KERA Arts Story Search

Looking for events? Click here for the Go See DFW events calendar.

Those Famous Robert Johnson Recordings in Dallas: Have We Been Hearing Them Wrong?

by Jerome Weeks 10 Jun 2010 12:00 PM

A British music writer has picked up an old rumor — that all this time, the recordings blues pioneer Robert Johnson made in Texas in the ’30s have been played too fast.


British music writer Jon Wilde picked up on an old rumor for his Guardian blog — that the 29 recordings blues pioneer Robert Johnson made in 1936 in San Antonio and in 1937 in Dallas at 508 Park Avenue have actually been played too fast. His post took off, partly because netsters had been putting up slowed-down versions and debating them for several years now.

Johnson tells WNYC’s Soundcheck that he actually put off investigating or writing about the theory because, as he says, he was horrified that he may have been misled by something so close to his heart. “It’s rather like hearing Citizen Kane was originally shot in color and we’ve been watching it wrong all these years.”

After a few listens, he says, “I wouldn’t say I’m absolutely convinced. But this isn’t about absolutes … Maybe there is a point.”

In his blog post, Wilde writes, “the common consensus among musicologists is that we’ve been listening to Johnson at least 20% too fast” — that “the recordings were accidentally speeded up when first committed to 78 [rpm], or else were deliberately speeded up to make them sound more exciting.” But as the Wikipedia entry on the controversy notes, Wilde “does not give a source for this statement.”

Wikipedia goes on,  however, quoting former Sony music executive Lawrence Cohn, who won a Grammy for the label’s 1991 reissue of Johnson’s works. He acknowledges “there’s a possibility Johnson’s 1936-37 recordings were speeded up, since the OKeh/Vocalion family of labels, which originally issued the material, was ‘notorious’ for altering the speed of its releases. ‘Sometimes it was 78 rpms, sometimes it was 81 rpms,’ he says. It’s impossible to check the original sources, since the metal stampers used to duplicate the original 78 discs disappeared years ago.”

In fact, other recording artists’ works have been released or re-mastered at the wrong tempo — including, Wilde says, the Rolling Stones and the Doors, and the artists only realized this later.

But Elijah Wald, author of Escaping the Delta: Robert Johnson and the Invention of the Blues, comments on the Soundcheck report, saying that when he wrote his book, he was familiar with this theory, and there’s no truth to it.

“First of all, anyone who understands the technology of the period knows that it is simply impossible that all of his recordings, made at two separate sessions, released and unreleased, would have been sped up a consistent amount, unless that was standard policy for the record company, which it was not.

Second, for seventy years there have been people alive who heard Johnson live, and none of them ever mentioned his recordings sounding too high or fast. This story begins and ends with the fact that some modern listeners think that Johnson sounds better if you slow him down. That’s fine, but it’s entirely about their taste, not about Robert Johnson.”

The Soundcheck interview with Wilde thoughtfully includes two different speed-versions of “Crossroads,” which Johnson recorded in San Antonio in 1936. The slower version does make Johnson sound throatier, even, well, bluesier. But then Soundcheck host John Schaefer demonstrates what even an 18 percent difference in speed can mean to his own voice and argues, it’d have been hard to overlook the Chipmunk-like change.

  • Along with Elijah, I find this notion, which is back in circulation, to be based more on contemporary ideas about ‘how a blues man should sound’ than on any historical facts.

    As a student of Johnson’s music and life and author of “Robert Johnson/At The Crossroads” as well as a teaching DVD on Johnson’s music, I find nothing out of the ordinary in Johnson’s tempos or pitches. And neither Johnny Shines, nor David ‘Honeyboy’ Edwards, nor Robert Lockwood, Jr. nor anyone else who we know heard Robert or played with Robert, has ever mentioned that the recordings were too fast. On the contrary, everyone of those folks have said, in so many words, ‘Yup. That’s little Robert,’ even his old girlfriend, Willie Mae, mentioned in ‘Love in Vain.’

    I believe that well-meaning amateur musicologists and blues lovers are simply trying to shape Johnson’s sound to their own liking, rather than taking the recordings at face value. Perhaps they are measuring his recordings against recordings of some of the elder statesmen of the tradition who survived later in the century and who we have had a chance to hear, like Honeyboy Edwards, or Johnny Shines, or Son House.

    Johnson wasn’t in his seventies, eighties or nineties. He was 25 when he cut his first recordings, including his Crossroads Blues.

    Here are my questions:

    Is anyone suggesting that the record producers sped up Louis Armstrong’s recordings a decade earlier? Or Blind Willie McTell?

    Are people repeating this idea suggesting that Johnson’s masters were all sped up, including the pieces that were never released?

    It seems to me that any objective – rather than subjective – evaluation of Johnson’s sound and recordings has to take them at face value.

  • Interesting notion and comments. Considering that House reputedly had to be re-taught some of his own tunes before returning to performance in the sixties, it may be possible that after 30 years and more memories of Honeyboy and other of Johnson’s contemporaries were not crystal clear. I wonder if anyone ever thought to ask Johnny or Robert Jr how often Johnson used his capo and whether his Spanish tuning was D-G-D-G-B-D or E-A-E-A-C#-E. Either way he’d have needed a capo to get Crossroads to the Bb key he seems to be playing in on the versions I’ve got. A good many of his other tunes would have required capos too if we believe the pitch of his recordings and assume he tuned roughly to A-440 as Blake, Lonnie Johnson and others seemed to do. Another factor may be whether some record companies were simply trying to fit some songs to a pre-conceived 3-minute, more-or-less, length requirement. If anyone’s able to speak with either Honeyboy or Pinetop in coming days, maybe they could ask them what they may remember about how frequently Robert used a capo or adjusted his tunings. Too bad Johnny Shines and Robert Jr aren’t still with us. It would be interesting to digitally adjust down pitch and tempos of tunes like Red Hot and Preachin’ Blues by a step or more and see what they sound like. Doubtful that this question will ever be resolved with 100% certainty but I’d probably lean towards the recordings being accurate since, as Scott pointed out, most other recordings of the era seem to be accurate in both pitch and tempo.

    On a separate note, besides clebrating Juneteenth, this weekend marks the 73rd anniversary of Robert Johnson’s Dallas Sessions. From 7:30 to 9:30pm this evening at the Alligator Cafe in East Dallas I’ll do my best to carry on the spirit of this event by performing Johnson tunes and the music of many other pre and post-war artists influenced by, and inspiring to Robert Johnson.

  • dasmb

    Let’s keep in mind, too, that fidelity was not a hallmark of early recordings. Just to hear ANY voice come out of that early equipment was pretty novel; I doubt if anybody was going to quibble over a bit of pitch control.

    As for keeping the slowdown consistent, all you need to speed up a recording is a cutting needle that’s more resistant than the playback needle. Add more weight to the cutting needle, or to the record platter itself, and you’ve got a “sped up” recording.

    Personally, when I hear Johnson “at speed,” I find his trilling voice unique almost to the point of being uncanny. When I hear it slowed down a little, he just sounds like a really talented musician.