Dynaxiom 3529

 

 Theory is a way to fully understand something, but not to the extent that you could successfully apply it. Applying knowledge requires giving shape so that you can examine it. You could have ideas for a sculpture, and you could know lots of theory on some other subject that could be applied conceptually, but attention has to be paid to that particular form, because that's what others see. In music, you can know theory, and you can apply it to form, but it's a question of theory serving the idea. 

I recall an anecdote about The Beatles discovering the B7 chord, and it opened up new possibilities. You could later learn that it's the dominant chord in E minor, and you could continue to study functional harmony because it's interesting, but it still has to be applied, or simply used in the flow of your work. Harmony is, in fact, functional and is ready to be applied, but how does it serve the art? Jazz is harmonic (or vertical) by its nature, and theory guides its composition, and improvisation rides over the top of it in real time horizontally, with theory also being applied in real time using the theoretical tool box of the player, who can decide if or when to use theory. 

Jazz wouldn't be jazz without knowledge of theory. AI-generated music requires no theoretical knowledge at all, and is (surprisingly) all art. But there's no reason to know anything. It can generate jazz, but there's no one who applies knowledge in real time. It doesn't matter to the listener because they are consuming it as art. The AI musical artist knows nothing about music, but knowledge is in how language (lyrics) shapes the music, even if made by a machine.

***

From the Dynaxiom LLM. It's nice how it summarizes the entries and groups them.

 

The following entries from the sources discuss the role of music theory, systematic frameworks, and the creative processes involved in composition and improvisation:

Music Theory and Systematic Frameworks

• 0055: Notes that while people spend time learning programming, few learn what "diatonic" means in music, even though such information provides useful linkages to other domains.
• 0150(05): Suggests that a musician should always want to compose as a way to explore the "operating system" of music.
• 0160: Observes that while jazz sounds like art, it is "really more like a science".
• 1231: Mentions that there are limits to the cultural usefulness of 19th-century counterpoint, voice-leading, and functional harmony in contemporary contexts.
• 1304: Notes that while the octave can be divided into ten equal parts, only certain intervals are "in-tune" and beautiful, existing infinitely in the electromagnetic spectrum.
• 1521: Argues that all sound intentionally constructed by humans shares a sequential logic, noting that "Theory has the answer" to why certain things sound harmonically wrong.
• 1576: Describes jazz as being about "shortcuts" or heuristics, such as the chord symbol.
• 1670: Discusses music education, stating that rote learning is about creating a sequence of patterns on a local map, whereas knowing the "operating system" of music (typical of jazz musicians) provides a universal knowledge of the overall system.
• 1783: Notes that while rock 'n' roll initially ignored formal concern with grammar, it eventually developed its own.
• 1790: Defines a scale as essentially a list, with chords representing a vertical stacking (portrait) and melody representing a horizontal selection (landscape) from that list.
• 1867: States that chord extensions function as ornaments for triads.
• 1929: Describes the "blueprint" for jazz as being comprised of chords and melody, allowing for the continuous transformation of seed ideas.
• 1991: Questions if jazz would still sound like jazz without its intricate harmonic rules.
• 2498: Asserts that modal music is not a free-for-all and still contains "wrong notes".
• 2508: Asserts that music theory provides the reasons why something might be wrong, and that avoiding it is often a form of passive aggression against rules.
• 2606: Mentions that learning a new chord like Dmaj7 allows a musician to unlearn the older patterns that previously created it.
• 2739: Notes that music is not necessarily in a single key but can have key centers that shift from bar to bar.
• 2740: Points out that diatonic structures do not automatically guarantee that something sounds good.
• 2774: Describes scales as the "connective tissue" between chord tones.
• 2998: Suggests that thinking in systems makes learning easier because one aspect is understood as part of a larger whole.

Composition

• 0157: Asserts that staging a musical performance should be like making a painting, though visual techniques are often omitted from music composition.
• 0164: Claims that "You don't write songs; you find them".
• 0458: Defines the process of music composition as essentially "making art that unfolds over time".
• 0742: Notes that composition requires 100% concentration rather than ostentation.
• 1398: Discusses the difficulty of writing a song that sounds finished the moment you begin.
• 1460: Defines composition as the "controlled current of sound and music".
• 1887: Explores the top-down control exerted when composing music from existing lyrics versus writing lyrics to existing music.
• 1926: Distinguishes between music heard as a style and the composition itself, comparing the former to interior decoration and the latter to an architectural blueprint.
• 1974: Suggests that composing music may be a built-in function of the brain, as structures often mirror natural patterns like fractals.
• 1982: Describes the composer's work as a process of exformation, creating a pathway to meaning from amorphous information.
• 2093: Discusses the traditional way of writing variations (theme first) versus leading toward the theme at the end.
• 2285: Describes song ideas as blocks of stone that are sculpted into a unique signature style over time.
• 2365: Compares writing music on score paper to solving a crossword puzzle.
• 2443: Likens composing music to designing a playground.
• 2738: Advises spending more time on the ending of a piece to inform the other parts.
• 2932: States that composition cannot always be driven solely by feelings.
• 3113: Reiterates that music is a matter of finding.

Improvisation

• 0070(05): Defines free will as a kind of "jazz"—improvising over a set of established "Changes" that cannot be changed.
• 0473: Notes that the jazz avant-garde attempted to break free of the grid of tonal music but created its own constraints, citing Charles Mingus: "You have to improvise from something—you can't improvise from nothing".
• 0606: Clarifies that jazz improvisation is not purely free like Abstract Expressionism but actually involves many rules and constraints.
• 1348: Defines improvisation as "taking set structures and using developed skills in rhythm, harmony, and melody to create organized sounds that comport with the set structures in a listener's brain".
• 1390: Discusses the serendipity and "spirituality" of improvisation, noting that the bandstand moment removes the over-analytical mind.
• 1448: Describes jazz as a vertical phenomenon informed by harmony, but notes it can also be completely linear, letting spontaneous counterpoints suggest chords.
• 1460: States that "Improvisation is extended composition".
• 1864: Asserts that listening to people improvising is equivalent to "listening to minds working".
• 1895: Argues that improvisation is effective because it uses rules as an "operating system".
• 1896: Compares improvisation to chess, being both strategic and tactical.
• 2376: Notes that free jazz often ignores musical structure or theory in favor of communion rather than listening.
• 2886: Observes that a musical score acts as a guarantee that a musician will not improvise.
• 2898: Notes that jazz is frequently a series of "happy accidents" that go unnoticed as such.