Pristine Classical: Preserving Priceless Historical Recordings, Part Two

Pristine Classical: Preserving Priceless Historical Recordings, Part Two

Written by Ted Shafran

In Issue 162, we profiled Pristine Classical, a company dedicated to improving the sound of historic classical music recordings, many of which were recorded using primitive equipment under less-than-ideal conditions. However, during the first half of the 20th century, some of the greatest musicians who ever lived gave performances that remain, today, the subject of legend. Andrew Rose of Pristine Classical is dedicated to making those sounds more accessible to modern ears. Our interview with him continues here.

Ted Shafran: So far we’ve talked about your early career, some of the technical details of your restoration processes, and the business realities of running your label. Let’s turn to the music. Do you have a personal preference with respect to period, specific artists, type of music, and so on?

Andrew Rose: (Laughing) I actually get so many e-mails and messages and requests, so a lot of the time I’m sifting through what people are asking me for. I’m also looking at building a catalogue in different directions. I’m aware of the artists that sell. I have to be clear that this is a business, where we really need to put out some Toscanini and Furtwängler and Callas at reasonably regular intervals in order to keep sales ticking over, and it’s a case of balancing it out across the genres as much for any particular artist or composer.

When we began there were obviously big gaps in repertoire. For example, someone e-mailed me and said you’ve not got Cosi fan Tutte in your catalog and I realized he was right. He suggested that the von Karajan recording [was] desperate for [restoration]. And I have various collectors who will come to me with suggestions and projects and recordings that they’ve managed to source, which they send to me. I’m working with the Busch Brothers Institute in Germany. And I’m also working with Misha Horenstein and doing a lot of Jascha Horenstein recordings from his private archive.

So, it’s a case of planning out roughly six to 10 months of weekly releases. I start by focusing on the big projects and then I look at the gaps and I might [think], I could really use some more solo piano recordings. Then I look at what we’ve got and what is missing from our catalogue. And if we find something that’s missing we look at who recorded it. Sometimes it’s obvious, but sometimes we ask ourselves: do we really need another set of Beethoven sonatas or another set of Chopin preludes? How about something different, like Erik Satie? But then we’ll realize that the reason we haven’t done it is because no one had recorded him until the late 1960s, apart from one album that was recorded in 1956, which we issued a few weeks ago. So, there’s no single strategy. I do try to see what and who we are missing, but also what will sell, and try to balance those out to something that looks like a reasonable set of releases over a six-to-eight-month time period.


TS: You mentioned that some of your materials come from private collectors. Are there are other, more common sources where you find these older recordings?

AR: (Laughing again). I’ve actually had several people donate big collections of LPs to us, 78s, and reel-to-reel tape, so I’ve got a big collection there. And if there’s something that I haven’t got, I will go out and find it. If that involves buying it and getting it to cross the world to here so that I can digitize it, then so be it. If there’s something that I really want, and I’ve asked around to various collectors and people that I know and I still can’t find it, then I’ll go online. And if I still can’t find it, well, there’s no shortage of other things to do until a copy comes along. But there are a number of people who are very good at this and have been in collectors’ circles for a long time. Sometimes they can find someone in their network who can furnish a copy of something that I’ve been looking for. Sometimes I honestly don’t know how recordings came to the person who got them to me. Going back quite a number of years ago, I worked on a Toscanini Bruckner 7 which was a fragmentary recording which I believe exists in the Library of Congress, and nobody knows how it escaped [from] there.

TS: Yes, that’s quite a performance; I’ve played it quite a few times and I don’t think I’ve ever heard anyone take the Scherzo that fast.

AR: It’s a shame there’s so much missing. One of the downsides of this job is that I’m always working on the next project and once something is finished, I move on to the next one. So, you often forget recordings, even ones you’ve done recently. And frankly, if I’ve been plowing through some tricky restoration work, I just prefer to have a bit of silence afterward. Most of the people that I communicate with are listening in a different way than I listen. If I listen to a recording, I’m looking for the faults rather than listening to the performance. I think back to when I was working in radio. And when you’re working in live radio and you’ve got a fast-paced show that’s about three hours long, there’s always the next item that’s about to come up, and you’ve got to get a lineup from somewhere and you’ve got jingles and you’ve got news inserts and you’ve got sports and all this kind of stuff going on. And you’ve got to make sure all the levels are right, and the equalization is right and everything is working sonically. So, I would get to the end of a three-hour news show and [would] have to put the TV on to find out what was going on in the world! And I think that’s actually helped me in this job, because I don’t get carried away by the performance. I concentrate on how it sounds, what needs to be fixed and how long it will take to fix it as I’m going through the recording. It also means that, for a while after I’ve finished something, I don’t really want to listen to it again, in case there’s something that I missed.

TS: Well, nothing in life is perfect.

AR: (Laughing) There’s always one more thing that you could have worked on.

TS: I understand what you mean. A lifetime ago, I did some sound work for an amateur theater and I remember that my entire focus was on the next cue. I’d have to think hard to tell you what the plays were.

AR: I can think of any number of very famous people that came through our studios when I was at the BBC and I can’t remember anything about any of them, except for [violinist] Nigel Kennedy. And that was because he arrived about 40 seconds before we went to air, and he was supposed to be playing live for us. But he thought we were [going to be] pre-recorded, so I had 40 seconds to mike him up do a sound check before we were on air.

TS: There are different copyright rules in different jurisdictions. Do you run into any copyright issues?

AR: Very, very rarely. I think it is an interesting situation because United States laws are the tricky ones. The rest of the world seems to be OK. The European Union where we’re located seems to have laws that are easy to understand and to follow. The US laws have been chopped and changed over the years, and Naxos Records got into trouble years ago in that respect. On the whole, we haven’t had an issue. Occasionally I have had contact from orchestras. I can remember an American orchestra and a European orchestra whose managing directors got in touch and said, “you can’t use our stuff.” I explained that we don’t actually sell many copies and we’re not making a lot of money out of these, and they went quiet because I think they realized we’re not stealing their revenue to any substantial degree. On the other hand, we have had very good working relationships with other orchestras and performers. The San Francisco Symphony asked if they could sell some of our recordings, and usually it’s the other way around, so we have a good relationship with them.

There was a recording I did a long, long time ago and I don’t remember the record company, other than that it was an American label. They got in touch and asked if they could issue it on their label, and asked about the copyright issue. I told them that if they had the masters and wanted to issue it, we would just take it off our catalog, since we weren’t selling that many copies anyway. We [did] have an issue with the Horowitz estate; there were a couple of recordings that we [had] issued [and] they weren’t happy about us selling downloads from Europe to the US. So, they [put] Sony’s lawyers on to us, and we ended up in a compromise whereby people can buy the CDs from Europe and import them to the States. Aside from that, I’ve always been very, very careful not to gamble in anything borderline copyright-wise. Anything published before 1963 is fair game.

There’s a sunset clause for more recent recordings, which allowed us to issue some Horenstein recordings with the Gothenburg Symphony that were made in the late 1960s. These were recordings that were sent to Horenstein’s cousin Misha but were never used. They were neither broadcast nor released, so they fell into the public domain after 50 years. But you’ve got to be very careful to make sure that you’re not stepping on anybody’s shoes and you’re absolutely watertight legally. I can’t afford to go to court over anything like this. There’s so much that’s in the public domain that we can [use]. It’s a shame that the European laws changed, because we used to have a nice time at the end of each year when we could access another year of recordings. But that stopped, probably to keep the Beatles in copyright.


TS: So, this is no longer a moving line?

AR: No. Basically, what has happened is that what used to be a 50-year rule is moving to a 70-year rule. In theory, in 2034, 1963 will fall into the public domain. I strongly suspect that by then there will have been further lobbying to increase the law from 70 to 90 years in much the same way as [things were done in the US] to keep Mickey Mouse in copyright. But by the time that happens, I probably won’t be doing this anymore..


Andrew Rose in the Pristine Classical archives.

Andrew Rose in the Pristine Classical archives.


TS: That leads into my next question. I know that you have restored a number of early stereo recordings. Do you plan to do more of that?

AR: I see no reason not to. For example, I’m working through the Maria Callas EMI recordings and [we’re] about to [get to the point where they] go stereo, and I will carry on for as long as we can make a difference. I don’t want to issue something just for the sake of putting it out. I won’t do it unless I think that there’s something that I can bring to it that makes it a worthwhile investment for anybody who likes that recording. What’s fascinating about the 1950s – and I’m into that decade at the moment – is that the pace of change technologically over that decade was enormous. Sometimes you feel that the engineers of the day couldn’t keep up with all the new things that were being thrown at them in terms of microphones, tape, and then stereo and so on. By the mid 1960s there were fewer problematic recordings being made. But during the ’50s, particularly the first half of the decade, there are recordings that can be vastly improved with the approach that we use. And that definitely takes us into the early stereo era.

TS: On the subject of stereo, I know that many, if not most of the recordings that you release are available in Ambient Stereo. As far as I know, no one else in the historical recordings field is using that technology. What is the reasoning behind this?

AR: It started out with a bit of software that was released quite a long time ago now; I think maybe it was 2007 – 2008, about the same time as the XR technology I developed, and I liked what it brought to the sound of a recording. We initially would offer this as an option, so you could buy a download in mono 16 or 24-bit, a mono MP3, and an Ambient Stereo 16-bit FLAC. It quickly became apparent that people liked the sound of [the process] so we basically followed the market in that respect. Other [technologies] came along which further enhanced what we were doing, and by following the technology and seeing how people responded, I’ve ended up in a position where all of the mono recordings that I do now go through these various processes of equalization, Ambient Stereo, and using convolution reverb to create the feel of the concert hall while [still] keeping a mono-ish sense of the original recording. It was really a case of [accommodating] what our public seems to want. Almost never do we have someone telling us they don’t like Ambient Stereo and want mono. I do get e-mails saying they wish that Mark (Obert-Thorn) would do it too. And from a business perspective, I think you’ve got to have a unique selling point.


TS: Personally, I like it [Ambient Stereo] very much. I agree that I think it adds something to those recordings.

AR: It’s something that has evolved over about 15 years. I do keep abreast of new technologies as they arrive. I’m beta-testing new versions of the restoration software that I use, and helping companies come up with new products. We try to keep as close to the cutting edge as I can, and hopefully that’s reflected in an improving product from us as the years go by.

TS: That actually leads into a question I was going to ask. Looking to the future, I’ve often wondered whether it would be possible to “rebuild” missing frequencies on older recordings by interpolating things like overtones. Do you think that kind of technology might be available in the future, and would you be inclined to use it?

AR: I think we’re not far away from that. There is software in development – not necessarily for this particular purpose [and] what I’ve seen so far has been working with speech rather than music – but it’s certainly using a degree of artificial intelligence processing. It is certainly something that is feasible. I wouldn’t have said that three or four years ago, but seeing some software in development I can see that being a possibility. The problem really is that what you and I would like in music is a very, very small field to try and market and develop software for. If you’re developing a suite of digital audio restoration software, a guy working with acoustically-[recorded] 78s is going to be a very small market for you; not something you’re going to devote a lot of time and effort to. But I think it’s certainly something that could probably be done in the next five to eight years.

There are other things that I likewise would like to see and would have [thought] impossible until quite recently, and one of them for example would be the idea of taking a recording of a string quartet or a piano trio and to be able to take an instrument and place it [in a stereo mix]. I think that’s almost possible now, but it’s something that the company that’s working on that technology hasn’t applied it to. They’re looking at how to remix a rock music track from just a stereo master by using artificial intelligence. For example, I can take a song and remove the bass, or increase the volume of the vocals, or change the basics of the mix, bring up the guitar, etc. That’s where the market is and I don’t see them yet addressing classical music. That would be fun to try and do at some point. These technologies are floating around and it comes down to the question of where the people who developing these technologies see their market. And I’m afraid that our small corner of historical classical music restoration is not really one that is big enough.

TS: True enough. One just needs to look at the death of record stores in past decades.

AR: Ironically, that kind of helped us. At first it initially nudged, and then forced people online. You can imagine that our customer base is at the older range of the spectrum, and not necessarily people who would have looked for their music online until they had no other alternative. We used to get record shops contacting us after someone read a review of our CD in such-and-such magazine, and [asked] how can we [could] get it to them. But that hasn’t happened in quite a few years because I don’t think most of those [kinds of] record stores exist anymore, and that is really sad.


TS: Turning back to music: what about so-called “pirate recordings” or air-checks? There are thousands of those floating around out there, at varying levels of audio quality, most of them really bad. What are your thoughts on restoring any of those?

AR: We have worked on some recordings that were privately recorded. It comes down to a number of things. First of all: what is it and who is it? And then, what is the sound quality, and if the quality is poor, is it worth [restoring]? If it’s [something from] a major artist and a very, very rare recording of them doing something that doesn’t otherwise exist, then maybe the quality threshold is a little lower than if it’s something like Toscanini doing a Beethoven symphony where there are dozens of [already-available] recordings. And then you have to ask: why do we want this specific one? These are all questions that get weighed when I’m trying to decide whether to go down the road with a particular recording. Something that came off shortwave radio and onto an acetate disk 80 or so years ago would generally not be of sufficient quality to bother with unless it was something very unusual.

TS: I’m familiar with some smaller companies who are releasing opera performances that, I assume, were recorded off the air and they’re often very interesting performances, but many of them are in very poor quality. I’m thinking of a Carlos Kleiber Elektra that’s been floating around and I thought could be something wonderful to restore. But I don’t know what the copyright issues might be.

AR: I know that there are other [companies] who are sailing close to the wind on copyright. But this is our livelihood, and if there’s any doubt in my mind, I don’t go there. The other side of it – sound quality – one of the things that people come to us for is [for] the highest-possible sound quality, so I don’t really want us to release anything that goes below a certain standard. There’s no shortage of recordings at a suitable standard that we can elevate to a higher standard, so I’m not looking to lower that, if I can possibly help it.


TS: That actually begs one of my future questions: how often do you run into a situation where you start a restoration and realize it just can’t be done?

AR: It becomes very apparent very quickly, but it’s very rare. This would normally happen with something a collector has sent to me, and I wouldn’t get very far into the restoration process before saying that [it’s] a non-starter. It would usually take me five minutes to tell you that something was impossible or not worth it. And if passes that five-minute test then I’m 99.9 percent sure that it will be a viable recording.

TS: You talked about emerging technologies. As restoration technology changes and improves, do you ever think about revisiting some of the older work you’ve done?

AR: This is a question that comes up from time to time and my general policy is not to do so. I think you can end up going round and round in circles with that approach. If we ran out of new material, and if I didn’t have months of weekly releases programmed out ahead with no apparent shortage of things to work on – well, maybe. But I don’t really want to go and do that. The exception to that rule was some recordings we [recently] released of Sir Adrian Boult conducting. I wanted to [reissue] a specific collection of recordings and one of them was among the very first 12 titles that we released back in 2005, and the new [collection we brought out] this year didn’t make sense without that as a part of it. [That said, the original 2005 release was sourced] from some 78s but I redid it from a very good vinyl transfer, so it was coming from a different source second time around. I don’t think it’s fair on customers who want to hear something new who have already bought this stuff, just because I can maybe make a marginal improvement on something.

TS: Do you have some specific favorite performances on Pristine?

AR: This is a question [where] I need to go away and think about [it] because my mind immediately goes blank. We have over 1,100 releases and to select any specific one is really difficult. There are some recordings that are special for sentimental reasons, going back to the formation of the company. The Edwin Fischer Bach Preludes. But it’s a very difficult question to answer..

TS: So, when you’re not in front of your equipment, do you listen to music?

AR: I do. I listen to a very diverse range of music. I listen to classical, jazz, rock, alternative rock, I’ve got very broad taste – outside of modern pop music and rap.

When I’m not doing [restorations] I have a little sideline. I’ve worked with a number of rock groups around the world doing mastering of new recordings. They are small, independent rock musicians and bands and if it’s something that catches my ear I will maybe ask them if they would like me to master an album or single for them. And it’s very, very different to this kind of [restoration] work and it’s interesting; so I get to listen to a bit of that. I wish I could have recorded Miles Davis, but alas, that wasn’t possible.


TS: Is there a performance or an artist you wish you could lay your hands on?

AR: Good question. There are gaps in repertoire of certain musicians. Along with Misha Horenstein, I would love to complete the Mahler symphonies with Horenstein conducting. A few years ago, we managed to find a 5th symphony which had never previously been issued and since then we’ve managed to issue two more. Stuff like that would be nice to find, but as far as anybody’s aware, no recordings exist, But then, we’ve heard that before and you never know.

TS: I guess you never know what’s hiding in the archives of radio stations and orchestras.

AR: Well, this is the question. You never know what’s hiding in the archives and it’s such a vast field and it’s hard to know what you might be missing. Completing missing cycles like Horenstein’s Mahler would be wonderful. But it may never happen.

TS: On the subject of imaginary recordings, I would have loved it if Mahler had been recorded as a conductor.

AR: (Laughing) Well, when you get into this sort of speculation you start saying, “it would have been nice if Beethoven had recorded this…” But more seriously, I think about Debussy, who recorded some piano rolls, but not actual recordings. I would have liked to hear him play his own music on an acoustic [pre-electrical] recording.

TS: I believe that Brahms actually did make an acoustic recording.

AR: The problem when you go that far back is that it becomes very, very difficult. The earliest recording we have dates back to 1899. It’s a military band playing Gilbert and Sullivan, and at that age, restoring those recordings is hard work. Classical disk recordings don’t really take off until you get Caruso in 1902. Prior to that there was very little, whether disks or cylinders; they were considered novelties.

TS: I have a very general question for you and if it’s too vague, please feel free to let me know. Any regrets? Anything you wish you had done or that you wish you could do?


AR: I don’t think so. If I haven’t done it yet, I feel like I can still do it. I regret that the copyright laws were changed [which put more limits on what we can do], but, of course, that’s out of my hands. [Our work] is an ongoing project I’ve no intention of drawing to a close anytime soon. I enjoy it, and I think Mark enjoys it. It pays the bills you know; it’s our living. And if there’s something I didn’t do a year or so ago, then maybe I’ll do it next year. There’s no reason to have regrets, because I don’t think any of these recordings are going to disappear out of reach.

TS: I want to thank you for taking so much time out of your day to speak to me.

AR: It’s been a pleasure.


About the Author:

Ted Shafran is the president of, a Toronto, Canada-based IT solutions company. He has studied piano, music theory, and voice, and sings with the Toronto Mendelssohn Choir, the oldest arts organization in Canada. He is also a long-time record collector and audiophile.

Back to Copper home page

1 of 2