Monday, September 24, 2007

Dead Musicians, New Performances

Zenph Studios is a North Carolina company that specializes in a very special area: what it calls re-performances of previously recorded performances. Its first product, a re-performance of Glen Gould’s famous 1955 Bach Goldberg Variations, is available here. I have had a number of versions of this famous performance: the LP; the original CD release; and, the recent triple CD set which couples the 1981 recording of the variations along with out-takes and commentary by Gould, available here. I highly recommend the triple set for Gould fans. His explanation for the tempi is insightful, and his off-beat humor is much in evidence (for good or ill).

The reviews of the Zenph recording of the 1955 performance on are decidedly mixed and I leave it others to form their own opinion after listening to it. One great advantage of the Zenph recording (aside from not having any of Mr. Gould's humming along) is that (unlike the other Gould releases), it is on the Hybrid SACD format, an amazing format which the record labels strangled on birth (along with DVD Audio).

Zenph, through its software attempts to reproduce the experience of hearing the original recording live. In the case of the Gould Goldberg variations, we are talking about a studio recording in monaural. The advantage then of the Zenph recording is that it attempts to expand the ambience of the sound experience. Some reviewers have applauded Zenph’s effort in this respect, while others deride it as artificial and note the differences between the piano used by Gould and by the Zenph engineers (e.g., a Yamaha Disklavier Pro grand). My own view is that a thousand flowers should bloom: the more the better, with everyone free to choose whatever version(s) they prefer.

Zenph made headlines in at Shrine Auditorium Los Angeles last night with a live recording of its re-performance of late jazz icon Art Tatum’s 1949 epic album "Piano Starts Here." Here is a link to Zenph’s description of the event. Zenph notes: “Jazz pianist Art Tatum was recorded live in Los Angeles on April 2, 1949. That recording is available today on the Sony CD Piano Starts Here. “He was the greatest soloist in jazz history, regardless of instrument,” jazz critic Leonard Feather wrote in the liner notes. It is one of the best loved jazz albums ever, and has been in print for more than fifty years”.

I am among the legions of people who are in awe of Tatum and I possess quite a number of the hissy, compressed Brunswick recordings for the sheer brilliance of Tatum’s playing buried somewhere in the recordings. If Zenph can make Tatum come alive and without the almost fatal defects in the originals, I’ll get down on my knees and praise them. The Tatum event led to an editorial in the Los Angeles Times Sunday by the always excellent Jon Healey, available here. (HT to Mr. Healey for the story, too). Mr. Healey observes:

Zenph, which also revived pianist Glenn Gould's 1955 recording of Bach's Goldberg Variations, has ambitions that stretch beyond re-performed piano works. The company's co-founder, John Q. Walker, says the technology will eventually be extended to other instruments -- and voices. And once enough data are gathered about a performer, living or dead, that person's distinctive playing or singing style could be applied to material he or she never recorded. Imagine adding Eddie van Halen's guitar pyrotechnics to your band's sound, without needing him in the studio. Or paying him. This is heady stuff, and frankly a little creepy. It also suggests a battle to come over who, if anyone, owns a playing style. Lawmakers didn't anticipate technologies such as Zenph's when they wrote the statutes governing copyrights and other intellectual property, so it's not clear how the courts might rule. Nor do we know how these technologies will develop and be used. The only sure thing is that because of Zenph, we have a lot more useful knowledge about how Art Tatum played the piano -- knowledge that could conceivably lead to a panoply of new creative works. As a result, Tatum's signature cascades of sound will reverberate out of the Shrine Auditorium long after his computerized self has left the building.
From a copyright perspective, there are a few points to keep clear. The Gould and Tatum re-performances are being done in conjunction with the labels and (at least in the case of Gould), the artists’ estates. They also involve recapturing the original sounds. The example of Eddie van Halen given by Mr. Healey, by contrast, is both hypothetical and would not involve recapturing the original sounds: instead, it would reflect an educated guess about how Mr. van Halen would perform something he had never performed. In such a case, there are no copyright implications at all. Section 114(b) of the Copyright Act states in relevant part for both the Gould/Tatum and van Halen scenarios:

The exclusive right of the owner of copyright in a sound recording under clause (1) of section 106 is limited to the right to duplicate the sound recording in the form of phonorecords or copies that directly or indirectly recapture the actual sounds fixed in the recording. … The exclusive rights of the owner of copyright in a sound recording under clauses (1) and (2) of section 106 do not extend to the making or duplication of another sound recording that consists entirely of an independent fixation of other sounds, even though such sounds imitate or simulate those in the copyrighted sound recording.

Under the final sentence, countless covers of performances have been permitted (such as Smith’s version of the Band’s “The Weight” in the original release of the soundtrack to Easy Rider, after The Band refused to grant permission to use their original. In the van Halen hypothetical there would be no original performance at all, and thus no infringement of a sound recording. Style has never been protected under copyright, and in the case of popular music, insuperable problems would be encountered. As wikipedia's entry on Tatum notes some of the dangers:

Tatum drew inspiration from his contemporaries James P. Johnson and Fats Waller, who exemplified the best stride piano style. Tatum's meteoric rise to the top began with his appearance at a cutting contest in 1933 that included Waller and others. Standard contest pieces included Johnson's "Harlem Strut" and "Carolina Shout," and Fats Waller's "Handful of Keys." Tatum was victorious, presenting his arrangement of "Tiger Rag." This was considered by Harlem musicians to be Tatum's ultimate contribution to stride piano, and taken as the most astonishing and original that would probably ever appear, in many respects, despite being an arrangement. … From the foundation of stride, Tatum made a quantum leap in terms of technique and theory, and honed a new style that would greatly influence later jazz pianists, such as Thelonious Monk, Oscar Peterson, Billy Taylor, Bill Evans, and Chick Corea. Tatum's extensive use of the pentatonic scale, for example, may have inspired later pianists to further mine its possibilities as a device for soloing. Moreover, as long as "new" performances via Zenph in the style of artists who never performed the pieces is marketed accurately, it is doubtful any state laws, such as right of publicity, would be violated, although I confess readily to not being an expert in state laws.


Christopher Fulmer said...

So, are you suggesting that the Zanph recording is an example of the first part of 114(b) (an "indirect recapture of the actual sounds fixed in the recording") or the second part (the "making . . . of another sound recording that consists entirely of an independent fixation of other sounds.")?

One view is that you may not actually get there: arguably, the capture of Gould's fingerings, pedal movements, and so creates a new musical work, derivative of the original Bach piece. (Can anybody deny that he added significant creative expression?)

So, for example, when a radio station plays the Zanph recording, they are no longer playing a sound recording of Bach (which they could do without a license), but a sound recording of Gould's derivative version of Bach. And, for that, the station would need a license.

William Patry said...

Chris, I am suggesting that the Gould re-performance is covered by the first part of 114(b): it is not an independent fixation, but a manipulation of the actual sounds. Whether it is a derivative sound recording of the original sound recording doesn't matter since 114(b) covers both rights in the Gould re-performance situation.

Anonymous said...

In regards to the Van Halen example, I agree that copyright concerns are likely irrelevant, but wouldn't there be some potential trademark issues relating to false endorsement/attribution? It seems to me that use of EVH's distinctive style of guitar would raise the same issues brought up by Tom Waits in his successful suit against Frito-Lays (978 F. 2d 1093).

Dilution might also be an issue...

William Patry said...

Anon: But what if the label on the recording made clear there was no sponsorship or association, and that this was someone else's guess of what he would sound like?

Anonymous said...

"Anon: But what if the label on the recording made clear there was no sponsorship or association, and that this was someone else's guess of what he would sound like?"

Good point...that would certainly diminish the likelihood of confusion. However, a savvy plaintiff might also claim dilution, in which case issues relating to actual confusion would become moot. The question would then be whether labels at the point of sale were adequate to prevent dilution. I think that the potential of radio play would diminish any effect the labeling might have.

I do not know whether these issues have been addressed by a court, so forgive me if I am raising mundane issues. In any event, this is certainly an interesting issue to ponder.

Christopher Fulmer said...

Today, we have machines which will take this information (or some part of it) as the pianist is playing and create sheet music. Voila, a new musical work. I would argue that if the pianist adds enough creative expression to an existing piece, the new musical work that comes out will be a derivative of the original work. And, if the original was in the public domain, a new copyright springs in the derivative work.

So, at some point, there is a computer data file created which contains information about the attack and sustain of each key on Gould's keyboard, when he pushed each pedal and so on. I suggest that these files are even more expressive than the sheet music, as they contain much more information.

So, I suggest that there is a new fixation, and Gould's performance is copyrighted as a musical work.

By analogy, suppose that a musician performs some free-form jazz, which he records, creating a sound recording. 20 years later, he listens to the recording and transcribes, note-for-note, exactly what he played 20 years earlier. Doesn't he now have a copyright in a musical work? (Or is the musical work fixed at the same time as the original sound recording? This is one of the corners of copyright law I don't recall very well.)

Anonymous said...

If Tom Waits' gravely voice is protected why not Tatum's stride? It's just a language thing. A jury can easily understand the attempt to copy the distinctive voice of Waits but is far less likely to understand a distinctive piano playing style unless heavily educated in music composition, performance, practice and theory over the course of a couple of weeks.

Joseph Gratz said...

Setting aside the musical-work question, why wouldn't the Gould re-performance be an independent fixation that merely simulates the sound of the original recording? My understanding of the process used by Zenph is that they use a computer (and their own skills) to reduce the Gould recording to a stream of notes (MIDI data), which is then played back by a player piano and recorded. This is markedly different from the usual case of a sound recording "restored" by technological means, in which the actual audio from the original is copied, in cleaned-up form, on the new recording. It would seem to be more analogous to asking a talented keyboard mimic to listen to the Gould recording her playing it on the piano. Does your opinion depend on the fact that the intermediate copy here -- the MIDI data -- is itself a phonorecord?

William Patry said...

Chris, creating sheet music of someone else's composition or performance does not result in a copyrightable composition; it is merely one form of fixation.

Joe, I may misunderstand the technology, but at bottom, my view of the Gould situation is that because they are starting with the original actual sounds, any manipulation of them is not an independent fixation.

Dean C. Rowan said...

The Sony triple disk Gould set is a gem, but so is the original vinyl. I have a fundamental problem with proponents of technology purportedly allowing one "to reproduce the experience of hearing the original recording live," but it's part of a larger audiophile rant I won't pursue. But keep those tizzy Tatums. You are no doubt aware of the remarkable Ward Marston, who transcribes old vocal recordings to CD. He has little concern for reducing surface noise, yet his work wonderfully delivers the glory of these old performances.

But if you gotta have Eddie Van Halen--that is, a superb parody of EVH--see this! The fellow who created this series of mash-ups is a comic genius.

William Patry said...

Thanks, Dean, I wish I had the money to still be a vinyl guy. I have never heard even a SACD on a great system ($100,000 +) that replicated the space and warmth of vinyl on a very good system I heard 30 years ago. Reports that recordings of music are now being dumbed down from the beginning to take into account the fact that they will be listened to as an MP3 file, makes me happy I do remember what recorded music used to sound like.

Anonymous said...

heady and creepy...

With respect to WKRP and Easy Rider, I get that. Sometimes we want to recreate our experiences with certain sound recordings and 114 gets us there.

However, "sometimes, dead is better." Do we trust what comes reverberating out of the Shrine Auditorium represents things as they were? What, exactly, are we preserving? Mark Katz's book, Capturing Sound: How Recording Changed Music, documents how recording changed audience expectations and performance styles when performances were altered to accommodate the recording process. The late Dwight Conquergood of Northwestern University also argued that certain fixations ("texts" to Conquergood) are used as "decoys" by some groups to lure away prying eyes of outsiders. Obviously, Conquergood wasn't exactly coming from any music scene. His concerns were ethnographic. Still, I think both these examples highlight a certain unreliability of any fixation no matter what its legal status. As a result, I agree that any attempt to protect style is problematic; because any protection based on fixation or use wouldn't be the kind of protection that fosters learning and sharing. I suppose one might argue this kind of protection encourages the production of new works (decoys?) but I think it would be for the wrong reasons. The Zenph technology is very interesting but I think it's important to know what we get out of it.