Proto-Munda language

Proto-Munda is the reconstructed proto-language of the Munda languages of the Indian subcontinent. It has been reconstructed by Sidwell & Rau. According to Sidwell, the Proto-Munda language split from Proto- Austroasiatic in Indochina and arrived on the coast of present-day Odisha around 4000 – 3500 years ago. Rau & Sidwell also strongly suggest the association of proto-Munda speakers with the Neolithic Sankarjang–Golabai Archaeological Complex offshore of the Mahanadi River Delta.

Phonology

The tables below show Proto-Munda consonant and vowel phonemes reconstructed by Rau and Sidwell & Rau.

Consonants

Vowels

Retroflexes

es are a major prominent feature of the South Asian linguistic area, and all modern Munda languages possess various sets of retroflexion. In languages with high degrees of contact such as Santali, Korku, Mundari,... retroflex consonants such as plosives often appear in loanwords or have permeated into some native words, while several Sora dialects lack retroflexes altogether. In some cases, Munda stems may have acquired retroflexes naturally via phonological changes similar to those exist in Swedish, Middle Chinese, Vietnamese,... The retroflex sonorants, which reflect contact influence from Dravidian languages, however, are rare and may not be considered phonemic in Munda consonant inventories.

Word structure

The proto-Austroasiatic canonical structure *CV was completely restructured to *CV shape in proto-Munda. Sidwell & Rau suggest a scenario that intense language contacts between few Austroasiatic speakers and large number of speakers of unknown language might have occurred immediately after the proto-Munda landfall to Eastern India.

Vocabulary

Austroasiatic roots

The following Proto-Munda lexical proto-forms have been reconstructed by Sidwell & Rau. Two asterisks are given to denote the tentative, preliminary state of the proto-language reconstruction.

Gloss	Proto-Munda
belly	**laɟ
big	**məraŋ
to bite	**kaˀp
black	**kEdE
blood	**məjam
bone	**ɟaːˀŋ
to burn	**gEˀp
claw/nail	**rəmAj
cloud	**tərIˀp
cold	**raŋ
die	**gOˀj
dog	**sOˀt
to drink	uˀt, uˀk
dry	**sAr
ear	lutur, luˀt
earth/soil	**ʔOte
to eat	**ɟOm
egg	**tAˀp
eye	**maˀt
fat/grease/oil	**sunum
feather	**bəlEˀt
fire	**səŋal
fish	ka, kadO
fly	**pEr
foot	**ɟəːˀŋ
give	**ʔam
hair	**suˀk
hand	**tiːˀ
to hear/listen	**ajɔm
heart, liver	rE, ʔim
horn	**dəraŋ
I	**iɲ
to kill	**ɡOˀɟ
leaf	**Olaːˀ
to lie	**gətiˀc
long	**ɟəlƏŋ
louse	**siːˀ
man/husband, person/human	**kOrOˀ
meat/flesh	**ɟəlU
moon	harkE, aŋaj
mountain/hill	**bəru
mouth	**təmOˀt
name	**ɲUm
neck	kO, gOˀk
new	**təmI
night	**ədiˀp
nose	**muːˀ
not	**əˀt
one	**mOOˀj
rain	**gəma
red	**ɟəŋAˀt
road, path	**kOrA
root	**rEˀt
sand	**kEˀt
see	**El
sit	**kO
skin	**
sleep	**gətiˀc
smoke	**mOˀk
to speak, say	sun, gam, **kaj
to stand	tənaŋ, tƏŋgə
stone	bərƏl, sərEŋ
sun	**siŋi
tail	**pata
thigh	**buluuˀ
that	**han
this	**En
thou/you	**Am
tongue	**laːˀŋ
tooth	**gənE
tree	**ɟiːˀ
two	**baːˀr
to walk, go	**sEn
to weave	**taɲ
water	**daːˀk
woman/wife	selA, kəni
yellow	**saŋsaŋ

Proto-Munda reconstruction has since been revised and improved by Rau.

Unknown substrate terms

A number of words which have uncertain origins are also found in Zide & Zide 's reconstruction of proto-Munda cultural terms.

*saXʔl ‘mortar’
*gaŋgay ‘sorghum’
*ə-rig' ‘small millet’
*deray ‘ragi’
*kodaXj ‘horsegram’
*rVm ‘black gram’
*uXli/uXla ‘mango’
*kaj'-er/*kag'-er ‘unripe mango’
*taŋ ‘cow’
*boŋtel ‘buffalo’
Indian blackberry
turmeric
tamarind
wild date
Morphosyntax

Syntactic shift

Although the modern Munda languages show a standard head-final subject–object–verb order in unmarked phrases, most scholars believe that proto-Munda was head-first, VO like proto-Austroasiatic. The first linguist to noticed this peculiarity, Heinz-Jürgen Pinnow, found strong evidence for a proto-Munda VO order. VO order has been found in compounds, noun incorporation verbal morphology in the Sora-Gorum languages, and to a lesser extent in Gutob, Remo, Kharia, and Juang. By any given verb conjugations, the Munda verbs always show internal head-first, V-P order, with two main overall syntactic orders of transitive verbs: A-V-P and V-P-A, corresponding to Austroasiatic clausal syntaxes SVO and VOS. Most Munda compounds are also head-first and right-branching, with new loan words from Indian languages following the Indian norm of head-final and left-branching.
Remo:
Sora:
Juang:
Gorum:
Winfred Lehmann reviewed,
Pinnow proposed that proto-Munda was SVO and that was the syntax of proto-Austroasiatic, which was also highly synthetic like Munda, whereas he attributed analytic and isolating typological features in modern Mon-Khmer to language contact in the Mainland Southeast Asia linguistic area.
Donegan & Stampe argued that proto-Munda was VO but non-agglutinative like its sister languages in Southeast Asia. According to Donegan & Stampe, there are some characteristics of the Munda languages such as head-marking and polysynthesis that are so distinct and attributable to neither Dravidian and Indo-Aryan influence. They believed that Munda synthesis and SOV order were clearly not stimulated just by language contact within the South Asian linguistic area, but by internal restructuring that caused the Munda word prosody to shift its rhythmic patterns from typical Austroasiatic rising, vowel reduction, iambic stressed to falling, vowel harmony, trochaic stressed profile, thus reversed the clausal syntactic structure from VO to OV and triggered word agglutination.
For the reason why the Munda languages keep head-first order in compounds and polysynthetic morphology, Donegan & Stampe believed that words, like the verb-noun compounds, are more resistant to internal changes of rhythms and ordering than phrases. Donegan & Stampe carefully stated that their explanation for the Munda rhythm-initiated synthesis drift does not include polysynthesis, and Donegan & Stampe also termed the Munda polysynthetic morphology as 'idiomatic and morphology.' Stanley Starosta explained that, during its early formation stage when Munda was still head-first SVO, verb-noun incorporation was facilitated as seen in modern Sora, but then it did a syntactic shift to head-final SOV and added more morphology. The polysynthetic verb phrases thus had become crystalized since that time.
Donegan & Stampe's prediction of Munda synthetic shift caused by change of rhythmic holism is contested by instrumental data of individual languages, which show that Munda prosodic characteristics are not what Donegan & Stampe described.

Proto-Munda predicate

Anderson and Zide (2001, 2007)

& Norman Zide reconstructed the head-marking bound predicate of Proto-Munda with A-V-P order as following:

Slot	+4	+3	+2	+1	core	-1	-2	-3
role	SUBJ	NEG	RECIP/CAUS	DERIV	verb stem	PASS/INTR	TRANS/TNS	OBJ

Anderson & Zide, together with van Driem positioned that the Munda languages are the most morphologically conservative Austroasiatic branch. Van Driem posited that the Austroasiatic languages dispersed eastward from Northeast India to Southeast Asia instead. Michael Witzel proposed an Austroasiatic homeland further west, in the Panjab region during the Indus Valley Civilization. The current general consensus considers these hypotheses unlikely. Peter Bellwood states that "the source for the whole Austroasiatic languages is still a mystery."

Rau (2020)

Felix Rau concludes that Proto-Munda predicate structure was verb-medial SVO, though he suggests that it might have been less inflected with fewer bound elements, which may cause the Munda predicate development to become divergent later.

Slot	+6	+5	+4	+3	+2	+1	core	-1	-2	-3
role	SUBJ	MOD/ASP	NEG	RECIP	CAUS	DERIV	verb stem	ASP	/valency	OBJ
reconstruction		A ''O Vj mO	əˀt Um	*kƏl	*Oˀp	bə- tA- **A-		=lə Perf* =tə Imperf*	n MID ˀt ACT
Morphological cognates		Preverbal in Khasi-Palaungic	*ʔət =m mǝ	kər-	op pa- pʌn -p-	*bə-, tA-, *p-		las lɛʔ li'' Postverbal in Bahnaric, Vietnamese, Khasi-Palaungic, Mon, Aslian, Bugan

At some points during their early development in South Asia, due to either language contact led to adoptions of South Asian areal features or internal rhythmic changes, the Munda languages presumably made a syntactic shift from head-first, prefixing SVO to head-final, suffixing SOV. Proto-North Munda restructured all prefixes and prepositions into suffixes. The situation is quite different in South Munda languages, especially Juang, Gtaʔ, and Sora-Gorum, where the original proto-Munda prefix slots are well-preserved, but later additional developments of their predicates are mostly suffixes or enclitics.