Skip to main content

Full text of "Studies in Bhāsa"

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 

V. S. Sfkthankae 


(Continued from JAOS 41. 107 ff.) 

III. On the relationship between the Carudatta and the 
Mrcchakatika. 1 

The close corbespondence between the anonymous fragment 2 
Carudatta and the celebrated Mrcchakatika, 3 attributed to 
King Sudraka, inevitably necessitates the assumption of a genetic 
relationship, and indisputably excludes the possibility of inde- 
pendent origin. 

It is commonly taken for granted 4 that the Carudatta is the 
original of the Mrcchakatika, a relation which does not, however, 
necessarily and immediately follow from the terseness or brevity 
of one, nor from (what amounts to the same thing) the length 
and prolixity of the other; for, in adaptation, abridgment is as 
common and natural a determining principle as amplification. 5 
In view of the intrinsic importance of the question, it seemed, 

' A paper presented at the One Hundred Thirty-third Meeting (Balti- 
more, 1921) of the Amer. Or. Soc, under the title: 'The Carudatta and the 
Mrcchakatika: their mutual relationship'. 

J See thereon my article, ' "Charudatta" — A Fragment' in the Quarterly 
Journal of the Mythic Society (Bangalore), 1919. 

3 Ed. N. B. Godabole, Bombay, 1896. 

* For instance, Ganapati Sastri in the Introduction to his editions of 
the Svapnavasavadatta (p. xxxviii), and the Carudatta (p. i); Lindenau, 
Bhdsa-Studien (Leipzig, 1918), p. 11; and Barnett (hesitatingly) Bulletin 
of the School of Oriental Studies, vol. I, part III (1920), pp. 35 ff. 

5 Some attempt has already been made in India to discredit the 
authenticity of the Carudatta; see, for instance, Rangacarya Raddl, 
Vividha-jHdna-vistara (Bombay), 1916, and P. V. Kane, ibid. 1920; Bhatta- 
natha Svamin, Indian Antiquary, vol. 45, pp. 189ff. 

60 V. S. Sukthankar 

therefore, desirable to undertake an unbiased and exhaustive 
investigation so as to remove (if possible) the haze of uncer- 
tainty surrounding the subject. 

Only the resemblances between the two plays appear hitherto 
to have attracted any attention; the differences between them 
are, however, equally remarkable and much more instructive. 
A careful comparative study of the two versions produces 
highly valuable text-critical results, which help further the 
understanding of the plays and throw unexpected light on the 
subject of our inquiry. 

Regarding their relationship there are only two logical 
possibilities: either, one of the plays has formed directly the 
basis of the other, or else both of them are to be traced to a 
common source. In the former case we are called upon to 
answer the question, which of the two plays is the original; 
in the latter, which of them is closer to the original. 

We cannot be too careful in deciding what is original and 
what is not. The original may have been concise and well- 
proportioned, and later clumsy attempts at improvement may 
have introduced digressions, tiresome repetitions and insipid 
elaborations; on the other hand, the original may have been 
prolix and loose, and subsequent revision may have pruned 
away the redundancies. Again, one may feel justified in assuming 
that the inaccuracies and inconsistencies of the original would 
be corrected in a later revised version; but one must also 
readily concede that a popular dramatic text like the Mrccha- 
katika, after it had been written down, during its migrations 
through centuries over such a vast territory as India, may 
have undergone occasional distortion and corruption. 

Every change, however minute, presupposes a cause; even 
the worst distortion was ushered in with the best of intentions, 
and though it may not always be possible to trace a given 
change to its proper cause, we are safe in assuming that in a 
limited number of favorable instances the intrinsic character 
of the passages under consideration may spontaneously suggest 
the cause for the change, and readily supply a clue to the 
relative priority and posteriority of two variations. In isolated 

6 See particularly Ganapati Sastri, SvapnavSsavadatta, Introduction, 
pp. xxxviii-xlii. 

Studies in Bhdsa 61 

instances we could say no more than that the change in a 
certain direction appears more probable than a change in the 
contrary direction. But the cumulative force of a sufficient 
number of analogous instances, all supporting one aspect of 
the question, would amply justify our giving precedence to that 
particular alternative and treating it as a working hypothesis. 
The problem, therefore, before us is to collect such instances, 
in which the motive for the change is directly perceptible 
and capable of objective verification. The cumulative effect of 
the indications of these scattered traces should not fail to give 
us the correct perspective. This digression was necessary in 
order to explain the methodology underlying the present 

The textual differences between the two versions comprise a 
large mass of details of varying importance. The selection 
presented below, though conditioned on the one hand by the 
requirements of the present inquiry, is by no means exhaustive; 
for lack of space, only a few typical examples have been singled 
out for discussion. 

A Selection oe Significant Textual Differences. 

"We shall now proceed to a discussion of the textual variations, 
roughly classified here under four headings: 1. Technique; 
2. Prakrit; 3. Versification; and 4. Dramatic incident. 

1. Technique. 

In point of technique the Carudatta differs from the Mrccha- 
katika (as from other classical dramas) in two striking parti- 
culars. In the first place, the usual nandi is missing, in both 
the available manuscripts of the Carudatta; in the second place, 
there is no reference to the name of the author or the play 
in the sthapana, which does not contain even the usual address 
to the audience. 

The Mrcchakatika, as is well known, begins with two bene- 
dictory verses; the name of the play is announced in the opening 
words of the sutradhara; then follow five verses which allude 
to the play, the playwright, 7 and other details not directly 
connected with the action. 

7 The verses in the prologue which refer to the death of the alleged 

62 V. S. Sukthankar 

Elsewhere 8 I have tried to show that the Carudatta is a 
fragment. I hold, accordingly, that we should not be justified 
in basing our conclusions regarding the technique of termi- 
nation on the data of the fragment preserved. 

Worth noting appears to be the fact that in the stage 
directions of the Carudatta, the hero is never called by his 
name or his rank, but merely by the character of the role he 
plays, nayaka. Professor Luders 9 has already drawn attention 
to two other instances of this usage (if it may be called a 
usage), namely, a drama belonging to the Turfan fragments, 
and the play Nagananda attributed to Harsa. Prof. Luders 
sees in it an archaism intentionally copied by the author of 
of the Nagananda. At present we can, it seems to me, do 
nothing more than record this third instance of its occurrence 
in a play of uncertain age and authorship. 

2. Prakrit. 

In the first article of this series, it was shown in a general 
way that the Prakrit of the whole group of plays under 
consideration was more archaic than the Prakrit of the classical 
plays. 10 This statement holds good also in the particular 
case of the Carudatta and the Mrcchakatika. A comparison 
of parallel passages in the two plays shows that the Mrccha- 
katika invariably contains Middle-Prakrit 11 forms in place 
of the Old-Prakrit forms of the Carudatta. Here are the 

The Absolutive of the roots gam and kr. Caru. has the 
Old-Prakrit gacchia and karia (kalia): Mrccha. gadua and 
kadua. Of. in particular Caru. 1 geham gacchia janami with 
the corresponding passage, Mrccha. 7 geham gadua janami. 
The form gadua, which never occurs in the Caru., is used uni- 
formly in the Mrccha. — For the absolutive of kr; 12 karia 

author are palpably later additions. This self-evident fact does not, 
however, necessarily justify the assumption that there was no reference 
whatsoever to the author in the prologue of the original draft. 

8 See above, footnote 2. 

9 Bruchstiicke Buddhistischer Dramm (Kleinere Sanskrit- Texte, Heft I), 
Berlin, 1911, p. 26. 

io Above, vol.40, pp. 248 ff. >' Luders, op. cit, p. 62. 

" See above, vol.40, p. 254. 

Studies in Bhasa 63 

(Sauraseni) Caru. 46, kalia (Magadhi) Caru. 23: kadua (&au- 
raseni and Magadhi) Mrccha. 53, 212, 213, etc. In the Caru. 
kadua never occurs; conversely karia is never met with in the 

Pronoun of the 1st Person; nom. sing. Caru. 23 we have 
the Old-Magadhi ahake 13 (but never hage or hagge): Mrccha. 
(passim) hag(g)e (but never ahake). Noteworthy is the following 
correspondence. Caru. I. 12 c aham tumam ganhia: Mrccha. 
I. 29 c ese hage genhia. — Nom. plu. Caru. 49. has the Old- 
Prakrit vaam: u Mrccha. (passim) amhe. The form amhe (nom. 
plu.) is never met with in the Caru., and conversely vaam 
never occurs in the Mrccha. 

Pronoun of the 2nd Person; nom. sing. Caru. (passim) we 
have Old-Prakrit tuvam: 16 Mrccha. (passim) tumam. Cf. 
especially Caru. 34 kirn tuvam, etc., with the corresponding 
passage Mrccha. 79 hanje tumam mae saha, etc. — Gen. sing. 
Caru. uniformly tava: 16 Mrccha. sometimes tuha. Cf. in parti- 
cular Caru. 25 tava geham pavittha with Mrccha. 59 tuha geham 

The Neuter plu. of nom. and ace. of thematic stems ends in 
the Caru. invariably in -ani {-ani in the Asvaghosa fragments): 
in the Mrccha. it ends in -dim. 

Treatment of the assimilated conjunct. Retained in Caru. 16 
dissadi" (as in the Turf an fragments): simplified in Mrccha. 

13 See above, vol. 40, p. 253. Dr. Truman Michelson has drawn my 
attention to an article of his (Indogermanische Forschungen, vol. 23, p. 129) 
in which he points out that the Magadhi ahake occurs several times in 
the DevanSgarf recension of the Sakuntala. The paragraph on this word 
in my article cited above needs modification in view of this fact. The 
statement that ahake is archaic is none the less correct. 

" See above, vol. 40, p. 258. 

15 See above, vol. 40, p. 257. In the references under no. 9 the last item 
'Caru. 2 (Natl)' is a mistake. Here tuvam is used for the ace. sing., and 
not for the nom. sing, as implied. Accordingly, on the same page, in 1. 6 
from bottom, read 'thrice' instead of 'twice', and add this instance. Caru. 
instances of tuvam (nom. sing.) are Caru. 34 (Ganika), 47 (Cetl), etc. 

'6 See above, vol. 40, p. 257. 

" See above, vol. 40, p. 258. — The form dis-, with the simplified conjunct, 
is met with on the same page (Caru. 16), spoken by the same character, 

64 V. S. Sukthankar 

41 disanti. The root-form diSs- (diss-) is never met with in the 
Mrccha., which shows uniformly dis- (dis-). 

Vocabulary. Caru. uniformly geha (Skt. grha): Mrccha. 39 
ghala. Cf. especially Caru. 16 edam tassa geham with Mrccha. 39 
vamado tassa ghalam. — The Old Prakrit affirmative particle 
dma, ls which occurs in Pali and the Turfan fragments and 
which figures so conspicuously in Caru. (e.g. pp. 4, 20, 64, etc.), 
is never met with in the Mrccha. — There is one other thing 
to be noted about the difference in the vocabulary of the two 
versions. While the Mrccha. contains a number of Desi words 
(not found in the Caru.), the vocabulary of the Caru. consists 
notably of pure tatsamas and tadbhavas. Here follow some 
of the Desi words which occur in the Mrccha. Mrccha. 17 
chivia, 'having touched', from root chiv (Hem. 4. 182) with the 
reflexes in the Tertiary Pkts., Hindi chuna, Marathi sivane, 
'to touch'; Mrccha. 104 dhakkehi, 'shut', from dhakkai, dhakkei, 
traced by Pischel (Grammatik 221) to a root *sthak, with 
reflexes in the Tertiary Pkts., Hindi dhdknd, Marathi dhdkne, 
'to cover'; Mrccha. 134 uddhehi, 'open', for which in the corre- 
sponding passage of the Caru. (p. 19) we have a tadbhava of 
the root apa + vr, 19 and which for that reason is particularly 
worthy of note; Mrccha. 207 karatta-daini, 'malevolent ogress' 
(cf. Marathi kdratd, a term of abuse, and ddkin, 'ogress'). 

3. Versification. 

In the verses common to the two plays the Mrcchakatika 
almost always offers better readings, of which a few are cited 

For Caru. I. 3 b yathdndhakarad iva dipadarsanam, we have 
Mrccha. I 10b, ghandndhakdresv iva, etc., in which ghana- is 
substituted for the tautologous yatha. 

Similarly, instead of the Prakrit line Caru. I. 10 b jahd 
sigdli via fciikkulehi, containing the same fault, we have Mrccha. I. 
28 b vane siMi via kukkulehim, in which vane takes the place 
of jahd. 

18 See above, vol. 40, p. 254. 

19 The text reading is avavuda, imp. 2nd sing., which is evidently 
incorrect. What the correct form should be I am unable to say. The 
initial letters avavu of the word show unmistakably that the root is 
apa + vr. 

Studies in Bhasa 65 

For Caru. I. 3 c yo yati dasam daridratam, we have Mrccha 
I. 10 c yo yati naro daridratam. It is correct to say dasam 
daridrdm, but dasam daridratam is clumsy, to say the least. 

Caru. I. 23 a begins esa hi vasu; instead, we have Mrccha. 
I. 41a esa si vasu. The si which takes the place of hi eli- 
minates the expletive hi, and adds moreover another sibilant 
to the row of alliterating syllables. In the same verse, for 
kujdhi kandahi of the Caru., we have akkosa vikkosa in the 
Mrccha., which serves better the purpose of the anuprasa, the 
dominating alamkara of this verse. Similarly in d, instead of 
mahessalam of the Caru., we have sambhum sivam in the 
Mrccha., which latter reading contains an additional sibilant 
as well as a pleonasm. 20 These are minor details, but they 
all tend in the same direction. 

For Caru. I. 25 a akamd hriyate 'smabhih, we have Mrccha. 
I. 44 a sakamdnvisyate 'smabhih. The reason for the change is 
not obvious, as in the foregoing instances. But a closer exa- 
mination of the context will show that the reading of the 
Mrccha. marks a distinct improvement, in so far as it implies 
a more minute analysis of character. In the Caru. the ingenuous 
Vi|;a inculpates Sakara and himself by admitting that they 
were engaged in carrying away forcibly an unwilling maiden. 
In the Mrccha. the artful Vita, readily inventing a plausible 
lie and explaining that they were following a girl who was 
willing, offers undoubtedly a much better excuse. 

Caru. I 29a describes the moon as klinnakharjurapdndu, 
'pale as the moistened fruit of the date': Mrccha. I. 57a has 
kdminigandapdndu, 'pale as a maiden's cheek'. The former is 
original and naive, the latter polished but hackneyed; the latter 
harmonizes better with the sentiment of srngara which pervades 
the last scene of the first act, and is more in keeping with 
the tradition of the later enervated rasa theory. 

For Caru. III. 3d visdnakotiva nimajjamdnd, 'like the tip 
of a tusk sinking in the water', the Mrccha. (III. 7d) has 
tiksnam visandgram ivavasistam, 'like the sharp tip of a tusk 
that alone remains visible'. As far as the sense goes there is 
not much to choose between them; but the line from the Caru. 

J0 According to Lalla Dikshita, commentator of the Mrcchakatika : 
vyarthaikdrtham apartham bhavati hi vacanam idkarasya (Mrccha. 28). 

5 JAOS 42 

66 V. S. Sukthankar 

contains one serious defect. In classical Skt. the root ni-majj 
is used exclusively with Paras, terminations; nimajjamana is, 
in other words, nothing less than a gross grammatical blunder. 21 

With Caru. III. 6 b saury am na karkasyata, cf. Mrccha. III. 
12b cauryam na sauryam hi tat. karkasyata of the Caru. is 
an anomalous word, being a double abstract formation. The 
Mrccha. eliminates this anomaly by substituting instead caurya, 
which, incidentally, rhymes with the succeeding saurya. 

These few instances 22 must suffice to illustrate the statement 
made above, that the Mrccha. verses are largely free from the 
flaws of the corresponding verses of the Caru. It should, 
however, be remarked that in a vast number of cases it is not 
possible to assign an adequate reason for the change: the 
different readings appear to be just arbitrary variations. 

4. Dramatic Incident. 

The Mrcchakatika shows a marked improvement in the 
selection and arrangement of the incidents of the action. 

The action of the Carudatta begins with a soliloquy of the 
Vidusaka followed by a lengthy dialogue between the Nayaka 
and the Vidusaka. The hero is conversing with his friend, 
deploring his poverty. This dialogue is brought to an abrupt 
end by the scene introducing Vasantasena, who appears on 
the street outside pursued by the $akara and the Vita (Caru. 10). 
In the Mrcchakatika (p. 25) the abruptness of the change of 
scene is skillfully avoided by the addition of the following 
words placed in the mouth of Carudatta: 

bhavatu \ tistha tavat \ aham samddhim nirvartayami, 
'Very well. Wait awhile and I will finish my meditation.' 
These words of Carudatta serve admirably to adjust the time 
relation of the different events. The playwright here unmista- 
kably indicates that the succeeding scene, which introduces the 
offers of love by Sakara, their indignant rejection by Vasan- 
tasena, and her subsequent escape, develops during Carudatta's 

21 Similar solecisms, met with in other dramas of this group, are 
discussed by me in the second article of the series (above, vol. 41, 
pp. 121 ff.). 

22 It may be remarked that there are no verses in the second act of 
the Carudatta, and only seven in the fourth act. 

Studies in Ehasa 67 

samadhi. Furthermore, as indicated by the subsequent words 
of Carudatta (Mrccha. 43): vayasya samcuptaja/po 'swi, 'Friend, 
my meditation is over', Vasantasena's reaching the door of 
Carudatta's house coincides exactly in point of time with the 
emergence of Carudatta from his samadhi. The words of 
Carudatta quoted above, which serve to link together these 
various groups of incidents, are missing in the Carudatta. 

Here is another example. In the fourth act of the Carudatta 

(p. 72), Sajjalaka comes to the house of the Ganika to buy 

Madanika's freedom. He stands outside the house and calls 

out for Madanika. Madanika, who is waiting on the heroine, 

hears him and, seeing that her mistress is musing on other 

things, slips away and joins Sajjalaka. The defect of this 

arrangement is obvious: it is inconsistent and illogical. With 

stolen goods in his possession Sajjalaka sneaks to the house 

of the heroine with the object of secretly handing over the 

spoils of his theft to Madanika. Under these circumstances 

it is the height of indiscretion to stand outside the house of 

the heroine and shout for his mistress at the top of his voice. 

Again, if Madanika is able to hear Sajjalaka, so should Vasan- 

tasena, who is sitting close by, be able to hear him. Apparently 

she fails to do so owing to her preoccupation; but this is a 

circumstance that could not have been foreseen even by a 

scientific burglar like Sajjalaka. The situation in the Mrccha- 

katika (p. 169) is much more realistic. On reaching Vasantasena's 

house, Sarvilaka, instead of calling out for Madanika, hangs 

about outside the house waiting his opportunity. The meeting 

of the lovers is brought about in the following manner. Soon 

after Sarvilaka reaches the house of Vasantasena, the latter 

sends away Madanika on an errand; on her way back, Madanika 

is discovered by Sarvilaka, whom she thereupon naturally joins. 

One more instance, which is the last. A time analysis of 

the first three acts of the Carudatta will show that the incidents 

developed in these acts are supposed to take place on three 

consecutive days, the sixth, seventh and eighth of a certain 

lunar fortnight. Here are the specific references. Carudatta 7, 

the Vidiisaka, in speaking of the Nayaka, applies the adjective 

sattMMdadevakayya to him, which incidentally shows that that 

day was the sixth. Latter on in the same act (Caru. 30), 

addressing the Cetl, the Vidusaka says: 

68 V. S. Sukthankar 

satthie sattamie a dharehi | ahaih atthamie anaddhde dharaissam. 

The arrangement he proposes is that the Ceti should guard 
the jewels of the Ganika on the sixth and the seventh, and 
that he should take over the charge of them on the eighth. 
In the third act we have a confirmation of the same arrange- 
ment. Caru. 53, Ceti remarks: 

iam suvannabhandam satthie sattamie (parivetthami?) J atthami 

khu ajja. 

The Ceti, appearing before the Vidusaka, with the jewels, on 
the night of the eighth, points out that she has guarded them 
on the sixth and the seventh, and adds that that day being 
the eighth it is the turn of the Vidusaka. Later on in the 
same act (Caru. 65), the Brahmani, the hero's wife, incidentally 
mentions that she was observing on that day the Fast of the 
Sixth 23 , to which the Vidusaka pointedly retorts that that day 
was the eighth and not the sixth 24 . These various references 
leave no doubt that the events that form the action of the 
first three acts are supposed to take place within the span of 
three consecutive days. 

There are in the play some further chronological data, which 
we must also take into consideration. They comprise two 
lyrical stanzas which describe respectively the rising and the 
setting of the moon. In that elegant little verse (Caru. I. 29) 
beginning with 

udayati hi sasdnkah klinnakharjurapanduh 

the moon is described as rising, late in the evening, after the 
lapse of a short period of darkness following upon sunset, during 
which Vasantasena escapes from the clutches of the evil Sakara. 
In the third act, on his way home from the concert, Carudatta, 
in a lyrical mood, recites another verse (Caru. III. 3), beginning 

asau hi dattva timirdvakds'am 
astam gato hy astamapaksacandrah, 25 

and having for its theme the setting moon. 

2J The words of the Brahmani are: nam satthim uvavasami. 

24 The Vidusaka observes: atthami khu ajja. 

25 Translation: 'For yonder the Moon of the Eighth, giving place to 
darkness, has sunk behind the western mount.' 

Studies in Bhasa 69 

This is the chronological material of the Carudatta. Let 
us turn for a moment to the Mrcchakatika and examine its 
data. Here also apparently the same conditions prevail. Appa- 
rently the events of the first three acts take place on three 
consecutive days, but only apparently so. There is nothing in 
the play itself from which the duration of the action could be 
precisely computed. 

To begin with, the reference to the sasthi is missing from 
the opening words of the Vidusaka in the first act. In place 
of satthikidadevakayya of the Carudatta, we have the reading 
siddhikidadevakajja, in which siddhi takes the place of satthi. 
Likwise we find that all subsequent references to the lunar 
dates are missing from the succeeding speeches of the Vidusaka 
and the Servant. An entirely different scheme has been 
adopted for the division of labor between the Vidusaka and 
the Servant. The Servant explains in the third act (Mrccha.137) 
the arrangement arrived at as follows: 

ajja mittea edam tarn suvannabhandaam mama diva tuha 

lattim ca, 

'Maitreya, here is the golden casket, that's mine by day and 
yours by nighf; no reference here to the satthi, sattami and 
atthami of the Carudatta. This is not all. The verse from 
the third act of the Caru. cited above, containing a reference 
to the date, has also been substantially modified. Caru. III. 3 b 
specifically states the date to be eighth: astam gato hy astama- 
paltsacandrah. In the Mrcchakatika version the line reads 
(Mrccha. III. 7 b): astam vrajaty unnatakotir induh. The phrase 
unnatakoti has taken the place of astamapaksa, which brought 
in its train, naturally, the change of gato to a word like 
vrajati. 26 It is true that later on, in the same act of the 
Mrcchakatika (p. 159), the Vadhu,Carudatta's wife, refers to satthi, 
saying that she is observing the raanasatthl (ratnasasthi). 27 
But here also a significant omission confronts us. The Vidusaka, 
instead of correcting her, accepts her statement with the necklace, 
and there the matter rests. 

26 The present tense vrajati gives better sense than the past gato, in 
regard to the simile contained in lines c and d. 

" Instead of the vague satthi of the Carudatta we have the more 
specific raanasatthl in the Mrcchakatika. 

70 V. S. Suktharikar 

As remarked above, apparently the joint duration of the 
first three acts of the Mrcchakatika is also three days. But 
I have grave doubts whether any strict proof can be brought 
forward to support such an assumption. I have read the drama 
carefully and I have failed to find any allusion that necessitates 
such a time scheme. However that may be, it is absolutely 
certain that the specific references of the Carudatta to the 
lunar dates are conspicuous by their absence in the other play. 

At this place it may be observed that the tithi-scheme of 
the Carudatta taken in conjunction with the references to 
moon-rise and moon-set in the verses already cited involves a 
chronological inconsistency, so minute and so latent as to be 
hardly noticeable. But the inconsistency is, nevertheless, an 
undeniable fact. For, the rising of the moon late in the evening 
and the setting of the moon at or about midnight 28 are 
phenomena that inherently belong to two different lunar fort- 
nights. Only in the dark fortnight does the moon rise late 
in the evening: and only in the bright fortnight does the 
moon set at or shortly after midnight. In other words, if the 
moon is seen rising late in the evening on any particular day, 
it is nothing less than a physical impossibility that after an 
interval of forty-eight hours the moon should be seen setting 
at or about midnight. 

The general time-scheme of the Carudatta has thus been 
shown to contain a latent contradiction from which the Mrccha- 
katika is wholly free owing to the absence therein of any 
specific references to the days on which the action takes place. 

Are these variations arbitrary; or are they directly or in- 
directly related; and if so how? 

Summary and Conclusion. 
Briefly summarized, the significant differences between the 
two versions discussed above are the following. Firstly, in point 
of technique, the Carudatta differs conspicuously from the other 
play in the absence of the nandi, and in having a rudimentary 
sthapana. Secondly, the Prakrit of the Carudatta is more 
archaic than that of the Mrcchakatika, in so far that the 

28 According to the words of the hero, just preceding the verse asau 
hi dattva, etc. (Caru. III. 3): updrudho 'rdhardtrah (Caru. 50). 

Studies in Bhasa 71 

former contains a number of Old-Prakrit forms not found in 
the latter. Thirdly, as regards versification, the text of the 
Mrcchakatika marks an advance upon the other play in the 
following directions: rectification of grammatical mistakes; 
elimination of redundancies and awkward constructions; and 
introduction of other changes which may be claimed to be 
improvements in the form and substance of the verses. Fourthly 
and lastly, because of suitable additions and omissions the 
Mrcchakatika presents a text free from many of the flaws, 
such as unrealities and inconsistencies, in the action of the 

These are the facts of the case. Do these facts enable us 
to decide the question of priority and anteriority? 

Let us assume first, for the sake of argument, that the 
Carudatta contains older material (at least in respect of the 
passages discussed above) which was worked up later into the 

The differences in the technique neither support nor con- 
tradict definitely such an assumption. The nandl, for all we 
can say, may have been lost. The words nandyante tatah 
pravisati sutradharah do not militate against such a supposition: 
they could be used with or without a nandl appearing in the 
text. Moreover, we cannot, in the present state of our know- 
ledge, rightly evaluate the absence of all reference to the name 
of the play and the playwright in the sthapana. 29 To say 
that in pre-classical times that was the practice is begging 
the question. The only technique of introduction with which we 
are familiar is the well-known classical model. Again the only 
play which is definitely known to antedate the classical plays 
is the Turfan fragment of ASvaghosa's drama. Unfortunately, 
as the beginning of the SWiputraprakarana 30 is missing, we 
we are not in a position to say whether the prologue of the dramas 
of Asvaghosa conformed to the standard of the classical dramas, 
or that of the dramas of the group under consideration. We 
are therefore bound to admit that at present we have no clear 
evidence that can aid us in placing with any degree of assurance, 

'» The references in the text-books of rhetoric and dramaturgy are 
obscure and partly contradictory. 

so Ed. Liiders, Sitzungsberichte d. kgl. preuss. Ak. d. Wiss. 1911. 

72 V. 8. SiMharikar 

chronologically or topographically, a drama with the technical 
peculiarities of the Carudatta. 

But the priority of the Carudatta version would explain, 
and satisfactorily explain, all the other differences between 
the two plays. It would explain the presence of archaisms in 
the Prakrit of the Carudatta. It would explain why many of 
the verses of the Mrcchakatika are free from the flaws of the 
corresponding verses of the Carudatta; the grammatical 
corrections one may be justified in regarding as an indication 
of an increasingly insistent demand for scrupulous purity of 
language. The hypothesis would lastly explain the reason for 
the differences in the incidents of the action of the play. 
All this is legitimate field of 'diaskeuasis', and is readily 

Let us now examine the other possibility, and try to explain 
the divergences on the assumption of the priority of the Mrccha- 
katika version. 

The question of the technical differences between the plays 
has been dealt with already. It was submitted that this part 
of the evidence was inconclusive; it supported neither one side 
nor the other. 

We will proceed to the next point, the Prakrit. 31 On the 
assumption of the priority of the Mrcchakatika version, it is 
at first sight not quite clear, how the Carudatta should happen 
to contain Prakrit forms older than those found in (what is 
alleged to be) a still older play. But a little reflection will 
suffice to bring home to us the fact that it is not impossible to 
account for this anomaly. We have only to regard the Carudatta 
as the version of a different province or a different literary tradition, 
which had not accepted the innovations in Prakrit that later 
became prevalent. In other words we have to assume merely 
that the Prakrit neologisms of the Mrcchakatika are unauthorized 
innovations and that the Carudatta manuscripts have only 

31 Until we have before us most carefully edited texts, any linguistic 
conclusion based upon minute differences in the form of Pkt. words, as 
appearing in the text-editions employed, must needs be regarded as 
tentative, a point not sufficiently emphasized in my article dealing with 
Prakrit archaisms (above, vol. 40, pp. 248 ff.). It may, however, be pointed 
out that no amount of critical editing can disturb the general inference 
that the dramas of this group contain quite a number of Old-Pkt. forms. 

Studies in Bhasa 73 

preserved some of the Old-Prakrit forms of the original Mrccha- 
katika. 32 ] This does not, however, necessarily make the Carudatta 
version older than the Mrcchakatika version. The Carudatta 
would become a recension of the Mrcchakatika with archaic 
Prakrit. Thus the Prakrit archaisms of the Carudatta may 
he said to he not irreconcilable with the general priority of 
the Mrcchakatika version. 

It is much more difficult to explain why the Mrcchakatika 
should consistently offer better readings of the verses. Some 
of the discrepancies could perhaps be explained away as the 
result of misreading and faulty transcript, but not all. We 
could not explain, for instance, why the excellent pada: 
tiksnam visdndgram ivdvasistam should have been discarded, 
and another, visdnakotiva nimajjamand, be substituted, forsooth 
with the faulty nimajjamand. Why should there be a change 
in the first place, and why should the change he consistently 
for the worse? We could not reasonably hold the copyists 
guilty of introducing systematically such strange blunders and 
inexcusable distortions. 

Let us combine the archaisms of the Prakrit with the imper- 
fections of the Sanskrit verses. On the assumption of the 
posteriority of the Carudatta, we are asked to believe that 
while the compiler of the Carudatta had carefully copied out 
from older manuscripts all the Prakrit archaisms, he had 
systematically mutilated the Sanskrit verses, which is a reductio 
ad absurdum! 

Let us proceed to the fourth point. The theory of the 
priority of the Mrcchakatika, which could with difficulty be 
supported in the case of the divergencies already considered, 
breaks down altogether when we try to account for the in- 
consistencies in the action of the Carudatta in general, and 
in particular the presence of the tithi-scheme, which latter 
serves no purpose, aesthetic or didactic, but on the other hand 
introduces gratuitously an indisputable incongruity. The deleting 
of the whole tithi-scheme admits of a simple, self-evident ex- 
planation, acceptable to every impartial critic. But, assuming 

33 Or that the Old-Prakrit forms had been substituted for the Middle- 
Prakrit forms, because the local tradition demanded the use of Old- 
Prakrit forms. 

74 V. S. Sukthankar 

that the original play contained no trace of it, can any one 
pretend to he able to give a satisfactory reason for the deliberate 
introduction of the tithi-scheme? 

Taking all things into account, we conclude, we can readily 
understand the evolution of a Mrcchakatika version from a 
Carudatta version, but not vice versa. The special appeal of 
this hypothesis lies in the fact that it explains not merely 
isolated variations, but whole categories of them: it implies 
the formulation of a single uniform principle to explain divers 

It may be that I have overlooked inconsistencies and flaws 
in the Mrcchakatika version, absent from the other, which 
could be better explained on the contrary supposition of the 
priority of the Mrcchakatika version. If so, the problem 
becomes still more complicated, and will need further investi- 
gation from a new angle. I merely claim that I have furnished 
here some prima facie reasons for holding that the Carudatta 
version is on the whole older than the Mrcchakatika version; 
hence (as a corollary) if our Carudatta is not itself the original 
of the Mrcchakatika, then, we must assume, it has preserved 
a great deal of the original upon which the Mrcchakatika 
is based.