The Turtle and the Geese: A Pañcatantra Fable in Sogdian

Nicholas Sims-Williams <> studied Iranian languages, Sanskrit and Syriac at Cambridge, obtaining his PhD with a thesis on the Sogdian manuscript C2, a miscellany of Christian texts translated from Syriac. In 1976 he joined the staff of SOAS, University of London, where he became Research Professor of Iranian and Central Asian Studies in 2004 and Emeritus Professor in 2015. His research focuses on Bactrian, Sogdian and other Middle Iranian languages of Afghanistan and Central Asia. He is a Fellow of the British Academy, the Austrian Academy of Sciences, the Académie des Inscriptions et Belles Lettres and the Academia Europaea, and an Honorary Member of the American Philosophical Society and the American Oriental Society. He is Chairman of the Ancient India and Iran Trust, Cambridge, and the Corpus Inscriptionum Iranicarum.

Olga Chunakova recently published a fragment of a previously unknown Sogdian fable under the title “A Sogdian Manichaean Parable.”[1] As she rightly indicates, the fragment (SI 5704) contains part of a tale which is well-known from sources both eastern and western, including the Pañcatantra, Kalila wa Dimna and Aesop’s fables, though in this last the content of the story is altered significantly. Chunakova has given an excellent survey of the various versions of this “migratory tale,”[2] extending as far as the Russian short story, “The Traveling Frog” by Vsevolod Garshin.[3] Sogdian-speaking Manicheans frequently employed such stories as parables, and it is very likely that this is the case here too; however, since the fragment is written in Sogdian rather than Manichean script, and since no moral or epimythion survives, one cannot be entirely certain of its Manichean origin.

The oldest surviving Indian version seems to be a story in the Buddhist Kacchapa Jātaka, according to which two geese make friends with a turtle (or tortoise) and invite him to visit their home in the Himalayas. They propose flying through the air, holding the two ends of a stick in their beaks while the turtle bites on the middle of the stick, but the plan goes awry when the turtle cannot restrain himself from speaking and thus falls to his death.[4] Essentially the same story is told in the Pañcatantra and Kalila wa Dimna, except that in these versions the reason for the journey is that the geese and the turtle need to escape because their pond is drying up.[5] Although only a small fragment of the Sogdian text survives, Chunakova’s convincing restoration of the word [p]twʾty “dried up” in the last line suggests that it is close to the Pañcatantra version.

According to Chunakova, the Sogdian version replaces the geese with falcons, a change which she is inclined to ascribe to reliance on a putative Buddhist text in Chinese.[6] However, it seems to me that the first word of line 5 is not zwš or nwš “falcon” but kws “side, limit,” here as elsewhere combined with its near-synonym kyrʾn in a hendiadys: compare, for instance, pw kws kyrʾn translating Middle Persian ʾqnʾrgwm[nd] “unlimited.”[7] If I am right, the species of the birds is not indicated in the surviving part of the text, so we are free to assume that they are geese as in the Pañcatantra and most other versions of the fable. Since Chunakova’s edition in my opinion contains several other misreadings, it seems worthwhile to give a new reading and translation of the whole text, together with a linguistic commentary which will, I hope, be of interest at least to our dedicatee, Professor Badri Gharib, the founder and leading proponent of Sogdian studies in Iran.



1                                            ]k(yš)ph

2                     ]…[                    ](t) kt

3     tγw myδ(ʾ)ny kwcʾky δʾrwkw

4     xns zγʾy [deletion] ZY mʾx ʾδw

5     kws ZY ʾδw kyr(ʾn) kwcʾky

6     xns zγʾy-ʾmkʾm βrwz-ʾny-(h)

7     šwymkʾʾm twʾ cym(ʾ)[yδ]

8     [p]twʾty zʾyh ny(š)[kʾwym]


“. . . [The geese said to] the turtle: ‘You bite the wood firmly in the middle with (your) mouth, and we will bite the two sides and the two ends firmly in (our) mouth(s); we will go flying (and) [will get] you out of this dried up place . . .’”


Linguistic commentary

  1. Sogdian kyšph can presumably refer to either a tortoise or a turtle, like the cognate forms in Indian languages, Sanskrit and Pali kacchapa– etc.[8]
  2. The particle kt no doubt introduces direct speech, a typical late Sogdian usage. It should be preceded by a 3rd-person plural verb meaning approximately “they (i.e. the geese) said,” but not enough of the word survives to justify making a choice between the various possible forms.
  3. kwcʾky (also in line 5) must stand for [kōčē] or [kōčā’ī], oblique case of kwcʾkh [kōčā] “mouth.” Compare to the variant spelling kwcʾkyh (with silent –h) in P3, line 295.[9]
  4. The verb zγʾy “to bite” (also in line 6) was not recognized by Chunakova. It is only known from one other published text, which describes the anguish of an animal which is about to be butchered: ZKw zβʾk ʿM xypδ δntʾk zγʾytw “it bites its tongue with its teeth” (P2, lines 297–98). In his edition of P2, Émile Benveniste read the initial letter as n-, which is in principle possible, but W. B. Henning, in an extended review of Benveniste’s work, noted that the verb is written with an unambiguous z- in unpublished texts.[10] One of the texts to which Henning refers is probably So 14410 /I/, a Manichean divination text, which contains the following prediction (verso, lines 5–7):

wyspw kwtyšt pr twʾ šrʾk kwnʾnt tγw ʾʾγʾz-ykʾn pckwyrt Lʾ( ktʾ) kwnʾ(n)t(k)ʾm δʾmcʾny kwtyšt twʾ z-γʾyʾʾt

“All the dogs (will) bark at you (and) you will begin to be afraid; (but) the dogs of the (material) world will not be able to bite you.”

The form of the verb in line 6 of our fable is apparently 1st person singular future but must be intended for 1st person plural “we will bite” (zγʾy-ʾmkʾm for *zγʾy-ʾymkʾm), while that in line 4 is 2nd person singular imperative zγʾy “bite!” Here too the scribe seems at first to have written zγʾy-ʾmkʾm, anticipating the later form, but the last five letters have been deleted.

Etymologically, zγʾy “to bite” seems likely to derive from the Iranian root *gah “to gorge,” Old Indian ghas “to eat, devour.”[11] For the semantics cf. in particular Choresmian *bγʾh- “to bite,” attested via forms of the imperfect stem bʾγy-, which is found in sentences such as bʾγyt ʾy xyr “the donkey bit” or bʾγytyc wsnʾ prmʾhʾcyc “he bit on it (= the wood) to test it (for hardness).”[12] The initial z– of the Sogdian verb probably represents the preverb *uz-, while the remainder of the stem could derive from *gahaya– (cf. ptxwʾy “to kill” from *pati-hwahaya-) or *gāh(a)ya-.

6–7. For βrwz-ʾny-(h) (whose final –h is perhaps just a line-filler) cf. βrwzʾnʾk “flying” (SCE, line 304), Christian Sogdian brwzʾny “bird” (E27/17V2).[13] Its use with šw– “to go” may be compared with the phrase prnʾʾyʾnn šw– “to go flying” in the Rustam text.[14]

  1. As noted above, [p]twʾty “dried up” was already restored by Chunakova. The same word occurs in a partially comparable Sogdian tale concerning a parrot who is obliged to move because the tree in which he has his nest loses its water supply and dries up.[15] My restoration of a form of the verb nyškʾw “to remove, bring out, take out” is hypothetical, but the third letter does appear more likely to be š rather than m as read by Chunakova.
Fig. 1. The manuscript SI 5704, verso. By permission of the Institute of Oriental Manuscripts, Russian Academy of Sciences

Fig. 1. The manuscript SI 5704, verso. By permission of the Institute of Oriental Manuscripts, Russian Academy of Sciences

Fig. 2. The Turtle and the Geese, carving from Nalanda, Temple 2. Photo by Michael Gunther, CC Attribution-Share Alike 4.0 International License

Fig. 2. The Turtle and the Geese, carving from Nalanda, Temple 2. Photo by Michael Gunther, CC Attribution-Share Alike 4.0 International License


[1]Olga M. Chunakova, “A Sogdian Manichaean Parable,” Written Monuments of the Orient, no. 2 (2017): 35–42. As I have been informed by Yutaka Yoshida and Pavel Lurje, a further Sogdian fragment of this tale, the text of which overlaps with that of the fragment discussed here, has now turned up in St Petersburg and will be published in a forthcoming book by Chunakova.

[2]Since Chunakova chiefly refers to works in Russian, it has seemed useful to give references here to English translations of the most important versions. For an informative survey of the many versions and illustrations of this fable see also (accessed 5 March 2019).

[3]Several English translations of Garshin’s short story are available, e.g. The Frog Went Travelling, trans. Olga Shartse (Moscow: Raduga Publishers, 1987),

[4]H. T. Francis and E. J. Thomas, Jātaka Tales (Cambridge: Cambridge University Press, 1916), 178–80.

[5]Patrick Olivelle, Pañcatantra The Book of India’s Folk Wisdom (Oxford: Oxford University Press, 1997), 51–52; I. G. N. Keith-Falconer, Kalīlah and Dimnah or the Fables of Bidpai (Cambridge: Cambridge University Press, 1885), 48–49.

[6]Chunakova, “A Sogdian Manichaean Parable,” 41. [See n. 1.]

[7]W. B. Henning, Sogdica (London: The Royal Asiatic Society, 1940): 27, fragment d, line 6.

[8]R. L. Turner, A Comparative Dictionary of the Indo-Aryan Languages (London: Oxford University Press, 1966), 130.

[9]Cited in B. Gharib, Sogdian Dictionary. Sogdian–Persian–English (Tehran: Farhangan, 1995), 199.

[10]E. Benveniste, Textes sogdiens (Paris: Librairie Orientaliste Paul Geuthner, 1940), 16; W. B. Henning, “The Sogdian Texts of Paris,” Bulletin of the School of Oriental and African Studies 11, no. 4 (1946): 734.

[11]Johnny Cheung, Etymological Dictionary of the Iranian Verb (Leiden: Brill, 2007), 93.

[12]Mahlagha Samadi, Das chwaresmische Verbum (Wiesbaden: Harrassowitz, 1986), 17.

[13]Ilya Gershevitch, A Grammar of Manichean Sogdian (Oxford: Basil Blackwell, 1954), section 1039; Nicholas Sims-Williams, The Christian Sogdian Manuscript C 2 (Berlin: Akademie Verlag, 1985), 58.

[14]Sims-Williams, “The Sogdian Fragments of the British Library,” Indo-Iranian Journal 18 (1976): 43–82, esp. 55, line 20.

[15]Enrico Morano, “Sogdian Tales in Manichaean Script,” in Literarische Stoffe und ihre Gestaltung in mitteliranischer Zeit, ed. Desmond Durkin-Meisterernst et al. (Wiesbaden: Reichert, 2009), 173–200, esp. 179, line 23 of the reconstructed text.