Pramāṇa · Search · About · API

The Indus Script: Texts, Concordance and Tables c9: Introduction

The Indus Script: Texts, Concordance and Tables c9

Language
English

Contents

Introduction

1. The Indus seals and the discovery of the Indus Civilization

c9.s1.1

The Indus Civilization ranks among the most ancient urban cultures of mankind. It covered an appreciably larger area than either the Early Dynastic Egypt or Sumer. Like the other Old World civilizations, the Indus Civilization seems to have grown from the skilful utilization of the fertile river valleys. Its distinctive characteristics - the gridiron layout of the cities, their elaborate drainage, and the puzzling pictographic script - are still best known from the excavations of Harappa and Mohenjo-daro.¹

c9.s1.2

¹ For the Indus civilization in general, its context and for the latest developments, see the works quoted in footnotes 2-26 and the following: Gregory L. Possehl (ed.), *Ancient Cities of the Indus*, New Delhi 1979; id. (ed.), *Harappan Civilization*, New Delhi 1982; A.H. Dani (ed.), *Indus Civilization - New Perspectives*, Islamabad 1981; B.B.Lal & S. P. Gupta (eds.), *Frontiers of the Indus Civilization*, New Delhi 1984; *Vergessene Städte am Indus: Frühe Kulturen in Pakistan vom 8. bis 2. Jahrtausend*, Mainz 1987; and the series *South Asian Archaeology*, from 1971 onwards.

c9.s1.3

Three distinctive seals - representing the object type that has remained most characteristic of the Indus civilization - were found at Harappa in the Punjab and published in 1875, 1886 and 1912.² But the full implication of these finds was not realized before excavations were started in 1920 at Harappa, and, by chance almost simultaneously, in 1922, at Mohenjo-daro in Sind, some 600 km south in the Indus valley. More seals of the same type were immediately found at both of these sites, and it became evident that an entirely unknown bronze-age civilization had come to light.³ This led to large-scale excavations at Mohenjo-daro and Harappa,⁴ followed by much more limited digs at Chanhujo-daro further south in Sind.⁵ The bulk of the Indus (or Harappan) seals and inscriptions available to research comes from these excavations in the 1920's and 1930's.

c9.s1.4

² Cf. A. Cunningham, Harappa, in: *Archaeological Survey of India, Report for the year 1872-73*, Vol. V, Calcutta 1875, 105-108 & pl. XXXII-XXXIII; M. Longworth Dames, Old seals found at Harappa, *The Indian Antiquary* XV: 179, January 1886, 1; J.F. Fleet, Seals from Harappa, *Journal of the Royal Asiatic Society* 1912, 699-701.

c9.s1.5

³ For a comprehensive history of Harappan studies, see Michael Jansen, *Die Indus-Zivilisation: Entdeckung einer frühen Hochkultur*, Köln 1986.

c9.s1.6

⁴ John Marshall (ed.), *Mohenjo-daro and the Indus Civilization*, I-III, London 1931; Ernest Mackay, *Further Excavations at Mohenjo-Daro*, I-II, New Delhi 1938; M.S. Vats, *Excavations at Harappa*, I-II, New Delhi 1940.

c9.s1.7

⁵ Ernest Mackay, *Chanhu-daro Excavations 1935-36*, American Oriental Series 20, New Haven 1943. Mackay's excavations were a follow-up of the pioneering explorations of N.G. Majumdar. The Indus seals and inscriptions from Amri, Jhukar and Lohumjo-daro included in this volume were discovered by Majumdar and reported by him in the *Annual Report of the Archaeological Survey of India for 1927-28* (1931), 76-83 & pl. XXVII-XXX (Excavations at Jhukar) and in his *Explorations in Sind*, Memoirs of the Archaeological Survey of India 48, Delhi 1934. The seals from Mehi and Shahi-tump were discovered by another famous early explorer, Sir Aurel Stein; see his report, *An Archaeological Tour in Gedrosia*, Memoirs of the Archaeological Survey of India 43, Calcutta 1931.

c9.s1.8

In the partition of British India in 1947, all the major sites of the Indus civilization known at that time became part of Pakistan. During the last four decades, due to constant efforts of the Indian archaeologists, more than 862 Early Harappan, Harappan and Late Harappan sites have been discovered in the Indian Union. In India the area of distribution of Harappan settlements runs broadly from Manda in Jammu (Jammu & Kashmir) in the north to Daimabad in Maharashtra in the south, and from Desalpur, District Kutch, Gujarat, in the west to Hulas in district Shaharanpur, U.P., in the east.⁶ Among the newly extensively excavated sites in India are Kalibangan, Lothal, Surkotada, Daimabad and Banawali. These important excavations have thrown fresh light on the cultural style of the Indus civilization and have produced the greatest number of seals and inscriptions in India.⁷ The recently started marine archaeology along the coast of Gujarat has already produced exciting results at Bet Dwaraka.⁸

c9.s1.9

⁶ For the progress of the Harappan discoveries in India, see the yearly reports in *Indian Archaeology - A Review*, from 1953/54 through 1984/85 and the following: Ancient India Vol. 9 (1953); Jagat Pati Joshi, Madhu Bala and Jassu Ram, The Indus Civilization: A Reconsideration on the Basis of Distribution Maps, in: B.B. Lal and S.P. Gupta (eds.), Frontiers of the Indus Civilization, New Delhi 1984, 511-530; Jagat Pati Joshi, Harappa Culture: Emergence of a New Picture, in: K.N. Dikshit (ed.), Archaeological Perspectives in India Since Independence, New Delhi 1985, 51ff.; B.M. Pande and B.D. Chattopadhyaya, Archaeology and History: Essays in Memory of Shri A. Ghosh, Vol. I-II, New Delhi 1987.

c9.s1.10

⁷ The final reports for only the excavations of Lothal and Daimabad have been published so far: S.R. Rao, *Lothal, a Harappan Port Town* (1955-62), Vol. I-II, Memoirs of the Archaeological Survey of India 78, New Delhi 1979-1985; and S.A. Sali, *Daimabad 1973-79*, Memoirs of the Archaeological Survey of India 83, New Delhi 1986. For the Late Harappan sites of Rangpur and Prabhas Patan (Somnath) and the non-Harappan site of Maski (relevant for the Corpus on account of its cylinder seal), see S. R. Rao, Excavations at Rangpur and other explorations in Gujarat, *Ancient India* 18-19 (1962-63), 5-207; J.M. Nanavati et al., *Somnath 1956*, Baroda 1971; and B. K. Thapar, Maski 1954: A chalcolithic site of the southern Deccan, *Ancient India* 13 (1957), 4-142.

c9.s1.11

⁸ See *The Statesman*, Delhi, April 17, 1985.

2. The Indus seals and the external contacts of the Indus Civilization

c9.s2.1

Immediately after the first news about the discovery of the Indus Civilization was published in 1924,⁹ it became apparent that the Harappans had been in contact with the ancient cultures of West Asia. Evidence for this was Indus seals coming from Susa, Ur, and other Mesopotamian sites;

c9.s2.2

⁹ See John Marshall, First light on a long forgotten civilization, *The Illustrated London News*, 20 Sept. 1924, 528-532, 548.

c9.s2.3

For the other Harappan sites in India with objects included in this volume, see the preliminary yearly reports in *Indian Archaeology - A Review* (= IAR, quoted here gratefully from the unpublished index by Gregory L. Possehl): Alamgirpur, 58-59: 50-55; Banawali, 83-84: 24-29; Chandigarh, 70-71: 7-8; Desalpur, 63-64: 10-12; Hulas, 82-83: 100-103; Kalibangan, 60-61: 31-33; 61-62: 39-44; 62-63: 20-31; 63-64: 30-39; 64-65: 35-39; 65-66: 38-41; 66-67: 31-33; 67-68: 42-45; 68-69: 28-32; Pabumath, 78-79: 67-68; 80-81: 14; Prabhas Patan, 55-56: 7-8; 56-57: 16-17; 71-72: 12-13; 75-76: 13; 76-77: 17-18; Rohira, 82-83: 65-66; Rojdi, 57-58: 18-23; 58-59: 19-21; 62-63: 8; 64-65: 12; 82-83: 28; 83-84: 19-20; Rupar, 53-54: 6-7; 54-55: 9; Surkotada, 70-71: 13-15; 71-72: 13-21. For Dholavira, cf. D.K. Vaidya, The Harappan seal from Kotda Timba near Dholavira in Khadir, *Bulletin of the Baroda Museum* 25 (1973-74), 33-36 & pl. XI; for Rakhigarhi, cf. Eric Grinstead and Asko Parpola, A Harappan seal from Rakhigarhi, Haryana, *Acta Orientalia* 35 (1973), 103-114 & pl. I; for Tarkhanewala-dera, cf. A. Ghosh, The Rajputana Desert - its archaeological aspect, *Bulletin of the National Institute of Sciences of India* 1 (1952), 37-42.

c9.s2.4

among these were both square stamp seals of a purely native Harappan type and seals combining Harappan and local elements such as the cylinder form.¹⁰

c9.s2.5

¹⁰ See C. J. Gadd, Seals of Ancient Indian style found at Ur, *Proceedings of the British Academy* 18, 1932, 406-422 & pl. I-III. See now also Dilip K. Chakrabarti, Seals as an evidence of Indus - West Asia interrelations, in: Debiprasad Chattopadhyaya (ed.), *History and Society: Essays in honour of Professor Niharranjan Roy*, Calcutta 1978, 93-116; Robert H. Brunswig, Jr., Asko Parpola and Daniel Potts, New Indus type and related seals from the Near East, in: Potts (ed.) 1983 (see fn. 12), 101-115 & pl. I-III.

c9.s2.6

Later, a few round Indus seals (a type rarely found in the Indus valley) were discovered along with a large number of local round stamp seals¹¹ on the islands of Failaka and Bahrain in the Gulf, where excavations since the 1950's have revealed a flourishing "Dilmun Civilization".¹² Furthermore when one purely "Dilmun-type" seal (L-123) was found at Lothal,¹³ much attention was paid to cuneiform sources dealing with the early maritime trade of Mesopotamia. Three foreign countries are referred to as participants of the sea trade: Dilmun (closest to Mesopotamia), Magan and (farthest away) Meluḫḫa. Magan is now widely identified with Oman and the opposite coast of Makran, and Meluḫḫa with the Harappan realm. Some tablets refer to a village of Meluḫḫans residing near Lagash for generations.¹⁴

c9.s2.7

¹¹ See Geoffrey Bibby, The 'Ancient Indian Style' Seals from Bahrain, *Antiquity* 32, 1958, 243-246 & pl. XXVI-XXVII; Brunswig et al. (see footnote 9); and Poul Kjærum, *Danish Archaeological Investigations on Failaka, Kuwait. Dilmun/Failaka - The Second Millennium Settlement*, Vol. 1:1 Stamp and Cylinder Seals, Jutland Archaeological Society Publication XVII:1, Aarhus 1983.

c9.s2.8

¹² See Geoffrey Bibby, *Looking for Dilmun*, Pelican Books, Harmondsworth, 1972; Daniel Potts (ed.), *Dilmun: New Studies in the Archaeology and early history of Bahrain*, Berliner Beiträge zum Vorderen Orient, 2, Berlin 1983; Amiet 1986 (see fn. 20), 171-180. For the cuneiform references to Dilmun, see Pettinato 1972 (see fn. 14).

c9.s2.9

¹³ Cf. S.R. Rao, A 'Persian Gulf Seal from Lothal, *Antiquity* 37, 1963, 96-99 & pl. IX-XI. Some further Dilmun seals are reported from the submerged city of Bet-Dwaraka off the Kathiawar coast (cf. *The Statesman*, Delhi, April 17, 1985), but they are yet to be published.

c9.s2.10

¹⁴ A. Leo Oppenheim, The seafaring merchants of Ur, *Journal of the American Oriental Society* 74 (1954), 6-17; Giovanni Pettinato, Il commercio con l'estero della Mesopotamia meridionale nel 3. millennio av. Cr. alla luce delle fonti letterarie e lessicali sumeriche, *Mesopotamia* 7, 1972, 43-166; Simo Parpola, Asko Parpola and Robert H. Brunswig, Jr., The Meluḫḫa village: evidence of acculturation of Harappan traders in late third millennium Mesopotamia? *Journal of the Economic and Social History of the Orient* 20:2, 1977, 120-165.

c9.s2.11

The gradual evolution of the Indus Civilization from the earlier neolithic cultures of the Indo-Iranian borderlands and the relationship of these cultures with those of the ancient Near East and particularly with the cultures of the Iranian plateau and Turkmenia have started being properly understood only during the past decade or so. The French excavations at Pirak (1968-74)¹⁵ and at Mehrgarh, Sibri and Nausharo (1974-87) have been really revolutionary in providing an unbroken stratigraphic sequence from the early 7th millennium to the middle of the 1st millennium B.C. in the Kachi plain, which leads from the Indus valley to the highlands of Baluchistan.¹⁶ The Italian excavations at Shahr-i Sokhta in Seistan,¹⁷ the American excavations at Tepe Yahya in Southeastern Iran,¹⁸ and the Soviet excavations in Central Asia¹⁹ are just some of the other crucial archaeological research projects of recent times which have created a veritable explosion of knowledge. The emerging new picture stresses the leading role played by the Proto-Elamites in the increasing cultural interaction in the Iranian plateau during the first half of the third millennium B.C.²⁰

c9.s2.12

¹⁵ J.-F. Jarrige, M. Santoni & J.F. Enault, *Fouilles de Pirak* I-II, Paris 1979.

c9.s2.13

¹⁶ See J.-F. Jarrige, Chronology of the earlier periods of the Greater Indus as seen from Mehrgarh, Pakistan, in: B. Allchin (ed.), *South Asian Archaeology 1981*, Cambridge 1984, 21-28; id., Continuity and change in the North Kachi

c9.s2.14

¹⁷ Maurizio Tosi (ed.), Prehistoric Seistan, Vol. 1, Roma 1983.

c9.s2.15

¹⁸ C.C. Lamberg-Karlovsky, Excavations at Tepe Yahya, Iran, 1967-1969, American School of Prehistoric Research Bulletin 27, Cambridge, Mass. 1970; id., Tepe Yahya 1971 - Mesopotamia and the Indo-Iranian Borderlands, Iran 10 (1972), 89-100.

c9.s2.16

¹⁹ See Philip L. Kohl (ed.), The Bronze Age Civilization of Central Asia: Recent Soviet Discoveries, New York 1981; Viktor I. Sarianidi, Die Kunst des alten Afghanistan, Leipzig 1986.

c9.s2.17

²⁰ Cf. C.C. Lamberg-Karlovsky, Third millennium structure and process: From the Euphrates to the Indus and the Oxus to the Indian Ocean, Oriens Antiquus 25: 3-4 (1986), 189-219; Pierre Amiet, L'âge des échanges inter-iraniens 3500-1700 avant J.-C., Notes et documents des musées de France 11, Paris 1986.

c9.s2.18

Another revelation is the expansion of the Bronze Age Civilization of Northeast Iran during the second half of the third millennium from the Gorgan plain (Tepe Hissar III and related sites) to Southern Turkmenia (Namazga V and related sites), to Seistan (Shahdad), to ancient Bactria (Dashly and Sapalli in Northern Afghanistan), to Baluchistan and to the Indus valley.²¹ The current excavations of Sibri and Nausharo near Mehrgarh have proved that intrusive NE Iranians became a dominant element in the lower Indus valley around 2000 B.C, and that their merging with the Indus Civilization started the Late Harappan period.²²

c9.s2.19

²¹ Cf. the literature in fn. 19 and 20. There are weighty reasons to assume that the NE Iranians represent the first wave of Aryan-speaking immigrants in South Asia, cf. Roman Ghirshman, L'Iran et la migration des Indo-aryens et des Iraniens, Leiden 1977; Homer L. Thomas, Archaeological evidence for the migrations of the Indo-Europeans, in: Edgar C. Polome (ed.), The Indo-Europeans in the Fourth and Third Millennia, Ann Arbor 1982, 61-86; A. Parpola, The pre-Vedic Indian background of the srauta rituals, in: Frits Staal (ed.), Agni: The Vedic Ritual of the Fire Altar, Vol. II, Berkeley 1983, 41-75; moreover, the three-walled fortress of Dashly suggests that the NE Iranians were the Dāsas mentioned as their enemies by the Rigvedic Aryans, cf. A. Parpola, The Sky Garment, Studia Orientalia 57, Helsinki 1985, 76-78.

c9.s2.20

²² Cf. Jean-François Jarrige, Les relations entre l'Asie centrale méridionale, le Baluchistan et la vallée de l'Indus à la fin du 3e et au début du 2e millénaire, in: L'Archéologie de la Bactriane Ancienne, Paris 1985, 105-120; id., Der Kulturkomplex von Mehrgarh (Periode VIII) und Sibri, Der "Schatz" von Quetta, In: Vergessene Städte am Indus: Frühe Kulturen in Pakistan vom 8.-2. Jahrtausend v. Chr., Mainz 1987, 102-111.

c9.s2.21

This immigration is reflected in the seals of the Jhukar period at Chanhujo-daro in Sind (C-41 to 50, especially C-49 and C-50) and at Shahi-Tump (Sht-1) and Mehi (Mehi-1) in Baluchistan. Distant interaction between the NE Iranian and Indus Civilizations is evidenced earlier during the Mature Harappan period. While two Harappan seals have been unearthed at Altin Tepe in southern Turkmenia,²³ one clearly NE Iranian type stepped seal comes from Harappa (see H-166).24 The

c9.s2.22

²³ See V.M. Masson, Seals of a Proto-Indian Type from Altyn-depe, in: Kohl (ed.) 1981 (see fn. 19), 149-162, with

c9.s2.23

plain (Baluchistan, Pakistan) at the beginning of the second millennium B.C., in: J. Schotsmans and M. Taddei (eds.), South Asian Archaeology 1983, Naples 1985, Vol. 1, 35-68; etc.

c9.s2.24

few cylinder seals found at the Indus sites have so far been thought to indicate connections with Mesopotamia, where this seal type is most characteristic. However, we now know that the NE Iranian civilization, too, used cylinder seals (which there, of course, ultimately go back to Mesopotamian inspiration);25 indeed, one cylinder seal from Mohenjo-daro (M-419) resembles more closely the NE Iranian type cylinders in having an engraved motif at the round ends, too. It seems that the cylinder seals of Daimabad (Dmd-4) and Maski (Msk-1) continue the NE Iranian tradition.

c9.s2.25

Thus the seals have played a leading role in the discovery of the Indus Civilization and its external relations. They continue to be centrally important in the archaeological study of the bronze age, not least as chronological indicators.26

3. The function and iconography of the Indus seals and tablets

c9.s3.1

Preserved ancient seal impressions prove that the Indus seals have served as instruments of control in administration and trade, as in ancient West Asia.27 Some seal impressions have been made, undoubtedly by the potter, on wet clay pots before firing (cf., e.g., M-420 to 424). Other impressions have survived on clay tags, once attached to bales of goods whose integrity they thus guaranteed. The most important collection of such labels comes from the burnt warehouse of Lothal (L-124 ff.).28 The study of the seals and seal impressions in combination with their archaeological contexts and details of style and manufacture can significantly contribute to the understanding of the economic and administrative aspects of an ancient civilization.29

c9.s3.2

The quality of the seal increases with its size, and the largest and most expensive seals must have belonged to important persons or institutions. Since the seals were probably worn in a visible

c9.s3.3

Kohl's note ibid., xix. 24 The unique T-shaped seal H-165 may also be of NE Iranian origin. 25 Cf. Marielle Santoni, Sibri and the South Cemetery of Mehrgarh: third millennium connections between the northern Kachi Plain (Pakistan) and Central Asia, in: Bridget Allchin (ed.), South Asian Archaeology 1981, Cambridge 1984, 57; I.S. Masimov, Novye nakhodki pecatey epokhi bronzy c nizovie Murgaba, Sovetskaya Arkheologiya 1981: 2, 132-150. 26 The dating of the Indus civilization continues to be a controversial issue. Based on the Near Eastern contacts evidenced by the Indus seals and by cuneiform references to sea-borne contacts with the far-off country of Meluḫḫa since the times of Sargon the Great (c. 2350 B.C.), as well as radiocarbon dates, the time bracket for the mature urban phase is conventionally placed between 2550/2300 and 2000/1700 B.C. Cf. Bridget and Raymond Allchin, The rise of civilization in India and Pakistan, Cambridge 1982; D.P. Agrawal, The archaeology of India, Scandinavian Institute of Asian Studies Monograph Series 46, London 1982. 27 Cf. McGuire Gibson and Robert D. Biggs (eds.), Seals and Sealing in the Ancient Near East, Bibliotheca Mesopotamica 6, Malibu 1977. 28 See S.R. Rao, Lothal: A Harappan port town (1955-62), Memoirs of the Archaeological Survey of India 78, Vol. I, New Delhi 1979, 111-114. 29 Cf. E. Fiandra and P. Ferioli, A proposal for a multi-stage approach to research on clay sealings in protohistorical administrative procedures, in: B. Allchin (ed.), South Asian Archaeology 1981, Cambridge 1984, 124-127, with further literature. Ute Franke-Vogt (Freie Universität, Berlin) is preparing a doctoral thesis on the seals of Mohenjo-daro. Cf. also the forthcoming study of Paul Rissman referred to in fn. 84.

c9.s3.4

fashion by their owners, as is suggested by the cord holes, they are likely to have secondarily functioned as indicators of the wearers' rank, seen at a distance by the size of the seal.³⁰

c9.s3.5

³⁰ Cf. Asko Parpola, The relative size of the seals and other clues to the royal titles of the Harappans, *Tamil Civilization* 1986: 3-4, special number on the Indus script edited by I. Mahadevan (in press). 31 See M.S. Vats, *Excavations at Harappa*, Delhi 1940, Vol. I, 324ff. 32 Cf. E. Mackay, *Further Excavations at Mohenjo-daro*, Delhi 1938, Vol. I, 349-351. 33 Cf. Asko Parpola, On the Harappan 'yoke-carrier' pictogram and the kāvaḍi worship, *Proceedings of the Fifth International Conference-Seminar of Tamil Studies* held in Madurai, South India, 5-10 January 1981, Madras 1981, Vol. I, 2: 73-89; cf. also B. Hrozny', *Inschriften und Kultur der Proto-Inder von Mohenjo-daro und Harappa* (ca. 2400-2100 B.C.): Ein Entzifferungsversuch, *Archiv Orientalni* 13 (1942), 36. 34 For an analysis of the copper tablets, see Asko Parpola, Tasks, methods and results in the study of the Indus script, *Journal of the Royal Asiatic Society* 1975:2, 196ff. with fig. 12, and Paul Yule, *Figuren, Schmuckformen und*

c9.s3.6

Some of the seals, such as M-319, are carved hollow and provided with a lid so that something - most probably a magic charm - could be kept inside. This has generally been taken to support the old hypothesis that the seals, besides their primary function as administrative instruments, also served as protective amulets. In addition to the script, the majority of the Indus seals contain iconographic motifs, whose clearly religious nature has suggested an amuletic function. The pictorial motifs not only rank among the very best preserved examples of Harappan artistic expressions but also provide some of the most important clues to the Harappan religion and to the accompanying inscriptions.

c9.s3.7

In addition to being found on the seals, iconographic motifs are found in particular on "tablets". An important general characteristic of this category of objects is that they comprise many identical duplicates. The incised "miniature tablets" from the lower levels of Harappa are the earliest known examples of the fully developed Indus script.31 Later, incised tablets give way to embossed ones, often massproduced in moulds. Sometimes great numbers of similar tablets (especially H-252 ff.) have been found together, or their find places are very close to each other. This has suggested that most of the tablets, both the embossed and the engraved ones, may have functioned as tokens of votive offerings or of visits to temples.32

c9.s3.8

The inscriptions of the tablets point to such a conclusion, too. Many of the tablets have on one side a U-shaped sign which looks like a pot drawn in profile; it is preceded by zero to four vertical strokes that clearly stand for numbers. In M-494 A and M-495 A, there is a sequence of three U-shaped signs in succession: this may be another way to write the sequence 3 + U occurring on numerous tablets and apparently meaning "three pots". Sometimes the U-shaped sign on the reverse of tablets is held in the hand of a kneeling or standing man-sign (cf. H-247 A). In the moulded tablets M-478 and M-479, the sign combination of 4 + U stands next to an iconographic scene where a kneeling worshipper extends a pot shaped like the U-formed sign towards a tree. Apparently the tree is sacred, and the man is presenting the pot (or according to the inscription, four pots) to it as an offering.33

c9.s3.9

The engraved copper tablets of Mohenjo-daro form an unusual class of inscribed objects, in that their inscriptions and iconographic motifs are clearly interrelated; this is not so obvious in other classes of Indus inscriptions, although cases like the above cited tablets M-478 and M-479 may occasionally be found.34

c9.s3.10

The interpretation of the iconography of the Indus seals and tablets constitutes a major scholarly challenge. Various comparisons have been made with the ancient West Asian glyptics as well as with the later art of classical India.³⁵ Although it is impossible to go into detail here, one further example may be briefly mentioned because of its intrinsic interest and also in order to point out that these two kinds of comparisons need not be mutually exclusive. Sir John Marshall's identification of a "Proto-Siva" in the buffalo-horned deity of a famous seal from Mohenjo-daro (M-304)³⁶ may well be correct, and so may be Alf Hiltebeitel's even more convincing identification of this figure as "Proto-Mahiṣa"37, although this deity and his "yogic posture" have close counterparts in the earlier glyptic art of the Proto-Elamites.³⁸ Comparative studies thus suggest that the Indus Civilization may have been an integral if marginal part of the West Asian cultural area and that there is an unbroken cultural continuity in South Asia from the Harappan times until the present day.

c9.s3.11

³⁵ The excavation reports contain a substantial discussion of the iconography of Indus seals and tablets. See further, also for art history appreciation of the Indus glyptics, Heinz Mode, *Indische Frühkulturen und ihre Beziehungen zum Westen*, Basel 1944; id., *Das frühe Indien*, Stuttgart 1959; E.C.L. During Caspers, Some motifs as evidence for maritime contact between Sumer and the Indus Valley, *Persica* 5 (1971), 107-118 & pl. VIII-XI; id., Harappan trade in the Arabian Gulf in the third millenium B.C., *Mesopotamia* 7 (1972), 167-191; id., Sumer, Coastal Arabia and the Indus Valley in Proto-Literate and Early Dynastic Eras: Supporting evidence for a cultural linkage, *Journal of the Economic and Social History of the Orient* 22 (1979), 121-135; A. Parpola, New Correspondences between Harappan and Near Eastern Glyptic Art, in: B. Allchin (ed.), *South Asian Archaeology 1981*, Cambridge 1984, 176-195; A. Ardeleanu-Jansen, Die Kunst der Indus-Zivilisation, in: Michael Jansen, *Die Indus-Zivilisation*, Köln 1986, 211-235.

c9.s3.12

³⁶ Cf. John Marshall, *Mohenjo-daro and the Indus Civilization*, London 1931, I, 52ff.

c9.s3.13

³⁸ See the last but one paper cited in fn. 35.

4. The enigma of the Indus script

c9.s4.1

From the very beginning, the pictographic Indus script has been the most tantalizing one among the many problems presented by the Harappan culture. Slightly more than 3500 short inscriptions hold an answer to the most debated question concerning this early urban culture, that concerning its language. Many attempts at deciphering this unknown writing system have been made ever since the first specimen was published in 1875, and all sorts of 'solutions' have been proposed.

c9.s4.2

The Indus script has been considered as genetically connected with the Brahmi script of early historical India.³⁹ Other hypotheses have connected the Indus script with the scripts of the ancient

c9.s4.3

³⁹ Cf. A. Cunningham, *Inscriptions of Asoka*, Corpus Inscriptionum Indicarum I, Calcutta 1879, 61f.; Stephen Langdon, in: John Marshall (ed.), *Mohenjo-daro and the Indus Civilization*, London 1931, Vol. II, 426; G.R. Hunter, *Mohenjo-daro - Indus epigraphy*, Journal of the Royal Asiatic Society 1932, 466-503; id., *The script of Harappa and Mohenjo-daro and its connection with other scripts*, London 1934; among the numerous Indian scholars maintaining

c9.s4.4

Täfelchen der Harappa-Kultur, Prähistorische Bronzefunde I: 6, München 1985.

c9.s4.5

³⁷ Alf Hiltebeitel, The Indus Valley "Proto-Siva", reexamined through reflections on the Goddess, the buffalo, and the symbolism of *vāhanas*, *Anthropos* 73 (1978), 767-797. Mahiṣa Asura seems to be identical with Siva in the terrible or Bhairava aspect, having the form of a fierce buffalo. Cf. A. Parpola, *The Sky Garment*, Studia Orientalia 57, Helsinki 1985; id., From Istar to Durgā, in: G. Sontheimer and M.L.K. Murty (eds.), *Durgā and the Buffalo*, Heidelberg (in press).

c9.s4.6

Sumerians, Proto-Elamites, Egyptians, Hittites and Chinese and even with Etruscan pot-marks and with script-like carvings on wooden tablets found in the Easter Island, in the middle of the Pacific Ocean. The language underlying the Indus script has been supposed to be Sumerian, Proto-Dravidian, Proto-Indo-European, Proto-Indo-Iranian, Sanskrit, Prakrit, and so on.⁴⁰

c9.s4.7

⁴⁰ For surveys of various attempts at deciphering the Indus script, see Arlene R. K. Zide, A brief survey of work to date on the Indus valley script, Papers from the 4th Regional Meeting, Chicago Linguistic Society, April 19-20, 1968, Chicago 1968, 225-237, reprinted in Journal of Tamil Studies II:1, 1970, 1-12; Jaroslav Vacek, The problem of the Indus script, Archiv Orientální 38, 1970, 198-212; N.V. Gurov, Izuchenie protoindiyskikh tekstov (kratkij obzor), in: Soobschenie ob issledovanii protoindiyskikh tekstov - Proto-Indica: 1972, Moskva 1972, I, 5-51.

c9.s4.8

But no unanimity has been reached even on the basic issues, and most literature on the Indus script requires a lot of sifting in order to pick up useful ideas. The main reason for this unfortunate state of affairs is the fact that all keys that opened other unknown scripts are unavailable here. There are no bi- or multilingual inscriptions giving the same text in both Indus script and some readable characters. There are no understood historical texts which could tell the names of the Harappan gods, kings or cities, or which would quote samples of the language spoken by the Indus people. Even the type of the writing system represented by the Indus script is debated. Moreover, all the texts are short and limited in nature: the average length is five signs, and the longest texts, two identical three-sided tablets (M-494 and M-495), contain 26 signs each. The longest inscription on any single side of an object is found on a seal (M-314) with 17 signs divided into three lines.

c9.s4.9

But students of the Indus script must face these formidable difficulties and the pessimistic prognoses based on them. In fact, some more objective work has been done also. There has been serious discussion of the methodology, and essential research tools in the form of documentation and concordances have been created. On one point, at least, most scholars agree: the direction of writing usually is from right to left (but in the seal stamps, engraved in mirror image, from left to right); however, in some texts (particularly in the early tablets from Harappa) the direction of writing runs from left to right, and in a few texts alternatingly, boustrophedon.⁴¹

c9.s4.10

⁴¹ Cf. Stephen Langdon, in: John Marshall (ed.), Mohenjo-daro and the Indus Civilization, London 1931, Vol. II, 427f.; C.J. Gadd and Sidney Smith, ibid. 409-411; G. R. Hunter, The script of Harappa and Mohenjo-daro and its connection with other scripts, London 1934, 37-43; B.B. Lal, The direction of writing in the Harappan script, Antiquity 40, 1966, 52-55 & pl. XII-XIII; id., A further note on the direction of writing in the Harappan script, Puratattva 1, 1967-68, 15-16 & pl. I; I. Mahadevan, The Indus script: Texts, concordance and tables, Memoirs of the Archaeological Survey of India 77, New Delhi 1977, 10-14.

c9.s4.11

We cannot enter into a detailed discussion of the Indus script and its study here. For this, the reader is referred to literature published elsewhere.⁴² In the sequel we shall only try to justify our this view, mention may be made of S.R. Rao (The Decipherment of the Indus script, New Delhi 1982), who derives the Brahmi script as well as the Semitic alphabet from the Indus script. However, it is a well-established fact that the Brahmi script is derived from the Semitic consonantal alphabet, and this in turn from the uniconsonantal signs of the Egyptian hieroglyphic writing. Cf. e.g. Georg Bühler, Indische Palaeographie, Grundriss der Indo-Arischen Philologie und Altertumskunde I:11, Strassburg 1896, 10ff.; A.H. Dani, Indian Palaeography, Oxford 1963, 23ff.; I.J. Gelb, A study of writing, 2 ed., Chicago 1963, 147ff, 197f.

c9.s4.12

⁴² Excellent general introductions to the ancient writing systems and the methodology of their decipherment are: I.J. Gelb, A study of writing, rev. ed., Chicago 1963; id., Written records and decipherment, in: T.A. Sebeok (ed.), Current Trends in Linguistics 11, The Hague 1973, 253-284; Ernst Doblhofer, Voices in stone: The decipherment of

c9.s4.13

belief that the present work will constitute an indispensable tool for research in this field: the *Corpus of Indus Seals and Inscriptions* endeavours to collect all the primary material necessary for the study of the Indus script and to make it available in as good form as possible.

5. Earlier documentation of the Indus seals and inscriptions

c9.s5.1

The collection, edition and careful indexing of all existing material is a basic requirement in the critical and methodical study of any unknown script.

c9.s5.2

A praiseworthy early undertaking in this task was the book by G.R. Hunter published in 1934⁴³. It contained drawings of all the Indus texts excavated by February 1927 (comprising 518 texts from Mohenjo-daro and 243 texts from Harappa), with a documentation of the excavation numbers, as well as a concordance to the occurrences of each individual sign within these inscriptions. Hunter further discussed this evidence and drew certain conclusions from it. Even if one disagrees with his general findings, Hunter is to be credited for a good number of pertinent observations and for the preparation of a valuable research tool.

c9.s5.3

⁴³ G. R. Hunter, *The script of Harappa and Mohenjo-daro and its connection with other scripts*, London 1934.

c9.s5.4

The official reports of the excavations at Mohenjo-daro, Harappa, Chanhu-daro and Lothal⁴⁴ have included photographs and very substantial and comprehensive descriptions of most of the seals and inscribed objects discovered. It must be noted, though, that the photographs of many duplicate inscriptions from Harappa (and a few from Mohenjo-daro) were omitted from the excavation reports, being replaced in the data tabulations by the short statement "similar to....". Objects in a bad state of preservation were also excluded. The reports of Mohenjo-daro and Harappa further comprise sign lists which record occurrences of the individual pictograms. The sign lists are valuable, even include a few unpublished texts, but are not always accurate and are limited to a portion of the material *ancient scripts and writings*, transl. by Mervyn Savill, London 1961, 1973; E.W.J. Barber, *Archaeological decipherment*, Princeton 1974; and Maurice Pope, *The story of decipherment*, London 1975.

c9.s5.5

⁴⁴ See above, footnotes 4,5,7. These reports remain essential, and the reader is referred to them also because the detailed catalogue of the material published in this volume will be published later, in the third volume of the *Corpus*. It is, however, useful to keep in mind that a few mistakes have crept into the excavation reports, especially that of Harappa. Thus, pictures of two-sided tablets have occasionally been mixed up, so that the two sides of a given object actually belong to two different objects. Sometimes two sides of one object have been separated from each other and given separate numbers. The tabulations are not fully reliable, either: in addition to misprints, some objects have excavation numbers which are quite different from those written on the respective objects themselves.

c9.s5.6

Entitled *Sealed Secrets of the Indus*, a monograph on the problems relating to the script, language and religion of the Indus Civilization is being prepared by A. Parpola. Many issues have been dealt with by this author in the following overviews, the first three of which include extensive bibliographies: Tasks, methods and results in the study of the Indus script, *Journal of the Royal Asiatic Society* 1975: 2, 178-209; The problem of the Indus script, in: D.P. Agrawal and Dilip K. Chakrabarti (eds.), *Essays in Indian Protohistory*, Delhi 1979, 163-186; The Indus script: a challenging puzzle, *World Archaeology* 17:3, 1986, 399-419; Zur Entzifferung der Indus-Schrift, in: *Vergessene Städte am Indus: Frühe Kulturen in Pakistan* vom 8. bis 2. Jahtausend, Mainz 1987, 196-205; and Religion reflected in the Indus script: Penetrating into long-forgotten picto+graphic messages, *Visible Religion* 6, 1987. See also K. Koskenniemi, *Syntactic methods in the study of the Indus script, Studia Orientalia* 50 (1981), 125-136.

c9.s5.7

only.

c9.s5.8

A computer-drawn concordance to the Indus inscriptions was published in 1973 by a group of Finnish scholars.⁴⁵ In the preparation of this work, Dr Asko Parpola visited the principal museums in Pakistan and India in 1971 in order to compare the readings based on the published photographs with the original objects. To his surprise he found more than 400 seals and inscriptions from Mohenjo-daro and Harappa that had never been published.⁴⁶ Most of them came from the digs carried out by the custodians of the site museums after the official excavations and reported only very briefly in the Annual Reports of the ASI.⁴⁷ (Fortunately, the unpublished fieldbooks of these as well as of the official excavations have since been discovered in Pakistan, and are in the process of being published.⁴⁸)

c9.s5.9

⁴⁵ Seppo Koskenniemi, Asko Parpola and Simo Parpola, *Materials for the study of the Indus script, I: A concordance to the Indus inscriptions*, Acta Academiae Scientiarum Fennicae B 185, Helsinki 1973.

c9.s5.10

⁴⁶ Cf. *Newsletter of the Scandinavian Institute of Asian Studies* 5, 1972, 12ff.

c9.s5.11

⁴⁷ Cf. *Annual Report of the Archaeological Survey of India*, 1930-31, 1931-32, 1932-33 & 1933-34, Calcutta 1936, 70-72; 1934-35, Calcutta 1937, 31-33 & pl. X; 1935-36, Calcutta 1938, 35-36; 1936-37, Calcutta 1940, 39-41. This was noted by I. Mahadevan, who had simultaneously made the same discovery.

c9.s5.12

⁴⁸ Michael Jansen and Günther Urban (eds.), *Mohenjodaro: Report of the Aachen University Mission 1979-1985. Section One: Data Collection. Vol. I: Catalogue and Concordance of the Field Registers 1924-1938. Part One: The HR-Area Field Register 1925-1927*. Leiden 1985. Five further parts of Vol. I are to follow.

c9.s5.13

Mr Mahadevan brought out his edition and concordance of the texts in the Indus script in 1977, improving upon the Finnish concordance in several respects.⁴⁹ Besides, Mahadevan could include more text material on the basis of the Photo Archive of the ASI, which preserves old unpublished photographs of objects since lost. On the other hand, however, Mahadevan excluded all material that Asko Parpola had discovered in the museums of Pakistan and that had been included in the Finnish concordance. Mahadevan's book further included a listing of the texts on which the concordance is based, good documentation, and several cross charts with interesting statistics of different kinds.

c9.s5.14

⁴⁹ Iravatham Mahadevan, *The Indus script: Texts, concordance and tables*, Memoirs of the Archaeological Survey of India 77, New Delhi 1977. One improvement was the general arrangement of the concordance, which took the single sign as the basis, as in Hunter's concordance, while the Finnish concordance indexed the pairwise combinations of signs and left the isolated occurrences of signs unindexed.

c9.s5.15

Dr Parpola had gone through the Photo Archive of the ASI in 1975 and identified most of its material. In collaboration with Dr Kimmo Koskenniemi, he brought out a revised edition of the Finnish concordance in three volumes in 1979-1982, since there was still scope for improving upon the reading of the inscriptions and upon Mahadevan's work. The new version was published in a preliminary limited edition as research reports, because the work on the present Corpus of Indus Seals and Inscriptions was expected to bring still further improvements upon the textual readings as well as new inscriptions.⁵⁰ The updated version will appear in print after the publication of the three volumes of the Corpus, for such a standardized and indexed text edition remains a necessary complement to the photographic Corpus.

c9.s5.16

⁵⁰ Kimmo Koskenniemi and Asko Parpola, *Corpus of texts in the Indus script*, Department of Asian and African Studies, University of Helsinki, Research reports 1, Helsinki 1979; id., *Documentation and duplicates of the texts in the Indus script*, Ibid. 2, Helsinki 1980; id., *A concordance to the texts in the Indus script*, Ibid. 3, Helsinki 1982. The reference numbers of the inscriptions in the Finnish concordance will be changed to those of the Corpus, so as to ease comparison with the photographs. The typological and iconographical classifications will be revised as well.

6. The purpose and scope of the Corpus

c9.s6.1

The texts in standardized editions and corcordances are based upon the subjective judgements of individual scholars, and they do not display all the intricacies of the originals. Moreover, they contain numerous admittedly doubtful readings. Objective photographic documentation of the original inscriptions thus is a necessary complement to such textual studies. Photographs of the original objects are equally indispensable tools for the historians of art and religion studying the iconographic motifs and for archaeologists engaged in a comparative study of the objects. In short, there is no replacement for good photographs of all the Indus seals and inscriptions.

c9.s6.2

A major part of the material has been published in photographs in the excavation reports of Mohenjo-daro and Harappa: they illustrate altogether roughly 2500 objects. These publications have long been out of print and difficult to procure. It is true that they have been reprinted in recent years, but the quality of the photographs in the reprints is so low that they are practically unusable. The published photographs of the rest of the material, on the other hand, are scattered in a number of publications, and their mere collection involves great difficulties for persons without access to specialized libraries.

c9.s6.3

It would have been simple enough to collect and reproduce the old photographs of the earlier publications. Such a procedure, however, would have resulted in a book that would not have fully satisfied the serious student of the Indus script and iconography. The size and quality of the illustrations, even in the original reports, is not always sufficient. Moreover, the available material is documented incompletely, for, as pointed out above, there are many hundreds of unpublished objects: objects coming from excavations conducted at Mohenjo-daro and Harappa after the conclusion of the official excavations; a large number of duplicate and broken or indistinct objects, especially from Harappa; and objects from excavations and explorations carried out in India and Pakistan during the past few decades but not yet published in full.

c9.s6.4

Apart from their inscriptions and iconography, the seals form an important category of artifacts in their own right, which we have seen to have much relevance for the study of the external relations of a culture as well as of its internal processes. Therefore, in addition to all the inscriptions in the Indus script, this Corpus will contain all the Harappan seals, including those without any inscription. In the case of other object types, 'inscription' has occasionally been understood rather liberally so as to include, for example, K-119, a most interesting 'terracotta cake' from Kalibangan, though its incisions form an iconographic motif rather than an inscription.

c9.s6.5

Moreover, the concept of 'Indus seals' is to be understood in its widest meaning. In addition to the Mature Harappan period or the Indus Civilization proper, the Corpus will, with certain restrictions, cover the Early and Late Harappan periods as well and also include all the imported seals of foreign types coming from Harappan sites.⁵¹ Furthermore, 'Harappan' is understood to include closely related cultures such as that of Kulli in Baluchistan or Prabhas Patan in Saurashtra. Thus the "Northeast Iranian" type seal coming from the Kulli site of Mehi has been included, and it would have been folly to exclude the NE Iranian type seal from Shahi-tump found in Indian collections. Although Maski is not a Harappan site, the cylinder seal found there is of great interest: obviously made in India - witness its elephant motif - it demands comparison with the cylinder seal found not so far from Maski, in Daimabad, in a Late Harappan context.

c9.s6.6

⁵¹ See the discussion in chapter 2. In the past, some clearly imported seals like H-166 have often been treated as Harappan; in order to prevent this from happening in the future, the word "foreign" has, space permitting, been put in the page caption of the Corpus at such seals.

c9.s6.7

Some objects kept in the museum collections together with Indus seals or inscriptions have been purposely excluded as irrelevant. In the case of this volume, these include some Kuṣāṇa coins from Mohenjo-daro,⁵² some Harappan ear studs (?) with geometrical motifs carved on them,⁵³ and some quite indistinct objects from Lothal.⁵⁴

c9.s6.8

⁵² (Sd 2756, ASI 63.10.294) Two copper coins, one round, one square; one copper coin found by R.D.Banerjee (ASI 63.10.301); and one round copper coin (Y 71, ASI 63.10.418).

c9.s6.9

⁵³ DK 8991 (PWM 350); DK 12204 (PWM 351); HR 225 (IM 10508, A 7978) and HR 822 (PTN Arch. 10259) from Mohenjo-daro; 336 (IM 11109, A 21202) and 3603 (IM 11101, A 22435) from Harappa; and one of uncertain provenance (PWM 352). Cf. E. Mackay, *Further Excavations at Mohenjo-daro*, Delhi 1938, Vol. I, 532f.

c9.s6.10

⁵⁴ The objects having the exc. nos. 2839 and 3750 could be remnants or elements of seals, but this seems most uncertain, and in any case they contain no writing. The clay lumps having the exc. nos. 1837, 1856, 1890, 1984, and 5242 contain no trace of a seal impression.

c9.s6.11

The relatively few seals and clearly Harappan-type inscriptions from the Late Harappan period have been included in the Corpus, but Late Harappan graffiti have been excluded, with a few exceptions. These graffiti are short and appear to be just "pot-marks" rather than real writing. Still, they are potentially interesting to the student of the Indus script, even though not to the same extent as the Early Harappan pot-marks. The problem is their great number, coupled with the difficulties of drawing a line between Late Harappan and Post-Harappan and of finding the original potsherds. For these reasons we have decided not to reproduce the graffiti from Rangpur in this volume; these have been collected and published (only partially in photographs) by S.R. Rao.⁵⁵ Only the most elaborate Late Harappan "inscription" from Rangpur (Rgp-2) has been reproduced in this volume along with the one original sherd that could be traced (Rgp-1). The "Late Harappan inscriptions" from Machiala Mota,⁵⁶ the signs painted on Jorwe pottery from Daimabad,⁵⁷ and, among other things, the graffiti on red pottery from Ganeshwar⁵⁸ have been excluded for similar reasons.

c9.s6.12

⁵⁵ S.R. Rao, *Excavations at Rangpur and other explorations in Gujarat*, *Ancient India* 18-19 (1962-63), 5-207, especially p. 130 and pl. XXV B - XXVIII.

c9.s6.13

⁵⁶ Cf. S.R. Rao, in: D.P. Agrawal and A. Ghosh (eds.), *Radiocarbon and Indian archaeology*, Bombay 1973, 329.

c9.s6.14

⁵⁷ *See Indian Archaeology 1974-75 - A Review*, pl. XXVI.

c9.s6.15

⁵⁸ See R.C. Agrawala, Aravalli, the major source of copper for the Indus and Indus-related cultures, in: B.B. Lal and S.P. Gupta (eds.), *Frontiers of the Indus Civilization*, New Delhi 1984, 157-162, especially pl. 71.

7. The documentation of the objects

Original objects and their present-day impressions

c9.s7.t1.1

Because the texts carved in mirror image on the seals are to be read as they appear in the impression, the reports of the excavations at Mohenjo-daro and Harappa published just the impressions. However, the impression may not faithfully reproduce all the features of the original, and the original always remains the ultimate authority. On the other hand, the impression is needed not only because it shows the inscription in its proper form but also because it sometimes reveals details not immediately visible by the inspection of the original. For example, it is harder to see an inscription on a rough or transparent or multicoloured surface than in an impression taken on a neutral and unweathered material (cf. M-221 and L-36). Thus the original and its impression complement each other and furthermore make a double checking possible.

c9.s7.t1.2

As a rule, an impression of an object is always published in this Corpus when the object was originally meant to produce one, as is the case with the stamp seals. Exception is taken to this rule, however, if it was not possible to get an impression, as for example if a seal was too brittle. In addition, an impression is published whenever it clearly helps in understanding an object meant to be read directly (e.g., H-176).

c9.s7.t1.3

The ASI has taken the responsibility for making the impressions of these unique and often fragile Harappan objects. The use of silicone rubber was considered, but in their tests the chemists of the ASI came to the conclusion that the condition of the objects does not allow this material to be used. Unfortunately the plasticine used instead is not sensitive enough, so that all details have often not been reproduced. Moreover, small crevices often form when a forceful impression is made on plasticine, with a result that is not aesthetically pleasing even if it may otherwise be adequate (cf., e.g., M-32 a & M-208 a). And in the case of large seals especially, it is difficult to obtain a good impression in which all parts of the inscription and the device are perfect. However, in the vast majority of the seals, the new impression is much better than the old one.

c9.s7.t1.4

Since an impression was taken and photographed twice for most of the objects, there was often the possibility to choose a second if one was not good, but in numerous cases neither version was publishable. In 1987, an effort was made to obtain a good new impression of such seals. Where this could not be done, recourse was taken to old impressions made soon after the excavations, either those published in the excavation reports or, if better, those available in the Photo Archives of the ASI.

c9.s7.t1.5

It would have been possible to replace missing or bad impressions by reversed prints of the original seals, but this procedure was strictly refrained from; it could have lead to serious misunderstandings, for some seals have a reversed direction of writing.

Broken objects

c9.s7.t2.1

Old photographs have been published besides the new if they clearly complement each other⁵⁹ and whenever they show an object in a state of preservation that is better than its present state. A broken object may have been restored afterwards, and in some cases the impression taken nowadays of the object may be quite misleading.⁶⁰

c9.s7.t2.2

⁵⁹ Sometimes the photograph of a new impression was received only after the printing was started, with the result, e.g., that M-81 a bis now makes M-81 a irrelevant.

c9.s7.t2.3

⁶⁰ Cf., e.g., H-141 and especially H-129. From M-277 a, an old impression, it can be seen that one corner was originally missing from this seal. The restoration evidenced already in another early impression, M-277 a bis, seems dubious, for the inscription in the added corner does not seem to fit.

c9.s7.t2.4

The excavation reports sometimes left one wondering whether the object depicted is complete, and if not, how much is missing. This can be checked by examining its back or sides. In the case of the regular square seals, this is often superfluous, because the estimate can be made from the front side itself, but for the rectangular seals without iconography it is indispensable to see the flank side and the position of the cord hole that is usually pierced through its centre.⁶¹

First 102 of 3,951 passages. Read on

Linked texts (34)