(Previously: Chapter 1, Chapter 2, Chapter 3, Chapter 4, Chapter 5.)

This the longest chapter so far, though Chapter 8 and Chapter 9 we’ll see are even longer. In this chapter Sediyapu presents one of his new theories. There are many things in this book that are perfectly obvious in hindsight, and enormously clarifying, and yet no one in the history of Indian prosody seems to have pointed out earlier.

And there are also a few things that are subjective, as we will see.

Table of contents

Chapter 6: On different kinds of chandogati (Āvarta-gatis)128 0.092 0.198

Let’s start with a footnote, as it spans 4 pages. :-)

[Start of Footnote 48] Practically any words can be sung to a rhythm; we call something a padya only if its syllables imply a rhythm, fit to a rhythm.128 0.297 0.547

Such padya are sama or viṣama gati. (Other to-be-sung compositions, like “pada”s, are always samagati.)128 0.524 0.671

But even many samagati / satāla-bandhas are sung without tāla. [Unfortunately, I know neither these metres nor these tālas, so I can’t understand this part at the moment.]128 0.647 0.928129 0.124 0.813

Something very perceptive, and one of several such remarks in the book:

Two spectrums (spectra?):
(1) harṣotpādaka v/s saumya, gambhīra, prasannate, ēkāgrate
(2) Pay attention to sound v/s Pay attention to meaning.
Laya, especially when druta, pushes the reader closer to the left end of both the above spectra.129 0.807 0.902130 0.130 0.353

This is the reason for traditions of singing vitāla.130 0.329 0.552

IMO this is all fine, but there is an unacknowledged tension with the reason for writing padya (rather than gadya) in the first place. After all, if one very much wants all attention to meaning and none to sound, then writing prose (gadya) would be even better. So this is a matter of balance, of taste and subjectivity, rather than absolutes. (Not that he’s stated it as such; just making it clear for myself…)

But many satāla are sung vitāla and vice-versa (e.g. Vande Mataram); it doesn’t change their basic nature. [End of footnote 48]130 0.534 0.907131 0.343 0.718

Now we can start reading the actual text part of the chapter :-)

Kinds of akṣara-gati and kinds of āvarta-gati

In Chapter 3, we saw all gatis are samagati (having tāla, laya) or not. And in Chapter 5, we saw druta, vilambita, etc., based on the proportion of laghus and gurus. (Orthogonal, and fuzzy – hard to precisely classify or say anything based on that).128 0.188 0.306131 0.105 0.325

Fn 49: But people have tried. Attempts at noticing gati, evidence from name “druta-vilambita”.131 0.710 0.907132 0.523 0.893

I claim that in samagati padyas, can actually say something about drutatva etc. No one has done it before, but let me try. First, some examples.132 0.093 0.496133 0.100 0.187

Group I.133 0.183 0.447

Group II.133 0.441 0.766

Group III.133 0.756 0.911134 0.102 0.258

[Aside: If some day this work is translated out of Kannada, having consistently Sanskrit examples everywhere may be nice…]

When we read them uniformly / naturally, we seem to perceive speed of I < II < III.134 0.252 0.424

You have to experience it for yourself. “idu anubhavaika-vedya” as said above – we’ll encounter this again.

But the reason here cannot be the “more G / more L” classification of Śārṅgadeva. So what is it?134 0.399 0.588

5, 4, 3!134 0.537 0.727

We perceive the repetition sooner/later, so faster/slower.134 0.700 0.878135 0.108 0.165

Fn 50: These are only apparent / perceived / “illusory” gatibheda, not actual change in speed of utterance as in music. [Repetitive IMO, but reread if you didn’t understand.]135 0.258 0.907

Conclusion from what we’ve seen so far: there are two kinds of druta etc: (1) Akṣara-gati-bheda (more Gs or Ls, as described by SRK), (2) Āvarta-gati-bheda, only for layānvita bandhas, being described here for the first time by SKB.135 0.157 0.244136 0.087 0.569

Fn 51 [Maybe this would have even fit in an earlier chapter, but it’s a fair question that may arise here]: Isn’t any samavṛtta layānvita, as it has repetition? E.g. in music we have tāla of duration 30 etc. Answer: Such long āvartas are not perceptible without special musical training; only up to ~8 are. …[Some philosophy about what padya should be]… Otherwise, not just all samavṛttas, but even ardhasama and even almost all padya (except a few like Anuṣṭup) could be called layānvita / said to have repetition/āvarta!136 0.571 0.899137 0.127 0.902138 0.571 0.789

In all layānvitabandhas, both akṣaragati and āvartagati present, but (especially other than satāna-bandhas, i.e. mātrābandhas or vitānavṛttas), the āvartagati is more prominent. [Because the akṣaragati has variations…]136 0.530 0.565138 0.107 0.383

There are some intricacies. See Fn 52 that follows.138 0.368 0.566

Also see the last sentence above; it’s one of the reasons this topic is so difficult and elusive. :-)

[Start of Fn 52.] When there are chandaḥkhaṇḍas that are full of laghus, the akṣaragati’s drutatva is perceptible (whatever the āvartagati).138 0.785 0.878139 0.130 0.501

Similarly, chunks with gurus $\implies$ vilambita akṣaragati perceptible.139 0.498 0.719

ja-gaṇa perceptible.139 0.713 0.821

Even though 4-4, the yati / word-break as G-LGL stands out: pause and jump.139 0.802 0.896140 0.385 0.916

[This would be a perfect time for me to be sceptical, but actually I perceived it…]

See examples of both gatis being apparent.140 0.555 0.681

Another summary: akṣaragati is prominent in case of (1) no laya, (2) something repeated more than once (including satāna-vṛttas), (3) special things like “pluti”, or jagaṇa. (Consider: Śikhariṇī, Mandākrāntā, Śālinī.) Else, āvartagati prominent. [End of Fn 52]140 0.675 0.922

Sounds perfectly correct once stated, but what a fine ear, and what clear thinking, are needed to state these, without getting confused!

Now back to the main text…

Gatis formed from combinations of these āvartagatis

Combinations of our 3, 4, 5 āvartagatis.140 0.116 0.366

4+5 is too long to be clearly perceptible/attractive. And 3+5 is confusable with 4+4, so needs some help from a G after 3, and/or yati. (See Fn 54.)141 0.110 0.463142 0.101 0.409

Fn 53. [To me this seems to be clarifying something that seems obvious, in fact I don’t understand how it could be different, so what is being pointed out. I’m probably missing something. Anyway…] mātrāgaṇa = āvarta = sequence of L/G with total duration $n$ for some $n$. Sanskrit works on prosody consider only $n = 4$, Kannada 3, 4, 5, Prakrit 2 to 6, Urdu up to 7 but with some constraints.141 0.479 0.917

[Start of Fn 54] $3 + 5$ in Rathoddhatā, using G.142 0.580 0.790

Examples of avoiding 4+4 through either G or word-break (yati).142 0.785 0.911143 0.139 0.608

Note: This shows that word breaks matter! A simplistic treatment of prosody says they shouldn’t matter, but I’ve long suspected they do – pleased to be right! (In fact I suspect they do even in Anuṣṭup, but we’ll come to that later.)

Sometimes not strictly followed in recent or lesser works. (And vice-versa: 4+4 metres can occasionally have 3+5.) Might be OK though.143 0.585 0.807

Example. One is 4+4 (2nd chunk of second line); rest are 3+5.143 0.800 0.906144 0.135 0.215

First pāda is 4+4 (madhyāvartagati); rest are 3+5 (druta-vilambitāvartagati).144 0.215 0.490

Some weirdness of “varaṇa” in verses(?) of the “pada” genre. [I’m ignoring for now.]144 0.490 0.903145 0.140 0.259

[Separate footnote, sharing same number] 8s are always confusable: 3+5 could also be 5+3.145 0.255 0.419

Examples (3+5 then 5+3)145 0.413 0.614

The distinctive gati comes not from being breakable as 3+5 or 5+3, but from being not 4+4, because of syllables 4 and 5 being blended by G (or start of word). So let’s call it “sankalita-madhyāvartagati” instead: new name. [End of Fn 54.]145 0.591 0.909

We saw why 4+5 is unlikely, and discussed 3+5. The other one, 3+4, is very common.146 0.100 0.212

Examples.146 0.207 0.307147 0.098 0.257

Fn 55: Someone says some “akṣara vṛttas” are “mātrāvṛttas”; this is not so. [Needless harsh criticism IMO; just a matter of definitions and someone being sloppy…]146 0.320 0.906147 0.652 0.900

A couple of points of note from the above:

…146 0.663 0.728

Most vṛttas are samagati, i.e. layānvita = satāla.146 0.744 0.812

This is another observation that becomes obvious in hindsight on examining the data, yet before Sediyapu Krishna Bhat no one seems to have made. Truly a mark of genius (IIRC something Huxley said about Darwin is similar).

Repeating the end of Fn 54: 3+5 = 5+3 chunks are formed by overriding 4+4, with a G after 3, i.e. occupying 4 & 5 — by blending/bridging (saṃdhāna) of the two 4s.147 0.246 0.637

(…contd) Even when achieved with yati rather than G, this is what is going on. [An innocuous-looking claim tucked away in brackets, but actually quite subtle and may not be easy to understand how 1+1— i.e. word break starting at 4, becomes saṃdhāna.]148 0.107 0.191

Beautiful: just as blended 4+4 is distinctive, so is blended 3+3.148 0.164 0.339

Examples I. [The second group is LLGG; the asterisk is a footnote.]148 0.334 0.544

Examples II. [In (a), Line 1, third group is 4+2, similarly first group of line 3. Typo in second group of line 2: it is LLGLL.]148 0.538 0.911

Examples III.149 0.098 0.414

Same: This gati comes when 3+3 is prevented, by 3&4 being joined by a G (as in I)…149 0.411 0.604

… or with a yati after 2 or 4 (as in II)…149 0.576 0.818

… or… in other ways. [This one is subtler and I didn’t hear it the first time.]149 0.789 0.899150 0.107 0.185151 0.105 0.242

[Read these examples carefully and convince yourself… or not.]151 0.207 0.435

[Start of Fn 56] One may ask: why is GLLG not 3+3? 150 0.194 0.339

Answer: Just listen to it. (Also: “…heccukaḍame…”)150 0.316 0.442

Aside: In mathematics books, sometimes the author writes that something is “obvious” or “trivial”, and sometimes readers get upset when it’s not obvious or trivial to them. I think something similar happens in this book in some places, and sometimes this comes off as a bit frustrating. It is one thing to say “unfortunately, this is hard to explain, and you have to experience it for yourself”. It feels different, possibly insulting, when stated as “If you have any sense at all, you will know this already”. :-) Could this be one of the secondary reasons this great book is not universally adored by everyone who’s read it?

The reason for sounding this way: GL↘ and LG↗, etc. 150 0.417 0.614

★A very subtle and important point [buried within parentheses in a footnote]!150 0.590 0.713

Similar with LLLLG150 0.689 0.899

Example. [End of Fn 56]151 0.693 0.896

Summary / reiteration.151 0.432 0.696

Like the 8 that is not 4+4, this one too requires special circumstances.152 0.106 0.251

And can sometimes accommodate 3+3.153 0.105 0.265

Example of 3+3 intruding (in fact, 5 times out of 14).153 0.255 0.537

Name: Similarly, let’s call it saṃkalita drutāvartagati.153 0.531 0.619

[Start of Fn 57] Many of the {3, 4} metres seem to have additional constraints (the 3 can’t be LG, and the 4 can’t be LGL); so is this really a generic class of metres? Yes it is.152 0.255 0.488

Just is. (Kannada doesn’t have gaṇas starting with LG.)152 0.461 0.714

Example of 3+4152 0.681 0.903

And 4+3. [End of Fn57]153 0.745 0.901

The six kinds of gati from changes in āvarta-gati

Collecting what we saw so far…153 0.606 0.739

3–8, so 6 (main) kinds.154 0.105 0.335

I. Drutāvartagati = 3 154 0.326 0.481

II. Madhyāvartagati = 4154 0.471 0.643

III. Vilambitāvartagati = 5154 0.643 0.793

IV. Druta-madhyāvartagati = {3, 4}154 0.793 0.902155 0.100 0.165

[Fn 58: Example.]155 0.295 0.877

V. Saṃkalita-drutāvartagati = 6 that is not 3+3155 0.159 0.284156 0.105 0.310

[Fn 59: Their prevalance, contra Ti Nam Shri’s Samālokana.]156 0.394 0.909

VI. Saṃkalita-madhyāvartagati = 8 that is not 4+4.156 0.301 0.383157 0.103 0.323

(IMO: From a mathematical point of view, we can quite simply classify these by “fundamental period” — if something has fundamental period 6, then it does not have period 3, etc., so these six classes are simply something like: period 3, 4, 5, 7, 6, 8, and it is not necessary to distinguish the last two. But he does…)

Names: 3, 4, 5, 7 v/s 6, 8.157 0.320 0.558

Fn 60: Ṣaṭpadi (LOL)157 0.750 0.879

Another: 4+3=7. Example.157 0.554 0.739

3+4 and 4+3 tend to cancel each other out. Example of Vasantatilakā!158 0.106 0.329

Fn 61: An example of 4+3 from “pada” genre from Yakṣagāna, needs stretching / shrinking so we’ll ignore for now.158 0.475 0.895

Also: trimūrti (aṃśa) gaṇa not discussed; they need to be read specially, and when done so, fits what we have.158 0.301 0.475159 0.105 0.187

[Start of Fn 62] trimūrti gaṇa, not read as written [I’m ignoring for now…]159 0.206 0.601

More…159 0.577 0.891

More (Yakṣagāna, etc.) (One thing of note: with pauses and so on, any layānvita chandobandha’s gati can be changed — that has no bearing.)160 0.127 0.884

Even more….161 0.113 0.800

Sounding vilambita161 0.776 0.887162 0.635 0.756

More. [End of Fn 62.]162 0.749 0.895

Variation in layānvita bandhas: endless variety

Variation (in the actual concrete LG pattern, and in yati) leads to great variety.162 0.106 0.367

Two examples of Udgatā162 0.367 0.634163 0.111 0.293

Aside: I used to struggle reading Udgatā, and now suddenly, magically, after having read so far into the book, it sounds good. That is after all the purpose of undertaking this study, to be able to appreciate beauty more.

Note: the terms samavṛtta and viṣamavṛtta are just about how broken into pādas, they are not about the gati of the composition.163 0.261 0.410

Notation.163 0.381 0.547

In Udgatā, some are 3+5 and some are 4+4.163 0.520 0.618

The two examples differ in first āvarta of 2nd half, because of yati: 3+5 [IMO 5+3 to be precise, but the point is the same] v/s 4+4.163 0.589 0.891164 0.099 0.168

On the gati of kanda-padya

5 kandapadyas from Kavi-rāja-mārga164 0.168 0.898

Kandapadya is the majority of Haḷegannaḍa. Constraints on 6th gaṇa, and 1357 can’t be jagaṇa. [IMO “can’t be jagaṇa” is equivalent to “can break as 2+2”… but maybe that’s not the intention.]165 0.101 0.329

Many variations manifest.165 0.303 0.399

Apparent even in the five examples just given.165 0.373 0.445

Example of (a): The āvartagati recedes to background. (b) seems to be of āvarta 6. In (c), the 4-gati remains strong [?] In (a) and (c), the akṣaragati are more prominent.165 0.416 0.907166 0.099 0.313

Fn 63: If you ignore the meaning & yati, and read by gaṇas, then the simple 4-gati stands out.166 0.361 0.652

Fn 63: Details of the further constraints on 6, 8 or 9 etc.166 0.631 0.905

There are definitely things in this chapter I have not understood. But what I’ve understood is already large and highly valuable. :-)