This essay is also available as a podcast on anchor.fm, Spotify, and other platforms
Hail and welcome. Over the month of December I put a great deal of time into working on a new music project, Cyrus Dark and the Symbols of Reverence, which I’ve mentioned a few times on the show and on my Patreon feed. During breaks between tracking parts and mixing songs, I surfed various musician subreddits (ostensibly as a way of keeping my brain in a musical mindset during a break; more likely it was just an excusable means of procrastination). I became fascinated in particular by certain patterns I was noticing in the questions posted on the music theory subreddit, patterns which matched up with sentiments expressed by many of my music students. I’ve been studying music and music theory for decades and worked primarily as a professional musician up until the pandemic, so this is an area in which I consider myself to be an expert, and I spent some time answering questions on the subreddit by way of correcting various misunderstandings about what music theory is and how it works. As I thought about it, I began to develop a theory about the nature of those misunderstandings, and that theory is the subject of today’s episode.
There are a few different ways to look at music theory as a field of study, and two in particular that I’d like to contrast: the first is general music theory, and the second academic music theory.
Let’s start with the concept of music. The question of what music is is an interesting philosophical topic on its own, but I think the intuitive understanding serves well enough for our purposes. The playing and enjoyment of music extends through every human culture in every time period, going back as far into history as we can see, and seems to be as much an intrinsic part of being human as spoken language. What is it about these particular patterns of sounds that leads us to treat them in this way, to hold them in such high regard and to categorize them separately from other sounds?
The answer, I believe, is intelligible structure. Sound itself has a particular physical structure, and our perception of sound, in terms of our physiology, has a structure as well. We perceive sound as the result of variations in air pressure hitting our ear drums, and perceive sounds as having different pitches based on the speed of those variations.
We can build on that structure in various ways, some of which hold closely to the physical structure of sound and some of which are more abstract. General music theory would be the general study of all possible musical structure as it has existed in all musical cultures across history and as it could potentially exist in the future. There are many common features shared between most musical cultures owing to the physical structure of sound and the structure of our perception, but much of the structure built on top of or around those common features differs from culture to culture, and a general music theory would include the study of all of those structures, and would also theorize about entirely new ways of structuring music. Bear in mind that this general music theory I’m describing doesn’t really exist as a field of study. The closest thing would be musicology, the general scientific study of music, but that also includes subdisciplines such as psychomusicology and sociomusicology which are concerned more with the perception and experience of music and focus on the structure of music in a more indirect way.
Someone studying general music theory would start with the physical structure of sound and the common features of human music and then would likely go on to specialize in the specific music theory of a specific musical culture or style. But academic music theory—the discipline of music theory as it’s typically presented in Western classrooms and textbooks—is a bit different. Academic music theory is, in reality, the specific music theory of a specific musical culture: European classical music as it existed over an approximately 300 year period, from about 1600 to about 1900, which is referred to by music theorists as the common practice period. Music theory, as it is typically taught, focuses on the specific musical structures common to that musical culture.
There are, of course, several significant problems with taking a particular musical culture—especially that of a historically colonial and imperialist people—and universalizing it in this way. The YouTube channel of musician Adam Neely—absolutely one of the best channels on YouTube—has a video about this, entitled “Music Theory and White Supremacy” (2020), which is based largely on the work of music theorist Philip Ewell and his paper “Music Theory and the White Racial Frame” (2020). I highly recommend both reading the paper and watching the video. The focus of both is, as the titles suggest, the racial implications of this universalization of music theory, and that’s an important point to make but not what I’m going to be focusing on here. What I’ll be focusing on is how music theory is approached as a means for musicians to abdicate their artistic voice.
Many of the questions I’ve observed in the music theory subreddit and in other music forums deal with what I’ll refer to as the procedural correctness of music. Academic music theory focuses primarily on a structural system called functional harmony, which is central to the music of the common practice period. I’m not going to go into the details of this system; what’s important for our purposes is just understanding that this is one particular way of structuring music out of a rich diversity of different ways that music can be structured. And academic music theory is simultaneously just one possible approach out of many possible approaches to analyzing that structure.
Functional harmony is a marvelous system of musical architecture, and one of the great things about it is that it provides a symbolic language that can be used to describe almost any music—and especially most of the music that is most popular today—with varying degrees of applicability, even if the music only loosely implements functional harmony or doesn’t implement it at all. That said, functional harmony does not, in itself, constitute a procedure for the composition of music.
There is a clear desire implied by these questions that music composition be a procedural process: that one begin with a selection of some musical parameter—key, for example—and then, having chosen the key, one has a range of selections of chords which are correct according to the key selection, and then once that first chord is chosen, there are options for the second chord which follow from the first, and options for the third which follow from the first two, and so on; and then, once the chord progression has been established, there is a procedure for writing a melody over it. In principle, this can be done without someone hearing what they are doing, and I think that’s exactly the thing that illustrates the difference that I’m talking about. Alternatively, one might begin with a melody, choosing the first note, and then procedurally chose the second based on the first, then the third based on the first two, and then, once the melody has been established, there is a procedure for harmonizing the melody with a chord progression.
But of the many musicians, composers, and songwriters I’ve worked with over the years, none have gone about writing their music in this way; nor was academic music theory intended to describe or direct such a process. This is then one of the clearest examples I’ve ever found of the process that the sociologist and philosopher Max Weber referred to as rationalization: the detachment of spheres of activity from the traditions of which they were once apart so that they can be refined through analytical procedures, even if the original meaning of the activity disappears in the process.
My own process for composing and producing music tends to be very bottom-up: I start with the drums, then write a bass line, then a chord progression, and add the melodic elements last. At no point am I doing this in any sort of procedural way. Certainly, every choice that I make is informed by the other choices that I’ve already made, and often my choices are informed by academic music theory as well, but at each stage, what I’m doing is much the same as how I, informed by the grammar of the English language, wrote the sentences of this essay: I start with some vague concept in my head, improvise a sentence to express that concept, and then type it. Even that makes the process sound more procedural than it really is: the conceptualization, the discursive improvisation of that concept into language, and the typing, are in fact all happening more-or-less simultaneously.
To be clear, approaching the creation of music in a procedural way is not in itself wrong or incorrect. Many of the greatest composers of the 20th century approached composition in exactly that way. I’ll name Arnold Schoenberg and Iannis Xenakis as some particularly notable examples, but even though they generated their raw materials in a procedural way, they were both clear that the ultimate deciding factor in compositional decisions was the ear and intention of the composer (cf. any of Schoenberg’s third period compositions, which regularly violated the underlying serial procedures; Xenakis, 1992). Furthermore, neither were operating within the confines of academic music theory, but rather created their own theoretical systems for the purpose of composing music in a procedural way. If it were one’s intention to do so, one could use academic music theory to compose music procedurally—for example, using Markov chains derived from statistical analyses of works in a certain genre (and I believe I’ve heard of this being done before); the problem comes only when one misunderstands music composition as being essentially procedural.
As an example, every day on the music theory subreddit you’ll see questions that fit the following template: “By way of studying music theory, I am analyzing the chord progressions of songs that I enjoy. I was analyzing this song and determined it to be in this key, but then a chord happened that was not in that key. Where did that chord come from?” I think at least a few times, the exact wording “where did that [chord, note, or other musical device] come from” has appeared in the post, and my answer is typically a snarky “It came from the person or persons who wrote the song.” My saying this deliberately misses the point of the original question, the implied intention of which is to say, “I understood this music as having been composed via this particular procedure, and then something happened in the music that seemed to me to violate that procedure; have I misunderstood the procedure at work here?” But this is a wrong question because, typically, no procedure was involved in the first place.
The regulars of the music theory subreddit will typically jump in at this point and offer the academic-theoretical language to describe what is happening in the music, but there’s a problem there as well, the conflation of naming with explanation. For example, say someone is analyzing a C major chord progression and was surprised by the presence of a D major chord, which contains the note F#, which is not in the C major scale and which thus violates the assumed procedure under which the note selection is drawn from the set of notes in the scale associated with the given key. Someone will respond, “Ah, it’s a secondary dominant, the V/V [pronounced, the five-of-five].”
Imagine that you, a friend of mine from some other country or planet, are visiting me, and ask me about cars, which you’ve never seen or heard of before. I show you my car and start talking about its usage and general operation. At one point, you say, “Wait, how does this big block of metal at the front of the car make the car move?” My response is, “Ah, that’s because it’s an engine.” You can see the error here: the engine does not make the car move because it is a thing called an engine; engine is just our word for that sort of thing, and so I really haven’t answered your question in any way. At most, I’ve given you a different way of asking it: “Okay, how does an engine cause a car to move?”
Going back to the prior example, a D major chord in a C major song is probably a secondary dominant, and its usage was probably informed by the songwriter’s study of secondary dominants in academic music theory, but not necessarily so. The Beatles, for example, made regular use of secondary dominants, but likely didn’t know that terminology or the function of those chords under an academic-theoretical understanding; it was likely just a sound that they heard in music they enjoyed and that they used whenever they felt it was appropriate, with consistently excellent success. Often, such as in the song “Eight Days a Week,” the Beatles would use chord progressions which simply cannot be reduced to an academic-theoretical procedure. The chord progression in “Eight Days a Week” begins with D major and E7. If you were to derive a procedure from academic music theory, the next chord would almost certainly be A major. There are other options, but G major, the chord that actually does come next, is at most, not likely according to the rules of that procedure. But there’s clear musical structure in that chord progression regardless, even if it doesn’t entirely match with the structure described by academic music theory. We can hear it. More specifically: that is what we hear when we listen to music and enjoy it or appreciate it in some way.
In music there is ultimately nothing beyond the pure experience of sound. All else is predicated on that or superficial to it.
Another example: in this sort of comment, someone is learning an instrument and accidentally landed on some sequence; a chord progression, for example. This chord progression struck them in some primitive way and they come to a music theory forum to try to get some information about what it is they heard. They might say something along the lines of, “I want to know what I was playing.”
The problem of course is that, better than anyone they’re asking, the person asking the question already knows what they’re playing. I think part of this is a very important need to find good language for new phenomena so as to facilitate conversation about them. “What is this kind of thing called?” But there also seems to be an implied question along the lines of, “I have found this thing but I do not know the proper procedure for using it.” Implicit in the way the question is asked is an assumption that this is not the real music, there is some other music beyond it. But that’s not the case. There is no other music. What this person has found, this experience of a series of chords or whatever, is not reducible to anything else. It’s sufficient in and of itself, and all the rest is mere description.
And I’m not trying to paint these as stupid questions. Given a certain (mistaken) understanding of what academic music theory is and how it works, they’re perfectly rational and evidence to an underlying intellectual curiosity about music, which is a good thing. Helpful responses in discussions I’ve read will often offer suggestions as to how these musical ideas can be used in the context of composition, songwriting, improvisation, or production. I’ve tried to offer such suggestions myself on several occasions. The thing is, I’ve found that, if the suggestions or the original idea aren’t reducible to an academic-theoretical procedure, the person making the inquiry might express disappointment, possibly even hostility. I’m not sure what is driving those kinds of responses, but my guess is that the person asking the question perceives the responses as telling them that whatever it was that they found isn’t real music, because “music theory” describes what music is—actually a perfectly reasonable interpretation if you just go off the name “music theory”—and so if music theory doesn’t describe it, it isn’t real music.
Part of the problem lies in the way that music theory and composition are taught. For example, early on in the music classes I took in high school, I was taught to not use parallel fifths. For those not familiar with music theory, a fifth is a kind of interval between two notes. If two notes are played together or in succession, the difference in their respective pitches is called an interval and one of those intervals is called a perfect fifth. If two melodies are playing at the same time and moving together—when one voice goes up the other goes up and when one voice goes down the other goes down—that’s called parallel motion, and if the distance between them is a fifth, then you have parallel fifths, which I was taught was a bad thing.
What I wasn’t taught was that parallel fifths are only a problem in a certain musical context: if you have two or more melodic lines that you want to be independent of one another, you have to follow certain guidelines or risk accidentally causing the melodies to merge. Parallel fifths destroy the independence of melodic voices. There are other reasons as well to avoid parallel fifths in this and other contexts, but there are also contexts where parallel fifths are a very useful musical device.
I’ve fallen into some of these traps myself, both early on when I was initially learning music and academic music theory and even within the last few years as I’ve been focusing more on music production using the popular digital audio workstation Ableton Live.
I mentioned in a prior essay, “Artificial Intelligence and Societies of Control” from March of last year, that there’s a piece of music production software that I use—Neutron, produced by the company iZotope, Inc.—which includes a technology that the company calls assistive audio technology. Neutron is a channel strip, a bundle of common audio processing tools intended to be applied to each track of a music production in process. I also use another piece of software from iZotope, Ozone, which does largely the same thing but which is intended to be applied to the entire mix (or large segments of the mix called stems) rather than to individual tracks. The assistive audio technology built into Neutron and Ozone ostensibly uses machine learning algorithms to analyze the audio signal to which it is applied, and then configures the software’s various settings so as to process the audio in a theoretically ideal way.
Both programs can be used without the assistive audio technology, but I was still using it at the time I wrote that essay, though after applying the ostensibly machine-learning-recommended settings, I would make my own adjustments. I don’t use it at all anymore. I still use Neutron and Ozone, but not their assistive audio technology features. For one thing, I’m not convinced that these features actually use machine learning algorithms at all, because I noticed that, one, they were applying largely the same settings from track to track, and two, it didn’t seem to matter what processing I was using in the signal chain before it hit the channel strip. If, for example, I applied a high shelf filter to a track to drastically cut its high frequency content, the channel strip wouldn’t then apply a dramatic boost to the high frequencies to compensate. After I first noticed this, I started testing the software by deliberately building problems into the signal before it hit Neutron to see if it would compensate for them in any way, but it never seemed to notice. I suspect that what “assistive audio technology” actually does is apply some generic tweaks which tend to make most music sound better in a generic way.
But more to the point, I realized that I was using the assistive audio technology because I was working under the assumption that, given the kind of music I was producing, there was some correct way to process it. In other words, given the initial constraints of the raw music I had created, I believed there was a procedure which, if properly followed, would result in the mix being “correct.”
It’s one thing to apply this sort of procedural thinking to mixing, but with increasing frequency I receive emails from music software companies describing their new products which apply machine learning algorithms to the process of composition itself. These products will (the marketing emails promise) create drum patterns, generate chord progressions, even write melodies or arrange songs. The products’ existence is testament to a market for such things, which conveys to me a kind of desperation to ameliorate the burden of freedom and the responsibility for saying something with one’s own voice.
What I think it comes down to is abdication of the will, a phenomenon I wrote about in an essay of that name that I published in July of 2020. In that essay, I described several instances in which people had given up their will—the driving force of their decisions and actions—to other people, to groups of people, or to procedural systems. For example, the essay opened with a description of the famous Milgram experiment. In the experiment, subjects participated in a study which they believed concerned punishment-reinforced learning. Ostensibly, they were reading questions to another research subject (actually an accomplice of the research team) and responding to wrong answers with electric shocks. In fact the accomplice was only pretending to be shocked, and would, as the shocks increased in voltage, become increasingly alarmed, eventually pounding on the walls and then ceasing to respond at all. Most of the research subjects continued to deliver shocks well past this point, choosing to follow a procedure even while they had every reason to believe that that procedure was causing harm.
Obviously, the abdication in the Milgram experiment is a much more serious matter than the abdication I’ve been talking about in this essay, but I believe that the underlying mechanism is the same. I think that the reason that people want music theory to provide the kind of procedural structure that I’m talking about is the same reason that people conform to various social expectations. People will do literally anything if they feel that there is a procedure being followed, so it seems reasonable to conclude that that’s something that people desire, and music theory certainly does a good job of presenting an illusion of procedural structure. Over the remainder of this essay, I’ll be exploring what the underlying mechanism might be. Please bear in mind, though, that this will be largely speculative.
Life on Earth began with single-celled organisms which, at some point, through the general process of evolution by natural selection, landed on collectivization as a viable solution to certain problems posed by the environment. This was the advent of multicellular organisms, though in some sense biological collectivization precedes even this, as unicellular organisms existed in colonies from which the first multicellular organisms may have evolved. In this sense, all life makes use of collectivization in some way. The general structure of biological collectivization is hierarchical and modular: ecosystems contain colonies, groups, societies, packs, and lone individuals; excepting unicellular organisms and very simple multicellular organisms, each member of these groups is a system of organs, which are in turn composed of cells. Even individual cells benefit from their symbiotic relationship with mitochondria.
Some groups within an ecosystem—take an ant colony as an example—function in a highly collective way. It almost makes more sense to think of the ant colony as being the unit organism of a given ant species, rather than the individual ant. At the other end of the spectrum are, for example, polar bears, who are almost entirely asocial. Based on what we’ve observed of chimpanzees, early humans probably lived in bands of perhaps 20 to 100 individuals which consolidated into larger hierarchical organizations: tribes, cities, nations, and civilizations. At any level of collectivization, there are benefits and there are drawbacks, and what benefits exist at a given level can be annulled through the contrary individual action of one or more of the collective’s members. For this reason, biology and sociality within collectives has evolved so as to inhibit certain manifestations of individual behavior. Humans in particular seem programmed, whether as a result of biology or socialization or a combination of both, to accept one or more other humans as “leaders” and to adhere to the behavior of groups of which they are a part.
What is the underlying mechanism that drives social cohesion? At some level, the best explanation seems to be reason: we begin with an observation of a situation or social dynamic and reason that the best thing to do in that situation is more-or-less what everyone else is doing. But this is not sufficiently explanatory in itself. I posit that the underlying mechanism is in fact what I’ll refer to as autogenic fear, a fear that arises from and is directed towards the self, a fear of the self.
As humans, we are from birth indoctrinated into a social context on which we rely to fill some of our needs. If we look at Maslow’s hierarchy of needs—hardly a perfect model but good enough, I think, to demonstrate what I’m talking about—we observe a pervasive integration with our social environment. For most humans, presently and historically, our needs, both basic and advanced, have been either provided by or acquired in cooperation with other humans. Our psychological needs often require other humans: we need to be friends with, esteemed by, loved by other humans. So it is entirely natural for us to want to defend our integration with this social context from threats. Certainly there are environmental threats to worry about: war, natural disasters, resource depletion; but it is also possible that the social context itself choose to remove us. A member of a given group may be shunned, ostracized, or imprisoned by that group, or exiled from it.
For what reason might such a thing happen? Well, one might act in a way that violates the group’s shared value system. If, for example, I were to eat a baby, and if this were discovered by any of the social groups of which I am a part, I would certainly be removed from those groups in some way. Nobody wants a baby-eater on the team (“Hey, welcome aboard; this is Jim, he’s our IT consultant, loves football; and that’s Susan, our financial coordinator, way into Dungeons and Dragons; oh, and that’s Todd, the ethics advisor, he eats babies, so… I guess keep an eye out on take your daughter to work day…”). Fortunately for me, and somewhat contradicting my frequent mention of baby-eating on this show, I don’t have any interest in eating babies, but I do frequently experience a primal urge to harm or even murder certain people. At the level of my executive reasoning, this is not something I ever want to do, but it’s possible for me to conceive of examples under which such harms would be, at least arguably, a rational and justified course of action were it not for the negative consequences that would fall back on me and my loved ones. But there is a part of me that desires this regardless, and so I think my fear of that part of myself is at least partially justified. It seems a chaotic and unpredictable other within me that might emerge and cause problems for me if I don’t keep it locked down.
A feature of group membership is the diffusion of responsibility. If I make a decision on behalf of a group, the group might blame me if the decision turns out to have been a bad one. But if a decision is made collectively, it is impossible to blame anyone. This isn’t to say that blame can’t be placed regardless, creating a scapegoat, but that actually serves to illustrate the problem. We know that we want someone to blame when things go wrong, and we fear being the person blamed, so we avoid giving anyone a reason for blaming us. And so we fear sharing our thoughts for fear that they may lead the group astray, or for fear that they may be perceived that way even if we are confident that we are correct. This is known to occur in corporate boardrooms, where, for example, all board members will privately support a change of strategy in the face of poor performance but say nothing because they believe that all of the other board members think that the company should continue to adhere to the status quo (Westphal and Bednar, 2005).
It’s also possible that we feel that we simply have nothing worth saying in the first place. I think many people approach songwriting or music production not because there is something particular that they want to express for which music is the proper medium, but because making music is just a fun thing to do. But while writing a poem or a story or an essay seems to require some semantic intention, semantics are typically, at most, secondary considerations of the non-lyrical elements of music. The opening melody to Beethoven’s fifth symphony, for example, does not mean something in the same way that this sentence means something. In the full context of the music, there is a great deal of emotional meaning, but the melody in itself, as an abstract composition, we can imagine having been composed without any particular emotion having been intended. And so, while writing an essay without actually saying anything seems absurd—though some of my early essays now seem to me to accomplish exactly that—it seems much more possible to do so with music, and so the possibility exists for musical neophytes to enjoy the process without having to worry about conveying anything through the music.
But if I’m writing an essay, what I write sentence-by-sentence is constrained in a significant way by what I’m writing about. If I were just composing grammatical sentences without worrying about their meaning, the degrees of freedom available for each choice would overwhelm me, and I would want for some procedure to guide me through it. In this case, autogenic fear becomes the fear not of the self but of the lack or absence of self. I suspect that many people do indeed have something meaningful and important to say but don’t know what it is or have not accurately evaluated it as such. But for many other people, perhaps most people, I think this fear may be justified. In conversations with people I meet, and especially with young people, I’ve found increasingly that they have no interests, no hopes or desires, no thoughts of their own whatsoever, and partake of no activities beyond content consumption and interaction with social media. More and more people, it seems, have become passive receptacles for YouTube and TikTok content. Music, then, becomes not a matter of vital expression, but a kind of trivial, procedural game, with songs a mere byproduct of the process of their creation.
I hope you’ve found this piece interesting and informative. If you’ve enjoyed it, I encourage you to look at some of my other essays, and if you find my approach to philosophy and religion at all valuable, I hope that you’ll stop in at my Patreon page, which features bonus content for patrons, and that you’ll stop back by to check on my new content.
Works Cited or Referenced
Ewell, P. A. (2020). Music Theory and the White Racial Frame. Music Theory Online, 26(2). https://doi.org/10.30535/mto.26.2.4
Neely, A. (2020, September 7). Music Theory and White Supremacy. https://www.youtube.com/watch?v=Kr3quGh7pJA&ab_channel=AdamNeely
Westphal, J. D., & Bednar, M. K. (2005). Pluralisitc ignorance in corporate boards and firms’ strategic persistence in response to low firm performance. Administrative Science Quarterly, 50, 262–298.
Xenakis, I. (1992). Formalized music: Thought and mathematics in composition (Rev. ed). Pendragon Press.