Prof. Dr. Klaus Bruhn
The essay was published in Berliner Indologische Studien No. 2. 1986, pp. 133-174.
§ 7. Slot-filler analysis and typology
The employment of slot-filler analysis, either in the general sense of “skeleton for the description“ or in more specific senses, must be adapted to the circumstances prevailing in a given case, and minimal observations on this point have already been made in IJI § 2, first para. There are two main differences: the slot-filler system is more rigid in some cases than in others; and the classification of the images (typology) is easier in some cases than in others. With respect to Deogarh, the following three facts have to be considered in this connection. Pārśva images are not very uniform and they cannot easily be described in terms of the standard system. This is not a conflict between two different systems, but rather we should say that the Pārśva images often appear as “intruders“ in the standard system. Again there is a difference between early-medieval and medieval images: the system is more tight in the case of the latter than in the case of the former (IJI § 2, first and last para). Finally, a certain confusion can be observed in the case of the later images of the early-medieval period. They all look “very similar“. But at the same time we observe a degree of disorder, due to eclecticism, which is in contrast with the more distinct iconographic conventions of the earlier images.
Under the circumstances we suggest the following procedure. We shall treat all the medieval images as one block or as one corpus (Images Nos. 227-319 minus non-Jinas; see JID § 330). The early-medieval images (Nos. 1 - 226 minus non-Jinas) will be split into three smaller blocks: standing non-Pārśvas (the present survey), standing Pārśvas, and seated non-Pārśvas. There are only a few late images showing Pārśva seated, and reference to them will be made where it is appropriate. A separate treatment of the Pārśva images (standing Pārśvas) is necessary for the reasons mentioned already, and the seated images must be described separately because they have lion-thrones and other motifs which are not found, or are at least rare, in the case of the standing images of the early-medieval period.
We shall consider the main Jinas as well as the subsidiary Jinas found on Jina images. Jinas in such other contexts as caumukhas etc. will normally not be taken into consideration. Our dates for the early-medieval period are A.D. 850-950, and for the medieval period 950-1150. Images before 850 and after 1150 are rare at Deogarh. The images JID 20a and 21 (see § 10) will not be considered.
The early images JID 24-26 could have been included into our survey if it were not that JID 26 is a fragment, and JID 24-25 differ from all the other images in their rendering of the pedestal. For these reasons it was advisable to exclude JID 24-26 from the corpus. Reference to early-medieval images at other places than Deogarh has been made occasionally. In the case of the medieval period, some pieces from neighbouring places have already been included into the numbering of JID (§ 331). It will be seen that the statements made in our survey, e.g. “cāmara-bearers are always male“, apply to our sub-corpus and there will not be any distinction between facts peculiar to the sub-corpus and facts of a more general distribution.
So far we have only used the expression “corpus“ instead of referring to a possible system of “types“ and “varieties“ (IJI § 2). We now call the Jina, i.e. the Jina as represented in sculpture, a “type“, just as Sūrya would be a “type“. This is not to say that the entire Indian pantheon as represented in plastic art resolves itself into a limited number of well-defined types. The simple case of Varāha, one zoomorphous and one hybrid type, already demonstrates difficulties inherent in typology. However, the greater part of the material can probably be subdivided into types or related classificatory units.
Naturally, the problem of classification recurs on a lower level when we try to organize the material within the corpus presented by one particular type. Here we use the terms “variety“ and “subvariety“. We would thus say that Deogarh demonstrates two varieties of the Jina, “early-medieval“ and “medieval“. The division of the former variety into the three subvarieties of standing non-Pārśvas, standing Pārśvas, and seated non-Pārśvas presents no problems. It should also not be too difficult to isolate further Jina varieties (Akota, Deccan, Orissa etc.). Difficulties will arise in connection with the “peripheral images“, e.g. Jina images from Madhya Desha which are in terms of time, locality, iconography etc. somehow distinguishable from the two Deogarh corpuses, but not so far removed that they could be used as a basis for postulating a separate type. A study of peripheral images of the early-medieval period could start from Gwalior and Amrol in the North and from Gyaraspur and Badoh in the South (see the map JID 390). Apart from that we must admit that there are isolated images within the corpus (JID 137) as well as transitional pieces (JID § 174). However, a limited typological deficit is no serious obstacle to the description of the material.
In order to smoothen the way to slot-filler analysis, it may prove useful to emphasize what we would call “variety taxonomy“. There is a tendency in literature on Indian art to make a sharp distinction between images, of a higher or lower quality, which follow more or less the beaten track, and images which include “new“ features and which are often labelled as “unique“. In this connection, we should not forget that the expression “new“ can also be used for features which were not noticed previously as well as for minor features in general. Images showing such features are equally interesting and a study of such pieces also deepens our knowledge of Indian art. In order to produce an exhaustive survey of minor or unnoticed features we do need the instrument of slot-filler analysis. Here, our first example is concerned with the “parasol-unit“ (§ 16 infra).
At Deogarh, the drum is invariably shown above the triple roof (JID 137), but in one case (JID 37) it is the other way round. The deviation may be attributable to the whim of an individual artist who saw here a chance for creating variety in a series of closely related images (compare JID 37 with JID 36). This is probably the correct explanation for that irregularity although the minute difference in the rendering of the parasol-unit could hardly have been noticed by the donor. It would, however, also be possible to explain the solecism of JID 37 by the influence of the canon prevalent in Orissa (e.g. compare Ghosh 325A: Musee Guimet Ṛṣabha, top left). Be that as it may, an irregularity like JID 37 is not likely to be recorded in a conventional description.
Our second example is JID 137. Here the rendering of the male cāmara-bearers is quite unusual. In addition, we notice two female figures on the pedestal which are without parallel in the Deogarh material. Taking both irregularities together we need not hesitate to call JID 137 “unique“. But, normally, even unusual features of this magnitude cannot be detected without a rigorously descriptive technique. And it is precisely such a technique which gives shape to variety taxonomy.
The varieties are defined by detailed analysis, but in the case of the type (Jina in general) we are also faced with the problem of definition. A systematic description of the Jina as represented in sculpture is not impossible, but this can be done in more than one way. A minimal definition just serves the purpose of identification, and here the label “naked human figure without attributes in the hands“ seems to be sufficient. Another type of definition is the isolation of the common denominator. Here we have to refer to the Jina's strict frontality, and to the rigorous limitation of the number of his possible postures. The third type of definition could be called hand-book definition. There we need a description of the Jina type which considers all the basic differences: with parikara / without parikara; completely naked / with dhotī (crown etc.); Pārśva / non-Pārśva; Ṛṣabha / non-Ṛṣabha. The handbook definition will in its turn have some influence on the two previous definitions. The definition “naked figure without attributes in the hands“ does not only ignore the Jina images showing a dhotī etc. For we may also ask whether the expression “human figure“ is an adequate description of a nāga-like motif like Pārśva.
From the point of view of “studies in patterns“ and “studies in canonization“ it is more interesting to compare different types (e.g. Varāha and Sūrya) than to study the internal organization of one single type. We refer in this connection to IJI § 2 (letter “b“) and to § 6 supra (third para). This inter-type comparison has again two aspects. It is possible to describe first all individual types, e.g. the main types of a particular region, and to state afterwards to what extent they show similarities in the patterning and in the selection of motifs. It is, however, also possible to select a few related types (e.g. Sūrya, Varāha etc.) and to compare them with special reference to relevant features (pedestal, parikara-top etc.). Such studies would not have the character of a survey. Rather, they would demonstrate the methodological scope of the slot-filler analysis.
The similarity of the images is reinforced when two types are not only similar as form motifs (e.g. standing human figure) but also related in terms of tradition (“Vaiṣṇava“, “Jaina“ etc.), see IJI § 2, letter “c“. In IJI § 2 we mentioned the slab of JID 210 (Jina: ācārya), in IJI § 7, eighth para, we gave a reference to the image JID 248 (Jaina goddess, to be compared with contemporary Jina images like JID 258). For the analysis we now suggest a triple scheme: comparison of the respective main figures (having a different identity but a measure of formal relationship), of the attributes (e.g. different motifs above the heads of the respective main figures), and of the non-attributes (many motifs are common to different types or transferred from type X to type Y, e.g. miniature Jinas in JID 248 and in JID 258). However, such an analysis is not part of our Jina project.
§ 8. Jaina art and Indian art
The observations in the present section are relevant to hermeneutics, but they are of a less general nature than the discussion in § 2 above. In the case of Jaina art, we are not only faced with the general problem of the relation between literature and art but also with the fact that certain developments in Jaina art are only recorded in comparatively late texts and that others are not mentioned at all. The relevant issues have been discussed repeatedly (U.P. Shah), but even so non-specialists are inclined to over-estimate the contribution made by Jaina literature to our understanding of Jaina art.
Another possible misunderstanding concerns the Jaina tradition in general. It is often expected that developments in the non-Jaina tradition, mainly on the side of non-Jaina art, can be better understood in the light of parallels taken from Jainism, mainly from Jaina art. But there we have to stress three facts:
The expression “shastric“ being too narrow, we prefer here the term “book-iconography“ (plastic art and miniature-painting as reflections of literary traditions). Book-iconography was accompanied by an increasing interest in monastic motifs which reached its climax in the miniature-paintings of the 15th century. The phenomenon cannot be described in a few simple words but it is obvious that the increased interest in literary sources, as they were available in those days, did not produce works of art which could be helpful for a better understanding of developments in the field of non-Jaina art. Thus all the three facts point in the same direction: The contribution of Jaina art, and of the Jaina tradition in general, to the understanding of non-Jaina art is limited. Jaina art is a part of, rather than a clue to Indian art.