The classical music world is in an uproar over the discovery that most or all of the recordings by acclaimed pianist Joyce Hatto are copies of recordings by other artists, either identical or digitally altered. Her career was fueled in recent years by Internet hype, and technology led to the discovery of the hoax. This is fun stuff.
Hatto’s career began in the 70s with a few live performances (relatively uninspired, according to one account), followed by a stream of more than a hundred CDs with performances of some of the most difficult classical pieces by a wide variety of composers – complete cycles of Beethoven, Mozart and Prokofiev sonatas, the concertos of Rachmaninov, Tchaikovsky, Brahms and Mendelssohn, most of the works of Chopin, and more. In the last few years mainstream classical magazines began reviewing the discs, giving them increasingly favorable reviews, until Hatto became a cause celebre for a year or more before her death in June 2006.
According to one Slashdot post, interest in Hatto was driven by posts in the rec.music.classical Internet newsgroup. Two people posted frequently about Hatto and flamed anyone critical or suspicious of her recordings. In hindsight, the favorable posts look like the work of shills.
A few days ago, a reviewer for Gramophone magazine inserted a Hatto CD into his computer. iTunes consulted the online CDDB database to identify the disc and inserted the name of a earlier recording by a different pianist on another label. CDDB (used by iTunes and other music software) is a database of CDs that primarily relies on the length of tracks to identify a disc; that doesn’t seem like it would be enough information to work but it’s surprisingly accurate.
This piqued the curiosity of the reviewer, who had access to the other recording and discovered that it sounded identical when he compared it to the Hatto recording. He tried another Hatto disc and had the same experience – a completely different pianist identified as the artist, an identical-sounding performance when he dug out the older recording. Gramophone hired a lab to do audio analyses of Hatto recordings and possible sources for those recordings – and so far, every single recording has been determined to be a copy of another recording. Some of them are simply identical, leading to the CDDB mixups; others are altered – slowed down or equalized or panned slightly. One amusing fact that’s come to light: the conductor and orchestra credited on some of the Hatto recordings are apparently completely fictitious.
Wikipedia is one of the clearing houses for information right now, as many people are poring over the Hatto recordings looking for more evidence of the hoax. More audio labs will be doing more analyses. The news media is jumping in. Hatto’s husband is professing to be shocked that anyone would be suspicious. (He is listed as the “producer” of virtually all the recordings, and runs the little mail order company pressing and distributing them.) A very audacious hoax, brought down by inserting a CD into the computer and idly wondering why it wasn’t identified correctly. Great stuff!