The Guardian recently ran an article titled “Can You Make Any Kind of Living as an Artist?” and the first sentence stated: “With the exception of household names, most people in the creative arts need a day job to make ends meet.”  This is not exactly news to those of us in the musical arts – so the more practical question for composers might be: “How can I get more done in less time?”  Can a 21st century composer forced to work a day job hope to produce the output of, say, a Bach or Haydn?  The answer, surprisingly, may be ‘yes’.

Let’s look at are some historical examples of composing productivity that we can use as a benchmark.

During 1724 and 1725 JS Bach wrote one cantata per week for the Leipzig churches where he had recently been installed as Kantor. Each cantata was about 20 minutes of music and consisted of choral, instrumental and keyboard parts. Bach had to compose, notate, rehearse and perform this music each week, so his productivity, assuming a 40 hour week, was something like 2 hours spent composing for each minute of music performed.

GF Handel raised the bar on composer productivity in 1741 by completing the ‘Messiah’ in just 24 days. This famous oratorio runs some 136 minutes and has 259 pages in the original score and works out to a rate of composition of almost 40 minutes per week. If we assume he worked 8 hours per day, this calculates out to about 1.4 hours of composing per minute of music. This does not include rehearsing or copying out of parts – so Handel and Bach were probably about equal in terms of composing efficiency.

Josef Haydn produced some 340 hours of music over a 42 year career. If we assume he did this in standard 40 hour, 5-day weeks, we get an output of something like 9.7 minutes per week – requiring an average of a bit more than 4 hours of composing per minute of music. Notice that Haydn worked at a rate about half that of the most skilled Baroque composers – but he is still considered very prolific.

As an experiment in 1998 I took one week of my vacation and tried to write a Baroque church cantata. And to my astonishment I actually succeeded in producing 8 minutes of usable material. Of course my 8 minutes weren’t as good as Bach, but it was performed during a church service very much as Bach would have done. Throw in a couple more hours for choir rehearsal and my composing efficiency was about 5.25 hours of composing per minute of music. Not a lot worse than Haydn!

But here is the thing: when I was writing my 8 minute cantata I noticed that only about half the time was spent actually creating music. The rest was spent breaking out parts and checking them, transposing for various instruments, making copies and organizing the pages into a rational format, etc. In other words my composing efficiency was cut in half because of the requirements of performance.

21st century composers now have the capability of realizing and delivering their music electronically – there is no need for notation and performance – and there are computers to increase our productivity over 18th and 19th century practices. I have been composing electronic ambient music for about 2 years. I do this by a combination of notation and processing – this music is not written for performance. In the first half of 2012 I have produced some 5.5 hours – realized by PC and delivered to the Internet. I have a full time job but even so I am producing over 12 minutes per week. So my composing output is something like 25% greater than that of Haydn. Now I’m no Haydn – and although I like my music it’s possible that I’m actually making 12 minutes of crap per week. Even so, I will only get better at what I am writing and can do so knowing that I don’t have to be a full-time composer to achieve a historically high output.

Contrast my situation with the composer who writes for performance – his efficiency will be only 50% of what I can achieve, and the number of new pieces performed each year for even a busy, well-connected composer is likely to be in the single digits. So his progress is restrained by the slower pace of writing for performance and his art will take longer to develop. Performance is presently deemed the successful end result of the composing process – but the lack of performance opportunities and the efficiencies to be gained via electronic music would seem to be compelling for those of us who are increasingly composing part-time by necessity. Sooner or later those among us with real talent – and the inevitable day job – will be working as I am, reaping the benefits of improved composing efficiency through electronics.

The most dramatic effect of the Internet on the art of music, therefore, may be the breaking of the historical chain of composer, performer and listener.  Music will henceforth be composed primarily for listening – rather than for performing.

What do you think?

14 Responses to “Is There Such a Thing as Composing Efficiency ?”
  1. adam says:

    As performing is the most interesting part of music for me I consider this to be a tragedy.

  2. Kraig Grady says:

    Can’t say that composing for the recorded medium is any faster. One has to think of different things and much more detail than with a score. I also side with Adam though in that people performing with and in front of others is the most satisfying environment for music.

  3. Greg says:

    Efficiency needs to be measured with quality included – it is a trivial matter to produce years of music that is awful with a computer. But if we think of efficiency as the production of music as good as a particular composer can make, then computers can be very efficient at least at the level of generating notes and sounds, if not so much at the level of articulating those sounds, where human performers are still far better, and ‘outsourcing’ that detail to the performers is far more efficient than trying to specify it with a computer. But for most people working with a computer is not as rewarding as working together with real people. Social animals and all.

  4. Not sure I agree with the mathematics you’ve got going there. And I don’t follow the relationship between music for performance and not. The mechanical stuff is trivial if you’re accomplished at the technical side.

    Also, you might have been able to look back on my piece about productivity … I did an extensive survey of 85 composers with detailed results that used to be available on NMBx but in their redesign apparently lost it. (If you’re interested and it doesn’t show back up on NMBx, I have it in .doc format.)

    My own productivity (efficiency) actually went up during “We Are All Mozart” — 100 pieces in one year, all commissions. It’s a matter of practice, I think, and the whole score-parts-performance deal is pretty much irrelevant. That’s busy work and during the process new ideas are in the mix. If you’re working on one piece and one piece only, then you stop one and start another. If you’re working on multiple pieces at various stages, then all that copying and proofing is just background chatter, a kind of noise out of which new pieces arise.


  5. Greg says:

    Hi Dennis – I would love to see your research on productivity – and it does not seem to be available via wayback machine

  6. Greg, give me your email address and I’ll send it to you. Follow the link to my webiste & you’ll find the contact form.

  7. I just realized that I had set up that productivity survey on my own site after it got edited in NMBx. So here are those results:

    The raw data was removed on request of one of those who filled it out. A ton of interesting comments and descriptions are included in the quotations in the results.


  8. Paul Muller says:

    Excellent comments.

    I believe that the human element in musical performance cannot be imitated electronically, and agree that it will never be replaced. Tragedy will be avoided because it will always be possible to hear a symphony by Beethoven live just as it is possible to watch a play by Shakespeare in the Globe Theater in London. But as live opportunities decline in frequency for both performers and composers, the centrality of this outlet to the art form will be reduced accordingly.

    Quality certainly should be considered in any discussion of efficiency, but I think we could be agreed that just because music is created quickly doesn’t mean it can’t be good. I expect there is plenty of indifferent music that was painstakingly written. Ultimately the skill of the composer will emerge, whatever the methods.

    The point Dennis makes about the nuts and bolts of notating and assembling a piece for performance is a good one – it may get faster and more routine the more often you do it. If Dennis wrote 100 pieces in a year – at an average length, say, of 10 minutes per piece – then his output over that year would be on the order of 20 minutes per week, not unlike that of the baroque. It would be interesting to know if this was done as a full-time effort or as something less, and to what extent the notation was by hand or automated.

    In my day job, we used to have six or seven drafting tables with people making drawings that they would give to the machinists who made the parts on a lathe, more-or-less by hand. Today we have one guy who draws the part on a computer and uploads a file into the programmable machining station. The efficiency has been increased by an order of magnitude. If our furniture is no longer made by hand, should our music?

    It would seem that the same thing will happen to the production of music, which is presently organized around production practices and criteria for success that date from the 19th century. A career in composing may eventually be defined by what one can do in 5 hours per week and not 40.

  9. Paul, it was a little more than 12 hours of music that could be roughly timed. The shortest is a minute, the longest 30 minutes. Some are flexible time, and a few are performance pieces with no specified time (including a book of 99 performance pieces as a single commission). There were 107 in all (The “We Are All Mozart” 100 plus an additional 7 for other reasons). The smallest orchestrations were solo pieces, the biggest a seven-minute piece for large orchestra.

    The commonality with Baroque music was the speed and the premise that I was in service to the folks commissioning the music — they would specify length, orchestration, etc., as well as some idea of the approach the music should take.

    Where it departed was that the range of music was more extreme than Vivaldi had to face with his melange orphan ensemble: electroacoustic (about a dozen) and one for just Midi; solos for cello, ukulele, trombone, soprano sax, theremin, violin, horn, flute, marimba, extended voice, accordion, piano, double bass, organ, clarinet, viola, tenor guitar, tenor pan, jew’s harp, alto sax, bass clarinet, piano with jingle bells inside; duos for voice/piano, cello/guitar, bass clarinet/bassoon, 2 celli (four of these), mandocello/piano, tenor pan/cello, trumpet/organ, cello/contrabass, 2 clavichords, piano four-hands, viola/bass clarinet, organ/computer; trios for three horns, three oboes, voice/anvil/piano, violin/cello/piano, voice/flute/guitar, voice/clarinet/vibraphone; quartets for voice/violin/viola/piano, string quartet, viola/horn/cello/piano; pieces for five guitars played by one person, clarinet & string quartet, brass ensemble, flute/bass clarinet/violin/cello/piano (that was the 30-minute one), several a cappella choral pieces, and orchestra.

    As for how they were written: Some were sketched on paper and input. Others were through-composed in Finale (I’ve been using it for 19 years and it’s like an instrument to me).

    And to the point on efficiency, some of these were being prepared for performance as they were being finished and others were being written. The ‘nuts and bolts’ simply had to be available to me, whether in mental pen & ink or via notation software.

    Finally, the question of quality. This part of the conversation bothers me. How can this be assessed? No, I’m not suggesting a bendy everything-is-okay approach. Do we overlook Beethoven’s early crap or Bach’s stuck-in-the-past last 20 years? Forgive Carter because of his age? Consider a young composer ‘practicing’ on the notation software ‘instrument’ to be developing? I just don’t know how to think about it other than to use my gut. I know those few of the 100 pieces that were weak and which were really good. Were they rough or not? I think almost all were polished in that internal speed with which artists can operate under pressure.

    Anyway, I gotta go make dinner. Still thinking about it 5 five years later….

  10. PS: It was not a full-time effort. I was still head of the Vermont alliance of country stores, went to the CMA conference in NYC, kept the house warm and the horses fed and the snow shoveled, built steps & railings for the house, consulted on a Linux audio project in California, was official photographer for a week-long chess camp, produced a percussion concert with Michael Manion, spent a month-long residency at Binaural Media in Portugal, engraved an orchestral score for Christian Wolff, attended several premieres of the WAAM pieces, and started the research for my “Country Stores of Vermont” book for The History Press … and joined Facebook. (And you bet I keep a freakin diary!)

  11. greg says:

    I’m enjoying this discussion and thanks for the research Dennis. re ‘quality’ I agree with where Dennis seems to be coming from – the composer is the one to make that determination (for the current discussion anyway). From that point of view, a technology helps composer productivity if they compose more with the same resources and are just as happy with the new compositions as with the old. Maybe even happier, but certainly not less happy.

  12. Kevin Clark says:

    I think we’re missing the metrics we should really be using: Excitement, attention, and love. Hours spent working is a huge thing, and minutes of music produced, certainly. But while those things are easy to measure in numbers, they don’t bring us very close to any way of aggregating the transformative artistic experiences that we’re trying to create.

    Maybe instead of minutes written we should be talking about adding up the enthusiasm for any given collection of music, spread out among the audience. Maybe we’d count someone who had a completely transcendent experience listening to it (whether live or online) a 1, and someone who didn’t even listen to the piece a 0, and rate everything in between, then add that up for the whole audience. That would probably give us a more helpful number to measure “output” than minutes of music or number of pieces. Comparing that number to hours put into writing would probably be more enlightening.

  13. Kevin, the question was about composing efficiency. The soft issues you bring up are the most often discussed among artists in a public forum, whereas hard issues of productivity, efficiency and money tend to be avoided, or swing into diversions about, well, ‘excitement, attention, and love’. Efficiency and related topics are actually critical to a composer’s day-to-day, don’t you think? How do you deal with efficiency?

  14. Dallas says:

    “…new pieces performed each year for even a busy, well-connected composer is likely to be in the single digits.”

    Do you mean less than 10 performances a year?

    Any composer with a few friends and even mediocre music writing skills should be able have this happening by the time they are out of their master’s degree.