4 1 Aṅgādhikāra

Before I examine examples of rule interaction at the same step in tiṄ and kr̥t derivations, I will examine rules 6.4.1 aṅgasya and 1.4.13 yasmāt pratyayavidhis tadādi pratyaye’ṅgam which play a pivotal role in running Pāṇini’s grammatical machine.

6.4.1 aṅgasya is an adhikāra (heading) sūtra, the jurisdiction of which continues all the way up to the end of 7.4. Pāṇini defines the term aṅga in

1.4.13 yasmāt pratyayavidhis tadādi pratyaye’ṅgam.

Sharma translates this as follows:

‘a form beginning with that after which an affix is introduced is termed aṅga when the affix follows’.

I think that the tradition has not correctly understood these rules, as a result of which it faces multiple problems in performing certain derivations. In this section, I will present my interpretations of these rules, and show how my interpretations enable us to perform these derivations correctly.

One anga per pratyaya per derivation

In my opinion, only one item can be called an aṅga with respect to a certain pratyaya in a derivation. I must admit that I am unable to support this statement using Pāṇini’s rules. However, through the examples discussed in section 4.2 of this chapter, I will show that it is not possible to correctly perform certain derivations without accepting this assumption.

Let me discuss an example from verbal inflection to explain what I mean. Consider the derivation of the present-tense third-person singular form of cit ‘to think’:

cit + LAṬ (3.2.123 vartamāne laṭ1) 🡪 cit + tiP (3.4.77 lasya, 3.4.78 tiptasjhi…2). According to the tradition3, cit is an aṅga with respect to tiP. Then, after we add the vikaraṇa ŚaP by 3.1.68 kartari śap4, we get cit + ŚaP + tiP. According to the tradition, cit + ŚaP too is an aṅga with respect to tiP.

[[109]]

Thereafter, we apply 7.3.86 pugantalaghūpadhasya ca5 to cit and get cet + ŚaP + tiP i.e., ceta + tiP. According to the tradition, ceta too can be called an aṅga with respect to tiP.

So, cit+++(at cit + tiP step)+++, cit + ŚaP and ceta can all be called aṅga with respect to tiP, in the tradition’s opinion. I disagree with the traditional perspective: in my opinion, we can have only one aṅga per affix per derivation[^6]. So, which one of the three options, namely cit, cit + ŚaP and ceta, should be called an aṅga with respect to tiP? I think ceta alone can be called an aṅga with respect to tiP.

I will present my arguments to support this claim below. But before we proceed, I must admit that I am unable to provide any strong evidence from Pāṇini’s rules to justify the arguments I make below. Nonetheless, I will discuss some derivations in section 4.2 of this chapter which will prove that if we accept any other item but ceta as the aṅga with respect to tiP, we risk getting the wrong form at the end of the derivation. That said, now let us consider all three possibilities, namely cit, cit + ŚaP and ceta.

Let us first look at cit. If Pāṇini wanted us to treat cit as an aṅga with respect to tiP, he could have simply said

yasmāt pratyayavidhis tad pratyaye’ṅgam ‘a form after which an affix is introduced is termed aṅga when the affix follows.’

Thus, I do not think that we should call cit an aṅga with respect to tiP. This leaves us with two options: cit + ŚaP and ceta. Let us closely consider 1.4.13 in the context of this derivation to decide which of the two should be called an aṅga with respect to tiP.

yasmāt – to (lit. after) cit
pratyayavidhis – (upon the) addition of tiP
tadādi – that which begins with cit
pratyaye – when tiP follows
aṅgam – (is called) aṅga.

[[110]]

‘Upon the addition of tiP to cit, that which begins with cit is called aṅga when tiP follows.’

The form which begins with cit is an aṅga with respect to tiP. Can we say that cit + ŚaP begins with cit? I do not think so. I think cit + ŚaP is still just a string of two separate items, namely the root cit and the vikaraṇa affix ŚaP. Only when they are fused into a single form that begins with cit, that form can be called an aṅga with respect to tiP. When can we fuse cit and ŚaP into a single form? I think we can do that after applying all possible rules to cit and ŚaP, except those that are triggered by tiP.

So here, we apply 7.3.86 pugantalaghūpadhasya ca to cit (an operation triggered by ŚaP, not by tiP) and get cet + ŚaP + tiP. Note that cet and ŚaP cannot undergo any other operations which are not triggered by tiP, so we can fuse cet + ŚaP into a single form, i.e., ceta. Ceta begins with cet and is followed by tiP, so it can be called an aṅga with respect to tiP. I summarize this information in this table:

Step Question Traditional opinion My opinion
cit + tiP Is cit an aṅga w.r.t. 7 tiP? Yes No
cit + ŚaP + tiP Is cit + ŚaP an aṅga w.r.t. tiP? Yes No
ceta + tiP Is ceta an aṅga w.r.t. tiP? Yes Yes

In my opinion, through 6.4.1 aṅgasya, Pāṇini teaches that, for any P + Q, a rule $$R_P$$ taught in the aṅgādhikāra which is triggered by Q is applicable to its intended operand P only if P is an aṅga with respect to affix Q. Similarly, a rule $$R_Q$$ taught in the aṅgādhikāra which is triggered by P is applicable to its intended operand Q only if P is an aṅga with respect to affix Q.

Also, note that I agree with the tradition that cit is an aṅga with respect to ŚaP. So, at the step cit + ŚaP + tiP, 7.3.86 pugantalaghūpadhasya ca, which belongs to the aṅgādhikāra and which is triggered by ŚaP, is applicable to cit.

Situation complexity

Before we go further, note that, we find vikaraṇas only in tiṅanta and kr̥danta derivations. So, in the rest of the derivations, it is very easy to determine what we should call an aṅga with respect to the affix. For instance, in deva + bhis (example 1 of section 2.7, chapter 2), deva is an aṅga with respect to bhis simply because the affix bhis has been added to deva.

[[111]]

Similarly, in sad + vas + Ṅas (example 2 of section 3.2, chapter 3), sad + vas is an aṅga with respect to Ṅas simply because the affix Ṅas has been added to sad + vas. In derivations involving vikaraṇas, because we add affix C to base A and then add another affix B between base A and affix C, the process of identifying the aṅga with respect to affix C becomes somewhat complicated, as observed above.

  1. Affix LAṬ occurs after a verbal root when the action is denoted at the current time (vartamāna). ↩︎

  2. Tip-tas-jhi-sip-thas-tha-mib-vas-mas-ta-ātāṁ-jha-thās-āthāṁ-dhvam-iḍ-vahi-mahiṅ. ↩︎

  3. Though the tradition does not explicitly state this, it becomes clear from the derivations we will examine below that such is indeed the case. ↩︎

  4. Affix ŚaP occurs after a verbal root when a sārvadhātuka affix which denotes kartr̥ ‘agent’ follows. ↩︎

  5. Guṇa replaces the iK of a verbal base which ends in the augment pUK or which has a laghu ‘light’ vowel as its penultimate sound when a sārvadhātuka or ārdhadhātuka affix follows. 6 I must clarify that, in my view, the modified version of an aṅga too can be called an aṅga, thanks to 1.1.56 sthānivad ādeśo’nalvidhau, which teaches that the substitute is treated like the substituendum, except when an operation relative to the original sound is to be performed. So, for example, in deva + bhyām, deva is an aṅga with respect to bhyām. By applying 7.3.102 supi ca, we get devā + bhyām. devā too can be called an aṅga with respect to bhyām by 1.1.56 sthānivad ādeśo’nalvidhau. ↩︎