Maintenance, Meaning, and Magic

Working notes, not prose. This page is a research packet for drafting the chapter: the brief, source material, questions, examples, reader perspectives, and the earlier draft.

Brief

Must: Take up the parts of life that don’t fit the MMM model, such as maintenance, meaning and magic, and give them their due without forcing them into the loops.
Serves: It keeps the model honest by showing where it stops, so the reader doesn’t mistake a tool for the whole of life.

Quote options

“The sourball of every revolution: after the revolution, who’s going to pick up the garbage on Monday morning?” — Mierle Laderman Ukeles, Manifesto for Maintenance Art 1969! (1969), typescript p. 1. [verified: https://kortina.nyc/notes/manifesto-for-maintenance-art-1969/ ; cross-checked https://www.makejusticenormal.org/resources/mierle-laderman-ukeles-manifesto-for-maintenance-art-1969] Why: Names maintenance as the work that learning, creating, and becoming all quietly depend on but never count.
“The place to improve the world is first in one’s own heart and head and hands, and then work outward from there. Other people can talk about how to expand the destiny of mankind. I just want to talk about how to fix a motorcycle.” — Robert M. Pirsig, Zen and the Art of Motorcycle Maintenance (1974), Part III. [verified: https://archive.org/details/zen-and-the-art-of-motorcycle-maintenance-an-inquiry-into-values-robert-m.-pirsi (full text)] Why: Treats plain upkeep as a serious practice in its own right, not a lesser stage of some bigger loop.
“One might paraphrase this by saying that, according to my conception, meaning arises when subjective attraction meets objective attractiveness, and one is able to do something good or positive about it.” — Susan Wolf, Meaning in Life and Why It Matters (Tanner Lectures, 2007; Princeton, 2010), Lecture 1. [verified: https://philosophy.northwestern.edu/community/nustep/08/papers/Wolf.pdf] Why: Gives meaning a precise shape that no measure-model-manipulate loop produces by itself.
“We feel that even if all possible scientific questions be answered, the problems of life have still not been touched at all.” — Ludwig Wittgenstein, Tractatus Logico-Philosophicus (1922, Ogden trans.), 6.52. [verified: https://en.wikisource.org/wiki/Tractatus_Logico-Philosophicus/6] Why: The classic statement that a complete model can still leave out what matters most, which is the chapter’s whole claim.
“This means that the world is disenchanted. One need no longer have recourse to magical means in order to master or implore the spirits, as did the savage, for whom such mysterious powers existed. Technical means and calculations perform the service.” — Max Weber, “Science as a Vocation” (1918/1919), Gerth & Mills trans. [verified: https://sociology.sas.upenn.edu/sites/default/files/Weber-Science-as-a-Vocation.pdf] Why: Frames “magic” as what calculation pushes out, so the chapter can ask what was lost along with it. (Note that “savage” is Weber’s period language.)

Research

Core argument

Start from the model’s own boast. Chapters 13 to 16 gave the reader three loops (Learning, Creating, Becoming) for meeting a problem. Every loop presumes a problem, a direction, and a kind of progress. Ask what happens to the parts of life that have none of those three things.
Maintenance: the loop that goes nowhere on purpose. Most of the work that keeps a life, a city, or a cell going doesn’t learn, make or become anything. It returns things to where they were: dishes, DNA repair, bridge inspections, taking a parent to dialysis. Arendt’s distinction helps here. Labor answers the body’s cycle and leaves nothing behind. Work builds a durable, artificial world, and Arendt herself ties work to “the unnaturalness of human existence.” MMM mostly describes work. Maintenance is labor, and the book’s word unnatural depends on it.
Maintenance is structurally invisible, and that is a political fact. It shows only when it fails. Innovation gets credit, and upkeep gets done, mostly unpaid and disproportionately by women and low-status workers (Ukeles, Russell and Vinsel). A tool built around improvement will rank upkeep below it unless someone corrects for that on purpose.
Maintenance isn’t a failed loop. Don’t try to rescue it by relabeling it (for example, calling cleaning “Becoming” or treating repair as a tiny “Make”). Some repair does teach (Jackson’s “broken world thinking,” kintsugi, the Ise rebuild), and that deserves a mention. The main claim is that keeping something going is a good in itself, and it keeps paying off only as long as you keep doing it. Arendt’s Augean stable makes the point: only the mythical one stays clean.
Meaning: something no loop produces. The loops can tell you how to do something. They can’t tell you whether it was worth doing. Wittgenstein’s 6.52 (already among the chapter’s quotes) and Wolf’s formula, subjective attraction meeting objective attractiveness, both put meaning in a relation between the person and the world, and a process can’t generate it. Camus’s Sisyphus is the limit case: a pure maintenance loop, a boulder that never stays up, and Camus asks us to imagine him happy. Meaning can be found inside a loop that makes no progress at all.
Magic, part one: the anthropology. Malinowski’s Trobriand fishing contrast is the chapter’s best single piece of evidence. There was no magic for the safe lagoon and a great deal of it for the dangerous open sea. Magic grows where knowledge and skill run out and the stakes stay high. It isn’t failed science. It is a response to the part of the world that Learning can’t make predictable. Gmelch found the same pattern in baseball, and it appears in exam-day rituals too.
Magic, part two: what calculation disenchanted, and what came back. Weber said technical means now “perform the service” that magic once did. Clarke’s third law turns this around: to most users, most technology is effectively magic again, because they can use it without being able to explain it. The open-label placebo trial shows that ritual and expectation have measurable effects even when nobody is deceived. The chapter should separate three things: magic as a false causal claim (reject it, per Chapter 4 and Chapter 7), magic as a practice under uncertainty (understand it, per Chapter 21), and wonder that something exists at all (Wittgenstein 6.44), which no explanation removes.
Why the model has to stop somewhere. A model that explains everything explains nothing, and it also starts to overrule people: grief becomes “stalled Creating” and rest becomes “strategic recovery.” The earlier draft’s list (repair, change the conditions or seek help, rest, live with a loss) is a list of things MMM doesn’t cover. Promote it into the chapter’s spine.
Close by giving each its due. Maintenance calls for thanks and fair distribution. Meaning calls for attention (Murdoch, Weil). Magic calls for humility, plus a clear line between wonder and fraud. Hand off to Chapter 18: once we admit loops that don’t aim at improvement, we can ask how any loop stabilizes, spirals or runs away.

Key questions

If you counted your waking hours for a week, how many went to learning, creating or becoming, and how many to keeping things from falling apart?
Is maintenance a kind of loop the model missed, or is it the ground every loop stands on?
Why does nobody throw a launch party for a bridge that didn’t fall down?
What is the difference between repair that teaches (you learn why it broke) and repair that only restores? Does the second kind matter less?
A child asks: “If we’re just going to make the bed dirty again tonight, why make it?” What is the honest answer?
A cynic says: “Talk of the dignity of maintenance is how people who don’t do it keep the people who do it from asking for more money.” How much of that is right?
Can a life be meaningful if it produced nothing new? (Test cases: a lighthouse keeper, a carer for a dying parent, a monk.)
Is meaning something you find, something you make, or something that happens between you and the world? What difference does the answer make on a Tuesday?
Why did Trobriand fishermen use magic on the open sea but not in the lagoon, and where are your own “open seas”?
Is a lucky ritual before an exam irrational if it measurably calms you down?
Clarke says advanced technology is indistinguishable from magic. If you can’t explain how your phone works, in what sense is your relationship to it different from a charm?
When does wonder turn into credulity, and what signs mark the crossing? This connects to the book’s “not being cheated” aim.
Is Weber’s “disenchantment” something we lost or something we escaped? Could it be both?
Does a framework that admits its limits become more trustworthy or less useful?
Which parts of your life have you (or others) tried to optimize that were never meant to improve, such as friendship, grief or sleep?
What would the Ise carpenters say about “progress,” given that they rebuild the same shrine every 20 years so that the skill survives?
Is rest maintenance, meaning, magic, or none of these?
Can a cell’s maintenance (autophagy, DNA repair) count as “caring for itself” in any sense, or does the word stop working below the level of a someone (Chapter 1)?
If the loops are “unnatural” in their sophistication (Act III close), is maintenance the most natural thing we do?
What does the book owe the reader who is too exhausted for any of the three loops?

Examples

The kitchen at 7:10, the next morning. Continue the dinner scene from Chapter 13. Nobody is learning, creating or becoming. Someone is scraping the pan. Ask who washed up last week, and the week before. (This draws on the earlier draft’s “noticing who has cooked and cleaned every week.”)
The cell that eats itself. Human cells take somewhere between tens of thousands and a million DNA lesions per cell per day, and they repair nearly all of them constantly. Through autophagy (“self-devouring”) cells break down and recycle their own damaged parts, and Yoshinori Ohsumi won the 2016 Nobel Prize for finding the genes involved. Life is maintenance before it is anything else. https://en.wikipedia.org/wiki/DNA_repair ; https://en.wikipedia.org/wiki/Autophagy
Sleep as night-shift cleaning (with a caveat). Xie et al. (Science, 2013) reported that in mice, sleep expands the brain’s interstitial space and speeds clearance of waste such as beta-amyloid. The glymphatic account is still disputed: some studies don’t replicate the widespread exchange, and a 2024 Nature Neuroscience paper (Miao et al.) reported reduced clearance during sleep in mice. Use it as a live question, not a settled fact. https://en.wikipedia.org/wiki/Glymphatic_system
Mierle Laderman Ukeles, Touch Sanitation (1979–80). Over nearly a year Ukeles shook hands with more than 8,500 New York City sanitation workers and said to each, “Thank you for keeping New York City alive.” She has been unpaid artist-in-residence at the Department of Sanitation since 1977. Her 1969 manifesto (already a chapter quote) asked “who’s going to pick up the garbage on Monday morning?” https://en.wikipedia.org/wiki/Mierle_Laderman_Ukeles
Ise Grand Shrine, rebuilt every 20 years. The Shikinen Sengū rebuilds Japan’s most sacred shrine on an adjacent plot every two decades. The 62nd rebuild was in 2013, the next is due in 2033, and each takes about 17 years of preparation. Less experienced carpenters build the Uji Bridge first to learn the skills. Here maintenance is the means of transmitting knowledge, which makes a good bridge to Chapter 19. https://en.wikipedia.org/wiki/Ise_Grand_Shrine
The Forth Bridge myth. “Painting the Forth Bridge” became an idiom for an endless task, though continuous end-to-end repainting never actually happened. A £130m repaint from 2002 to 2011 used a coating meant to last about 25 years, and the project manager declared “Job done.” It’s a good example of how we mythologize maintenance, and how an innovation can reduce maintenance without ending it. https://en.wikipedia.org/wiki/Forth_Bridge
I-35W bridge, Minneapolis, 1 August 2007. Thirteen people died and 145 were injured. The bridge had been rated “structurally deficient” since 1990. The NTSB’s primary cause, though, was undersized gusset plates in the original design, made worse by added load. It’s an honest case: failure makes maintenance visible, but the neat “they didn’t maintain it” story is too simple. https://en.wikipedia.org/wiki/I-35W_Mississippi_River_bridge
The maintainers. Russell and Vinsel (Aeon, 2016) estimate that inventors and innovators are “perhaps somewhere around one per cent” of the technical workforce, and argue that upkeep has more effect on daily life than most innovation. https://aeon.co/essays/innovation-is-overvalued-maintenance-often-matters-more
Kintsugi. One origin story has the shōgun Ashikaga Yoshimasa (late 15th century) sending a cracked Chinese tea bowl away for repair. It came back stapled, and Japanese craftsmen developed gold-lacquer repair as a more beautiful answer. Collectors were even accused of breaking pots deliberately. Here repair adds meaning. https://en.wikipedia.org/wiki/Kintsugi
Trobriand fishing (Malinowski, 1925/1948). The inner lagoon was fished reliably by poisoning, with no magic. The open sea was dangerous and unpredictable and came with extensive ritual. Magic tracks uncertainty and stakes, not stupidity. (Quoted below.)
Baseball magic (George Gmelch, “Baseball Magic,” Trans-action 8(8), 1971). Gmelch tested Malinowski’s theory on professional ballplayers. His well-known finding is that rituals cluster around hitting and pitching, which are high-variance, and are rare in fielding, which is reliable. Check the exact fielding and batting figures in the original before citing numbers. https://en.wikipedia.org/wiki/George_Gmelch
Placebos without deception (Kaptchuk et al., PLoS ONE, 2010). Eighty IBS patients were told plainly that they were getting sugar pills. The open-label placebo group reported significantly more global improvement than no-treatment controls (5.0 vs 3.9 at endpoint, p = .002). Ritual and expectation work even when nobody is fooled. That’s a case of “magic” that is neither fraud nor physics. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0015591
Sisyphus as the patron saint of chores. Camus explicitly compares Sisyphus to “the workman of today,” who does the same tasks every day. It’s a thought experiment for the reader: is your laundry absurd, and does noticing that make it worse or better?
The Sabbath as maintenance of the soul, and refusal of the loop. Heschel’s The Sabbath (1951) treats one day in seven as “a palace in time” in which you stop making and simply are. A whole civilization built a weekly stop into its calendar. It is neither Creating nor Becoming-for-later-productivity.
Thought experiment: the perfectly optimized life. Imagine an app that routes every hour into Learning, Creating or Becoming, and schedules grief to a 12-week recovery plan. What does the reader feel? Use it to show the model overreaching and to justify the chapter.
Clarke’s third law in a child’s hand. A four-year-old talks to a smart speaker and the lights go off. For her it is magic in the full sense. For most adults it is magic in practice. Where does knowing that something works without knowing how leave us? (Chapter 4’s warning about trusting authority applies here.)

Source material

“Work is the activity which corresponds to the unnaturalness of human existence, which is not imbedded in, and whose mortality is not compensated by, the species’ ever-recurring life cycle.” Hannah Arendt, The Human Condition (University of Chicago Press, 1958), Prologue/Ch. I §1, p. 7. [verified: https://archive.org/download/the-human-condition-hannah-arendt/The%20Human%20Condition%20-%20Arendt%20Hannah_djvu.txt] Use: Arendt uses the book’s own keyword. MMM (especially Creating) is work in her sense, the unnatural part. Maintenance is labor, which she defines just before this (“The human condition of labor is life itself”). This could anchor the whole chapter.
“the endurance it needs to repair every day anew the waste of yesterday is not courage, and what makes the effort painful is not danger but its relentless repetition. The Herculean ‘labors’ share with all great deeds that they are unique; but unfortunately it is only the mythological Augean stable that will remain clean once the effort is made and the task achieved.” Hannah Arendt, The Human Condition (1958), Ch. III “Labor,” §14, pp. 100–101. [verified: https://archive.org/download/the-human-condition-hannah-arendt/The%20Human%20Condition%20-%20Arendt%20Hannah_djvu.txt] Use: The clearest statement of why maintenance doesn’t fit a progress loop: it never finishes, and it isn’t heroic. Pair it with Ukeles’s garbage-on-Monday quote.
“It is most significant that in the Lagoon fishing, where man can rely completely upon his knowledge and skill, magic does not exist, while in the open-sea fishing, full of danger and uncertainty, there is extensive magical ritual to secure safety and good results.” Bronisław Malinowski, “Magic, Science and Religion” (1925), in Magic, Science and Religion and Other Essays (Beacon Press, 1948), p. 14. [verified: https://archive.org/download/dli.ernet.476121/476121-Magic%2C%20Science%20And%20Religion%20And%20Other%20Essays_djvu.txt] Use: The center of the “magic” section. Magic starts where the Learning loop’s reach ends, which makes it a boundary marker of MMM rather than a rival to it.
“The workman of today works every day in his life at the same tasks, and this fate is no less absurd. But it is tragic only at the rare moments when it becomes conscious.” Albert Camus, The Myth of Sisyphus (1942; trans. Justin O’Brien, 1955), final essay “The Myth of Sisyphus.” [verified: https://archive.org/download/CamusMythOfSisyphusAndOtherReadings/Camus%20Myth%20of%20Sisyphus%20and%20other%20readings_djvu.txt] Use: Camus himself links Sisyphus to repetitive daily labor. Follow it with the famous closing line, “The struggle itself toward the heights is enough to fill a man’s heart. One must imagine Sisyphus happy.” (verified at the same URL; the reference entry says it is currently slated for the epilogue, so coordinate), to show meaning found inside a loop with no progress.
“Not how the world is, is the mystical, but that it is.” Ludwig Wittgenstein, Tractatus Logico-Philosophicus (1922, Ogden trans.), 6.44. [verified: https://en.wikisource.org/wiki/Tractatus_Logico-Philosophicus/6] Use: Separates wonder from explanation. No completed Learning loop removes the that. This keeps “magic” in the sense of wonder without licensing superstition. (The chapter already uses 6.52; 6.44 and 6.522, “There is indeed the inexpressible. This shows itself; it is the mystical.”, are on the same page.)
“Any sufficiently advanced technology is indistinguishable from magic.” Arthur C. Clarke, “Hazards of Prophecy: The Failure of Imagination,” Profiles of the Future (rev. ed., 1973); the formulation also appears in a 1968 letter to Science. [verified: https://en.wikipedia.org/wiki/Clarke%27s_three_laws] Use: Pair it with Weber’s disenchantment quote: our world is disenchanted in principle and re-enchanted in practice, and the gap between the two is where people get cheated.
“Maintenance and repair, the building of infrastructures, the mundane labour that goes into sustaining functioning and efficient infrastructures, simply has more impact on people’s daily lives than the vast majority of technological innovations.” Andrew Russell and Lee Vinsel, “Hail the maintainers,” Aeon, 7 April 2016. [verified: https://aeon.co/essays/innovation-is-overvalued-maintenance-often-matters-more] Use: Gives the maintenance section an empirical and political edge, and supports the claim that a progress-centered model misallocates credit.
“Attention is the rarest and purest form of generosity.” Simone Weil, letter to Joë Bousquet, 13 April 1942, in Correspondance (Lausanne: L’Age d’Homme, 1982), p. 18. [verified: https://en.wikiquote.org/wiki/Simone_Weil (secondary source; check the printed edition before publication)] Use: Frames maintenance-as-care and meaning as forms of attention rather than production, and ties back to Chapter 16 without collapsing into it.
“if we consider what the work of attention is like, how continuously it goes on, and how imperceptibly it builds up structures of value round about us, we shall not be surprised that at crucial moments of choice most of the business of choosing is already over.” Iris Murdoch, The Sovereignty of Good (Routledge, 1970), p. 36 (from “The Idea of Perfection”). [verified: https://en.wikiquote.org/wiki/Iris_Murdoch (secondary source; check the printed edition)] Use: Moral attention works like maintenance, slow and continuous and mostly invisible, and it is where meaning actually gets built. It is a strong counterweight to loops that picture life as discrete problems.
“Six days a week we live under the tyranny of things of space; on the Sabbath we try to become attuned to holiness in time.” Abraham Joshua Heschel, The Sabbath (1951), Prologue, p. 10. [UNVERIFIED: Wikiquote gives only the first clause, “Six days a week we live under the tyranny of things of space…”, and the rest is from memory. Quote only the first clause unless checked against the book.] Use: An inherited practice that deliberately stops all three loops each week. It connects to Chapter 21.
“a species activity that includes everything that we do to maintain, continue, and repair our ‘world’ so that we can live in it as well as possible.” Berenice Fisher and Joan Tronto, “Toward a Feminist Theory of Caring,” in Circles of Care (SUNY Press, 1990), p. 40; restated in Tronto, Moral Boundaries (1993), p. 103. [UNVERIFIED: only the core phrase “everything that we do to maintain, continue, and repair our world” was confirmed, at https://en.wikipedia.org/wiki/Ethics_of_care; check the full wording and page.] Use: Defines care as maintenance, which connects this chapter back to Chapter 3 (“love … lived mostly through ordinary care and maintenance”).

Also worth pursuing (no quotation pulled): Steven J. Jackson, “Rethinking Repair,” in Gillespie, Boczkowski and Foot (eds.), Media Technologies (MIT Press, 2014), on “broken world thinking,” the view that erosion and breakdown, not novelty, are the normal condition; Hartmut Rosa, The Uncontrollability of the World (2018; Eng. 2020), on how trying to make everything available and controllable kills resonance, which would be a strong modern ally for the “magic” section; Susan Wolf (already quoted) for meaning.

Counterarguments and limits

“This is just MMM with extra steps.” A careful reader will say maintenance is Manipulate at steady state, meaning is Marvel, and magic is a bad Model. The chapter has to say plainly why relabeling fails. The loops are defined by revision toward something, and maintenance, meaning and magic have no toward. If the author can’t hold that line, the chapter becomes a patch rather than a boundary.
Romanticizing drudgery. Praising the dignity of maintenance can sanctify unfair distributions of it (the cynic’s point, Question 6). The chapter should pair respect with a demand for fair shares and fair pay, and should admit that some maintenance really should be automated or abolished (the Forth Bridge coating).
Overstating invisibility. Plenty of maintenance is visible and prized: surgeons, pilots doing checklists, the Ise carpenters. The “invisible” claim is about credit relative to innovation, not total invisibility.
The Malinowski model is dated and contested. Functionalist “anxiety-reduction” theories of magic have been criticized (for example by Evans-Pritchard and later anthropologists) for being too psychological and for treating other peoples’ practices as compensations. Also avoid a “they did magic, we do science” hierarchy. Gmelch shows the same pattern in American ballparks, and that example helps.
Magic and harm. Giving magic “its due” must not blur into endorsing astrology-for-decisions, homeopathy or manipulative gurus, since the book promises to help readers not be cheated. Keep three senses apart: wonder (keep), ritual under uncertainty (understand and sometimes use), and false causal claims (reject).
Placebo overreach. Open-label placebo effects are real but mostly shown for self-reported, subjective symptoms (IBS, pain), in small trials. Don’t claim ritual heals disease.
Science claims in flux. The glymphatic sleep-cleaning story is contested (see Example 3), and the DNA-lesion counts vary a lot by source (10^4 to 10^6).
Meaning isn’t necessarily outside all models. Psychologists do model meaning (for example Steger’s Meaning in Life Questionnaire, and Frankl’s logotherapy). The claim should be that MMM doesn’t produce meaning, not that meaning is unknowable.
Scope creep. Three “M” topics could each fill a book. The chapter risks becoming a grab-bag. It needs one unifying claim, perhaps “things that matter without improving,” to hold them together.

Connections

Ch. 1 (Before Thought, There Is Want): Does a cell’s self-repair count as wanting? It tests where “someone” begins.
Ch. 3 (Love, Growth, and Meaning): Ch. 3 already says love is lived “mostly through ordinary care and maintenance.” This chapter makes good on that and returns to Wolf.
Ch. 4 (From Subjectivity…): Wonder versus credulity, and not trusting magic-in-practice (Clarke) on authority.
Ch. 6 (Reality and Representation): Magic as the belief that manipulating a representation directly changes reality, the second blade misapplied. Technology as representation that does loop back.
Ch. 7 (Good, Bad…): Why magical practices survive whether or not they’re true, as survival of ideas.
Ch. 9 (The Cost of the Blade): Measures replacing purposes: counting innovation and not upkeep is exactly that failure.
Ch. 10 (Thingification): Institutions and infrastructure are “things” only while someone maintains them.
Ch. 13–16 (MMM): This chapter is their boundary. Especially Ch. 16, whose “refuse to justify by later productivity” is the nearest neighbor, so the author must keep Becoming and Meaning distinct.
Ch. 18 (Loops That Learn and Loops That Don’t): Maintenance is the plainest stabilizing loop (homeostasis, negative feedback). Hand off directly.
Ch. 19 (What Knowledge Needs to Stay Alive): Ise as knowledge kept alive by rebuilding. Knowledge itself needs maintenance.
Ch. 20 (Raising the Floor): Maintenance and care work are the floor. Unpaid care is a program-theory blind spot.
Ch. 21 (Practices Before Patterns): Sabbath, rituals and fishing magic as inherited practice that carries more than pattern-matching would predict.
Ch. 22 (A Better Tuesday): A maintenance audit or rota could become one of the practical practices.
Act III close and Epilogue: Arendt’s “unnaturalness of human existence” is a possible line to reuse. Camus’s Sisyphus is currently slated for the epilogue, so decide where it lives.

Exercise ideas

The upkeep ledger (one day, one page). Instruction: For one ordinary day, write down every task you do in two columns. Column A: tasks that leave something new or changed (you learned, made or grew something). Column B: tasks that only put something back the way it was (cleaning, charging, refilling, replying, repairing, checking on someone). Add a third column for who else did Column B work for you that day, as far as you can tell. Notice: How long Column B is. Which Column B tasks you resented and which you didn’t. How many names appear in the third column, and whether you have ever thanked them. Why: It makes the chapter’s first claim concrete: most of life is maintenance, MMM mostly describes the other column, and maintenance is unevenly shared and poorly seen.
Your open sea. Instruction: List three small rituals or superstitions you actually keep, such as a lucky shirt, knocking on wood, a pre-exam routine, or saying “drive safe.” For each, write (a) what uncertain, high-stakes outcome it attaches to, and (b) what you would lose if you stopped, even knowing it has no causal power over the outcome. Notice: Whether your rituals cluster where your control is lowest, as Malinowski’s fishermen and Gmelch’s pitchers did. Whether any ritual has crossed from harmless comfort into a costly or exploitable belief (paying for it, deciding by it). Why: It tests the chapter’s anthropology on the reader’s own life, and it trains the book’s key distinction between ritual that steadies us and claims that cheat us.
One thing, not improved. Instruction: Choose one thing you keep going purely by maintaining it: a plant, a friendship, a tool, a habit. This week, care for it once without trying to improve, optimize or learn anything from it. Water the plant, oil the hinge, or call the friend with no agenda. Afterward, write two sentences: what it asked of you, and whether it felt meaningful. Notice: The pull to turn it into a project (“I could repot it,” “I should track this”). Whether meaning showed up anyway, without any loop. Why: It lets the reader feel directly that maintenance and meaning can be complete without progress, which is the chapter’s central claim.

Open questions for the author

Is this one chapter or three? Maintenance, meaning and magic could each carry a chapter. Choose one unifying claim, such as “what matters without improving,” or cut one M (magic could move to Ch. 21).
Is “Magic” in the title an asset or a liability? It draws readers, but the book’s anti-cheating mission requires carefully separating wonder, ritual and false claims. Is “Mystery” or “Wonder” a better third M?
Where does Sisyphus live? The reference data slates Camus for the epilogue, but the epilogue is supposed to be a page or less. Moving him here may be better.
Should Arendt’s “unnaturalness” line be used here or saved? It is almost a thesis statement for the whole book, and could serve the Act III close or the epilogue instead.
Does the earlier draft’s “Choosing a Mode” material belong here or in Ch. 13? Its list of non-MMM responses (repair, seek help, rest, live with loss) fits this chapter’s brief best, but the mode-switching advice fits Ch. 13 or 22.
How political should the maintenance section be? Ukeles and care-work economics push toward gender and labor politics. Decide whether to name it directly or leave it to Ch. 20.
Is rest here or in Ch. 16? Both chapters claim it, so pick one home.
How far down the scale of life? Including cellular repair strengthens the “maintenance is prior to everything” claim but risks stretching “care” beyond Ch. 1’s definition of a someone.
Is the chapter’s own framework allowed a fourth M? If the author ends by suggesting “Maintain” as a missing mode, the chapter undercuts its own point that some things shouldn’t be absorbed into the model. Decide deliberately.

Reader perspectives

Curious young child

First reactions:

“MAGIC? Is there going to be real magic in this chapter?” The title promises wands, and the earlier draft delivers three grown-ups staring at a fridge. Big letdown.
Maintenance sounds like the word on the sign when the swings are broken. They know it: “It means you can’t play.”
They like the fridge-gazer best: “That’s me when I’m hungry but nothing looks good.”
The “garbage on Monday morning” quote (Ukeles) makes total sense to them. Somebody has to take the bins out, and in their house it’s always Dad.
Wittgenstein’s line about the “problems of life” confuses them. “Which problems? Like when your fish dies?”

Questions they’d ask:

“Is magic real or not? Just tell me.”
“If you brush your teeth every day and they don’t get better, they just stay the same, is that still learning?”
“Why does the fish tank get green again if you already cleaned it?”
“Who cleans the stuff that the cleaners use?”
“What does ‘disenchanted’ mean? Did someone take the spell off the world? Who?”
“Why do I have to make my bed if I’m just going to mess it up tonight?”
“Is a bedtime story learning, making, or the other one? Because I already know the ending and I still want it.”
“What does ‘meaning’ mean? Is that a trick question?”
“If a rainbow is just light and water, is it still magic?”
“Why is grief not in the loops? What’s grief? Is it like when Grandpa’s dog died?”
“If nobody notices you tidied up, did it count?”

Where they’d get lost, bored, offended or unconvinced:

The earlier draft is really a chapter about choosing a mode. It hardly mentions maintenance or magic, the two words the title promised. The child would say, “You said magic and then you didn’t do any.”
The bulleted list of seven responses (“Repair a relationship… Change the conditions or seek help…”) is where they’d start flicking pages. Too many grown-up words in a row.
“Openness becomes a noble name for postponement” is lost on them completely. What they’d hear is “you’re just stalling,” which is what their mum says about getting dressed.
They’d be offended if the chapter says maintenance is boring. They’re proud of feeding the cat every morning.
The Weber quote uses the word “savage.” They’d ask why that’s allowed, and the chapter should say why it’s quoting it.
An example about rent and unsafe workplaces is real, but it’s outside their world. A small bridge would help, like “sometimes you can’t fix it yourself, you need a grown-up.”

Examples they’d bring:

Feeding the goldfish or cat every day. Nothing gets “better”, but if you stop, something very bad happens. That’s maintenance in its purest form.
Losing a tooth and the tooth fairy. They half-know it’s their parents and still want it to happen. That’s magic you keep on purpose, which is meaning.
A favorite blanket that has to be washed. After washing it smells wrong, and that matters more than it being clean.
Blowing out birthday candles and making a wish. It doesn’t “do” anything, but nobody skips it.
Refilling the bird feeder. The birds eat it all, so you fill it again, forever.

What would win them over:

An opening scene about something that has to be done every day and never finishes: the cat bowl, the bins, brushing teeth. Say out loud that it counts.
An honest answer to “is magic real?”, something like “not the wand kind, but here’s the kind that is,” given early on, not dodged.
One small ritual, like the birthday wish or the goodnight song, shown to matter even though it teaches nothing and makes nothing.
Short sections, each with a clear heading: Maintenance. Meaning. Magic. Right now the title’s three words are missing from the body.

Cynical adult

First reactions:

A chapter titled with three alliterative M-words, in a book already built on nine alliterative M-words, announcing that it will cover “what the M-words leave out.” I can smell the branding.
“Maintenance, meaning and magic” are three wildly different things. Grouping them because they start with M is exactly the kind of pattern-matching the book elsewhere warns against.
The earlier draft is not about maintenance, meaning or magic at all. It is a “how to choose a mode” decision guide with a bulleted list. The actually good line is buried: “A person unable to afford rent does not chiefly need a richer inner model of money.”
Ukeles’ “who’s going to pick up the garbage on Monday morning?” is the best thing on the page. If the chapter lived up to that line I’d keep reading.

Questions they’d ask:

If your framework needs a whole chapter explaining what it doesn’t cover, why should I adopt it rather than just use common sense?
Is “maintenance” really outside Learning/Creating/Becoming, or did you just not think about it until now? Changing the oil involves measuring and making.
Who does the maintenance in your household examples? Are you going to say it, or just gesture at “the same tired person”?
What do you actually mean by “magic”? Stage magic, religious ritual, the feeling of falling in love, or “technology I don’t understand”? Pick one.
Are you defending magic, or doing Weber’s disenchantment story with a wistful sigh?
Susan Wolf says meaning needs “objective attractiveness.” Who decides what’s objectively attractive? You, the author?
Wittgenstein said the problems of life aren’t touched by science. Then he also said to be silent about them. Are you going to be silent, or write a chapter?
Isn’t “give them their due without forcing them into the loops” just a way to say “these things matter” without saying anything about them?
Why is grief in a chapter with magic? That feels flippant toward grief.
What should I do differently on Monday after reading this?

Where they’d get lost, bored, offended or unconvinced:

The dinner scene with three people “doing competent but incompatible things” is a sitcom setup that has now appeared in several chapters. By chapter 17 it’s wallpaper.
The seven-bullet list (“Understand more / Try something small / Make room / Repair / Change conditions / Rest / Live with a loss”) is a flowchart pretending to be humility. It reads like a corporate wellness slide.
“Rest need not be a clever technique for producing more later” is correct, and it’s also said by every productivity influencer in 2024 while selling a rest course. The book needs to earn it rather than assert it.
“Magic” risks sounding either condescending (quaint rituals) or woo (the universe conspires). The Weber quote uses “savage”; a parenthetical note won’t stop a reader from wincing.
“Meaning” is the word most abused by self-help. Any sentence of the form “meaning arises when…” will make me close the book unless it’s followed by a hard case.

Examples they’d bring:

Ukeles’ Touch Sanitation (1979–80), in which she shook hands with New York City sanitation workers and said “Thank you for keeping New York City alive.” It’s a real, specific act that makes maintenance visible without mysticism.
Car maintenance nobody thanks you for: the brake job that means nothing happens. Maintenance succeeds when there’s no story to tell, which is why frameworks built on stories and loops miss it.
Stewart Brand, How Buildings Learn (1994): buildings that get used and maintained adapt, and “magazine architecture” that’s photographed and never maintained fails. That’s a book about maintenance beating design.
Arthur C. Clarke’s third law (“Any sufficiently advanced technology is indistinguishable from magic,” Profiles of the Future): your phone is magic to most of its users. What does the book say about living among magic you depend on but can’t inspect?
A parent’s bedtime routine that “means” something to a kid even though it’s a sequence of teeth, pyjamas, book. Meaning comes from repetition, not insight, and no loop improves it. It works because it doesn’t change.

What would win them over:

Admit plainly that MMM is a tool for problems, and most of life isn’t a problem. Say it in one sentence, early, then show three hard cases rather than a taxonomy.
Give maintenance its own concrete, unglamorous scene (dishes, the water main, a chronic illness regimen) with no redemptive twist at the end.
Define “magic” narrowly and defensibly, or drop it from the title.
Name who does the invisible work, and don’t moralize about it.

Believer / spiritual reader

First reactions:

Relief that the book finally admits its model has edges. A believer has been waiting since Chapter 13 for someone to say that prayer, grief and washing the dishes are not “stalled Creating.”
Wariness at the word “magic” sitting next to “meaning.” If prayer, blessing or sacrament ends up filed under magic, a Catholic, Orthodox Jew or Muslim reader will feel the book has put their deepest practice in the same drawer as fairground tricks.
The Weber quote, including “savage,” reads as a signal of whose side the book is on. Even with the note about period language, a reader from an Indigenous or African traditional religion will see it as the old evolutionary ladder (magic, then religion, then science) coming back.
The earlier draft (“Choosing a Mode”) is about switching modes at dinner. It never actually takes up maintenance, meaning or magic. A spiritual reader will feel the chapter title promises something the draft doesn’t deliver.
They’d be moved by “Grief is not stalled Creating” and “Rest need not be a clever technique for producing more later.” That is nearly Sabbath theology, and it’s worth saying so outright.

Questions they’d ask:

When you say “magic,” do you mean what anthropologists mean (ritual acts meant to compel outcomes), or do you mean prayer, blessing and sacrament? Those are very different, and my tradition insists on the difference.
Is “meaning” something humans make, or something we find, or something given to us? The Wolf quote (“subjective attraction meets objective attractiveness”) leaves room for the third option. Will you?
If maintenance doesn’t fit the loops, is that because it isn’t a loop, or because it’s a loop whose purpose isn’t improvement? A daily office or a daily puja is a loop that isn’t meant to get better.
Who does the maintenance in your examples? In my congregation it is mostly older women, and no one thanks them. Will the chapter name that?
Weber says the world is disenchanted. Is it, or have we just stopped paying attention? Where do you stand?
Wittgenstein says the problems of life “have still not been touched at all” by science. Do you agree with him, and if so, what does touch them?
Is there a place in this book for gratitude, which is neither learning nor making nor quite becoming?
Can something be meaningful if no one ever finds out about it, like a kindness done in secret (Matthew 6:3–4, “let not thy left hand know what thy right hand doeth”)?
Why is magic lumped in with maintenance? Is the claim that both are things rational people look down on?
If I light Shabbat candles every Friday, is that maintenance, meaning, magic or all three? Can you tell me without making me feel foolish?
Does “giving them their due” mean respect, or does it mean a polite pat on the head before getting back to the real framework?

Where they’d get lost, bored, offended or unconvinced:

Offended: “Magic” used as a catch-all for anything non-instrumental, and the Weber passage as the frame. The chapter should separate (a) magic as coercive technique, (b) religious practice as relationship or obedience, and (c) enchantment as a quality of attention. Collapsing them will lose every religious reader in the room.
Unconvinced: The earlier draft’s list (“Repair a relationship… Rest… Live with a loss”) treats these as exceptions to MMM, like leftovers. A believer would say they are the center of life and MMM is the exception.
Bored: The dinner-with-three-modes scene is being reused yet again. By Chapter 17 they want a different scene: a hospital room, a funeral, a Friday night.
Lost: “Meaning” is never defined in the draft. The quote options offer Wolf, but the draft doesn’t use her. Without a working definition the chapter can’t show where meaning escapes the loops.

Examples they’d bring:

The Rule of St. Benedict, ch. 31: the cellarer is to regard all the utensils and goods of the monastery “as if they were the sacred vessels of the altar.” Maintenance treated as holy, not as preparation for something else. That’s the chapter’s point, said 1,500 years ago.
Dōgen, Tenzo Kyōkun (Instructions for the Cook, 1237): a Zen master writing that cooking for the monks is itself full practice, not a lesser job. It pairs well with the Pirsig quote and shows an Eastern tradition refusing the maintenance/meaning split.
Brother Lawrence, The Practice of the Presence of God (compiled 1690s): a Carmelite lay brother who worked in the monastery kitchen and said he felt as close to God among the pots as at the altar. Ordinary upkeep as the place where meaning happens, not a break from it.
Sitting shiva and saying Kaddish (Judaism): a structured, repeated mourning practice that doesn’t try to fix or learn from the loss. The Mourner’s Kaddish doesn’t even mention death. It shows grief held by practice rather than solved by a loop.
Wittgenstein, Tractatus 6.522 (Ogden trans.): “There is indeed the inexpressible. This shows itself; it is the mystical.” It follows directly from the 6.52 quote the chapter already has, and a believer would want the book to include it rather than stop one line early.
Charles Taylor, A Secular Age (2007): his “porous self” versus “buffered self” is a more careful account of disenchantment than Weber’s line and doesn’t carry the “savage” baggage. It lets the chapter talk about what changed without saying who was primitive.

What would win them over:

An honest taxonomy: here is magic in the coercive sense, here is devotion, here is enchantment as attention, and the book treats each fairly and separately.
Admitting that for billions of people meaning is received, not constructed, and that the book can’t settle the question either way, only mark it as unknown. That would honor the humility contract from the Prologue.
Showing maintenance as sacred in some traditions and simply dignified in secular terms, with Ukeles and Benedict side by side, not one used to explain away the other.
Letting at least one example stand without translating it into MMM at all.

Skeptical scientist

First reactions:

The brief promises three things (maintenance, meaning, magic) and the earlier draft delivers none of them by name. It is a “when to switch modes” chapter with a list of non-MMM responses (rest, repair, help, grief). Useful material, wrong chapter shape. The Ukeles and Weber quotes are doing all the work the draft doesn’t.
Maintenance is the easiest of the three to ground in physics and biology, and the brief undersells it. Every living thing spends most of its energy budget not learning, creating or becoming, but staying the same against entropy. That is the strongest possible argument that MMM is not the whole of life, and it’s currently absent.
“Magic” is used without a definition. Weber’s Entzauberung (disenchantment) is a sociological claim about calculability, not a claim that wonder vanished. If the chapter slides between “magic” as enchantment, as ritual efficacy, and as the unexplained, a scientist will stop trusting it.
“Meaning” is the hardest to handle without hand-waving. Wolf’s formula is good because it has two testable-ish parts (subjective attraction, objective worth). Keep it and resist anything vaguer.

Questions they’d ask:

What exactly counts as maintenance here: keeping a state constant (homeostasis), restoring it after damage (repair), or replacing parts while keeping the pattern (turnover)? Those are different mechanisms.
If maintenance is “outside” MMM, why isn’t it just Learning with a fixed setpoint? A thermostat measures, models (crudely) and manipulates. Where’s the line?
Is the claim that maintenance can’t be described by the loops, or that describing it that way misrepresents what matters about it? Those are different claims and need different support.
Which definition of magic is in play: Frazer’s sympathetic magic, Weber’s enchantment, stage magic, or “effects we can’t yet explain”?
Is the chapter saying ritual works, feels like it works, or matters whether or not it works?
When you say something has “meaning,” is that a psychological state people report, a property of the activity, or both? What would count as evidence either way?
Placebo effects are real but mostly on self-reported outcomes. Does the chapter want to lean on them as “magic that works”? If so, with what caveats?
Who does the maintenance in your examples, and is the chapter about the work itself or its invisibility? Ukeles is about the second.
Is “the model stops here” a permanent limit or just the current reach of the model? The chapter should say which.
If grief “is not stalled Creating,” what is it, mechanistically? Is the book comfortable saying “we don’t have a good model”?

Where they’d get lost, bored, offended or unconvinced:

Unconvinced by the bullet list of seven responses in the draft. It reads like a taxonomy, but there’s no rule for when a situation is “material” versus “informational”; the categories overlap (rest can be repair; seeking help is often Creating).
Lost at “How do you know when to switch? Look for evidence, not a personality label.” Which evidence? “New observations keep changing the explanation” is a decent operational criterion; the others (“repeated effort is narrowing perception”) aren’t measurable at all.
Mildly offended if “magic” gets a respectful treatment that slides into implying efficacy for things that have been tested and failed (homeopathy, prayer-at-a-distance: the STEP trial, Benson et al. 2006, American Heart Journal, found no benefit of intercessory prayer for cardiac bypass patients, and slightly more complications in the group told they were being prayed for).
Bored if Wittgenstein 6.52 is used as a trump card. It’s an assertion, not an argument, and it cuts both ways.

Examples they’d bring:

The energy budget of staying the same. A resting human spends a large share of daily energy on basal metabolism: ion pumping (Na+/K+-ATPase alone accounts for a substantial fraction of resting ATP use, especially in neurons), protein turnover, and repair. Most of what a body does, it does to not change. That is maintenance as the default activity of life, not a side chore.
DNA repair. The 2015 Nobel Prize in Chemistry (Lindahl, Modrich, Sancar) went for mechanisms of DNA repair. Lindahl’s starting point was that DNA is chemically unstable and decays measurably in water; life persists only because repair runs constantly. A precise, uncontroversial case of maintenance as a condition for everything else.
Homeostasis as a named idea. Walter Cannon coined “homeostasis” in The Wisdom of the Body (1932). Sterling and Eyer later proposed “allostasis” (1988), stability through anticipatory change, which is a nice case of maintenance that does use prediction, and so a real test of where the chapter’s boundary lies.
Infrastructure failure from deferred maintenance. The I-35W Mississippi River bridge collapse in Minneapolis (2007); the NTSB attributed it to undersized gusset plates in the original design, compounded by added load over the years. It’s a good corrective example: not every collapse is neglect, and the chapter should get the cause right if it uses it.
Open-label placebo. Kaptchuk et al. (2010, PLoS ONE) gave IBS patients pills they were told were placebos; they reported more symptom improvement than no-treatment controls. Small trial, self-reported outcomes, but a genuinely strange result about ritual and expectation that doesn’t require deception. Useful for “magic” handled honestly.
Rituals and anxiety. Norton and Gino (2014, Journal of Experimental Psychology: General) reported that performing rituals after losses reduced grief; Brooks et al. (2016, Organizational Behavior and Human Decision Processes) reported rituals reduced anxiety before performance tasks. Both are lab studies with modest samples; flag that they haven’t been widely replicated.

What would win them over:

One precise definition for each of the three words, stated early, and held to.
Maintenance grounded in thermodynamics and physiology first, so the chapter’s claim (“most of life is keeping going, not getting better”) rests on something solid before it moves to the social and personal.
A clear statement of which of these the model cannot handle in principle versus which it just handles badly, with the book admitting when it doesn’t know.
On magic: separate “the ritual has measurable effects via expectation or anxiety reduction” from “the ritual does what it says it does.” Take both seriously, conflate neither.

Earlier draft

From “Choosing a Mode”

At ten minutes past seven, three people are doing competent but incompatible things. One keeps reading recipes. One has begun chopping without checking what anyone else planned. One is gazing into the open refrigerator, hoping dinner will reveal itself.

No mode is bad. Each can be wrong for the moment.

Learning helps when a consequential uncertainty can be reduced. Why did the sauce split? What does the guest actually avoid? It becomes less useful when more information will not change the decision, or when research is protecting us from making one.

Creating helps when contact with reality would teach more than another plan. Cook a small batch. Move the chairs. Send the first honest sentence. It becomes less useful when action is scattering effort, imposing on someone else, or disguising the absence of a purpose.

Becoming helps when force and premature categories are closing attention. Sit with the music. Let several possibilities remain alive. Notice the person rather than assigning them a role in your solution. It becomes less useful when openness becomes a noble name for postponement.

These are situational judgments, not stages of development. Stress does not send everyone down a universal ladder, and age does not confer receptivity. A child may notice what every adult missed. An expert may need deliberate measurement. Any person may need two modes at once: make an apology while remaining open to a response they cannot control.

A useful question is: what kind of response could this situation support?

Understand more when relevant evidence is available and the stakes justify looking.
Try something small when feedback would be informative and the attempt is safe enough.
Make room when the outcome depends on other agents, living processes, or meanings that cannot be forced.
Repair a relationship when harm or broken trust is the central fact.
Change the conditions or seek help when the obstacle is material, institutional, medical, or beyond one person’s capacity.
Rest when effort itself has become part of the damage.
Live with a loss when no mode can restore what is gone.

The last four prevent MMM from swallowing the world. A person unable to afford rent does not chiefly need a richer inner model of money. Someone in an unsafe workplace may need allies, protection, or an exit. Grief is not stalled Creating. Rest need not be a clever technique for producing more later.

How do you know when to switch? Look for evidence, not a personality label. If new observations keep changing the explanation, continue Learning. If the explanation has stabilized but nothing has met the world, try Creating. If repeated effort is narrowing perception or violating another person’s agency, pause and receive. If exhaustion, danger, or isolation dominates, the next move may be support.

Sometimes the goal deserves challenge before any mode is chosen. Becoming faster at answering every message may intensify an unreasonable expectation. Improving the dinner may matter less than noticing who has cooked and cleaned every week.

The modes are instruments. Judgment includes knowing when to put them down. The next chapter follows what happens when we use them badly, lose what we learned, and begin again without pretending every return is upward.