The Chakhar (Mongolian: ᠴᠠᠬᠠᠷ Čaqar; Chinese: 察哈爾; pinyin: Cháhār) dialect is a variety of Mongolian spoken the central region of Inner Mongolia. It is phonologically close to Khalkha and is the basis for the standard pronunciation of Mongolian in Inner Mongolia.
Location and classification 
There are three different definitions of the word Chakhar. First, there is Chakhar proper, spoken in the banners Shuluun Köke, Shuluun Köböötü Chagaan, Höböötü Shira, Taipusi, Chakhar Right Wing Rear, Chakhar Right Wing Middle, Chakhar Right Wing Front, Shangdu, Dolonnur and Huade with a number of approximately 100.000 speakers. In a broader definition, the Chakhar group contains the varieties Chakhar proper, Urat, Darkhan, Muumingan, Dörben Küüket, Keshigten of Ulanqab. In a very broad and controversial definition, it also contains the dialects of Xilin Gol such as Üjümchin, Sönit, Abaga, Shilinhot. The Southern Mongolian normative pronunciation is based on the variety of Chakhar proper as spoken in the Shuluun Köke banner.
Excluding the phonology of recent loanwords, Chakhar has the pharyngeal vowel phonemes /ɑ/, /ɪ/, /ɔ/, /ʊ/ and the non-pharyngeal vowel phonemes /ə/, /i/, /o/, /u/ that adhere to vowel harmony. All have long counterparts and some diphthongs exist as well. /ɪ/ has phoneme status only due to its occurrence as word-initial vowel in words like ɪlɑ̆x ‘to win’ (vs. ɑlɑ̆x ‘to kill’), thus /i/ (<*i) does occur in pharyngeal words as well. Through lexical diffusion, /i/ <*e is to be observed in some words such as /in/ < *ene ‘this’, rather than in /ələ/ ‘kite (bird)’. However, long monophthong vowels also include /e/ < *ei. The maximal syllable structure is CVCC. In word-final position, non-phonemic vowels often appear after aspirated and sometimes after unaspirated consonants. They are more frequent in male speech and almost totally disappear in compounds. The consonant phonemes (again excluding loanwords) are
/n/, /nʲ/, /ŋ/, /b/ /bʲ/, /p/, /x/, /xʲ/, /ɡ/, /ɡʲ/, /m/, /mʲ/, /l/, /lʲ/, /s/, /ʃ/, /t/, /tʲ/, /d/, /dʲ/, /t͡ʃ/, /d͡ʒ/, /j/, /r/, /rʲ/
while [k] (<*k) and [w] (<*p) that occur in loanwords and native words alike are only allophones of /x/ and /b/ in native words. Palatalized vowels have phoneme status only in pharyngeal words.
Word classes and morphology 
The case system of Chakhar has the same number of morphemes as Khalkha with approximately the same forms. There is a peculiar Allative case suffix, -ʊd/-ud, that has developed from *ödö (Mongolian script <ödege>) 'upwards' and that seems to be a free allomorph of the common -rʊ/-ru. The reflexive-possessive suffixes retain their final -ŋ (thus -ɑŋ<*-ban etc., while Khalkha has -ɑ).
Large numbers are counted according to the Chinese counting system in powers of 10.000. Collective numerals can be combined with approximative numeral suffixes. So while ɑrwɑd 'about ten' and ɑrwʊl 'as a group of ten' a common in Mongolian, ɑrwɑdʊl 'as a group of about ten' seems to be peculiar to Chakhar.
The pronominal system is much like that of Khalkha. The colloquial form of the 1. person singular accusative (in which the idiosyncratic accusative stem is replaced) can be nadï instead of nadïɡ, and the alternation of i ~ ig does occur with other pronominal stems as well. This does not lead to confusion as the genitive is formed with mid-opened instead of closed front vowels, e.g. the 2. person singular genitive honorific is [tanɛ] in Chakhar and usually [tʰanɪ] in Khalkha. The 3. person stems don't employ any oblique stems. The 1. person plural exclusive man- has an almost complete case paradigm only excluding the nominative, while at least in written Khalkha anything but the genitive form <manaj> is rare.
Chakhar has approximately the same participles as Khalkha, but -mar expresses potentiality, not desire, and consequently -xar functions as its free allomorph. On the other hand, there are some distinctive converbs such as -ba (from Chinese 吧 ba) 'if' and -ja (from 也 yè) 'although' which seem to be allomorphs of the suffixes -val and -wc of common Mongolian origin. The finite suffix -la might have acquired converbal status. Finally, -xlar ('if ... then ...') has turned into -xnar, and the form -man ~ -manjï̆n 'only if', which is absent in Khalkha, sometimes occurs. Chakhar has the same core declarative finite forms as Khalkha, but in addition -xui and -lgui to indicate strong probability.
- Sečenbaγatur 2003: 6
- Janhunen 2003: 179-180
- Janhunen 2003: 179, Sečenbaγatur 2003: 7; judging from a map provided by Sečenbaγatur et al. 2005: 565, this extension relates to administrative areas in the following way: next to Xilin Gol and Ulanqab, Chakhar in its broadest extension would be spoken in Bayan Nur, Baotou, the northern part of the Hohhot area and the very western part of the Ulanhad.
- Sečenbaγatur et al. 2005: 85
- The analysis used in this article follows Svantesson et al. 2005: 22-25 in assuming that short vowels in non-initial syllables are non-phonemic.
- Sečenbaγatur et al. 2005: 209-213
- Sečenbaγatur et al. 2005: 224
- Köke and Sodubaγatur 1996: 10-14. They assume voicedness and not aspiration to be distinctive and give the following rations: males after voiceless consonants: 94% (n=384; “n” is the number of appropriate words uttered by speakers during a phonetic test), after voiced consonants: 53% (n=371), females after voiceless consonants: 87% (n=405), after voiced consonants: 38% (n=367). In contrast: only 7% of words in compound uttered by informants of both sexes contained such vowels. The authors try to link this with historical vowels, but as they don’t take into account that aspirated vowels were historically disallowed in word-final position, this line of argument is not very convincing.
- MKBAKKSKND 2003: 30
- Norčin 2001: 148
- MKBAKKSKND 2003: 31-35
- Sečenbaγatur 2003: 43-44, 49-50
- Sečenbaγatur 2003: 70, 75
- Sečenbaγatur et al. 2005: 390
- Sečenbaγatur 2003: 92-95, 100 for Chakhar; see Poppe 1951: 71-72 for Khalkha. He also gives ~tanä and a full paradigm (excluding nominative) for man-, but at least the full exclusive paradigm is exceedingly rare at least in writing and not reproduced in more recent (while somewhat prescriptive) grammars, e.g. Önörbajan 2004: 202-217. The exceedingly frequent form [natig] (instead of <namajg> doesn't seem to be mentioned anywhere.
- Sečenbaγatur 2003: 126-127
- Sečenbaγatur 2003: 130-131
- Sečenbaγatur 2003: 134 asserts this, but doesn't provide conclusive evidence. However, compare Ashimura 2002 for Jarud where such a process has taken place
- Sečenbaγatur 2003: 132-133
- Sečenbaγatur 2003: 140-141; forms like <-lgüj dee> are quite frequent in Khalkha, but it is difficult to assess their exact grammatical status without some research.
- Sečenbaγatur 2003: 16-18
- Ashimura, Takashi (2002): Mongorugo jarōto gengo no -lɛː no yōhō ni tsuite. In: Tōkyō daigaku gengogaku ronshū 21: 147-200.
- Janhunen, Juha (2003): Mongol dialects. In: Juha Janhunen (ed.): The Mongolic languages. London: Routledge: 177–191.
- Köke and Sodubaγatur (1996): Čaqar aman ayalγun-u üge-yin ečüs-ün boγuni egesig-ün tuqai. In: Öbür mongγul-un yeke surγaγuli 1996/3: 9-20.
- Mongγul kelen-ü barimǰiy-a abiyan-u kiri kem-i silγaqu kötülbüri nayiraγulqu doγuyilang (2003): Mongγul kelen-ü barimǰiy-a abiyan-u kiri kem-i silγaqu kötülbüri. Kökeqota: Öbür mongγul-un arad-un keblel-ün qoriy-a.
- Norčin (2001): Barim/ǰiy-a abiy-a - Čaqar aman ayalγu. Kökeqota: öbür mongγul-un arad-un keblel-ün qoriy-a.
- Önörbajan, C. (2004): Orčin cagijn mongol helnij üg züj. Ulaanbaatar: Mongol ulsyn bolovsrolyn ih surguul'.
- Poppe, Nicholaus (1951): Khalkha-mongolische Grammatik. Wiesbaden: Franz Steiner.
- [Sečenbaγatur] Sechenbaatar (2003): The Chakhar dialect of Mongol - A morphological description. Helsinki: Finno-Ugrian society.
- Sečenbaγatur et al. (2005): Mongγul kelen-ü nutuγ-un ayalγun-u sinǰilel-ün uduridqal. Kökeqota: Öbür mongγul-un arad-un keblel-ün qoriy-a.
- Svantesson, Jan-Olof, Anna Tsendina, Anastasia Karlsson, Vivan Franzén (2005): The Phonology of Mongolian. New York: Oxford University Press.