This article is taken from the May 2022 issue of The Critic. To get the full magazine why not subscribe? Right now we’re offering five issues for just £10.

I first heard of the indivisible connection between Bach’s Goldberg Variations and Glenn Gould when I was around 16. It was that time in high school when people began to separate themselves by tribe: the sporty types (constantly eating), the theatre crowd (constantly screaming), the artists (constantly painting each fingernail a different colour), and so on. A member of the tribe that skipped class to read Patti Smith and Sylvia Plath over cups of burnt coffee scoffed at my copy of Trevor Pinnock’s recording of the variations and pronounced that the right recording was by someone named Glenn Gould. How could I not know this? 

Maybe they were too light, too pleasant

There have been many reprises of this scenario. I have no recollection in this period of ever sitting down to closely listen to Gould’s recordings of he Goldberg Variations of 1955 or 1981. It was not that I had anything against him but in that period I was drawn to other influences when it came to Bach. In my twenties I was busy practising and playing catch-up as a musicologist who decided a bit late to actually play music. But when it was time to perform and record the variations, I came to learn of the dangers of failing to show sufficient enthusiasm for Gould.

This musical mark of Cain manifested itself in different ways. During post-recital Q&As, there is always an audience member who asks about Gould. I usually point out tactfully that the work predates the life of the Canadian pianist (1932-82) and move on. 

Now and then a review of my disc would mention the association of the Goldbergs with Gould. A few would note I had not mentioned Gould either in my liner notes or in interviews. Other experiences were less amusing. During an onstage talk with a journalist in Vancouver, I admitted to not being terribly familiar with either of his recordings. The audience gasped, people walked out, and I proceeded to deliver one of my more mediocre performances. 

Afterwards, I sat in silence with the hall programmer. He later wrote my then-agent a complaint about how insulted audience members felt about my “transgression”. For years afterwards that Vancouver journalist made considerable efforts in articles and in social media to remark on how my arrogance was but one of my many faults as a second-rate musician. 

After this, I figured it was time to take apart the public perception of Bach’s work into “Before Gould” and “After Gould” (B.G. and A.G.) phases, and to examine the myths and traditions surrounding the Goldberg Variations. One would do well to ask how Gould’s association might be a result of various historical and social factors. What did the Goldberg Variations mean to the public before 1955, the year of Gould’s first studio recording? 

They occupied a minuscule space in the nineteenth century’s engagement with Bach. This is not because they were unknown. At a time when Bach was considered secondary to his sons, his first biographer Forkel (1802) referred to the “air with several variations … with the most easy and flowing melody”. Forkel is also the source for the story that led to their moniker — the commission from an insomniac count who commanded performances of Bach’s variations from his court harpsichordist, a young Mr Goldberg. But they failed to capture the Romantic imagination, which prized Bach’s keyboard music chiefly for its perceived virtues of edification and technical improvement. Maybe they were too light, too pleasant. 

Busoni dared to treat Bach on similar terms as himself

The few mentions of them in this period were related to their unsuitability for piano, as the score specifies a harpsichord with two keyboards and thus requires two independent sets of strings at the same pitch. Apart from mysterious references to excerpts in the repertoire of virtuosi such as Franz Liszt and Ignaz Moscheles, a few of the variations were trotted out by musical antiquarians in that most Victorian of institutions, the public lecture-recital. 

The task of reviving the Goldberg Variations in the performing repertoire was left to a phenomenon worthy of such a responsibility: the composer and pianist Ferruccio Busoni, who was once as readily associated with Bach’s keyboard music as Gould is in our age. Busoni’s 1915 edition of the Goldberg Variations lies somewhere between the philological discipline of the editor, the creative art of the transcriber, the erudition of the musicologist, and the transcendent insight of the composer. Busoni dared to treat Bach on similar terms as himself, approaching him with a creative license that never transgressed the boundaries of respect. His edition is a work of art in its own right. 

Busoni’s transcription for piano was based on a keen understanding of the harpsichord. He owned one of the earliest modern copies of a historical harpsichord, a double-manual concert instrument with four registers built by the early music guru Arnold Dolmetsch. His approach to Bach was shaped by the attitudes of the early stages of the Baroque music revival. It was thought that the harpsichordist made terraced dynamics through the combination, addition, and subtraction of various sets of strings, some at different octaves, as on an organ. 

In order to hear Bach’s music, the score should undergo surgery so that the music could be heard “authentically.” Thus he only hardened the narrative that the score of the Goldberg Variations was best suited to the harpsichord. 

Enter harpsichordist Wanda Landowska (1879-1959), the first modern concert performer on the instrument. The oversized influence of Landowska (above, right) on the prewar public imagination of Bach’s music cannot be overstated, and it was she who established the Goldberg Variations as a universally recognised work. In a 1981 interview, Gould even admitted that in his youth the work was seen as being “the private preserve, perhaps, of Wanda Landowska”.

Perhaps due to Landowska, recordings of the Goldberg Variations were virtually non-existent in this period, with a piano roll for Welte-Mignon by a 25-year-old Rudolf Serkin being the one interpretation in the catalogue to predate Landowska’s 1933 recording for French HMV. A 1942 recording by pianist Claudio Arrau was shelved by RCA — the label had already recorded Landowska’s interpretation (her second one) in 1945 in an originally limited release for subscribers which became an unexpected overnight sensation. 

Though it is difficult to believe now, the popularity of Landowska’s recording and the association of the Goldberg Variations with the harpsichord were sufficiently strong to dissuade RCA’s executives from releasing Arrau’s version. Arrau, who had made his European reputation in the 1930s mainly as a Bach interpreter, was so in awe of Landowska that he decided “that I should be a pianist [and] that I should only play things that were conceived for the piano and piano sound”. His Goldbergs remained in the RCA vaults until 1988. 

In 1951, the 40-year old Ralph Kirkpatrick (the survivor of eight uncomfortable months as Landowska’s student), made the first of his two recordings, using the 1908 harpsichord once owned by Busoni. The result is Kirkpatrick at the height of his powers — unsentimental, uncompromising, at once dry and quietly poignant. His recordings marked a watershed in the revival of the harpsichord in the performer’s rejection of the rabbit holes of the terraced dynamics school in favour of a language of articulation and timing closer to that which Bach would have known. The idiosyncrasies of Landowska and the dour amateurishness of the early music movement were discarded in favour of a sober stance that allowed the immanent qualities of the music to shine through. 

Kirkpatrick’s subtleties did not find a wide audience in a time stuck between the aging ideal of the Romantic virtuoso and the anti-virtuoso mantle taken up a few years later by Gould. The critic Harold Schonberg, in thrall to the variegated colours and mystical persona wrapped up in Landowska’s playing, wrote with some disappointment of Kirkpatrick’s “effort to avoid sensationalism or anything that might be construed as an unstylistic romanticism”. 

The American public was more taken with pianist Rosalyn Tureck, who first recorded the Goldbergs in 1947. A Chicagoan who acquired an English accent so plummy it made Margot Asquith sound like Dot Cotton, Tureck was a progenitor of the Bach interpreter-as-guru model which dominates attitudes today. 

An erstwhile harpsichordist who made her debut as a performer on the theremin, Tureck is the source for the glib notion of Bach’s music as existing on a plane beyond questions of medium; in an essay from the 1970s, she wrote “Bach music is fundamentally abstract: not abstract in the theoretical sense, but in that its existence and endurance [are] based on its fundamental musical terms, rather than that of a confined period or instrumental style.”

Even before she and her acolytes came to believe in her own legend, Tureck’s playing, at least on the 1947 Goldbergs, is defined by a degree of calculation and premeditation of each and every gesture. In other words, it is arch. 

By taking cursory stock of the era “Before Gould”, we can see that, as with Busoni before him, Gould’s historical position vis-à-vis the Goldberg Variations is as much a result of the stultification of the public discourse around the work that preceded him as it is to do with his own epoch-making artistry. 

The decision of a gifted, eccentric 22-year-old Canadian pianist to make his debut with a work associated with a somewhat uneven ride in the modern era was a brilliant stroke; it is no disrespect to Gould to see this as an intelligent calculation.