Brian Kehew is well known as the co-author (with Kevin Ryan) of the massive book Recording The Beatles [Tape Op #53]. But he's also been a fill-in keyboardist (and live tech) for The Who and a member of the charming group The Moog Cookbook (with Roger Joseph Manning, Jr.). Oh, and he's played on, engineered and produced a number of albums, and has been the Archives Historian for the Bob Moog Foundation. But, as far as I can tell, he also has the coolest job around; combing the vaults of the big record labels looking for unreleased tracks and outtakes to mix them. I met up with Brian at his modest studio in North Hollywood, California.
What is your job?
My job is to find and mix what's not been heard or released. I started as an engineer/producer. I co-produced Fiona Apple's Extraordinary Machine, among other things. I study recording history and I love that stuff. I'm much more comfortable with traditional styles — be it '50s, '60s, '70s, or '80s — than I am with recording styles now. One day I was in the studio working on an album for a friend of mine, Andrew Sandoval — who's also a producer and archivist. The album's producer was Bill Inglot, who's worked with Rhino and Warner Bros. for 25 years. Bill said, "I like the way you record stuff. It's fast; it sounds traditional. Could you make something sound like an older time period?" And I said, "That's one of my favorite things!" So he said, "Great, we're going to do some Doobie Brothers stuff at Capitol Records. We're going through the vaults." They took us to Capitol's Studio B and we went through the outtakes and found some amazing stuff. There are all those Ted Templeman tapes, which are amazing to hear. People with those kinds of skill...
Good quality, huh? [laughs]
Ted, George Massenburg and Bill Szymczyk are about the best I've ever heard. I put up their tapes and think, "How could George get a bass sound like this? I don't know anyone that could record a bass guitar that sounds as good as that, even with all the gear we have now!" How did they do it? They are amazing. So I began with that one project and made the Doobie's sound like 1974 or 1980; whatever was needed. And I keep on doing those archival projects to this day, 15 years later, for major and indie labels.
These are alternate versions of songs, as well as unreleased songs?
Yeah. To make it clear, one of the things we do not do is remix.
What about taking an album that's been released and re-tooling it?
I haven't really done it. Warner Bros., WEA and all their subsidiaries have this amazing catalogue that is perhaps unrivaled in the world. We go through and find what's there that has not been released. If it's usable, let's say an alternate take with a vocal, we'll mix it to sound good; but we do not remix the released master tapes to make existing albums sound "better," or to sound current. I'm happy not to do that — I call it putting a t-shirt on the Mona Lisa. The only time we've ever done remixes of released masters was when Andy Zax and I went through the Talking Heads' tapes — multitracks for every album — and we found two songs, "Drugs" and "Cities," that had been recorded with a bunch of Brian Eno-isms [Tape Op #85] all over them that were used in the mix only a little bit. This was the only time where were found a remix of the released song to be okay. Not to make it better sounding, but to do a different version of those songs.
Because it could be something different.
It's kind of neat because certainly the Talking Heads' "Drugs" is a very psychedelic song. There were a lot of these cool, weird synth effects on the master, so it came out quite different with those mixed in loudly. The band really liked those two remixes and chose to put them on the box set instead of the originals. Generally, I don't want to put much of a personal stamp on things. If someone used a lot of reverb — like Echo and The Bunnymen, for example — we do copy their style, even if it's not so cool to do that now. We once went through hundreds of Ramones outtakes; I would mix something so it sounded really good to me, then we'd compare to the album and say, "Oh, we have to thin the guitars. We've got to make the drums a little more papery" to match their sound. Not the sound I would choose; but matching their choices from that era.
Turn the vocals way up. [laughs]
And we do, because it's not about your ego, it's about matching what existed before so it sounds accurate. The nice part is getting a fresh project, like when I mixed Black Sabbath's Live at Hammersmith Odeon. That was probably one of my favorite things. It had not been released yet, so I got to mix it the way I hear it. It's a very traditional sound that I like. There are a lot of records from the Stones, Neil Young or John Lennon that have that sound — it could have been recorded yesterday, or could be in the future. The opposite [example] is gated reverbs, '80s chorus, tuning vocals, fast limiting, and squashing the hell out of things; those are all very trendy or dated sounds that I'm usually trying to avoid. That being said, if something is from the '80s we will turn that snare drum up real loud and put some gated reverb on it. It's a style choice of that era. For example, in 1969 people were really getting into wide stereo. They'd do hard panning, where the vocals are on one side and drums are on the other. That stereo thing was new to people, and they used it asymmetrically then; we don't now. I try to preserve the history of things. Like stereo reverb wasn't usually used until the '70s. When I was mixing a 1958 Coasters track you can't put a stereo reverb on that; it just sounds wrong. Keeping it true to the style is a really interesting thing to learn. Bill Inglot and I started working on stuff together initially, and he taught me a lot. I've worked with several different producers from Warner Bros. More recently Andy Zax and I have done a lot of stuff like the Talking Heads reissues, the Woodstock box set [Woodstock: 40 Years On: Back to Yasgur's Farm], and Rod Stewart [The Rod Stewart Sessions 1971-1998]. We went through hundreds and hundreds of Rod's tapes, for three months, and it made for an amazing four-disc set of unreleased studio outtakes. Even in his '80s and '90s era, which I wasn't such a big fan of, we found good music hiding. Sometimes we took off extra overdubs that didn't need to be there and stripped it down to the essentials. We've all heard so many records from different periods that have some awful thing from that style period that we now agree is heinous! But we don't necessarily grab a master tape and remix it just to remove some dated trend. We're looking for unheard material; we're not looking to change the past.
Do you do a bit of research on where artists were working and on what?
Yeah, I always try to look up the history as much as possible. We did some Dr. John tracks and it happens that Eric Clapton was the guitarist and Mick Jagger came along and sang. You need to feature those parts. If I'm looking at the track sheet and see just "Guitar 1, Guitar 2" I might not notice Eric was there. The history helps you understand what people would want to hear. Research is good, but even more revealing is when I actually pull the physical materials and get the tapes out to explore. It's surprising how badly people labeled boxes and tapes back then! We rarely get track sheets, and they're sometimes wrong or vague. We get a reel of four or five takes; I have to decide which one is the good one and then mix it. It's usually fun. It's really a quest. Audio spelunking!
Are you working directly off the tapes or...
There's no computer in this room. [laughs]
Yeah, I noticed that.
We always work from the master tapes, and we work very quickly. A mix is done in 15 minutes to a half an hour. Period. There are almost no exceptions. I do remember once spending about two hours on a song mix and it was, to me, an exceedingly long time. The stuff I do sounds good and it sounds comparable to the released records because we often have a guidepost as to the original sound. I can mix from scratch very quickly, following their guide. Most good tracks can be done that way, especially when the track layout and sounds are similar from tape to tape. If somebody was tracking in the old days — let's say Fleetwood Mac, Van Halen, or the Ramones — they have a fairly similar setup from song to song. You get to work with what they've got. When we do a live album, it's got to be done in a day. I'm going to have to do maybe 25 complete mixes; but the setup is the same throughout, with small changes. It's mixed by hand and we rarely edit mixes together. It's all full passes, usually.
It is a certain limitation when you're presented with these tracks that need mixing. You can't do overdubs. And, for the sake of historical accuracy, you're not going to go in and Auto- Tune or move beats around.
There are people who do. I mean they've gone through a lot of the Jimi Hendrix tapes. They've tuned him and made takes out of pieces that didn't originate together, like moving a solo and so on. I have a real problem with that. Some people don't mind it.
On the posthumous Elliott Smith work that I've done, occasionally there's been a track that was being sketched out, or a rough take, and I have to look for parts that work. Do you find instances like that?
Tones on Tails
I'd say one out of 20 albums has a tones reel. Even when you have them, I rarely rely on tones the way some people do. They are not a ruler set in stone. Tones are an EQ balance between lows, mids and highs. If you don't get a tones reel, you put the music on... how does it sound? If it's too bright you could recalibrate to take down the high end 2 dB on the output of the tape machine. Even if you've got tones on a tape and it lines up properly, you may listen to it and think the low end is boomy. Then take it down at the machine. Maybe they had bad EQ in the studio, or weird speakers, so I try to start from a reference point by listening rather than just believing that tones give us a perfect sound.
Labeling
There was a period when we treated the materials themselves, like the tape box, as museum artifacts. "Don't write on it!" But now we're getting around that with a pencil, just adding helpful information one would want to find. We'll put, "No vocals on here from Elvis Costello," "No complete backing takes," or "Take two is really good." You want to help people of the future who might have to struggle with tape machines from 200 years ago, especially if there's nothing to find. "There's no Miles Davis on any of these Miles Davis tapes" is a good thing to tell people. It would have saved us three days in the studio if someone had written that on those tape boxes originally!
Perfect Imperfections
In doing the work, I learned about how I want to make records. Like the Talking Heads; when we were going through their tapes we found their rhythm tracks are really not tight, and sometimes even sloppy. I was appalled at listening to them soloed up. But I put it together and it's some of the funkiest, grooviest music I've ever heard. It proves that when you get into that nitpicking detail, like when you want to line up the kick drums to the bass hits, it actually can make music worse. Music that has this amazing feel is not precise and is not perfect.
Bootlegs
I love bootlegs. You collect them too, probably. We're digging into something that was left behind, or rejected, hoping to find something good out in the garbage. You know, in all parts of life, people do throw nice stuff in the trash. Sometimes this is material that is superior to what came out. It's clearly a better version, when we're lucky. Many times the artists have heard something they rejected and agreed, "This is better than what we put out. It's more along the lines of what I had in mind."
Subtractive is a common technique, where we find what just doesn't work and it has to come out. There was a Tom Waits outtake that had never been released where he sang "Auld Lang Syne" with a choir against the song in the background. It absolutely fits harmonically with the song that he did, but it totally detracts from what he was doing in the front with the lead vocal. We put it up and did a version with it, but it doesn't work and I can see why they took it out. It may never be released. When we did the first Stooges record [The Stooges] there's an alternate vocal track that's an octave lower on "I Wanna Be Your Dog." It's very different and not as good. I can see why they didn't use it. We did a mix of that. Now I realize that I think they meant them to be layered, two octaves together, but it was not obvious. "Vocal, Vocal," is all the track sheet said. With "Life During Wartime" by the Talking Heads, they had Robert Fripp do this frantic, attacking, slashing, amazing guitar track and he stops three- quarters of the way through the take. And, though I love Mr. Fripp's work, we did a mix of it and it did come out; but I can see why they didn't use it originally.
They probably were at that point in the song going, "Ah, I dunno." And Fripp was like, "I can keep doing this but..."
You can tell. It's very clear. Another odd choice that didn't serve people was when John Cale produced the first Stooges album. The song "Ann" starts beautiful and quiet before kicking into this heavy fuzz-rock guitar solo. It goes on for seven or eight minutes on the multitrack and it's amazing. Then, after the full take, Cale said, "That was good for two minutes," and he cut it off at three minutes on the original album. The longer outtake we found is so much better — that's one where they missed the ball and it was a great thing that got lost. I won't name names, but there are two very famous producer/engineers who I find pretty much terrible — all the mastering guys and engineers who have since worked on their stuff agree. They're legendary in the business, but they record terribly, with bad sounds and levels; and they often choose the wrong takes. We've heard an artist do a truly amazing take on the tape and they'll say, "Nope! One more!" We're falling down hearing this unheard version that was rejected. Then they do two more takes and the real boring, flat one is the one they choose to work on — where the artist has actually lost the spark. In some cases it's nice for us later, because it means we have a really exceptional outtake to find and release. Good or bad, I do find the behind-the- scenes workings fascinating.
I imagine you hear other producers that certainly did pick the best take.
Like the two names I mentioned earlier, George Massenburg and Bill Szymczyk. When we listened to the Eagles tapes, the Bill Szymczyk stuff is just jaw- droppingly good-sounding; plus amazing performances and takes. He must have a detailed method of working that's so strong. Same with Massenburg; he not only captures and maximizes the vibe of the band, but it sounds stellar too. It's all the good things rolled into one, and the arrangements are impeccable. Definitely great producers. I have nothing but respect for the work of those people. You put up any four channels of the song and it already sounds great. Put up another 20 and it sounds like a mix. Different methods of working are sometimes very revealing. In going through the Ramones tapes — we have this illusion of them being the ultra-real punk band rocking together — when actually a lot of it was a producer adding guitar or bass tracks, and some other voice doubling Joey's vocals to make them sound better. Even the famous "one, two, three, four" count offs were Dee Dee Ramone overdubs from a distant mic!
And just dropped in?
Yeah, they added it to the track. Whereas with the band Yes — when we were listening to some Fragile and Close to the Edge tapes — they're in the room playing together with a scratch vocal and it's all there; they're even leaking on each other's tracks. Oddly, it was recorded in segments; they'll do the verse over and over again six times and then pull the one they like. It's like, "Let's hear the tempo... okay... three, four" and then they play the verse of "Roundabout." And then they stop and do it again, but it's all there live in the room. They might add a Minimoog solo, as well as vocals and background vocals; but what you heard is a real performance. That band, which sounds like layers and layers of overdubbed parts, was much more in the room than the Ramones were, which was amazing to me. Their engineer and producer, Eddie Offord, is obviously another genius who can make things sound good. He can make that complexity work and sound cohesive.
You have some tape decks here. But what do you do if it's a format you don't have?
We do pick up and move now and then. If somebody's got weird digital formats, or they want to link multiple machines together, we'll go to Capitol Studios. They have an amazing tech staff. Those guys know how to set up a dbx [noise reduction] decoder and they have Dolby A and Dolby SR. It's easy for them and very few studios have that anymore. Plus they have a massive vintage console, which is great. They've got real reverb chambers and all kinds of cool outboard stuff that we can use on projects. We do most work here at my place because it's my home field advantage. My favorite stuff within arm's reach. But we love Capitol's Studio B.
Do you get paranoid about having rare tapes stored here?
Paranoid? No. I don't keep anything here longer than it needs to be. I do get very uncomfortable if something is just sitting here for two days. When it's done, we take it back to the vault. It's very low-key once you've done this daily; you see important tapes all the time, so you can't treat things with crazy fear. You have to get the job done. We're not going to do something stupid either, though. We're cautious. We are very careful, and things are under lock and key, always. You just have to be attentive and treat it with the value it deserves.
You can't let this music out of your hands.
Ha! I certainly could be the king of bootlegs, as could the producers I work with: Andy Zax, Bill Inglot, Andrew Sandoval, Mason Williams and Cheryl Pawelski. All these good producers have amazing stuff, but we would not work if we were letting things out to anyone but the label. Remember, this is a dream job for me. So it's not worth the risk of losing this great job.
Yeah, it might mean you never work in the field again.
It's that simple.
I would guess that a lot of songs that you've mixed are never going to be heard.
Sadly. Some of the best stuff is still not out, for political or band reasons. There are lots of amazing early Van Halen tracks that could come out.
What do you foresee in the future for this kind of work?
I dread doing this job in 10 or 15 years when I have to search through stem after stem of vocal parts, and outtakes in folder after folder. Which version of Pro Tools were they using? Which version of Waves Gold plug-ins did they use? It will be an almost impossible task to go through stuff. It's daunting enough to go through reel after reel of multitrack on tape. You may have the wrong operating system in 50 years, or it might be very hard to pull up a hard drive that doesn't want to spin up. I'm really worried about that.
Do you have cases where it's really hard to guess what the vision might have been with a song or parts?
Maybe it works well because we're approaching it as fans. We're approaching it as the consumer. How many times have we heard an artist say, "This is finally the album I want to make," and nobody wants to hear it? We're approaching it as, "This is Twisted Sister, Dion [DiMucci] or Nico — what was good about them? What do people like or want from them? What do we want to know about them?" We're approaching it from that aspect of a fan saying, "I can see why they wouldn't choose this originally, as it shows them being a little frail or a little bit human. But it's really nice for us to finally hear that." It's an intimate, fly-on- the-wall kind of thing. And yes, some artists might sit here saying, "That's not a perfect performance." But usually we don't put out anything embarrassing. There's certainly nothing that's damaging to an artist's credibility released; we censor a lot of unflattering stuff!
www.curvebender.com
