The ‘Liber scriptus’ rewritten

As we have seen, in May 1875 Verdi replaced the original setting of the ‘Liber Scriptus’ section of the Dies irce - a fugue for chorus and orchestra - with the mezzo-soprano solo performed ever since. Why did he find the original setting wanting? We need not enquire into the defects of this fugue, as Verdi did not try to improve it but simply replaced it with a piece of a very different type. I doubt that his decision was influenced by the few critics who did not find the fugue effective or appropriate, and it is unlikely that he had somehow discovered and been troubled by the resemblance of the fugue subject to that of the second Kyrie of Donizetti's Requiem, premiered and published in 1870. To discover why Verdi decided to replace the piece, we must consider it in its context.

One clue comes from the letter of early April 1874 in which Verdi responds to one of Ricordi's queries about the first shipment of the Requiem score: ‘ … the transition from A to G minor is very ugly, but more in theory than in practice. It would be easy to prepare it with a simple chord, but I prefer it as it is.’ The ‘Mors stupebit’ builds powerful dominant preparation for a cadence in D minor that, in the first version, never happens: the ‘Liber scriptus’ fugue instead races off in G minor, leaving the preceding section hanging on its dominant, and thus in a movement where sections are otherwise cogently linked, makes the ‘Liber scriptus’ seem a new independent piece.