As much as I respect Süssmayr’s completion of Mozart’s requiem, I still think the Osanna fugue is very short (musically speaking). Liturgically speaking, the Sanctus makes perfect sense and might be the perfect example of how church music should correspond with the text of the Catholic liturgy. If the requiem is ever used for liturgical purposes (though it’s unlikely because the Catholic Church has stopped using the Dies Irae as the Sequence since the Second Vatican Council), I would vote on keeping the Osanna that short to avoid delaying Communion.

My musical instincts took over and I thought, after analyzing the Osanna fugue, what a great fugue subject. Too bad Süssmayr was rushing to finish the requiem so Constanze could collect her payment!

There are already many “extensions” to the Osanna fugue by Robert Levin and Duncan Druce, but every time I listen to them, I get a strange feeling of dissatisfaction. Both, to me, seem to stray too far from what may have been Süssmayr’s original intention (and Druce’s cadence is unconvincing). Knowing that Levin’s intention was to create a new edition of the requiem that retained the familiarity of Süssmayr’s version, I wondered why he didn’t keep Süssmayr’s final five bars.

Just for the fun of it, and as a contrapuntal exercise, I decided to try extending the fugue. I keep all of Süssmayr’s material. The entire development section fits between the first 23 bars and the final 5 bars. The cadence is perfectly logical; why did everybody else change it?

First things first. I’ll show you the first 23 measures of Süssmayr’s original, so that my explanations of the development will make more sense.




For those of you familiar with the requiem, it’s obvious that after the 23rd bar comes the cadence. I stopped copying the original here and started wondering how to proceed. My attention turned to the phrase where the soprano has an octave leap in bar 20. Where else did this motif occur? I noticed it also was presented, an octave lower, in the tenor line from measures 12-15:

The three notes A, B, and C# in the soprano at bar 23 was meant to lead into the cadence starting on D, since C# is the leading tone here. I studied the tenor line for a while and then I knew exactly how to proceed. Why not have the sopranos continue, at measure 24, with what the tenors are singing at measure 16?

The phrase the altos were singing at from measures 15-19 was sung by the bass at measures 23-27, which I added. Since the tenors come in with the subject on the second beat of measure 21 but don’t finish it because of the cadence, I thought, just let them finish. Sure enough, this also corresponds to first entrance of the soprano voice on the second beat of measure 13.

So here is the beginning of my continuation:


(Note: the notes that are yellow or red are nothing to worry about; the program is just trying to say that these notes are around the upper limits of the ranges.)

I made use of the motif introduced in the bass at measure 22 in Süssmayr’s original in both the bass and soprano lines here. This motif is later seen in F# minor here:


As you can see, as short as Süssmayr’s Osanna fugue (or rather, fughetta, as I would say) may be, it has some very promising material that can be developed.

And, of course, I kept his original cadence:


I could go on and on, but I think I should end here with the YouTube video of the fugue that I made:

I know the fugue isn’t perfect, and that artistic opinions are subjective, but I tried my best as a self-taught thirteen-year-old composer!

Update: By the way, I also did a completion of Mozart’s Amen fugue sketch:


One thought on “My Extension of the Osanna Fugue

  1. Levin actually did keep the original cadence at the end of his fugue – except it’s kind of hidden, and it comes before the final cadence. Not too sure about Druce and the others though.

    By the way, there is a logical explanation as to why the fugue is so short – as you can see, in the beginning of the extension, the tenors are singing the subject, the sopranos are singing the tenors’ countersubject, and the bass is singing a countersubject that previously appeared in the alto voice (see explanations above). What we end up with, as a friend pointed out, is a case of parallel 6/4 chords (which is not good because 6/4 chords are technically considered “dissonant” – it’s a topic of debate, but this is another story).

    As he wrote: “You did a very creditable job in your creative efforts with the Requiem. A few slips of grammar, at least ‘historically speaking.’ For example, in bars 23 /24 of your ‘Sanctus’ rewrite, you have some descending parallel 6/4 chords. While there’s no inherent problem with this—-as many composers over the last 100+ years have shown—it does create a sound that was not accepted as ‘proper’ in the late 18th century. (On the other hand, chains of parallel 1st inversion triads certainly were).”

    So here’s the problem: Sussmayr himself wrote parallel first inversion triads!!! So when I “inverted” the voices (you know – invertible counterpoint is just so much fun!), the 6/4 chords appeared. That’s probably the reason why Sussmayr stopped continuing the fugue, and why so many modern completers change the countersubject. Well, and yes – he was probably out of time, too, or he probably would have gone and re-thought his idea. Not sure why he didn’t “foresee” the problem – kind of like avoiding parallel fourths with invertible counterpoint to avoid parallel fifths – but that is another story.

    But come on! Give Sussmayr a break!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s