Yankee Doodle, Boulezian Amnesia, Imperfect Replicas
In this eleventh issue, we recommend a cultural analysis of American revolutionary war and traditional music, a musical paradox contained within the work of Boulez, and the anticipated Part II of a inquiry into the commercial and aesthetic stakes of AI.
Recommendation: “Yankee Doodle,” John Cage’s Thirteen Harmonies (Selections from 44 Harmonies From Apartment House 1776), as performed by Cop Tears
This past week I crushed Ken Burns’ The American Revolution documentary series. Also in the spirit of ‘76, I’ve been revisiting my favorite John Cage piece Thirteen Harmonies (Selections from 44 Harmonies From Apartment House 1776) in preparation for a forthcoming recommendation on Derek Baron’s new album The Holy Restaurant (Derek’s Cop Tears ensemble released my favorite recording of the Cage work on their Reading Group label). The Harmonies is a beautiful, spare arrangement of the inner voicings of psalms and hymns from the 18th century American choral tradition (mostly William Billings). Omitting the original melodies altogether, hollow voice-leadings drift along in incidental counterpoint and new harmonic permutations. Cage’s chance procedure situates these metamelodic forms into new interactions, where the dressings of modal harmony graze alongside each other in an uncanny valley, quietly reforming the disparate characters into a novel and permeable unity. Beneath Cage’s interventions, the work is in large part so incredible because the source material remains historically transcendent, uncorrupted and fossilized in its propositions.
This general curiosity about the early American period has inspired me to listen to revolutionary era jigs, hymns and marches more broadly. Many have an addictive draw to them. The propulsive circularity of the melodies, mirrored in the rhythmic suspension of dotted note syncopations and the hammering of rising fifths and fourths lock in with a fluctuating consistency that feels infinitely repeatable. Walking to the grocery listening to the tune “Jefferson and Liberty,” I was genuinely fired up; and, I was compelled to imagine how and why these dances circulated with global popularity centuries before music technology began to extract the resources of sub-bass and refine the mid-range of the fiddle with compression.
Cage’s study of material from this tradition as a permeable and combinatory set grasps its special, engineered qualities of durability and adaptable - almost universal - functionality. The same tunes were deployed in taverns and military marches by populations and forces with directly opposed political interests. If I may invoke for you a bit of 6th grade social studies: the iconic “Yankee Doodle,” was a tune sung by British occupying forces to mock the backwater American colonists, before they reinterpolated the mockery as a proud expression of collective identity, and later developed myriad other American heroic ballads over the melody. Well before any of that, the tune was known as a swashbuckling jig, “All the Way to Galway,” sung by Irish sailors:
“I swept a few grogs at Monroe’s Inn
(All the way to Galway, aye)
Then I took a piss at the gutter stones
(All the way to Galway, aye)
But I got a whack in my drunken head
(All the way to Galway, aye)”
There is a deeper cultural and psychological formality that resonates behind this evident structure of popular culture. In the Burns documentary, I was struck by many strangely whimsical and contemplative soldiers’ accounts of brutal and disturbing circumstances. One which stuck with me, was an American militiaman recounting total defeat and near death in a battle around New York City: amidst a wild bombardment of canons, disembowelments and decapitations of his friends, he said, “I made a frog’s leap into a ditch, where I stayed as still as I could be and wondered which part of my carcass would go first.” This wry matter-of-factness – awaiting annihilation, but evoking the folksy natural imagery of a frog’s leap away from danger – invites a couple of interpretations. He is either so familiar with a general condition of violence, that even his own humble destruction contains an endearing and disturbing element of whimsy; or, the conventions of “common people’s” speech and literacy are so relatively new, that the images of oral folktales and provincial life come to cohabitate with descriptions of global-scaled atrocity, moral and religious conviction and high-minded, new formations of the common political imaginary. I can only guess here, but both seem plausible.
This cultural signature is found across the literature. “Success to America Says Granuille” (presumably a reference to the Irish pirate-queen Grace O’Malley, but also ironically, perhaps another frog reference) is a jolly song in which 500 die of their wounds, and the British come ”...to spill my Guinness and my children’s blood.” It feels too easy to understate or ignore how psychedelic this combination is of formalized humor, normality of mass death and emergent social dimensions of political idealism and nationalistic heroism. There’s a type of corporeal investedness, grand historical scale and an existential smallness that is hard to relate to, I think, from within the standpoint of America’s last couple centuries of global imperial success.
I plan to further explore and problematize some of these historical and cultural queries in future recommendations, but for now, I’ll return to John Cage’s perceptive musical reckoning with the problems and vitalities of American identity. As Thomas Paine wrote of the revolutionary epoch, reflecting on the “precariousness of human affairs:” “it is wiser and safer, to form a constitution of our own [...], than to trust such an interesting event to time and chance.” The “time and chance” of the American revolution folded over into ever more complex layers of imperialism, and likewise, the complicated “time and chance” of music has similarly altered, adapted and reformed the basic social constructions on which its expressions depend. I recommend Cage’s particular “constitution” in Thirteen Harmonies as both culturally instructive in the analogies of radical American musical and social forms, and as a more abstract and perhaps spiritual codex from which we can connect to our own whimsical, existential humility.
—Alec Sturgis
Recommendation: Pierre Boulez: Music Lessons, The Collège de France Lectures
A few weeks ago, I picked up a copy of Music Lessons, the first English translation of Pierre Boulez’s Collège de France lectures, written while he was teaching between 1976 and 1995. Boulez is perennially on my mind through his demanding, often contradictory worldview. Philosophical inconsistency is a meaningful aspect of his great musical thought—almost running parallel to my favorite pre-Socratic philosopher Heraclitus—an inconsistency centered around how a deep structure (Logos) underlies extreme change and variability. Music is forgotten, remembered, and in Boulez’ hope, interpreted and reinvented through structural modification. Throughout his lectures, Music is at once an object, a text, and a physics. Boulez was known to disavow his own work, revise his positions while staking incendiary claims, taking up the whole of music as a material to be worked with in all its discrepancies.
I admire Boulez’ seeming hypocrisies as essential to an authentic musical thinking. Boulez famously called for opera houses to be blown up, while simultaneously conducting the New York Philharmonic Orchestra and BBC Symphony, founding IRCAM, amongst innumerable other accolades—an anti-institutionalist working starkly within institutions. He advocated for controlled aleatoricism in music, while also striking up anti-chance positions in the post-war musical landscape documented in The Boulez-Cage Correspondence. He spoke of modernism as an ethical obligation, while offering strained outlines for avant-garde values often at odds with themselves—attempting to liberate music from tradition, but also to bind it into strict musical order. He was both a polemicist and an architect. I find all these contradictions to be what music is all about.
I remember discussing Boulez with Jack Callahan at length in the late 2010s, around the same time I was writing an essay Against Worldbuilding for Tiny Mix Tapes—swapping texts and PDFs of Wagner, Boulez, and Feldman as we were contending with various conceits of electronic music during that particular moment of music culture. That essay also contained a lot of contradictions, but ones that felt like footholds to theorize music for me. Against Worldbuilding also sought to emphasize musical inconsistency colliding with various attempts to build internal consistency through systemization and notation, recording, particular spatialization, or certain symbolic or social efficacy. Despite these attempts at consistency, style, aesthetic appreciation, and the technological production and social consumption of sound all incongruently condition our reception, ideas, and agency for listening. Throughout Boulez’ lectures, we see a profound reckoning with this.
Flipping through Boulez’s writings, I was pulled into a passage from his 1988 lectures on “Memory and Creation.” Many of his writings in this particular era are concerned with the act of composing music, referencing the complications of the immutability of the “text” of music in contrast to its performative mutability. On this point, Boulez lectured:
“Shall I once again sing the praises of amnesia? One feels that in the midst of an era ever more imbued with memory, to forget becomes so urgent … and yet not only do we not forget, but we gild all the libraries in all the Alexandrias—reference being integral to discovery, and the source of the only kind of renewal still possible. The era of avant-gardes and exploration being definitively over, what follows is the era of perpetual return, consolidation, citation.”
I’m drawn to this at the close of 2025. So much of my own musical listening has been operating within a tension between musical memory and a quest for amnesia. Particularly having spent this past Fall writing about music again, I’ve been thinking about musical activism and passivity—and in a way 2020MG has become a space for chronicling a cadence between active and passive stances in musical experience. As Boulez lectured, musical memory’s gravity over the present, and any musical gesture, is pervasive. To pay attention to music, and develop active processes for our reflection on music, also reveals music’s latent beginning and disappearing, forgetting and remembering. Boulez shows us this kind of implicit musical paradox: active amnesia, passive amnesia, active memory, passive memory—all postures hiding within the medium of musical time, and musical text.
Ultimately, Boulez’s praise of amnesia reads as a kind of consolation and acknowledgement in the failure of the avant-garde, much of which he tried to spur onward. His use of the term “perpetual return,” a nod to the Nietzschean “eternal return,” shows that he’s trying to conjure an existential affirmation within the memory and amnesia that enframes our work with music.
In this spirit, and as the great interpreter of musical systems he is, Boulez finds again his active form within the “fragment” of music. He pops off in his final 1995 lectures on “Memory, Writing, and Form”:
“One can now, after all one’s experiences of the twentieth century, arrive at this completely provisional and doubt-filled conclusion: the work can be only a fragment of an imaginary whole. The work thought of as a whole is never more than a convenient illusion, but as with light through a prism, it deconstructs into fragmentary components, which when given a temporal continuity regain the appearance of a whole [...] In other words, and provisionally to complete my remarks: having no reality but the fragment, the whole is nothing but an endless renewed, endlessly sought-after illusion.”
—Nick James Scavo
Recommendation: Invest Like the Best with Patrick O’Shaughnessy [Episode 430] - The Chainsmokers
Last week, we recommended an interview with the Chainsmokers on the Invest Like the Best podcast, in which they discuss their artistic practice and experience running a VC firm. There was one passage in particular that we focused on, spoken by bandmember Drew Taggart:
How do you define success in an age of abundance of art? How do you define success with us, where we’ve had a bunch of songs [with] over a billion streams, we’ve had chart-topping songs, but what is the one thing that we do that no one else can do? Can that even be replicated by AI? Can that be replicated by another person? Because when that person does it, it will be them doing it, and it’ll be at a different time, and it’ll look different, and it will feel different.
When you think about AI, you can’t be scared of it, because it’s just gonna hit different, for better or for worse at some point. We just gotta focus on being the most Chainsmokers as possible.
In order to understand what’s going on here – I mean this in a very general sense – it would be useful to situate Taggart’s thoughts in the context of current debates within AI research in order to evoke a small slice of the conceptual and technical milieu in which his statement operates.
We can begin with the questions of the definition and replication of intelligence, and refer to a recent conversation between the computer scientist Richard Sutton and the podcaster Dwarkesh Patel. Sutton is one of the founders of reinforcement learning (RL), a branch of machine learning concerned with how a computational agent learns to make decisions in order to maximize reward signals. RL is generally used in contemporary AI products to fine-tune LLMs after their initial training, functioning as a refinement layer on top of their core next-token prediction mechanism.
Sutton is also known for his influential 2019 essay “The Bitter Lesson,” in which he argues that “[AI researchers] have to learn the bitter lesson that building in how we think we think does not work in the long run.” Citing examples of research progress in the fields of computer chess, speech recognition, and computer vision, he makes the following assessment:
The bitter lesson is based on the historical observations that 1) AI researchers have often tried to build knowledge into their agents, 2) this always helps in the short term, and is personally satisfying to the researcher, but 3) in the long run it plateaus and even inhibits further progress, and 4) breakthrough progress eventually arrives by an opposing approach based on scaling computation by search and learning. The eventual success is tinged with bitterness, and often incompletely digested, because it is success over a favored, human-centric approach.
Patel and Sutton disagree about the basic nature of intelligence, and how it might be constructed through computational means:
Patel: We’re trying to replicate intelligence. If you want to understand what it is that enables humans to go to the moon or to build semiconductors, I think the thing we want to understand is what makes that happen. No animal can go to the moon or make semiconductors. We want to understand what makes humans special.
Sutton: I like the way you consider that obvious, because I consider the opposite obvious. We have to understand how we are animals. If we understood a squirrel, I think we’d be almost all the way there to understanding human intelligence.
This is an interesting claim, and it is taken up in a response essay written by another influential AI researcher, Andrej Karpathy, who is known for his work as a founding member of OpenAI and as the former Director of AI at Tesla. In “Animals vs Ghosts,” Karpathy offers the following assessment of whether LLMs are sufficiently “bitter lesson-pilled”:
Frontier LLMs are now highly complex artifacts with a lot of humanness involved at all the stages - the foundation (the pretraining data) is all human text, the finetuning data is human and curated, the reinforcement learning environment mixture is tuned by human engineers.
The implications of Sutton’s “Bitter Lesson” are quite psychedelic, and Karpathy directly draws them out. He goes on to address Sutton’s claim that “If we understood a squirrel, I think we’d be almost all the way there”:
I still think it is worth to be inspired by animals [sic]. I think there are multiple powerful ideas that LLM agents are algorithmically missing that can still be adapted from animal intelligence … Stated plainly, today’s frontier LLM research is not about building animals. It is about summoning ghosts. You can think of ghosts as a fundamentally different kind of point in the space of possible intelligences. They are muddled by humanity. Thoroughly engineered by it. They are these imperfect replicas, a kind of statistical distillation of humanity’s documents with some sprinkle on top.
For the moment, let’s bracket the question of whether LLMs genuinely “summon ghosts,” and consider the terms and implications of the disagreement between these two researchers, focusing on the level of signification. Returning to the Chainsmokers quote with this discipline-specific conceptual scaffolding in mind, let’s simply consider the proposal that on some level Taggart is talking about spirits. That he is implicitly telling a ghost story, thinking across many different times and places, and that it is in the milieu of spirits that we consider the question of being the most Chainsmokers as possible. What should we make of this?
—Alexander Iadarola