Music technology is stuck in the 19th century. The technologies that we musicians use everyday are simply digitized versions of past musical technologies —a video of an incredible performance, a score or musical textbook in digital form, or, even a lecture on music by a distinguished professor, freely accessible online. Such things may be convenient, but they aren’t new musical technologies. If technologies are tools created by humans to solve problems, then what musical problems do these technologies solve? The above technologies solve distribution and scarcity problems, not musical ones. In an ideal world, a young musician would have in-person access to the real versions of all these things. And besides, real-life experiences are far superior to digital ones!
A Definition of Music
Music is one of those topics that has been philosophized to death, and because I have no patience for that kind of thing, I will ignore existing definitions and supply my own practical definition. Music is the use of melody, rhythm, harmony, counterpoint, form, texture, arrangement, and orchestration to create an art object. A musical technology then is a tool that operates on the domain of music to solve a musical problem. If you agree with the above definition then you are likely to agree with the premise of this article. Some may disagree with the above because it doesn’t include sound as a musical element. Adherents of post-modern musical aesthetics as described by Karlheinz Stockhausen or composers of musique concrète would disagree strongly with the definition as provided. They would likely get rid all of those old-fashioned techniques in favor of a definition of music as organized sound.
What are Music Technologies
By this definition, almost none of the technology developed in the 20th century can be classified as a musical technology. MIDI, DAWs, and DSP algorithms are fundamentally about manipulating sound, not music. Music notation software is about engraving and editing the musical score and doesn’t meaningfully interact with the domain of music. But there are still some loose ends here. What about generative AI models that generate music? The models that generate audio can clearly be eliminated from contention. But what about AI models that generate MIDI or some kind of musical encoding format? For a technology to operate on the domain of music it should maintain all of the semantic meaning of music as defined above. When an AI model looks for patterns in its training data and generates data exhibiting similar patterns, the AI model certainly isn’t operating on anything resembling a musical domain. All of the semantic meaning of the music has most certainly been lost during the training process. And this is obvious to verify if you listen to any music generated by AI. It simply can’t understand music, because it can’t understand anything. It’s just a fancy machine that looks for patterns in data.
Examples of Music Technologies
Good examples of musical technologies developed recently are musical analysis toolkits such as the Humdrum Toolkit, and music21. They provide powerful analytical facilities for analyzing musical datasets. And because these are tools created specifically for this kind of musical analysis, they do a good job of analyzing the music while maintaining the semantic meaning of the musical data being analyzed. Though these toolsets can be really useful to some, the vast majority of musicians aren’t technical enough to be able to use them effectively. If a musician develops the technical skillset to be able to analyze music with tools such as these, they would be better off to build their own toolset. In building their own tool, they could define the musical domain model as they wish and perform any musical data transformations that they desired.
Music Technology for the 21st Century
Now that we are a quarter of the way through the 21st century, we should think about new ways to create musical technologies that genuinely interact with the domain of music in an intelligent way. We may be lagging behind, but one great thing about lagging behind other fields is that we could look to them for inspiration! There have been incredible advances in fields like mathematics with computer algebra systems, in engineering with computer-aided engineering tools, and in chess with engines like Stockfish. Why shouldn’t we musicians have similarly powerful tools?