Notes from an eyewitness

Theses, monographs and books have been written about orchestral and instrumental performance practice in the Baroque, Classical, and even the Romantic eras.  Indeed, the entire authentic performance and authentic instrument movement, the ‘period performance’ movement, is an attempt to recreate performances of the past.  This movement exploded into academic and musical popularity in the ’60s and early ’70s, and resulted in a great deal of textual research and, on the whole, progress in the authentic recreation of historical instruments.

The recreation of performance practice, however, was disappointing.  Thousands of recordings were made by ‘academically informed’ conductors and soloists.  Too many were thin, metronomic, and dry as dust.  After three decades a reaction set in.  The younger generation pointed out that few historical documents existed on performance practice and those that did, including autograph scores, were usually ambiguous.

Recent research has turned to early recordings, on the theory that performance practice from the first decades of the last century must have reflected 19th century practice . . . which in turn might have included some survivals of 18th century practice.  Unfortunately, electrical recording did not come in until 1926, and acoustical records from the previous three decades are dim and distorted.  Nevertheless, the early electricals were examined.  The latest scholarly opinion suggests that late 19th century and early 20th century performance practice was probably flexible in tempo and interpretatively free by modern standards.

But in an effort to find out exactly how something was generally played, say, during the classical period, might we be chasing a chimera?  Isn’t it possible that there existed a wide variety of performance practice in every period?

Elizabeth Mittler-Laudy is 101 years old and living independently in Toronto.  In the ’30s she was a professional violinist in Holland.  In 1940 in Banyuls-sur-Mer, while in flight from the Nazis, she performed publicly with Wanda Landowska.  I recently asked her to describe performance practice during that period, and particularly her experience performing with Landowska.  Here is her reply:

Concerning (her) tempo there were slight variations within the flow of the music.  It was just her deep feeling for the score that told her when to apply this and how much.  At the time it seemed completely natural to me and I took it as just the way Bach should be
played.

In Holland I had played in a special Bach orchestra with a conductor who believed in a strict, almost metronomical beat, hardly slowing down at the last bars.  To me it was very unsatisfying.  I played a number of St Matthew Passions under different conductors and I found that the tempos and the general approach was a bit different with each one.  It is difficult to point to a general accepted style used in the thirties.  I believe that it mostly depended on the conductor or the soloist.

While thinking about Wanda Landowska it came to me that I once differed with her about the last bars in Bach’s Concerto for two Violins.  She wanted them to be drawn out considerably.  The word she used was “majestueusement”.  I thought it was just a bit too
much, but who was I to question her judgment?

Landowska continued to perform for another 18 years on both the harpsichord and the piano.  Her early recordings on the harpsichord can show great liberty in tempo and ornament; her late recording, below, of Haydn’s F Minor Variations, done in her own home in Lakeville, Connecticut, on her own Steinway, in 1957, shows by comparison a classical restraint.  When playing Mozart and Haydn, she attempted to reproduce on the modern piano the sonority and dynamic range available to the fortepiano.  I think she did a pretty good job.

Note: this entry on performance practice was inspired by reading The Krupp Secret, a privately printed memoir by Elizabeth Mittler-Laudy.

of her recording of the Goldberg Variations

Landowska’s first recording of Bach’s Goldberg Variations, which was not only the first recorded on a harpsichord but the very first recording of the piece ever made, is variously listed as made in 1931, 1933 and 1935; I’m going with the earlier date.  I first listened to it on an LP about 1980.  It made an indelible impression.  She was trained in the classical and romantic repertory, and I’ve read that she played Chopin on the piano all her life, but clearly her heart–or at least a significant part of her heart–was with Bach.  She was also a serious musicologist and researcher with the interest and the languages and the cultural background to do original research, and the luck to be active at a time when you could still collect original manuscripts and instruments–before the the looting of Leipzig, the fire bombing of Bremen and the destruction of Berlin and much of western Europe.

Many reviewers describe her Bach as romantic, at least one as Gothic.  For me, her Bach lives.  There is something amateurish about many of the ‘informed’, ‘historical’, performances of the past forty years.  They sound like an academic’s attempt to recreate something long dead.  They are marked by a musical version of textual criticism, a scholarly activity well-suited to the Bible and the Koran and other literary texts, but ill-suited to musical performance practice.  An Urtext is the starting point, not the end point of a performance.  Landowska’s performances on the other hand have real blood flowing in them: the red blood of an informed, authentic, performance artist.

It is true that the recorded sound of her ‘revival’ Pleyel harpsichord is surprising to the modern ear.  But then, so is the sound of an authentically constructed modern clavichord.  Most musical sounds from the past sound strange at first to us today; we have to learn them anew, as we learn a foreign language.  (This is true even of ‘contemporary’ music: try to keep from cringing upon first hearing Ginger Rogers sing ‘In the Money’, in ‘Gold Diggers of 1933′.)  And, strangely, the latest scholarship is catching up with Landowska: her instrument’s 16 foot register, so denigrated as ‘unauthentic’ for four decades, is now admitted to have been more common in the 18th century than previously thought, and in fact to have been used by Bach; it is now being provided to new, large harpsichords.

How, financially, was the Goldberg even recorded on Landowska’s Pleyel harpsichord, in Paris, at the height of the worldwide depression in 1931?  It’s an interesting story.  The record companies were suffering like everyone else.  Walter Legge, at HMV, came up with the idea of collecting subscriptions for important recordings, and then actually making them once the amount collected made the project feasible and profitable for the company.  His idea was a success, and many of the records made endure as classics today.

Below is the first recording ever made of the 25th variation of Bach’s Goldberg Variations.