The Analysis of Jina Images [Part 3]

The essay was published in Berliner Indologische Studien No. 2. 1986, pp. 133-174.

§ 5. “Changing combinations“ as a category

In the previous section we discussed “changing combinations“ and other aspects of the material (pool concept etc.) in connection with the parallelism of image-description and statistics. However, we cannot ignore the fact that the category under discussion is also interesting from the point of view of a general structural analysis of the material. This applies both to Jaina iconography and Jaina literature, more particularly to Jaina narrative literature. ^[1]

Our § 4 did not contain any comprehensive demonstration of “changing combinations“. We merely mentioned JID 157 (or JID 156-158) as an abstract paradigm.

The situation is somewhat different in the case of the closely related images JID 140 and 141. Here we observe a sharp contrast between the high artistic standard of the two pieces and the mechanical methods of differentiation or variation. JID 141 had standing male figures to the left and to the right of the lion-throne, whereas the corresponding places of 140 are left empty; JID 141 has, to the left and to the right of the head of the Jina, a standing figure in addition to three hovering celestial genii, while 140 only has the three genii; the garland-bearing couples of JID 141 follow the standard formula whereas 140 presents a rare formula where both partners carry the garland together. Further differences can be easily mentioned. We thus observe that we are concerned with “variation“, in the sense of it being a workshop stratagem, and not merely with a natural development within one and the same artistic tradition. Another such case is supplied by the five images (all provided with abhiṣeka-elephants) of JID § 122. Here we notice, and this is a rather extreme case, that all the images of the group have different attendant figures (§ 13). Strictly speaking, even JID 157 is a fairly good example of the specifics of variation. However, we could not demonstrate the general phenomenon of “changing combinations“ with common-run examples. Rather, we needed an unmistakable case. Naturally, such a case falls at the same time in the present category which is more specialized.

The term “variation“ can be used whenever similar images are in certain respects strikingly dissimilar. As a corollary we can mention such instances where dissimilar images are in certain respects strikingly similar. In order to minimize the number of technical terms we suggest for such cases the word “parallel“, just as we speak of a “parallel“ if two different texts have a verse or part of a verse in common. Our examples are not pretentious but are sufficient to demonstrate this phenomenon: three images from our corpus share the combination of the abhiṣeka-elephants with the duplicated garland-bearer motif (JID 8A, 44, 147A: §§ 17 and 18); the Throne-Frame Images JID 140 and 143 share a specific rendering of the seat-lotus (carved in full relief); three images of the Early-Medieval Style share the combination of drummer protome and rich double-leaf (JID 8A and JID 140/141).

§ 6. Slot-Filler Analysis

In continuation of § 2 of IJI we have to say a few words about slot-filler analysis in order to describe its place in the present study. This analysis is necessitated by the specific form of canonization which we observe in large sections of Indian art. ^[2] The relevant canonization is characterized by the operation of the slot-filler principle or, as we may also say, by the effect of the slot-filler pattern or system. As a method, analysis answers to the principle much in the same way as image-description and statistics answer, as separate approaches, to changing combinations. There is also some analogy to § 5. Just as we distinguished in the case of changing combinations between the general principle and its methodic consequences on the one hand and certain more specialized phenomena on the other, so we can also distinguish, in the present case, between the slot-filler principle in general and the more specific phenomenon of substitution. A further general observation may be added. While it is difficult to describe the relation between different categories such as changing combinations, canonization, and slot-filler principle, it is easy to describe the methodic tetrad outlined in § 4. The tetrad has thus priority as our basic methodic pattern although our various different categories are instrumental in deciding how exactly the tetradic programme is to be implemented.

Canonization in general and the slot-filler principle in particular become at once evident if we study their adequate examples as found in the medieval art of Northern India. We mention only the fact that the images normally show a central figure of comparatively large dimensions which is surrounded by a host of smaller figures and other subsidiary motifs. The whole scheme is dominated by a more or less pronounced grid of vertical and horizontal lines. This impression is enhanced if we study images where the various motifs are not interwoven but rendered in a more paratactic manner. This paratactic tendency depends to some extent on the region - good examples may be found in Orissan art - and to some extent on the period: motifs on early-medieval images are more clearly separated than motifs on medieval images.

It is difficult to translate the bare impression of slot-filler patterning into a rigid method. A critic may even say that, in the case of very similar images, the analysis seems superfluous and, in the case of different images, too complicated. There are nevertheless instructive examples for this patterning in several art provinces (when we compare e.g. the Hindu goddess Chandra, A1.235 with the Sūrya ibid. 397). Such examples indicate that there is scope for methodic efforts. A definition of the slot-filler analysis must, however, be an ideal-typical construct rather than an analytic description of existing images. We first subdivide the pattern of the image in different slots (e.g. attendant figure slot) and then we “list“ the fillers which occur in each individual slot (e.g. male cāmara-bearer, female cāmara-bearer etc.). The system has three factors of limitations:

1. limited number of slots;
2. limited number of fillers for each individual slot (even the number of all the fillers is limited); and
3. restriction of a filler to one specific slot (one slot admits of several fillers, but not vice versa: there is as a rule no merry-go-round effect).

The addenda that follow are as important as the basic description. In the case of the Deogarh Jinas we have for example very little filler-variation: one and the same slot normally has one and the same filler. We therefore have to distinguish between the more numerous cases where the fillers do not differ very much - here we simply describe the fillers and their variants on the basis of the slot-system - and cases where the fillers do differ considerably. In the second instance, we are faced with the special phenomenon of substitution. Of course, substitution indicates more than a general similarity of the pattern. It implies that we once get filler x and once, in exactly the same slot, filler y, so that we may also say that in the case of one image filler x appears, instead of filler y as found in the case of another image.

It is nevertheless difficult to distinguish clearly between similarities in the general pattern and special cases of substitution. We have, e.g., to remember that one and the same filler may have different iconographic features. We already noticed that two of the cāmara-bearers in JID 157 wear crowns while the other four are bare-headed. We have no established terms for such cases (sub-motif? sub-slot?) but they also belong to the field of filler-variation. If “more than one“ feature of a subsidiary figure is subject to such changes we may already speak of substitution. It is thus necessary and possible to make pointed observations, but ultimately all the relevant cases form one continuum.

The cāmara-bearers are amongst the more important human figures to appear in the composition, and this leads us to another problem. We have described the slot-filler system as a grid formed by horizontal and vertical divisions. Naturally, this description does not apply to those parts of the system which are based on human figures. Attributes in the hands of the figures (main figures and subsidiary figures) are fillers in their own right (the slots being provided by the respective hands). We thus have a tension between a rectilinear pattern and one or more sub-patterns based on the form of the human body (elliptical, circular).

Deviations from a specific local slot-filler system as used for a specific god or goddess are normal. Here we mention only the fact that the pedestals of the medieval Jinas at Deogarh (both seated and standing images) show more variation in their patterning than one would expect under the circumstances (IJI § 6, last para). But the question of deviation already takes us into the next section.

If a slot is empty, then we can say that the filler has the value of “zero“ (e.g. compare JID 140/141: absence and presence of the standing figures on the pedestal). But the value of “zero“ is dependent on the context. If one image is less developed and less rich than another comparable image we have to be cautious when using the expression “zero“ in any positive sense, i.e. in the sense of an opposition.

As a technical aid we have prepared the graph of fig. 1. This is not a demonstration of slot-filler analysis but a guide in the sense in which numismaticians use the figures on the dial of a clock as a means of orientation. Since the compositions are largely symmetrical, it is in most cases sufficient to quote the left side of the scheme. For example we assign squares 39-50 to the pedestal of the image without mentioning 39'-50'