Owners&#39; graffiti on pottery from Tissamaharama

Harry Falk

Owners' graffiti on pottery from Tissamaharama

Abstract

The legends on sherds from the archaeological site Tissamaharama in southern Sri Lanka are presented and their contents examined with regard to the composition of laypeople and Buddhist clerics from the late 3rd century BC onwards. The role of women is particularly noteworthy.

Figures (110)

Fig. 1 No.1. Sherd of a dining plate from layer “a2”.

Fig. 3. No.3, Phase “b”, dinner plate, BRW. /// pathara ///, “Dinner plate (of NN?)”.

Fig. 2. No.2, Phase “a2”. Cooking pot, Tissa Form A, BRW. No. 79 writes de+(initial)e+vaha, “of Deva”. Under that number cases are cited from Tamil sites where a medial vowel is followed by an

Fig. 4. No.4, phase “a2”, dining plate, Tissa Form G1, BRW.

Fig. 5. No.5, phase “cl”, small cup-shaped bowl, Tissa Form I, BRW.

Fig.7. No.7, phase “b”, large cooking or serving vessel, Tissa Form A, BRW. Fig. 6. No.6, phase “b”, water-pot, Tissa Form D.

Fig. 8. No.8, phase “b”, water-pot, Tissa Form D. Fig. 9. No.9, phase “b”, storage jar, Tissa Form E. The mi is disturbed in its lower left part. No. 9 (fig. 9). b2(k)[z]n1-Sa[ga] ///, “(Vessel for the com)munity of nuns”.

Fig. 12. No. 12, late part of phase “c2” or early “di”, small serving or cooking pot, Tissa Form A in transition to B2. Fig. 11. No. 11, later part of phase “cl”, serving pot, Tissa Form B1.

Fig. 15. No. 15, phase late “b” or early “cl”, dining plate. Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype A. Fig. 14. No. 14, late phase “b” or early phase “cl”, dining plate. Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype A.

No. 14 (fig. 14). ctta-Samaniya, “(Plate) of the = 99 nun Citra

Fig. 13. No. 13, phase “b” or early “c1”, small bowl. Fine Grey Ware from Northern India.

Fig. 19. No.19, phase “cl”, small water-pot. Tissa Form D or E

Fig. 17. No.17, late phase “b” or early “cl”, small storage vessel, Wheeler Type 18, Fine Grey Pottery from Northern India.

Fig. 18. No. 18, phase early “c1”, dining plate. Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B.

Fig. 21. No. 21, phase early “cl”, dining plate. Tissa Form rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B. Fig. 20. No. 20, phase “b” or early “cl”, serving bowl, Tissa Form B1.

Fig. 23. No. 23, phase early “cl”, water-pot or storage jar, Tissa Form D.

Fig. 22. No. 22, phase early “cl”, dining plate, Tissa Form G1, BRW.

Fig. 24. No. 24, later part of phase “c1”, serving bowl, Tissa Form B1.

Fig. 26. No. 26, phase “c2”, storage pot, Tissa Form D.

Fig. 25. No. 25, later part of phase “cl”, small water- pot, Tissa Form D.

Fig. 27. No. 27, phase “d1”, water-pot, Tissa Form D.

Fig. 31. No.31, phase late “cl”, dining plate, Tissa Form G1, Fine Red Ware, polished.

Fig. 28. No. 28, phase “b”, cup-shaped small bowl, Tissa Form I, BRW.

Fig. 29. No.29, phase “b”, water-pot, Tissa Form D.

Fig. 30. No.30, phase late “b” or early “cl”, small carinated bowl, Wheeler Type 18, Fine Grey Ware from Northern India, NBP-like.

Fig. 33. No.33, phase “cl”, small conical beaker, Wheeler ‘Type 10, late form of Fine Grey Pottery from Northern India.

Fig. 35. No.35, most probably phase “cl”, cooking or serving pot, Tissa Form A.

Fig. 32. No. 32, phase “ct”, dining plate, ‘Tissa Form G1, BRW, legend on underside.

Fig. 36. No. 36, phase “c2”, dining plate, Type G2; inscribed on bottom inside, Coarse Red Ware.

No. 34 (fig. 34). pusa-sa[man]i ///, “(Vessel of the (novice?) nun Pusya”.

Fig. 38. No. 43, phase “a3”, dining plate, Tissa Form G1, BRW.

Fig. 42. No. 47, phase late “b” or early “cl”, small globular vessel, Tissa Form I, BRW.

Fig. 41. No. 46, phase late “b” or early “cl”, dining plate, Tissa Form G1, rim type 9; type “Rouletted Ware” in the fabric of Fine Grey Pottery from northern India, subtype A.

Fig. 40. No. 45, phase late “b” or early “cl”, storage jar, Tissa Form E.

Fig. 44. No. 49, phase “b”, water-pot or storage jar, Tissa Form D. Fig. 43. No. 48, phase “b”, dining plate, Tissa Form G1, BRW.

Fig. 45. No.50, phase “b”, water-pot or storage jar, Tissa Form D.

Fig. 46. No.51, phase “b”, small water-pot, Fine Red Ware.

Fig. 47. No.53, phase “b”, dining plate, Tissa Form G1, BRW.

Fig. 51. No.57, phase “a2-a3”, dining plate, Tissa Form G1, BRW. Uncertain male personal name in the geni- tive, probably containing iva as a part and ending in -ta. No. 57 (fig. 51). + + + + la Siva + + + taSa ///

Fig. 48. No. 54, phase “c2/d1”, large globular storage jar, Tissa Form D.

Fig. 30. No. 56, phase “a2”, dining plate, Tissa Form G1. Fig. 49. No. 55, phase “a2”, cooking pot, Tissa Form A, BRW.

Fig. 54. No.60, phase “a2-a3”, dining plate, Tissa Form Gi, BRW.

Fig. 35. No. 61, most probably phase “a”, dining plate, Tissa Form G1, Black Ware, inscribed on the underside.

No. 58 (fig. 52). <symbol> sumana[ha?] ///, “(Vessel) of Sumana”. The name, Skt Sumanas, is widespread and also found in Tissa-1 (1H,no. 18,23/93,16) reading <moon-symbol> <swastika> su[manasa].

Fig. 60. No. 68, phase “cl”, dining plate, Tissa Form G1, BRW.

Fig. 36. No. 64, phase “b”, small cup-shaped bowl, Tissa Form I, BRW.

Fig. 57. No. 65, phase early “cl”, dining plate, Tissa Form G1, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B; letters incised with small bored holes.

Fig. 59. No. 67, phase “c1”, dining plate, Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware from Northern India, latest subtype C.

Fig. 64. No. 73, phase early “c1”, globular bowl with beaked rim similar to “Rouletted Ware”; type of Fine Grey Pottery from Northern India. Fig. 63. No. 72, phase “b” or early “cl”, small bowl. Type of Fine Grey Ware of Northern India.

Fig. 62. No. 70, phase “a3”, dining plate, Tissa Form G1, BRW.

Fig. 61. No.69, phase “a2-a3”, dining plate, Tissa Form G1, BRW.

Fig. 66. No.75, most probably phase “d2”, found in phase “e”; storage jar, Tissa Form D or E; Coarse Red Ware. Vihara at Anuradhapura, to be dated at the beginning of the third century."

Fig. 65. No. 74, later part of phase “c1”, dining plate, Tissa Form G1, BRW.

Fig. 68. No.77, phase “a3”, dining plate, Tissa Form Gi, BRW.

Fig. 71, No. 80, later part of phase “b”, tiny small carinated bowl, Wheeler Type 18, type of Fine Grey Ware from Northern India. Fig. 70. No.79, phase early “c1”, dining plate, Tissa Form G1, BRW.

Fig. 69. No. 78, later part of phase “b”, dining plate, NBP, Fine Grey Ware from Northern India.

Fig. 75. No. 84, phase “b” or early part of “cl”, small bowl. Type of Fine Grey Ware from Northern India.

Fig. 72. No. 81, phase “b” or early “cl”, small bowl, type of Fine Grey Ware from Nothern India.

Fig. 76. No. 85, phase “cl”, dining plate, Tissa Form G1, BRW.

Fig. 80. No. 89, early part of phase “b”, water-pot or storage jar, Tissa Form D. Fig. 79. No. 88, phase “a3”, dining plate, Tissa Form G1, BRW.

Fig. 77. No. 86, phase “cl”, serving vessel, Tissa Form B1.

Fig. 78. No.87, phase “a”, dining plate, Tissa Form G1, BRW.

Fig. 84. No. 94, phase “a2”, dining plate, Tissa Form Gi, BRW. Fig. 83. No. 93, phase “c1”, dining plate, Tissa Form G1, BRW. Incised on the underside.

Fig. 82. No. 91, phase “b”, cooking pot, Tissa Form A. No. 90 (fig. 81). ///.. ta-baginiya-tisaha, “(Vessel) of Tisya, the sister-son (bhagineya) of ..ta”.

Fig. 81. No.90, phase “b”, dining plate, Tissa Form G1, BRW.

Fig. 85. No. 95, phase “a2”, dining plate, Tissa Form Gi, BRW.

Fig. 87. No. 97, phase “d1”, storage jar, Form Tissa D1. Fig. 86. No. 96, phase “a2”, dining plate, Tissa Form G1, BRW.

Fig. 88. No. 98, phase “b”, cooking-pot, Tissa Form A, BRW.

Fig. 89. No.99, phase “cl”, small conical beaker, Wheeler Type 10, late type of Fine Grey Pottery from Northern India.

Fig. 92. No. 102, phase “a3”, dining plate, Tissa Form G1, Red Ware.

Fig. 91. No.101, most probably phase “a3”, dining plate, Tissa Form G1, BRW.

Fig. 94. No. 104, phase “cl”, serving vessel, Tissa Form B1. Fig. 93. No. 103, early part of phase “cl”, water-pot or storage jar; Tissa Form D.

Fig. 95. No.105, phase “b”, small carinated bowl, Wheeler type 18, Type of Fine Grey Pottery from Northern India, NBP-like.

Fig. 97. No.107, phase “b”, serving bowl, Tissa Form B1.

Fig. 100. No. 110, phase “c1”, dining plate (Tissa Form G, rim type 9), type “Rouletted Ware” in the fabric of Fine Grey Pottery from northern India. Subtype C. Fig. 99. No. 109, phase “cl”, small cup-shaped bowl, Tissa Form I, BRW.

Fig. 98. No. 108, phase “b”, serving bowl, Tissa Form B1.

Fig. 104. No.114, phase “di/d2”, globular vessel, Tissa Form D or E Fig. 103. No. 113, phase “c1”, water jar, Tissa Form D2.

Fig. 101. No. 111, phase “b”, water jar, Tissa Form D2.

Fig. 102. No.112, phase “b”, water jar, Tissa Forr Dork

Fig. 106. No.117, early part of phase “b”, dining plate, Tissa Form G1, BRW. Fig. 105. No. 116, find context phase “e”, water pot (Tissa Form D or F) dating within the range of phases c2-e.

Fig. 107. No. 118, phase “b”, dining plate, Tissa Form Gi, BRW.

Fig. 109. No. 120, phase “a3”, dining plate, Tissa Form G1, BRW.

Fig. 108. No. 119, phase “b”, dining plate, Tissa Form G1, BRW.

Fig. 113. No. 124, phase “d1-d2”, large serving vessel, Tissa Form B2. Fig. 112. No. 123, phase “a2”, cooking vessel, Tissa Form A), BRW.

Fig. 111. No.122, phase “d1/d2”, water-pot, Tissa Form D2.

Fig. 114. No.126, phase “d2”, cooking or serving vessel, Tissa Form B2, Coarse Red Ware. glyphs, and b) regarded as a precursor to Brahmi. The latter proposal is certainly not helpful, as both symbols and Brahmi script can occur at the same time, even on the same vessel, as in Tissamaharama. Occasionally a stratum with only symbols on pottery overlays an earlier stratum with nothing but Brahmi, as at Uraiyur, Tiruchirappalli (Raman 2011: 75). This shows, to my mind, that symbols and script represent rather different social groups than different time levels.

Fig. 115. Sherd from a globular storage jar (Ono. 60,12), phase “c2”, with cognate forms at Harappa (after Mahadevan 1977: 33) and at Uraiyur (after Raman 2011: 76).

Table 1. Types of owners or donors during the phases a/b/cl1/c2/d. Owners’ graffiti on pottery from Tissamaharama

ZEITSCHRIFT FÜR ARCHÄOLOGIE AUSSEREUROPÄISCHER KULTUREN BAND 6 2014 REICHERT VERLAG  WIESBADEN Sigel der Zeitschrift für Archäologie Außereuropäischer Kulturen: ZAAK 372 Seiten mit 378 Abbildungen Herausgeber Burkhard Vogt – Josef Eiwanger Wissenschaftlicher Beirat Peter Breunig, Frankfurt Ian Glover, London Nikolai Grube, Bonn Thomas Höllmann, München Eric Huysecom, Genf Peter Kaulicke, Lima Paul Yule, Heidelberg Günther A. Wagner, Heidelberg Dorothee Sack-Gauss, Berlin Die Zeitschrift für Archäologie Außereuropäischer Kulturen erscheint in Fortsetzung der Zeitschrift Beiträge zur Allgemeinen und Vergleichenden Archäologie der ehem. Kommission für Allgemeine und Vergleichende Archäologie (bis Band 24, 2004) Bibliograﬁsche Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliograﬁe; detaillierte bibliograﬁsche Daten sind im Internet über http://dnb.ddb.de abrufbar ISSN 1863-0979 ISBN 978-3-89500-075-6 © Kommission für Archäologie Außereuropäischer Kulturen des Deutschen Archäologischen Instituts, Bonn, 2014 Redaktion: Heiko Prümers, KAAK. Satz und Gestaltung: Linden Soft Verlag e.K., Goetheweg 10, D - 73773 Aichwald, www.lindensoft.de Druck und Vertrieb: Reichert Verlag, Tauernstr. 11, D - 65199 Wiesbaden, www.reichertverlag.de Printed in Germany Printed on fade resistant and archival quality paper (PH 7 neutral) · tcf INHALT Brigitte Borell The Power of Images – Coin Portraits of Roman Emperors on Jewellery Pendants in Early Southeast Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Harry Falk Owners’ grafﬁti on pottery from Tissamaharama . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Heidrun Schenk Tissamaharama Pottery sequence and the Early Historic maritime Silk Route across the Indian Ocean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Hans-Joachim Weisshaar Legged Saddle Querns of South Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Oscar Quintana Nakum – Ciudad Maya, Petén, Guatemala . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Gabriela Ortiz / Luis Nieva Morir en el Valle de San Francisco. Prácticas funerarias, termoalteración y estratégias de memorización en la selva pedemontaña de las Yungas del Noa . . . . 247 Berichte der Projekte der Kommission für Archäologie Außereuropäischer Kulturen des Deutschen Archäologischen Instituts Johannes Moser / Lawrence Kiko Die archäologischen Ausgrabungen in ‘Apunirereha’ und ‘Ria’ auf der Insel Malaita, Salomonen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 Markus Reindel / Johny Isla / Heike Otten / Hermann Gorbahn / Jennifer von Schwerin Archäologische Forschungen in Peru und Honduras im Jahr 2013 . . . . . . . . . . . . . . 289 Heiko Prümers / Carla Jaimes Betancourt Die frühen Siedler von Jasiaquiri (Bolivien) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Josef Eiwanger / Sonja Tomasso Forschungen in Ifri n’Ammar und ihrem Umfeld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Thorsten Behrendt / Andreas Reinecke Die Petrographie keramischer Grabbeigaben und Steinwerkzeuge aus der Deltaebene des Mekong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341 Christina Franken / Ulambayar Erdenebat / Tumurochir Batbayar Erste Ergebnisse der Grabungen des Jahres 2013 in Karabalgasun und Karakorum / Mongolei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 Harry Falk Owners’ grafﬁti on pottery from Tissamaharama Keywords: Sri Lanka; Tissamaharama; buddhism, onomastics, nunneries, monastic life Abstract: Almost twenty years of excavation at Tissamaharama, southern Sri Lanka, have produced an immense number of pottery sherds. More than one hundred carry comprehensible legends in Brāhmī letters, dating back to centuries before and after Christ. They allow for the ﬁrst time to accompany the Buddhist monks and nuns in their numerical development over centuries and their interaction with sponsors and supporters. Nunneries are otherwise almost untraceable, while here we have a domination of nuns and laywomen over their male counterparts. Tamil speaking components of society are few but perceptible, particularly during the foundation phase. 1. Chronological considerations 1.1 Radiocarbon and chronology In the centuries around the turn of times people in ancient India marked their vessels either with logograms or with short inscriptions in early forms of the Brāhmī script. This script is forever linked to the edicts of king Aśoka, ruling from Patna, the ancient Pātaliputra, in the years 268 to 232 B. C. There have been attempts in India and in Sri Lanka to date Brāhmī epigrams before the time of this ruler. So far, no inscribed piece in mainland India can be demonstrably dated before Aśoka. In Sri Lanka 14C dates from material found with or near inscribed sherds in Anuradhapura show very similar possibility ranges from 100 B. C. to 400 B. C. over a long stretch of strata (called G2 down to J3, Coningham et al. 1996: 78), showing to my mind that these strata built up in a rather short time. By a mathematical trick, including the lower strata which did not produce a single inscribed sherd, the oldest inscribed sherd is then dated to the “early part of the fourth century B. C”. (Coningham et al. 1996: 77b).1 From this arose the claim by Sri Lankan nationalists that Brāhmī was created on the island, centuries before Aśoka and that it migrated north. Against this epigraphists opine that under those premises parts of Sri Lankan Brāhmī should show primary features which later developed into the northern script, while a survey of all features in all forms of early Brāhmī (Ceylonese, Tamil and Aśokan) shows just the opposite, i. e. that all differences can 1 The authors admit not to have used Southern Hermisphere Correction (Coningham et al. 1996: 79a). As long as there is no special calibration deﬁned for Sri Lanka in the ﬁrst millennium B. C. it remains to point out that the southern hemisphere’s load of radiocarbon is signiﬁcantly different from the one expected for northern Pakistan, cf. the curve in McCormac et al. 2004: 1089. Sri Lanka may occupy a medium position, but as long as there is no ﬁxed curve for the island, even vague dates should be taken with a pinch of salt. Zeitschrift für Archäologie Außereuropäischer Kulturen 6 (2014): 45–94 46 Harry Falk only be explained once the Aśokan script is taken as primary and the two others as derivations. But regional chauvinism will never abate.2 The vexed question if Brāhmī is older than Aśoka or not melts down to an evaluation of radiocarbon data. Are they reliable or not? The material from Anuradhapura can now be supplemented by data from the excavations at the so-called citadel at Tissamaharama, locally known as Akurugoda or Akurugodalla,3 a minor and elevated part of town bordering the eastern bank of the age-old tank in the Ruhuna region of southern Ceylon. The excavations were guided by Hans-Joachim Weisshaar over almost 20 years from 1992 until 2010 and have exposed a hitherto undisturbed site through all of its phases, down to the virgin soil. A meticulous stratigraphy in 10 cm steps occurred under the scrutiny of the expert in local pottery, Heidrun Schenk, aided by gifted local draftsmen and account keepers. This excavation therefore evokes hopes that ﬁnally typology and stratigraphy will work together to help settle the chronological conundrum. Three places were examined by the excavators. One, Tissa-1,4 near the ascent to the citadel, proved to be a sort of workmen’s quarters, Tissa-2 had soon to be given up, and Tissa-3, near the bank of the tank, showed a number of vessels belonging to the Buddhist clergy, laypeople and “parish workers”, predominantly females. The upper layers from phase “d” onwards are credibly regarded as being part of a hospital (Weisshaar 2004: 139 f.). This interpretation was based mainly on the enormous number of saddle querns, which the locals on inspection instantly traced back to pharmaceutical necessities. Since monasteries are often in charge of medical services the same purpose may also have prevailed which the lower strata built up, although the earlier layers “a” to “c2” disclosed nothing except that which looks like ordinary living quarters. Even if the monastic site was not in the conﬁnes of the excavation it existed not far from it. In the sequel I will present a collection of ca. 125 sherds, read their legends, analyze the group of people producing this waste over centuries and putting the names and functions in a relative chronological sequence. A relative chronology is possible, absolute only as far as 14 C dates can provide absolute dates. The strata yielding Brāhmī legends on pottery sherds have been labelled “a (1–3), b, c1, c2”, and “d”. On the basis of 14C results these strata have been allotted to the 4th century B. C. to the 2nd century A. D.5 More recent research with reworked time brackets are found in Pavan / Schenk (2012: 196, footnote 18)6. As said above, dates for inscribed sherds in the 5th and 4th centuries B. C. are principally in conﬂict with the well-founded view that Brāhmī was invented during the time and on the order of king Aśoka, so that no grafﬁto with that script can be older than the 3rd century B. C. This view, which I share and for which others and me (Falk 1993: 205–218) have given a range of palaeographical reasons, was also expressed by Ceylonese (Paranavitana, 2 3 4 5 6 Particularly ill-informed appears Rajan (2008: 48), according to whom “it is almost clear now that Aśoka did not developed (sic) the Brahmi script. The origin or evolution of a script is a social process and it could not be associated with a particular individual or dynasty.” Misinterpreting megalithic logograms for Brāhmī writing at Porunthal he “reads” va-y-ra (“meaning diamond”) on a ring-stand 14C-dated to 490 B. C. by an American laboratory. The non-letters put aside, the presence of stirrups in the same tomb sounds suspicious too. Cf. http://porunthal.blogspot.fr/2011_10_14_archive.html as of May 2014. Meaning “letter(aksara)-hill” because of the prominent inscribed pillar at the southern entrance. Instead of akurugodalla also akurugoda is used with the same meaning. I thank Osmund Bopearachchi for this information. Sherds from the ﬁrst excavation called “Tissa–1” have not been photographed as meticulously as those of Tissa–3, which are the subject of this paper. There are 27 sherds with Brāhmī aksaras copied by draughtsmen as eye-copies; the relevant ones are named in this paper with their respective legends. Görsdorf (2001: 280b) and Schenk (2001: 61) on the basis of samples from Tissa–1. Samples from Tissa3 put phase “a” and “b” even one century earlier. Here, phase “a” belongs to 5th cent. to 300 B. C., “b” to 300 to 200, “c1” to 200 to 100, “c2 ” to 100 to 1 B. C., and “d1” to 1 to 100 A. D., “d2” to 100 to 200 A. D. Owners’ grafﬁti on pottery from Tissamaharama Karunaratne, Emmanuel) as well as South Indian (Mahadevan) authors using additional and similar arguments. But science also lives through considering new or controversial ideas. In principle, the discrepancy could be solved in two ways: either Brāhmī is older than Aśoka and was invented in Sri Lanka, or the 14C samples provide unreliable results. In the sequel only a few reasons are presented or repeated why early Sri Lankan Brāhmī must be derived from Aśokan Brāhmī, but new perspectives should never be excluded. Hardly any attention will be given to the “megalithic” logograms, although they are ubiquitous at Tissamaharama until around the 2nd century A. D. This excavation for the ﬁrst time allows them to be grouped stratigraphically and we can now see how they distribute and develop over time, a promising task requiring an interested research student. The Tissamaharama excavations prove at least that these logograms have no ritualistic aspect and that linking them to burial-related practices (Boivin et al. 2003: 31b) is based on a superﬁcial impression. Here, only a few logograms will be dealt with, to hopefully inspire comprehensive studies. There is almost no hint in Indian literature of people inscribing their vessels. A story from the Buddhist law book (vinaya) of the Mūlasarvāstivādins is told in Schopen (2007: 293 f.) where a disciple of Aniruddha washed his and his teachers bowl together, confusing them quite often. To prevent further confusion he wrote on Aniruddha’s bowl: “The bowl of the Preceptor Aniruddha”. Monks and the Buddha discussed this unprecedented case and ﬁnally made a decision: “A monk should not inscribe his private property” (2007: 294). Against this Buddhist rule we have a great number of violations at Tissamaharama. In fact, apart from clearly Buddhist owners there is very little evidence which points to other religions. One name containing śiva is there (no. 53, cf. 72), being the only occurrence. About half of the legends may not have anything to do with Buddhism and most uninscribed sherds will once have been part of vessels belonging to private citizens. On the 47 other hand, the Buddhists discernible comprise a wide array of monks, nuns, helpers and laity. The number of Buddhists, clerics and laypeople, might anticipate the discovery of a settlement for monks and nuns near to where the sherds were found. But expecting some sort of a classical monastery may be misleading. The buildings of phase “d” could have belonged to a kind of hospital, where private individuals and clerics alike were treated, by monks as physicians and by śramanerīs as nurses. Even then, the Buddhist physicians and nurses need their meetings, education and their prayerhalls. Amongst the buildings excavated none gives the impression as having served purely religious purposes. Vessels serving the common dinners on the platform (prāsāda, nos. 105–110 below) are difﬁcult to bring in line with the needs of a hospital. Thus, the true nature of the buildings remains an open question from a buddhological point of view. In any case, Buddhists are around and they play a major role at the excavation site and in its vicinity. All legends, here as at other Buddhist monastic sites in the Indian cultural sphere, document individual possession, or, if the local samgha was to receive the vessels, collective possession with an emphasis on the non-clerical person who effected the donation. Mercantile usage, such as naming trade goods, sender or receiver, is absent in the excavated part of Tissamaharama and on the whole extremely rare at other places.7 1.2 Buddhism and chronology The habit of furnishing pottery with owner’s inscriptions is met with all over southern India and Sri Lanka.8 Datings are always somewhat 7 8 This contradicts Ray (2006: 124) who misinterprets culinary vessels as “pots . . . used for transporting commodities”. There is no “monastic context” visible linking Buddhist monasteries with their inscribed vessels and stamped (not inscribed!) containers for trade goods. For an early sherd from Nepal found at Tilaura-Kot reading ///nayadi/// see Mitra (1972: 100; pl. XXVII,5). 48 Harry Falk vague. Even in the revised chronology of Arikamedu most pieces showing forms of Tamil or Sri Lankan Brāhmī have an origin in the ﬁrst or second century A. D. (Mahadevan 1996a: 295). As so many sherds presuppose a Buddhist clergy at Tissamaharama, it is self-imposing that the sherds cannot be older than the Buddha himself. Few would today date his parinirvāna to 486 B. C., as was accepted for a long time. Archaeological reasons (Erdosy 1993: 47), as well as literary ones (Gombrich 1992: 251), speak rather in favor of the beginning of the fourth century for his demise. Since missionaries left Magadha only after the death of the Buddha, a community in Sri Lanka during his lifetime, that means in the ﬁfth century B. C., looks inconceivable. 1.3 Steps of Brāhmī palaeography The Brāhmī script occurs in northern India ﬁrst in the rock inscriptions of king Aśoka from the Maurya dynasty. His earliest texts incised on rock faces date from his 10th regnal year, that is in about 258 B. C. There is evidence to show that his scribes in diverse parts of the subcontinent soon found a number of ambiguities in this script which they remedied out of necessities which can be easily understood. 1.3.1 The ra-la-problem Aśoka’s own language knew only la where other vernaculars had preserved the original distinction between ra and la. One example is lājā, “king”, where others said rājā. When his texts were brought to areas with the distinction, the need to introduce a sign for ra was felt. A simple vertical stroke was used for this purpose. Originally, Aśokan Brāhmī came without punctuation. Some clerks very near to Aśoka’s capital in modern Bihar, however, introduced vertical or slightly bent strokes to separate sentences and parts of sentences. Outside the main Bihar area, pre- or postconsonantal Sanskritic -r / r- was preserved, while in Aśoka’s language it had disappeared, occasionally changing the consonant to which it was originally attached. For instance, Skt dharma became dhamma, or Skt priyadarśī became written piyadasi. Particularly in the western parts of the subcontinent, such an r- or -r was not necessarily elided. The texts came to the west without the rs and scribes reﬂected on how to represent them in writing, e. g. to express their spoken / priyadassi / . The earliest solution we can ﬁnd comes from Girnar in Kathiawar, where the upper part of the concerned consonant was given a wavy form, irrespective of whether the r preceded or followed the consonant, that means that rka and kra looked the same. This ambiguity was remedied in the ﬁrst century B. C. when the wavy line of the r-element was moved to the lower part of the letter for following -rs, while it remained above the letter for the preceding r-. We see a threefold development: 1) no ra at all, 2) introducing a straight vertical for ra, 3) introducing a wavy line for pre- or post-consonantal -r / r-, and this was followed by 4) a wavy vertical line for ra which did away with the possibility to confuse a straight ra with a straight line dividing sentences. Even with this useful new shape the wavy ra could not completely supplant the older straight one, which made it into the precursors of modern Indian scripts. In Ceylon from the start we have both forms, straight and wavy occurring side by side. That seems to indicate that all straight and wavy forms in Sri Lanka can hardly be older than Aśoka’s earliest edicts. In our sherds the wavy ra is rare while an angular zigzag ra prevails. But in some of the oldest pieces with links to Dravidian speaking sites in Southern India (utatiran, no. 1; pathirā, no. 3) a plain vertical is found, as well as in all later legends which use Indian Brāhmī. 1.3.2 Tamil Brāhmī notation I. Mahadevan managed to classify the different stages of Brāhmī used for Tamil in Tamilnadu. His ever-growing experience allowed him to deﬁne the stages of this particular script in a deﬁnite way. In 1996(a) he showed that the northern, Aśokan Brāhmī was adapted in the South in two ways. A Tamil name like Cātaó Owners’ grafﬁti on pottery from Tissamaharama cannot be properly expressed, as the northern Brāhmī does not know closing consonants apart from a nasal (anusvāra). One solution, called TB-I by him, would write cātāna, where every -ā-stroke says that a short or long a / ā is there, and its omission would signal the absence of a vowel, either at the end of the word or before a second consonant. Written cātāóa could thus be pronounced / cataó / , cātaó / , / cātāó / or / catāó / . This ﬁrst stage of the ﬁrst solution solves the problem of the ﬁnal consonants, but cried out for a change to do away with the multiple ambiguity regarding the presence and length of vowels. The solution is found at Bhattiproòu in Andhra Pradesh, where the texts on reliquaries use a vowel-hook for long -ā, a standard stroke for the -a, and no stroke for no vowel (Mahadevan 1996a: 293). This ﬁnal form of TB-I was not widely accepted and died out except for a few traces (cf. below 1.3.3.1), but TB-I in its original form lived on in certain quarters. The second spelling mode is called TB-II and would write Cātaó as cātaóa, where the -ā-stroke indicates a long -ā as in the North, and the strokeless letters ta and óa would indicate a short -a or, alternatively, no vowel at all. The context, which was known to the writer, would decide, but there was no guarantee that the reader guessed at the same. Written cātaóa could then, theoretically, be pronounced / cātaó / or / cātó / . In this example only the ﬁrst possibility makes sense, but when used for geminates this system too shows an ambiguity, which needs to be removed. The solution is the so-called pulli, a dot indicating “no vowel”. In this ﬁnal form of TB-II “Cātaó” would be written cātaµa with a dot over the óa and thus nothing but / cātaó / can be read. That means, that the northern Brāhmī produced problems in the South on account of the absence of vowel-less characters and this problem met with two answers in the early Tamil-Brāhmī, both very imperfect. Both were subsequently perfected but only one unambiguous system, the pulli, did survive to this day. To ﬁnd Tamil names or titles on Sri Lanka in early cave inscriptions or on pottery is not 49 Fig. 1 No. 1. Sherd of a dining plate from layer “a2”. surprising, but to ﬁnd one of Mahadevan’s TB systems in Ceylon is new, to my knowledge. The case is a sherd (No. 1, ﬁg. 1) of a dining plate9 from layer “a2” which therefore belongs to the oldest stratum that provided script, right above the stratum of the earliest settlement (phase “a1”). It is truncated on both sides: No. 1: /// [śi][ma?] utatira[ó] /// The ﬁrst two letters are not absolutely clear. The initial u- should commence a word, but utatiraó· does not make sense in a Sanskrit derived vernacular, while it can be read as uttiraó in Tamil, in the initial and ambiguous stage of TB-II. This reading provides a name which is attested in at least one pottery inscription from Arikamedu (Mahadevan 1996a: 314 f. no. 30), spelled in the same way (u-ta-tira-óa), only the u has only one bar and not two. The same name is also found on a lead coin from Tissamaharama in the spelling of 9 This term builds on the profane usage where dining after noon is no offence. For monks a term “alms bowl” seems more appropriate, but these “dining plates” are no deep bowls, but rather ﬂat plates, like a modern thālī. Whether such ﬂat vessels were used for begging too may be questioned. All “dining plates” in this paper are wide vessels with a low rim and a rather ﬂat bottom, local type “G1”, unless stated otherwise. 50 Harry Falk utirāna (Bopearachchi, Falk & Wickremesinhe 2000: 126 no. 21; Falk 2008: 56 no. 3), where the u again comes with two bars. The spelling utirāóa can be explained as belonging to the ﬁrst and ambiguous stage of TB-I, where every a, short of long, needed an -a-stroke. We thus see that both Tamil spellings in their initial forms were known in Tissamaharama at some time. On our sherd the term preceding uttiraó ends most likely in śi and another cup-shaped letter, which resembles a ma. At least śa is not found in early Tamil epigraphy and thus this term must be local Prakritic. Whatever the complete grafﬁto said, it was expressing a Tamil name in TB-II and used a Prakritic term as well, that means it did not contain a complete Text in Tamil written in TB-II. It must have been written by a man who maintained the spelling of his own name the way he was told to do in a Tamil ambiance and the same man was also able to express himself in speaking and writing in the local vernacular and script. It would be highly welcome if the Uttiraó from Arikamedu could be used to date the same spelling at Tissamaharama. The sherd from Arikamedu was dated by Mahadevan (1996a: 315) to the 3rd century A. D. on palaeographic grounds, dangerous grounds indeed. The sherd from the Tissamaharama, however, came from the phase “a2”, several hundreds of years away from Mahadevan’s guess. 1.3.3 The initial-u problem In the text dealt with above we ﬁnd an initial u in utatitan, being a vertical with two strokes protruding to the right. In the oldest script of Aśoka an initial u is just a hook with a vertical and a short horizontal line to the right starting at the lowest end of the vertical. Such an u was necessary for Aśoka’s own language, while a consonant ra was not, as said above. Therefore, Aśoka was not aware of the problem that a ru, made from the straight ra, would look exactly like an initial u. One way out would be to use the wavy ra, another one is ﬁrst found at the stūpa railings at Bharhut, built from the 2nd century B. C. onwards. The depiction of a celestial elephant coming down to earth to enter the womb of the future Buddha’s mother carries the legend: bhagavato ukramti. Because of a Prakritic law regarding the metrical value of closed syllables, this initial u can only be short, still it shows two horizontal lines as would a long initial ū, which, however, is not attested at this early time. A second case comes from the cave of Pabhosā near Kauśāmbī, ﬁrst century B. C., where the term udakūpa, “water-well”, is written with the same two-stroked initial sign (Falk 2013b: 275 fn. 27). Apart from these Prakritic inscriptions, we have a number of Tamil Brāhmī legends (Mahadevan 2003: 318 f., 332 f., 335, 384 f.) where nothing but the short u- can be meant by the two-pronged letter. This means that the two-pronged initial u, found on the “Tamil-Brāhmī” sherd from Tissamaharama presupposes a development which is visible in northern India as well as in Tamilnadu, but only from the 2nd century B. C. onwards. Clinging to the reading of a long ū in early Ceylonese Brāhmī (Sirisoma 1990: 12) leads nowhere. 1.3.3.1 More Dravidian afﬁnities? Apart from Uttiraó some more legends point to Tamil people around, partly by their use of letters only found in Tamil Brāhmī. One case is again from a very early stratum (No. 2, ﬁg. 2). No. 2: /// [ka] òa-vela ///, “(Vessel of) the ..kaòa Vela”. The ka is probable, and could be ku, only because the lower part is missing. What looks like a da written mirrorwise is problematic and a solution may be sought considering that the last visible letter to the right is a Dravidian la, which is found in Arikamedu as well and poses no difﬁculties.10 A personal name 10 Karunaratne (1984: 73) sees his completely retrograde cave inscriptions nos. 51 and 52, both with non-reverted forms of da, at the root of the “use of the reversed form of ‘da’ to represent the ‘la’ sound”. But in both cases (cudadata, badakarika) no Dravidian sound is necessary. Owners’ grafﬁti on pottery from Tissamaharama Vela is frequent in early rock inscriptions, which Paranavitana (1970: 122a) continuously misread as velu to avoid admitting that vela is a Dravidian clan name. With regard to the problematic inverted da, Mahadevan (2003: 223) lists such a form as the Tamil letter òa from Arikamedu with strokes for -ā and -u. The ﬁrst part of the compound in no. 2 may be compared to the “layman, the artisan Vela” (upaśaka-devata-vela), whose brothers donated a drip-ledge at Naggala in Rohana (Falk 2007: 98 f.), which would allow the incomprehensible ﬁrst part /// kala- to be regarded as referring to a profession.11 Tissa-1 produced a unique piece (1E, 21 / 66,25) of pseudo-writing with 6 to 7 na in a row, from phase “d”. No. 79 writes de+(initial)e+vaha, “of Deva”. Under that number cases are cited from Tamil sites where a medial vowel is followed by an 51 Fig. 3. No. 3, Phase “b”, dinner plate, BRW. /// patharā ///, “Dinner plate (of NN?)”. initial one. This habit could have led to the spelling as found at Tissamaharama. No. 3 (ﬁg. 3) shows three letters, the last of which is a ra with a hook starting on top and bending down to the right. Such a hook is only known from the script at Bhattiprolu close to the famous stūpa-site of Amarāvatī on the Kistna. The stūpas at Bhattiprolu furnished reliquaries with inscriptions in a script which was one of several trials to make use of the Brāhmī script for the local Dravidian vernaculars. Their solution to the said problem of closing vowels was to use the ordinary horizontal stroke for a short -a, a hook for the long -ā and no stroke for no vowel at all. Until recently the legends from the reliquaries were the only occurrence of this method. In 2010 Mahadevan (Raman & Mahadevan 2010) published a gold ring with a short legend reading sivadātasa, “of Śivadatta”, where the long-ā is 11 Fig. 2. No. 2, Phase “a2”. Cooking pot, Tissa Form A, BRW. Bypassing the rarely attested Dravidian la, what looks like an inverted da could also be taken as an inverted ri, leading to °kari, “maker”, as in kumbhakārin / kumbhakāri, “potter”, or carmakārin / cammakāri, “shoemaker”. A few early Ceylonese Brāhmī ris, non-inverted but with the i-hook attached to the top of the vertical can be found in IndoSkript. 52 Harry Falk indicated by such a hook, which shows that Bhattiprolu had at least some afthermath. Now, our sherd could furnish yet another evidence. The translation presupposes that pa could as well be pa with the simple horizontal stroke, or pā with yet another hook. The surface at the upper end of the pa is abraded and everything is possible. The tha is an empty circle, distinguished from a tha by having just a dot in the middle. A dot may be there, or it may have been forgotten, which is a frequent mistake made by many scribes for centuries to come. In Tamil speaking countries, unaspirated letters are often aspirated, because Tamil does not know aspiration and its speakers often use the more unusual aspirate variants in writing to show off their skills. The tha / tha has no vowel sign and thus p(ā)thrā could be written, where pātrā in Sanskrit, “plate”, was intended. This term would require a personal name in the genitive, broken away to the left. The only point which seems certain is the presence of the Bhattiprolu vowel-hook attached to the ra. There may have been more than one Buddhist settlement around, but only one sherd (no. 126) leaves a trace of its name on a rather recent piece from phase “d2”. In no. 54 on a large storage jar it is possible to read that it was a donation to the order of monks or nuns at the Jalevihāra, that is “the monastery near the water”. No comparable number of classiﬁed clerics has ever been presented. Mahadevan (2003: 129 ff.) can list titles of Jaina monks and nuns from rock inscriptions in Tamilnadu, and (1996a) a number of names on sherds from Arikamedu, but none of the corpora of sherds derived from Buddhist communities show grades of monks with this clarity. Although at Salihundam a number of clerics of various grades must have lived, no thera, bhiksu or śramana occurs, only one bhadanta (Subrahmanyam 1964: pl. LV, no. 51). 2.2 Titles etc. and their position 2.2.1 Status-irrelevant specifications precede the name 2. Composition of the owners The mass of sherds by the side of the tank show that there was a place where dishes were cleaned. Servants from all sorts of households would have gone there. The fact that so many grafﬁti can be traced back to clerics probably simply shows that writing was a more common practice in the monasteries than in profane circles. 2.1 Clerics This afﬁnity towards writing in Buddhist monasteries is also reﬂected in the number of Buddhist laymen, particularly laywomen, who adopted the practice. No religious agency or adherents of any other religion are represented in the names on the sherds, which reconﬁrms the impression won through other means before (Falk 1993) that it was the Buddhist centres which were open to this new technique of communication. The laypeople appear in Paranavitana’s cave inscriptions as upaśaka-NN (nos. 488, 491, 542, 590, 618) and upaśika-NN (nos. 534, 542, 581, 673). The same initial position of the title is found on our sherds. Present in the caves but not found among the grafﬁti are the roommates (śadivihariya-NN, nos. 608, 614, 641) and the cohabitants (antevaśika-NN, nos. 670, 682, 787). All these terms specify secondary living conditions and do not exclude that the persons described hold titles of a worldly or monastic nature. 2.2.2 Status-relevant titles follow the name Positioned after the name a term expresses status, as seen in the worldly “minister” which always follows the NN-amata formula (Paranavitana 1970, nos. 997, 1064). Ordinary clerics as a rule place their name before their title and so a thera always comes last in NNtera (Paranavitana nos. 506 ff., 570 ff., 593, 682 etc.) without exception. The “nuns” also come last in NN-śamani (nos. 8, 161b, 224, 725, 857, Owners’ grafﬁti on pottery from Tissamaharama 969) with just one exception śamani-NN in no. 159. Also, two “monks” seem to apply the seemingly less prestigious śamana-NN system in nos. 109, 376. The standard system is perfectly mirrored at Tissamaharama with the NN-tera and NNśamani sequence. At Termez on the Oxus, the term bhiksu is used several times, and śramana (samane) only once. But here the Ceylonese rule does not apply and bhiksu as well as śramana is either preﬁxed (bhiksu-NN, no. 20a-KT) or ﬁrst positioned (bhiksusya NN-sya; nos. 20bKTff., 170-KT etc.) “Monks”, bhiksu / biku, and “nuns”, bhiksunī / bikini, do occur at Tissamaharama but only on broken sherds were the term following biku- or bikini- without exception starts with śa and breaks after it. If this was the ﬁrst letter of a personal name after the clerical title it would be a strange coincidence over all four cases. However, as the examples in the cave inscriptions show, biku- is always followed by -saµgha, written -saga in the “modern” way with sa in the caves, and with śa as in śagaśa on most of our sherds. That means, in all biku- and bikini-cases the śa- starts -śagaśa and so there is no breaking of the rule regarding the naming formula of higher clerics. This survey provides three levels of clerics, where all initiated “begging” males are called bhiksu, stratiﬁed as thera, śramana and śrāmanera (if the last-mentioned could occur). The latter two occur only in their female forms, where we have a bikiniśaga, stratiﬁed into therī at the top (if that could occur), and śramanīs and śramanerīs. That means the highest level is only present on the male side while the two lower levels are present mainly through females. A similar division is found in the Dīghanikāya (DN III 125) where the Buddha subdivides the general term bhikkhu into “old” (therā bhikkhū), “middle” (majjhimā bhikkhū) and “new” (navā bhikkhū), and does the same with the nuns (therā etc. bhikkhuniyo). 53 Fig. 4. No. 4, phase “a2”, dining plate, Tissa Form G1, BRW. Fig. 5. No. 5, phase “c1”, small cup-shaped bowl, Tissa Form I, BRW. 2.3 The community of monks, bhiksu / biku No. 4 (ﬁg. 4). /// [b]iku-śa /// “(Vessel for) the community of monks”. As shown above this should be biku-śa(gaśa); what looks like an -u in śa is either a clerical mistake or a secondary scratch. 2.3.1 Unclear cases No. 5 (ﬁg. 5). bi ///, “(Vessel of the community of) monks (or nuns)”. 54 Harry Falk 2.3.2 Advanced monks: sthavira / tera The few theras occur very early and no successor, if present, in later phases had consigned a broken dish with his name to the deposits. Since communities of nuns are dependent on male teachers and their administrative guidance advanced monks are a prerequisite for a successful order of nuns (Hüsken 1993). No. 6 (ﬁg. 6). /// data-tera[śa]///, “(Vessel of) the Thera ..datta”. No. 7 (ﬁg. 7). /// mitadata-t[e]raśa, “(Vessel) of the Thera Mitradatta”. Fig. 6. No. 6, phase “b”, water-pot, Tissa Form D. The -e-stroke in theraśa is in a disturbed surface area. Since theras are rare and the name in both cases ends in -datta an identity in person is likely. No. 8 (ﬁg. 8). ///.. tera[śa], “(Vessel) of the Thera . . .”. 2.4 The community of nuns, bhiksunī / bikini As shown above the -śa in all three cases starts -śagaśa, one further case would start with -saga. Fig. 7. No. 7, phase “b”, large cooking or serving vessel, Tissa Form A, BRW. Fig. 8. No. 8, phase “b”, water-pot, Tissa Form D. No. 9 (ﬁg. 9). bi(k)[i]ni-śa[ga] ///, “(Vessel for the com)munity of nuns”. The ni is disturbed in its lower left part. Fig. 9. No. 9, phase “b”, storage jar, Tissa Form E. Owners’ grafﬁti on pottery from Tissamaharama 55 No. 10 (ﬁg. 10). bikini-ś· ///, “(Vessel for the com)munity of nuns”. No. 11 (ﬁg. 11). /// [b]ikini-śa .. ///, “(Vessel for the com)munity of nuns”. No. 12 (ﬁg. 12). /// ni-saga .. .. /// The writing is in a style of the early centuries C. E. If this really belongs to this category, then (*biki)ni-saga was not followed by sa, but by a symbol or by another term starting with re or initial o. Fig. 10. No. 10, phase “c1”, dining plate, Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware from Northern India. Subtype C. Riveted fragment. 2.4.1 Nuns – śramaNī / śamaNi Apart from the threefold division of monks and nuns alike as cited above, there is a threefold division of nuns alone in the Vinaya (Vin I.135), where nuns (bhikkhunī), probationers (sikkhamānā), and sāmanerīs are forbidden to attend the prātimoksa recitation of the monks. Of the two lower ranks only the sāmanerīs have a counterpart in the sāmaneras on the male side, who likewise are forbidden to attend. The probationers are not found everywhere, regarded as a theoretical concept at least in China (Heirman 2008), and they are also not found in Ceylonese rock inscriptions and not on pottery from Tissamaharama. Most likely the samani (Skt śramanī) refers to the ordinated nun in early Ruhuna, and bikini (Skt bhiksukī) is the generic term covering all females, *teri, samani and samaneri, that is. Linguistically, all sibilants can be simpliﬁed to spoken / s / in the Old Sinhalese vernacular, but are then mostly written with the sign that expresses spoken / śa / in the North. The spoken sa can also change to ha in the terms under discussion, as well as in the genitive masculine ending. Since -i and -u may be interchanged, we ﬁnd the following graphical forms on our sherds: śamana, śamani, śamanu, hamina and hamini. Fig. 11. No. 11, later part of phase “c1”, serving pot, Tissa Form B1. Fig. 12. No. 12, late part of phase “c2” or early “d1”, small serving or cooking pot, Tissa Form A in transition to B2. 56 Harry Falk Fig. 13. No. 13, phase “b” or early “c1”, small bowl. Fine Grey Ware from Northern India. Fig. 16. No. 16, late phase “b” or early “c1”, dining plate, bottom of Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype A. No. 14 (ﬁg. 14). cita-śamaniya, “(Plate) of the nun Citrā”. The legend appears on the black underside. Fig. 14. No. 14, late phase “b” or early phase “c1”, dining plate. Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype A. No. 15 (ﬁg. 15). /// [śama] nuya, “(Vessel) of the nun NN”. What looks like an -ā-mātrā at the right side of the ya is only a scratch. The ending in -nu instead of -ni or -na is surprising, but also found in no. 23 below. No. 16 (ﬁg. 16). /// reva-śamanaya [pa] ///, “Di(ning plate) of the nun Revā”. For the name cf. Paranavitana 1970 no. 405. Fig. 15. No. 15, phase late “b” or early “c1”, dining plate. Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype A. No. 13 (ﬁg. 13). /// tiśa-hama[na]ya, “(Vessel) of the nun Tisyā”. There are some vertical scratches above the na, but too weak to take them as an i-mātrā. No. 17 (ﬁg. 17). śamanaya a[t]·///, “(Vessel) of the nun NN . .”. The two letters to the right are unclear. An initial a does justice to the bend at the upper left, while the lower circle is either a shorthand form of the lower bend, or the whole letter is a va with the i-hook attached to the wrong side. The last letter could be ta or śa with a number of attached vowels possible. If these letters after śamanaya start a name, then this position after the title would be against Owners’ grafﬁti on pottery from Tissamaharama 57 Fig. 18. No. 18, phase early “c1”, dining plate. Tissa Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B. Fig. 17. No. 17, late phase “b” or early “c1”, small storage vessel, Wheeler Type 18, Fine Grey Pottery from Northern India. the general formula of the type NN-śamani. At one place in the cave inscriptions (Paranavitana 1970 no. 159) the form śamani-NN is found, but in a regular compound, whereas here we had an inﬂected form plus an inﬂected name following. Two offenses against the standard seem too much. Although the space is slightly too large it should not be ruled out that the nun’s name was found in front of śamanaya the regular way. No. 18 (ﬁg. 18). /// hamaniya, “(Vessel) of the nun NN”. The script is ﬂuent and atypical, displaying Indian inﬂuence. In the classiﬁcation of Karunaratne (1984: ﬁgs. 16, 26), ma and na / na belong rather to the 4. / 5. cent. A. D. No. 19 (ﬁg. 19). /// śamani(y?) ///, “(Vessel of) the nun NN”. 2.4.2 Novice girls – śrāmaNerā / śamaNira The male śrāmanera and female śrāmanerī are novice monks and nuns before their ordination. It may be doubted that this deﬁnition was applied everywhere, since at Sanchi a male śrāmanera calls himself a banker, śresthin,12 Fig. 19. No. 19, phase “c1”, small water-pot. Tissa Form D or F. probably implying that the afﬁliation to the monastery was only partial or part-time, with his duties in civil society still maintained. From a ﬁrst century inscription from Gujarat a śramanerī is known who also functioned as a kudumbinī, “housewife” (von Hinüber 2009: 42 with fn. 25). At Tissamaharama the female 12 Majumdar (1940: 320, nos. 211–2) translates sāmanerasa ābeyakasa sethino as if sāmanera was the personal name and ābeyaka a local derivative, “from Abā”. 58 Harry Falk Fig. 20. No. 20, phase “b” or early “c1”, serving bowl, Tissa Form B1. Fig. 23. No. 23, phase early “c1”, water-pot or storage jar, Tissa Form D. (śama)neri, forms which seem to show that the femin of the novice ine form was rather a local version of śramanerā, and only exceptionally derived from its variant śramanerī. Fig. 21. No. 21, phase early “c1”, dining plate. Tissa Form rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B. No. 20 (ﬁg. 20). /// muda-śamaniraya, “(Vessel) of the novice Samudrā”. No. 21 (ﬁg. 21). tiśa-śaminiraya, “(Vessel) of the novice nun Tisyā”. The scribe loves to draw horizontal lines with an additional line leading back to the right side. No. 22 (ﬁg. 22). guta-śamaniraya, “(Vessel) of the novice nun Guptā”. The i-stroke is very faint. No. 23 (ﬁg. 23). ///.. śamanuraya, “(Vessel) of the novice girl NN”. Fig. 22. No. 22, phase early “c1”, dining plate, Tissa Form G1, BRW. śrāmanerās are relatively numerous, while male śrāmaneras are absent from the legends. On our sherds the spellings are śamanira, śamanara, śamanera, śaminira, śamanura and For the nu instead of ni, cf. no. 15 above. No. 24 (ﬁg. 24). /// da-śamanaraya ///, “(Vessel) of the novice ..dā”. The female genitive ending deﬁnes this as the property of a novice. No. 25 (ﬁg. 25). /// da-śamanara ///, “(Vessel) of the novice ..dā”. Owners’ grafﬁti on pottery from Tissamaharama 59 Fig. 26. No. 26, phase “c2”, storage pot, Tissa Form D. Fig. 24. No. 24, later part of phase “c1”, serving bowl, Tissa Form B1. Fig. 25. No. 25, later part of phase “c1”, small waterpot, Tissa Form D. Since nos. 24 and 25 come from the same ﬁnd context, it is likely that they belonged to one and the same novice girl. No. 26 (ﬁg. 26). ///(m)[i]nira[ya], “(Vessel) of the novice girl NN”. Fig. 27. No. 27, phase “d1”, water-pot, Tissa Form D. No. 27 (ﬁg. 27). /// neraya ku ///, “Water-pot of the novice nun NN”. What looks like an -ā-stroke on the ra was not considered to belong to the original writing by the excavation drawers. 60 Harry Falk Fig. 31. No. 31, phase late “c1”, dining plate, Tissa Form G1, Fine Red Ware, polished. Fig. 28. No. 28, phase “b”, cup-shaped small bowl, Tissa Form I, BRW. 2.4.3 Unclear cases Clerics seem to appear in the following cases, but either the gender is unclear or it cannot be decided between śramanīs and śramanerīs: No. 28 (ﬁg. 28). /// gupa-śaman· ///, “(Vessel of) the monk / nun / novice ..gopa / -gopā . . .”. Possibly a name like dharmagopa has to be expected. For the change from -o to -u cf. Paranavitana 1970: xxix § 18. Fig. 29. No. 29, phase “b”, water-pot, Tissa Form D. No. 29 (ﬁg. 29). /// śama[ra] /// Unclear if this is a clerical omission of a ni or another term altogether. No. 30 (ﬁg. 30). ruvala-śam[i] ///, “(Vessel of) the (novice?) nun Rūpalā”. The name is met with in cave inscriptions (Paranavitana 1970 nos. 896a, 1110, 1133). No. 31 (ﬁg. 31). /// naga-śama ///, “(Vessel of) the monk / nun / novice Nāga / Nāgā”. No. 32 (ﬁg. 32). /// [la]-śa[ma] ///, “(Vessel of) the monk / nun / novice ..la / lā”. Fig. 30. No. 30, phase late “b” or early “c1”, small carinated bowl, Wheeler Type 18, Fine Grey Ware from Northern India, NBP-like. No. 33 (ﬁg. 33). puśa-śamini ///, “(Vessel of) the (novice?) Pusyā”. Owners' graffiti on pottery from Tissamaharama 61 k\ Fig. 33. Fig. 32. No. 32, phase "cl ", dining plate, Tissa Form G 1, BRW, legend on underside. No. 33, phase "cl", small conical beaker, Wheeler Type 10, late form of Fine Grey Pottery from Northern India. Fig. 34. No. 34, phase ;;:t,::'"' ｾＺｩＧ＠ｾＢＺｯｉＧ＠ｾ＠ ed Ware", type of Fine Grey Pottery from northern India. t t' I I r I" \-,I Fig. 35. No. 34 (fig. 34). pusa-sa[manJi Ill, "(Vessel of) the (novice?) nun Pu;;ya". No. 35, most probably phase "cl", cooking or serving pot, Tissa Form A. The name is Pu;;ya, as so often; the term following it could be samani or samanira. No. 35 (fig. 35). yahala-sa[maJ, "(Vessel) of the (monk or nun) Yahala". The name is singular, possibly related to Skt yasas, cf. sramarza ---. hamarza and the name yahasini, Skt Yasasvinl (Paranavitana 1970 no. 89, 422). No. 36 (fig. 36). naga-sami Ill, "(Vessel) of the nun I novice Naga". Although there is space enough no trace of the concluding (Sami)rziya is seen. Fig. 36. No. 36, phase "c2", dining plate, Type G2; inscribed on bottom inside, Coarse Red Ware. 62 Harry Falk Fig. 37. No. 42, phase “a2-a3”, small cup-shaped bowl, Tissa Form I, BRW. There are more unclear cases. A good number needs not be illustrated and is only listed here: No. 37, phase “c1”, dining plate, Tissa Form G1: /// mana ///, No. 38, phase “b”, dining plate, Tissa Form G1, BRW: /// ta-śama[n] ///, No. 39, phase “b / c1”, water-pot, Tissa Form D: /// [?]śama[n] ///, No. 40, phase “b”, dining plate, Tissa Form G1. BRW: /// guta-śa[ma] ///, No. 41, phase “c1”, small conical beaker, Wheeler Type 10, late type of Fine Grey Pottery from northern India: /// ta-śama ///. 2.5 Laypeople No monastery can survive without laypeople providing food and other basic commodities. At Tissamaharama, male laymen are totally absent from the sherd legends. 2.5.1 Laywomen – upāsikā / upaśika Fig. 38. No. 43, phase “a3”, dining plate, Tissa Form G1, BRW. Fig. 39. No. 44, phase “b”, storage pot, Tissa Form E. It is remarkable that laywomen are present from the ﬁrst layer onwards, but until the end not a single layman. The compounds are predominantly of the upāsikā-NN-type, while NN-upāsikā is less used. Instead of the clerical designation of the NN-title-type, the laywomen’s formula seems to put more stress on their non-clerical nature, by putting the designation ﬁrst. No. 42 (ﬁg. 37). [u]paśika-ti ///, “(Vessel) of the laywoman Ti(syā)”. No. 43 (ﬁg. 38). /// [da] upaśika ///, “(Vessel of ) the laywoman (NN)”. The da at the left side is much smaller and may belong to another or an older legend. No. 44 (ﬁg. 39). /// [u]paśi<ka>-guta[ya] ///, “(Vessel) of the laywoman Guptā”. Owners’ grafﬁti on pottery from Tissamaharama 63 No. 45 (ﬁg. 40). upaśi<ka>-śude ///, “(Vessel of) the female laywoman Sude. .”. The ka of upaśika was forgotten and added below the śi. The name may be emended to Sudevī. There are parts of letters on other fragments of this pot, only one ka-śu can be read with certainty. Possibly, the text was incised for a second time, with the ka in its regular place. No. 46 (ﬁg. 41). /// [la?]ya upaśika .. ///, “(Vessel) of the laywoman NN ..la”. Fig. 40. No. 45, phase late “b” or early “c1”, storage jar, Tissa Form E. According to the standard formula the proper name of the laywoman should follow her designation as upaśika. The term here preceding it thus remains unclear. No. 47 (ﬁg. 42). upaśika-anu[ra]di ///, “(Vessel of) the laywoman Anurādhā”. The ra is rather obscure, but the female name spelled anuradi is frequent in the cave inscriptions (Paranavitana 1970: 102b). A princess (abi) spelled anuradh· is known from Anuradhapura.13 Tissa-1 adds one doubtful case, 1F, layer 12 reading ///[ś]·ka-u///, most likely another laywoman. 2.5.1 Samudrā / Śamuda and the “Tamil” dining plate Fig. 41. No. 46, phase late “b” or early “c1”, dining plate, Tissa Form G1, rim type 9; type “Rouletted Ware” in the fabric of Fine Grey Pottery from northern India, subtype A. One legend from Tissamaharama received international attention when I. Mahadevan (in Falk 2008: 63 f.) read a text in the Tamil language on it. With some distance in time and more reasoning the sensation disappears and we end up with an ordinary name and some sort of clerical “funny” scribbling. First, we need to get used to the name, being Samudrā in Sanskrit, hardly in use in northern India, but known already from the oldest cave inscriptions. Paranavitana (1970) notes in his nos. 1005 and 1010 the feminine genitive śamudaya, and in nos. 69, 774 and 1096 śamudaha as 13 Deraniyagala & Abeyratne 2000: 786 no. 16b and 784 fn. 11. As the title shows this is not a “clan” name. Fig. 42. No. 47, phase late “b” or early “c1”, small globular vessel, Tissa Form I, BRW. 64 Harry Falk Fig. 43. No. 48, phase “b”, dining plate, Tissa Form G1, BRW. the masculine genitive singular. In Tamilnadu the name occurs twice at Alagankulam on sherds with Sinhalese phonology, once only its beginning śamu///, and once in a “mixed” orthography as cāmutaha, being the Tamil ca for Sinhalese / sa / , written śa, with the long -ā of TB-II, Tamil / t / for Sinhalese / dd / , and the Sinhalese genitive masculine singular -ha (Mahadevan 1996b: 57). The genitive form samudaha is also found at least on one lead coin from the island (Bopearachchi, Falk & Wickremesinhe 2000: 131 no. 42). So far, Tissamaharama has provided only feminine and undeﬁnable forms: No. 49 (ﬁg. 44). /// śamu[da] ///, “(Vessel of) Samudra / Samudrā”. No. 50 (ﬁg. 45). /// [śa]muda ///, “(Vessel of) Samudra / Samudrā”. No. 51 (ﬁg. 46). śamudin[i] ///, “(Vessel of) Samudranikā”. Fig. 44. No. 49, phase “b”, water-pot or storage jar, Tissa Form D. Fig. 45. No. 50, phase “b”, water-pot or storage jar, Tissa Form D. The name, possibly akin to Skt *samudranikā, is derived from the common śamuda / samudrā. The form samudamnikā is found in Nagarjunikonda in Tamilnadu (Mahadevan 1996b: 57). The common use of the sufﬁx -anaka / -anikā in South India and Kanganhalli in particular is dealt with in von Hinüber 2014: 15. Fig. 46. No. 51, phase “b”, small water-pot, Fine Red Ware. Owners’ grafﬁti on pottery from Tissamaharama 65 No. 52. There is a further piece from the later part of phase “c1”, a dining plate (Tissa Form G1, BRW) inscribed on the outside of its rim, reading śamu ///, not shown here. No. 53 (ﬁg. 47). This is the piece with the alleged Tamil legend. It cannot be simply transcribed and read. The drawing shows in black the basic letters and in gray the vowels attached. There are seven graphs which I all take to be letters of the Brāhmī script. One of them has no vowel-stroke, being a clear ra, second from left. Another one, second from right, has only one vowel, the u-stroke, at its lower end. All other ﬁve letters have two vowel- or other strokes together. First, on the left, comes a la with strokes for -i and -u; following is a simple ra; next is a vertical again with an ihook above and a slanting stroke to its lower left; following is a na with i-hook above and two u-strokes below. Next is a śa with two slanting strokes on its left side. Next is mu and the closing letter is a da again with i-hook above and a slanting stroke below. This is not writing meant to convey any sense but rather a series of letters enlarged by vowel- and other strokes. As one consonant sign can only be furnished with one single vowel stroke we see ﬁve out of seven letters producing clerical nonsense. Mahadevan took letters 4 and 5 as symbols, placed inside a running text as nowhere else. There are two symbols in Paranavitana 1970 nos. 1051 and 1052, but both end a full sentence. Mahadevan took the l+i+u as a miswritten Dravidian alveaolar l+u Ý lu, and he took the d+i+u as alveolar retroﬂex ôa+i. But the form of the ôa with a forked lower end always starts with a C-bend above (Mahadevan 2003: 221 chart 5B), not with a vertical as our letter da does. That means that Mahadevan’s reading of a retrograde Tamil text (lirati Ý tirali + + muôī) with its alleged meaning “Written agreement of the assembly”) is excluded as it presupposes too many exceptions: l+u+i hardly stand for li; Fig. 47. No. 53, phase “b”, dining plate, Tissa Form G1, BRW. if ti would have to be read, the letter would have been inscribed retrograde with an -i-hook placed on top of the vertical instead of lower down the vertical as in li, ni and di; ôa would have a form which does not yet exist. Symbols in the middle of a sentence are unknown, as are Brāhmī texts on vessels written from right to left. His “text” constructs a word (tirali) which is not found anywhere else and the alleged meaning has absolutely nothing to do with a dining plate. On the other hand, if we start from a standard Ceylonese Prakrit in ordinary Brāhmī running to the right, we get only one single meaningful word, śamuda, produced by the last three letters. This would be a lady named Samudrā here as in all the other cases where this name occurs. A genitive ending is missing, but there are more cases where a nominative or stem-form is found (e. g. no. 73 supāra), – as is the rule in the Tamil-Brāhmī cases (Subbarayalu 2008: 230–236). I suspect that after śamuda was written, that either the owner or someone else “enhanced” the legend with vowel signs at the da, then to the left of the śa and then added more and more letters to the left of śa, ﬁrst na+i+u+u, then a straight ra with -i and with a stroke slanting to the left, then the wavy ra, then la with -i and -u. All additions with no meaning at all. Strange as the case is, it is not singular. One more sherd from Tissamaharama provides such “enhancements”: 66 Harry Falk water” (jalevihāra) would perfectly describe the situation by the side of the tank. For the ﬁnal bi / bā a self-imposing solution would be biku or bikini, and, as we have seen, at least bikini always is followed by śaga or another representative of saµgha. Then the intended legend could have been: jalevihāra-bi(ku / kini-sagasa), “for the community of monks / nuns at the Jalevihāra”. 2.6 Undeﬁnable males- clerics, laymen or civilians Fig. 48. No. 54, phase “c2 / d1”, large globular storage jar, Tissa Form D. Many legends are truncated in a way that it is impossible to say if the male persons are clerics, laymen or neither of both. No. 55 (ﬁg. 49). /// hadevaha, “(Vessel) of ..hadeva”. No. 56 (ﬁg. 50). /// [ka]śapa ///, “(Vessel of) Kaśyapa”, or “Din(ner plate) of ..ka”. Fig. 49. No. 55, phase “a2”, cooking pot, Tissa Form A, BRW. Either a personal name kaśapa, spoken / kassapa / , Skt kaśyapa, or a dining plate (pati), abbreviated to pa, of a person with name ending in -ka. No. 57 (ﬁg. 51). + + + + la śiva + + + taśa /// Uncertain male personal name in the genitive, probably containing śiva as a part and ending in -ta. Fig. 50. No. 56, phase “a2”, dining plate, Tissa Form G1. No. 54 (ﬁg. 48). jilevihira-bi / bā This is the reading at face value. It seems as if someone wrote jalevahara-bi and then received the order to place an -i-stroke on the va and a long ā-stroke on the ha. But instead of producing jale-vihāra-bi, the three letters ja, va and ha received -i-hooks, and bi already having one vowel expressed receiving a second one meant for another letter. Despite the clerical nonsense the term as such is remarkable, since a “monastery by the Fig. 51. No. 57, phase “a2-a3”, dining plate, Tissa Form G1, BRW. Owners’ grafﬁti on pottery from Tissamaharama 67 Fig. 52. No. 58, phase “a2-a3”, small water-pot, Tissa Form D2. Fig. 53. No. 59, phase “a2-a3”, dining plate, Tissa Form G1, BRW. Incised on the underside. No. 58 (ﬁg. 52). <symbol> śumana[ha?] ///, “(Vessel) of Sumana”. The name, Skt Sumanas, is widespread and also found in Tissa-1 (1H,no. 18,23 / 93,16) reading <moon-symbol> <swastika> śu[manaśa]. No. 59 (ﬁg. 53). [t]i[śa] ///, “(Vessel of) Tisya”. No. 60 (ﬁg. 54). /// tipaduka[sa] ///, “(Vessel) of (*A)tipānduka”. The reading is clear, the restoration unsafe. The left margin allows more letters at the beginning, for the proposed restoration Skt atikrsna and atipāndura can be compared, the latter with the same meaning “very pale”. Fig. 54. No. 60, phase “a2-a3”, dining plate, Tissa Form G1, BRW. No. 61 (ﬁg. 55). ///.. nabutiya .. ///, “(Vessel) of (Dha)nabhūti”. Male personal names ending in -bhūti, “abundance”, are frequent, cf. agibuti, idabuti, gobuti. tiśabuti, puśabuti, mahabutaya, śivabuti, none of which end in -na, while dhanabhūti is known from the stūpas at Sanchi and Bharhut. Rudimentary legends are also found on the following pieces, not illustrated: No. 62, phase “a2”, dining plate, Tissa Form G1, BRW: /// naha ///. No. 63, phase “a2”, dining plate, Tissa Form G1, BRW: /// ·iśaha. Fig. 55. No. 61, most probably phase “a”, dining plate, Tissa Form G1, Black Ware, inscribed on the underside. 68 Harry Falk Fig. 56. No. 64, phase “b”, small cup-shaped bowl, Tissa Form I, BRW. Fig. 57. No. 65, phase early “c1”, dining plate, Tissa Form G1, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B; letters incised with small bored holes. No. 64 (ﬁg. 56). /// kadevaha, “(Vessel) of ..kadeva”. The legend could be upaśaka-devaha, or compounded with a term ending in -ka, like yajñikadeva. Fig. 58. No. 66, phase early “c1”, dining plate, Tissa Form G1, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B. No. 65 (ﬁg. 57). /// taha, “(Vessel) of ..ta”. The technique is a copy of the punching of metal vessels with a pointed instrument. For a similar case in Kharosthī on a schist reliquary from Gandhara cf. Falk 2010: 16, ﬁgs. 2–3. No. 66 (ﬁg. 58). /// puśa-[śa]///, “(Vessel) of Pusya”. No. 67 (ﬁg. 59). homaha ///, “(Vessel) of Soma”. Twice the sibilant of a Prakritic *somassa has been changed to ha. Fig. 59. No. 67, phase “c1”, dining plate, Form G, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware from Northern India, latest subtype C. No. 68 (ﬁg. 60). upajhadaka ///, “(Vessel of) Upajātaka”. The / ja / is expressed with the old-fashioned jha and looks as if it was scratched out. 2.7 Undeﬁnable females – clerics, laywomen or civilians The same uncertainty as under 2.6 applies to the females in the following cases. No. 69 (ﬁg. 61). <double cup symbol> nadika .. ///, “(Vessel of lady) Nandikā . .”. Fig. 60. No. 68, phase “c1”, dining plate, Tissa Form G1, BRW. Owners’ grafﬁti on pottery from Tissamaharama 69 Fig. 62. No. 70, phase “a3”, dining plate, Tissa Form G1, BRW. Fig. 61. No. 69, phase “a2-a3”, dining plate, Tissa Form G1, BRW. The name is well-known, nasals are never written in this script. No. 70 (ﬁg. 62). /// nagaliya jha, “(Vessel) of Nāgalī”. Fig. 63. No. 72, phase “b” or early “c1”, small bowl. Type of Fine Grey Ware of Northern India. Nothing comes after jha, which is used in Ceylon for spoken / ja / ; occasionally it resembles a straight ri. Its meaning is not obvious here and maybe it was thought to be a symbol. The name is frequent in cave inscriptions (Paranavitana 1970: 113a). No. 71. Another naga+liya is found on two sherds of a dining plate of phase “b”, not shown here. No. 72 (ﬁg. 63). śumanaya <beaker symbol> ///, “(Vessel) of Sumanā”. What looks like a pu with a right vertical slightly too large could also be a logogram, which is known from other sherds in this form with two horizontals lines crossing the lower part of the vertical, which is here broken away. For a full sign cf. no. 106 and sherd 3A59-74,29 with nothing but this sign. Fig. 64. No. 73, phase early “c1”, globular bowl with beaked rim similar to “Rouletted Ware”; type of Fine Grey Pottery from Northern India. No. 73 (ﬁg. 64). ///[ka] ut·laya [m] ///, “(Vessel) of the (lady) ..ka Uttilā m . . .”. Uttilā could be an abbreviated form of a name starting with uttara-. Neither the preceding nor the following word can be reconstructed. 70 Harry Falk Fig. 65. No. 74, later part of phase “c1”, dining plate, Tissa Form G1, BRW. No. 74 (ﬁg. 65). sivala ///, “(Vessel of) Śivalā”. Fig. 66. No. 75, most probably phase “d2”, found in phase “e”; storage jar, Tissa Form D or E; Coarse Red Ware. The old letter śa, pronounced / sa / , has given way to the proper “continental” sa. For the name, throughout female, cf. śivala and sivali in Paranavitana (1970 nos. 1178, 1201) and śivali in Karunaratne (1984: 76). On further early occurrences of sīvalā, sivalā, and śivalā cf. Lüders (1963: 148). The script is the one current during the Sātavāhana dynasty, commencing in the late 1. cent. B. C. No. 75 (ﬁg. 66). /// [a?]jariya-devasama[ya] ///, “(Vessel) of the female teacher Devaśarmā”. Several divisions are possible. If the ﬁrst truncated letter is an initial a- then ajariya can only be Skt ācārya, written acariya, aciriya, ajariya and ajiriya in cave inscriptions (Paranavitana 1970: 101b). In contrast to Sanskrit, where it can follow the name, this title is preﬁxed to names in all Ceylonese cave inscriptions, in line with all other “worldly” titles. The personal name could be deva followed by a masculine genitive -sa. This would leave the following ma[ya] unexplained. A second interpretation would take devasama[ya] as the female genitive of Skt devaśarmā, a female name comparable to agisamaye at Sanchi, Skt agniśarmāyai.14 This is formally impeccable and allows us to expect female “teachers”. Although a female ācāryā is known already to Pānini (4.1,49), in Sri Lanka the term needs not be linked to religious communities, but refers also to “elephant trainers, archers and horse trainers” (Paranavitana 1970: xcv). Further disciplines may have required female teachers as well. Script and phonology compare perfectly with the inscription no. 82 at the Dakkhina Fig. 67. No. 76, phase “a2”, dining plate, Tissa Form G1, BRW. Vihāra at Anuradhapura, to be dated at the beginning of the third century.15 2.8 Plain names – males Here, names are assembled which deﬁnitely come without a title or any other speciﬁcation. Most likely, these are non-clerics, and private individuals. No. 76 (ﬁg. 67). supāra, “(The owner is) Supāra”. The -u is a stroke crossing the sa vertically. The “Indian” sa is remarkable as is the long -ā in pā. If this really is one of the oldest pieces it would show that Mauryan habits of 14 15 Lüders List no. 302; Majumdar (1940: 324), no. 245 reads erroneously agisimaye. Karunaratne (1984: 99–106; pls. 82a-q); cf. IndoSkript no. 574. Owners’ grafﬁti on pottery from Tissamaharama 71 Fig. 69. No. 78, later part of phase “b”, dining plate, NBP, Fine Grey Ware from Northern India. Fig. 68. No. 77, phase “a3”, dining plate, Tissa Form G1, BRW. expressing / sa / and long -ā in writing came ﬁrst and at the same time the ra missing in Aśoka’s ﬁrst set of letters was newly invented in a short zigzag-form on the island. Fig. 70. No. 79, phase early “c1”, dining plate, Tissa Form G1, BRW. No. 77 (ﬁg. 68). <symbol> utaraśa, “(Vessel) of Uttara”. Paranavitana (1970: cxxiv) derives this name from a naksatra like Uttaraphālguna. No. 78 (ﬁg. 69). sidhalakhitasa, “(Vessel) of Siddharaksita”. Written in a very energetic and particular hand; the Māgadhī-form with la matches the northern origin of the ware, as does the use of sa and the two aspirates. The name as such is also found as siddharaksita on a seal in the Patna Museum, acc.no. 7504. No. 79 (ﬁg. 70). de<e>vaha, “(Vessel) of Deva”. After de, the scribe wrote an initial e, which looks like a triangle and was erased subsequently. The method of combining consonants with a vowel by an initial vowel sign is known from Tamil Brāhmī, for -a+a cf. Mahadevan (2003: 315, 381) for -i+i 381. A case -e+e is probably found at Mangudi (Shetty 2003: 52 no. 4 rate-e). The name may be an abbreviation and is singular in our collection but frequent in the caves, cf. Paranavitana 1970: 112a. Fig. 71. No. 80, later part of phase “b”, tiny small carinated bowl, Wheeler Type 18, type of Fine Grey Ware from Northern India. 2.9 Plain names -females The legends are complete and deﬁnitely contain neither title nor any other designation. No. 80 (ﬁg. 71). utara[ya], “(Vessel) of Uttarā”. The name is common all over South-Asia, including one companion of Sanghamitrā (Dīpavamsa 15,77; 18,12) on her way to Sri Lanka. 72 Harry Falk Fig. 75. No. 84, phase “b” or early part of “c1”, small bowl. Type of Fine Grey Ware from Northern India. Fig. 72. No. 81, phase “b” or early “c1”, small bowl, type of Fine Grey Ware from Nothern India. Fig. 73. No. 82, phase “b”, water-pot, Tissa Form D2. Fig. 76. No. 85, phase “c1”, dining plate, Tissa Form G1, BRW. For the male name roniguta cf. Paranavitana 1970 nos. 350, 790, 926. No. 82 (ﬁg. 73). /// <beaker symbol> ron· ///, “(Vessel) of Rohin(īguptā)”. Fig. 74. No. 83, later part of phase “b” or early “c1”, dining plate, Tissa Form G1, type “Rouletted Ware” in the fabric of Fine Grey Pottery from Northern India, subtype A. Although this legend is incomplete it is likely on account of the arrangement and the style of writing that this pot belonged to the same lady as the item listed under no. 81. No. 83 (ﬁg. 74). aya-deviya pati, “Dining plate of the noble Devī”. No. 84 (fig. 75). śumanaya, “(Vessel) of Sumanā”. No. 81 (ﬁg. 72). <beaker symbol> ronigutaye, “(Vessel) of (lady) Rohinīguptā”. No. 85 (ﬁg. 76). tiśaya pati, “Dining plate of (lady) Tisyā”. Owners’ grafﬁti on pottery from Tissamaharama Fig. 77. No. 86, phase “c1”, serving vessel, Tissa Form B1. 73 Fig. 78. No. 87, phase “a”, dining plate, Tissa Form G1, BRW. No. 86 (ﬁg. 77). gu[ta]ya, “(Vessel) of (lady) Guptā”. 2.10 Undeﬁnable gender In the following cases we could have male or female owners; titles are likewise possible but not preserved, as are the expressions of relationship towards the Buddhist order. Fig. 79. No. 88, phase “a3”, dining plate, Tissa Form G1, BRW. No. 87 (fig. 78). śivika ///, “(Vessel of) Śivika / Śivikā”. No. 88 (ﬁg. 79). dāga ///, “(Vessel of) Dāga. .”. The ﬁrst letter looks like an ordinary da with a stroke to the right, a combination which would be dā on the continent. The cave inscriptions don’t use da at all in initial position. Enigmatic. No. 89 (ﬁg. 80). śagarak[i] ///, “(Vessel of) Sangharaksita / Sangharaksitā”. A common name, not used by non-Buddhists. 3 Owners mentioning their relationship to other persons Most likely these donors make a presentation to the monastery but refer to one of the monks or nuns being part of it. Fig. 80. No. 89, early part of phase “b”, water-pot or storage jar, Tissa Form D. 74 Harry Falk Fig. 81. No. 90, phase “b”, dining plate, Tissa Form G1, BRW. Fig. 82. No. 91, phase “b”, cooking pot, Tissa Form A. No. 90 (ﬁg. 81). ///.. ta-baginiya-tiśaha, “(Vessel) of Tisya, the sister-son (bhāgineya) of ..ta”. No. 91 (ﬁg. 82). /// [a]ya-tiśaha jhita-a ///, “Vessel of ) the daughter A... of the honorable Tisya”. No. 92. A further but unclear case from phase “b” is: /// jhita / / “daughter (of)”; not illustrated. No. 93 (ﬁg. 83). /// [na]śa mataya ///, “(Vessel) of the mother of ..na”. Fig. 83. No. 93, phase “c1”, dining plate, Tissa Form G1, BRW. Incised on the underside. 4 Owners of state All of the following epigraphs mentioning holders of a title are truncated and we cannot tell owners from donors. Any relationship to the Buddhist establishments may therefore be put in question. 4.1 “Big man / woman” – parumaka, parumakal The term is the Tamil parumakan, “big / important man”, many of which have made natural rock-shelters inhabitable for Buddhist monks. The wife of such a person is called parumakal. No. 94 (ﬁg. 84). parumaka-u ///, “The great man, U. .”. Fig. 84. No. 94, phase “a2”, dining plate, Tissa Form G1, BRW. It is not clear if u- starts a personal name or the term upāsaka, “lay follower”. Owners’ grafﬁti on pottery from Tissamaharama 75 No. 95 (ﬁg. 85). parumakala p· ///, “The great woman P. .”. In contrast to the preceding entry, the last letter before the break is a pa, not an u. 4.2 “Lord” – bata The term bata and its meaning is not completely clear. It occurs frequently in cave inscriptions and was explained by Paranavitana (1970: cv) as derived from Skt bhartr, nominative bhartā, with two forms in Prakrit, barata and bata, both applied to monks and laymen, translated by “Lord”.16 Examples like no. 225, bata-upatiśateraha lene, would mean that monks were still remembered as deriving from a noble family, receiving the “worldly” epithet before and the clerical one after their name. According to Sitrampalam (1990: 286) and Seneviratne (1985) bata goes back to barata / baratavar, being “a mercantile community who lived in the southern Pandyan country of Tamil Nadu”. This is the same as the parataó, “an ancient community of ﬁshermen and traders principally in pearls, chank and salt”, taken by Mahadevan (2003: 143) as standing at the root of the Ceylonese barata. Perera (2001: 90) takes it as a “title with a religious signiﬁcance”, and thus it could only be derived from bhadanta. For the sake of simplicity we prefer the solution of Paranavitana. Fig. 85. No. 95, phase “a2”, dining plate, Tissa Form G1, BRW. Fig. 86. No. 96, phase “a2”, dining plate, Tissa Form G1, BRW. No. 96 (ﬁg. 86). bata-ra[k]· ///, “(Vessel of) Lord Rak. .”. Names starting with rak (raka, raki, rakita) are frequent in cave inscriptions. No. 97 (ﬁg. 87). /// [b?]·ta-subaha .. ///, “(Vessel) of Lord Śubha (?)”. Sātavāhana script, ca. 100 C. E. Possibly the title bata and a name starting with sub-, probably a Prakrit form of Skt śubha. Tissa-1 furnished one (1C, 25 / 46,13) bataka[ra?] ///. 16 Tissa–1 (1G,24 / 88,11) reads bharatha[n / r]·/// in a carefully written post-Aśokan Brāhmī. If related it would be another case of a sanskritising “hyper-aspiration”. Fig. 87. No. 97, phase “d1”, storage jar, Form Tissa D1. 76 Harry Falk has an absolute counterpart at Kodumanal in Southern India where it is taken by Mahadevan (2003: 223 no. 89) to denote a Tamil ôa. 4.4 Princess – abi Fig. 88. No. 98, phase “b”, cooking-pot, Tissa Form A, BRW. Not found in Tissa-3 is the “princess”, abi or “hyperaspirated” abhi, Skt ambā / *ambī, while she occurs four times in Tissa-1. The ﬁrst occurrence is (1A,23 / 24,4) reading ///<symbol> abi/// with the name broken off. A complete legend (1Hno. 36,27) reads abi-[pa]liya <symbol>, “of princess Pālī”, an almost complete one (1Hno. 18,23 / 94,14) reads abi-tiśa///, “of princess Tisyā”. With a different spelling we ﬁnd abhi-[d / bh] [ā / i]/// on sherd (1H,23 / 97,15). 5 Personal property Occasionally, monks call their bowls or plates or other vessels “personal property”, Skt pudgalika. Other ways of expressing private property are not unknown. The late occurrence of such phrases may be accidental. No. 99 (ﬁg. 89). tiśaśa dhaniya, “Cereals of Tisya”. Fig. 89. No. 99, phase “c1”, small conical beaker, Wheeler Type 10, late type of Fine Grey Pottery from Northern India. 4.3 Treasurer – badagarika No. 98 (ﬁg. 88). /// dakara-ti .. ///, “(Vessel of the treasurer Ti[sya?])”. The letters, reading and sense are not wholly clear. If we assume a sloppy way of writing then the text may have been badakara-tiśaśa or similar, as translated above. Besides the standard badakarika also badakaraka and badakari are found in the caves (Paranavitana 1970: 116a; Karunaratne 1984 nos. 10, 59), but also batakarika (Karunaratne 1985 no. 11). The zigzag ra The reading is clear, but the letters are strange. The ta of ti can only be explained on the basis of a Sātavāhana ta, transformed into the angular Ceylonese Brāhmī type. The use of the aspirate dha likewise looks rather continental. What is written as dhaniya is Skt dhānya, “corn, grain, cereals”. On the frequent svarabhakti in Old Sinhalese cf. Paranavitana 1970: xxviii § 8. No. 100 (ﬁg. 90). /// taha pugi[la] ///, “personal property of ..ta”. The writing is careless, and pu could also be an initial a. Agila would be a name derived from Agni. If pugila/// really is a faulty spelling of pugalika, then polika (von Hinüber 1991: 122) at Salihundam can be compared, where not only vowels are transposed but a whole letter has disappeared. Owners’ grafﬁti on pottery from Tissamaharama 77 Fig. 91. No. 101, most probably phase “a3”, dining plate, Tissa Form G1, BRW. Fig. 90. No. 100, phase “c1”, peculiar type of Tissa Form I. 6 Donations The following cases present donations. The oldest of them use the term dāna, “donation”, later ones can be anonymous, saying only that the vessel is given for the use inside the monastery. 6.1 Personal donations In the cases cited below (§ 7.2.1), the dining plate pātrī / pati was regarded as private property by the nuns. Here, however, we see that such plates could be donated to the order through non-clerics. A similar case is found at Salihundam, where (Subrahmanya 1964: no. 28, pl. L upasikānam dane pātī) laywomen as a group donate a dining plate. No. 101 (ﬁg. 91). /// liya dane, “Donation of ..li / lī”. The scratches made large patches of the slip ﬂake off; however, the original lines are visible Fig. 92. No. 102, phase “a3”, dining plate, Tissa Form G1, Red Ware. and show that da in dane was not provided with a long-ā stroke. No. 102 (ﬁg. 92). /// ya danu ///, “Donation of lady NN”. The -ya is north-Indian, in that the vertical line in the middle touches the lower cup. No. 102a, not shown, from the early part of phase “c1” presents exactly the same reading with one illegible letter to the left of ya. 78 Harry Falk No. 103 (ﬁg. 93). śagaśa, “(Property) of the Sangha”. No. 104 (ﬁg. 94). śagaśa, “(Property) of the Sangha”. Another ///[śa]gaśa is found at Tissa-1 (1Hno. 18,23 / 93,16). 6.3 Donations linked to a special occasion, prāsādaparibhoga: Fig. 93. No. 103, early part of phase “c1”, water-pot or storage jar; Tissa Form D. Fig. 94. No. 104, phase “c1”, serving vessel, Tissa Form B1. The following vessels are predominantly deep and voluminous containers for serving, only no. 110 is a dining plate, but of a very peculiar type, possibly with a special use. These communally owned vessels were used on the elevated platform or assembly building, prāsāda, where food is served as well (cf. Mahāvastu 1.325). The same place serves for the halfmonthly recitation of the prātimoksasūtra (Hu-von Hinüber 1994: 265). In the Pali literature, more intimately connected with Sri Lanka, the term pāsādaparibhoga occurs only in two passages, and it does not occur at all in Buddhist texts from the North. It ﬁrst occurs in the Vinaya (Vin II 169, 24 ff.), Cullavagga 6.14, where a woman plans to have a pāsāda built “with an entrance terrace” (sālindam pāsādam), including a particular enhancement called hatthinakhaka.17 Answering to irritated monks, who obviously do not know if such an enhancement is allowed or not, the Buddha allows sabbam pāsādaparibhogam, “every object of use on the platform”.18 The second occur17 6.2 Anonymous donation The donation to the samgha as such is common in cave dedications. It is found on pottery too; a case from Salihundam is mentioned in Subrahmanya (1964: 34 no. 3) in Sātavāhana script. At Tissamaharama just an abbreviation of the standard formula is used, which commonly mentions the donor as well. 18 Verbally “elephant’s nail(s)”. Starting from the literal meaning we can expect something which looks like the series of half-circles which the nails form around an elephant’s foot. A suitable part of a building could be the merlons along the edge of a platform or battlement. Traditional and recent explanations of Arthaśāstra 2.3,39 and Śiśupālavadha 3,68 prefer to see here instead some sort of staircase in a city wall. paribhoga in its ordinary connotation as an abstract noun expresses the idea of “enjoyment, use”. A phrase requiring the instrumental case, °paribhogena paribhuñjati, is found frequently in the Pali literature. A second meaning occurs already in the Majjhimanikāya PTS s. 369 with a semantic shift towards an “object of enjoyment”: mamsam paribhogan ti vadāmi, “I Owners’ grafﬁti on pottery from Tissamaharama rence is in the Mahāvamsa chronicle (27,40), where in the household of a king the rinsing vessels (ācāmakumbhī) and ladles (uluµkā) are said to be made of gold, pāsādaparibhogesu, “amongst the objects used on the platform”. This latter example clearly links food and the building. All legends on our sherds where the end is preserved use only the nominative pariboge, speaking of an “object of use”. They do not use a dative or genitive *paribogaya or *paribogatha, which presupposed an abstract noun, “for the use of / at”. This reference to the object itself is in line with the scanty evidence from the Pali literature cited above. The legends from the excavation site prove that a prāsāda was close by, and the nature of the vessels shows that it was used communally for serving food. No. 105 (ﬁg. 95). /// [ya] paśadha-pariboga [b·] ///, “(Vessel donated) by (NN) as an object of use on the platform”. 79 Fig. 95. No. 105, phase “b”, small carinated bowl, Wheeler type 18, Type of Fine Grey Pottery from Northern India, NBP-like. Fig. 96. No. 106, phase “b”, serving bowl, Tissa Form B1. No. 106 (ﬁg. 96). /// [bo]ge <beaker symbol>, “... (object of) use”. No. 107 (fig. 97). /// śamaniya paśadapari[bo]:ge, <beaker symbol>, “(donation) by the nun (NN) as an object of use on the platform”. declare meat a (legal) means of subsistence”. The accusative is also found in the term vihāraparibhogam, which denotes an object offered for use by a certain monastery: (Cullavagga 7.18.1) aññatarassa upāsakassa vihāraparibhogam senāsanam aññatra paribhuñjanti, “beds and seats for the use of (a certain) monastery, (gifts) of a layman at one place, are used somewhere else”. This is why I. B. Horner in her translation (Vin translation III 216) of the Cullavagga passage uses “things which appertain” and “appurtenances” to render the crucial term °paribhoga in a compound with pāsāda°. In texts of a later date, pāsāda denotes a palace, often with many storeys. In early Ceylonese Buddhist communities nothing more than an elevated sitting platform is to be expected, either cut from the rock, as preserved in Sigiri, or built with stone slabs, as preserved in Satdhara or Sonari near Sanchi. Willis (2000: 67a) takes them to be “monastic platforms”, i.e. bases for superstructures “made of brick , wood, thatch”, something veryunlikely for reason of statics. Fig. 97. No. 107, phase “b”, serving bowl, Tissa Form B1. 80 Harry Falk No. 108 (ﬁg. 98). /// da-paribogā, “as an object of use on the ter(race)”. The legend is regarded as a truncated form of the preceding text; the -ā-stroke is not so rare a case of a vowel-stroke attached to the wrong side, so that gā has to be read as ge. No. 109 (ﬁg. 99). /// [pa]riboge, “Object of use . . .”. No. 110 (ﬁg. 100). ///.. [pa]ribo ///, “(object of) use . . .”. Fig. 98. No. 108, phase “b”, serving bowl, Tissa Form B1. Another case from Tissa-1 (1A,4–22 / 25) reading ///bo[g]e/// seems to belong to this group. 7 Names of vessels Fig. 99. No. 109, phase “c1”, small cup-shaped bowl, Tissa Form I, BRW. On the whole the local terminology is not far removed from the one at other early Buddhist monastic sites. At Termez (Fussman 2011) on the Oxus no ﬂat dining plate was inscribed or found, while water pitchers and globular pots are frequent. The globular water-pots there (nos. 43, 44, 45) are called ghada, close to Skt ghata, a term not found in Tissamaharama. 7.1 Water-pot – kunda, kundaka, The standard water-pot in Termez is spelled (paniya)-kundiya or kundika in Kharosthī (Fussman 2011: 63 f.) or pāni-kundikā in Brāhmī (2011: 69). At Termez-Čingiz-Tepa a large globular pot is named kudaga written in Kharosthī (Fussman 2011: 128 no. 3). At Salihundam for a water beaker (Subrahmanya 1964: pl. LII no. 42, kathikasa kudi rāhulasa) the shorter kudi form is used. Such and similar pots are named kuta19, kutaka or kuda on the island. Fig. 100. No. 110, phase “c1”, dining plate (Tissa Form G, rim type 9), type “Rouletted Ware” in the fabric of Fine Grey Pottery from northern India. Subtype C. 19 Cf. /// tayā kute at Anuradhapura (Deraniyagala & Abeyratne 2000: 786 ﬁg. 15, b), with dates beyond measure. Owners’ grafﬁti on pottery from Tissamaharama 81 Fig. 103. No. 113, phase “c1”, water jar, Tissa Form D2. Fig. 101. No. 111, phase “b”, water jar, Tissa Form D2. Fig. 102. D or F. No. 112, phase “b”, water jar, Tissa Form No. 111 (ﬁg. 101). /// ·i .. ha kute, “Water-pot of .i...”. No. 112 (ﬁg. 102). (e?) kutake aya-ti[ś]· ///, “Water-pot of the noble Tisya”. In front of kutake a triangle, or parts of a triangle, seem to be scratched, but because of many scratches this is not certain. No. 113 (ﬁg. 103). /// kutake, “Water-pot”. Fig. 104. No. 114, phase “d1 / d2”, globular vessel, Tissa Form D or F. No. 114 (ﬁg. 104). /// ya-kuda saga ///, “Waterpot for drink-water (for the) order . .”. Most likely, the ﬁrst part has to be emended to paniya, but see next. The old spelling kuta for kunda is here changed into kuda. The script is Sātavāhana style, as in the 1st cent. A. D. 82 Harry Falk Modern calligraphic Indian script, Andhra style of the ﬁrst / second century. More cases come from Tissa-1. 1A,4–22 reads /// kute+ ///; 1B,2–22 / 31 reads uta[raya] kuta, “Water-pot of Uttarā”. 7.2 Dining plate – pātrī / pati Fig. 105. No. 116, ﬁnd context phase “e”, water pot (Tissa Form D or F) dating within the range of phases c2-e. Dining plates (cf. fn. 9) are spelled pati in a number of places. Von Hinüber (1991: 123 f. / 817 f.) has assembled the evidence from Salihundam in Orissa, and felt that pātra and its vernacular variants most likely denote some different kind of vessel. In Salihundam the term is spelled pāti several times and clearly feminine (Subrahmanya (1964: no. 24, pl. L pāti polikā, cf. no. 35); in another case it ends in long -ī Subrahmanya (1964: no. 28. pl. L pātī), so that a derivation from pātrī cannot be doubted. In Akurugoda, patis are also donated by nonclerics; in Bactria they are found at Kara-Tepe (von Hinüber 1991: 122 / 816) while they seem to be absent from Termez. Deraniyagala (1972: pl. 1d, 129) found one rim sherd of a dining plate in Gedige, at Anuradhapura, where I see pati ///, read as pata by the excavator, who knew that it “signiﬁes bowl”. Their local distribution or absence seems to be dependent on the use of rice as a staple food, which is mixed up with condiments in it, as in a thālī. 7.2.1 Female owners No. 117 (ﬁg. 106). śonaya [p]· ///, “Di(ning plate) of Śronā”. Fig. 106. No. 117, early part of phase “b”, dining plate, Tissa Form G1, BRW. No. 16 (ﬁg. 16). /// reva-śamanaya [pa] ///, “Di(ning plate) of the nun Revā”. For the name cf. Paranavitana 1970 no. 405. No. 83 (ﬁg. 74). aya-deviya pati, “Dining plate of the noble (lady) Devī”. No. 115 (ﬁg. 27). (*NN-śama) /// neriya ku ///, “Water-pot of the novice girl NN”. No. 116 (ﬁg. 105). ///.. kudak· ///, “Water-pot (of NN)”. No. 85 (ﬁg. 76). tiśaya pati, “Dining plate of Tisyā”. Owners’ grafﬁti on pottery from Tissamaharama 83 Fig. 108. No. 119, phase “b”, dining plate, Tissa Form G1, BRW. Fig. 107. No. 118, phase “b”, dining plate, Tissa Form G1, BRW. 7.2.2 Male owners No. 118 (ﬁg. 107). /// manaśa pat[i] ///, “Dining plate of (Su)mana”. Because of the dental na, the plate belongs rather to a private Sumana than to a cleric -śramana monk, which would be written śamana-śa. Fig. 109. No. 120, phase “a3”, dining plate, Tissa Form G1, BRW. No. 119 (ﬁg. 108). puśa <letter ha ﬂaked off> ha pati, “Dining plate of Pusya”. The third letter has ﬂaked off during incising and was repeated at a safe distance. 7.2.3 Undefinable owners Since masculine -i-stems use a genitive ending -ya, as do all female stems, the following cases cannot safely be attributed to either gender. No. 120 (ﬁg. 109). /// naguliya pa[t](i), “Dining plate of Nāguli”. Unclear if the name starts with na or not, but a male naguli is found six times in the cave inscriptions (Paranavitana 1970: 113a). No. 121 (ﬁg. 110). /// [n]iya pati, “Dining plate of ..ni”. Fig. 110. No. 121, early part of phase “c1”, dining plate, Tissa Form G1, rim type 9, type “Rouletted Ware” in the fabric of Fine Grey Ware of Northern India, subtype B. 84 Harry Falk Fig. 112. No. 123, phase “a2”, cooking vessel, Tissa Form A), BRW. Fig. 111. No. 122, phase “d1 / d2”, water-pot, Tissa Form D2. The ta of ti has been scratched twice, one over the other. 7.3 Water dispenser – pānikā Globular water pots for use all through the day can be called pānikā or pāniyā, a term probably different from kundikā. No. 122 (ﬁg. 111). ///.. nuhara-hamanaha panika [ya?], “Water-pot of the monk (*A?)nusāra”. Only two eye-copies remain, differing at the right end, which makes the letter ya doubtful. The legend presents a number of singularities: The pot is only here called pānika, this is the only male śramana, and the script is “modern” and reminiscent of the cave inscriptions on the Indian west coast. 7.4 Serving vessel – bhājana / bajana Fig. 113. No. 124, phase “d1-d2”, large serving vessel, Tissa Form B2. 8 Unclear contents No. 123 (ﬁg. 112). ///.. bajhane, “Vessel”. No. 124 (fig. 113). /// [b]·ya [µa / ja / da?] rethā[b]· /// Pronounced / bajane / with old style spelling; the term derives from Skt bhājana, a “vessel”. Sātavāhana script, after C. E. 100. Meaning obscure. Owners’ grafﬁti on pottery from Tissamaharama 85 No. 125, phase “a2”, dining plate, not shown, is similarly obscure, reading /// maśaśa ///. 9 Place names and forms of incising No. 126 (ﬁg. 114). /// patanaya, “In the town . ”. Inscribed with a white liquid. Skt pattana or pattanī is a frequent second member in placenames all over eastern and southern India. If not miswritten we would have here a feminine form pattanā in the locative (pattanāyām). On a detailed map one village spelled “Patana” is found at the north-western corner of the Tissamaharama tank (Rieger 2001: 2008 ﬁg. 202). This legend is the only painted one; all others are incised after ﬁring. It is tempting to contrast the sherds from Peshawar written in Kharosthī published by Chhabra (1949 / 50: 125): all but one of 24 grafﬁti were painted while only one was incised prior to ﬁring. Does this latter go back to a man (no. 24 budhamitrasa) from the South or East? In Gandhara, inscribed water-pots generally show painted letters; at Salihundam one carefully inscribed legend was incised “before burning” (Subrahmanya 1964: 86 no. 29). On the possibility of a reconstructed term jalevihāra, “Monastery by the water-side” cf. above no. 51. 10 “Megalithic” symbols and logograms As said above, Tissa-3 as well as Tissa-1 yielded an enormous mass of those symbols which are often called “megalithic” because they ﬁrst or foremost occur on pottery found within settlements of the so-called megalithic cultures on the Deccan in the second half of the ﬁrst millennium B. C. (Falk 1993: 158 ff.). 10.1 Symbols and the glyphs from the Indus Valley Culture Similar forms occur from Gujarat down to Sri Lanka, and by various authors the system has been a) linked to or derived from Harappan Fig. 114. No. 126, phase “d2”, cooking or serving vessel, Tissa Form B2, Coarse Red Ware. glyphs, and b) regarded as a precursor to Brāhmī. The latter proposal is certainly not helpful, as both symbols and Brāhmī script can occur at the same time, even on the same vessel, as in Tissamaharama. Occasionally a stratum with only symbols on pottery overlays an earlier stratum with nothing but Brāhmī, as at Uraiyur, Tiruchirappalli (Raman 2011: 75). This shows, to my mind, that symbols and script represent rather different social groups than different time levels. Lal (1960) has mapped the occurrences of the symbols and also demonstrated the great number of similar or identical shapes. However, a great number of symbols graphically show such basic forms, as occur in almost all early scripts on the globe. Such forms cannot be used to prove dependency of any sort. A few non-basic forms have been found at Daimadabad and Navdatoli, ca. 1700 B. C., both sites in Maharashtra (Parpola 1994: 55) and show indisputably that at least some Harappan signs survived a movement to the Deccan. At Tissamaharama, there is at least one graph, not listed by Lal (1960), which seems to provide a possible further link to the Harappan sign 86 Harry Falk 11 Results 11.1 Monastic development Fig. 115. Sherd from a globular storage jar (3Ono. 60,12), phase “c2”, with cognate forms at Harappa (after Mahadevan 1977: 33) and at Uraiyur (after Raman 2011: 76). system. It consists of a pointed triangle with crossbars, topped by a horizontal line with short hanging verticals ending either side (ﬁg. 115). It is no. 206 in Mahadevan (1977: 33), it was found only in the large cities of the Indus culture. The number of cross-bars can vary (ibid.: 789). It is found in Tissa-3 with one cross-bar and in Tissa-1 (1H,22 / 95,22) with at least three, and also in Tamil Nadu in “megalithic” contexts at Uraiyur, with two cross-bars (Raman 2011: 76 f. no. 56). At least the similarity is surprising. 10.2 A new symbol Untouched by the Harappan question is a sign which seems to have been developed anew in Tissamaharama. It looks like a beaker on a stand occurring at the beginning or at the end of Brāhmī legends in our cases nos. 71, 79, 82, 89, 90.20 Although looking like a pictogram it may as well be just another logogram of a clan or family. There deﬁnitely is scope for further research. 20 On a water-pot of phase “b”, 3H,70–72,37, on a large bowl (no. 106, ﬁg. 96) from phase “b”; at the end of text (no. 72, ﬁg. 63) in phase “b”. Most of the sherds at all sites of Akurugoda are uninscribed. However, the majority of the inscribed sherds have links to Buddhist establishments at the site. For contrast we can look at Tissa-1, the workers’ quarter east of the southern ascent to the citadel. There, not a single cleric or layperson left a legend, only one legend (/// śagaśa; cf. nos. 103, 104) marks a donation to a saµgha and one vessel was meant for the platform (cf. no. 110). The following table makes the distribution of clerics, laypeople and other terms during the phases a / b / c1 / c2 / d obvious. In cases of doubt, the earliest phase is chosen for reasons of simplicity, and every sherd produces one entry, neglecting the reasonable assumption that one single person could have produced several sherds (tab. 1). There are three sherd legends mentioning a thera in phase “b” with a bhiksusaµgha already in phase “a”. A bhiksunīsaµgha leaves traces in the phases “b” to “c1”, consisting of a good number of śramanīs starting likewise in phase “b”, as do the novices, śramanerī, lasting longer, from “b” to “d”. The earliest phase has theras and one bhiksusaµgha, after that only one single indisputable corresponding male śramana occurs in phase “d”, and not a single śramanera. The laypeople consists completely of women. They occur only in the two earliest phases “a” and “b”. Donors with the Tamil title parumaka and parumakal occur only in the ﬁrst phase, as does one indisputable bata. A treasurer follows in phase “b”. Monastic activities on the platform pāsāda supporting the distribution of food occur only in phases “b” and “c1”. This material tells an amazingly consistent story: The installation of one or several Buddhist establishments, was done in phase “a” by male teachers, sthavira / teras, without further monks apparent through inscribed sherds. Support came from politically inﬂuential people Owners’ grafﬁti on pottery from Tissamaharama 87 male female unclear bhiksu / (nī)-saµgha 1/-/-/-/- -/1/3/-/- -/-/1/-/- thera / ī -/3/-/-/- - - bhiksu / nī - - - śramana / ī -/-/-/-/1 -/6/2/-/- -/1/2/1/- śramanera / ī - -/1/5/1/1 -/5/5/-/- upāsaka / upāsikā - 2/4/-/-/- - parumaka / l 1/-/-/-/- 1/-/-/-/- - bata / batagarika 1/1/-/1/- - - Tamil Brāmī 2/1/-/-/- - - Plain names 2/3/3/-/- - / 11 / 3 / - / - 3/4/3/-/- Undeﬁnable names 9/1/4/-/- 2/2/2/-/1 2/-/1/-/- Table 1. Types of owners or donors during the phases a / b / c1 / c2 / d. with Tamil titles, and a number of private women who constituted the lay community. The heyday of the order was in phases “b” and “c1” and ended with nothing but a few female nuns. Strictly speaking we do not know if there was just one monastery, vihāra, for monks, and an attached upāśraya / upāssaya for nuns, or if the nuns enjoyed a certain independence. In later times, nunneries could be subordinate to the monks’ vihāras (Perera 2005: 171), although donations to “the order of nuns in the śramanī-monastery” (hamani-vihara-bukanisagaya, Paranavitana 2001: 175) are known at least once in the 2nd century A. D., a rare exception indeed. The oldest cave and rock inscriptions are more or less silent on the organization of nun; therīs and śrāmanerīs are completely absent, and the śramanīs are few, so that Perera (2001) summarizes the early times without ever mentioning a nunnery. Paranavitana (1970: cxvii) saw 10 nuns compared to 300 monks mentioned in his collection and concludes that, “nuns in ancient Ceylon were not numerous”. The numerous and dominant groups of nuns and upāsikās make Tissamaharama a true exception not to be left out when painting the picture of the early history of Buddhism on Sri Lanka. The evidence is unmistakable provided that all people once present inscribed and broke their vessels to a similar extent. The Ceylonese Buddhist tradition has it that the ﬁrst missionary came to the island in the days of king Aśoka and Devanampiya Tissa, and so even the oldest phase should not predate ca. 250 B. C. However, as Emmanuel (2000, 83) has remarked, this story need not be absolutely accurate and a date somewhere between the parinirvāna and Aśoka is possible, at least in principle. Also, the same tradition has it that Buddhism ﬁrst came to Anuradhapura and Mihintale, but this tradition was shaped in Anuradhapura and an older tradition from and about Ruhuna could have gone lost (Emmanuel 2000, 214 f.). Theras with a separate bhiksunīsaµgha not far away reminds of the so-called schism edict of Aśoka, where he warns bhiksus as well as bhiksunīs not to attempt to break up their respective local saµgha. In case of transgression the culprits would have to wear white garments as non-clerics do and would have to live outside the residences of the monks, āvāsa. The term vihāra is not yet used in those days and the standard “monastery” was still to be designed. Aśoka calls his own 88 Harry Falk donation21 at Pātaliputra an ārāma (aśokārāma). In the Schism Edict the crucial term is āvāsa, a “place to live”, a legal term with relevance for the admission to the purifying uposatha ceremony (Bechert 1961: 23, 39). 11.2 Vessels and migrations practiced by non-clerics as well. On the whole the language is basic and lacks every trait of a living idiom. The texts are stereotypical and inelegant, and the phonology shows the heavy inﬂuence of Dravidian languages. Some letters, like ma, show the close link to Tamil Brāhmī, which nobody would date before Aśoka. On the other hand, one of the oldest sherds from Tissamaharama (no. 76) displays Mauryan traits, like sa for spoken / sa / and a written long vowel -ā, which otherwise is absent on the island for all early phases. Taking every single aspect known to me into account, I cannot see the slightest possibility that the earliest inscribed sherds can date before the middle of the third century B. C. Radiocarbon dates apparently speak a different language and this dichotomy is something I am incompetent to explain. The Fine Grey Pottery found at Tissamaharama was an import from the Mauryan homeland on the Ganges (Gogte 2001). Apart from one case in phase “b” all other sherds of this ware come from the phase called “c”. All full legends on this ware are clearly of a Buddhist nature. Time and ambiance could point to a change in attitude towards Buddhism after the take-over of the Brahminic Śunga dynasty in Pātaliputra in the course of the second century, forcing clerics to move to more forthcoming areas.22 Difﬁculties in the North may have driven monks south after the downfall of the Mauryas. Later, the monastery at the lake site seems to have been given up in the second century A. D. One sherd from Salihundam in the script of that period seems to bear witness of a move away from Ceylon. It reads (Subrahmanya 1964: pl. LI; 86 no. 29 without understanding the text): /// ·ya dana pātī [?]iya parivenam dakhina pāhāya puvo bāvītāya///, “this pātrī is the gift of (Lady) NN . . . a domicile23 after having left behind the South and (vyāvrtya?, moving to?) the East”. No reason for the shift is given, but is is apparent that a relocated Buddhist monk from the South, most likely Ceylon, was in need of a residence and equipment, which he received from this lady. The cave inscriptions have been classiﬁed (e. g. Karunaratne 1984: 51, 75, 87, ) partly according to paleographic considerations: the idiomatic use of śa for spoken / sa / and jha for / ja / was regarded as unique, old and original, while the spoken -ha for original sibilants was regarded as a phonetic development and thus younger. In our texts the standard sa for spoken / sa / is found from phase “a” onwards. The “new feature” of written and spoken ha replacing an original spoken sa is again found in phases “a” and “c”. If we take the relative sequence of strata for granted then the certainly correct graphical and phonological development found in genitive endings as śa Ý sa Ý ha cannot any 11.3 Palaeography 21 Compared to Tissamaharama, Salihundam has not a single legend in the oldest forms of the Brāhmī script, while our site has many and yields very few in clearly Sātavānaha or related scripts from the 1st / 2nd century A. D. The oldest sherds from stratum “a” contain a number of legends which are not incontrovertibly Buddhist. When writing was introduced it was seemingly 11.4 Orthography 22 23 In fact, in the Schism Edict at Sanchi and in the Minor Rock Edict at Ratanpurwa (Falk 2013a: 43 f.) he calls the saµgha at Pātaliputra “my” (me, mama) saµgha. This ﬁnds an elucidation in Bechert’s (1961) treatment of the Schism Edict(s) as such. Cf. Verardi (2011: 98): “There is a complete halt in the patronage of Buddhist monuments between Pusyamitra’s coup [in 186 B. C. HF] and c. 100 B. C.” The term parivena describes living quarters of a particular nature, not being an integral part of a vihāra, cf. Perera (2005: 168 f.). Owners’ grafﬁti on pottery from Tissamaharama longer be used for a historical stratiﬁcation and we must now say that the two forms of expressing / sa / in writing (śa and sa) and the two forms of pronouncing a genitive masculine singular ( / ssa / and / ha / ) occur together at the same time. And this same time starts right in the earliest “literate” phase “a2”. Acknowledgement This report developed over many years. During two seasons Jo Weisshaar invited me to be part of the team on the site, in other years I received excellent photographs soon after the team had returned. Heidrun Schenk never got tired explaining the different wares and their historical background. My heartfelt thanks for more than ten years of an exciting cooperation go to both of them. Thanks to Gunatilaka behind the wheel for unforgettable days in the jungle. Many thanks as well to Oskar von Hinüber for pointing out ﬂaws in an earlier version and to Blair Silverlock who again looked after comprehensible English. Anschrift: Prof. Dr. Harry Falk Joachim-Friedrich-Str. 48 10711 Berlin Germany email: [email protected] 89 References Bechert, Heinz 1961 Aśokas “Schismenedikt” und der Begriff Sanghabheda. In: Wiener Zeitschrift für die Kunde Süd- und Ostasiens 5: 18–52. Boivin, Nicole / Korisettar, Ravi / Venkatasubbaiah, P. C. 2003 Megalithic Markings in Context: grafﬁti marks on burial pots from Kudatini, Karnataka. In: South Asian Studies 19: 21–33. Bopearachchi, Osmund / Falk, Harry / Wickremesinhe, Rajah M. 2000 Earliest Inscribed Coins, Moulds, Seals and Sealings from Tissamaharama (Sri Lanka). In: The Numismatic Chronicle 160: 117–134, Pl. 10–12. Chhabra, B. Ch. 1949 / 50 Peshawar potsherds with Kharoshthi writings. In: Epigraphia Indica 38: 125–129. Coningham, Robin A. E. / Allchin, Frank Raymond / Batt, C. M. / Lucy, D. 1996 Passage to India? Anuradhapura and the Early Use of the Brahmi Script. In: Cambridge Archaeological Journal 6(1): 73–97. Deraniyagala, Siran Upendra 1972 The Citadel of Anuradhapura 1969: Excavations in the Gedige Area. In: Ancient Ceylon 2: 48–170. Deraniyagala, Siran Upendra / Abeyratne, M. 2000 Radiocarbon Chronology of Anuradhapura, Sri Lanka: A Revised Age Estimate. In: M. Taddei / G. De Marco (eds.), South Asian Archaeology 1997. Proceedings of the Fourteenth International Conference of the European Association of South Asian Archaeologists, held in the Istituto Italiano per l’Africa e l’Oriente, Palazzo Brancaccio, Rome 7–14 July 1997, pp. 759–791. Rome: Istituto Italiano per l’Africa e l’Oriente. Emmanuel, W. D. J. Benille Priyanka 2000 Civilization in its Own Words: Inscriptions and Archaeology in Ancient Sri Lanka. Dissertation, University of California, Los Angeles. Erdosy, Georg 1993 The Archaeology of Early Buddhism. In: N. K. Wagle / F. Watanabe (eds.), Studies on Buddhism in Honour of Professor A. K. Warder, pp. 40–56. Toronto: University of Toronto. Photo credits: 1–110, 112–115 Photos H. P. Wittersheim, drawings grafﬁti H. Falk; 111 H. Piyadasa. Falk, Harry 1993 Schrift im alten Indien. Ein Forschungsbericht mit Anmerkungen. Script Oralia 56. Tübingen: Gunter Narr. 90 2007 2008 2010 2013a 2013b Harry Falk Ten rock-shelter inscriptions from Ruhuna, southern Sri Lanka. In: Zeitschrift für Archäologie Aussereuropäischer Kulturen 2: 91–104. New lead coins and other inscriptions from Tissamaharama, Sri Lanka. In: Airāvati – Felicitation volume in honour of Iravatham Mahadevan, pp. 53– 66. Chennai: Varalaaru. Signature Phrases, Azes dates, Naksatras and some new reliquary inscriptions from Gandhara. In: Annual Report of the International Research Institute of Advanced Buddhology 13: 13–33. Remarks on the Minor Rock Edict of Aśoka at Ratanpurwa. In: Jñāna-Pravāha Research Journal 16: 29– 49. Hariśyenalekhapañcāśikā – Fifty Selected Papers on Indian Epigraphy and Chronology, ed. B. Schneider / I. Strauch. Bremen: Hempen. Fussman, Gérard 2011 Catalogue des inscriptions sur poteries, I,1, Introduction, catalogues, commentaires. In: S. Pidaev / T. Annaev. / G. Fussman (eds.), Monuments bouddhiques de Termez = Termez Buddhist Monuments. Publications de l’Institut de Civilisation Indienne, 79,1. Paris: Collège de France. Görsdorf, Jochen 14 C Datings of the Tissamaharama (Aku2001 rugoda) Excavation. In: H.-J. Weisshaar / H. Roth / W. Wijeyapala (eds.), Ancient Ruhuna. Sri Lankan – German Archaeological Project in the Southern Province, Vol. 1 (= Materialien zur Allgemeinen und Vergleichenden Archäologie, Band 58), pp. 279–282. Mainz: Philipp von Zabern. Gogte, Vishwas D. 2001 XRD Analysis of the Rouletted Ware and Other Fine Grey Ware from Tissamaharama. In: H.-J. Weisshaar / H. Roth / W. Wijeyapala (eds.), Ancient Ruhuna. Sri Lankan-German Archaeological Project in the Southern Province, Vol. 1 (= Materialien zur Allgemeinen und Vergleichenden Archäologie, Band 58), pp. 197–202. Mainz: Philipp von Zabern. Gombrich, Richard Francis 1992 Dating the Buddha: A red herring revealed. In: H. Bechert (ed.), The Dating of the Historical Buddha / Die Datierung des historischen Buddha, Part 2. Symposien zur Buddhismusforschung, IV,2, pp. 237–259 (= Abhandlungen der Akademie der Wissenschaften zu Göttingen, Philologisch-Historische Klasse 194) Göttingen: Vandenhoeck & Ruprecht. Heirman, Ann 2008 Where is the probationer in the Chinese Buddhist nunneries? In: Zeitschrift der Deutschen Morgenländischen Gesellschaft 158: 105–137. Hinüber, Oskar von 1991 Inscribed Vessels from Buddhist Monasteries in Termez and Salihundam. In: Pakistan Archaeology 29: 120–124 [repr. Kleine Schriften 814–818]. 2009 Cremated like a King: The funeral of the Buddha within the ancient Indian cultural context. Journal of the International College for Postgraduate Buddhist Studies 13: 33-66. Hinüber, Oskar von / Nakanishi, Maiko 2014 Kanaganahalli Inscriptions. In: Annual Report of the International Research Institute for Advanced Buddhology 17 (Supplement). Hu-von Hinüber, Hayan 1994 Das Posadhavastu. Vorschriften für die buddhistische Beichtfeier im Vinaya der Mūlasarvāstivādins, aufgrund des SanskritTextes der Gilgit-Handschrift und der tibetischen Version sowie unter Berücksichtigung der Sanskrit-Fragmente des Posadhavastu aus zentralasiatischen Handschriftenfunden herausgegeben, mit den Parallelversionen verglichen, übersetzt und kommentiert. Studien zur Indologie und Iranistik, Monographie 13. Reinbek. Hüsken, Ute 1993 Die Legende von der Einrichtung des buddhistischen Nonnenordens im Vinaya-Pitaka der Theravādin. In: R. Grünendahl / J.-U. Hartmann / P. Kieffer-Pülz (eds.), Studien zur Indologie und Buddhismuskunde – Festgabe des Seminars für Indologie und Buddhismuskunde für Professor Dr. Heinz Bechert. Indica et Tibetica 22, pp. 151–170. Bonn: Indica et Tibetica Verlag. Karunaratne, Saddhamangala 1984 Lithic and Other Inscriptions of Ceylon. Epigraphica Zeylanica 7. Colombo: Archaeological Survey of Ceylon. Lal, B. B. 1960 From the Megalithic to the Harappa: Tracing back the Grafﬁti on the Pottery. In: Ancient India 16: 4–24. Lüders, Heinrich 1963 Bharhut Inscriptions, revised by E. Waldschmidt and M. A. Mehendale. Corpus Inscriptionum Indicarum, II,2. Ootacamand: Archaeological Survey of India. Owners’ grafﬁti on pottery from Tissamaharama Mahadevan, Iravatham 1977 The Indus Script – Texts, Concordance and Tables. Memoirs of the Archaeological Survey of India, 77. New Delhi: Archaeological Survey of India. 1996a Pottery inscriptions in Brahmi and Tamil Brahmi. In: V. Begley / P. Francis Jr. / N. Karashima / K. V. Raman / S. Sidebotham / E. L. Will (eds.), The Ancient Port of Arikamedu. Volume One. New Excavations and Researches 1989–1992. Memoirés Archéologiques 22, pp. 287–315. Pondicherry: École Française d’Extrême-Orient. 1996b Old Sinhalese inscriptions from Indian ports: New evidence for ancient India-Sri Lanka contacts. In: Journal of the Institute of Asian Studies 14(1): 55– 65. 2003 Early Tamil Epigraphy. From the Earliest Times to the Sixth Century A. D. Harvard Oriental Series 62. Chennai / Cambridge (MA). Majumdar, N. G. 1940 Inscriptions. In: J. Marshall / A. Foucher, The Monuments of Sanchi. With the texts of inscriptions edited, translated and annotated by N. G. Majumdar. London; reprint Delhi 1982: Swati Publications. McCormac, F. G. / Hogg, A. G. / Blackwell, P. G. / Buck, C. E. / Higham, T. F. G. / Reimer, P. J. 2004 SHCa104 Southern hemisphere calibration, 0–11.0 cal kyr B.P. In: Radiocarbon 46(3): 1087–1092. Mitra, Debala 1972 Excavations at Tilaura-Kot and Kodan and Explorations in the Nepalese Tarai. Calcutta Paranavitana, Senarat 1970 Inscriptions of Ceylon, Vol I: containing cave inscriptions from 3rd century B. C. to 1st century A. D. and other inscriptions in the early Brāhmī script. Colombo: Department of Archaeology. 2001 Inscriptions of Ceylon Vol. II, Part II: containing Rock and Other Inscriptions from the Reign of Kanitthatissa (164–192, A. D.) to Mahasena (276–303, A. D.), editor: M. Dias. Colombo: Archaeological Survey Department. Parpola, Asko 1994 Deciphering the Indus Script. Cambridge: University Press. Perera, Lakshman S. 2001 The Institutions of Ancient Ceylon from Insriptions (from 3rd century B. C. to 830 A. D.), Vol. I. Prepared for Publication with Introduction and Supplementary Notes by S. Kiribamune and P. Senanayake. Kandy: International Centre for Ethnic Studies. 2005 91 The Institutions of Ancient Ceylon from Insriptions (from 831 to 1016 A. D.), Vol. II, Part II – Economic and Religious Institutions. Kandy: International Centre for Ethnic Studies. Rajan, K. 2008 Situating the beginning of early historic times in Tamil Nadu: Some issues and reﬂections. In: Social Scientist 36: 40–78. Raman, K. V. 1988 Excavations at Uraiyur (Tiruchirapalli) 1965– 69. Madras: University of Madras. Raman, Sankaran / Mahadevan, Iravatham 2010 A unique gold ring from Karur – and some reﬂections on the Bhattiprolu script. In: Studies in South Indian Coins 20: 33–36. Ray, Himanshu Prabha 2006 Inscribed Pots, Emerging Identies: The Social Milieu of Trade. In: P. Olivelle (ed.), Between the Empires: Society in India 300 B. C. E. to 400 C. E. New York (Oxford University Press): 113–143. Rieger, Andreas 2001 From Topography to Digital Terrain Models in the Tissamaharama Area. In: H.-J. Weisshaar / H. Roth / W. Wijeyapala (eds.), Ancient Ruhuna. Sri Lankan – German Archaeological Project in the Southern Province, Vol. 1 (= Materialien zur Allgemeinen und Vergleichenden Archäologie, Band 58), pp. 283–290. Mainz: Philipp von Zabern. Schenk, Heidrun 2001 The Development of Pottery at Tissamaharama. In: H.-J. Weisshaar / H. Roth / W. Wijeyapala (eds.), Ancient Ruhuna. Sri Lankan – German Archaeological Project in the Southern Province, Vol. 1 (=Materialien zur Allgemeinen und Vergleichenden Archäologie, Band 58), pp. 59–195. Mainz: Philipp von Zabern. Schopen, Gregory 2007 Art, beauty, and the business of running a Buddhist monastery in early northwest India. In: D. M. Srinivasan (ed.), On the Cusp of an Era – Art in the Pre-Kusāna World. Brill’s Inner Asian Library 18. Leiden / Boston (Brill): 287–317. Shetty, Ashok Vardhan 2003 Excavations at Mangudi. Chennai: Department of Archaeology, Government of Tamilnadu. Sirisoma, M. H. 1990 Brāhmī Inscriptions of Sri Lanka from 3rd century B. C. to 65 A. D. In: A. D. N. Wijesekara, 92 Harry Falk (ed.), Archaeological Department Centenary (1890–1990) – Commemorative Series, Vol. 2, Inscriptions, pp. 1–54. Colombo. Subbarayalu, Y. 2008 Pottery inscriptions of Tamil Nadu – A comparative view. In: Airāvati – Felicitation volume in honour of Iravatham Mahadevan, pp. 53– 66. Chennai. Varalaaru. Subrahmanyam, R. 1964 Salihundam. A Buddhist Site in Andhra Pradesh. Andhra Pradesh Government Archaeological Series 17. Hyderabad. Weisshaar, Hans-Joachim 2004 An der „Seidenstrasse des Meeres“: Forschungen in der Königsresidenz von Tissamaharama (Sri Lanka). In: Expeditionen in vergessene Welten. 25 Jahre Archäologische Forschungen in Afrika, Amerika und Asien (= Forschungen zur Allgemeinen und Vergleichenden Archäologie, Band 10), pp. 117–150. Aachen: Linden Soft. Willis, Michael 2000 Buddhist Reliquaries from Ancient India. London Verardi, Giovanni 2011 Hardships and Downfall of Buddhism in India. New Delhi: Manohar. Appendix 1: Provenience of the illustrated pottery sherds with Brāhmī letters from Tissamaharama discussed in this paper, assembled by Heidrun Schenk. No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 No. 9 No. 10 No. 11 No. 12 No. 13 No. 14 No. 15 No. 16 No. 17 No. 18 No. 19 No. 20 No. 21 No. 22 No. 23 No. 24 No. 25 No. 26 No. 27 No. 28 trench 3P, context no. 126, square 40 / 77, layer 20 trench 3B, context no. 422, square 56 / 89, layer 35 trench 3G, square 62 / 79, layer 30 trench 3E, square 50 / 73, layer 43 trench 3M, square 35 / 79, layer 16 trench 3G, context no. 248, square 63 / 79, layer 34 trench 3A, context no. 305 + context no. 306, square 59 / 74, layer 32 trench 3H, context no. 210, square 67 / 73, layer 36 trench 3E, context no. 180 below, square 47 / 74, layer 28 trench 3P, square 45 / 79, layer 9–11 trench 3G, context no. 214, square 63 / 80, layer 28 trench 3M, square 29 / 79, layer 12 trench 3B, context no. 251 below, square 59 / 86, layer 27 + trench 3G, square 62 / 79, layer 29 (3B 2006 / 5) trench 3A 2007 / 5+3C 2007 / 1 trench 3D 2007 / 2 trench 3A, context no. 231, square 59 / 75, layer 28 trench 3D, context no. 241, square 52 / 89, layer 34 trench 3O, context no. 53, square 23 / 94, layer 13 trench 3E, context no. 208 below, square 50 / 76, layer 30 trench 3A, context no. 229, square 59 / 79, layer 30 trench 3B, square 58 / 85, layer 25 trench 3M, square 29 / 70 N-Proﬁle trench 3A, context no. 229, square 60 / 78, layer 28 trench 3G, square 62 / 79, layer 30 trench 3G, square 61 / 79, layer 30 trench 3B, context no. 238, square 57 / 87, layer 26 trench 3L, context no. 29, square 25 / 79, layer 13 trench 3A, square 59 / 73, layer 30 Owners’ grafﬁti on pottery from Tissamaharama No. 29 No. 30 No. 31 No. 32 No. 33 No. 34 No. 35 No. 36 No. 42 No. 43 No. 44 No. 45 No. 46 No. 47 No. 48 No. 49 No. 50 No. 51 No. 53 No. 54 No. 55 No. 56 No. 57 No. 58 No. 59 No. 60 No. 61 No. 64 No. 65 No. 66 No. 67 No. 68 No. 69 No. 70 No. 72 No. 73 No. 74 No. 75 No. 76 No. 77 No. 78 No. 79 No. 80 No. 81 No. 82 No. 83 No. 84 93 trench 3A, context no. 117, square 58 / 76, layer 30 trench 3B, square 56 / 86, layer 26 trench 3E, square 46 / 75, layer 29 trench 3A, context no. 229, square 60 / 79, layer 30 trench 3M 2006 / 4 trench 3D, context no. 174, square 54 / 90, layer 25 trench 3R, square 39 / 83, layer 10 trench 3M, square 29 / 78, layer 16–17 trench 3E, context no. 305, square 47 / 79, layer 38 trench 3D, context no. 243below, square 55 / 90, layer 32 trench 3A, context no. 241below, square 57 / 78, layer 32 trench 3B, context no. 291, square 58 / 86, layer 31 trench 3D, context no. 174, square 53 / 90, layer 28 trench 3G, context no. 219, square 65 / 77, layer 34 trench 3A, context no. 229, square 59 / 79, layer 30 trench 3A, context no. 229, square 59 / 79, layer 30 trench 3A, square 59 / 79, layer 30 trench 3A, context no. 229, square 58 / 78, layer 31 trench 3F, context no. 211, square 46 / 86, layer 26 trench 3D, context no. 166, square 54 / 84, layer 23 trench 3R, context no. 142, square 37 / 84, layer 20 trench 3A, square 58 / 79, layer 35 trench 3A, context no. 340, square 59 / 76, layer 34 trench 3B, square 58 / 81, layer 36 trench 3A, square 59 / 73, layer 35 trench 3A, square 58 / 75, layer 34 trench 3A, square 59 / 77, layer 25 trench 3A, context no. 233 beside, square 60 / 73, layer 34 trench 3D, context no. 174, square 55 / 88, layer 26 trench 3H, square 70 / 72, layer 30 trench 3P, square 40 / 77, layer 11 trench 3B, context no. 254, square 59 / 85, layer 28 trench 3O, context no. 127, square 20 / 91, layer 21 trench 3D, context no. 326A, square 52 / 82, layer 35 trench 3G, context no. 236, square 62 / 78, layer 32 trench 3I, square 26 / 83, layer 8 trench 3G, context no. 196, square 64 / 80, layer 27 trench 3A, context no. 46, square 56 / 71, layer 9 trench 3A, context no. 352, square 59 / 75, layer 35 trench 3E, context no. 258, square 46– 47 / 77, layer 36 trench 3C, context no. 275, square 52 / 71, layer 35 trench 3A, context no. 117, square 58 / 76, layer 29 trench 3A, context no. 305, square 59 / 73, layer 32 + context no. 232 belowA, square 59 / 71, layer 32 (3A 2008 / 1) trench 3G, context no. 219, square 65 / 74, layer 34 trench 3H, square 70 / 72, layer 37 trench 3G, context no. 286, square 65 / 80, layer 34 (3G 2007 / 6) + 3H, square 68 / 72, layer 37 trench 3H, context no. 231, square 67 / 72, layer 37 94 No. 85 No. 86 No. 87 No. 88 No. 89 No. 90 No. 91 No. 93 No. 94 No. 95 No. 96 No. 97 No. 98 No. 99 No. 100 No. 101 No. 102 No. 103 No. 104 No. 105 No. 106 No. 107 No. 108 No. 109 No. 110 No. 111 No. 112 No. 113 No. 114 No. 115 No. 116 No. 117 No. 118 No. 119 No. 120 No. 121 No. 122 No. 123 No. 124 No. 125 No. 126 No. 127 Harry Falk trench 3A, context no. 229, square 58 / 79, layer 29 trench 3H, context no. 230, square 66 / 73, layer 37 trench 3D, context no. 236, square 53 / 88, layer 32 trench 3L, context no. 48, square 22 / 77, layer 20 trench 3R, context no. 116, square 43 / 84, layer 18 trench 3A, context no. 229, square 58 / 79, layer 29 trench 1F, context no. 89, square 25 / 76, layer 18 trench 3D, context no. 136, square 51 / 89, layer 29 trench 3R, context no. 116 below, square 43 / 83, layer 18 trench 3R, context no. 116, square 43 / 84, layer 18 trench 3G, context no. 351, square 65 / 80, layer 39 trench 3I, context no. 5, square 24 / 82, layer 3 trench 3M, context no. 144, square 27 / 78, layer 21 trench 3P 2008 / 1 trench 3I, square 22 / 83, layer 3 trench 3G, context no. 270, square 62 / 76, layer 34 trench 3B, context no. 338, square 56 / 85, layer 33 trench 3B, context no. 254, square 59 / 85, layer 29 trench 3P 2008 / 5 trench 3G, square 64 / 72, layer 34 trench 3A, context no. 227, square 58 / 78, layer 27 trench 3A, context no. 229, square 58 / 77, layer 29 trench 3A, context no. 229, square 59 / 79, layer 28 trench 3G, context no. 143, square 64 / 78, layer 27 trench 3D, context no. 174, square 54 / 90, layer 26 trench 3E, context no. 269, square 46 / 72, layer 37 trench 3B, context no. 282 belowA, square 58 / 85 trench 3M, context no. 170, square 35 / 76, layer 24 trench 3H, context no. 109, square 69 / 78, layer 27 = no. 27 trench 3G, square 63 / 80, layer 11 trench 3B, context no. 378, square 58 / 90, layer 33 trench 3A, context no. 302, square 58 / 73, layer 32 trench 3R, square 43 / 81, layer 16 + context no. 69 below, square 43 / 81, layer 18 trench 3D, context no. 315, square 53 / 89, layer 34 trench 3M, context no. 80in, square 27 / 79, layer 16 trench 3D, square 51 / 84, layer 11 trench 3F, context no. 276, square 48 / 83, layer 33 trench 3D, square 55 / 88, layer 17 trench 3E, square 49 / 76 layer 39 trench 3I, context no. 76, square 27 / 83, layer 13 trench 3O, context no. 60, square 22 / 91, layer 12

Log In

Owners' graffiti on pottery from Tissamaharama

Sign up for access to the world's latest research

Abstract

Related papers

Related topics