Stayin’ Alive: Preserving Electroacoustic Music

Max Patch

In preparation for a performance at an electroacoustic music festival this weekend, I’ve been revisiting a slightly older piece that hasn’t been performed in its live electronics version for a few years. Several updated releases of Max/MSP software have come and gone since I last fired up the interface for this piece and, as one might expect, when I tried it out for the first time, it didn’t work correctly. Fortunately the changes needed to get it properly up and running were small, but there was still a significant time suck involved. And indeed, unless I keep on top of software upgrades and changes, pieces that incorporate live electronics could end up hopelessly broken for no reason other than years pass and technology changes. The same is true for much older electroacoustic compositions that involve very specific pieces of gear—stomp boxes, synthesizers, a particular delay unit, or a drum machine.

Because the way I am using the software in the case above is relatively straightforward, I could easily transfer my performance system to more accessible commercial software like Ableton Live or Logic; certainly more people own those programs than Max/MSP, and it makes sense to do so, to ensure that the music continues to be performed. But eventually there will be different software that replaces those programs, and what will happen then? Chances are I’m the only one who is ever going to undertake such a conversion task, and when I’m not around anymore, the performance materials for these pieces may well turn to dust. It’s a question for anyone creating electroacoustic music. What on earth is going to happen to compositions that are painstakingly crafted for effective live performance at the time of their creation, but which become increasingly difficult to mount live, simply due to the march of time?

A good clarinetist friend has taken it upon himself to recreate a number of older electroacoustic works that were originally made using equipment that is no longer available. He has resuscitated Thea Musgrave’s Narcissus, for example, translating the electronics from a very specific gear list into a more modern software situation. But I seriously doubt many people out there are really willing to take on projects of that nature, and I worry for the life of a lot of truly wonderful music. David Tudor’s Rainforest can’t really be recreated (darn good thing there are recordings, though there’s no replacing the live experience), nor can so many other works that rely on custom-designed electronic gear or software. It’s not just an issue for works that are considered classics now; there are many, many artists making great work incorporating live electronic performance today that should be able to be experienced by people decades from now. How is that going to happen?

This is an issue for individual composers and publishers alike, since it’s not any easier for the big publishers to deal with the sale, rental, and distribution of live electroacoustic music. It’s difficult and inconvenient to wrangle performance materials of this type, no matter how you slice it. Although I know that publishers do not encourage (and may actively discourage) their composers to create electroacoustic music, what I would really love to see is some sort of technology manager/archivist position become standard at publishing houses in order to deal with these issues going forward. Because like it or not, a lot of contemporary music involves technology.

Another argument could be presented that addresses the sheer volume of creative work that is being produced at this time. Should this music even be preserved for continuing performance? Is documentation in the form of audio recording and/or video enough? There is so much of everything. It’s easy to dismiss work and not be uncomfortable with its erasure because of the volume of creation and recording in this era we live in. Should these performance experiences be viewed more as ephemeral events from a very specific place and time, an expanded view of site-specific work? Is it possible that there could ever be electroacoustic “war horse” pieces that continue to be performed centuries later?

22 thoughts on “Stayin’ Alive: Preserving Electroacoustic Music

  1. scott worthington

    I wonder if the sort of score prefaces that Saariaho’s and Nono’s works have, for example, might be a bit of a model, where the specific technical processes are described along with list of equipment. Perhaps publishing code or similar would help with live computer music. Then if someone wants to revive a piece years after the software has become obsolete, there is enough information to reconstruct what once was. Any playback files should be able to last for some time as wavs or aiffs.

    Reply
  2. Benjamin Mawson

    I understand this familiar problem also, only too well. in my case the issue was not MAX but files generated in older (non-transferable) DAW formats like Cakewalk’s *.wrk. Thousands of sketches and dozens of finished compositions that used to run on CW, a pc and old modules now have to be comprehensively remade in Logic to be heard.
    My question though is that if there is relatively little real-time intervention and we are working in effect with the finished object, the sculpted sounds themselves, perhaps the whole distinction between performance and recording is outdated by our own compositional practices?
    If we can create the “recording”, the heard work, exactly as we intend it to be heard then, barring inevitable vagaries of ambient acoustical properties of the audition space and quality, quantity and placement of loudspeakers, surely this is sufficient?
    WIth a variety of tools like Sonic Visualiser and our own bespoke transcription systems – I do hybrids of scores and annotated screen captures – we can represent the work’s sound.
    Sure, this doesn’t facilitate reproduction in the way of the old notated score (whose ability to convey compositional intent has also forever been hotly debated) but perhaps the score’s function has changed, from one for purposes of reproduction to one of alternative representation, conveying additional information that might not at first hearing be perceived?
    I would welcome a debate on digital music preservation in non-digital formats and to hear what other composers and sound artists think the value of the work’s graphical representation might be in this.

    Reply
    1. Dennis Bathory-Kitsz

      Cakewalk .wrk files don’t need to be rebuilt — they load fine into the latest iteration of Sonar. Unless some custom hardware has failed, they will function just as well as they did two decades ago.

      Reply
  3. Kyle Gann

    Yeah. I was trying to analyze, in 2009, Robert Ashley’s early operas via the 1980s software files they were preserved in. The original machines the music was played on were long gone, and I found myself staring at strings of numbers or random rhythms that didn’t make any sense compared to the recording I was hearing.

    Reply
    1. Benjamin Mawson

      I quite agree, Kyle, with your implication that reduction to the bare data of original instructions leaves us with less than the essence of the intended artifice.

      After many attempts recently to literally translate a composition into a set of ‘instructions’ for the sound, I found too that the closest I could get was exporting the MIDI data into an impossibly long text file, followed by a specification of sample timbres, also as data.

      Could this be read with the ease that we translate Latin epitaphs or, more likely, the difficulty we face in deciphering Saxon and Pictish standing stone inscriptions?

      The latter , I suspect.

      It reminds me too of the forlorn hopes of NASA in the 1980s, sending recordings of Elvis along with pictures of a Pythagorean triangle into space as proof that intelligent life is waiting to make new contacts with other worlds.

      Elvis? Earthling Geometry?

      What can our conception of meaning and significance possibly convey to those whose frames of reference are entirely other?

      Back to music tech problems – should we be thinking about a more robust protocol than MIDI in considerably more serious terms?

      And the far greater difficulty of translating complex, multi-channel audio into a format that can be (if not recouped) read in some way?

      Who remembers the brief television celebrity of the 1980s who went onto chat shows to read the grooves on 33 rpm records and describe the content of them to a keenly fascinated audience?

      He seemed to be able not only to differentiate style or genre but musical detail, simply by close reading of the impressions upon a vinyl disc under magnifying lens.

      Maybe we could all learn new reading skills?

      Meanwhile, how about Berners-Lee’s brilliantly simply HTML solution – still effective, with developments that don’t outdate earlier iterations whereby code representing meaning and presentation can be simply transferred from platform and operating system and user and language to other, without critical loss.

      Imagine if a parallel to HTML (for text) could be derived for both recreating and for representing the quality of sound.

      Who’s up for starting the research and development on a universal, obsolescence-proof language for transcribing sound?

      Reply
  4. Dennis Bathory-Kitsz

    I’ve been trying to keep my own stuff preserved from the first electronic pieces in 1969. The hardware (not just digital, for sure) had full maps of how to build the sounds, but the hardware has already begun to fail. So several years ago I did VSTi versions of the instrument, but even those will vanish as software goes through a few more cycles.

    I’ve simply given up trying to preserve any of the software/hardware-driven projects. In Bocca al Lupo, for example, used an array of six small computers and my firmware, but the firmware self-erased in the EEPROMs over the years. http://maltedmedia.com/people/bathory/bocca/

    So Bocca (and several other projects) are simply gone. No point in worrying about it; only the famous will get preserved anyway, right? And even those (considering Kyle’s comment above) will have to be very lucky.

    The CEC has been involved in this for a while. Here is a special issue of eContact from 2009: http://cec.sonus.ca/education/archive/10_x/index.html

    Reply
  5. Dennis Bathory-Kitsz

    PS to the above: Back in 1977, I produced a series of concerts at the New Jersey State Museum, one of which was David Behrman and his “Figure in a Clearing” … a piece, running on a KIM-1 interfaced with a homebrew synth, that has little chance of survival without substantial curation. I remember the setup was incredibly delicate and difficult, but what a gorgeous piece!

    Reply
    1. Matt Sargent

      David has upgraded most (if not all) of the Kim-1 works to MAX. As I remember, he has a single patch that would run nearly all of his “vintage” pitch follower pieces.

      Reply
        1. Matt Sargent

          I’m sure that you could contact him about the specifics of the new patches. I worked with him in the summer of 2011 and his 21st century MAX work had (to me) all of the lovingly homemade qualities of his 1970s music.

          Reply
  6. John Driscoll

    I need to correct your quote:” David Tudor’s Rainforest can’t really be recreated (darn good thing there are recordings, though there’s no replacing the live experience), nor can so many other works that rely on custom-designed electronic gear or software”.

    As I’ve been performing Rainforest IV since 1973 as part of Composers Inside Electronics with David Tudor and after his passing, there has never been a problem recreating the work. It has been in regular performance now for a total of 40 years with over 125 performances in over 40 cities worldwide. In fact, performances are being planned for NYC and Wesleyan Univ. for 2014. As part of the David Tudor Trust we are recreating a number of David’s works including: Rainforest I, Microphone, Pepscillator, Pepsibird, Anima Pepsi (with others to come in the future). These involve a combination of hardware and software solutions.

    I agree with you that we are in an interesting time with the challenge as to how to perpetuate electro-acoustic works. Some can be recreated (as above) and others will simply pass away. With some of Tudor’s works we have created software to replace antiquated analog hardware, in other cases we are rebuilding some of the hardware to take advantage of new technologies. We are also going back into the Tudor instrument collection at Wesleyan University to use original equipment. Each requires a unique approach.

    In David’s case the new “Art of David Tudor” recordings provide wonderful insights into the sound materials and performances, but do not convey the spatial character of a number of the works. Tudor often used between eight to thirty seven output channels. The Rainforest IV recordings you mentioned were recorded in Binaural to try maintain the dimensional quality of the work. Ultimately live performance is the best way to perpetuate his work.

    Reply
    1. Alexandra Gardner Post author

      John, thank you for your comment, and for setting the record straight. I stand corrected, and look forward to more opportunities to experience Rainforest live. I agree that the binaural recordings aren’t able to capture the total experience, and that live is the way to go for the work.

      David was so fortunate to have people like you and the other Rainforest folks to keep it going with so much care and attention! Thanks to you all for that.

      Reply
  7. Phil Edelstein

    OhI can’t resist to rumble in here as well. I’m not much for nostalgia but there’s some very substantial material in so called dead electronics and obsolete computer languages. Let’s try to strip away the idea of newness, of oldness having any innate indicator of value. To the extent we can find the time and the path to keep electronic works of value alive, let’s make the investment. There is a bit of an contradiction in live electronics in any event. The liveness was always in the discovery of the composers, designers, players and the audience . Always seemed to me to be the compositional act that animated the circuitry – that is the arrangement . the animation was from the creative discovery possible from the innate properties of a circuit or an algorithm and the act of playing. Though, i’ll admit that some works will hold up better under translation than others.

    The essence of some works transcends their original technology implementations. Beautiful resonances that can be found in a Chaladni plate seem variously interesting if excited by bowed with catgut, analog circuitry, digital doodads or tweaked by genetic driven algorithms that go off searching for resonant nodes based on visual feedback.

    Isaac Bashevis Singer’s speech at the Nobel Banquet, December 10, 1978
    ….
    ‘People ask me often, ‘Why do you write in a dying language?’ And I want to explain it in a few words.

    Firstly, I like to write ghost stories and nothing fits a ghost better than a dying language. The deader the language the more alive is the ghost. Ghosts love Yiddish and as far as I know, they all speak it.

    Secondly, not only do I believe in ghosts, but also in resurrection. I am sure that millions of Yiddish speaking corpses will rise from their graves one day and their first question will be: “Is there any new Yiddish book to read?” ‘

    Reply
  8. Pingback: Stayin’ Alive: Preserving Electroacoustic Music | NewMusicBox | Jeff Albert's Research

  9. ryan ingebritsen

    I think the elevation of electronic performer from “technician” to “artist” would somewhat help facilitate the convention of being able to reconstruct older works as well as laying the groundwork for creating slightly more standard conventions for future works.

    In general, throughout history, there have been technological advancements that made certain pieces of music more difficult to perform. When the piano became the standard rather than the harpsichord, there was some work that went into translating the great keyboard works for the new instrument. But in comparison to our current situation, this was fairly easily done. As technology has become more and more available, and more diverse, the possibility of really understanding all technical resources has become completely impossible. No one person can anymore be that person that knows “everything” there is to know about electronically produced sound. But to have an identifiable set of “artists” who have the basic skills necessary to sift through all the possibilities, translate ideas, either notated, written verbally, or otherwise inscribed, and then find a way for those ideas to come to fruition for the specific situation in mind by a performing or presenting entity, would be a good starting point.

    More institutions need to support this kind of artist. It means more than just investing in “technicians”. It means creating the support systems that other artists (composers and performers) rely on to make it a serious and recognized artistic discipline. Publishing companies having professionals who deal with this on a regular basis could be a very valuable step, but I think first and foremost, the art producing (and more importantly, the arts funding) community needs to understand and appreciate the importance of that role. All music is becoming more technological weather we like it or not. Why not view this as an aspect of creating new music as important as that of a composer, performer, or conductor?

    Reply
    1. Dennis Bathory-Kitsz

      This is an excellent concept. With publishers and other institutions struggling, how would you make the case to do it? Technological curation is terribly expensive.

      Ever with software conversion, obsolescence sets in fairly quickly. Materials created as late as the 1990s are vanishing, and only a few artists have the personal wherewithal, institutional support, or fame to preserve their work and up-transfer to new media.

      And what about performance variations or even failure as modes of presentation? When teenagers can do in minutes what took me weeks in the 1970s removes the entire struggle and ‘dark side’ of electronic works. As mentioned with both Behrman and Tudor, some of their works are in software; I recall the Behrman vividly because we were on the edge of our seats with both sound and technology. Is something lost when the performance cannot fail? (I’m having a Ferneyhoughsian moment, I think…)

      Reply
  10. Michael Gogins

    If you use open source software (like Pure Data) instead of closed source software (like Max/MSP) your chances are better. It is a good idea if the software saves pieces in text format, also. Even if the text format changes, you at least have the chance of editing it to get it to work again.

    Unfortunately, nobody teaches artists or musicians about these things when they are getting started. Every artist I know goes back to earlier work at some point and wants to do something new with it.

    Commercial notation software has begun to use MusicXML which should remain readable by future notation software, but synthesis software is more problematic.

    I am a composer and a contributor to Csound, and I can say that the Csound community takes backward compatibility very, very seriously. Csound pieces are text files. Csound today runs all pieces that ran on the original Csound. This will continue to be the case.

    Reply
    1. FLOSS

      Exactly Michael.

      Pure Data is free software [libre] and I see no reason why – or how – it could one day just disappear. It runs on GNU/Linux and has incredibly powerful libraries, capable of doing almost anything MaxMSP can do. I don’t think most casual users are aware of just how powerful Pure Data can be.

      CSound and SuperCollider are great too. And there’s LilyPond even if you’re so inclined. These are all unstoppable and future-proof by the very nature of being free software. (FLOSS)

      Users of these applications should not worry about their music as they compose into the future. They’ll be fine. Safe and CSound.

      Reply
    2. Kevin Ernste

      ^THIS.

      And this too: the Open Source license means PD patches (and the software to run them!!) can be freely distributed along with the score, for example in a library copy. Yes, I know, “Max runtime”. The first time you have a performer or ensemble director who needs/wants to change a value they can’t access , you will know runtime is a cul-de-sac. And the non-text format means you are always a versoin change away from being dead in the water, as with every other piece of proprietary software.

      I also wanted to note Miller’s PD Repertoire Project (http://msp.ucsd.edu/pdrp/latest/files/doc/), an attempt at building a core set of modules for implementing old (and new) works. It has an events system, a score follower, and other common “objects” already implemented. There is a *long* ways to go and it’s been in a stand-still for years, but it’s an interesting start and points to a real possibility for implementation and performance, preservation, and future study of electroacoustic “rep”.

      Reply
  11. Matt Sargent

    I have recently become interested in FAUST (http://faust.grame.fr/). It’s an open source coding language that outputs into standalone applications, MAX or PD, VST, or others.

    Most interesting, the code can also be output in the form of a PDF, which provides a step-by-step set of instructions and equations on how the patch could be replicated. This information in the PDF is universal (only audio math or DSP information, avoiding any language-specific idiosyncrasies), so one could theoretically remake any obsolete FAUST patch in the future using other technology.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

Conversation and respectful debate is vital to the NewMusicBox community. However, please remember to keep comments constructive and on-topic. Avoid personal attacks and defamatory language. We reserve the right to remove any comment that the community reports as abusive or that the staff determines is inappropriate.