A Raga Is Not a Scale

Ask what a raga is and you will usually be told it is a scale, or a mode, or a scale with feelings attached. All three are wrong in the same way: they describe a set, and a raga is a set plus the rules for moving in it, and the rules are most of the content.

Start with the plainest demonstration. Two ragas can contain exactly the same pitches and be different ragas. Not variants, not regional names for one thing: different, recognisably so to a listener inside the tradition, performed differently, named separately, taught separately.

If the pitch set were the identity, this could not happen. So the pitch set is not the identity.

What else is there?

First, the ascent and the descent differ. The aroha is the way the raga goes up and the avaroha is the way it comes down, and they are frequently not mirror images. A note that is permitted coming down may be skipped going up. A note may be reachable only from above. This asymmetry is structural and no single list of notes can express it, which is why a single list is not an identification.

Second, notes have weight. Some are dwelt on, some are passed through, one functions as a resting place. Two ragas with identical pitch sets and identical ascent can differ in which note the melody keeps returning to, and that difference is audible and is the point.

Third, there is a phrase. The pakad is the catch phrase of a raga, and its function is identification: play it and a listener names the raga before any melody has developed, the way you recognise a familiar voice saying two words. This is recognition by gesture rather than by inventory, and it is how the tradition actually works in practice.

Now the thing that trips up almost every outsider, and should be said early rather than late.

Hindustani and Carnatic music are not two dialects of one system. They share ancestry and they diverged a long time ago, and the differences are not cosmetic.

Carnatic music organises its ragas under seventy-two parent scales, the melakarta system, generated systematically. Hindustani music has nothing equivalent; its organising scheme, the thaat system, has ten, and it is a looser classification that plenty of ragas sit awkwardly inside.

Worse for the beginner, shared names are unreliable. Hindustani Todi and Carnatic Todi are unrelated music. Hindustani Nat Bhairav corresponds to the Carnatic Sarasangi, the twenty-seventh melakarta, and not at all to the Carnatic Natabhairavi, which is the twentieth. The names look like a bridge and are a trap.

Finally, the time of day.

Hindustani practice assigns ragas to prahars, segments of roughly three hours, so that a raga has a part of the day it belongs to. Bilawal in the first prahar of the morning; Purvi in the late afternoon; Shankara late at night. This is real tradition, documented and taught, and it is also a convention rather than a law: performances happen when concerts happen, and many musicians treat the assignment as heritage rather than obligation.

Both halves of that are worth holding at once, which is the general skill this material teaches. The tradition is specific, it is not arbitrary, and it is not a rulebook handed down intact. It is what a large number of serious people did, variously, for a long time.