39 Joining (sandhi) of diphthongs (eC): Section 2 of the Laghu-Siddhanta-Kaumudi (contd.)

After dealing with the sandhi of vowels, the Laghu Siddhanta-Kaumudi proceeds further with diphthongs:

Laghu 29. Eco ‘yavāyāvaḥ (Panini 6.1.78)

This actually follows right after 6.1.77 iko yaṇ aci, which we saw earlier (“the semi-vowels yaṇ= y, v, r, l substitute for the corresponding vowels iK= i, u, ṛ, ḷ (long and short), when follwed by a vowel aC”). The first term in 6.1.78, eco = ecaḥ, in genitive case singular (6/1); it denotes the entity which is being replaced, i.e. the substituend. It refers to the sounds contained in the pratyāhāra eC, or Siva-sutras 7 and 8: e, o, ai, au (what are called diphthongs in English grammar). The entity which replaces, the substitute, is the set of sounds listed in the sutra: ay, av, āy, āv (the whole collectively given a plural ending).

We have to understand carry-over by anuvṛtti of the word aci (7/1) from the preceding sutra; that is, in the environment of a following aC (vowel), the respective substitutions are made. Examples for the addition of suffixes:

Ce + ana = cayana(ṃ)
Lo + ana = lavana(ṃ)
Lau + aka = lāvaka(ḥ)

The Laghu interposes a meta-rule to control the substitution among the two sets (eC and ay-av-āy-āv):

Laghu 30. Yathā saṃkhyam anudeṣaḥ samānām (Panini 1.3.10)

The paraphrase: Yathā saṃkhyam (0, indeclinable= ‘as the number’) anudeṣaḥ (nominative singular 1/1 = ‘assiignment’) samānām (genitive pluural 6/1 = ‘of the same, similar’). Or,
“Assignment of equivalents for equal numbers of elements follows the order of enumeration” (Sharma).

In other words, assignment from one list (the substituends) to another (the substitutes) of equal number of elements, is done by matching their numerical or ordinal positions: the first with the first, etc. Here, it matches the elements ay with e, av with o, āy with ai, and āv with au.

More examples (from the Laghu):
Hare + e = haraye ‘to Hari’
viṣṇo + e = viṣṇave ‘to Vishnu’
nai + aka = nāyaka ‘leader’
pau + aka = pāvaka ‘purifier; fire’

As we can appreciate, the S-K brings together the relevant sutras where it will add to clarity; we can then begin to develop our awareness of how the different parts of the grammar hang together right from the initial stages. Here, a sutra from Book 6 is being elucidated with another from Book 1, Chapter 3.

The next succeeding sutra from Panini is also given in the Laghu:

Laghu 30. Vānto yi pratyaye (Panini 6.1.79)

The item denoted by vānto = va-antaḥ are those ending in –v, i.e. av and  āv. “The substitution of” is to be understood. The word pratyaye (‘affix’, locative case), being in locative case (7/1), denotes the right-context in which the substitution takes place; the qualifier yi denotes an affix starting with the sound y. The sutra means:
“The substitution by v-final items (i.e. av and āv) of o and au [to be understood by carry-over], when followed by a y-initial affix (an affix beginning with the letter y).

That is, the substitutions come into operation not only when a vowel or diphthong follows, but also when a y- sound follows. Examples:

Go + yam = gavyam ‘pertaining to cow’
Nau + yam = nāvyam ‘pertaining to boat’

Vasu (p.1075) helpfully throws further light on the conditions required:
1) it happens only with av and āv, which means to the sounds o and au, and not to e and ai;
2) the succeeding affix has to start with y-, so the substitution does not take place in forms like gobhyām, where the affix does not have a y-;
3) the following element has to be an affix. However, there is one exception which is illustrated by the commentator, in the vārttika; this is in words referring to distance measures,

adhva parimāṇe ca
‘And in distance measures’

go + yũti = gavyũti, ‘a distance of about four miles’; here the substitution by av has taken place, even though yũti is not an affix.

It happens in the Vedas; in normal parlance, we use goyũti.

38 Vowel joining (ac-sandhi)-II: Section 2 of the Laghu-Siddhanta-Kaumudi

Proceeding further with the Laghu S-K, we come to this seemingly connected sutra, which actually introduces a totally new concept:

Laghu 24. Anaci ca (Panini 8.4.47)

It looks connected because of the ca ‘and, also’, which makes it look like a continuation of a preceding chin of thought (calling for us to furnish the missing words from preceding sutras by anuvṛtti (ellipsis). We will be right in thinking so, but the preceding sutra from which the missing words need to be brought in, is not that in the Laghu, but that in the panini sutras:

Panini 8.4.46. aco rahābhyām dve
The words are:
Acah (genitive case, ‘of the vowels aC’) dve (‘two, reduplication’) rahābhyām (ablative case dual number, ‘from the r and h sounds’)

We may note here the use of the ablative case (rahābhyām) denotes something to be done to the right, i.e. to a following entity (Panini 1.1.67 tasmād ity uttarasya, see previous post here http://readingpanini.blogspot.com/2016/06/37-vowel-joining-ac-sandhi-i-section-2.html).

Even this, however, requires completion by bringing in words from previous sutras:

anuvṛtti: samhitāyām (from 8.2.108, ‘in samhitā = close proximity, contact) yarah (from 8.4.45, ‘of a sound denoted by yaR’), vā (‘optionally’).

The yaR sounds, as per the siva-sutras, comprise of all the consonants from the ya of siva-sutra 5, to the sa of no.13: it omits only the consonant ha. (R stands for the end-marker, so we do not stop at the ra of siva-sutra no.5 itself). The whole thing is rendered thusly by Sharma:
“A sound denoted by the abbreviatory term yaR, when occurring in close proximity after a vowel followed by r and h, is optionally replaced with two” (Sharma, Vol.VI, p.771).

In a nutshell, where a vowel followed by r or h, is again followed by any consonant (except h), it results (optionally) in duplication; but of what? The accepted interpretation is: of the consonant (except the h sound). Thus, we get (Sharma, ibid.)

Arkkah
Markkah
Brahmmā
Apahnnute

Having retraced this from the Panini, we can now get back to the Laghu:

Laghu 24. Anaci ca (Panini 8.4.47)

This sutra follows immediately upon 8.4.86 aco rahābhyām dve. Thus it is assumed to continue the idea of reduplication (dve).

An-aci (locative case, ‘in [the context of] a non-vowel’) ca (‘also’)

We have to supply the following by anuvṛtti (ellipsis): samhitāyām (from 8.2.108, ‘in samhitā = close proximity, contact) yarah (from 8.4.45, ‘of a sound denoted by yaR’), vā (‘optionally’), aco dve (from 46). The paraphrase (vṛtti) is as follows:

Acah uttarasya yaro dve bhavatah anaci paratah
“A sound denoted by yaR” [that is, all the consonants except h] “and occurring after one denoted by aC” [a vowel or diphthong] “is, optionally, replaced with two, even when aC does not follow” ((Sharma).

A consonant (except h) coming immediately after a vowel is reduplicated:

Daddhy atra ‘milk here’ [from dadhi atra]
Maddhv atra ‘honey here’ [from madhu atra]

We notice that in both these examples, the reduplicationn of the aspirated dh has resulted in the first being reduced to the unaspirated:  -ddh- instead of –dhdh-. This is, actually, explicitly covered by a subsequent sutra:

Laghu 25: jhalām jaś jhaśi (Panini 8.4.53)

This beautifully alliterative aphorism says that:

jhalām (genitive case, ‘[in place] of jhaL = consonants in pratyahara jhaL from jha to ha ‘) jaŚ (‘sounds represented by pratyahara jaŚ =  j, b, g, ḑ and d’) jhaśi (‘in [the presence of] the sounds represented by pratyahara jhaŚ’).

To correctly interpret the three words with their case endings, we have, of course, to recall the meta-rules for genitive case (sixth case) (ṣaṣṭhī sthāne yogā) and locative (seventh) case (tasminniti) already covered. Where a jhaŚ follows, a jhaL is replaced by a jaŚ. This implies that the first of a pair would be de-aspirated:

Labdha (not labhdha)
Dogdhā (not doghdhā
Boddha (not bodhdha)

This is obviously the result of a natural propensity in articulation (aspiration cannot be articulated without an intervening vowel sound, however tiny, between the two consonants), and also demonstrates the extremely close attention paid by the grammarians.

In the example given for a previous sutra:

sudhī (‘the intelligent’) upāsya (‘to be worshipped’) = sudhyupāsya (‘the intelligent fit to be worshipped, or God’.

We are now told that the we can optionally derive

Sudhdhyupāsya or suddhyupāsya

By the optional reduplication. Panini feels that this does not reflect the actual practice, so he provides the rule

Laghu 26. Samyogāntasya lopah (Panini 5.2.23)

Samyoga- (‘conjunction of consonants’) –antasya (genitive, ‘of its ending’) lopah (‘elision, deletion’)

The last element in a combination of consonants (without any inter-vocalic) is to be elided. The commentary interprets this as saying that the entire pada is to be deleted:

Laghu vŗtti: Samyogāntam yat padam tadantasya lopah syāt

Obviously this will a drastic, and counter-productive, culmination. (Actually, one does not see why the rule should be interpreted in such a drastic manner). In the example of

suddhyupāsya
 the laghu feels that the “pada suddhy- ends with a compound consonant; hence, according to the rule, the whole padsa ought to disappear”. However, the grammarian provides the following rule to limit this action:

Laghu 27. Alo’ntyasya (Panini 1.1.52)

We have already come across this one (here: http://readingpanini.blogspot.com/2015/11/26-further-rules-on-substitution.html). This says that a substitution ordered for any letter aL (by showing it in the genitive case, aLah), only acts on the final letter (antya). Thus any elision in suddhy- will operate only on the –y-, which would presumably yield suddh-upāsya. But here “Katyayana interferes”, and states:

Lagu 28. Yaņah pratişedho vācyah

“The prohibition of the rule (No.26) in the case of yaŅ should be stated” (Ballantyne, p.11). So “the elision does not take place, and the formation of the word suddhyupāsya (a name of God – ‘He who is to be worshipped by the intelligent’) is completed” (ibid).


In broad terms, we see that the S-K (and more so the Laghu) does not always provide all the links in the logical chains. We may have to bring in additional material from the full version. In this context, the S-K or Laghu S-K may be thought of more as an aid to memory, by citing key formulas. These are like triggers to bring up the fuller logical trains from the Astadhyayi.