Pramāṇa · Search · About · API

Indian Epigraphy: A Guide to the Study of Inscriptions in Sanskrit, Prakrit, and the Other Indo-Aryan Languages c12.s1: 7.1 Epigraphy and History

Indian Epigraphy: A Guide to the Study of Inscriptions in Sanskrit, Prakrit, and the Other Indo-Aryan Languages c12.s1

Language
English

Contents

7.1 Epigraphy and History

7.1.1 Political and dynastic history

c12.s1.t1.1

SIE 13–30; David P. Henige, “Some Phantom Dynasties of Early and Medieval India: Epigraphic Evidence and the Abhorrence of a Vacuum,” BSOAS 38, 1975, 525–49; J. F. Fleet, “The Present Position of Indian Historical Research,” IA medieval period, the picture is much brighter; in the case of such dynasties as the Rāṣṭrakūṭas, for instance, where epigraphic remains are relatively abundant, we can reconstruct a reasonably detailed account of their history, chronology, and geography. The difference between the earlier and later periods lies not only in the amounts of epigraphic material available but also in their contents. Thus in the early period we find such extraordinarily vague formulations as a dating merely “during the reign of the Śuṅgas” (*suganaṁ raje* = *śuṅgānāṁ rājye*; Bhārhut pillar ins., l. 1; SI I.87–8), whereas in medieval inscriptions the dating formulae and historical situation are often detailed and explicit. As an ideal example, in the inscriptions of the Eastern Cālukyas the genealogical introduction not only includes the usual eulogistic description of the forefathers of the current king but also states the number of years of their rule (SIE 19–20), enabling historians to reconstruct that dynasty’s chronology with particular precision.¹

c12.s1.t1.2

In most cases, of course, the situation lies somewhere between these two extremes. More typically, for dynasties for which we have a reasonable number of records dated in a continuous era, we can reconstruct a dynastic chronology with reasonable accuracy. The main problem is usually that for determining the earliest and latest dates of any particular ruler we are at the mercy of chance finds of dated inscriptions, and the precise date of transition of rule can be known only in such cases as we are fortunate enough to have records of two kings dated in the same or at least in successive years. Otherwise, we can only estimate these dates based on the range of attested dates for each king. The result of this situation is that for many of the better-represented dynasties of the medieval period, for example, the Rāṣṭrakūṭas or the Gurjara-Pratīhāras, we can usually date each king at least within a range of a few years, and frequently to the year.²

c12.s1.t1.3

But we are less well informed with regard to earlier, less abundantly attested dynasties or to those which dated their records in regnal years only. The chronology of the Pālas, for example, is considerably less certain than that of the contemporary Rāṣṭrakūṭas, not because of any scarcity of materials but because their records are dated only in regnal years. Regnally dated records require some type of historical synchronization which can provide absolute chronological “anchor points.” These are typically provided (if at all) by reference to historical personages or events known from other sources, the classic case being the absolute dating of Aśoka’s regnally dated inscriptions provided by chronological correlation with the known dates of the five Hellenistic kings (Turamāyo = Ptolemy II Philadelphus of Egypt, etc.) referred to in his rock edict XIII, which enabled historians to definitively fix Aśoka’s reign at around the middle of the third century b.c.

c12.s1.t1.4

With regard to the genealogy of the various dynasties too, the state of our understanding varies considerably. In general, the succession of rulers can be determined from genealogical accounts in the inscriptions themselves, especially from the early medieval period onward, when it became customary to provide a quasi-historical preface to inscriptions (cf. SIE 14–15, and sections 4.¹.1 and 4.1.².1c). But the record is often less clear, especially in earlier inscriptions which lack full genealogical accounts. In such cases, the recurrence of the same name in different generations can be troublesome, as in the case, for example, of the Kuṣāṇas or the Guptas, where the repetitions of the names Kaniṣka and Kumāragupta,³ respectively, have caused no end of frustration to modern historians.⁴ Equally troublesome is the practice, especially in later inscriptions, of referring to the same king by two or even more different names in different records, which sometimes produces serious or even insoluble genealogical problems; see, for example, the problems of Gurjara-Pratīhāra genealogy discussed in R. C. Majumdar, ed., *The Age of Imperial Kanauj*, 37.

c12.s1.t1.5

¹ For other examples of precise dating, see Fleet, “The Present Position,” 9–10.

c12.s1.t1.6

² For examples of the types of chronological difficulties which can arise with era-dated inscriptions, see R. K. Dikshit’s *Candellas of Jejākabhukti*, 17–9.

c12.s1.t1.7

attested in inscriptions are generally deduced on the grounds of (1) the findspots of the inscriptions themselves and (2) claims of territories and conquests made in the inscriptions. The first method has two main limitations. First, the geographical record preserved by inscriptions, like their chronological record, is necessarily incomplete, providing merely random spots in a larger field; and also as in the case of chronological reconstruction, it is only in such cases (which are the minority) where we have a reasonably large number of records for a particular ruler that we can determine anything like a clear picture of the extent of his realm. Second, it cannot always be assumed that the findspots of inscriptions are their original location, particularly in the case of copper plates, which were sometimes removed from their original location by grantees who migrated to a different region.⁶ Thus a single instance of a portable inscription found outside the known limits of a ruler’s realm should not be automatically accepted as evidence of his vast conquests, but a consistent pattern of this may be historically.⁷

c12.s1.t1.8

⁶ For examples, see Fleet, “Epigraphy,” IGI II.28.

c12.s1.t1.9

⁷ But such cases must be treated with caution. Thus the apparent conquest of Bihar and Bengal by the Gurjara-Pratīhāra king Mahendrapāla has recently been shown to be an illusion (G. Bhattacharya, *South Asian Studies* 4, 1988, 71–3).

c12.s1.t1.10

The territorial and military claims put forth in the inscriptions themselves are even more problematic. That many rulers (or rather their court poets), especially in medieval times, shamelessly exaggerated the territorial extent of their empires or spheres of influence is well established. The question is how to distinguish fact, or at least a core of fact, from the conventional bombast and rhetorical exaggeration of the *praśasti* style. A classic contrast is provided by the claims of *digvijaya* or conquest of the whole world (i.e., of India) made on behalf of Samudragupta in the Allahabad pillar inscription and of the Aulikara ruler Yaśodharman in the Mandasor pillar inscription. In the case of the former, the extensive and detailed list of kings and territories in the south, west, and north of India whose conquest or submission Samudragupta claimed lends a semblance of credence, so that historians generally accept that he did in fact rule over or at least raid the places in question, despite the absence of any corroborative evidence. In Yaśodharman’s *praśasti*, on the other hand, the claims of sovereignty from the Lauhitya (Brahmaputra) River in the east to the ocean in the west and from the Mahendra Mountain in the south to the Himālayas in the north are obviously formulaic, and no historian would be willing to credit Yaśodharman with anything like a pan-Indian empire.⁸ Thus the rule of thumb followed by most historians with regard to such claims is that the more specific and nonformulaic they are, the more likely they are to be historically justified; in Sircar’s words, “Vague claims are generally less reliable than definite statements involving the mention of the personal names of adversaries” (SIE 27).

c12.s1.t1.11

⁸ For other examples, see SIE 25–7.

c12.s1.t1.12

Especially in later inscriptions, such grandiose and obviously spurious claims to vast territories become almost routine, even to the extent that poetic considerations overrule historical plausibility; thus in the Khajurāho inscription (EI 1, 126, v. 23) the Candella king Yaśovarman is described as *kośalaḥ kośalānāṃ naśyatkaśmīravīraḥ śithilitamithilaḥ kālavan mālavānāṃ*, “who seized the treasuries of Kośala, who destroyed the heroes of Kashmir, who weakened Mithila, and was like Death to Malwa.” Here, in Sircar’s words, “It seems that the lure of alliteration . . . carried the poet far away from historical accuracy” (SIE 27).⁹ Moreover, even those boasts of victory which are based on historical fact are almost invariably couched in highly rhetorical style in which mundane details of time, place, and military strategy are rarely specified.

c12.s1.t1.13

⁹ For other examples, see D. K. Ganguly, *History and Historians in Ancient India* (New Delhi: Abhinav Publications, 1984), 51–2.

c12.s1.t1.14

Given this situation, historians will naturally wish to evaluate the claims put forth in the inscriptions in the light of corroborative evidence, whether in the form of material from other inscriptions or from literary or other alternative sources. For instance, Pulakeśin II’s claims in the Aihoḷe inscription (SI II.443–50, v. 23) of a victory over Harṣa finds explicit corroboration in the testimony of the Chinese pilgrim Hsüan Tsang, and hence can be considered reliable. Corroboration from other inscriptions is often harder to establish, due not only to the general problems of the historical evaluation already alluded to but also to the tendency of the poets to gloss over, distort, or simply ignore their patrons’ military defeats (SIE 27–8). In some cases, however, such corroboration can be established, for example, in the case of Gurjara-Pratīhāra Nāgabhaṭa’s victory over his Pāla enemy referred to in the Gwalior *praśasti* (SII II.242–6) which is confirmed by similar statements in the separate records of three of Nāgabhaṭa’s subordinate allies: “[T]he combined testimony of the four different records, coming from four different sources, . . . leave no doubt that Nāgabhaṭa scored a great victory over his Pāla rival Dharmapāla.”¹⁰ “The Pāla records,” on the other hand, not untypically “make no reference to this struggle.”¹¹

c12.s1.t1.15

¹⁰ Majumdar, *The Age of Imperial Kanauj*, 26.

c12.s1.t1.16

¹¹ Ibid., 48.

c12.s1.t1.17

Occasionally we do find cases where a conflict is reported from both sides, and here we are faced with special problems of analysis. For instance, the same Gurjara-Pratīhāra King Nāgabhaṭa’s war with Indra, the Rāṣṭrakūṭa king of Gujarat, seems to be alluded to in inscriptions in which victory is claimed by both sides; in such a case, if no further determination is possible, “It may be concluded . . . that no party gained a decisive victory.”¹² A similar problem has been raised by the recent discovery of the earliest inscription of the imperial Gurjara-Pratīhāras (New Delhi ins.; EI 41, 49–57), in which King Vatsarāja’s forces are said to have conquered all of Karnataka (*avajitāśeṣa-Karṇṇāṭa-*, ll. 5–6), while his contemporary in that region, the Rāṣṭrakūṭa king Dhruva, explicitly claims in his inscriptions to have defeated Vatsarāja. Here the editors comment, “These claims and counterclaims of victory are, more or less, a conventional part of epigraphic poetry and may indicate either the uncertain nature of the outcome of the battles . . . or may pertain to different battles in which the results were successively reversed and for which we do not have tangible evidence” (52). Such conflicting claims, in other words, can be settled or reconciled, if at all, only when other evidence, for example, the locations of subsequent inscriptions of the rulers concerned, provide clear corroboration of the claims of one or the other of them.

c12.s1.t1.18

¹² Ibid., 25.

c12.s1.t1.19

Such, in short, are some of the typical problems which confront the historian in attempting to evaluate the highly rhetorical, subjective, and often intentionally vague statements of eulogistic inscriptions. The results, not surprisingly, are more often than not less than entirely satisfactory. For the earlier periods of Indian history we often have little more than the vaguest outlines of events and chronologies. In the classical and medieval periods we are somewhat better off, but even here we have, with rare exceptions, only the bare outlines of chronology and events and relatively little in the way of details to flesh out the skeleton of history. There will always be many uncertainties, and any historical study based on epigraphic sources will inevitably be filled with qualifications like "probably," "possibly," and "it would appear that." Indeed, any such study that is *not* provided with such stipulations should be viewed with suspicion.

7.1.2 Administrative, economic, and social history

c12.s1.t2.1

In recent decades, several scholars have begun to undertake more detailed and sophisticated analyses of various aspects of political science, economics, and social history on the basis of inscriptional evidence. Such studies are primarily based on the copious documentation provided by later southern Indian inscriptions, especially those of the Coḷa and Vijayanagara kingdoms, which are sufficiently numerous and detailed to permit significant statistical analysis, as is generally not the case with the sparser data available from the north. The inscriptions in question include very large numbers (in the thousands) of stone temple inscriptions in Sanskrit and/or Tamil (or other Dravidian languages) recording various transactions, typically endowments or donations to the temples, which provide a wealth of information on virtually all aspects of social and economic life. These inscriptions have given rise to a sizable body of scholarly literature on medieval southern India which combines the study of inscriptions with modern methods of historical and social scientific research. In view of the ever-increasing volume and diversity of the literature in question, it can be introduced here only in outline; the studies cited are meant only to provide representative or outstanding examples of the genre, especially of more recent studies and of those which concentrate particularly on information derived from epigraphic sources.

c12.s1.t2.2

Among the several topics which have recently been addressed via the study of inscriptions, one of the most important and controversial is the nature of the state and of kingship among the traditional Indian kingdoms, with the Coḷa empire serving as the focus for most studies. The central figure in these studies is Burton Stein, whose works, most notably *Peasant State and Society in Medieval South India*, have stimulated extensive discussion and further research. Although Stein's characterization of the Coḷa empire (and by extension of other traditional Indian kingdoms) as a "pyramidal segmentary state" with a king whose principal function is more ritual than executive has not been universally accepted,¹³ it has come to constitute the central focus of subsequent discussion of the subject.

c12.s1.t2.3

¹³ See, e.g., the reservations expressed by Noboru Karashima in *South Indian History and Society: Studies from Inscriptions A.D. 850–1800*, xxv–xxviii.

c12.s1.t2.4

Numerous studies of administrative and bureaucratic systems in southern India and elsewhere have also been undertaken on the basis of epigraphic data. Noteworthy recent efforts along these lines include the studies of Noboru Karashima and his collaborators, which are conveniently compiled in his *South Indian History and Society: Studies from Inscriptions A.D. 850–1800* (esp. 55–68). Studies of these subjects are typically based on official terms and titles culled from the inscriptions. Such titles, however, are often unknown from other sources and are not always etymologically transparent, so that their significance must be inferred, with varying degrees of success, from the contexts in which they appear in the inscriptions.

c12.s1.t2.5

Systems of land tenure and modes of agricultural production have also been the subject of numerous studies based on or making extensive use of epigraphic data. Much of the argument has focused on the persistent controversy as to whether or not the term "feudalism" can be properly applied to the medieval states; see, for example, D. C. Sircar, ed., *Land System and Feudalism in Ancient India*. Some recent studies, such as those of Stein and Karashima (see esp. *South Indian History and Society*, 3–35), have succeeded in raising the discussion of land tenure to a higher level by avoiding ideological preconceptions and concentrating on analysis of the actual data provided by the inscriptions.

c12.s1.t2.6

Revenue systems are another important aspect of historical and economic studies for which inscriptions constitute a principal source of information, especially, once again, for South India. Among recent works, those of Karashima (compiled in *South Indian History and Society*, 69–158) are again noteworthy for their scrupulous attention to epigraphic details; although here too (as in the case of the official titles) many of the technical terms for types of taxes remain to be satisfactorily elucidated.

c12.s1.t2.7

Patterns of trade, both external and internal, can also sometimes be traced with some precision from the epigraphic records. Here, too, the Coḷa empire has been the focus of attention, as, for example, in Kenneth R. Hall's *Trade and Statecraft in the Age of Cōḷas*.

c12.s1.t2.8

Studies of traditional Indian society also look primarily to inscriptions for clues about the actual social situation at various times and places, as a check to the highly normative and theoretical formulations of social ideals provided in the Dharmaśāstras and other literary works. For the earlier periods, and in general for northern India, the information is, as usual, sparse, consisting for the most part of odd scraps of data from which it is difficult to draw anything like a comprehensive overall scheme.¹⁴ Once again, it is in the medieval south, and particularly in the Coḷa period, that epigraphic data attain a critical mass that allows for detailed studies of social structures and caste relationships and their historical developments. Here too the studies of such scholars as Stein (e.g., *Peasant State and Society*, 173–215) and Karashima (*South Indian History and Society*, 36–68) are representative of the application of sophisticated modern methods to the study of Indian society on the basis of epigraphic data. Also worthy of note is K. Sundaram's *Studies in Economic and Social Conditions of Medieval Andhra* (A.D. *1000–1600*).

c12.s1.t2.9

¹⁴ Though this is not to deny that a good deal of interesting information is available; see, e.g., D. C. Sircar's *Studies in the Society and Administration of Ancient and Medieval India*. See also the general comments by Aloka Parasher in "Epigraphy as a Tool for Writing Social History," JESI 18, 1992, 62–73.

Linked texts (1)