Brian Micklethwait has been quite a fountain of insights into classical music lately. Among other things, he has an absolutely scathing review of Thomas Adès' The Tempest; and unlike me, he's actually heard the thing.

He also has some very interesting speculations about what might happen if the modern practice of enhancing old recordings through computer technology is taken to its extreme:

[O]ne can foresee the day when you will be able to put an ancient recording into a super-computer, and a modern recording made by a similar or even identical group of musicians in the same place (say), and give the computer the general instruction to make that sound like that. This is bound to cause rows, but personally, if I could listen to that old Dvorak symphonies recording, but in up-to-date sound, I'd be very happy.

If and when such modern recreation of ancient recordings becomes possible, it will cause particularly ferocious rows in the world of opera recordings. "Make it sound like she's doing the phrasing and the intonation and the 'interpretation', but have her do it with her voice, and with him conducting." The logical end point would be something like a Wagner Ring Cycle, with a dream caste, cherry picked from the entire back catalogue of all recordings, of everyone, done anywhere. Musical purists will go berserk, but why not?

I have strong doubts whether such procedures would produce quality artistic results, but it would be very interesting to hear the experiment attempted. I think the musical and dramaturgical interactions between singers, conductor and orchestra would be thrown all out of balance, and the result would be almost unimaginably peculiar. Take a moment to think how Act I of Walküre would sound with oh, say, Jessye Norman's phrasing with Gundula Janowitz's timbre, opposite James King, under Furtwängler's baton with Karajan's dynamic balances. I grow operatically seasick just contemplating it. A Siegmund will act and react differently to a Sieglinde who is ardently lovelorn than he would to one who is lividly angry or poignantly broken, and the conductor and orchestra will (or at least should) be similarly influenced. I think a recording which lacked such chemistry and spontaneity among its performers would sound cold at best, perhaps utterly alien.

Micklethwait continues his discussion of historical recordings here, and links to a fascinating article on the subject by Peter Gutmann; if you're sufficiently interested in these matters to have read this far, it's a must read. Gutmann gives a detailed overview of which historical recordings are really worth hearing, the goal being to find the ones which give us the most insight into how classical music was performed in the heart of the romantic era. To take one of many examples:

[Fanny] Davies was one of the last pupils of Clara Schumann, who was not only a famous virtuoso in her own right but also the widow of the great composer. For forty years following Robert's death in 1856, Clara devoted herself to perpetuating her husband's way of performing his music (much of which she had inspired), insisting that her pupils observe her detailed instructions exactly, just as she had absorbed them from Robert in the 1830s. Thus Davies's earnest but relaxed elegance and unusual phrasing are presumably those of the composer himself and transport us back nearly a full century before her records were made.

Gutmann also points us toward the recorded artists who had the most direct connections to Chopin and Liszt, and to the tenor in the first performance of Verdi's Otello. There are lots of fascinating tidbits here:

[Soprano Adelina] Patti's contributions to the recording industry were not wholly aesthetic. The most flaming egos of our age pale in comparison to hers. She commanded up to $5,000 per performance, more than the President of the United States made in a year; when chided about this, she reportedly challenged the President to sing as well as she did! She refused to set foot in a studio and made engineers come to her home; she insisted that her records bear a distinctive pink "Patti" label; and she set their price at 21 shillings. (That's right – about 3 minutes of music for the modern-day equivalent of $50. And we complain about CD prices?)

No wonder classical music is still perceived as elitist.

Gutmann finally argues that the most interesting historical recording out there is one of the violinist Joseph Joachim (friend of Mendelssohn, Schumann, Liszt and Brahms, and by general consensus one of the greatest musicians of his century) playing Brahms' Hungarian Dance #2:

Here is a style of playing absolutely unknown in our time. Every note bursts with passion. Every gesture throbs with meaning.

Joachim doesn't sharpen or flatten certain notes because he can't reach them, but rather to emphasize the force of a melodic progression or to shade the impact of a chord; indeed, his fingering is so fluid that the individual notes of his passagework are barely apparent. His rhythm is so constantly dynamic and alive that it belies the very notion of tempo. And his bowing– the first downbeats slash with splintering force and soon subside into a whisper.

Joachim tears the notes right off the page. After hearing this astounding performance, no classical artist should ever feel embarrassed to play the romantic repertoire with the same unfettered passion as a hard rocker.

Gutmann argues that improvisation and performers' impulses were once as central to western classical music as they are to almost every other musical genre, and that only in the twentieth century have they been forgotten and frowned upon.

I have doubts as to how much I'd enjoy the application of this stance to the nineteenth century repetoire. There are many passages in romantic concertos which I think come off better if the soloist blends into the orchestral textures rather than constantly wringing every possible drop of passion from her instrument. But Gutmann's point about improvisation certainly resonates with my love of the obscure seventeenth century violin repetoire. All performers and most fans of pre-Baroque music are well aware that, given the paucity of interpretive guidelines in such old scores, the performer must be trusted to give shape to and half invent the performance. When I listen to the works of Biber (whom Micklethwait has recently discovered) or Paolo Pandolfi, I have no doubt that what I am hearing are solidified improvisations congealed from the spontaneous experiments of master violinists. This is especially palpable in Pandolfi's music, where not only melodic motives, but also flourishes and ornaments are repeated Baroque fashion in different positions throughout the scale*. Modern composers in search of inspiration and new sounds would do well to heed Gutmann's and Micklethwait's advice, and spend less time pretentiously composing and more time passionately playing. Micklethwait puts it very well:

What is needed in the classical world is not a steady trickle of Fake Great Composers, but a healthy flow of genuine lesser ones (from which posterity can be left to pick the great ones at its leisure), who can make use of all those violin and cello skills by writing entertaining music that will pay the rent.

*Pandolfi's music is so oddly creative that, though all of it which survives dates from 1660, it often reminds me of the Brahms symphonies. Like Brahms, he frequently transforms his primary melodic material into shimmering textures which can sound very orchestral; and both composers' most beautiful passages often develope out of harsh and somewhat exasperating beginnings.