Tritone Substitutions.

Ok, here's the deal on tritone subs...

Take any 7th chord. I'll use C7 for example: C-E-G-Bb. This chord contains a tritone - the distance from E to Bb is three whole steps... which is a dissonant interval (a b5)... which gives the chord its tension.

A tritone happens to be exactly half an octave.

And that means the exact same interval will occur in some other 7th chord. Where C7 had E as the 3rd, some other chord will have it (or the enharmonic equivalent) as the 7th, and that same chord will have Bb as the third.

So... if Bb is the third, what's the root? It's two steps lower - which is Gb. And Gb7 is Gb-Bb-Db-Fb (enharmonic to E). If you prefer, the whole thing can also be written as F#-A#-C#-E.

It's a tritone substitution only when the tritone - the tension in the chord - remains exactly the same. So maybe a better name for it is the "all the notes except the tritone" substitution, since the other ones are the notes that change :)

I'm still not sure if that means that if I changed the F7sus4 in the chord progression above to a B7 the B7 would be considered a tritone substitution or not. Logic tells me no, because surely then any 7th chord would be a tritone substitution because they all have the tritone in them and thus have a 'tritone partner'. And the F7sus4 seems to function in terms of resolving and being resolved to much better than a B7 would there - so I'm not sure if you'd call the F7 a tritone substitution, but not call the B7 (if in F7's place) a tritone substitution also. :?

Or maybe I'm just being a plank and missing something really obvious.

The ugly timbred cow has been kicked. Hard. But not slaughtered just yet, I'm afraid.

The thing is - F7sus4 doesn't contain a tritone. :?

The other thing is that, yes, any 7th chord can be part of a tritone substitution but ONLY if you actually substitute it. If you take the dominant 7th of C, which is G7, and replace it with Db7 (or C#7) then you've made a tritone substitution. G7 contains the tritone B-F and Db7 contains the tritone Cb-F, which sounds exactly the same.

The neat thing about it is that in order to substitute the 7th chord which contains exactly the same pitched tritone, you have to use the 7th chord which is a tritone away from the original chord. So there are two tritones at work here. The tritone within both chords and the tritone between the roots of the two chords - The root of Db7 (or C#7) is a tritone away from the root of G7 - clever huh?

So in your example, F7 sus4 is indeed a tritone away from B7, but as F7sus 4 doesn't contain a tritone, I can't see how substituting B7 would qualify as a true tritone substitution.

The other thing is that, yes, any 7th chord can be part of a tritone substitution but ONLY if you actually substitute it. If you take the dominant 7th of C, which is G7, and replace it with Db7 (or C#7) then you've made a tritone substitution. G7 contains the tritone B-F and Db7 contains the tritone Cb-F, which sounds exactly the same.
.

So, if you're playing in C, and you see a Db7 you may say it's a tritone substitution for G7, but if in C and you see a G7 it would be really stupid to say that the G7 is a tritone substitution for a Db7? I thought so, but that example had me really confused - the F7 seemed more appropriate in the context than the B7 it was supposed to have substituted for. I'm thinking the example/explanation of it might have included errors.

And presumably 7th chords with other extensions could also be called tritone substitutions, but only provided they substitute for a chord a tritone away, and they feature a tritone interval within them? Does the chord they substitute for have to have a tritone interval within it also? Surely subbing a Db7 for a G chord when playing in the key of C would also yield a tritone substitution?

Thanks Fretsource, I'm totally clear on tritone substitutions now! Consider the cow sacrificed. :wink:

[Surely subbing a Db7 for a G chord when playing in the key of C would also yield a tritone substitution?

Not really. Yes, you would have substituted the G for a chord (Db7) which is a tritone away, but it wouldn't work. The idea of tritone substitution is to take advantage of the same tritone existing within 7th (or extended) chords that happen to be a tritone apart. The Db7 can be used because it allows you to keep the same tension that G7 has as it contains the same tritone, even though it's spelled differently. It means you're getting exactly the tritone that's required of the dominant 7th chord (G B D F), but dressed in a whole new set of clothes (Db F Ab Cb).

Oh yes, good point. Sorry, I had an idiot moment there. Thanks!

"The IV7 chord (F7) is in fact a substitute for the B7 chord and could resolve to the Eminor (iii) chord also. This type of substitution is known as a tritone substitution."

I find this odd for several reasons. The progression has every appearance of being in the key of Bb (it starts on a Bbmaj7 and resolves to it in a V-I cadence which is just about the strongest way of establishing a key) and yet he refers to the F7sus4 as the "IV7" chord when it's in fact the V7. It would be the IV chord in the key Ab. He then says the F7 is a sub for "the B7" chord which could resolve to E-. This would be fine if we were planning on going to E- but there is no indication of this (also, the chord would typically be a 7 not a 7sus4). In fact the only chord here relevant to E- without a large stretch of the imagination is D7 and it leads into G-7. He also labels the E- as iii, which it would be in the key of C, which is, again, nowhere to be found. The text sounds like it's referring to some other chords entirely.