Natural Language, Computer language, and Music

I have been contemplating the similarities and differences between natural language, computer language, and music. In all three cases there is a dichotomy between the aspects of performance and notation. In the case of natural language, there is at least one case of one notation informing three different performances, that is, the classical form of pictographic writing is shared between Chinese, Japanese, and Korean although the spoken languages are totally different. A similar case is that of a higher level computer language that can be compiled to run on different hardware cpu architecture.  In the case of musical notation, while the pitch and duration of sound is firmly established, many aspects of the intonation of the music is left to the interpretation of the musician.

Although the three examples above have a number of similarities, they also have a number of significant differences. For instance, music has a rich emotional semantics but its abstract intellectual semantics is somewhat constrained. On the other hand, while mappings between computer notations and intellectual semantics abound, code is notoriously devoid of inherent emotional semantics.

Natural language brides the other two examples by providing adequate mechanisms for both intellectual and emotional semantics. It is interesting that in so many cases those that embrace the exploration of one of these three media so often dabble in all three. This has not always been the case. Only recently has fundamental computer programming literacy become common enough that many, if not most, educated adults are capable of writing simple programs in some computer language.

Considering these three as separate domains becomes somewhat confused when you consider that you can describe music and programs in prose. You can write programs that produce music and text for that matter. While I haven’t heard of music producing prose or code per se, it is certainly not much of a stress to imagine some form of experimental mapping that might do just that.

All three domains are subject to detailed formal syntactic and semantic analysis. I plan to tabulate the similarities and differences between these domains and will share it with you when I have done so. I expect to find some more subtle correlations between them.


Sweet dreams, don’t forget to tell the ones you love that you love them, and most important of all, be kind.