Lee Barry Composition (And Photo) Diary

In many ways, the method by which we now compose and record music, is very much like the editing of film. Film editing takes the linear experience of real life and rearranges the sequence of events to construct a story line, and places things in specific contexts (“reel” life). This is much different from real-time performance, where there are no second chances, and the only option is to do it right the first time. You could film a theatrical event and later go back, rearrange the sequence of events, include a soundtrack and you’d have something completely different from the actual performance. By simply adding a sound to a visual event you can change its meaning. Placing one event before another changes the viewers interpretation of the moment. And in music, making recordings gives us the illusion of things taking place in one moment, but in fact could have taken place days, weeks or even years apart.  [10/19/2026: This is exactly what's happening in the film (or documentary withi...

What was great about pre-20th century creativity was that the pace of technology was slow or even nonexistent. Things simply stayed the same all the time and was a stable environment for creating things—we didn’t feel like we had to be on the bandwagon about everything—have the Medium and the Substack, the Twitter following, and so on. All you need are the basic materials and perhaps a standardization of thinking. I know that has overtones of totalitarianism but what you want to do is to slow down enough to where you can work on one thing for several hours. In my experience, just warming up the creative process can take 3 hours, before anything gets racked up as being creative. In Beethoven’s time, people made things without technological change. Everything was local, and you had a stable atmosphere for creative work. I think we can still go back to something like that: where you have strictly local “sceniuses”. That’s probably hard to do now because things are all online and local com...

Back in the early 1990s, I was a fanatic for world music, and for a while, I ran a world music mail-order company called Intersect Music . It had several sections for different regions from around the world. It's now interesting to think about all that music in context with what's happened since the early 90s and through the end of the century: Rodney King, the first World Trade Center attack, Waco, the OJ trial, the Oklahoma City bombing, and so on. They were events that you watched on TV for hours and hours, all of which for me were in the background and were all unnerving--something I had never experienced before. For boomers, there was a palpable shift. What was particularly interesting to me was to think about how music can or can't be used--especially music that comes from Islamic countries. I always liked the music just for the music but there's there's a subtext that I have since realized--certainly post-9/11--that music can exist on its own and you could us...

There was an interesting study that was done a few years ago on the complexity of lyrics. But there’s a difference between those who write scientific papers about lyrics and those who actually write and sing them, which has an interesting connection with AI-generation of songs where the “singers” are just LLMs shuffling patterns to match riffs. When I work on a lyric, I’m focusing on how the quality of words to be used musically, especially at the rhythmic level. How syllables are mapped over note values is crucial. I think lyrics should be as economical as possible, which is the issue with translations from English because you’re “deracinating” the natural speech rhythms. With the Esperanto lyrics, the music sounds fine on its own with the vocals just being another instrument. If you don’t understand the words, then it’s essentially an instrumental. Most people don't tune into lyrics. The music is most important on first hearing, but lyrics can soak in over time.

Why am I so attracted to AI for generating material? I had been involved in generative music in the ‘90s creating music from seed structures, and I’m still interested in that, but sometimes I find myself wasting too much time going to websites, logging in, thinking of text prompts to generate music–when I can generate music from an idea or simply pick up a guitar or a bass and I’ll have an idea within seconds. Like this morning: There was a Bill Clinton quote in my diary on 9/22/2010: “They always like you better when you’re gone.” I heard a 12/8 minor blues in it. For the past 4 or 5 months I have been generating music from words and phrases that are in my diary and I jot them down in my Songday notebook. I wouldn’t necessarily want to generate music algorithmically as a first choice. It would be interesting to have my own large language model that would generate music based on the inputs that I give it, such as a rhythm or melody and see what happens. A lot of creativity is “Let’s tr...

Improvisation is a form of spontaneous composition that uses predetermined structural elements, usually chords and scales. Jazz improvisation is understood as being free-form, but is controlled with rules. In Arabic music, there is Taqsim, guided by a different system of rules for improvisation—and similarly with Indian Raga. The eastern versions of improvisation have an additional cultural element that is more important than the West’s focus on improvisation for art’s sake. Interestingly, Taqsim may possibly have some connection with algorithms (simply a series of rules or instructions). The word algorithm is eponymous with Algoritmi, a Persian mathematician. The fact that jazz musicians can relate to physics (and vice versa) is perhaps not just coincidental!             The curious thing about improvisation in anything is that there are always some rules or constraints that are not universal, e.g. a jazz player would be confused by the rules of Taqsi...

I have always worked combinatorially, i.e., each creative act is somehow connected to others, either spontaneously (when I am aware of them) or later when I associate them with existing works. This works the same for my music as well as art projects. I also have always kept lists of ideas, titles, strategies, and so on, which I go back to very often. Electronic notebooks, such as Evernote have made this easier (perhaps too easy), but I used to keep paper journals and sketchbooks. There are two basic ways to add cohesion to creativity: 1) create the work then serialize; 2) create similar pieces in an existing series. This also works in music where you can compose first, then add to a collection (album) or define the framework for the album and fill it in accordingly. I don’t think there’s anything that I do, either through design methodologies, or through improvisation that doesn’t fit with something I’ve done before. I know there are some artists who like to make something they’ve neve...

 I’ve always been restless and eclectic. I never really stick with one thing for very long. It’s not that I lose interest and never want to do it again; it’s just that there are too many other things to explore. You could say that it’s kind of impulsive and that I’m not thinking things through, but as an artist, I don’t think it would be very enjoyable or interesting to want to stick to one thing. There are artists who have only done one thing, and I admire their work. The artist I always cite is Peter Halley , who’s done essentially the same piece his entire career. I know lately he’s been doing other things, and there have been other singularly focused artists who have branched out. But I’m different because I do lots of different things and alternate between them, sometimes doing them all simultaneously. Over the past 5 or 10 years, I’ve kind of returned to skill, having spent quite a bit of time doing things that are more abstract and postmodern, which don’t involve a lot of sk...

I still think AI music is in its talking head stage; we really don’t know what to do with it. It usually takes about 10 years for a new technology to get its sea legs. Eventually, lots of people will be using it legitimately, unless the courts decide otherwise. In the meantime, there's going to be a lot of smaller shifts that will increase the options, such as choosing keys and meters. If there was an option to create a song in 5/8 or 7/8 it would be interesting to see what it does with language. Would people generate prog? Perhaps. Would you subscribe to Rush AI for $50 a month?

  If it wasn't for AI music I wouldn't have had an interest in writing lyrics. I have kept a running list of lines and couplets for decades but there was no incentive to use them. With the  Songdays, I had started to use them just for the speech rhythms, to be converted to instrumental riffs for use in classical music. I've always been a "singer-songwriter" who hated singing, and now I can experiment with many different ways of creating potentially singable words and phrases, drawn from prose, e.g. lines from letters as cryptic "black-out" poetry (or libretto), distilled to pop lyrics.  The source for this was a June, 16, 1895 William James letter.  

  I started experimenting with AI on some of my orchestral pieces. At first, it's interesting, but it's never what I want, and it's always the same clichés–the John Williams-style scoring.  It's always the case with AI-generated music that it's an approximation of what you might want. Sometimes it suffices, but it will never be the one that you originally thought of. With the Curios, I only want a piece of music that's a minute or less in duration, and AI extends it. It can extrapolate what you've already written and can suggest ideas, but you need to get into the weeds and do it yourself. I played a number of these for a friend, and he immediately liked them. I can see why because it's a complete work, and when you play mine against it, it pales in comparison. With the 250-series pieces, I’m scoring things only for brass ensembles, and AI is making a complete orchestral arrangement out of them. That's not what I want. I know that you could probably ...

  Sometimes I like to interact with AI-generated music to see what's under the hood. AI music is the new radio that musicians play along with, but they are also generating the songs. I think it's a great way to learn music and start writing. This is a pop song I generated from my lyrics in an early-mid 70s pop style--a style I grew up with on AM radio, so this is somewhat of a nostalgia trip. It's like the late '70s when I would turn on the radio and play along. So I attempted to put my own bass part on it as I would on any track. The existing part was almost imperceptible. My approach to bass parts is always compositional in nature. You can't ride the root all the time, and you have to think about the overall shape of the lines, what range they are played in, and how it is a counterpoint with the other parts. For pop music like this, you don't want too much below the staff, keeping in the low range of a guitar, then changing octaves for contrast. The playing o...

This is a passage from the book The Story of Stories where I’m swapping our “story” for “song” as a corollary between stories and song lyrics. I don’t think lyrics have to tell stories, but rather suggest them. Even if the songwriter explains what it’s about, listeners will find different stories which might be triggered by words, lines, or couplets. I like them to be about something if possible, but music has a top-down control on what you can get away with in a lyric, and also how the singer sings it. A bunch of random lines that means nothing in themselves can work well with the right singer. With AI you can experiment with this. It’s much harder with human singers because of the differences in vocal apparatus.  *** And that raises a question: Is a song what is written or what is listened to? This is not a new concern. Literary theorist [Roland Barthes[, among many others; wondered about it. [Barthes] concluded a song is not what is written but what is listened to; so much so t...

  To me, using AI in music is essentially a very conceptual thing, and is an extension of conceptual art—not mainstream. Music itself is a conceptual art even if you’re not doing conceptual work.   I just read an article about the Endel and Plastikman collaboration. Basically what they’re doing are remixes but they’re running them through neural networks and using Alexa Skills, which is interesting. But I don’t see it as that innovative. It’s people trying to be innovative and they’re covering the same ground of people who are now their grandparents or great-grandparents. If you consider the threads of classical music, we’re sort of still doing that as well. It’s a way to be conceptual-cool as a musician. I did it too at age 35.   The problem I have with AI music is that it doesn’t really give you anything to do over longer periods. Once the Beatles were using the studio as an instrument their time in the studio increased. If all they ever did in the studio was let machin...

Another distinction between AI-generated music and music that we make organically is that we go to the art store, we buy the paint, we buy the brushes, we take them home, and we make something with just those tools, which is drastically different from using some software or app that millions of other people are using. Take for example, there's one art store in a city of millions of people, and those are the only materials that anybody could ever use. What would the art exhibition be like with the results of millions of people using the same materials from that one store? Actually, I've been doing it in art for some time with my Black White and Red series. Sometimes constraints can make you more creative.

  Writing lyrics in the AI era is a new form of writing. Obviously, lyrics are a form of poetry, but ultimately, music shapes the meaning--if it has any at all. With AI-generation, the process of idea to a final mix is very fast--too fast in fact--which drives forms of writing that could become lyrics. It's another form of "the medium is the message": the medium makes messages possible things that weren't possible before.  It also makes it more difficult to find relevance: It may mean something to you because you wrote them, but is it a song as relevant as The Streets of Minneapolis in 2026--or any protest song? An AI-generated protest song is actually a strange idea and highlights the difference between AI and humans, primarily an awareness of a galvanizing moment. It can generate a protest song and use Springsteen's voice as a model and might actually have the same emotional valence, but there aren't 10,000 people gathered in a park listening to someone perf...

  If you’re a musician or artist for long enough various things will overlap with your music. What happened to me was that I studied jazz and classical then started exploring other areas: spirituality, psychology, and philosophy. Many of the electives I had taken in college were taking place at the same time as the music classes. For example, I’d have a harmony class in the morning and a philosophy class in the afternoon. So there was a slippage between the two and began to affect my lyric writing around 1990.  There was a piece that I wrote called Path of Salvation , partly inspired by the book Communion by Whitley Streiber which came out around 1989. I had not been interested in UFOs although I knew people who were interested in them, and I had drawn parallels between ufology and spirituality because I think they’re related in some ways. That was the first new influence that changed my approach to music. There were other things that happened along the way but fell by the way...

  AI music creates an absurd situation for artists and musicians. The longer you've been using it, the more you've gotten used to using it, and it's in the flow of your creative activities. But since we don't have control over the variables in manual art making, we are slaves to whatever options are available. The results are pretty interesting, and I like using AI music generators as I can experiment with lyric ideas. AI doesn’t parse them the way that I personally would. Since I'm a trained musician, I have lots of options. But when you use AI music generators, you have limited options, but the results are interesting, which is different from manual creativity, where you have a lot of options, and the results aren't always so interesting–and it's more work to shape them so that you can feel satisfied with the results.  Since the 1980s, I've juggled writing classical music and pop music, and I'm still continuing to do it today. it can create conflic...

Vellum Prints : Reality Frames My composition techniques are derived more from general creative strategies, or borrowed from other domains:  Frameworks/Concepts. This is my favorite way of being productive. Typically when I sit down to create, my intention is to work on something in progress, but I keep an ear open for little “happy accidents” and record them right away. I have hundreds of these little snippets. Eventually I go back and put them into the existing frameworks or concepts. Composition Strategies. I have been keeping a running list of Strategies since the early 90s, some of which I made into decks of playing cards. I made some other “virtual” decks that I use for the concepts. I have subsets just for jazz, that only use flat keys for example, and other attributes idiomatic of jazz. Feel free to use them and/or contribute to Cycles and Symmetries. What I call “cycles” are repeating symmetries in the harmony, rhythm or melody to create an armature from which to compose a...

  I am still of the opinion that too many gimmicky tools make creative activities seem empty. When the tool is designed to make one thing, it puts you on a creative hamster wheel. (The software preset is the electronic version of this, where the sound has only one use.) But when the tools are there to solve specific problems, then they become more useful. Better yet is to design the tool to solve the problem, then discard it. Metaphors are another example of a this kind of temporary heuristic. It's the lowest tech thing one can do with most payoff. "We used to varispeed the tape machines for tape echo. The original ADT [Automatic Double Tracking] we did was developed as a result. It was Ken Townshend who came up with a way of taking the signal off the sync head and then delaying it through other varispeeded recorders, so the gap between the two could be adjusted If you got it slightly out, you got phasing. But if you got it out to about 25 to 30 msec, it became a form of ADT....