Rather late in the process, somebody in the administration of the film had the presence of mind to ask if we had the rights to use the recording of The Valkyries that we were using in the film during the attack on the beach. And this had been in the original script that we were going to use The Valkyries; this is John Milius's idea. It was Francis [Ford Coppola], I believe, who chose that particular recording which is a famous recording. It's Sir George Solti's recording with the Vienna Philharmonic in 1966, and it's the go-to classic of The Valkyries. And... so as I was editing those scenes using that music, that was just given; that's the music that we're using. And I never thought about it, and if I had thought about it, I would've thought, 'Well, somebody has already secured the rights to this.' But the question was asked with probably a month or two months before the release of the film, and the answer was, no, we don't have the rights to it. Who owns the rights? Well, it's a record company Decca. So a letter was drafted, sent to Decca and a week later the answer came back, 'No, you cannot use this for the film.' So another letter was written, 'Please, can we use it?' 'No, you cannot use it.'

So panic set in, what are we going to do? And a triage response was setup. The first fork of the triage was: Walter, go to the record store – and this was long before the internet – go to Tower Records and get every recording of The Valkyries that there is. Listen to them all, and there were 17, I think, and find out if there's one that is sort of as good as, maybe, the Solti recording and we'll see if we can get the rights to use that. The other fork was: continue to pound away at Decca. And the last fork was, let's record it all again with The San Francisco Symphony. And so Francis's father who was a composer was assigned the task of wrangling The Valkyries and arranging to record The Valkyries with singers. And I mean, it was a huge production when you think about it, with the San Francisco Opera and Chorus and whatever.

And I did listen to all 17, and I developed a grid system analysing Solti's metrics every five seconds. Because what seems like a regular tempo is not, in fact, regular, it only seems that way. And good conductors are constantly modifying the time base of their rhythm to give you the impression of regularity. But if they were actually regular, the music... if they were metronomically regular, the music would not be alive. It's similar to the solution that the Greeks came up with for the columns of the Parthenon that they actually bulge a little bit, but that gives the impression of rectilinear. If they look... if they were actually rectilinear, they would look pinched at the waist. So you have to distort things in order to give the impression of regularity. So this is the musical equivalent of this, so I mapped all of this out, and then judged the recording... the other recordings by how well they tracked Solti's decisions. Of course, they're not going to be the same in every case by any means, but are there ones that come closer to the others? So I was able to get rid of three-quarters of the recordings almost immediately. And finally, it came down to one single recording which was Erich Leinsdorf's recording with the LA Philharmonic that had been recorded the year previously, 1977.

So the next step is take that record, get it transferred to magnetic film, and then put it on the editing machine, line the two things up and play it. I did that, and as soon as I turned the switch, I was optimistic. It actually sounded pretty good, and the metrics were actually pretty close, so close that I thought for about half a minute, 'Yes, I can make this work, I can make little changes that will keep it all... It would be a lot of work, but we were in a desperate situation.'