The single objects of transitive verbs and both objects of ditransitive verbs are unmarked, while the subject is morphologically marked by a change in tone when present after the verb, and is often omitted entirely. Subjects of intransitive and transitive verbs receive the same case marking. Thus, in terms of morphosyntactic alignment, Turkana is both a marked-nominative language and a double-object language. With ditransitive verbs, the recipient/indirect object is required to be animate, and it always precedes the theme/direct object. Neither object can be promoted to be the subject of a basic sentence, so that Turkana has no real passive construction.

Turkana features six cases: a nominative, an absolute, a genitive, an instrumental, a locative, and a vocative.[3] This makes it typologically unusual as one of the only verb-initial languages attested to have more than two or three cases.


English Turkana
singular form
plural form
face ereet ngiReetin
body akwaan ngaWat
clothes eworu ngiWorui
food akimuj ngaMuja
tobacco etaba ngiTab
goat akine ngaKinei
cattle aite ngaAtuk
donkey esikiria ngiSikiria
camel ekaal ngiKaala
water ngakipi ngaKipi



  1. ^ Turkana at Ethnologue (18th ed., 2015) (subscription required)
  2. ^ Laurie Bauer, 2007, The Linguistics Student’s Handbook, Edinburgh
  3. ^ a b c d Dimmendaal 1983.