Iacobus Leodiensis [Iacobus de Montibus, Iacobus de Oudenaerde]

(c) Origins, number and change.

Rāgas are traditionally considered to be apaurusa, ‘non-human’, in origin. Mythology attributes them, for example, to the five heads of Śiva, the patron deity of music, and his consort Pārvatī, or to the songs sung in praise of Krsna by his 16,000 milkmaid consorts (gopī). Sources from the Brhad-deśī onwards attribute the origin of the rāgas to the earlier system of jāti (themselves created by Brahma) by processes of variation and mixture. However, the names of many early rāgas imply an origin in regional (deśī) musics, including the musics of ‘tribal’ peoples (e.g. Todī, Śabarī) and foreign invaders (e.g. Śaka, Turuska-(‘Turkish-’)gauda). Many rāga names still suggest regional or ethnic origins, even (since the Muslim period) in areas outside India (e.g. Yaman, Hijāz).

The number of rāgas in existence cannot be defined exactly. A traditional number is 36, deriving from the medieval rāga-rāginī systems, but the actual number has probably always been greater. As an example, 260 were classified by the 13th-century theorist Śārngadeva. For Hindustani music, Bhatkhande gives 186 in his Kramik pustak-mālikā (1953–5), but in that tradition a performing musician will have a working repertory of approximately 40 or 50 rāgas. Kaufmann lists some 2000 Karnatak rāgas (1976), compiled from various sources, but again the working repertory of an individual musician would be considerably less than this.

Uncertainty as to the total number of rāgas is due partly to the absence of any single authoritative source, written or oral, and partly to changes in the repertory as new rāgas are introduced and old ones forgotten. The ideology of rāga exercises a check on the creation of new rāgas, however, since they are considered to be of divine rather than human origin and, according to some, all viable rāgas exist already. The invention (or discovery) of certain rāgas is attributed to or claimed by particular musicians, and in the south, rāgas have been created to exploit all the 72 scalar possibilities of the melakarta system (of which only 19 were in use in the 17th century). But many ‘new’ rāgas are in fact combinations of elements from existing ones. Another method of introducing novelty without presuming on divine prerogative is to import Karnatak rāgas or scale-types into Hindustani music, or vice versa (see §(v) below). In the north, each gharānā has a number of ‘secret’ (acchop) rāgas that, although claimed to be old, are in effect new when first revealed in public (Neuman, 1980).

Out of respect for the divine origins of rāga, many musicians also hold that individual rāgas are not susceptible to change but are still performed (by those who know them correctly) in their original forms. Documentary evidence shows that although some significant continuities can still be traced between modern and earlier rāga forms, profound changes have occurred. Many rāgas have changed their scale-type or have generated variants in different scale-types, as the difference in scale-type between many similarly named Hindustani and Karnatak rāgas suggests. Such changes have been attributed in part to the combination of attributes from different rāgas (Powers, 1970) and in part to the inherent asymmetry of musical scales, which may have induced musicians to make compensatory adjustments in scale-type, melodic movement, intonation, ornamentation etc. (Jairazbhoy, 1971).

India, Subcontinent of, §III, 2: Theory and practice of classical music., Rāga.

(ii) Historical development of rāga systems.

The term rāga occurs in the Nātyaśāstra only in its non-technical sense of ‘passion’. The term for melody-type or modal category in that source is jāti, ‘class’. A jāti was distinguished from the grāma pitch-collections and the mūrcchanā scales on which it was based (see §1(ii)(a) above) by the attribution of special functions to particular svara. These functions included graha, the initial note; amśa, the predominant note; nyāsa, the final note; and apanyāsa, a note or notes on which intermediate cadences might fall. In addition, particular svara might be frequent (bahutva) or infrequent (alpatva), and melodic movements between certain svara were specified. The primary or ‘pure’ (śuddha) form of each jāti was heptatonic, and the same svara served as both predominant and final. However, each jāti also admitted a number of variants (vikrta), in which specified svara other than the final could serve as predominant, and one or two specified svara might be omitted. The jāti was thus a modal category under which a number of related melody-types, not individually named or identified, might be classified.

Some clue as to the origins of the jāti, and their relationship with later modal systems, may be provided by their names. The seven ‘note-born’ (svara-jā) jāti, considered to form a primary set, are each named after the svara that is their final (e.g. Sādjī, with sadja as final). They form a sub-system in which scale-type, as defined by the final, is the primary criterion of classification. The remaining 11 ‘mixed’ (samsargajā) jāti are explained as combinations of two or more primary jāti. Since several of the mixed jāti have the same final, scale-type is no longer the sole criterion of classification, and these jāti begin to resemble melody-types. Some of their names hint at regional origin (thus Sadjodīcyavā, ‘northern Sadja’ etc.).

A number of sources from the middle of the first millennium ce, including a passage in the Chinese Sui History referring to the late 6th century and the Kudumiyāmalai inscription in south India dating from the 7th or 8th century, refer to a different set of seven melody-types, to which the term grāma-rāga became attached. By the time of Matanga (8th–9th century) these formed the basis of an extended repertory of 32 grāma-rāga together with numerous variants called ‘dialects’ (bhāsā). These early rāgas are defined using terms borrowed from the jāti system (amśa, nyāsa etc.), but the alternatives inherent in the jāti are not allowed. Each rāga comprises a specific and unique combination of structural characteristics, to which its unique aesthetic effect is attributed. A number of the grāma-rāga bear names suggesting an origin in northern regions of the subcontinent, while the bhāsā include names referring to peoples and places from both the north and the south (Widdess, 1993). The basis of classification was now performance style (gīti); as with the mixed jāti, scale-type was not a criterion.

By the 12th century the old grāma-rāga were no longer used in concert music, though they continued to be described and illustrated with notated examples in theoretical works. The current repertory became known as the deśī (‘provincial’) rāgas, of which some were based on earlier grāma-rāga or bhāsā, some were hybrid forms, and some originated in local traditions. The Sangīta-ratnākara (13th century) classifies rāgas according to their place in a quasi-genealogical paradigm: grāma giving rise to jāti, giving rise to grāma-rāga, giving rise to bhāsā and deśī-rāga. But the earlier generations of this paradigm were abandoned by most later authors, since they had long ceased to be relevant in practice. The Sangīta-makaranda (attributed to Nārada, ?14th century) classifies the rāgas according to various criteria: performance time (morning, midday or night), number of pitches (five, six or seven), threefold gender (male, female or neuter), twofold gender (eight male rāgas with three ‘wives’ each) and degree of ornamentation (much, moderate or none, a throw-back to the gīti classification of grāma-rāga). Note that scale-type is not treated as a basis for rāga classification at this period.

Various male-female rāga systems (mata) were current in different regions of north India during the 16th to 18th centuries, and some were popular with painters of rāga-mālā (see §II, 3(iii) above). A common pattern comprised six male rāgas each with five female rāginī, totalling 36, to which sons (putra) could be added. There is little evidence that these systems were based on any scalar similarity between a rāga and its dependants.

Scale-type as an aspect of rāga definition and classification was introduced by Rāmāmātya (c1550) and became the standard method of classification in the south. However, Rāmāmātya also classified rāgas according to their potential for melodic development in ālāpa and prabandha compositions. The Caturdandī-prakāśikā of Venkatamakhin (17th century) expanded the number of theoretical scale-types to 72, but only 19 were required for classifying the rāgas then current. In later Karnatak music scale-type has become the overriding criterion of classification, such that the principal rāga in each scale-type is called the ‘parent’ (janaka) and the others its ‘children’ (janya), and rāgas have been developed in all the 72 scale-types of Venkatamakhin's system.

From the 16th century onwards some Hindustani treatises advocated scalar definition and classification on the model of the southern scale-type system, while others clung to rāga-rāginī typology (see §II, 3(iv) above). The latter is now obsolete, but Hindustani musicians vary in their attitude to scalar classification, some accepting Bhatkhande's thāt system, others not. An alternative northern approach to typology associates rāgas in groups or ‘rāga-families’ (rāg-kul), where each group comprises variant or related rāgas characterized by shared melodic features but not necessarily a shared scale. The continuing importance of motivic relationships between rāgas has been discussed in historical and contemporary contexts by Powers (1970).

India, Subcontinent of, §III, 2: Theory and practice of classical music., Rāga.