(Note: This turned out to be a lengthy topic. Since I ramble endlessly even for small topics, I will deliver this in a couple of installments, and this is the first. Also, all audio samples are MIDI (i.e. computer) generated – so may sound “lifeless”, and may have glitches and imperfections. Please accomodate.)
Please take a listen to these samples:
Sample 1:
Sample 2:
(If you have trouble listening to the streaming audio, please click on these links to download the samples: Sample 1, Sample 2)
Now, consider the following:
- Both are based on the same western music scale, the major scale, which has the same notes (swarasthanams) as the carnatic raga Sankarabharanam. Were you able to perceive this? Please listen again if needed.
- The two samples uses all the notes of the major scale. Are you able to spot any resemblance (even if you have stretch things a lot and it is still faint) between the samples and Sankarabharanam?
- If so, among the two, which one is kind of, sort of,”relatively closer” to Sankarabharanam? If you think it is sample 2, that is a good guess.
So what is the point of this exercise?
First, people with any exposure to carnatic music will hopefully conclude that neither sample can be considered to be even in the same planet as Sankarabharanam! This is perhaps obvious as both samples are in a western style, played with a distinctively western instrument – the piano. However, would you agree that the first sample is further removed from Sankarabharanam than the second – like it is not even in the same solar system as that raga?
If you don’t, consider this: the first one is a (slightly modified) version of the famous children’s rhyme Frère Jacques, while the second follows (approximately) the tune/melody of the first stanza of the varnam in Sankarabharanam. So the second sample is actually based on the tune of a carnatic composition in Sankarabharanam, and hence it better have a closer resemblance to it compared to the first! But it is still does seem “so different”, “so far” from anything that is Sankarabharanam and for that matter, anything Indian classical – right?
So why is a raga so different from anything in western music? Why does a tune in the raga Sankarabharanam sound so different from a tune in the western music major scale even though both the raga and the scale have pretty much the same notes/tones i.e. swarasthanams? That is the point of this exercise.
I will try to demonstrate some of the reasons behind this. Of course, this is not meant to be a comprehensive, exhaustive analysis and so I won’t cover all possible differences. And again, I am not an expert – neither in carnatic nor in western. Please let me what you think by posting a comment.
Let’s begin …
There are several differences between how Sankabharanam uses its constituent swaras vs. how a western music piece uses the notes in the major scale. Some are probably obvious, but some critical ones are actually not until you look closer.
Western Music: Key Shifts, Scale Shifts, Accidentals and Skipped Notes
It should first be noted that most western music pieces involve key shifts (shifts in sruthi), and/or scale shifts (like shifts in raga). Even among western music pieces that are in one key and one scale, I think it is not uncommon for some notes in the scale to be omitted, or “accidentals” (notes outside the scale i.e. anya swaras in carnatic lingo) to be present. All of this adds color to the western music piece. However for Sankarabharanam, none of this can be true – all are forbidden.
These differences do make it difficult for us to correlate a western music piece to any carnatic raga. This is a very significant difference between carnatic raga based pieces and western pieces. In fact, this probably means that comparing the western music major scale to Sankarabharanam raga is not that much meaningful.
However, let us limit ourselves to only those western music pieces that stick to one scale – the major scale, do not use accidentals, and use “pretty much” all the notes of the scale. There are such pieces – Frère Jacques being a simplistic example (although it actually does not use the seventh i.e. ni, and hence sample #1 above is a “modified” version above where I included it at the tail-end). In such pieces, where all notes in the scale and only those notes are used, this difference is negated, thus allowing us to do a better compare/contrast exercise against Sankarabharanam.
It’s the gamakas – stupid!
Another immediate and very obvious difference between Sankarabharanam and a western music piece in the major scale is that while in the western piece notes are held flat almost all the time, they are not commonly done so in carnatic. In carnatic music, it is way more common to modulate a note (i.e. render it with gamaka) than not. In short – it’s the gamakas stupid!
Let us take the major scale based on say D and take a simple, contrived phrase were one simply walks up the octave starting from the tonic (i.e. D or Sa) and then back down. This is how it would typically sound in western:
(Download by clicking here)
In Sankarabharanam, there are various ways of doing this, but we can safely say one must not do it as above i.e. with all notes held flat. Instead, one could do it as follows:
(Download by clicking here)
It should be obvious why I used guitar as the instrument above, and why piano would not worked. Also, as mentioned at the beginning, the pitch modulations are generated using MIDI i.e computer generated, and the results are not that stellar. Sorry.
Now, the difference should hopefully be obvious. In the second sample, while walking up the scale, some notes (2nd/ri, 4th/ma, 6th/da) are prominently modulated and not rendered flat. Similarly, while walking down also some notes are subtly modulated – differently from the modulations going up.
(Note: There is also the difference that most western instruments use equi-tempered tuning while carnatic uses natural tuning. I will just note that here but move on to other differences)
I should again emphasize that this is just one way of rendering the above phrase in Sankarabharanam. Each note can be modulated differently depending on the phrase, and sometimes differently depending on musician preference for the same phrase. It can be rendered flat too depending on the context. But no matter what, the fully flat version of sample 1, which would be quite acceptable in western music, would be completely unacceptable for Sankarabharanam.
Swaras vs. Notes: This may sound a bit technical and you can skip this para if it is hard to follow. In general, a swara in carnatic music does not represent always a fixed pitch/interval. Swaras are not notes in practice. One way to look at them is each swara can be considered as a “set of movements” that fall within a certain pitch range in the octave. Depending on the swara and the raga, this set may or may not include the “flat version”, which would be around a specific interval or swarasthanam that is roughly equivalent to that of the corresponding western note. The non-flat movements typically spans on either side up to the neighboring swara in the raga. A given swara in a raga (and in specific contexts) may have only the flat movement – at that swarasthanam. However, in another context it may be modulated.
In any case, the modulations i.e. gamakas, make a huge difference to the melody. They are easily the most characteristic difference between a melody that is raga based, vs. a melody that is not raga based. They are a big reason why sample #2 above did not really give out the feel of Sankarabharanam even though it sort of followed a tune of a song in Sankarabharanam. There were no gamakas in sample #2. So how would sample #2 really sound in carnatic style if we introduce gamakas? Take a listen:
(Download by clicking here)
Again, this was completely generated with MIDI, and the limitations of my effort are sort of more exposed by more complex modulations. Also I should note that here I have “struck” every swara – that is really not the way this song would be sung/played. There are words of the songs behind the swaras, and one really should be striking at most only those notes that coincide with syllables. Note also that here, every “note” in sample #2 is interpreted as a “swara” in that some involves a modulation/movement. If we were to notate this in western style (if possible) such that we account for the pitch modulations of the swaras, then we would say that the tune of this is quite different from that of sample #2. No wonder it sounds so different from sample #2! But then again just points to why/how Sankarabharanam is different from the major scale, and more generally why a raga is different from a scale.
Update: Ok, the popular consensus among readers is that the above sample sucked as bad as the original sample #2 in representing Sankarabharanam. They were saying without saying “Come on Arun, What kind of a cr$p was that?” (I understand but still – sob, sob! I sweated for you guys and this is what I get? 😉 ). So here is a professional version of that stanza of the song sung by the late (and great) Shri. K.V. Narayanaswamy:
(Download by clicking here) Of course in this version, we can hear the actual words of the song, and so you may not find an automatic, ready made correlation to sample #1. An instrumental version would have been better – but I could not find one. Also, note that this version has more differences beyond just rendering swaras with gamakas – he has added extra swaras in the form of flourishes. This is of course typical, and for any song there are always such variations. Also, please note that his pitch (i.e. tonic/sa) is lower than that of sample #2 – it is about 2 semitones lower.
In any case, this is how the song that sample #2’s tune was (very crudely) based upon should sound like. This is how Sankarabharanam typically sounds like.
Is that it?
The above are perhaps very obvious differences. They do help making ragas sound very different from their western music counterparts. However, there are other not so obvious, but nevertheless key differences that further underline and embolden the dividing line between Sankarabharanam (ot any raga for that matter) and its western music counterpart. I talk about these in Part 2.
August 20, 2007 at 1:54 pm
Good one Arun!
-neelanjana
August 21, 2007 at 9:26 am
thanks neelanjana
August 22, 2007 at 8:17 am
Way off on a tangent, maan!
Even I can’t comment on this!!
🙂
Change topic soon??
August 22, 2007 at 10:07 am
sorry doc – but unfortunately you are going to have to put up with such stuff quite often in this blog :)! I think I “warned” about this in my very first post. What can I say? I am just too interested in obscure details about music, which many may not care about 🙂
August 23, 2007 at 5:14 am
Excellent, excellent post!
To supplement the first sample and make it easier for us Carnatic-challenged folks, it would have helped if you would’ve included a typical rendition of “Sankarabharanam”.
What you call “gamakas” in Carnatic, is what we refer to as “shruti” in Hindustani. Many learned folks say that this is the one thing that makes Indian Classical music superior to Western.
However, I’m not sure what exactly you mean when you refer to “western music”. Do you mean western classical or western popular? The samples I’ve seen so far on your blog are the western popular ones. If you’re only referring and studying western popular, there’s a much greater universe out there for you to compare and draw parallels.
For one, western classical music focuses more on “Harmony”, unlike our ragaas, that focus on Notes and Melody. The western study of harmonies is I believe more profound than ours, while our studies of individual notes, their shrutis, and the ensuing melodies is more profound in a more spiritual fashion.
For more on exactly what I’m expressing, see: http://cnx.org/content/m12459/latest/.
And thanks for the wonderful post. I admire your passion…
August 23, 2007 at 5:19 am
For more on shrutis and why I recommended that link, here’s a quote from there:
“In some ragas, some notes may be flattened or sharpened by one shruti, in order to better suit the mood and effect of that raga. So, for tuning purposes, the octave is divided into 22 shrutis. This is only for tuning, however; for any given that or raga, only twelve specifically-tuned notes are available.
In spite of the fact that these tunings are based on the physics of the harmonic series, Indian music can sound oddly out of tune to someone accustomed to equal temperament, and even trained Western musicians may have trouble developing an ear for Indian tunings.”
August 23, 2007 at 5:41 am
hey arun,
Amazing post.Really enjoyed every bit of it.
“\\In any case, the modulations i.e. gamakas, make a huge difference to the melody. They are easily the most characteristic difference between a melody that is raga based, vs. a melody that is not raga based. They are a big reason why sample #2 above did not really give out the feel of Sankarabharanam even though it sort of followed a tune of a song in Sankarabharanam.//”
Very much true.
Music,in performance, is a type of sculpture. The air in the performance is sculpted into something.
Nice post.keep them coming 🙂
August 23, 2007 at 10:26 am
@sowmya – thanks for visiting again Sowmya! That MIDI piece took forever to construct, with painful retouches and it still sounds “fake” with blemishes 🙂 That is a testament to how much detail is required in music – just like sculpture, painting – arts in general.
@mahendra – thanks man! Yeah – I wanted to include a typical rendition of that particular song – but couldn’t find one. (But what?!? You didn’t think my MIDI version was typical Sankarabharanam 😉 ?) I didn’t have guts to sing it myself and use it. But now I realize, a proper Sankarabharanam would have been better. I will try to dig around more and include a “pro version” if I find it.
Regarding Gamakas and Shruthi:
I think Hindustani also has concept of gamakas (gamaks?). Carnatic also has the extremely important concept of Shruthi. But it is an overloaded, and in some contexts one could dare even say misused/misunderstood term. It is used to indicate
1. The tonic/key i.e. shadjam/sa (what Shruthi is the singer singing in or your instrument tuned to?)
2. Whether the singer is “singing in key” (so a slip of a swara is a Shruthi slip).
3. And, yes it is used to indicate microtones i.e. 22 sruthis – although some late scientific research confirms that there is lot of misconception here (i may post on this later – it is a can of worms).
But gamakas (at least in carnatic) are not the same as Shruthis – one could say they are a “higher level concept” that uses the underlying Shruthis. They involve specific modulations of pitch with some modulation in volume/emphasis – so slides/glides, shakes etc. with varying emphasis would be the gamakas.
Western music
Yes my samples were indeed more “popular music” oriented although I meant western classical too (but I confess that my exposure to it is minimal). However, with both, I consider only certain aspects, since once you bring in harmony and chords, you no longer can make a meaningful comparison with Indian classical music. By “certain aspects” – I mean that for many western pieces, even if they contain lot of harmony, it would be possible to focus only on the melody line of say a solo, and see how it uses the underlying scale(s). I was implying that even there, we will find differences (a bit more on this in part 2). That is another reason why a melody line of a western piece sounds so different from the raga. The harmony/chords, key-shifts, scale-shifts and flat notes are of course the more obvious differences. I am not ignoring them but taking those differences as granted.
Sorry for the long comment but I hope I am making myself clear here! Wow, the way things are going, soon my comments will be longer than the main post 😉
August 23, 2007 at 11:52 am
I came to blogroll you, then saw more of Carnatic music stuff, turned back till the door and then thought, what the heck, the world’s not perfect!
So here you are: blog-rolled! 🙂
BTW, I don’t see your blogroll?
August 23, 2007 at 12:20 pm
Oh, my browser hadn’t managed to load the right side of the page when I typed the comment…now I see it.
August 23, 2007 at 1:02 pm
arun, mahendra,
I think one of the gamakas (of the 10 or 15 standard gamakas described generally)is also called shruti 🙂 I am sure having read it in one book or the other. I can cite the source later after some search.
mahendra said:
>>To supplement the first sample and make it
>> easier for us Carnatic-challenged folks, it
>> would have helped if you would’ve included a
>> typical rendition of “Sankarabharanam”.
and arun replied:
>>(But what?!? You didn’t think my MIDI version
>>was typical Sankarabharanam ?) I didn’t have
>>guts to sing it myself and use it. But now I
>>realize, a proper Sankarabharanam would have
>>been better. I will try to dig around more and
>>include a “pro version” if I find it.
I have to agree with mahendra too – The midi version of sAmi ninne is almost from another planet, if not from another universe 😉
-neelanjana
August 23, 2007 at 1:10 pm
@rambodoc – 🙂 thanks. I blogrolled you too.
@neelanjana – Ah! Didn’t know about it. That would add yet another interpretation of term ‘Shruthi’ (although i was only referring to how it is used today – I dont think we use shruthi for gamakas in cm today do we?)
Regarding MIDI version sAmi ninnE 🙂 – see this is precisely why I posted the MIDI version. I have something else to put the blame on! If it was mine and it sounded as bad (a good possibility) …
August 24, 2007 at 2:30 pm
In another few years, I will take printouts of your blog, and tell Sapthaswara music in Mylapore to publish it and give Prof Samabamurthy a rest 😉
That said, you have got to pick up a guitar and try those gamakas out. Piano just doesnt cut it for indian classical music
August 24, 2007 at 3:24 pm
mahendra, neelanjana, krishasok – okay you purists 😉 – I updated and uploaded a real Sankarabharanam.
Ashok – thanks. Yes piano is useless of Indian classical stuff. BTW, the last sample (i.e. the one before the KVN sample) was not piano – it was actually a MIDI guitar instrument based on real guitar sample. I can play the guitar (the real one) a bit but mainly rhythm. In fact, the reason I gave up guitar is because ones I got into carnatic, it reminded me constantly how much I sucked. I could never play much lead, and to play carnatic gamakas requires a lot of slides with precise timing – I just could’nt do it. It was easier to sing it which is what I (try to) do nowadays.
August 29, 2007 at 1:15 pm
Arun,
Found the source of “shruti” as a gamaka. Parameshwara, in his treatise called “Veena Lakshana” (Probably post 1700 AD) describes it as one of the gamakas. He does not specify the number of gamakas to be 10 or 15, but mentions most of the standard ones like humphita, ullasita, orika etc.
Prof. R Satyanarayana has written an elaborate discussion on this book.
Suruti is one rAga described in the book, which takes the ‘shruti’ gamaka on the rishabha.
-neelanjana
Arun: Thanks!
August 30, 2007 at 12:19 pm
Hi Arun.. This is really very educative.. Can you please come up with the raga Attana this way(the way you have come up with Kaanada, to be more precise)?
August 30, 2007 at 3:33 pm
thanks ramakrishna. 🙂 – I would love to do it aTANA, another favorite raga of mine. So far the requests have been
1. rItigowLa/Anandabhairavi
2. kApi
3. and now aTANA.
All ragas I love – the trouble of course whether I will able to do proper justice. I am definitely not an expert but I will definitely want to try to the best of my abilities. However, I first need finish part 2 of this, and then there is another raga I love which I place ahead of these. So in time, yes.
Thanks again.
August 31, 2007 at 4:08 pm
[…] by Arunk under Carnatic Ragas , Carnatic , Carnatic Music This is a continuation of the Ragas and Scales post covering why a raga based melody sounds so different from a melody that is western melody even […]
September 6, 2007 at 3:22 pm
Hello Arun.. Definitely you will be able to.. Part 2 was also very good.. and one more thing.. Where can i find the complete rendition of MMI Sukhi Evvaro? I searched every nook and corner of the net, but couldn’t find.. If you dont mind, can you please mail me the same? my email id is:
ramakrishna_paramahansa@yahoo.co.in
Arun: Thanks ramakrishna. I will email it to you as soon as possible.
September 9, 2007 at 5:29 am
hi Arun.. Thanks.. But, i still haven’t received any mail..
Arun: I was a bit busy the last couple of days. I have sent it to you now
September 27, 2007 at 4:41 pm
[…] sophisticated melodic concept. Treating it like standard “scalar structure” (I know, wrong term) would result in something that sounds different with some vague and fleeting resemblance to […]
February 27, 2008 at 7:04 am
Great Stuff !! Thanks man !
Arun: Thanks Sanjit. Welcome to my blog!
May 27, 2008 at 2:56 am
Arun, if there is a Pulitzer prize equivalent in carnatic music, consider it awarded to you. Great and very involved piece of research to enlighten the masses like me.
July 9, 2008 at 2:48 pm
Very educative.
July 10, 2008 at 6:33 am
Hallo Arun
There is a pentatonic raga that I really like but have never been able to find out the name of. In western notation the notes are C, E, F,G Bb. Do you know this one?
Thanks. Great piece by the way.
Arun: Thanks Dirk. I think this scale is not that popular in carnatic music. This seems to be called saavithri – but I think this is not an established raga (in terms of compositions, expositions etc.) and is perhaps a modern one. There are a couple of known ragas which are sort of close to it. One is bahudaari which includes A in the ascent (so C E F G A Bb C going up and C Bb G F E C coming down), and the other is thillang which has B in the ascent but Bb in descent (i.e. C E F G A B C going up, and C Bb G F E C coming down).
July 16, 2008 at 7:28 am
I think it’s quite a common one – maybe I got the pitches wrong. If the root was F rather than C, what would that be called? (With C root the notes would then be CDFGB – megh with a B natural)
Many thanks
Dirk
July 16, 2008 at 11:28 am
Hi Dirk,
From http://en.wikipedia.org/wiki/Megh, it looks like Megh has fifth (pa) in ascent and fourth (ma) in descent(?). In any case, both C D F G B or F G Bb C E (i.e. your previous scale but with F as root) would be equivalent indeed, and represented in carnatic notation as
S R1 M1 P N3 S | S N3 P M1 R2 S.
Unless I am mistaken, this again is not popular as a symmetric scale.
But it could be that you are referring to either:
madyamaavathi:
C D F G B C | C B G F D C
or
S R1 M1 P N2 S | S N2 P M1 R2 S
(seventh note is minor seventh i.e. komal-ni like megh)
This is very popular in both carnatic and hindustani. It is a major raga in carnatic.
or
brindavanasaranga:
C D F G Bb C | C B D G D C
or
S R1 M1 P N3 S | S N2 P M1 R2 S
Correction: Most flavors actually include minor third (G2 in CM) in descent)
Here the seventh note is major-seventh in ascent and minor-seventh in descent. This is also very popular in carnatic music, and is an import from hindustani. It is not as major a raga as madhyamaavathi, but nevertheless popular.
I should also note that there are some subtle variations with respect to the seventh note in various flavors of this raga – in carnatic music circles, even in ascent it is considered a minor-seventh (kaisiki nishadham) with a very perceptible tinge of major-seventh (kakali nishadham). Some versions skip the kakali altogether – which although may make it similar to madhyamaavati above, is still entirely different in terms of intonations of other swaras.
July 17, 2008 at 5:20 am
Arun
Many thanks indeed for your generous attention. Now I have the information I wanted – madyamaavathi.
Brilliant stuff!
Best wishes
Dirk
Arun: You are welcome Dirk!
October 10, 2008 at 3:29 am
wow, nice post.
Arun: Thanks and welcome to my blog.
October 8, 2009 at 6:05 am
hey, arun.
great post! i enjoyed reading about it because i’ve always wondered about the differences. very illuminating. 😉
March 9, 2011 at 12:32 pm
Very nice! Namaste!
May 19, 2015 at 8:15 am
Thanks,very interesting
September 2, 2023 at 12:52 pm
Excellent work. By the by I still confused about that the tune of come September resembles Sankarabaranam . I would like to be enlightened.