Appendix E: Interview with Bob Irwin, September 2002
Bob Irwin, owner of Sundazed Records, recently
remastered the mono Blonde On Blonde for reissue on vinyl. In
preparation he spent a long time trawling through Sony's tape archives looking
for the right tapes to use. He was kind enough to answer my questions about what he found . . .
RF: For the new Sundazed release of the album, were you able to cut directly from the original 1966 mono master tapes, or did you have to create a cleaned-up copy for any reason and cut from that?
BI: No - cut entirely from the original analogs, no clean up necessary, just ever-so-slight eq applied to maybe two or three cuts. Everything else was a straight analog cut.
RF: Was there a "Do Not Use" tape for the mono mix, or would Columbia back in 1966 just have cut straight from the mono mixdown master without producing an adjusted cutting master? The latter is what you seem to imply in your interview in Stereophile magazine.
BI: No adjusted cutting master was originally created. The final mono mix downs were reel-assembled to create the cutting master, ie, a first-generation tape.
RF: So "reel-assembled" means they actually cut the required mixes out of the reels of alternate mixes for each song, and spliced these together into a new master reel?
BI: Yes.
RF: When you researched the available tapes, did you find more than one set of mono masters? The reason I ask is that when the album was originally released, European countries got a very different mono mix from the US release. The European mix sounds less polished - one or two edits of vocal and guitar slips are missed, for example.
BI: There were actually three (partial) sets of mono masters created for the album before the official release. Very confusing, very difficult to sort out. At first glance, it would just appear to be a slight EQ difference between mixes, but eventually a very subtle change in the mix itself would become apparent... a guitar raised a half db or so from the previous mix, or a very slight level tweak in the keyboards, etc..
RF: So the official release was made from the fourth (and only complete) set of mono masters?
BI: Only side three went to the fourth version.
RF: In connection with this, do you know anything about how Columbia allocated matrix numbers back then? I've noticed that all released mono copies of Blonde On Blonde appear to have matrix numbers in the pattern
XLP113761-2a
XLP113762-3a
XLP113763-4a
XLP113764-2a
where the a on the end is some alphabetic character in the range A-J.
BI: This shows that side one went to a second version, side two went to a third version, three went to a fourth, etc.. In each case, there was at least one song per side that was remixed. Obviously, the third side was worked long and hard! Earlier mixes had Bob's guitar level down on some tracks, keyboard levels up, etc.. Perhaps some of these earlier versions are what made it to the U.K. pressing... Sometimes these numeric indications are not clear on the tape boxes themselves.
That letter on the end of the matrix number is the lacquer-cutting info, "A" being the first cutting, then generational-cycling through the letter "J". Then they would start over again with double-letters such as "AA", "AB", etc..
RF: The fact that the third side of the mono Blonde on Blonde got up to a -4 suffix by the time of release would fit with your finding three previous generations of master. I suspect CBS in Europe got sent a copy of the -1 tapes.
BI: Exactly.
RF: Presumably there was a separate master reel for each side of the album?
BI: Yes, on the final side masters - there are four final tapes (all earlier versions are intact as well). But not necessarily sequenced on the mix reels, because the album's sequence wasn't determined until at least the first round of mixing was completed. So, songs were out of order on the original mix reels. I can't remember if there were any sequence changes after the "dash 1" reels were assembled.
RF: You said in your interview with Stereophile magazine that for each song on the album there was a reel (or maybe more than one) of alternate mixes, from which they picked the best to pull out to the master. Would this have applied to mono mixes as well as stereo? Would the mono mixes have been kept on separate reels from the stereo?
BI: Yes, but from what I can remember, not as many reels as the stereos. There were more mono revisions, but less tape generated, because they were working toward a very deliberate end result. That end result changed a few times, but there was clear vision throughout the mono mixing. I can let you know, without a doubt, the mono mix was the one that was considered most important to everyone associated with the album at the time. The final mono mix is much, much more complicated and deliberate than the stereo. It would be nearly impossible to replicate!
RF: Was it more complicated because in mono it was much harder to avoid instruments getting "buried" in the mix, whereas in a stereo mix they'd be audible anyhow because of the stereo separation? Maybe that's putting it over-simply, but is that the essential point?
BI: I think it was not so much an issue of things getting buried, but the desire to make things more pronounced, other things less pronounced. A remix to bring up more of the guitar-figure on "I Want You", then yet another remix to bring that track up even more is one set of revisions that stand out. Another thing that I remember is them tweaking the snare drum sound on "Absolutely Sweet Marie" to get a more high-end "snap" out of it. It was comparatively dull on the multi's. LOTS of those type of changes and revisions.
RF: I realise that the primary target of your researches would have been the mono masters, but did you come across anything that might have shed light on a weird fact in the history of Blonde On Blonde: that the original stereo mix disappeared after a couple of years, replaced by pressings cut from clearly remixed masters? It's generally been assumed that this remixing was done as some sort of later artistic revision (the original stereo mix certainly had some peculiar features). But what you said in the interview suggests another possible explanation - that the original tapes just got worn out. Is it possible the original masters could really have been worn out in just a couple of years, and they maybe then pulled out the next-best mixes from the 1966 multiple-mix reels for each song, to make a set of second-generation masters? It just seems funny that this didn't happen with Bringing It All Back Home or Highway 61 Revisited - all LP pressings of those albums seem to have the same mono or stereo mixes.
BI: Hmmm... Was the original Blonde album remixed within just a couple of years of release? If so, I wasn't aware of that. If that was indeed the case, then I would bet that it was due to artistic revision. The original stereo masters, to the best of my knowledge, were around for many, many years...not worn out until the mid or late 70's, in fact.
RF: Did you find any documentation with the Blonde On Blonde tapes regarding who produced the original mono and stereo mixes, and when and where?
BI: Much of the mixing took place in Nashville - I don't really know exactly how much input Bob Johnston had in the mixes, but clearly Dylan was involved, and I would strongly suspect that most changes came from directly him. In my experience, I don't believe it could have been anyone other than Bob that would have been concerned about, or dealt with, the minutiae of the slight mix revisions... What I'm saying is that, these subtle tweaks and improvements were clearly artistic in nature - not the type of thing an A&R guy would suggest...
RF: Yes, Al Kooper has confirmed that they did some mixing in Nashville, while he was still around following the recording sessions. However, I've also heard it said that some mixing work was done a bit later in Los Angeles, just before Dylan went off on his big tour. It would be interesting to see if your documentation confirms this, but maybe you just don't have this much detail.
BI: I think the original mixing was done in Nashville, the revisions were done in LA. The original Nashville mixes (most unused, except for "Rainy Day Women" and maybe one or two others) were done to 'Audiotape' reels, the LA mix sessions to 'Scotch' brand.
RF: Were the stereo mixes all done in LA? I imagine the original Nashville mixes were mono, or am I wrong there?
BI: I think that's right. Original mono's in Nashville, stereo's done in LA.
RF: Getting back to the question of the vinyl matrix numbers, would each of Columbia's mastering rooms have had their own series of numbers? Is there any way of telling them apart? I heard that back then they had three mastering facilities, in New York, Los Angeles and San Francisco, and each would have had a set of tape masters to cut from. If the final mixing was done in LA than maybe it was only the LA lacquers that were cut from the original masters, and the other facilities worked from copy tapes.
BI: Traditionally, New York would get the masters, LA would get a copy. I've never seen the situation where San Francisco received or held masters, until later years, when original mixdowns were left at that facility, such as certain Sly Stone masters from the later albums, etc.. With Dylan, etc., all masters went to New York.
RF: Finally, do you think that in 1966 the mono release of Blonde On Blonde would still have been the primary, mass-market format, with the stereo more of a specialist product? From listening to the original US mono and stereo releases I'd agree with you that more artistic care was taken over the mono mix. They certainly sound as though they were mixed by different people altogether.
BI: You are absolutely correct. As I said above, there was an unbelievable amount of care and attention given to the mono mixes. The stereo mixes were much more straightforward. It was a great learning lesson on just how much information could be carefully tweaked and expertly coaxed out of a four-track recording to create a wonderful mono mix!
Last updated October 2002