Topic: Are Robert Johnson's recordings fraudulent?

[dj]

Lindy,

You're right to point out variations by region/style, etc.  One of the things that jumps out at me is that almost half of my lost is made up of Texas pianists.  It's certainly possible that this was a stylistic preference.

[banjochris]

DJ:

One thing I've noticed over the years is a difference between white performers/groups (including hillbilly, early country, and what we refer to as white country blues) and black country blues performers. I'm talking about recordings in the 1930s, maybe a stray tune from the 1920s. The former tended to end their songs between 2:55 and 3:10 and the latter would stop anywhere from 2:10 to 2:50. This wasn't true for all black performers--black jazz bands tended to have set arrangements that took them right up to the 3:00 mark.

Admittedly my sample is small for the white performers, the Columbia compilation called "White Country Blues". For the black performers I have multiple compilations from Catfish, JSP, etc. etc.

Let it be known that I think this means absolutely nothing, I don't feel any need whatsoever to ponder why it might be true, don't really care. It's one of those small observations I made one day many years back, thinking "Maybe someday I'll be on a truly esoteric game show and this will be worth some prize money, or whatever's behind door #3." But your post dredged it up from my memory.

Lindy

Lindy:
It has seemed to me that the time limit depended on the label/producer as well as the performer. It seems to me that Victor had a lot of blues records that really push it to the limit, like Canned Heat, T-Bone Steak Blues, I Do Blues (that was Victor, wasn't it?), some of the Bo Chatman early string band sides -- those are all over 3:30. Mississippi Sheiks records for both OKeh and Paramount occasionally push the length past 3:30. Maybe it had to do with the producer sometimes aying close attention to how long a verse took and knowing they had enough time before they turned on the red light. I don't think it means much either, but it is sort of interesting.

[dj]

I have 643 songs recorded in 1937 loaded into iTunes.  106 of them, or a little over 16%, were 2:35 or less in length.  So while Robert Johnson's songs were on the short side of things, they were not without company. 

Chris, interestingly enough, a lot of the short ones were recorded on Vocalion field trips, which would have been supervised by Don Law.  I guess he liked to make sure there was a lot of room on the disks.

[Kokomo O]

Could be, and there are legitimate technical reasons for wanting to have that room--you get more low end, more dynamic range, generally better sound.

[banjochris]

Another reason songs may have gotten shorter into the 30s: the rise of the jukebox -- the shorter the song, the more folks have to put in to keep the music going.

[repeater]

Repeater, you present something as a conspiracy from the start, refuse to accept logical and musical and technical explanations of why the idea of Johnson's recordings deliberately being sped up to the degree suggested is extraordinarily improbable (to be polite), and demand a scientific explanation for something that in all probability never happened. How are people supposed to react? If you think we need a scientific explanation, then why ask, go hire a scientist.

Other than the truth.  I presented nothing as a conspiracy.  And while I received "explanations," they are hardly watertight, and I quite justifiably found them less than satisfactory when they lacked either clarity or support.  Of course, when I asked for explication or citations, I was rebuffed. 

Also, I most certainly did not "demand a scientific explanation."  I said that I think one is needed, in the hope that perhaps some frequenter of this Web site might know of such an appraisal having been done, or might be able to do one.

I'll thank you not to distort the record (no pun intended).

[repeater]

I have 643 songs recorded in 1937 loaded into iTunes.  106 of them, or a little over 16%, were 2:35 or less in length.  So while Robert Johnson's songs were on the short side of things, they were not without company. 

Chris, interestingly enough, a lot of the short ones were recorded on Vocalion field trips, which would have been supervised by Don Law.  I guess he liked to make sure there was a lot of room on the disks.

Great to know.  Thank you.

[repeater]

Slack: Leave the thread open for the time being. Let's all try to focus on the question: Are the recordings we have of RJ accurate reproductions of how he actually sounded at the time he made them?

Gibbens states, "If the theory I?ve advanced is not completely crazy, a possible motive for speeding up Johnson?s records might have been to try to make them more exciting for an age in which the Delta tradition he came out of was already a thing of the past."

His article posts a link to the IGS thread, which I suggest you review. Dave Rubin has some interesting things to say as he actually listened to the 78s.

But it is a theory--he says so himself. What are it's presuppositions? And what do the other statements Gibbens makes presuppose?

Where and what is the objectively verifiable evidence (not possible speculative reasons) that the specific engineers who recorded RJ made the deliberate and conscious decision(s) to have the equipment that recorded RJ's individual songs (either one, some or all of them at both sessions) turn at a slower RPM speed so that the resulting 78s would make RJ's song(s) sound faster and pitched higher than they actually were when he preformed them when they were recorded?

Gibbens also says that "[p]erhaps there are scientific tests that could be applied to the sound that might establish its original frequencies ? to the qualities of the voice, for example, like the vibrato, which at full speed sounds to me like an alien nasal flutter but at slower speeds like a proper musical ornament; or perhaps to the decay time of the guitar notes."

Might something like this be feasible?  Conclusive?

As for whether any speeding up was deliberate or not, I think that's tangential at best to the real issue (not to mention, perhaps impossible ever to know with certainty), which is, as you say:  "Are the recordings we have of RJ accurate reproductions of how he actually sounded at the time he made them?"  I'm reluctant to believe that there isn't some way of analyzing the recordings to enable us to know for sure one way or the other.

[Stuart]

This tangent just came to mind:

http://www.pbs.org/wgbh/nova/sciencenow/0401/03.html

http://www.pbs.org/wgbh/nova/sciencenow/0401/03-perf-flash.html

http://www.antarestech.com/

http://www.npr.org/templates/story/story.php?storyId=97312511

Lots more on the web--Andy Hildebrand is Dr. Harold Hildebrand.

Maybe Andy Hildebrand might know of something that can be applied to mid-to-late 1930s recordings.

More to follow...


RJ's recordings may not be fraudulent, but I think we're now in an age that, if he were alive and recording today, "only his hairdresser (recording engineer) would know for sure."  

[Lyle Lofgren]

RJ's recordings may not be fraudulent, but I think we're now in an age that, if he were alive and recording today, "only his hairdresser (recording engineer) would know for sure."

It does seem strange to be arguing about RJ, when a number of recordings released today (i.e., those which have had pitch correction applied, and other computerized doctoring) are obviously fraudulent, even though you can't detect the frauds by listening to them.

Lyle

[waxwing]

Gibbens also says that "[p]erhaps there are scientific tests that could be applied to the sound that might establish its original frequencies ? to the qualities of the voice, for example, like the vibrato, which at full speed sounds to me like an alien nasal flutter but at slower speeds like a proper musical ornament; or perhaps to the decay time of the guitar notes."

Might something like this be feasible?  Conclusive?

As for whether any speeding up was deliberate or not, I think that's tangential at best to the real issue (not to mention, perhaps impossible ever to know with certainty), which is, as you say:  "Are the recordings we have of RJ accurate reproductions of how he actually sounded at the time he made them?"  I'm reluctant to believe that there isn't some way of analyzing the recordings to enable us to know for sure one way or the other.

The best test which has been suggested is to examine an actual 78 recording sonically and find the spike at 60Hz. All electricity at the time of RJs recordings was supplied at 60Hz and light bulbs and other electric appliances in the room, including the recording equipment, would give off a hum at this frequency and it should be detectable. If there is a spike at about 72Hz than it has been sped up by 20%. To my knowledge no one has put together an original RJ 78 and an oscilloscope. Andrew, at Pristine Sound, who has done remasterings of most of RJ, used existing transfers but claims he discerned that they were approximately at the correct playback speed. You can find discussion about that either at the Pristine site or on the Blind Man's Blues Forum, which I think you are familiar with.

Obviously you didn't even bother to read my post before you demanded citations of my assertions. This is quintessential troll behavior and I suggest you be banned from the site.

You obviously also never read any of the threads at the IGS forum which Stuart posted or you would have seen this same suggestion made by a poster there who uses the logon "Adrian". Adrian is Adrian Freed, the Director of Research at UC Berkeley Center for New Music and Audio Technologies. If you take an original RJ 78 to him he will gladly perform the test for you. Now all you have to do is wait for the next one that comes up on ebay, bid somewhere over $5000, and you can have your proof, one way or the other.

Wax

[Stuart]

Gibbens is not alone in his question regarding whether or not some kind of science based method (something that might be better referred to as "technology") for analyzing recordings so as to determine the original recording speed exists or could be developed. If such a method is available, it would go a long ways towards resolving many of the uncertainties. I would hope that such a scientific method or technology would be evaluated with sufficient rigor so that we could have faith and confidence in the results of its application.

My question about the deliberate and conscious decision(s) of the recording engineers is specifically aimed at the hypothetical case that Gibbens mentions, one that has been around for years. I believe the IGS thread dates from 2004 or so. I have not heard of any evidence that suggests that this was the case. Speed variations and a certain amount of slop may or may not have occurred when RJ was recorded. (And if so, on what sides and how much?) I wouldn't say that my question is tangential to what is the main question, but derives from a more general question and is specifically applied to address the idea that RJ's recordings "were speeded up by 20%." I guess the general form of the main question would be, "Are the recordings we have accurate reproductions of how what was recorded actually sounded at the time they were made?" (Referring to the set of all extant recordings--awkward at best, but I tried to keep it close to the specific RJ application.)

In your reply you said, "(not to mention, perhaps impossible ever to know with certainty)"

This is a very important point. Trying to arrive at a decision and/or reach a conclusion in the absence of a preponderance of objective, verifiable evidence that really "closes the case" sometimes means that we have to look at everything we have, as well as the context. We have to critically analyze and evaluate the quality of the evidence as well as assess the quality and soundness of the reasoning and argumentation?both pro and con, as well as that of alternative explanations. And regardless of what we do and how hard we try, there is always an element of uncertainty.

Regarding the guitar tunings, 14 fret clear guitars, capo positions and other factors: As I mentioned and as dj followed up on, there are many people who can play RJ's music on 14 fret clear guitars. It doesn't suggest that any of us have talent that in any way approaches RJ's, but it does address the view that that what we hear on the records would have been impossible to play. As a result, I do not think that this is a valid objection to the recordings being accurate reproductions of RJ. My personal opinion is that the voice we hear on the recordings is an accurate reproduction of RJ's voice. I do not think that the recordings are "speeded up," although there may be some slop due to natural variations in RPM speed given the technology of the times. One thing that reinforces my view is what Andy Hildebrand said in the NOVA piece about the pitch changes that can be made using Auto-Tune software. He said that there are many sonic properties to the human voice (and musical instruments as well), and that pitches cannot be raised  without there being a "chipmunk" effect. Many other adjustments have to be made as well, or else the resulting sound is not natural. Obviously it's just my subjective, impressionistic opinion, but RJ's voice and his guitar do not sound like the recordings have been manipulated. I don't hear the chipmunk effect.

Finally, I've always liked RJ and never have let all of the hype sour me on his music. As I've said before, he never hyped himself after his death. Enjoy and appreciate what he has left us.

In closing, thanks for returning the tone of this thread to a level of civility. I encourage others to do the same.

[jpeters609]

Many of us who are now great admirers of the pre-war blues masters were first introduced to the music through the hagiography, hoopla, and (ultimately) hype surrounding the recordings of Robert Johnson. This was especially true during the 1960s and again in the 1990s, but nevertheless the Robert Johnson myth remains a potent and popular entry to the world of pre-war blues.

As someone who admittedly enjoyed the myth many years ago, I admit to a certain sadness at its diminishment. There's magic in them thar myths, though they bespeak a bit of childishness. For non-musicians, especially (whose attachment to the music is perhaps more esoteric and less felt at the fingertips), the explaining away of the Robert Johnson lore steals away a bit of the magic, despite the abundant wonders of the music itself.

I think we sometimes replace one mystery with another. In this case, if Robert wasn't really a mere teen who sold his soul to the Devil at some mythical crossroads, and who didn't really compose every lyric and guitar phrase he employed, then maybe there's some other mystery we can seize upon. Perhaps it's an evocative photograph of two unknown men purchased on Ebay. Or maybe his recordings were secretly manipulated, and the "true" Robert Johnson yet to be discovered, hidden in some yet-to-be calculated rpm.

Much of the pre-war blues (I'm thinking mainly about Charley Patton, Geechie Wiley, Willie Brown, and Robert Johnson, but of course there are so many others) is bound up in mystery and magic, conjured I suppose by the sometimes mystifying lryics, but mainly by the simple mists of time. Personally, I don't discount these mysterious aspects. I think they add color and texture to the story. But I also think they are largely a fiction, and I know what really draws me to the music is the mastery of these very talented individuals. Ultimately, there's enough magic in that mastery. We don't really need to add any more.   

[repeater]

Gibbens also says that "[p]erhaps there are scientific tests that could be applied to the sound that might establish its original frequencies ? to the qualities of the voice, for example, like the vibrato, which at full speed sounds to me like an alien nasal flutter but at slower speeds like a proper musical ornament; or perhaps to the decay time of the guitar notes."

Might something like this be feasible?  Conclusive?

As for whether any speeding up was deliberate or not, I think that's tangential at best to the real issue (not to mention, perhaps impossible ever to know with certainty), which is, as you say:  "Are the recordings we have of RJ accurate reproductions of how he actually sounded at the time he made them?"  I'm reluctant to believe that there isn't some way of analyzing the recordings to enable us to know for sure one way or the other.

The best test which has been suggested is to examine an actual 78 recording sonically and find the spike at 60Hz. All electricity at the time of RJs recordings was supplied at 60Hz and light bulbs and other electric appliances in the room, including the recording equipment, would give off a hum at this frequency and it should be detectable. If there is a spike at about 72Hz than it has been sped up by 20%. To my knowledge no one has put together an original RJ 78 and an oscilloscope. Andrew, at Pristine Sound, who has done remasterings of most of RJ, used existing transfers but claims he discerned that they were approximately at the correct playback speed. You can find discussion about that either at the Pristine site or on the Blind Man's Blues Forum, which I think you are familiar with.

Obviously you didn't even bother to read my post before you demanded citations of my assertions. This is quintessential troll behavior and I suggest you be banned from the site.

You obviously also never read any of the threads at the IGS forum which Stuart posted or you would have seen this same suggestion made by a poster there who uses the logon "Adrian". Adrian is Adrian Freed, the Director of Research at UC Berkeley Center for New Music and Audio Technologies. If you take an original RJ 78 to him he will gladly perform the test for you. Now all you have to do is wait for the next one that comes up on ebay, bid somewhere over $5000, and you can have your proof, one way or the other.

Wax

What the hell is your problem?  I read your post carefully, and it was your claim about electricity in particular that I would've liked a citation for.  Blues fora, incidentally, are not in and of themselves credible sources.  And a vague mention of some discussion at some other Web site isn't a helpful citation.  Nor is it incumbent on me to know who "Adrian" is.  Dick.

But get me banned, big man.  And all other insolent "trolls" who don't know their place.

[repeater]

Gibbens is not alone in his question regarding whether or not some kind of science based method (something that might be better referred to as "technology") for analyzing recordings so as to determine the original recording speed exists or could be developed. If such a method is available, it would go a long ways towards resolving many of the uncertainties. I would hope that such a scientific method or technology would be evaluated with sufficient rigor so that we could have faith and confidence in the results of its application.

My question about the deliberate and conscious decision(s) of the recording engineers is specifically aimed at the hypothetical case that Gibbens mentions, one that has been around for years. I believe the IGS thread dates from 2004 or so. I have not heard of any evidence that suggests that this was the case. Speed variations and a certain amount of slop may or may not have occurred when RJ was recorded. (And if so, on what sides and how much?) I wouldn't say that my question is tangential to what is the main question, but derives from a more general question and is specifically applied to address the idea that RJ's recordings "were speeded up by 20%." I guess the general form of the main question would be, "Are the recordings we have accurate reproductions of how what was recorded actually sounded at the time they were made?" (Referring to the set of all extant recordings--awkward at best, but I tried to keep it close to the specific RJ application.)

In your reply you said, "(not to mention, perhaps impossible ever to know with certainty)"

This is a very important point. Trying to arrive at a decision and/or reach a conclusion in the absence of a preponderance of objective, verifiable evidence that really "closes the case" sometimes means that we have to look at everything we have, as well as the context. We have to critically analyze and evaluate the quality of the evidence as well as assess the quality and soundness of the reasoning and argumentation?both pro and con, as well as that of alternative explanations. And regardless of what we do and how hard we try, there is always an element of uncertainty.

Regarding the guitar tunings, 14 fret clear guitars, capo positions and other factors: As I mentioned and as dj followed up on, there are many people who can play RJ's music on 14 fret clear guitars. It doesn't suggest that any of us have talent that in any way approaches RJ's, but it does address the view that that what we hear on the records would have been impossible to play. As a result, I do not think that this is a valid objection to the recordings being accurate reproductions of RJ. My personal opinion is that the voice we hear on the recordings is an accurate reproduction of RJ's voice. I do not think that the recordings are "speeded up," although there may be some slop due to natural variations in RPM speed given the technology of the times. One thing that reinforces my view is what Andy Hildebrand said in the NOVA piece about the pitch changes that can be made using Auto-Tune software. He said that there are many sonic properties to the human voice (and musical instruments as well), and that pitches cannot be raised  without there being a "chipmunk" effect. Many other adjustments have to be made as well, or else the resulting sound is not natural. Obviously it's just my subjective, impressionistic opinion, but RJ's voice and his guitar do not sound like the recordings have been manipulated. I don't hear the chipmunk effect.

Finally, I've always liked RJ and never have let all of the hype sour me on his music. As I've said before, he never hyped himself after his death. Enjoy and appreciate what he has left us.

In closing, thanks for returning the tone of this thread to a level of civility. I encourage others to do the same.

Agreed.

As I've said before, I would very much like to believe that what we have on CD and vinyl is reasonably close to what Johnson actually sounded like.  I don't, however, believe that it is wrong, or absurd, or trolling to ask whether that's indeed the case, given the doubts raised by Gibbens and others.  Nor is it wrong, absurd, or trolling to challenge incomplete or unsubstantiated claims from self-proclaimed experts and bouncers.