Pramāṇa · Search · About · API

Karnatakabhashabhushana 1

Karṇāṭakabhāṣābhūṣaṇa 1

Language
Sanskrit

Contents

1.1.1

atha karṇāṭaka-śabda-siddhiḥ

Now begins the formation of Kannada words.

English (machine translation) (machine translation)

1.1.2

atha śabdō ’yam ādau māṅgalyārthaḥ. māṅgalyādīni śāstrāṇi pradanta iti prasiddhēḥ. adhikārārthō vā. mayā karṇāṭa-viṣaya-śabda-siddhir adhikriyata ity arthaḥ.

Here the word "atha" is used at the beginning for the purpose of auspiciousness, according to the well-known dictum: "Treatises with auspicious beginnings flourish." Or it has the sense of a governing rule. The meaning is: the formation of words in the domain of Kannada is being undertaken by me.

English (machine translation) (machine translation)

1.2.1

akārādayaḥ prasiddhā varṇāḥ.

[1] The phonemes beginning with *a* are well known.

English (machine translation) (machine translation)

1.2.2

akārādayō varṇā hakāra-paryavasānāḥ prasiddhā eva vēditavyāḥ. tēṣām anādi-kāla-prasiddhatvād anyathā nōpadēṣṭavyāḥ.

The phonemes beginning with *a* and ending with *ha* are indeed to be understood as well established. Since they are established from beginningless time, they need not be taught otherwise.

English (machine translation) (machine translation)

1.3.1

tēṣv ādau caturdaśa svarāḥ.

[2] Among them, the first fourteen are vowels.

English (machine translation) (machine translation)

1.3.2

tēṣv akārādiṣu varṇēṣv ādau caturdaśa-vyavasthitāś caturdaśa-varṇās svara-saṁjñā bhavanti. tē kē? a ā, i ī, u ū, r̥ r̥ r̥̄, l̥ l̥ l̥̄, ē, ai, ō, au.

Among those phonemes beginning with *a*, the fourteen phonemes situated at the beginning receive the technical term "vowel". Which are they? *a ā*, *i ī*, *u ū*, *ṛ ṝ*, *ḷ ḹ*, *e*, *ai*, *o*, *au*.

English (machine translation) (machine translation)

1.4.1

ēkadvitrimātrās tē hrasvadīrghaplutāḥ.

[3] Having one, two, or three morae, they are short, long, and protracted.

English (machine translation) (machine translation)

1.4.2

tē iti tacchabdōpādānaṁ vyaktyarthaṁ. akārādisvaravyaktyau yē varṇā ēkadvitrimātrāḥ tē khalu yathāsaṅkhyaṁ hrasvadīrghaplutā bhavanti.

The employment of that word "they" (*tē*) is for the sake of individual instances. Among the individual instances of vowels beginning with *a*, those phonemes having one, two, or three morae are, respectively, short, long, and protracted.

English (machine translation) (machine translation)

1.4.3

a

*a*

English (machine translation) (machine translation)

1.4.4

ā

*ā*

English (machine translation) (machine translation)

1.4.5

ā3

*ā3*

English (machine translation) (machine translation)

1.4.6

i

*i*

English (machine translation) (machine translation)

1.4.7

ī

*ī*

English (machine translation) (machine translation)

1.4.8

ī3

*ī3*

English (machine translation) (machine translation)

1.4.9

ity ādi.

and so forth.

English (machine translation) (machine translation)

1.4.10

sāmānyāpēkṣayā saṅkhyāpi na virudhyatē.

From the standpoint of generality, the count also is not contradicted.

English (machine translation) (machine translation)

1.5.1

ēdōtau hrasvā ca.

[4] *e* and *o* are also short.

English (machine translation) (machine translation)

1.5.2

ēkaukārau dvāv api kvacid bhāṣāyāṁ hrasvā ca bhavataḥ. nityaṁ sandhyakṣarāṇi dīrghāṇi ity upadēśād idaṁ vacanaṁ.

Both *e* and *o* are also short in some instances in the language. This statement is made because of the instruction that "diphthongs are always long."

English (machine translation) (machine translation)

1.5.3

ele

*ele*

English (machine translation) (machine translation)

1.5.4

ēle

*ēle*

English (machine translation) (machine translation)

1.5.5

sele

*sele*

English (machine translation) (machine translation)

1.5.6

sēle

*sēle*

English (machine translation) (machine translation)

1.5.7

ole

*ole*

English (machine translation) (machine translation)

1.5.8

ōle

*ōle*

English (machine translation) (machine translation)

1.5.9

gole

*gole*

English (machine translation) (machine translation)

1.5.10

gōle

*gōle*

English (machine translation) (machine translation)

1.6.1

kādayas trayastriṁśad vyañjanāni.

[5] The thirty-three beginning with *ka* are consonants.

English (machine translation) (machine translation)

1.6.2

kakāraprabhr̥tayaḥ trayastriṁśatparimitāḥ yē varṇāḥ pratyēkaṁ tē vyañjanasaṁjñā bhavanti. tē kē? k kh g gh ṅ, c ch j jh ñ, ṭ ṭh ḍ ḍh ṇ, t th d dh n, p ph b bh m, y r l v ś ṣ s h.

Those phonemes beginning with *ka* and numbering thirty-three individually receive the technical term "consonant". Which are they? *k kh g gh ṅ*, *c ch j jh ñ*, *ṭ ṭh ḍ ḍh ṇ*, *t th d dh n*, *p ph b bh m*, *y r l v ś ṣ s h*.

English (machine translation) (machine translation)

1.7.1

vargās tē māntāḥ pañcaśaḥ pañca.

[6] Those ending in *ma* are five consonant classes of five each.

English (machine translation) (machine translation)

1.7.2

tē punaḥ kakārādayō varṇāḥ makārāntāḥ pañca pañca bhūtvā pañcaiva vargasaṁjṇā vēditavyāḥ. tē kē? k k g gh ṅ, c ch j jh ñ, ṭ ṭh ḍ ḍh ṇ, t th d dh n, p ph b bh m.

Furthermore, those phonemes beginning with *ka* and ending with *ma*, grouped five by five, are to be known as five consonant classes. Which are they? *k kh g gh ṅ*, *c ch j jh ñ*, *ṭ ṭh ḍ ḍh ṇ*, *t th d dh n*, *p ph b bh m*.

English (machine translation) (machine translation)

1.8.1

anunāsikās tēṣāṁ pañcamāḥ.

[7] The fifth of these are nasals.

English (machine translation) (machine translation)

1.8.2

tēṣāṁ vargāṇām antē vyavasthitā varṇās tē anunāsikasaṁjñā bhavanti. tē kē? ṅa-ña-ṇa-na-mā.

The phonemes situated at the end of those consonant classes receive the technical term "nasal". Which are they? *ṅa*, *ña*, *ṇa*, *na*, *ma*.

English (machine translation) (machine translation)

1.9.1

vibhāṣayā yavalāḥ.

[8] Optionally, *ya*, *va*, and *la* [are nasals].

English (machine translation) (machine translation)

1.9.2

yavalā ity ētē varṇāḥ vikalpēna anunāsikasaṁjñā bhavanti.

These phonemes—*ya*, *va*, and *la*—are optionally given the technical term "nasal".

English (machine translation) (machine translation)

1.9.3

tad yathā —

For example:

English (machine translation) (machine translation)

1.9.4

saṁyutaṁ

*saṁyutaṁ*

English (machine translation) (machine translation)

1.9.5

samyutaṁ

*samyutaṁ*

English (machine translation) (machine translation)

1.9.6

sēṁve

*sēṁve*

English (machine translation) (machine translation)

1.9.7

sēve

*sēve*

English (machine translation) (machine translation)

1.9.8

kaṁlene kanaldu

*kaṁlene kanaldu*

English (machine translation) (machine translation)

1.9.9

kallene kanaldu

*kallene kanaldu*

English (machine translation) (machine translation)

1.9.10

yaṁlōkaṁ

*yaṁlōkaṁ*

English (machine translation) (machine translation)

1.9.11

yallōkaṁ

*yallōkaṁ*

English (machine translation) (machine translation)

1.9.12

jaṁvaṁ

*jaṁvaṁ*

English (machine translation) (machine translation)

1.9.13

javaṁ

*javaṁ*

English (machine translation) (machine translation)

1.9.14

gāṁve

*gāṁve*

English (machine translation) (machine translation)

1.9.15

gāve

*gāve*

English (machine translation) (machine translation)

1.9.16

ōṁvaṁ

*ōṁvaṁ*

English (machine translation) (machine translation)

1.9.17

ōvaṁ

*ōvaṁ*

English (machine translation) (machine translation)

1.9.18

kāṁvaṁ

*kāṁvaṁ*

English (machine translation) (machine translation)

1.9.19

kāvaṁ

*kāvaṁ*

English (machine translation) (machine translation)

1.9.20

oṁlaneyonaldu

*oṁlaneyonaldu*

English (machine translation) (machine translation)

1.9.21

ollaneyonaldu

*ollaneyonaldu*

English (machine translation) (machine translation)

1.9.22

ity ādi.

and so forth.

English (machine translation) (machine translation)

1.10.1

laraḍā ēva ḷaṟaḻāḥ prayatnāśrayāt.

[9] *la*, *ra*, and *ḍa* become *ḷa*, *ṟa*, and *ḻa* through dependence on articulatory effort.

English (machine translation) (machine translation)

1.10.2

laraḍā ity ētē varṇāḥ prayatnaviśēṣāśrayāt ḷaṟaḻā bhavanti. tēṣāṁ bhāṣāyāṁ pr̥thapkr̥thakprayōgārhatvād ayam ārambhaḥ.

These phonemes—*la*, *ra*, and *ḍa*—become *ḷa*, *ṟa*, and *ḻa* on account of a particular articulatory effort. This rule is undertaken because they are fit for distinct usage in the vernacular.

English (machine translation) (machine translation)

1.10.3

tad yathā —

For example:

English (machine translation) (machine translation)

1.10.4

vole

*vole*

English (machine translation) (machine translation)

1.10.5

voḷe

voḷe,

English (machine translation) (machine translation)

1.10.6

kale

kale,

English (machine translation) (machine translation)

1.10.7

kaḷe

kaḷe,

English (machine translation) (machine translation)

1.10.8

more

more,

English (machine translation) (machine translation)

1.10.9

moṟe

moṟe,

English (machine translation) (machine translation)

1.10.10

mare

mare,

English (machine translation) (machine translation)

1.10.11

maṟe

maṟe,

English (machine translation) (machine translation)

1.10.12

kaḍal

kaḍal,

English (machine translation) (machine translation)

1.10.13

kaḻal

kaḻal,

English (machine translation) (machine translation)

1.10.14

kūḍi

kūḍi,

English (machine translation) (machine translation)

1.10.15

kūḻi

kūḻi,

English (machine translation) (machine translation)

1.10.16

ity ādi.

and so forth.

English (machine translation) (machine translation)

1.10.17

saṁskr̥tapadēṣu tēṣām aikyam ēva —

In Sanskrit words, however, they are identical:

English (machine translation) (machine translation)

1.10.18

salilaṁ

salilaṁ,

English (machine translation) (machine translation)

1.10.19

saḷiḷaṁ

saḷiḷaṁ;

English (machine translation) (machine translation)

1.10.20

khaḍgaṁ

khaḍgaṁ,

English (machine translation) (machine translation)

1.10.21

khaḻgaṁ

khaḻgaṁ.

English (machine translation) (machine translation)

1.11.1

nātra prāyēṇa vargāṇāṁ dvitīyacaturthāḥ.

[Sūtra 11:] Here [in Kannada], the second and fourth [consonants] of consonant classes generally do not occur.

English (machine translation) (machine translation)

1.11.2

asmin karṇāṭakaviṣayē vargāṇāṁ dvitīyacaturthākṣarāḥ kha-cha-ṭha-tha-phā gha-jha-ḍha-dha-bhāś ca prāyēṇa na vidyantē. prāyēṇēti vacanēna kvacit saṅkhyānukaraṇādau dr̥śyantē. tat kathaṁ?

In this domain of Kannada, the second and fourth letters of the consonant classes—kha, cha, ṭha, tha, pha and gha, jha, ḍha, dha, bha—generally do not occur. By the statement "generally," they are seen in some instances, such as in numerals, onomatopoeia, and so forth. How so?

English (machine translation) (machine translation)

1.11.3

saṁkhyāyāṁ tāvat —

First, in numerals:

English (machine translation) (machine translation)

1.11.4

irchāsiraṁ

irchāsiraṁ,

English (machine translation) (machine translation)

1.11.5

eṇchāsiraṁ

eṇchāsiraṁ,

English (machine translation) (machine translation)

1.11.6

nūrchāsiraṁ

nūrchāsiraṁ.

English (machine translation) (machine translation)

1.11.7

anukaraṇē —

In onomatopoeia:

English (machine translation) (machine translation)

1.11.8

khammene

khammene,

English (machine translation) (machine translation)

1.11.9

ghaḻelene

ghaḻelene,

English (machine translation) (machine translation)

1.11.10

challene

challene,

English (machine translation) (machine translation)

1.11.11

paṭhilene

paṭhilene,

English (machine translation) (machine translation)

1.11.12

duḍhummene

duḍhummene,

English (machine translation) (machine translation)

1.11.13

thokkene

thokkene,

English (machine translation) (machine translation)

1.11.14

dhoppene

dhoppene,

English (machine translation) (machine translation)

1.11.15

bhōṅkene

bhōṅkene.

English (machine translation) (machine translation)

1.11.16

ādigrahaṇād anyatrāpi —

From the mention of "and so forth," elsewhere also:

English (machine translation) (machine translation)

1.11.17

ghaḻiyaṁ

ghaḻiyaṁ,

English (machine translation) (machine translation)

1.11.18

jhaḷaṁ

jhaḷaṁ,

English (machine translation) (machine translation)

1.11.19

ṭhakku

ṭhakku,

English (machine translation) (machine translation)

1.11.20

ḍhokkaraṁ

ḍhokkaraṁ,

English (machine translation) (machine translation)

1.11.21

dhaḻaṁ

dhaḻaṁ,

English (machine translation) (machine translation)

1.11.22

bharam

bharam.

English (machine translation) (machine translation)

1.12.1

śaṣā ca.

[Sūtra 12:] And śa and ṣa [do not occur].

English (machine translation) (machine translation)

1.12.2

śakāraṣakārau dvāv apy asmin karṇāṭakaviṣayē na staḥ.

Both the phonemes śa and ṣa do not exist in this domain of Kannada.

English (machine translation) (machine translation)

1.12.3

sasi

sasi,

English (machine translation) (machine translation)

1.12.4

sisu

sisu,

English (machine translation) (machine translation)

1.12.5

sēse

sēse.

English (machine translation) (machine translation)

1.13.1

r̥kārādayaś catvāraḥ svarāḥ.

[Sūtra 13:] The four vowels beginning with ṛ [do not occur].

English (machine translation) (machine translation)

1.13.2

r̥, r̥ r̥̄, l̥, l̥ l̥̄ ity ētē catvāraḥ svarāś ca karṇāṭakaviṣayē na santi.

These four vowels—ṛ, ṝ, ḷ, and ḹ—also do not exist in the domain of Kannada.

English (machine translation) (machine translation)

1.13.3

risi

risi,

English (machine translation) (machine translation)

1.13.4

basavaṁ

basavaṁ,

English (machine translation) (machine translation)

1.13.5

ēvam ādi.

and so forth.

English (machine translation) (machine translation)

1.14.1

anusvārō binduḥ.

[Sūtra 14:] The anusvāra dot is the anusvāra.

English (machine translation) (machine translation)

1.14.2

nistuḷōdāttamuktāphalānukārī binduḥ anusvārasaṁjñō bhavati. kēvalasyāsya bindōḥ prayōgābhāvād akāraḥ sannidhīyatē — aṁ iti.

The dot resembling a perfectly round, prominent pearl receives the technical term anusvāra. Because this dot cannot be used alone, the letter a is placed near it, as: aṁ.

English (machine translation) (machine translation)

1.15.1

sa uparyadhastāc ca visargaḥ.

[Sūtra 15:] That [dot placed] above and below is the visarga.

English (machine translation) (machine translation)

1.15.2

sa ēva binduḥ uparyadhastāc ca vyavasthitō visargasaṁjñō bhavati — aḥ iti.

That very dot, arranged above and below, receives the technical term visarga, as: aḥ.

English (machine translation) (machine translation)

1.16.1

vibhaktyantaṁ padaṁ.

[Sūtra 16:] An inflected word is that which ends in a case inflection.

English (machine translation) (machine translation)

1.16.2

nāmākhyātavibhaktyantaṁ yacchabdarūpaṁ tat padasaṁjñaṁ bhavati.

Whatever word form ends in a nominal or verbal inflection receives the technical term of inflected word.

English (machine translation) (machine translation)

1.16.3

keyvolaṁ divadante sasiyindesedattu

keyvolaṁ divadante sasiyindesedattu.

English (machine translation) (machine translation)

1.17.1

parasparasannikarṣaḥ saṁhitā.

[Sūtra 17:] Continuous phonological text is mutual proximity.

English (machine translation) (machine translation)

1.17.2

śliṣṭānāṁ varṇānāṁ parasparamanyōnyaṁ sannikarṣaḥ saṁślēṣaḥ saṁhitā sandhir ity ucyatē.

The mutual, reciprocal proximity and close contact of adjacent phonemes is called continuous phonological text or euphonic combination.

English (machine translation) (machine translation)

1.1_2

iti karṇāṭakvyākaraṇavivaraṇē saṁjṇāvidhānaṁ nāma prathamaḥ paricchēdaḥ

Thus ends the first chapter, called "Rule of Technical Terms," in the Commentary on Kannada Grammar.

English (machine translation) (machine translation)

Translations (1)