Listen to this audio clip: (A warning: If you only like “soft music” you may want to turn your headphones down a bit 
. And of course all rights belong to BOC and the record company. This is here just for illustrative purposes!)
Aside: Oh and by the way, what’s up with wordpress not allowing mp3 file upload but will allow videos?
It is a guitar solo featured in a famous classic rock song . This is one of my favorite songs in spite of the possible creepy interpretation of the lyrics, and it came on the radio a couple of days ago. For the first time, I started thinking Hmm. This has that feel of a carnatic raga. What raga is this?
I think it is dharmavati although I am not 100% sure. It comes off much darker in mood for some reason. But if it is, that is pretty cool. I mean a really hard rocking band doing a blistering guitar solo in a scale that is very rare in their neck of the woods, and I would say rare even in our carnatic world.
If you listen or have listened to classic rock actively you may recognize it instantly as both the song and this solo are quite popular. For people who don’t – the song is called Don’t fear the reaper by the band Blue Oyster Cult (1976). The song is quite notorious for its lyrics as it seems like a guy telling his lover that they should end their lives together. In fact, that is the popular interpretation and the band was deemed satanic and what not. It seemed to me that they may have played up to that notoriety (judging my some of the album covers) but as far as this song goes – I now see from http://en.wikipedia.org/wiki/(Don’t_Fear)_The_Reaper that this was not the songwriter’s intention at all. In fact it is philosophical.
Update #1:
A second thought: I noticed that the guitarist’s name is Donald “Buck Dharma” Roeser. Is that a strange coincidence or what ? But, I must admit, I still cannot say for sure if it indeed (even loosely) based on dharmavati. Also, for the record, I did not know his name when I first thought it could be dharmavati !
Update #2:
Krish said (see comments) that this is hEmavati/haimavati and not dharmavati and I agreed it is possible. Both differ in one swara – the nishAdam – kaisiki i.e. minor seventh for hEmavati, and kAkali i.e. major seventh for dharmavati. After that I did some googling and guitar tabs I could find seemed to indicate that perhaps both are used in different contexts.
I think I understand why it is so after doing some math based on info gathered from the website of the Blue Oyster Cult guitarist Buck Dharma’s site itself: http://www.buckdharma.com/Guitar/DFTR_guit.pdf.
Before I offer the nutty gritty details which most of you may not be interested, the short answer is that it is a mix of two scales – one of which is indeed dharmavati. The mix of the other does seem to introduce dharmavati plus the other nishAdam
If you are interested in why/how read on.
The PDF document at the URL says:
Middle Eastern in nature, the guitar solo played over the alternating F minor and G7 tonalities (sic) is more rhythmic than melodic. The scale used over F minor is like melodic minor but with a sharp 4.
The melodic minor scale (see here) is gowrimanOhari. Raised fourth means fourth i.e. madyamam is raised by a semi-tome and hence becoming prathimadhyamam and thus dharmavati
Basically in western scale, here shadjam/tonic is F and you have F G G# B C D E F
It also says:
Over the G7 chord a scale based on C harmonic minor starting on the 5th scale step is used. Characteristic of this scale are the flat 2nd and the flat 6th scale steps.
Harmonic minor is kIravANi and starting on 5th scale kIravAni starting from pancamam (i.e. the fifth). I think this means grahabedham/tonic-shift from pancamam, since that leads to vakuLabharaNam, which has suddha-rishabam (flat 2nd) and suddha-dhaivatam (flat 6th).
Now since kIravANi based on C is: C D D# F G A B C and such a kIravANi starting from fifth i.e. F is F G A B C D D#
If you compare this with the dharmavati on F, you find D# instead of E, which is the hEmavati nishAdam instead of dharmavati nishAdam. There is also A (antara gAndaram) as opposed G#(sadharaNa gandaram). I am sure it appears somewhere.
The end
PS: Let it be told that never has so much said about something so obscure!
July 12, 2007 at 10:08 am
Arun,
Oh yeah. Don’t fear the reaper. Love that song. Hemavathi though. Not Dharmavathi. “Sree Kaanthimatheem” by Muthu Dikshatar is one of the finest in that raga
July 12, 2007 at 10:14 am
Yay! The intersection set of “people who like carnatic music” and “people who give a hoot about BOC” has just been doubled
!
Hemavathi? Possible. I do know about kAntimatIm – a favovite of mine. The only difference between the two being ni – need to listen again. I thought I felt a major-7th i.e. N3 feel but I am not that good on this anyway. So even if it is hemavathi – 6 out 7 ain’ bad
July 12, 2007 at 12:37 pm
Actually this seems like chakravakam (with D1 thrown in).
Conversely you can say it is vakulabharanam (with D2 thrown in).
It has the swaras S R1 G3 M1 P D1 D2 N2
Actually IMO ‘rare’ carnatic ragas are not all that infrequent in rock. For eg. Kashmir (Led zeppelin) has traces of nasikabhushani.
And Chakravakam, I think is also a favourite rocker scale — I have a guitarist friend who frequently plays this tune.
July 12, 2007 at 12:49 pm
krish – please check last update.
vijay thanks – and please also check last update. Your view of vakuLAbharaNam is also implied
July 12, 2007 at 1:35 pm
Yes vakulabharanam, Kiravani, Hemavati and Kosalam form a group of ragas which are transformable into each other based on the sruti bedham relation.
July 12, 2007 at 1:42 pm
I think melodies like kashmir are more off-beat rather than the norm in rock/pop music world. But Led-Zep did do some off-beat stuff.
July 12, 2007 at 1:45 pm
…while chakravakam, Sarasangi and Dharmavati form a similar group!!
This is very interesting. For me it sounded like vakulabharanam/chakravakam — but for you/krishashok it sounded like hemavati/dharmavati.
Each of us apparently deciphered the notes correctly, but we had differing opinions on what shadjam (sruti) is.
I had written about this in this post (scroll towards the end)
http://nvijayanand.blogspot.com/2007/05/partitioning-melakarta-ragas-into.html
“4. These equivalence classes of ragas have an important consequence for an automated raga identifier. If a naive identifier worked by finding the swaras in the composition, supposing it found the notes corresponding to say kharaharapriya, then how can the identifier be sure its is not any of the other 5 ragas from the same class? Thus I claim, any automated raga identifier must first identify the sruti (the shadjamam) of the composition, before it can identify the raga.”
I never thought this (deciphering the sruti) would be ambigous as far as humans are concerned!!!
July 12, 2007 at 1:56 pm
“automatic raga identifier” – brings back memories of my “grand scheme” (before I came back to earth). But yes identifying shadjam is of paramount importance for such a system.
As far as this solo goes, my theory is that since the chords alternate along with the bass lines, and other rhythm sections, the two “shadjams”/tonics are alternatively emphasized. The emphasis can be accentuated by the nature of the bass, rhythm sections. So some latch on to one and other to other – this is not uncommon with film songs also.
I dont know if you have noticed, once you latch on to a particular pitch for shadjam, to shake that off is quite difficult. This can happen even if you sing and go a tad off-shruthi (happens often for me
. You almost have to shake your head to “clear” it!
July 23, 2007 at 6:09 am
Hey Arun ..
Thanks for visiting my Blog !! and yeah pretty interesting post.
On Raaga identification, i have been wondering if we could identify the raaga (among its equivalence class) based on the sequence of swara patterns that have come so far .. (??) and with a serious bit of Machine Learning and Probabilistic reasoning ..
If the artist actually was singing Shankarabharanam, the swara patterns would deem fit a Shakarabharanam. Yup, We need to go thru all the possible raagas within the equivalence class and find which one suits best (read most probable) for the swara patterns that have come so far .. probably a Hidden Markov Model would do the trick .. psssst ! This could obviate the need to identify the shruthi.
PS(i): You need some way of encoding (what is a swara pattern ?) and serious training for this to work..
PS(ii): What am saying might be utter nonsense
July 23, 2007 at 6:09 am
Oops the last post was from me !!
July 23, 2007 at 8:09 am
Hi krishna – thanks for visiting!
The trouble I think is that as you know swaras are not defined in our Music by fixed, flat tones because of gamakas. So if you take say the rishabam of sAvEri, and compare how it is rendered by 2 musicians, or even the same musician on say few occasions, there will be minor differences in the “pitch contour” (pitch vs time during the duration of intonation of the swara). So the rishabam of sAveri, is not only a fixed/flat tone, it is not even a fixed contour. It is a class of contours that conform to certain parameters. And of course that can change depending on the context – e.g. ga of kanada when coming from pa, takes various sets of contours, but coming from ri would be a different set etc.
So a real reliable swara identification (which is a prerequisite for reliable raga identification) itself seems like a hugely complex task.
But if one were to disregard this, aim lower and ask if we can devise a system that will come up a set of ragas that match a input swara pattern entered in textual form (say SRPMGRGMDN …). I think that could be possible and in itself is an interesting problem.
Although I dont understand it fully, I see what you are implying by obviating need for sruthi – but again all the dynamics involved in cm seems to make me skeptical as to how reliable it would be to be of any use.
Arun
July 31, 2007 at 9:25 pm
[...] post is a contrived attempt at continuing the “theme” of the earlier Blue Oyster Cult post i.e. Carnatic Music concepts in the Western Music [...]
September 5, 2007 at 12:03 pm
Hi Arun,
listened to the BOC clip and was spell bound.. man! i am BOC-ignorant, but that sure sounded similar to dharmavati/hemavati.
fantastic blog you have man! most of though is over the head for me… stumbled onto your blog searching for more info on Vakulabharanam. I am a huge Led Zeppelin fan and have always wondered about their music in terms of how they blend Indian Classical music shades into their hard rock/metal themes.
glad to find yet another person who likes such parallelisms..:)
cheers!
Arun(k): Thanks Arun! Zeppelin has at least Kashmir which has an Indian/Middle-eastern feel. I don’t remember now but perhaps also “No Quarter”?
May 20, 2008 at 7:08 am
Great to see someone going into such details of Carnatic music, without taking an essentially purist or overly technical approach. Your blog looks interesting for sure.
On Dharmavathi, well maybe not an exact match , but check out “Behind my Camel” by The Police..definitely some shades of it there I think.
Also “Voices inside my head” from the same album (Zenyatta Mondatta) sounds to me like PURE Suddha Saveri….I even do mental remixes of the song with Darini telusukonti
May 20, 2008 at 7:46 pm
Thanks sughosh and welcome to my blog!
Thanks for the police references. I listened to Behind my Camel – it does use a subset of dharmavati. From what I can tell, sa-ri-ga-ma being the main pattern and that of course being s r2 g2 m2 – same as many ragas including dharmavati. Then there is then r2 (in higher octave), and then pa I think. I wonder why the guys didnt expand on it. It almost seems like they had a snippet of tune which seemed fine but couldn’t build upon it more
. BTW, the wiki page for it is interesting – how it won the grammy but both Sting and Stewart copeland hated it!!
The other one “Voices inside my head” – yep it indeed does follow the s.saveri *scale* although I wouldn’t go far as qualifing as PURE s.saveri – mainly because I would then expect carnatic specific gamakas for the swaras as appropriate