20Chapter 20: Raising the Floor

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: Argue that society and civilization are about raising the floor as much as the roof, using the parable of the long spoons. Present program theory, which spells out how a collective effort is supposed to produce its outcomes so the assumptions can be tested, as a way to be objective about collective subjectivity.
• Serves: It turns the book’s tools on collective life, and gives the reader a way to judge institutions by what they do for everyone.

Quote options

1. “The message that we should feed and care for each other, and that only in this way can we overcome the curse of selfishness, is such a high and holy and universal message that the story deserves to be claimed by all.” — Rabbi Marc Gellman, “Famous spiritual story varies among faiths and cultures, but carries the same message,” The God Squad column (Tribune Media, 2012), on the long spoons parable. [verified: https://www.arcamax.com/healthandspirit/religion/godsquad/s-1191693-115819] Why: A specific, citable telling of the long spoons that names the parable’s multi-tradition, authorless status, which avoids a false attribution.

2. “Justice is the first virtue of social institutions, as truth is of systems of thought. A theory however elegant and economical must be rejected or revised if it is untrue; likewise laws and institutions no matter how efficient and well-arranged must be reformed or abolished if they are unjust.” — John Rawls, A Theory of Justice (1971), §1. [verified: https://oyc.yale.edu/node/1545 (first sentence); full passage matches https://quotefancy.com/quote/1378711/John-Rawls-Justice-is-the-first-virtue-of-social-institutions-as-truth-is-of-systems-of] Why: Pairs truth-in-thought with justice-in-institutions, the chapter’s move from judging claims to judging institutions.

3. “Development requires the removal of major sources of unfreedom: poverty as well as tyranny, poor economic opportunities as well as systematic social deprivation, neglect of public facilities as well as intolerance or overactivity of repressive states.” — Amartya Sen, Development as Freedom (1999), Introduction. [verified: https://tcagley.wordpress.com/2025/01/18/development-as-freedom-by-amartya-sen-week-2-preface-and-introduction/] Why: States “raising the floor” concretely: progress measured by what is removed from beneath the worst-off, not by the height of the roof.

4. “Grounding evaluation in theories of change takes for granted that social programs are based on explicit or implicit theories about how and why the program will work. The evaluation should surface those theories and lay them out in as fine detail as possible, identifying all the assumptions and sub-assumptions built into the program.” — Carol H. Weiss, “Nothing as Practical as Good Theory,” in New Approaches to Evaluating Comprehensive Community Initiatives (Aspen Institute, 1995), pp. 66–67. [verified: https://aea365.org/blog/memorial-week-in-evaluation-the-classic-nothing-as-practical-as-good-theory-by-carol-h-weiss-by-sharon-rallis/] (source reproduces “lay then out”, an evident typo for “them”) Why: The founding statement of program theory: making a collective effort’s hidden assumptions explicit so they can be tested.

5. “The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.” — Donald T. Campbell, “Assessing the Impact of Planned Social Change” (1976). [verified: https://en.wikipedia.org/wiki/Campbell’s_law] Why: The warning label for program theory: objectivity about collective life degrades once the measure becomes the target.

Research

Core argument

1. Start at the table, not the summit. The long-spoons parable: heaven and hell have the same room, the same food, and the same unwieldy spoons. The only difference is whether people feed each other. The tools and the constraint are identical, and the outcome depends on arrangement. This is the book’s systems lens (Ch. 5) turned on a group: the environment is the same, and what changes is how the agents relate.
2. Civilization has two measures, a roof and a floor. The roof is the best that anyone can do: the cathedral, the moon landing, the richest person’s life. The floor is the worst that anyone is allowed to fall to: famine, untreated smallpox, sleeping rough. Histories usually celebrate roofs, but most of what people actually live inside is set by the floor. Smith already said a society cannot be “flourishing and happy” if most of its members are “poor and miserable.”
3. The floor is real but it moves. Some of it is biological (calories, clean water, no variola). Some of it is social: Smith’s linen shirt, which is “not a necessary of life” but whose absence marks disgrace. The floor is a collective representation (Ch. 6, Ch. 10), and it still bites physically. That is why floors have to be argued over and can’t just be looked up.
4. Raising the floor is often how the roof gets built. Curb cuts, sanitation, literacy, vaccination and Haber-Bosch fertilizer all widened who could take part, and so widened who could contribute. Cooperation is old: vampire bats share blood meals, and Ostrom’s irrigators and herders keep commons going for centuries. The “unnatural” part is the scale and deliberateness with which humans do it.
5. Good intentions are not a program. Collective efforts run on stories about causes: “if we scare kids with prison, they’ll stop offending.” Scared Straight and the Cambridge-Somerville Youth Study show that caring, confident programs can make the floor lower for the very people they target.
6. Program theory is objectivity applied to collective wanting. Write down the causal chain (inputs → activities → outputs → outcomes), name each assumption at each link, and decide in advance what would show a link is broken. It is Ch. 14’s measure-model-manipulate loop, done in public with other people’s lives at stake. It turns shared subjectivity (what “we” want) into claims that can be checked.
7. Measure the floor, but watch the measure. Campbell’s law (already in the quote options) and Scott’s “legibility” warn that the indicator becomes the target and that schemes imposed from above crush local knowledge. Ostrom’s design principles are the corrective: the people living under a rule help make it, monitor it and revise it.
8. Judge institutions by who they hold up. A usable test for any institution (school, app, law, company, church, family): what does it do for the person with the least power inside it, and how would we know? This is Rawls’s difference principle in plain terms, and Sen’s “remediable injustices” turn it into something you can act on.
9. Close humbly. No committee controls the whole (the printing and Internet material from the earlier draft). Repair happens in layers, and the chapter should offer a way to look rather than a blueprint. That hands off to Ch. 21, where inherited practice often already encodes floor-raising wisdom.

Key questions

1. In the long-spoons story, why doesn’t anyone in hell just try feeding a neighbor? What stops the first move: pride, fear of not being fed back, or simply not noticing?
2. A child’s question: if everyone has enough food, why are some people still hungry?
3. Is a society better if its best life gets better, or if its worst life does? Can you have one without the other?
4. Where is the floor, and who decides? Is it calories, dignity, or Smith’s linen shirt: whatever you would be ashamed to go without?
5. Does raising the floor lower the roof (taxes, incentives, “tall poppy”), or build it (curb cuts, literacy)? When has each happened?
6. The cynic’s question: aren’t most “floor-raising” institutions really for the people who run them?
7. Why do well-meant programs sometimes hurt the people they aim to help (Scared Straight)? What did their designers believe that was never tested?
8. What would it mean to be “objective” about something that exists only because many people want it, such as a school or a currency?
9. If you had to draw the arrows from “we spend money” to “children read better,” how many links would there be, and which is the weakest?
10. Who gets to write the program theory: funders, staff, or the people it is meant to help? Does it change the answer?
11. When a measure becomes a target (test scores, arrest counts, engagement), how can a group keep caring about the thing itself?
12. Do other animals raise floors? Vampire bats share blood, but only with bats that shared before. Is reciprocity care, or insurance?
13. Is the welfare state a long spoon, or a waiter who feeds everyone?
14. Why do commons sometimes last for centuries (Törbel’s alpine meadows) when theory said they would collapse (Hardin)?
15. What does the floor owe to people outside our borders? King’s “whatever affects one directly, affects all indirectly” versus the plain fact that nations budget for citizens.
16. Can a floor rise too fast, for example when a top-down plan skips local knowledge (Scott’s high modernism)?
17. What is the floor in a family, a friendship, a workplace, an online community?
18. If you would not know in advance where you’d land in a society (Rawls’s veil), which one would you pick, and is that thought experiment even possible for a creature who is always somebody (Ch. 4)?
19. How do we tell that an institution is learning (Ch. 18) rather than only getting better at preserving itself?
20. What is an institution’s “exercise for the reader”: what could one person check this week about something they belong to?

Examples

1. The long spoons (cross-tradition parable). Often attributed to Rabbi Haim of Romshishok. Variants appear in Jewish, Christian, Buddhist, Hindu and Chinese tellings, some with chopsticks and some with splinted arms (Wikipedia summary). Modern echoes include the film The Platform (2019), a vertical prison where food descends floor by floor and there is literally a floor at the bottom. It could open or close the chapter.
2. Vampire bats (animals). Gerald Wilkinson (1984) showed that wild Desmodus rotundus regurgitate blood to roostmates who failed to feed, based both on kinship and on the chance to reciprocate. A bat that misses two or three nights of feeding can starve, so the group sets a survival floor (Nature 308:181).
3. Cells and cancer (cells). Multicellularity is the ultimate floor-raising pact: cells give up reproducing independently in return for a supply of oxygen and nutrients. Cancer is a defector. Link this to Maynard Smith and Szathmáry’s Major Transitions (in references).
4. Smallpox eradication (civilization). The WHO estimated 15 million cases and 2 million deaths in 1967 alone. The last natural case came in October 1977, and eradication was certified in May 1980 (Wikipedia/WHO). The floor was raised for everyone forever, and it only worked because surveillance-and-containment reached the worst-covered villages. A disease cannot be eradicated just for the rich.
5. Haber-Bosch (civilization). Synthetic ammonia underlies inorganic fertilizer. Smil wrote that without it “nearly half the world would go hungry” (Nature 400:415, 1999). It is a roof achievement that became a floor, and it has a nitrogen-pollution cost, which matters for the “no free floor” point.
6. Curb cuts (city). Kalamazoo piloted curb cuts in the 1940s for disabled veterans. Berkeley activists poured their own ramps in the early 1970s, the city cut its first official curb at Telegraph Avenue in 1972, and the ADA (1990) made them national. Angela Glover Blackwell’s “curb-cut effect” names how designs made for the margin serve everyone: strollers, carts, cyclists (SSIR, 2017; Wikipedia).
7. Scared Straight (program theory failure). The 1978 Rahway prison documentary won an Oscar. Finckenauer’s controlled study the same year found that participants were more likely to offend. The Petrosino et al. Campbell/Cochrane review concluded that such programs increase offending, and a 2004 Washington State analysis estimated about $203 of social cost per dollar spent (Wikipedia summary with sources). The unstated theory (“fear deters”) was never written down, so it was never tested before scaling.
8. Cambridge-Somerville Youth Study (program theory failure). From 1939, about 500 Massachusetts boys were matched in pairs and one of each pair received years of counseling, tutoring, camps and medical care. Joan McCord’s 30-year follow-up (American Psychologist, 1978, doi:10.1037/0003-066X.33.3.284) found the treated group did no better and on several measures worse. Participants themselves reported that the program had helped them. Warmth is not evidence. Numbers should be checked against the paper before use.
9. Progresa/Oportunidades (program theory success). Mexico, 1997. Cash went mostly to mothers, conditional on school attendance and clinic visits, and it was rolled out with a randomized evaluation by IFPRI (1997–2000). The phased rollout became the control group, which made the theory testable, and it helped bring RCTs into policy worldwide (Wikipedia).
10. Microcredit’s modest results (hype check). It was celebrated with a Nobel Peace Prize in 2006. Six randomized evaluations across six countries (Banerjee, Karlan and Zinman, AEJ: Applied 2015) found modest, not transformative, effects (AEA). The theory “credit → enterprise → escape from poverty” broke at the middle link for most borrowers.
11. Finland’s Housing First (floor as policy). Since 2008 Finland has given people a home first and treated addiction and mental-health problems second. Homelessness fell by roughly 30–50% and long-term homelessness by more than 35%, with estimated savings of at least €15,000 per person per year. It rose again in 2025 after benefit cuts (Wikipedia with sources). It shows the floor held up by upkeep, and falling when the upkeep stops.
12. Törbel and Ostrom’s commons (village). Swiss alpine villagers have governed shared meadows and forests under written rules since at least the 13th century, and similar cases include Japanese villages and Spanish huertas. Ostrom’s eight design principles came from comparing long-lived and failed commons. Across more than 100 later studies, about two-thirds confirmed the pattern (Ostrom Nobel lecture, pp. 421–422).
13. Seeing Like a State (high-modernist floor-raising gone wrong). Tanzania’s compulsory ujamaa villages, Soviet collectivization and Brasília’s planning each tried to raise living standards from above by making people legible, and each broke local know-how (Yale UP description).
14. The garden committee (from the earlier draft; small scale). The watering rule “works” (the tank lasts all week), yet the far beds starve and the wheelchair user can’t reach the valve. It is a ready-made program theory exercise: goal, assumption (“equal pressure at every plot”), unmeasured outcome.
15. Internet access (global floor). The ITU (2024) found 93% of people online in high-income countries against 27% in low-income ones (from the earlier draft). The roof (frontier tech) races ahead while the floor lags.
16. Thought experiment: the veil at the dinner. You’ll be seated at a random place at the long-spoon table: next to a generous neighbor, a greedy one, or at the end with no neighbor at all. Which seating rules would you vote for before you know? That is Rawls in a sentence.

Source material

1. “No society can surely be flourishing and happy, of which the far greater part of the members are poor and miserable.” — Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (1776), Book I, ch. 8 (“Of the Wages of Labour”). [verified: https://www.gutenberg.org/cache/epub/3300/pg3300.txt] Use: This comes from the supposed father of market economics, not a socialist, so it disarms the “floor-raising is a left-wing idea” reading.

2. “A linen shirt, for example, is, strictly speaking, not a necessary of life. The Greeks and Romans lived, I suppose, very comfortably, though they had no linen. But in the present times, through the greater part of Europe, a creditable day-labourer would be ashamed to appear in public without a linen shirt” — Adam Smith, Wealth of Nations (1776), Book V, ch. 2, Part II, Art. IV (taxes on consumable commodities). [verified: https://www.gutenberg.org/cache/epub/3300/pg3300.txt] Use: The floor is partly social: it is defined by what shames. It ties to Ch. 10 (thingification) and supports the point that the floor moves.

3. “Sociability is as much a law of nature as mutual struggle.” — Peter Kropotkin, Mutual Aid: A Factor of Evolution (1902), ch. I. [verified: https://www.gutenberg.org/cache/epub/4341/pg4341.txt] Use: It is a counterweight to “nature red in tooth and claw” and connects to Ch. 2 (conflict) and Ch. 3 (care). Flag that Kropotkin overstated his case.

4. “Designing institutions to force (or nudge) entirely self-interested individuals to achieve better outcomes has been the major goal posited by policy analysts for governments to accomplish for much of the past half century. Extensive empirical research leads me to argue that instead, a core goal of public policy should be to facilitate the development of institutions that bring out the best in humans.” — Elinor Ostrom, “Beyond Markets and States: Polycentric Governance of Complex Economic Systems,” Nobel Prize Lecture, 8 December 2009, pp. 435–436 (also American Economic Review 100(3), 2010). [verified: https://www.nobelprize.org/uploads/2018/06/ostrom_lecture.pdf] Use: This is the long spoons in policy language. Institutions should be judged by whether they make feeding each other possible, not only by whether they coerce. It runs a little over 50 words and could be trimmed to the second sentence.

5. “They are to be to the greatest benefit of the least-advantaged members of society (the difference principle)” — John Rawls, Justice as Fairness: A Restatement (2001), §13, pp. 42–43, as quoted in Leif Wenar, “John Rawls,” Stanford Encyclopedia of Philosophy, §4.3. [verified: https://plato.stanford.edu/entries/rawls/] Use: This is the formal name for “judge by the floor.” Check the page number against the print edition. The quotation from A Theory of Justice §1 in the Quote options is a separate passage.

6. “What moves us, reasonably enough, is not the realization that the world falls short of being completely just – which few of us expect – but that there are clearly remediable injustices around us which we want to eliminate.” — Amartya Sen, The Idea of Justice (2009), Preface. [verified: https://en.wikiquote.org/wiki/Amartya_Sen] (Wikiquote gives Preface without a page; confirm against the print edition, p. vii.) Use: It links to Ch. 1 (“what moves us” is wanting) and frames floor-raising as fixing particular, nameable things rather than waiting for utopia. That is exactly the program-theory temperament.

7. “Injustice anywhere is a threat to justice everywhere. We are caught in an inescapable network of mutuality, tied in a single garment of destiny. Whatever affects one directly, affects all indirectly.” — Martin Luther King Jr., “Letter from Birmingham Jail” (16 April 1963). [verified: https://www.africa.upenn.edu/Articles_Gen/Letter_Birmingham.html] Use: It is the moral version of the curb-cut effect and of Ch. 5’s claim that no system is isolated.

8. “food sharing by regurgitation of blood among wild vampire bats ( Desmodus rotundus ) depends equally and independently on degree of relatedness and an index of opportunity for reciprocation.” — Gerald S. Wilkinson, “Reciprocal food sharing in the vampire bat,” Nature 308 (1984): 181–184, abstract. [verified: https://www.nature.com/articles/308181a0] Use: It shows floor-raising below the human level, with a mechanism (reciprocity plus kinship) the chapter can contrast with human institutions that feed strangers.

9. “Without ammonia, there would be no inorganic fertilizers, and nearly half the world would go hungry.” — Vaclav Smil, “Detonator of the population explosion,” Nature 400 (1999): 415 (standfirst). [verified: https://www.nature.com/articles/22672] Smil’s Enriching the Earth (2001) is in references.json. Use: A single chemical process is the floor under billions. It shows that roof tech can become floor infrastructure, with costs attached.

Counterarguments and limits

• The long spoons assume abundance. In the parable the food is plentiful and the only problem is coordination. Real scarcity (triage, lifeboats) is harder. The chapter shouldn’t imply every floor problem is a failure of goodwill.
• Floors vs. incentives. Critics (Hayek, public-choice theorists) argue that guaranteed floors can dull effort, grow bureaucracy and crowd out mutual aid. Evidence is mixed: cash-transfer studies mostly find little reduction in work, but some programs create poverty traps at benefit cliffs. Present this honestly.
• Who defines the floor? A floor set from above can be paternalistic, and Scott’s cases show this. Nussbaum’s threshold list has been criticized as parochial. Sen deliberately refuses a fixed list. The chapter has to pick a position or show the tension.
• Program theory can become box-ticking. Logic models are often written after the fact to satisfy funders, and theories of change can be “fashion,” with arrows standing in for thought. Campbell’s law applies to evaluation itself.
• Not everything valuable is a program. Care, friendship and ritual (Ch. 16, 17, 21) resist input→outcome diagrams. Applying program theory to a family would be a category error. Say where the tool stops.
• RCT limits. Randomized evidence says what worked on average, somewhere, once (Pawson and Tilley’s realist evaluation asks “for whom, in what circumstances”). Deaton and Cartwright have argued that RCT results don’t travel automatically.
• Survivorship in Ostrom’s cases. Long-lived commons are visible because they survived. The design principles are “best practices,” as Ostrom herself says in footnote 5 of the lecture, not laws.
• Kropotkin overreach. Nature has plenty of cheating, parasitism and infanticide. Mutual aid is one pattern, not the moral of evolution. Avoid “nature is cooperative, therefore…” (the naturalistic fallacy, and ironic for a book called Unnatural).
• Roof versus floor may be a false split. Some roof gains come from concentrated resources (patrons funding science, monopoly profits funding R&D). Claiming the two always align would be a just-so story.
• “Judge institutions by what they do for everyone” can erase differences. Universal policies can miss people with specific needs. The curb cut was targeted, and it helped everyone only as a side effect.
• Apocryphal anecdotes. The popular “Margaret Mead said the first sign of civilization was a healed femur” story has no verified source. If it is used at all, present it as folklore.

Connections

• Ch. 1 (Before Thought, There Is Want): collective wanting is still wanting, and Sen’s “what moves us.”
• Ch. 2 (Will Creates Difference): conflicting wants at the table, and hell as everyone reaching for their own mouth.
• Ch. 3 (Love, Growth, and Meaning): feeding the other is care writ large, and this chapter scales Ch. 3’s ethics up to institutions.
• Ch. 4 (From Subjectivity…): the veil of ignorance as an attempt at objectivity about shared subjectivity, and whose account gets heard.
• Ch. 5 (The First Blade): no system in isolation, as in King’s “network of mutuality.”
• Ch. 7 (The Third Blade: Good, Bad): floors require value cuts, deciding what counts as unacceptable.
• Ch. 9 (The Cost of the Blade) / Ch. 10 (Thingification): indicators and categories (the “poverty line”) become things. See Campbell and Scott’s legibility.
• Ch. 14 (Learning): program theory is measure-model-manipulate for groups, and one study rarely settles a cause (Cambridge-Somerville vs. participant testimony).
• Ch. 15 (Creating): stopping rules and cheap attempts, such as Progresa’s phased rollout as a cheap, revisable test.
• Ch. 17 (Maintenance, Meaning, and Magic): floors are maintenance. Finland’s homelessness rose again when the upkeep was cut.
• Ch. 18 (Loops That Learn and Loops That Don’t): institutions that learn versus those that only preserve themselves (the garden committee).
• Ch. 19 (What Knowledge Needs to Stay Alive): the long-spoons parable spreading laterally across religions, as an example of a meme with no single author.
• Ch. 21 (Practices Before Patterns): tithing, zakat, gleaning laws (Leviticus 19:9–10) and potlatch are inherited floor-raising practices that predate program theory.
• Ch. 22 (A Better Tuesday): the exercises below scale down into one person’s week.

Exercise ideas

1. Find the floor of something you belong to. Instruction: Pick one group you’re part of (workplace, school, household, club, online community). Write down (a) its roof: the best outcome anyone in it gets, (b) its floor: the worst outcome it currently permits for a member, and (c) one person, by role rather than name, who is closest to that floor. Then ask that person, or imagine as specifically as you can, what one change would raise their floor. Notice: How easily you could name the roof compared with the floor. Whether the floor was ever discussed in the group. Whether the fix would also help people who aren’t at the floor (a curb-cut effect). Why: It makes “judge institutions by what they do for everyone” a thing you do rather than a slogan, and it shows that floors are usually invisible to those standing well above them.

2. Draw the long spoon: a five-box program theory. Instruction: Choose one effort meant to help people (a charity you give to, a school policy, a company wellness program, your own New Year’s plan for your family). On paper, draw five boxes in a row: Resources → Activities → Immediate result → Change in behavior → Better life. Fill each box in one line. Under each arrow, write the assumption that must be true for the next box to follow (“parents will read the newsletter”). Circle the weakest arrow, and write one piece of evidence that would show it is broken and where you could find it. Notice: Which arrows you had never thought about. Whether the “better life” box is measured at all, or only the activities are (“we held 12 workshops”). Whether the people in the last box had any say in drawing it. Why: This is program theory at home scale. It shows how collective good intentions carry hidden, testable claims, which is the chapter’s route to being objective about shared subjectivity (and a guard against another Scared Straight).

3. The veiled seat (a five-minute thought experiment). Instruction: Take one rule you would like to change or keep (a tax, a house rule, a school admissions policy). Write down your position. Then list three people the rule affects very differently, and imagine being assigned one of them at random tomorrow. Would you keep your position? Write the sentence you’d need to add to make the rule acceptable from every seat. Notice: Where your view changed, and where you couldn’t really imagine another seat, which is the limit Ch. 4 warns about. Why: It puts Rawls’s difference principle into practice and ties floor-raising back to the book’s recurring theme of objectivity as deliberate work.

Open questions for the author

• Tone of the parable: open with the long spoons as a fable, or with a real case (Scared Straight or curb cuts) and reveal the parable later? The Platform is a vivid modern hook but a grim one.
• How political to be: the chapter can stay at “judge by the floor” without taking positions on taxes and welfare, or it can argue for specific floors. Decide whether you want Rawls or Sen (or neither) as the anchor.
• Which floor concept: absolute (calories, disease), relative (Smith’s shirt), or capability-based (Sen/Nussbaum)? Each leads to a different chapter.
• How much program-theory jargon: “logic model,” “theory of change,” “program theory” and “realist evaluation” overlap. Pick one term and define it once.
• Scope of scale: include cells and bats (continuity with Act I) or keep strictly to human institutions (more “unnatural”)?
• Keep the printing/Internet material from the earlier draft here, or move it to Ch. 19 (knowledge persistence), where it may fit better?
• The garden committee: keep it as a recurring small-scale case alongside a large one (smallpox or Housing First)?
• Global versus local floor: address obligations across borders (King, ITU data), or keep to the reader’s own communities?
• The “unnatural” hook: is the chapter’s claim that deliberate floor-raising for strangers is uniquely human? If so, state it and defend it against the bat and ant counterexamples.

Reader perspectives

Curious young child

First reactions:

• “Raising the floor? Like jacking up the house?” They’ll take it literally, which is useful. Picture a floor lifting so short people can reach.
• The long-spoons parable is a perfect children’s story (people who can’t feed themselves with long spoons but can feed each other), and it’s their favorite part of the brief. They’d be upset to learn it’s missing from the earlier draft.
• The wheelchair gardener who can’t reach the new valve makes them indignant right away: “That’s not fair, move it down!”
• Printing presses and the Internet are “history class,” and they’ll glaze over unless there’s a hook.

Questions they’d ask:

1. “Why were the spoons so long? Who made them that long? Was it a mean person?”
2. “Why didn’t the people in the bad place just feed each other too? Were they stupid?”
3. “Can you just cut the spoons shorter?”
4. “What’s the roof in ‘raising the roof’? Is that the rich people?”
5. “If the tank lasts all week, why isn’t that good enough?”
6. “What’s a ‘program theory’? Is it a computer thing?”
7. “Why does the Internet have a book of how it started? Did one person invent it?” (The draft’s line about there being no big red INTERNET button would make them laugh.)
8. “Why do some countries not have Internet? Can we send them some?”
9. “If you give a gold star for reading lots of books, won’t people just read tiny books?” (This is Campbell’s law, straight out of the child.)
10. “Who decides what’s fair: the teacher, or everyone?”
11. “Is it fairer if everybody gets the same, or if everybody gets what they need?”

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

• The earlier draft never tells the long-spoons story, even though the brief is built around it. That’s the thing they’d understand best.
• Program theory is missing from the draft too. The Weiss quote (“surface those theories… identifying all the assumptions”) makes no sense to them without a worked example like “we think X will cause Y because Z.”
• The Gutenberg and Dittmar sections are long, with dates, cities and a journal name. Unless there’s a picture of a printing press, they’re lost by “financiers accepted risks.”
• “Capability is not moral progress with better stationery” is a grown-up joke that goes over their head.
• The draft ends with “The next chapter asks how far the book’s pattern can travel,” but the next chapter is about rituals. That’s a broken promise, and a kid does notice when “next time” is wrong.
• The Rawls and Sen quotes are two heavy sentences in a row. Keep one, or give a kid-sized paraphrase.

Examples they’d bring:

• The playground where the monkey bars are too high for the little kids. Do you build higher bars for the big kids or a lower set for everyone?
• The class reward chart. Campbell’s law in the wild: kids who game “most books read” by choosing the thinnest ones.
• Sharing snacks at a party. Cutting the cake so the last kid in line still gets a piece.
• A step stool at the bathroom sink, so the smallest person can wash their hands. Floor raised, literally.
• A group project where one kid does all the work. That’s the “who carries the labor” question.

What would win them over:

• Opening with the long-spoons story told simply, then asking “so which room is our town?”
• One tiny program theory a child could write: “We think a step stool → little kids reach the sink → fewer germs. Let’s check if the germs part happened.”
• Returning to the wheelchair gardener at the end and showing the valve actually moved.
• Cutting printing and the Internet down to one vivid picture each, with the stats saved for grown-ups.

Cynical adult

First reactions:

• “Raising the floor, not just the roof” is a good slogan. It’s also a slogan, and I’ve seen it on political mailers.
• The long spoons parable is lovely and so familiar it’s been embroidered on pillows. Using it means the chapter has to do something with it that the pillow doesn’t.
• “Program theory” is what NGOs put in grant applications as a “theory of change” diagram that nobody reads after funding. If the book presents it as a fresh tool for objectivity, I’ll laugh.
• The earlier draft is really a history of printing and the internet. It’s well sourced and competent, but it isn’t about the floor, the spoons or program theory. The wheelchair-and-valve detail in the opening is the only floor-raising moment.

Questions they’d ask:

1. Who decides where the floor is? Is it a poverty line, a set of rights, “capabilities,” or whatever feels decent to the author?
2. Is this chapter left-wing politics dressed as philosophy? If not, what would a conservative or libertarian find persuasive in it?
3. The long spoons: in real life, who makes the spoons long? Isn’t the parable letting the rule-maker off the hook?
4. Program theory claims to make a collective effort’s assumptions testable. Name one program that was killed because its theory was tested and failed.
5. Campbell’s law says measures get corrupted once they’re targets. So isn’t program theory just a new set of metrics to game?
6. What does “being objective about collective subjectivity” mean in plain words?
7. Rawls says unjust institutions must be reformed or abolished. That’s a big step from a garden valve. Which institutions is the author thinking of?
8. Is the claim that civilization is about raising the floor (a description), or that it should be (a prescription)?
9. The printing and internet material shows capability isn’t goodness. Fine. What does that have to do with floors?
10. What am I, a reader with a job and no committee, supposed to do with this?

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

• “Capability is not moral progress with better stationery” is a nice line, but the chapter mostly lectures at that point. Lines like “Consent, fairness, care, and future choice need reasons of their own” sound like a mission statement.
• The printing and internet sections run long and turn into Wikipedia-grade history. The ITU statistic is good, but it arrives with no argument attached.
• “No lone engineer pressed a large red INTERNET button, disappointing though that may be to museums”: the jokes are charming but read as filler here.
• The garden committee again. By chapter 20 it’s become the book’s mascot, and it’s too small to carry claims about civilization.
• The Gellman quote (“high and holy and universal message”) is the preachiest line available. Starting there would confirm every suspicion.

Examples they’d bring:

• Scared Straight: the juvenile prison-visit program had an obvious theory of change (scare kids, and they won’t offend). Systematic reviews for the Campbell Collaboration (Petrosino and colleagues) found it was associated with more offending. This is program theory at its most useful: an intuitive program that failed the test.
• Microcredit: sold as a way out of poverty, then six randomized evaluations published together in American Economic Journal: Applied Economics (2015, Banerjee, Karlan and Zinman, eds.) found modest, not transformative, effects. It’s a floor-raising promise that was tested and trimmed.
• Wells Fargo’s fake-accounts scandal (2016): sales targets produced millions of unauthorized accounts. Campbell’s law with a corporate logo.
• The curb-cut effect (Angela Glover Blackwell, Stanford Social Innovation Review, 2017): curb cuts built for wheelchair users also helped parents with strollers, delivery workers and travellers. It’s a floor-raise that also lifted everyone else, and it connects directly to the wheelchair user at the garden valve.
• Finland’s basic income experiment (2017–2018): a real test of a floor-raising theory, with mixed results (better wellbeing, small employment effects). The cynic respects that it was tested.
• The office kitchen: everyone agrees it should be clean, nobody’s job is to clean it. A long-spoons problem in miniature.

What would win them over:

• Show program theory working and failing on a real case, step by step: here is the chain of assumptions, here is where it broke.
• Retell the long spoons with a twist that earns its place, for example by asking who made the spoons, or what happens when one diner refuses to feed anyone.
• State the chapter’s normative claim openly as a claim, and give the strongest objection a fair hearing.
• Trim printing and the internet to one short example, or move them.

Believer / spiritual reader

First reactions:

• Raising the floor is the most religious idea in the book, and a believer will notice that. Gleaning laws, tithes, zakat, the langar and “the least of these” are all floor-raising institutions, thousands of years old. They’d be glad to see them credited.
• The long spoons parable is theirs. The quote options handle it well (Gellman notes it belongs to many traditions and has no single author). But if the chapter uses it without saying it’s a religious story about heaven and hell, readers will feel it’s been borrowed and stripped of its setting.
• Program theory will feel cold at first. But a believer who has served on a church outreach committee or a mosque food bank board will recognize the problem it solves: good intentions with no way to tell if they’re working.
• The garden case, where the tank lasts all week but the gardener who uses a wheelchair can’t reach the valve, is exactly what the prophets complained about: a system that works for most people and quietly fails the vulnerable. It will move them.
• The printing section will make them wince if it treats the Gutenberg Bible only as an economic event. For believers, printing Scripture in the vernacular (and the Reformation that followed) was about who could read God’s word directly. That’s a floor being raised, and it’s the best example the chapter has.

Questions they’d ask:

1. Where does the obligation to raise the floor come from? My tradition says it comes from God, or from the image of God in every person. Where does the book say it comes from?
2. The draft says “Those commitments do not emerge automatically from a successful feedback loop.” Agreed. So what grounds them? If it’s just shared preference, why should the strong accept them?
3. Is charity raising the floor, or does it keep the floor where it is and make the giver feel better? Maimonides already ranked these. Will you?
4. The long spoons story says heaven and hell have the same food and the same spoons. The difference is whether people feed each other. Is the book saying heaven is a social arrangement we can build?
5. Can program theory measure dignity? A food pantry that makes people line up in public may hit its targets and still shame the people it serves.
6. Campbell’s law warns that measures get corrupted. Isn’t that why the Gospel warns against giving alms to be seen (Matthew 6:2)?
7. Religious institutions have also kept floors low, by excluding women, lower castes and outsiders. Will the chapter say that plainly, and credit the reformers inside those traditions who fought it?
8. Is there room in your account for the jubilee idea, that debts and land should periodically reset so that inequality can’t compound forever?
9. Who counts as “everyone”? Future generations? Animals? The dead, whose wishes built these institutions?
10. If a religious charity does better work than a government program, does the book’s framework let you say so?

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

• Offended: If the book presents floor-raising as a modern, secular discovery (Rawls, Sen, Ostrom) after its goal statement warns about being “cheated by religions.” A believer will point out that the religious institutions came first and still run much of the world’s charity work.
• Unconvinced: The draft admits that care, consent and fairness “need reasons of their own,” then doesn’t give any. A believer will say that’s exactly where faith supplies an answer and secular ethics has to argue for one. The book should at least lay out the options.
• Bored: The Internet section, with ITU statistics and TCP/IP, is long and unrelated to the parable or to program theory. The chapter’s two required elements (long spoons, program theory) are both missing from the draft.
• Lost: Program theory is named in the brief but never explained in the draft. A reader who has never seen a logic model won’t know what it is.

Examples they’d bring:

• Gleaning, Leviticus 19:9–10: “thou shalt not wholly reap the corners of thy field… thou shalt leave them for the poor and stranger.” A floor built into how the harvest itself is done, not handed out afterwards. The Book of Ruth shows it working in a story.
• Maimonides’ eight levels of tzedakah (Mishneh Torah, Laws of Gifts to the Poor 10:7–14): the highest level is helping someone become self-sufficient, for example with a loan, a partnership or a job. Lower levels include giving unwillingly and giving where the giver and receiver know each other. It’s a twelfth-century program theory: it asks how the giving actually produces the outcome.
• The Sikh langar: a free community kitchen at every gurdwara, where everyone sits on the floor and eats the same food regardless of caste or religion. The Golden Temple in Amritsar serves very large numbers daily. It raises the floor and flattens the roof at the same time, literally on the floor.
• Zakat: one of the Five Pillars of Islam, an obligatory annual alms (commonly 2.5% of qualifying wealth above a threshold), with recipient categories set out in Qur’an 9:60. It is a floor written into the definition of the faith.
• The jubilee, Leviticus 25: every fiftieth year, land returns to its original families and those in bonded servitude go free. Whether it was ever fully practiced is debated by historians, which is itself a useful program theory question: a design whose assumptions were never tested.
• Catholic social teaching on subsidiarity (Pius XI, Quadragesimo Anno, 1931): decisions should be made at the smallest competent level. It’s a direct match for the garden committee versus the distant designer, and it comes from a religious source.

What would win them over:

• Crediting religious floor-raising institutions as the oldest working examples, and applying program theory to them honestly: what works, what shames, what entrenches.
• Telling the long spoons parable in full, with its heaven-and-hell setting kept intact, and noting that the book is borrowing it.
• Admitting that the book can’t derive “everyone counts” from loops alone, and naming the religious, Kantian and other sources that people actually rely on.
• Using the vernacular Bible and the Reformation as the printing case: a real floor raised, with real violence following, both parts told.

Skeptical scientist

First reactions:

• The brief asks for the long spoons and program theory; the earlier draft has neither. It’s a well-researched section on printing and the Internet, about capability not choosing its own good. That’s a fine argument, but it’s not “raising the floor,” and it doesn’t show a reader how to test an institution’s assumptions.
• Program theory is the most scientifically respectable tool in Act III, and the chapter can make it concrete. Evaluation research has a well-documented record of good-sounding programs that did nothing or did harm. Those cases are the chapter’s best evidence.
• “Objective about collective subjectivity” is a nice phrase that needs cashing out. In practice it means: write the causal chain down, name the assumptions, measure the intermediate links, and compare against a counterfactual.
• The long spoons is a parable, not evidence. That’s fine, but pair it with the real biology and experimental economics of cooperation so it isn’t carrying the argument alone.

Questions they’d ask:

1. How do you measure “the floor”? Poverty headcount, the bottom decile’s income, child mortality, life expectancy at the bottom? Each gives different rankings.
2. Is the claim normative (we should raise the floor) or empirical (societies that raise it do better on X)? If empirical, better on what, and with what evidence?
3. What’s the counterfactual in the program-theory examples? Without one, you’re describing outcomes, not effects.
4. How does program theory handle effects that come through channels nobody drew on the diagram?
5. Can you give one case where writing out the theory of change exposed a false assumption before money was spent?
6. How does Campbell’s law interact with program theory? If you measure each link, don’t the links get gamed?
7. The Dittmar printing result: is it causal? What was the identification strategy? (He used distance from Mainz as an instrument, which is worth saying.)
8. Does “raising the floor” trade against “raising the roof,” and is there evidence either way?
9. The long spoons assumes everyone feeds everyone. What about free riders? What does the evidence say about how cooperation is stabilized?
10. Who gets to decide what the program’s intended outcome is? That’s the “collective subjectivity” part. How does program theory make it objective rather than just explicit?

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

• Lost immediately: the chapter’s title and brief promise spoons and program theory, and the draft opens with a garden valve and then spends most of its length on Gutenberg and TCP/IP.
• Unconvinced by the ITU statistic being used to make a floor argument without saying what internet access does to outcomes. Access is an input, not a floor.
• Mildly bored by the Internet-versus-Web distinction. It’s correct but irrelevant to the brief.
• Would bristle if the chapter cites The Spirit Level (Wilkinson and Pickett, 2009) as settled: its inequality-to-outcomes correlations are contested, and it should be presented that way if used.
• Would bristle at the Soviet nail-factory anecdote for Goodhart’s law if it appears; it’s likely apocryphal. Use a documented case.

Examples they’d bring:

• A program that did harm, found only because someone tested it. The Cambridge-Somerville Youth Study (1930s–40s) paired at-risk boys with mentors and services, with a randomized control group. Joan McCord’s 30-year follow-up (1978, American Psychologist) found the treated group had worse outcomes on several measures. The theory of change looked obvious; the assumptions failed.
• Scared Straight. Petrosino et al.’s Campbell Collaboration systematic review (2013 update) found that programs taking juveniles to visit prisons increased later offending compared with no intervention. A perfect case of a plausible causal chain (fear leads to deterrence) that the data reject.
• Targets and gaming. Bevan and Hood (2006, Public Administration, “What’s measured is what matters: targets and gaming in the English public health care system”) documented gaming of NHS targets such as ambulance response times. A real, documented Campbell’s-law case.
• Raising the floor in health. Removing lead from gasoline was followed by large falls in population blood lead levels in the US (NHANES data from the late 1970s onward). It’s a floor intervention: it helped most the children with the highest exposure, who were disproportionately poor. The causal chain (lead exposure to cognitive harm) is well supported.
• Cash as a test of assumptions. Haushofer and Shapiro (2016, Quarterly Journal of Economics) evaluated GiveDirectly’s unconditional cash transfers in Kenya with a randomized design: gains in consumption, assets and psychological wellbeing, with no measured increase in spending on alcohol or tobacco. It tests a common assumption about the poor directly.
• Cooperation needs enforcement. Fehr and Gächter (2002, Nature, “Altruistic punishment in humans”) showed cooperation in public-goods games collapses over rounds unless participants can punish free riders. For biology, Gerald Wilkinson (1984, Nature) documented vampire bats regurgitating blood to roost-mates who failed to feed. The long spoons works in real life only with some mechanism that handles defectors.

What would win them over:

• Build the chapter the brief describes: parable, then a real program’s theory of change drawn out step by step, then the evaluation that tested it.
• At least one example where the theory was wrong and the program did harm. That’s what makes program theory a tool for objectivity rather than a planning diagram.
• A clear, measurable definition of “floor,” with the admission that choosing the measure is itself a value choice.
• Keep Campbell’s law in view as the built-in limit, with a documented case.

Earlier draft

From “Changing a World We Share”

The garden committee discovers that its new watering rule works. The tank lasts all week. Then someone notices why: the plots nearest the gate are thriving while the far beds receive weak pressure, and the gardener who uses a wheelchair cannot reach the new valve.

The system has adapted. It is not therefore good.

Feedback matters. A change that cannot register its harms cannot correct them. Reversibility matters too: pencil marks on a planting plan are easier to revise than concrete poured across the only accessible path. Participation matters because people living with consequences often know what a distant designer misses. Yet an exploitative institution may learn efficiently how to preserve itself. Capability is not moral progress with better stationery.

The committee might install an accessible handle now, test pressure at every plot, compensate the person doing the measurements, and set a date to reconsider allocation. Its reasons come from care, consent, preventable harm, fairness, and future choice. Those commitments do not emerge automatically from a successful feedback loop. They tell us what the loop should notice and what it must not trade away for a larger harvest.

A garden is small enough that the valve and the affected people can meet in the same afternoon. Changes at civilizational scale are harder to see. They spread through many decisions, cross borders, acquire uses their makers did not imagine, and settle into ordinary life before anyone can render a final verdict. Two communication technologies make the difficulty visible.

Printing: the same page, many different worlds

In Mainz in 1454 or 1455, craftspeople produced the Gutenberg Bible with movable metal type. The surviving books resist the tidy story that printing instantly replaced an older world. The Library of Congress notes that copies were still finished by hand, with initials and headings supplied by different artisans. A new capacity arrived inside an existing culture of paper, ink, skill, money, religion, and trade.

What changed was not simply that one inventor had a thought. Workshops could make many closely similar copies; printers chose what might sell; financiers accepted risks; carriers moved sheets; authorities licensed or seized them; readers gathered around them. Over decades, printing became an infrastructure for preserving and circulating claims. The result was more reach for scholarship, administration, devotion, argument, fraud, satire, and abuse. The press did not inspect a pamphlet and prefer truth.

Even apparently clean measures tell only part of the story. Economist Jeremiah Dittmar found that European cities adopting printing in the fifteenth century later grew faster than comparable cities in his analysis (Quarterly Journal of Economics). That is evidence of economic consequence, not a verdict that printing caused every later intellectual or political transformation. Readers, institutions, conflicts, and local conditions carried the change.

Could the transformation be repaired? A mistaken sheet could be corrected in another edition. Laws, schools, libraries, and professional standards could respond to new problems. But printed accusations could travel beyond recall, and investments in presses and distribution altered who could speak at scale. Repair remained possible, but it was rarely a rewind. It meant building another practice in the world printing had helped make.

The Internet and the Web: layers of possibility

The Internet is not the Web. The Internet connects networks through shared protocols; email and many other services use it. The Web is one system built on that infrastructure, using addresses, links, browsers, servers, and HTTP.

That distinction matters because neither arrived as a single finished object. An Internet Society history written by people involved describes research groups, public agencies, universities, companies, independent networks, evolving protocols, and a carefully planned transition to TCP/IP in 1983. No lone engineer pressed a large red INTERNET button, disappointing though that may be to museums.

Tim Berners-Lee proposed the Web at CERN in 1989 to help people working across different systems share information. CERN released its Web software into the public domain in 1993. Berners-Lee later explained that a person could start a server without registering it with a central authority; this minimal constraint helped the Web scale (W3C). Open standards and permissionless publishing expanded what distant people could learn, make, coordinate, sell, and preserve.

Again, more capability did not choose its own good. The same reach can support mutual aid or harassment, public knowledge or organized deception, small publishers or systems that concentrate attention and surveillance. These outcomes were not all contained in the 1989 proposal, waiting to unfold. Governments, firms, standards bodies, maintainers, advertisers, communities, and billions of users kept changing the environment and one another.

Nor were the gifts or costs evenly shared. The International Telecommunication Union estimated that in 2024, 93 percent of people in high-income countries used the Internet, compared with 27 percent in low-income countries (ITU Facts and Figures). Being offline can close routes to work, education, public services, and relationships. Being online can expose a person to extraction, manipulation, fraud, and control. “Connect everyone” and “protect everyone” are not rival slogans so much as unfinished obligations.

Repair is possible, but no committee controls the whole. A harmful interface can change quickly; a dominant business model, installed cable, legal regime, or lost archive may be harder to alter. Open protocols can preserve room for alternatives, yet technical openness alone does not supply affordability, safety, democratic authority, or fair working conditions. Correction happens in layers, and sometimes one layer protects the problem in another.

The garden, printing, and the Internet differ enormously. Together they sharpen the same ethical questions: Who defines success? Who gains capacity? Who carries the labor and risk? Can affected people contest the arrangement? What evidence of harm can travel back? Which choices remain open to people who arrive later?

Learning helps detect consequences. Creating gives an arrangement a form the world can answer. Becoming can keep unplanned voices and values present. None guarantees goodness. Changing a shared world also requires care, consent, fairness, limits on preventable harm, and the humility to leave someone else a real chance to repair what we made.

The next chapter asks how far the book’s pattern can travel. These cases offer a warning before it does: resemblance across scales may guide attention, but it must not erase differences in mechanism, agency, or power.