The Analysis of Jina Images [Part 1]

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

The Analysis of Jina Images

§ 1. Two Kinds of iconographies

Publications on Indian art often employ the terms iconoplastic (in “iconoplastic art“) and narrative (in “narrative panel“ etc.). This distinction suggests itself naturally since the differences between these two idioms are perhaps more pronounced in Indian art than in other art forms.

We propose in this connection a dichotomic distinction between “narrative“ and “non-narrative“. The latter term includes developments which, though they are not narrative, can also not be called iconoplastic. Strictly speaking, it is only a sort of negation but we understand it mainly in the positive sense of motifs belonging either to iconoplastic art or to its ambience, e.g. human figures with little characterization on the outer walls of temples. By contrast, vyāla figures are more decorative in character, so that the term “non-narrative“ would be meaningless.

Our dichotomy implies the distinction between two or rather three different fields of research:

1. narrative art,
2. non-narrative art,
3. their interface (border-line cases, overlapping, interferences etc.).

The problems of the first category have already been studied by various scholars; ^[1] those of the third category are not our main problem but we will make reference to them in § 2. Our main concern is the study of the second category. Inquiries in this area are greatly facilitated by two lines of research which have been mainly pursued outside the field of Indology: studies in traditional stories ^[2] and studies in the canon issued It is of course primarily the first line which is relevant to the actual elaboration of a method for enquiries in non-narrative art.

In order to establish a sound basis for the study of non-narrative Indian art, we have to introduce a number of distinctions:

1. figures with clear identity (identification on the basis of the ancient texts) vs. figures without clear identity;
2. main figures vs. subsidiary figures (including similar figures in compositions without main figures);
3. part vs. whole (main figure vs. subsidiary figures and other subsidiary motifs);
4. true systems vs. ad hoc systems: this refers in the first place to the difference between series of gods and goddesses as described in the ancient texts and series as de facto found on the outer walls of temples;
5. corpus vs. individual images: a corpus consists of a fairly large number of related images (whatever the exact character of this relation); it therefore requires special methods of description.

As a consequence of these distinctions, traditional subjects now appear in a new light. Thus a zoomorphous Varāha is not only interesting by traditional criteria (style, inscriptions, conspicuous peculiarities) but also on account of the numerous miniature figures covering the entire body of the animal. These figures must be subjected to a systematic study. In addition to that, we can establish new subjects of enquiry, like “zoomorphous Varāhas / slabs with miniature figures / pillars with miniature figures“. It is, however, obvious that the catch phrase “new light on traditional subjects“ is at least as important as the catch phrase “new subjects stricto sensu“. The five distinctions mentioned above (but demonstrated only in connection with “b“) can also be described as elements of an “extension of the methodic canon“.^[3]

In IJI § 3 we have proposed a distinction between five main traditions in Jina (not Jaina) iconography. A more detailed survey is possible on the basis of a more or less systematic typology (§ 7 infra). Apart from that we can also try to make the field of Jaina iconography (sculpture as well as painting) more manageable by isolating a number of fields, some of greater and some of lesser extent: āyāgapaṭas, narrative panels at Abu and Kumbharia, narrative panels at Sravana Belgola (Candragupta Basti), narrative paṭlīs, Western Indian miniature painting of the 14th century (limited corpus of cliches, formative period of the canon of miniature painting), painting of the 15th century (diversification of the cliches: Uttarādhyayanasūtra etc.), cosmographic representations, mantrapaṭa's, and “gyān caupaḍ's“.

So far we have always used the word “distinction“ although it is obviously necessary to distinguish (in theoretical contexts) between the use of the word “distinction“ in the sense of conceptually separating different planes (e.g. style and iconography) from that in the sense of distinguishing different subjects (e.g. Northern and Southern Jaina iconography). But we feel that there are so many possible cases and constellations of distinction that a restriction to two “pure“ forms would be a hindrance to, rather than a guarantee for more clarity. It is needless to add, that we feel free to use in non-theoretical contexts always those terms which seem to be best suited for the occasion: speaking on different levels, separation, segmentation, classification, typology, isolation, selection etc. However we think that systematization of these terms should be avoided as long as it is not absolutely necessary to do otherwise.

§ 2. Hermeneutics

The concept of hermeneutics is useful for the discussion of a number of issues.

First of all we have to admit and even emphasize that the understanding of narrative representations is more rewarding than the mere identification of pantheon deities. In the latter case, and in connection with all the “figures without clear identity,“ we must, however, focus attention on the difference between mere identification of a figure and its description. Identification always suggests a very limited approach, although of course it has some internal dimensions that are of theoretical interest (see below). By contrast, the scope for description, i.e. description beyond mere identification, is quite considerable, and prima facie evidence already shows that description is an important category (difference between Deogarh Jinas and Akota Jinas, and so on). All these arguments are significant independent of the purely quantitative argument that the bulk of Indian art is iconoplastic rather than narrative.

But apart from the competition between identification and description we have to consider a number of hermeneutic issues. Starting from the hermeneutics of narrative art, we have to refer the reader again to the line of research as mentioned in the previous section. ^[4] We may add that the phrase “hermeneutics of narrative art“ which indicates a variety of possible methodic strategies is by no means exaggerated. In the great Vessantara cycle in Cave No. 17 at Ajanta we notice for example that a figure like Jūjaka is better characterized (specified) than for example Vessantara himself. Again, an episode such as the massacre in the palace (Siṃhala cycle in the same cave) is better characterized than the numerous different episodes of different stories which are represented in a somewhat monotonous way by “palace scenes“ (undercharacterized from the point of view of hermeneutics). It would thus appear that the methodic spectrum is considerably more broad than appeared at first sight.

Here and in the following discussion we feel free to mention side by side facts of greater and lesser significance. Differences in the frequency of the phenomena (e.g. adequate or inadequate characterization in cycles of paintings) are not basic to our methodic discussions even though they must be considered as far as possible or necessary.

Hermeneutic problems in non-narrative art are much more numerous than would appear at first glance. In Bruhn, Āc. we have studied the Jaina ācāryas at Deogarh (a type without clear identity), and there we were faced with the problem engendered by a twofold opposition:

1. guru-like ācārya vs. śiṣya-like ācārya,
2. ācārya vs. monk (see also to § 10 infra).

Obviously, the latter opposition was created by the appearance of additional monk-motifs in the medieval period. As a consequence we now had two oppositions instead of one. In IJI we studied the opposition between Ṛṣabhas and non-Ṛṣabhas (IJI § 5B, see also § 10 infra), the issue of personal attributes and generic attributes (IJI § 4), the difference between attributes and non-attributes (IJI § 7 and 8A; see also § 9 infra), and the meaning of identification (IJI § 9). - An additional problem is the description-cum-identification of elements in cases where this is not relevant to the identification of the main figure. Often this is dismissed as “unnecessary“ because it would overburden the general description without supplying substantial information. But such elements do require closer scrutiny, for apparently “irrelevant“ features may occasionally turn out to be more important than would appear at first sight. It is possibly not incorrect to say that “hair-locks falling on the shoulders“ always identify Ṛṣabha, but that does not render a closer study of the different forms of the hair of the Jinas unnecessary (IJI § 5B, AJI § 10).

Hermeneutic problems of identification can also be demonstrated with the help of the question “what was to be represented?“ (fingendum = x). Here, “x“ becomes a slot which admits of quite different fillers, even as far as persons alone are concerned. The general formula “this is x“ produces statements such as “this is St. Peter“, “this is a disciple,“ “this is a Roman soldier“; or in Jaina iconography “this is a Jina“, “this is Ṛṣabha“, “this is Abhinandana“, “this is an ācārya“, “this is the Sacred Couple“ (so-called parents of the Jina), “this is Kubera“ (Jaina Kubera; mostly in combination with Ambikā).

The last category to be considered is the relation between narrative and non-narrative art (§ 1 supra). There is action in non-narrative art, and there is a static element in narrative art. In the first case, we have to refer to cāmara-bearers, celestial genii etc. which, as part of the general stylistic ethos, are in motion and convey a general feeling of movement. In the second case, which is not the true opposite of the previous one, narrative and iconoplastic motifs appear side by side. We refer not only to narrative panels in Gandhāra with the usual Buddha type in the centre, but also to the Kṛṣṇa cycle in Cave No. 3 at Badami with its occasional employment of the Viṣṇu cliché. In addition, we find representations of gods and goddesses (in the sense of main figures) which are midway between “narrative“ and “non-narrative“ representation. It may be objected that this is to be expected in the formative phase of an iconographic type. But on the one hand, we do not know whether we shall ever see an early Viṣṇu image where the god is actually using his club or his śaṅkha, and on the other hand we do not notice a gradual transition from narrative to non-narrative (iconoplastic) forms but long-lasting parallelism and interferences until the iconoplastic canon was fully established, thus putting an end to narrative experiments. ^[5]

A discussion of hermeneutics makes it necessary to mention also the problem of ambiguity (used here in its widest sense). In this connection we may mention the following possibilities: Gods-with-unknown-identity (early art), gods-without-established-identity (textual description nowhere traced), gods-with-arbitrary-characterization (on the outer walls of medieval temples), and gods-with-weak-characterization (again on the outer walls). More involved is the problem of the substratum. “Tree and Serpent Worship“ ^[6] did exist, and strictly speaking we are free to read a certain snake motif according to the context as “Buddha“ (in symbolic form) or against the context (but following the substratum) as “Nāga“. Closely linked with the problem of ambiguity is the problem of interferences between literature and art (“Wort und Bild“, “Erzählung und Bild“). The śilpaśāstras do not simply confirm the artistic conventions of the day, nor can it be assumed that the artists merely translated shastric descriptions into stone. We require different working hypotheses for different cases, and often the situation will necessitate very careful analyses. - Further hermeneutic problems will be discussed in § 8.