Appayya Dīkṣita on rūpaka (“metaphor”)

Among the works of the great 16th century savant Appayya Dīkṣita is the Kuvalayānanda (“Joy of the Water-lily”), which became the standard introductory textbook on figures of speech throughout India, particularly in the South. The Kuvalayānanda focuses entirely on arthâlaṅkāras, “ornaments of meaning”. My non-existent knowledge of Western rhetoric prevents me from translating the term precisely, but I suspect it may correspond to trope.

Prof. Yigal Bronner has a series of articles on Appayya Dīkṣita’s prodigious œuvre in literary theory, of which “Back to the Future: Appayya Diksita's Kuvalayananda and the Rewriting of Sanskrit Poetics” describes the structure of the Kuvalayānanda. Specifically, the text is organized in three layers: (1) kārikās, around 170 easily memorizable verses in the anuṣṭubh meter that define and exemplify exactly 100 alaṅkāras; (2) examples of these alaṅkāras drawn from kāvya; and (3) an auto-commentary of sorts, in which Appayya Dīkṣita explains his motivation for defining the alaṅkāra in that fashion and sets it into its context of over a millennium of alaṅkāra-śāstra. I shall draw solely from layer (1), the kārikās, here.

For now, I want to focus on just one trope, rūpaka, usually translated “metaphor”. Its definition takes up one verse; examples of its subtypes each take up half a verse. I’m not going to translate the definition literally, because the compactness of Sanskrit technical writing does not render well in English. In what follows, [the stuff inside brackets] isn’t part of the Kuvalayānanda itself.

[rūpakam]

viṣayayy-abheda-tādrūpya-rañjanaṃ viṣayasya yat |
rūpakaṃ tat tridhâdhikya-nyūnatvânubhayôktibhiḥ ||


[1 abheda-rūpakam]
[1a anubhayâbheda-rūpakam]
ayaṃ hi dhūr-jaṭiḥ sākṣād yena dagdhāḥ puraḥ kṣaṇāt |

[1b nyūnâbheda-rūpakam]
ayam āste vinā śambhus tārttīyīkaṃ vilocanam |

[1c adhikâbheda-rūpakam]
śambhur viśvam avaty eṣa svīkṛtya sama-dṛṣṭitām |

[2 tādrūpya-rūpakam]
[2a anubhaya-tādrūpya-rūpakam]
asyā mukhêndunā labdhe netrânande kim indunā |

[2b nyūna-tādrūpya-rūpakam]
sādhvîyam aparā lakṣmīr asudhā-sāgarôditā |

[2c adhika-tādrūpya-rūpakam]
ayaṃ kalaṅkinaś candrān mukha-candro ’tiricyate ||

“Metaphor”

The trope rūpaka is a portrayal of the subject’s (1) identity with, or (2) similarity in form to, another object. Each of these two can be broken down into three subtypes, depending on whether the portrayal conveys the subject’s (c) superiority over, or (b) inferiority to, the object, or (a) remains indifferent to the nature of the relationship between the subject and the object. Thus:

1. Identity of the subject with some object [where the viṣaya is some unnamed king, and the viṣayin is Śiva]
1. Indifference: “He indeed is Śiva of matted locks, by whom the [three] cities were instantaneously incinerated.”
2. Inferiority: “He sits, Śambhu (Śiva) without a third eye.”
3. Superiority: “This Śambhu (Śiva) protects the world, having accepted evenness of sight.”
2. Similarity of form of subject and object
1. Indifference: “My eyes are delighted by her face-moon; what need have I of the moon?”
2. Inferiority: “She is a chaste woman, a second Lakṣmī not arisen from the Ocean.”
3. Superiority: “This face-moon surpasses the stained moon.”

One major question emerges: how are abheda and tādrūpya different from each other? The kārikā does not make this explicit, but the terminology and the examples seem to indicate that with abheda, the viṣaya and the viṣayin are actually asserted to be the same thing, whereas with tādrūpya they are seen as two different things. Perhaps a first-order approximation would be to consider abheda as asserting numerical identity between viṣaya and viṣayin, and tādrūpya as asserting qualitative identity (i.e., treating them both as instances of the same class). Furthermore, since it is possible to have two instances of the same class that have different properties (if your model permits object-level methods, like Ruby does), this allows us to easily accommodate relationships of superiority and inferiority for tādrūpya.


However, this raises a serious further problem for subtyping abheda. If viṣayin and viṣaya are numerically identical, then it makes no sense to talk about superiority or inferiority: something cannot be both numerically identical to something else and still be superior or inferior to it. This would mean that we would be left with just four kinds of rūpaka: abheda, which would be a direct assertion of numerical identity, and the three subtypes of tādrūpya.


Now, I don’t know Appayya Dīkṣita’s response to this objection; perhaps he does address it somewhere. One thing is clear: he would reject the idea that there are no subtypes of abheda, if only because he supplies three examples of abheda-rūpakam!


So what are some ways to defend the idea that abheda does have three subtypes?

• We could argue that abheda isn’t really about numerical identity at all, or at least that to sharply insist on such an ontological claim in a text on literary tropes is to entirely miss the point. Fair enough, but somewhat unsatisfying to me.
• We could also argue that what distinguishes the superior and inferior abheda-rūpakam from the superior and inferior tādrūpya-rūpaka is that the former does begin with an assertion of identity, or of non-difference, which is later qualified. In the case of tādrūpya, there is not even the pretence that the viṣaya and the viṣayin are the same, only the claim that they have the same form. But in abheda, there is, at least at the very beginning, an assertion of identity, which is then subsequently qualified by identifying a difference between viṣaya and viṣayin. And even then, there is never an explicit statement that separates the viṣaya from the viṣayin. Merely stating that “Śambhu has accepted evenness of sight” does not mean that the king is asserted as being different from Śiva.