Skip to content
May 2, 2012 / A

Panini and Kalidasa

For a change of pace I thought I would linger on the beautiful words of Kalidasa’s Meghaduta:

कश्चित्कान्ताविरहगुरुणा स्वाधिकारप्रमत्तः
शापेनास्तंगमितमहिमा वर्षभोग्येण भर्तुः ।
यक्षश्चक्रे जनकतनयास्नानपुण्योदकेषु
स्निग्धच्छायातरुषु वसतिं रामगिर्याश्रमेषु ॥ १ ॥

Sanskrit poetry is notoriously difficult to translate, so I will not try to tackle this verse. But I’m partial to Ryder’s rendition:

On Rama’s shady peak where hermits roam,
Mid streams by Sita’s bathing sanctified,
An erring Yaksha made his hapless home,
Doomed by his master humbly to abide,
And spend a long, long year of absence from his bride.

Like most old poetry, the Meghaduta has some variant readings and extra verses. This verse only has one: in the last part of the first line. One reading has स्वाधिकारप्रमत्तः . The other has स्वाधिकारात् प्रमत्तः . Both roughly mean “negligent of (one’s) duty.” But which one did Kalidasa actually use?

In textual criticism there are notions of lectio difficilior and lectio facilior, the “harder reading” and the “easier reading.” The idea is that as texts pass through the hands of scribes through the centuries, they often change in a few places. These changes occur for many reasons, but a small variance like this indicates that one reading was made as a correction or improvement on the other. Thus we can infer here that the more “difficult” reading is probably the correct one.

So which reading is correct here? Those of you who are familiar with Sanskrit literature will be well aware of how common compounds are. So it seems reasonable that स्वाधिकारप्रमत्तः is the easier reading and therefore that स्वाधिकारात् प्रमत्तः is the more difficult one. Thus स्वाधिकारात् प्रमत्तः is probably the original reading.

Unfortunately for us, we made a bad conclusion: स्वाधिकारप्रमत्तः is almost certainly the original reading. And we can demonstrate this with our knowledge of Panini.

Consider स्वाधिकारात्. It is in what is called the “ablative case,” or “case 5” for short. In Sanskrit this case is called पञ्चमी (“the fifth”), and it is usually added in the sense of अपादान:

2.3.28 अपादाने पञ्चमी
The fifth case is added (to the end of a प्रातिपदिक “bare stem”) in the sense of अपादान.

But what is अपादान? We see this defined in 1.4.24:

1.4.24 ध्रुवमपाये ऽपादानम्
When moving away from something, the fixed point is called अपादान.

So far so good. Now according to a वार्त्तिक (“comment”) on this rule, we have the following:

We must also include (verbal bases in the sense of) disgust, cessation, and neglect.

Patanjali says this is implied by the rule already, as long as we take the rule metaphorically. But in either case we find that the expression स्वाधिकारात् प्रमत्तः is grammatically valid. We even have प्रमाद in the वार्त्तिक itself!

Now things get interesting.

Chapter 2.1 defines the various rules of compound formation. In the तत्पुरुष section, it has only 3 rules on compound formation with a word in case 5:

2.1.37 पञ्चमी भयेन
A word in case 5 (is preferably compounded with the case-inflected word) भय (and the compound conveying the same meaning as the corresponding word group is called तत्पुरुष);
2.1.38 अपेतापोढमुक्तपतितापत्रस्त्रैरल्पशः
likewise with अपेत, अपोढ, मुक्त, पतित, and अपत्रस्त in a few instances,
2.1.39 स्तोकान्तिकदूरार्थकृच्छ्राणि क्तेन
and when words meaning स्तोक “a little,” अन्तिक “near,” or दूर, “far,” or the word कृच्छ्र, (are in case 5 and are compounded with) a word ending in क्त (the past participle suffix).

So we have compounds like चौरभयम् “fear of thieves,” सुखापेतः  “parted from pleasure,” दूरागतः “come from far,” and so on. The वार्त्तिकs allow a few other words, but not many.

Under no other circumstances can a word in case 5 be in a तत्पुरुष compound. स्वाधिकारप्रमत्तः is clearly a तत्पुरुष compound. And it clearly follows none of these rules. Thus we make a big discovery: स्वाधिकारप्रमत्तः cannot be produced by the Ashtadhyayi, at least in the clearly intended sense of “negligent of one’s duty.” The word is ungrammatical.

Kalidasa wrote for an educated audience that knew Sanskrit. His commentators also knew Sanskrit, and scribes probably knew Sanskrit to some extent. The Meghaduta is one of Kalidasa’s most admired poems. So it is very likely that an educated reader would try to correct the poem and use the proper Paninian form: स्वाधिकारात् प्रमत्तः. Thus this is the easier reading. स्वाधिकारप्रमत्तः is more difficult. And by basic principles of textual criticism, it is most likely that स्वाधिकारप्रमत्तः was what Kalidasa wrote centuries ago.

But why would Kalidasa do such a thing? It is possible he slipped up, but I think that’s unlikely. The end of the first line creates a lovely effect when written as स्वाधिकारप्रमत्तः. rap-ra-mat-tarap-ra-mat-ta. It sounds soft and much more beautiful than the harshness of rāt-pra-mat-taḥ that we have otherwise.

So I applaud Kalidasa’s small audacity, for just as the Meghaduta‘s hero neglected his duty, Kalidasa himself neglected grammatical fidelity, all for the sake of creating beautiful poetry.



Leave a Comment
  1. S / May 2 2012 1:48 am

    This is a great blog; thanks for the posts!

    Excellent analysis here. Assuming it is indeed the case that स्वाधिकारप्रमत्तः got turned into स्वाधिकारात् प्रमत्तः, I have another (less possible, more outlandish) theory for why it may have happened:

    Background: The Mandakranta metre, in which the Meghaduta is written, requires a ‘guru’ in that position. By the laws of prosody, the first syllable in रप्र is a guru since it’s followed by a conjunct consonant (two consonants in a row). However, there are instances of some controversy and confusion by medieval times, in which some authors make an exception for conjunct consonants like प्र, saying that since they’re so easy to pronounce (don’t take the time of two matras to pronounce, for them), they can be counted as not elongating the previous syllable (turning it into a guru).

    In other words, depending on how smoothly you pronounce प्र, you may find yourself with a laghu at that point, and someone may have preferred स्वाधिकारात् प्रमत्तः to re-elongate it.

    • A / May 2 2012 1:21 pm

      An interesting insight. I’d like to consider it from two angles.

      First, from the perspective of commentary. I am familiar with two commentaries on the Meghaduta: one by Mallinatha and one by Dakshinavartanatha. Dakshinavartanatha’s is older, according to T. Ganapati Sastri, and I see no reason to disagree.

      Mallinatha’s commentary is straightforward:

      स्वाधिकारात् स्वनियोगात् प्रमत्तः अनवहितः ॥ “प्रमादोऽनवधानता” इत्यमरः । “जुगुप्साविरामप्रमादार्थानामुपसंख्यानम्” इत्यपानत्वम् । तस्मात्पञ्चमी ॥ अत एवापराधाद्धेतोः ।

      स्वाधिकारात् “of one’s own station” प्रमत्तः “negligent” . “प्रमादोऽनवधानता” (“neglect and inattention”), according to Amara. “We must also include (verbal bases in the sense of) disgust, cessation, and neglect” for अपादान. Thus the fifth case (is used). Hence (the phrase specifies an action) on account of some transgression.

      But Dakshinavartanatha’s commentary is more interesting:

      कीदृशो यक्षः, स्वाधिकारात् प्रमत्तः स्वस्याधिकारात् स्थानात् प्रमत्तः भ्रष्टोऽनवहितः । “प्रमादोऽनवधानता” इत्यमरः । “जुगुप्साविरामप्रमादार्थानाम् — ” (वा. १.४.२४) इत्यपादाने पञ्चमी । स्वाधिकारप्रमत्त इति पाठे समासः कृच्छ्रलभ्यः ।

      What sort of Yaksha is this? One स्वाधिकारात् प्रमत्तः , or प्रमत्तः (“inattentive”, “negligent”) of one’s own अधिकार (“station.”) “प्रमादोऽनवधानता” (“neglect and inattention”), according to Amara. “We must also include (verbal bases in the sense of) disgust, cessation, and neglect,” hence the use of the fifth case in the sense of अपादान. स्वाधिकारप्रमत्तः, in (another?) reading, is a compound that can be allowed (only) with difficulty.

      Given the high degree of similarity between the two explanations, I imagine that Mallinatha read Dakshinavartanatha’s commentary and dismissed स्वाधिकारप्रमत्तः on grammatical grounds. This would be in keeping with a grammatical preference. It is clear that the reading was bothersome to Dakshinavartanatha, and it lends evidence to a grammatical basis for स्वाधिकारात् प्रमत्तः.

      (I just used this as an excuse to go through commentaries. I love commentaries for some reason.)

      Second, on metrical grounds in the text itself. I’ve gone through the first 5 verses of the Meghaduta and found all instances where प्र is ambiguously heavy, given the exception you mention:

      विप्रयुक्तः स कामी
      आषाढस्य प्रथमदिवसे
      कण्ठाश्लेषप्रणयिनि जने
      स प्रत्यग्रैः
      कामार्ता हि प्रकृतिकृपणाः

      I might have missed some instances, but I think this is a good set of evidence to examine. Among these instances, I do find the variant reading संप्रत्यग्रैः , which would address the issue you mention. But the others don’t have any variant readings I can find. (Disclaimer: my copy of the poem is missing the variant readings for verses 2 and 5). And we would expect more, especially since some of these can be fixed somewhat easily:

      कण्ठाश्लेषं प्रणयिनि
      कामार्ता वै प्रकृतिकृपणाः

      (वप्रक्रीडा is नित्यसमास by 2.2.17 नित्यं क्रीडाजीविकयोः “always in the sense of play or livelihood” (?), and रिक्तप्रकोष्ठ can’t be split because this is a बहुव्रीहि.)

      So I think it is unlikely that the exception for प्र was a motivating factor here.

      As an aside, I’m surprised to read that there is such an exception at all. Something similar happens in Greek all the time, but I thought Sanskrit metrical rules were well established enough that there would be no doubt. I’m not familiar with छन्दस् beyond basic concepts like वृत्त and गण, but even in Panini we have a pretty unambiguous statement:

      1.1.7 हलोऽनन्तराः संयोगः
      Consecutive consonants are called संयोग “conjunct.”
      1.2.27 ऊकालोज्झ्रस्वदीर्घप्लुतः
      Vowels as long as उ are called ह्रस्व “short,” those as long as ऊ are called दीर्घ् “long,” and those as long as ऊ३ are called प्लुत “prolated.”
      1.4.10 ह्रस्वं लघु
      Short vowels are लघु.
      1.4.11 संयोगे गुरु
      In front of conjuncts they are called गुरु,
      1.4.12 दीर्घं च
      and likewise if they are long.

      So I wonder what happened to cause confusion in medieval times.

      • S / May 2 2012 7:31 pm

        Awesome. There goes my theory, then. 🙂

        Yes, reading commentaries is great practice; I love it too! The पाठान्तर-remark of Dakshinavartana is interesting. I think explanations of grammatical points are inevitably similar though (the same Panini sutras to quote, etc.), so I’m not convinced by just this one instance that one followed the other, but it’s certainly possible.

        And I think I went too far in saying there was “controversy and confusion”; rather what I vaguely remember seeing somewhere is a disdainful remark like “Some people who don’t know grammar think that consonants like प्र are not heavy; this is totally wrong and the rules of Sanskrit prosody are clear enough in this matter so I don’t know what their confusion is.” 🙂


  1. Is this rule in the Ashtadhyayi or not? « अवग्रह

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: