Bibliography
In Hindi:
Bhatkhande, V.N., Hindusthani Sangit Paddhati, Vols. I-IV, Hathras 1951-57
Bhatkhande, V.N., Kramik Pustak Malika, Vols.I-VI, Hathras 1954-59
Thakur, O., Sangitanjali, Vols. I-VI, Banaras 1954-62
Roy Chaudhury, V., Rag Vyakaran, New Delhi 1981
In English:
Subba Rao, B., Raga Nidhi, Vol.I-IV, Poona 1956-66
Kaufmann, W., The Ragas of North India, Calcutta 1968
Bor, J. (ed.), The Raga Guide, Nimbus Records 1999.
In 1982 M. Levy published his Intonation in North Indian Music, A Select Comparison of Theories with Contemporary Practice (New Delhi). This work, largely inspired by his mentor N.A. Jairazbhoy, completely demystified the shruti theory. At the time it came out Bernard Bel, Jim Arnold and the present author were working on the Melodic Movement Analyser (MMA), a filter based pitch extractor. With the help of computers enormous amounts of MMA output could be analyzed. In articles published in the Istar Newsletter (No. 2, New Delhi 1984 and No.3-4, New Delhi 1985), Bel cs. criticized Levy's methods and results. In those days, we were mesmerized by the mathematical elegance of the classical shruti theory and felt they simply must be applicable. In the following years I rewrote the software for pitch extraction and went on processing large amounts of data. Slowly I came to the conclusion that in essence Jairazbhoy and Levy were closer to the truth, and that contemporary practice at best paid verbal tribute to ancient shruti theories. I would go one step further than they did however, since I no longer believe the measurement of "shruti" can be made operational (see the Todi sample). In fact, the concept of shruti as a unidimensional pitch (a static frequency) should be discarded and replaced by a multidimensional gestalt of pitch, timbre and volume that change in a time frame.
Other references
Bose, N.K., Melodic Types of Hindusthan, Bombay 1960.
Deva, B.C. and P.S. Nair, Forms in Music, in Sangeet Natak 2, Apr. 1966, p. 106-117.