In Hindustani classical music, we do not notate music for performance purposes because a classical music artist is by definition one who is capable of extemporaneous raga development, and a classical music performance, by definition, is an act of extemporaneous raga development. We do, however, use notation to teach and learn music, and as an aid to memory. When you learn a new raga, you notate a few basic melodic phrases, patterns and simple compositions in that raga so you can recall them later.
Traditional Notation System
There have been many systems of notation in Hindustani classical music over the centuries, but a system proposed by musicologist Vishnu Narayan Bhatkhande (1860-1936) gained widespread acceptance during the early 20th century, and is commonly used to this day. It uses the Devanagari (Hindi) script for the notes and the lyrics, and a few simple symbols for other musical markers. The Devanagari script is ideal because the lyrics of Hindustani music compositions are almost always in a dialect of Hindi. But with the globalization and digitization of Indian music, it is becoming more common to romanize for the sake of convenience. Below is a romanized example of the Bhatkhande system. Explanations follow the image.
A bandish in Raag Kedar notated using the Bhatkhande system
Teentaal is a 16-beat rhythm cycle, comprising four sections of four beats each. The sheet is, therefore, divided into four sections by drawing three vertical lines, so that each line of lyric can be neatly fitted into a single row.
The entire bandish has only four lines of lyric, two in the first stanza (sthayii), and two in the second (antara). For each line of lyric, there are three rows of notation. The first row notates the melody, the middle row contains the lyrics, and the last row provides rhythm markers.
The melody is notated using note names and a few other symbols. The note names are Sa, Re, Ga, Ma, Pa, Dha and Ni (romanized from Devanagari). Flat versions of notes are shown by an underline, the sharp version is denoted by a vertical line above the note. If a note belongs to the octave above or below the main octave, a dot is placed above or below the note. A hyphen indicates that the previous note is to be elongated. When two notes have to be sung within the space of a single beat, they are joined underneath by a curved line.
The lyrics are written in the middle row, syllable by syllable, to show what syllable must be sung to what note. A symbol resembling a large "S" is used to indicate that a syllable must be elongated or sustained.
The third row of notation contains rhythm markers. Not all the beats in the rhythm cycle are notated. Only the important beats are notated, using just three kinds of rhythm markers. A cross (x) indicates the starting point of the rhythm cycle (the first beat of the first section). This does not usually coincide with the starting point of the bandish. Some sections in the rhythm cycle start with an empty beat. These sections are denoted by a small circle (o). If a section is neither the first section, nor starts with an empty beat, it is simply denoted by the section number. Please see my page on Rhythms to understand this better.
The Bhatkhande system works well for those who notate music by hand in Devanagari, but it poses problems for those trying write or share notations digitally. You will, therefore, find that the various websites on Hindustani music online (including this site) have tweaked the traditional system slightly to adapt it to the digital medium. Which is fine, except that there is no uniformity, as the digitization of Indian music is still in its infancy and there is a lot of experimentation going on. Until a new system suited for the digital medium is perfected and popularized, we may have no choice but to put up with this lack of uniformity. Most websites, however, provide at least a cursory explanation of their notation system. Below is the system I use. It is quite similar to what you will find elsewhere.
A Simple System for Romanized Digital Notation of Hindustani Music
Music cannot be written without a time signature, so let us start with that. Hindustani music uses a very intuitive system for encoding time - by structuring the melody and lyrics around the rhythm (taal). Take Jhaptaal as an example. A standard Jhaptaal cycle has four sections and a total of ten beats:
Standard Jhaptaal - 10 beats (2/3/2/3)
dhin naa / dhin dhin naa / tin naa / dhin dhin naa
A melody set to standard Jhaptaal could, therefore, be notated in a table with 10 columns, bold lines used to clearly separate the four sections of the rhythm cycle (see example below).
In classical music, the tempo is only specified to the extent of whether it is slow (30 to 70 bpm), medium (70 to 140 bpm), fast (140 to 350 bpm) or rapid (350 bpm and above).
The lyrics of the bandish or any other song that needs to be notated are broken up into lines based not on meaning or number of syllables, but on rhythm. Whatever fits into one taal cycle is considered one line of lyric for purposes of notation (see example below).
I use Pundit Ravi Shankar's Notation IDs to encode the melody. See my page on Tones to understand this better. Or if you simply need to refresh your memory, here you go:
Key Signature (name of the raga)
Technically speaking, in the Indian system of notation, you do not need a key signature, as each note in the chromatic scale already has a unique notation ID. But in fact, it is extremely important to specify the name of the raga, because a raga is much more than just the sum of its notes. Many ragas share note combinations with other ragas, but sound very different either due to a difference in microtones or the way the tones are traversed (directly/circuitously) and so on. It is important to allow an artist to prepare for the emotional content of the raga by specifying the name of the raga.
Here is an example of my system of musical notation. The information you need to know beforehand is provided above the table. The first row of the table gives you the name and structure of the taal, the second row gives you the vocalization of the beats in the taal. This is followed by the first stanza of the bandish. The black rows are the lyrics, the melody for which is encoded in red below. The third portion is the second stanza of the bandish, which also has lyrics in the black rows and melody encoded for those lyrics in red below. And so on.
All the notes in a box must be sung within the space of one beat. Most of the boxes are crammed full of notes, but since this is a slow-tempo bandish, four quarter notes can be sung easily in the space of one beat. A hyphen indicates that the previous note must be sustained for the space of a quarter note. An exclamation mark (not used in this example) would indicate a sharp break/pause before going on again. "Rest" indicates for the singer to remain quiet for the space of the beats in question.
Composition type: Bandish
sakhii morii, ruuma jhuuma baadala garuje baruse
reinaa andherii kaarii, bijurii chamake, keise jaauun piiyaa paasu, baadala garuje baruse
Tempo: Slow (approx. 60 bpm)
Here is a simpler score in the same raga (Raag Durga) for practice. Use any instrument of your choice, decide where your root is, figure out the positions of the other notes relative to your root. I have given this score a very simple time signature - an 8-beat rhythm cycle, no divisions. Follow the score to play the ditty first, then compare your results with the audio in the link. If you don't play any instruments, listen to the audio in the link first, then try to sing along in sargam (solmization) using the score.
Rules: Each row, representing one whole cycle of the rhythm, is divided into eight boxes. Each box represents one beat. No matter how many notes are in the box, they must all be played/sung within the duration of one beat. A hyphen indicates for the previous tone to be sustained, while an exclamation mark indicates a sharp break. This melody is best suited for a slow tempo (up to 70 bpm).
Rhythm: 8-beat rhythm
Tempo: slow (approx. 70 bpm)
Scale: Raag Durga