STORY LINKS:
Hatto Hoax evidence - Part 1: Liszt (Simon, Nojima) |
|
Hatto Hoax evidence - Part 2: Godowsky (Grante, Hamelin) |
HERE |
Hatto Hoax evidence - Part 3: Rachmaninov (Bronfman) |
|
Hatto Hoax evidence - Part 4: Chopin (Indjic) |
|
Hatto Hoax evidence - Part 5: Brahms (Ashkenazy) |
|
Hatto Hoax evidence - Part 6: Latest claims of matches |
|
Hatto Hoax questions - Part 7: A case for Hatto? |
|
Part 8: How it's done: A fake we prepared for this site |
Joyce Hatto - The Ultimate Recording Hoax - Part 2 |
This CD is not what it seems:
It's been heavily doctored to disguise at least three sources
Example 3: Joyce Hatto plays Godowsky - or rather, no she doesn't, Carlo Grante does, as does Marc-André Hamelin and Ian Hobson .
Manipulating Time |
Using off-the-shelf audio editing software it's easy to manipulate time. Here we present one of our own recordings, from Edwin Fischer's classic reading of Bach's Well Tempered Clavier, in both its original state, and digitally manipulated to make it exactly 15.112% shorter and faster: The original (5'38") Time-shrunk 15.112% (4'54") |
15.2.07: A much trickier one for me to unravel, as it's a much harder recording to disguise - there simply aren't many of them around, and so there are far fewer candidates to choose from.
I was immediately suspicious about the sound of the second track on the 'Hatto' CD - it didn't sound like a straight CD copy; rather it had some of the smearing and blurring you'd expect from a poor MP3.
In fact, the reason for this is that the original, by Carlo Grante, had been subjected to more than 15% of time stretching manipulation - way over the 3-4% that most software is designed to achieve without audible flaws. When we compress the original Grante to match the already time-compressed Hatto, once again a perfect match is found.
To find fakery on one CD seems evidence enough. To find it so quickly and easily on a second, picked at random from the Hatto catalogue, seems to offer the final proof of what may turn out to be the greatest recording hoax ever perpetrated on the music-loving public
22.2.07: The Plot Thickens as another source is confirmed.
After proving the Grante connection with Track 2 of the Godowsky Studies, I was approached by Hyperion Records in the UK with the suspicion that the third track of the Hatto CD was in fact taken from their own multiple-award-winning 1998/9 recording by Marc-André Hamelin. They sent me an uncompressed file of the Hamelin to compare with the 'Hatto' and asked for my conclusions.
Initial findings indicate a number of minor changes - the left and right channels appear to have been reversed, there's been some re-equalising, with minor boosts in bass and mid-range and a slight cut in the treble, and once again it's been speeded up, though this time by a 'mere' 2.364%
This information has been passed back to Hyperion Records, and with their permission we have reproduced the entire recording below, once again with the source material speeded up to match the 'Hatto' and placed in the right speaker channel, alongside the 'Hatto' in the left channel.
24.2.07: A tip-off on rec.music.classical.recordings yields the third source .
On Saturday morning 'Kevin' noted on the Usenet newsgroup, referring to the first study: "Ian Hobson on Arabesque Z6537. I don't know about waveforms but "her" fingerslips on measure 42 (1:06) is identical to Hobson's (1:08)". He's absolutely correct - Hobson had been speeded up about 7.5% to become 'Hatto'
26.2.07: Intensive audio detective work unmasks a complex deception on Track 4.
I'm grateful here to Peter Harrison at disk2disc - a regular colleague of mine and good friend - for spending the time needed to crack a major mystery: the fourth track on CD1. Pianist Carlo Grante was convinced it was taken from his Altarus CD, but I simply couldn't get the two recordings to match in the way the others had.
What Peter discovered, however, was that this recording had undergone multiple time tampering in a manner not yet detected on any other 'Hatto' recording, and did indeed come from Grante's Altarus CD. In an e-mail to Carlo he wrote the following, which I've tested independently and confirm to be the case:
"The 'Hatto' Godowsky Etude No 4. is definitely from your Altarus CD. What 'someone' did to it was: (1) Take the first 1'13" and speed it up by 114.3%; (2) take the remainder except for the last chord and speed it up by 127.55% (!!!); (3) take the last chord and slow it down by 50% - that is, make it last twice as long on the fake as on the original (!!!!); (4) as usual add reverb, fiddle with levels, and so on. The result, as you might expect, is a grotesque parody of your original recording."
The implications of this discovery are alarming - it is much, much harder to detect this degree of time doctoring. If this is repeated throughout the 'Hatto' CDs there may be a number of recordings which go undetected as a result.
26.2.07: Final piece of the puzzle falls into place on CD1 - Track 14 is also by Ian Hobson
I'm grateful to Farhan Malik, who e-mailed me with the results of his own findings, including the Hobson tracks and the Grante Track 4 - the latter a conclusion more easily reached by ear, as the pianist himself had done, than confirmed in the studio.
In the case of the Hobson which found itself onto Hatto Track 14, I admit I'd missed his recording in the rather poorly annotated track listing I'd found on emusic.com, from where I'd sourced his Arabesque CD recordings - Hobson recorded a selection, presented out of order, rather than the whole set. Once I had located his recording and lined it up, I quickly confirmed that, with an increase in speed of just over 8%, this was the one.
Track 2: 'No 2 Op.10 No.1 D Flat Major-Left Hand' - DOWNLOAD MP3
|
Track 3: 'No 3 Op 10 No 2 (1st version), A minor, left hand' - DOWNLOAD MP3
|
Track 1: 'No 1 Op 10 No 1 (1st version), C major' - DOWNLOAD MP3
|
"Hatto" Godowsky Chopin Studies - sources for CD1:
1 - Hobson, Arabesque CD
2 - Grante, Altarus CD
3 - Hamelin, Hyperion CD
4 - Grante, Altarus CD
5 - Grante, Altarus CD
6 - Grante, Altarus CD
7 - Grante, Altarus CD
8 - Grante, Altarus CD
9 - Grante, Altarus CD
10 - Grante, Altarus CD
11 - Grante, Altarus CD
12 - Grante, Altarus CD
13 - Grante, Altarus CD
14 - Hobson, Arabesque CD
15 - Hamelin, Hyperion CD
16 - Grante, Altarus CD
17 - Hamelin, Hyperion CD
18 - Grante, Altarus CD
19 - Hamelin, Hyperion CD
20 - Hamelin, Hyperion CD
21 - Hamelin, Hyperion CD
22 - Grante, Altarus CD
23 - Grante, Altarus CD
24 - Hamelin, Hyperion CD
25 - Hamelin, Hyperion CD
26 - Hamelin, Hyperion CD
27 - Hamelin, Hyperion CD
As expected, all sorts of tricks have been used to disguise origins, including a variety
of tempo changes, stereo channel swapping, EQ and so forth.
"Hatto" Godowsky Chopin Studies - sources for CD2:
Marc-André Hamelin's Hyperion recording used for tracks 1, 5, 6, 12, 16, 19, 22
The Grante CD used to fake Hatto:
The Hamelin CD used to fake Hatto:
The Hobson CD used to fake Hatto:
STORY LINKS:
Hatto Hoax evidence - Part 1: Liszt (Simon, Nojima) |
|
Hatto Hoax evidence - Part 2: Godowsky (Grante, Hamelin) |
HERE |
Hatto Hoax evidence - Part 3: Rachmaninov (Bronfman) |
|
Hatto Hoax evidence - Part 4: Chopin (Indjic) |
|
Hatto Hoax evidence - Part 5: Brahms (Ashkenazy) |
|
Hatto Hoax evidence - Part 6: Latest claims of matches |
|
Hatto Hoax questions - Part 7: A case for Hatto? |
|
Part 8: How it's done: A fake we prepared for this site |
We have yet to investigate a Hatto recording that has not proved to be a hoax.
|
*Sleevenotes included wherever present in original CD booklet. We provide sleevenotes either direct from record companies' digital source material or from our own scans. In the latter case we normally scan English language content and libretti only. Scanned notes pass through OCR software and can be treated as text files - all are offered as Adobe Acrobat PDFs.
Pristine Classical - bringing you DRM-free classical MP3 downloads
