Krishnamurti Bhadriraju. The Dravidian Languages

Подождите немного. Документ загружается.

10.6

Neologisms

481

Zvelebil (1975: 439) endorses these attempts at nativization of general and technical

terms and says that they ‘often increase efﬁciency, economy and precision of the lan-

guage’. Other linguists may consider this reform as an unnecessary interference with

the normal growth of a language through time-honoured processes of borrowing. Much

time would be wasted if assimilated English loanwords like radio, telephone, station

etc. were to be replaced by newly coined native terms, which have to be learnt afresh by

everybody. The other languages do not have any trace of this trend, which means that

speakers of these languages consider borrowing as a natural corollary of growth.

10.6.1 Language of news media

In 1978 a seminar was held at Osmania University on the modernization of Indian

languages in news media and a publication was brought out with adequate case material

from nine major Indian languages (Krishnamurti and Mukherji 1984). Telugu, Kanna

and Tamil were included in the study. It was noticed that there were eight processes

underlying new coinages in different languages, based on English models. These have

developed independently in each language (see Krishnamurti 1984; 1998: 227–48).

They are: (i) total borrowing as tatsamas (with phonetic adaptation); (ii) loan blends, i.e.

partial borrowing combining native and non-native constituents; (iii) loan translation or

calquing; (iv) ‘adaptive coining’, i.e. coining a new expression with native or Sanskritic

elements which strike the native writers as the best translation of a given expression in

English; (v) paraphrasing; (vi) assigning a new meaning to some traditional expressions;

(vii) extending the meaning of existing words (loan shift) by using an old word in a new

context, or

creating translation equivalents

within the native stock; (viii)

totally coining

new terms within the cultural matrix of native speakers after assimilating a non-native

concept into native consciousness. A few examples, mainly from the major literary

languages, will illustrate these processes.

(22)(I) Loanwords

English Telugu Kanna

da Tamil

bank by¯anku [b ¯æ

ŋku] by¯anku [b ¯æŋku] b¯enk

hotel h¯o

tal h¯o

t¯elu ¯o

tal

police p¯ol¯ısu p¯ol¯ıs p¯ol¯ıs

(II) Loan blends

writ petition ri

t darakh¯astu ri

t arj¯ı –

[Eng. + Pers.]

World Bank prapanca by¯anku wi´sva by¯anku ulaku vanki

[Skt. + Eng.] [Skt. + Eng.] [Ta. + Eng.]

Here the Persian originals have been taken through the route of Dakkhini (Urdu of south India).

482 Lexicon

small-scale Te. cinna-tarh¯a-pari´srama [Te. small-Ur. type-Skt.

industry industry]

state lottery Ka. r¯ajya l¯a

tar¯ı [Skt. state-Eng. lottery]

(III) Loan translations

English Telugu Kanna

da Tamil

black money nalla dhanam kappu ha

na karuppu pa

nam

[Te. + Skt.] [Ka. + Ka.] [Ta. + Ta.]

black market nalla baj¯aru k¯a

la sante –

[Te. + Ur.] [Skt. + Ka.]

(IV) Adaptive coining

English Telugu Kanna

da Tamil

agitation ¯and¯o

na ¯and¯o

na –

[Skt.] [Skt.]

agriculture s¯edyam [Te.] kr

si [Skt.] makas¯ul [Ur.]

transport raw ¯a

n¯a [Ur.] s¯arige [Ka.] p¯okkuvaruttu [Ta.]

self

-sufﬁciency

Te. swayam

samruddhi

[Skt. + Skt.]

-patient

Ka. o

r¯ogi

[Ka. + Skt.]

cross

-examination

Ta. karukku

vis¯ari

nai

[Ta. + Skt.]

inquiry

committee

Ka. tanikh¯ı

samiti

[Ur. + Skt.]

The difference between (iii) and (iv) is subtle; (iii) is a literal translation of the model,

word by word; (iv) is also a translation but not literal; each language has a way of

considering what would be the best way of rendering the model in such a way that it

would sound natural in the language concerned. Note that the resource language for

coining can be native or any other language which has become nativized like Sanskrit or

Urdu.

(V) Paraphrasing

‘non-aligned’ : Te. ¯e r¯ajya-k¯u

t¯ani-k¯ı cend-a-ni

[whichever nation-group-dat belong-neg-adj]

‘not belonging to any groups of nations’

(VI) Semantic reinterpretation

Some established Sanskrit expressions are given a new context and mean-

ing, e.g.

radio Te. ¯ak¯a´sa-w¯a

ni [Skt. + Skt.], Ta. v¯a

n-oli [Ta. + Ta.] ‘sky-voice’

(originally ‘an invisible voice from the sky foreboding future

events’)

fuel Te. indhanam (originally ‘dry wood used in sacriﬁcial ﬁre’)

missile Ka. brahm¯astra (‘Brahma’s arrow’, epic meaning ‘an arrow

which never fails to destroy the target, a powerful weapon’)

10.7

Structured semantic

ﬁelds 483

(VII) Loan-shifts

Using established words in new contexts:

New meaning Traditional meaning

reserve (of funds) Te. nilawa ‘saving’(lit. ‘that which is stationary’)

industry Ka. kai-g¯arike ‘hand-work’

seniority Ka. jye

thate ‘elderliness’

aim Ta. kur

i ‘aim to shoot at’

(VIII) Idiomatic translation or creative coining

smuggled goods Te. donga saruku [stolen goods]

black-marketing Te. donga wy¯ap¯aram [thief business]

injection Te. s¯udi mandu, Ka. cuccu maddu [needle medicine]

television Ta. pa

ta r¯e

diy¯o [picture radio]

The foregoing examples give a fair idea of the way the major languages are building

different registers for modern communication.

10.7 Structured semantic ﬁelds

Proto-Dravidian had only four basic terms for colour, white, black, red, yellow–green

(see section 8.2.1: items 5–8). Among kinship terms there are expressions for ‘man,

woman, father, mother, daughter, son, younger and elder brother, younger and elder sis-

ter’ (see section 1.2.2.3). The remaining are restricted to certain languages or subgroups.

In modern languages, many new combinations have been innovated for other kin rela-

tions, particularly for the second ascending generation (grandfather, grandmother) or the

second descending generation (grandson, granddaughter), for parallel and cross-cousins

and their parents etc. These have been discussed in section 1.2.2.3. A brief survey of

body parts is stated below. Only words denoting the following are reconstructable for

Proto-Dravidian.

(23) Reconstructed words for body-parts

body

∗

may-(mt)- SD I, II, >

∗

m¯e-nd- SD II, CD, ND [5099];

∗

talSD I and SD II; also means ‘chest, stomach’ [585].

bone

∗

el-V-mp/-nk SD I; Te. [839],

∗

z-an, SD II [4418] ‘bone,

stone of fruit’.

breast

∗

mul-ay SD I [4985], Te. cannu;

∗

p¯al ‘milk’; SD II (Go.

da); CD (P

a.)

‘woman’s breast’ [4096]; the

seman-

tic shift is witnessed only in some languages of central

India.

ear

∗

kewi SD I, SD II, CD, ND[1977a];

∗

kir-V- Tu. of SD I

and Kui–Kuvi of SD II [1977b].

eye

∗

n SD I, SD II, CD, ND [1159a].

ﬁnger

∗

wir-al SD I, Te.; wir-and /-anj SD II, CD [5409].

484 Lexicon

foot

∗

t-V,SDI,

∗

t-V-k- SD II ‘foot, bottom, base’, CD [72].

hand

∗

kay [2023] SD I, SD II, CD, ND ‘hand, arm’ [2019].

head

∗

talay ‘head, hair, top’ SD I, SD II, ND; also ‘honeycomb’

Pa., Nk., Go. [3103].

heart

∗

ti ‘heart, kidney’ SD I, SD II, CD, ND [1693].

intestines

∗

t-al SD I [1652],

∗

z-ul CD, Go. [4193].

knee

∗

z-V- ‘joint’, mo

z-am k¯al/-kay ‘knee, elbow’ SD I, SD

II, CD, ND [4990].

leg

∗

k¯al SD I, SD II, CD, ?Brahui trikkal ‘tripod’; ‘leg/foot’

[1479].

mouth

∗

w¯ay ‘mouth (edge, beak,

mouth of vessel, aperture, b

lade

of sword)’ SD I, SD II, CD, ND [5352].

liver

∗

z-V-nk/-nkk SD II, SD II (except Telugu), Tu. and CD

[3120],

∗

ir-Vl/w SD I [546].

nail

∗

ukir ‘ﬁnger nail, toe-nail’ SD I,

∗

g¯or/

∗

k¯or SD II, CD, Br.;

∗

or-kk-Ku

r.–Malt. [561].

navel

∗

t-V- SD I, SD II, ND [4460b],

∗

pok-ku

z-V-SDI,SD

II, CD [4460a].

neck/throat

∗

z-u-tt-SDI,ka

z-nt(t)- Tu., CD, SD II [1366]; ku-nt-/ntt-

∼ ko-nt-/ntt- ‘throat, voice, tone’ SD I, SD II [1718].

nose

∗

m¯u-nk(k)u SD I, SD II, CD, m¯u-nc- ND ‘nose, beak’

[5024]

skin

∗

tokk-/

∗

tokal/

∗

t¯ol SD I, SD II, CD [3559].

skull

∗

put-V ‘gourd shell, skull’ SD I, SD II, CD [4332].

stomach

∗

t-/po

t-V- ‘belly, stomach, pregnancy, intestines’ SD I,

SD II, CD, ND [4494];

∗

wac-Vt SD I, SD II [5259],

wac-V-(kk)Ko

da–Kui–Kuvi–Pe.–Man

da; was-Vr-,

way-nc- Tu. [5259].

tendon,

nerve

∗

˜nar-ampu ‘tendon’ (‘nerve, sinew, vein, pulse’) SD I,

SD II, CD [2903].

thigh

∗

kut-u-k/u-w SD I, SD II, CD, ND ‘thigh (of human or

animal)’,

∗

kut-a-nk/-nkk SD I [1840].

tongue

∗

n¯a-l, n¯a-l-k(k) SD I, SD II, n¯a-nk-(<

∗∗

n¯al-nk-k)CD

[3633].

tooth

∗

pal SD I, SD II, CD, ND; by extension ‘harrow’ in Ta.

Te. [3986a].

A comparative listing of basic vocabulary is not attempted here, since many Proto-

Dravidian reconstructions occur in different places throughout this volume: kinship

10.8

Strategies of expr

essives

485

terms (chapter 1), personal pronouns (chapter 6), numerals (chapter 6), body parts

(above) and basic verbs and adjectives (chapters 7 and 8) and many widely represented

etymologies under comparative phonology (chapter 4).

There is one other aspect of interest in the Dravidian lexicon, i.e. expressives, which

include onomatopoetic expressions, echo words and lexical reduplicatives. There is

overlap among these three categories. All of these are found in the literary languages

from the time of early records and need to be considered as a heritage of Proto-Dravidian.

10.8 Strategies of expressives

10.8.1 Onomatopoetic expressions or ideophones

Emeneau (1969a: 274–99; repr. 1980a: 250–93) has discussed this phenomenon in

comparison with Indo-Aryan counterparts. Abbi (1992) has dealt with ‘reduplication’

as a cover term for ‘expressives, echo-words, grammatical and lexical reduplicatives’

comprehensively for the whole of South Asia. Apparently, the phenomenon of ono-

matopoeia is found in many language families and is not an areal feature of South Asia

alone. However, the phonological and morphological aspects for each family are unique.

In Dravidian Emeneau proposes two types of onomatopoeia, namely non-reduplicated

and reduplicated.

The non-reduplicated ones generall

y have the structure (C)VCVCC-/

CVC-

VC- with the second syllable longer than the ﬁrst which is not a favoured type of

lexeme in non-ideophones. Both the types can be verbalized with an inﬂected form of

the verb

∗

aHn- ‘say’ (> ¯an/an-/en-/in-).

(24) Non-reduplicated

A. Telugu

(i) Adverbials marked by -na, e.g. digg-ana ‘suddenly’ (OTe.),

gabh ¯al(u)-na ‘urgently’, dabukku-na ‘suddenly’, bha

lu-na ‘com-

ing of dawn’, bh¯oru-na ‘weeping loudly’, gu

tukku-na ‘swallowing

fast’, gammu-na ‘silently’, phe

d¯el-na ‘cracking up’, gappu-na ‘ﬁre

ﬂashing’, cappu-na ‘quickly’,

thappu-na ‘sounding thap’ etc.

(ii)

With -

an ‘say’: curukku-m-an ‘a body

part burns with a heated object

suddenly’, tha

lukku-m-an ‘ﬂash like lightning’ etc.

B. Kota

(i) With the verb -in ‘say’, e.g. da

nak in- ‘become limp with fatigue’,

lak in- ‘body feels cool’.

(ii) With an adverbial marker -n, e.g. burak-n ‘suddenly’, di

dak-n ‘in a

ﬂash’, dabak-n ‘with noise of falling’, pa

t¯ar-n ‘with noise of break-

ing’.

(iii) With -

Vl: da

d-¯al ‘with a sudden jerk’, car-¯ıl ‘with the noise of snake’s

motion’.

486 Lexicon

The reduplicated ones are more numerous and widely represented. Several of these

can be reconstructed for Proto-Dravidian. A few examples from DEDR:

(25) PD

∗

kala-kala- > SD I: Ta. kalakala (-pp-, -tt-) ‘rustle, tinkle, rattle’,

kalakalam ‘chirping of birds’, ka

laka

l(a)-e

n-al [sound-say-noml] ‘tink-

ling’, Ma. ka

laka

la ‘confused noise’, Ka. kalakala ‘buzz of a crowd’,

galagala/ga

laga

la ‘rattling, clinking’, Tu. kalakala ‘hum’, galagala ‘noise

of bracelets etc.’; SD II: Te. kalakala ‘confused noise, buzz of a crowd’,

kalakala-l-¯a

du ‘to shine forth, glow on face’, galagala-l-¯a

du v.i ‘sound

of bracelets, metal ornaments, coins, etc.’, Go. kalla ‘uproar, commotion’,

Kuvi k¯alori ¯a- ‘to shout’; CD: Nk. (Ch.) kalla ‘noise’; ND: qal-qal-tre

‘shake water or cowries’| Skt. kala- ‘soft melodious sound’; CDIAL 2914

[1302].

Note that there is a meaning difference between kalakala and galagala in Telugu and

Kanna

da.

(26)

∗

tapa-tapa or?

∗

daba-daba ‘imitating

walking fast

’ > SD I: Ta

tapukk-

n ‘[tapukk-sa

y], rashness, haste

’, Ko

dab¯al in- ‘noise

of falling from

height’, Ka. tapp-ane ‘suddenly’, dabakk-ane ‘noise of heavy body falling’,

dabb-ane ‘suddenly’, Tu. dabakka ‘suddenly’, dabadaba ‘palpitation of

heart’: SD II: Te. dabbu-na ‘suddenly’, dabadaba ‘walking fast’, Go.

topne ‘quickly’, Kuvi toppe, tobbe ‘quickly’, Pe. tapp in-ji [tapp say-ppl]

‘suddenly’| CDIAL 6170 with

∗

dab- [3069].

Emeneau (1969a: Appendix) has cited 43 entries from DED (1961) and compared the

Dravidian cognates with comparable entries in Indo-Aryan from Turner (1966) and could

ﬁnd parallels

in almost all cases. He suggests that

Rgvedic budbuda- ‘bubble’ could

be the earliest loanword based on PD

∗

dabu

da- (1980b: 282–3), disputed by Hock

(1996: 29, 39–40). Emeneau says, ‘It is probable that in reconstructed PDr. no isolated

word should begin with a voiced stop, except for some members of the onomatopoetic

class’ (1980b: 287–8). Emeneau and Kausalya Hart (1993) have pursued this question

further, taking data from Modern Tamil in which many of the onomatopoetic expressions

are rendered with voiced initial stops, although in script they are represented voiceless.

They suggest on the basis of comparative evidence that ‘in expressive items there are

phonetic developments different from those that are regarded as regular for the language

that is being treated’ (Emeneau and Hart 1993: 83). They examine a total of 76 entries

from DEDR (1984) and propose 24 items out of these as probably beginning with voiced

stops.

10.8

Strategies of expr

essives

487

10.8.2 Echo words and intensives

In echo-word formation, reduplication occurs partially and not fully, and the echo word

is not onomatopoeic. For any base word of any part of speech, beginning with either

a vowel or a consonant with variable vowel length, i.e. (C)V/(C)

V-, an echo word is

formed replacing this syllable by gi-/g¯ı- respectively (the vowel length in the echo word

corresponds to the vowel length of the model word); the rest of the model is repeated

after gi-/g¯ı. The echo word means ‘that and the like, or that etc.’, e.g. Te. illu gillu ‘house,

etc.’, puli gili ‘tiger, etc.’. In other words, for a given word of the structure (C)

VX, there

can be an echo word of the structure g

¯ıX where X = that part of the word excluding the

initial (C)

V. The echo word need not occur immediately after the model word, but can

be coordinated with it in a compound sentence, e.g.

(27) Te. w¯a

d-i-ki illu l¯e-du, gillu l¯edu

[he-obl-dat house be not-3neu-sg , such a thing be not-3neu-sg]

‘he has no

house and nothing of that sort

’

This pattern is available in Tamil, Malay¯a

lam and Kanna

da deﬁnitely. It is reported for

some tribal languages also (for Toda, see Emeneau 1938, for

Alu Kur

umba, see Kapp

1985; a short bibliography occurs in this last reference). This pattern can be reconstructed

for Proto-South Dravidian with a voiced stop as part of the ﬁrst syllable of the echo

word.

Intensives are a restricted set involving a model and a partly reduplicated component,

which intensiﬁes the meaning of the model word. Here the model is a nominal or

adjectival and

the reduplicated part becomes a

morphological extension of the

base word

and is not separable from it, as in the case of echo words. Emeneau (1987a) discusses

this phenomenon and cites twenty-ﬁve examples drawn from DEDR (1984). Intensives

are mainly found in the four major literar

y languages and therefore the construction

is reconstructable for Proto-South Dravidian. However, it is not a productive type like

onomatopoeic expressives, which are an open set admitting possible new formations.

First the examples may be examined. C

a- is the reduplicated part added before

the model in Tamil and Kanna

da; in Telugu C

ta- is added. C

- are identical

with #CVC- of the base word (also see section 8.3.1).

(28) a. Ta. putta(m)-putiya ‘brand new’ (putu, putiya ‘new’) [4275].

b. Ka. tutta-tudi/tu

ta-tudi ‘the very end’, Te. tu

ta-tudi id. (tudi ‘end’)

[3314]

c. Ka. motta-modalu ‘the very beginning’, Te. mo

ta-modalu id. (modalu

‘beginning’) [4950].

d. Ta. ve

rra-veritu ‘absolute worthlessness’ (veri-tu ‘uselessness’, veru

‘empty’) [5513].

488 Lexicon

e. Ta. na

ta-na

tu ‘the very middle’, Ka. na

ta-na

duve id., Te. na

ta-na

adj ‘the very central’ (Ta. na

tu, Ka. na

du,Te.na

di ‘middle, centre’)

[3584].

The principle is that Tamil and Kanna

da geminate the C

of the base word in each

case and Telugu uses C

ta- in all cases. Three cases from Tamil and Kanna

da also

show C

ta- as in Telugu irrespective of C

of the model word. Ta. pa

ta-ppakal, Ma.

t¯a-pakal,Te.pa

ta-pagalu ‘broad daylight’ (pagal(u) ‘day’). In all cases, the resultant

reduplicated morpheme functions as a bound adjective of the model. Tamil has cases

where C

is not a stop; it can be a nasal /n

n/, or a lateral /

l/. This pattern is not found in

Kanna

da and Telugu.

Another intensive with partial reduplication is found in one interjection in all major

languages, e.g. Te. ayy¯o ‘alas’ : ayy-ayy¯o (denotes greater intensity of sympathy). Tamil,

Malay¯a

lam and Kanna

da also have the same pair of expressions. We cannot explain the

last one by

sandhi since a long vowel

+ short vow

el will not result in the loss of

the long

vowel. It is a case of partial reduplication of the model V

- by dropping the ﬁnal

vowel. Old Telugu has another interjection ¯oh¯o (surprise) → ¯oh-¯oh¯o (denotes more

surprise) with reduplication of #

1-.

Another type of expressive is the reduplication of words of different parts of speech in

distributive or intensive meanings, e.g. Te. nouns: illu ‘house’: ill-illu ‘every house’, in

ti-

ki [house-obl-dat] ‘to a house’; in

t-i

ti-ki ‘to each house’; pronouns: interrogative, ewaru

‘who’: ewar-ewaru ‘who particularly?’ (distributive); numerals: iddaru ‘two persons’:

iddar-iddaru ‘two each,

by twos

’; demonstrati

ve adverbs:

akka

da ‘there’: akka

d-akka

‘here and there’, appu

du ‘then’: appu

d-appu

du ‘now and then’; adjectives: pedda ‘big’:

pedda-pedda ‘very big’, cinna ‘small’ : cinna-cinna ‘very small’; adverbs: tondara noml

‘quickness’, tondara-g¯a ‘quickly’: tondara-tondara-g¯a ‘very fast’; verbs: tin- ‘eat’, tin-i

[eat-ppl] ‘having eaten’: tin-i tin-i ‘having eaten a lot’; OTe. cuc-/c¯u

d- ‘see’, c¯u

d-an

[see-inf] ‘to see’: c¯u

d-a-c¯u

d-a ‘as one keeps seeing’.

These patterns are found in all Dravidian languages. Particularly the reduplication

of inﬂected verbs, like the gerund, is a very ancient feature in Dravidian occurring

in early texts of the literary languages. It is also found in Indo-Aryan in the post-

Rgvedic period. Hock (1996: 18, fn.1) says, ‘Non-verbal iteration is found to varying

degrees in many

early Indo-European languages, but verbal iteration (of the type

Skt.

utplutya utplutya ‘continually jumping up’) ﬁrst appears in late Vedic.’ The extensive

use of reduplicated verbs in Modern Indo-Aryan and other languages in South Asia is

illustrated by Abbi (1992: 31–54). The variety of uses between Indo-Aryan and Dravidian

is striking. Diachronically, this is a feature that goes back to Proto-Dravidian and must

have spread to Middle and Modern Indic by diffusion. Examples from different Dravidian

languages occur in earlier chapters under different parts of speech.

Conclusion: a summary

and overview

11.1 Introduction

A summary of conclusions from different chapters, which throw light on the subgrouping

of the Dravidian languages adopted here, will be presented with a short review of earlier

work on this topic. I will brieﬂy speculate on the probable date of Proto-Dravidian. I

have left out several questions like the original home of the Dravidians and whether

they were the same people who ruled the Indus valley around 2500 BCE. The Proto-

Dravidian culture that I have reconstructed, based on comparative vocabulary in sec-

tion 1.2.2, must help archaeologists and linguistic archaeologists to sort out this problem.

I have hinted at the future direction for research in Dravidian studies, comparative and

typological.

11.2 Earlier attempts at subgrouping the Dravidian languages

Serious attempts at the subgrouping of the Dravidian languages proceeded hand in

hand with the study of a number of new languages, mainly in central India, namely

Kolami, Parji, Naiki, Ollari, Gadaba, Ko

da, Pengo etc. during 1950–75. Earlier, L.

V. Ramaswami Aiyar, E. H. Tuttle and T. Burrow (till 1950) placed Telugu in South

Dravidian (our South Dravidian I). I have noted that Telugu is genetically closer to

its northern neighbours, Gondi, Ko

da, Kui, Kuvi, to which Pengo and Man

da were

added later. Burrow and Bhattacharya (1953: xi) have pointed out the close relationship

among Parji, Ollari, Gadaba, Kolami and Naiki. They also speak of ‘many signs of

special connection between Gondi–Ko

da and Kui–Kuvi’. In TVB (ch. 4: 236–74), I

argued for placing Telugu genetically as a close sister of Gondi–Ko

da–Kui–Kuvi, but

its geographical and cultural contact with the other literary languages of south India was

responsible for the diffusion of some of the features of South Dravidian I to Telugu, e.g.

loss of PD

∗

c > s > h > Ø, and having ‘he’, ‘she/it’ in the singular (in common with the

other Central Dravidian languages) and ‘they’ (human plural) and ‘they’ (non-human

plural) in common with South Dravidian I in gender. Consequently, Central Dravidian

has emerged as another branch of Proto-Dravidian with two sub-branches, Telugu–Kuvi

on the one hand and Kolami–Gadaba, on the other. This scenario was accepted by all

490 Conc

lusion: a summary

and overview

Tamil

Malayalam

Toda

Kota

Kodagu

Badaga

Kannada

Figure 11.1 Subgrouping of South Dravidian

by Emeneau (1967b)

Dravidian scholars until the mid 1970s (see my ﬁrst review article on Dravidian studies

in Krishnamurti 1969b/2001a: 114–16).

In the case of the South Dravidian languages Emeneau (1967b/1994) has proposed

a subgrouping (differently conﬁgured) as in ﬁgure 11.1. Emeneau set up two interme-

diate stages: Proto-Toda–Kota node connecting Proto-Tamil–Malay¯a

lam node, which

is brought out by the diagram. I have given morphological arguments to place Ko

dagu

closer to Tamil–Malay¯a

lam than to Toda–Kota. Three other languages have since been

studied, namely Iru

la, Ku

rumba (with several dialects) and Koraga. The position of Tu

in the Southern group still has difﬁculties, although it seems to be part of South Dravid-

ian I, perhaps the very ﬁrst to branch from it. Koraga is close to Tu

lu. Iru

la and Ku

rumba

are closer to Tamil–Malay¯a

lam (see section 4.4.4.2).

North Dravidian consisting of Ku

rux, Malto and Brahui has not presented any prob-

lems, with Ku

rux–Malto being one branch and Brahui the other.

11.2.1 Studies in lexicostatistics

Andronov (1964c) attempted a glottochronological study based on Swadesh’s 100-word

list with cognates from major and known Dravidian languages. He puts the modern stan-

dard varieties of Tamil–Malay¯a

lam, Kanna

da and Telugu in the Southern group. He put

Kolami–Parji and Gondi as another subgroup, w

hich he designated Central Dravidian.

He has all the three languages from North Dravidian. The procedure suffers from several

shortcomings and inadequacies. Except for the literary languages, vocabularies of the

remaining ones are not comprehensive. It is difﬁcult to decide if a cognate has failed to

be recorded or it has not existed, in several cases. The basic word list has been ques-

tioned by many scholars for decades. For instance, the ﬁrst item ‘all’ has different lexical

representations between human/non-human, e.g. Telugu ant¯a, given in the list, refers to