Concept · updated 2026-09-09

Retirement

A 140-year-old invention, taught from the evidence. Ten chapters, every claim tagged.

Status: complete first pass

V verified — safe P primary read — safest S secondary — hedge it ? unsourced — never on air

How to Use This

This is the prose layer. The evidence layer is Retirement — every claim there carries a confidence tag and a source. Web version: https://concepts.rycolston.com/retirement/

Rule before you record anything: if a sentence here matters, open Retirement and check its tag. V and P are safe. S needs a hedge. ? does not get said out loud.

Ten chapters. Chapters 1–7 are the history and the health evidence. Chapter 8 is the psychology — filled this session, and the part you most wanted. Chapter 9 is retirement now. Chapter 10 is the honest list of what is still unverified.

Table of Contents

  1. The Thing Is 140 Years Old — the frame that makes everything else land
  2. Before — how people actually stopped working, and how the nursing home got invented by accident
  3. The Invention — Germany 1889, and the most-repeated false fact in the topic
  4. America Picks a Number — how 65 was really chosen, and it is dumber than you think
  5. The Rule — when 65 stopped being a custom and became a law, and what happened when it was repealed
  6. Selling the Dream — Sun City, the 401(k), and the quiet swap from pensions to accounts
  7. What It Actually Does To You — the evidence, and why it looks contradictory
  8. The Psychology — Freud's fake quote, Jung's afternoon, Erikson, the midlife crisis, the retired husband
  9. Retirement Now — the numbers, the myths from Okinawa, and the plan to abolish the three-stage life
  10. What We Still Don't Know

Chapter 1: The Thing Is 140 Years Old

Start here, because it reframes everything that follows.

Retirement — the idea that a healthy adult stops working at a set age and is paid to not work — is a social invention roughly 140 years old. It is younger than the telephone. It is younger than the light bulb. Nobody in the ancient world retired. Nobody in the medieval world retired. Your great-great-grandfather almost certainly did not retire.

This matters because of how people talk about retirement now. They talk about it the way you talk about weather: a fixed condition of the world you plan around. It is not. It is a policy, invented by specific people, for specific reasons, at a specific moment — and every one of those reasons is worth knowing, because most of them no longer apply.

The three things to hold in your head for the rest of this:

  1. Retirement was invented for a political reason, not a humane one.
  2. The number 65 was chosen almost arbitrarily, and then justified afterward.
  3. Before it could be sold as a reward, old age first had to be made into a problem.

Chapter 2: Before

People did not retire. They faded.

The honest picture of the pre-pension world is not "everyone worked until they dropped dead at the plough." That is too clean. What the data show is a slow drift downward.

Ransom and Sutch went back and re-estimated how many American men over 60 were working between 1870 and 1937. What they found, and named, was "on-the-job retirement": older industrial workers moving into less demanding and less well-paid jobs late in life. The transition out of work, they argued, was less sudden then than it is today.

Sit with that. We tend to assume modern retirement is the gentle option and the past was brutal. On this reading it is the reverse. The past had a ramp. We built a cliff.

Teach the fight, not just the finding. Jon Moen published a rebuttal arguing that what Ransom and Sutch scored as voluntary gradual retirement was actually unemployment — old men who could not get their old jobs back, taking whatever was left. Nobody has cleanly settled it. And the reason nobody has settled it is itself the lesson: it is extremely hard to see "retirement" in data collected before retirement existed. You cannot count a thing that has no name yet.

The nursing home was created by subtraction

This is the best story in the early history, and almost nobody knows it.

American poorhouses were not old-age homes. They were general-purpose dumping grounds for everyone destitute — children, the mentally ill, the disabled, the drunk, the unemployed, and the old, all in one building.

Then, one group at a time, reformers pulled people out:

  • 1875 — New York's Children's Act orders children aged 2 to 16 removed
  • 1890 — New York's State Care Act moves the insane poor to state hospitals
  • After 1880 — the able-bodied are pushed out by police-lodging bans and municipal wayfarers' lodges

Nobody set out to build an institution for the elderly. They set out to rescue everyone else. Michael Katz's sentence is the one to read aloud:

"By stripping away one group after another, policy makers came closer and closer to their goal: the transformation of poorhouses into old age homes. When everyone else had been siphoned off elsewhere, the only group remaining was the elderly."

The modern old-age home is a residue. It is what was left in the building after everyone worth saving had been saved out of it. That is the institutional inheritance retirement was born into.

Old age had to be made into a problem first

Before you can sell people a reward for stopping, you have to convince everyone that older workers should stop.

On 22 February 1905, William Osler — the most celebrated physician in the English-speaking world — gave his farewell address at Johns Hopkins. He called it "The Fixed Period," after an Anthony Trollope novel in which men retire at 67 and are then "peacefully extinguished by chloroform." The press ran with it: "Osler recommends chloroform at sixty." The verb "to Oslerize" entered the language.

Osler said it as a joke. But here is the part that makes it worth teaching — his own written reply a year later, in the preface to the next edition of his book:

"I jokingly suggested for the relief of a senile professoriate an extension of Anthony Trollope's plan... to every man over sixty whose spirit I may have thus unwittingly bruised, I tender my heartfelt regrets. Let me add, however, that the discussion which followed my remarks has not changed, but has rather strengthened my belief that the real work of life is done before the fortieth year and that after the sixtieth year it would be best for the world and best for themselves if men rested from their labours."

He apologised for the offence and kept the position. The most respected doctor alive believed, on the record, that useful life ended at 40 and men should be done at 60. That is thirty years before Social Security.

Retirement did not arrive into a neutral world and offer old people a gift. It arrived into a world that had already decided old people were finished.


Chapter 3: The Invention

Germany, 1889

The world's first national old-age insurance program was Germany's, in 1889, under Otto von Bismarck.

Bismarck's motive was substantially political. He had passed the Anti-Socialist Laws in 1878. Social insurance was the other half of the pincer: crush the socialist movement with one hand, take away its best argument with the other. The Social Security Administration's own history puts it plainly — he acted "to promote the well-being of workers in order to keep the German economy operating at maximum efficiency, and to stave-off calls for more radical socialist alternatives."

The German socialists of the day read it exactly that way, and opposed it.

Hedge this one on air. The anti-socialist motive is documented. Whether Bismarck also genuinely cared about workers is live historical argument — some historians read it as pure authoritarian calculation. Say "the political motive is documented; the sincerity is debated." Do not pick a side you cannot defend.

The most-repeated false fact in the entire topic

Here is the payload. This is the thing your audience thinks it knows, and it is wrong.

Everyone says: Bismarck set retirement at 65, and he picked it because that was his own age / because almost nobody lived that long.

The truth:

  • The 1889 law set the qualifying age at 70
  • Bismarck was 74 when it passed
  • It was not lowered to 65 until 1916
  • By then Bismarck had been dead for 18 years

Not a small correction. Every element of the popular story is wrong — the number, the reasoning, and the man.

The SSA says it flatly. A peer-reviewed correction in The Gerontologist is literally titled "Otto Von Bismarck Is Not the Origin of Old Age at 65." It quotes the 1889 act itself: it "determined the time of retirement at the age of 70 years."

Do not overcorrect. Germany did get to 65 first — in 1916, nineteen years before the US. The German precedent is real. It just is not Bismarck's.


Chapter 4: America Picks a Number

This is the chapter that will surprise people most, and it is now sourced straight from the institution that made the decision.

The assumption is that 65 came from somewhere — an actuary's table, a medical finding, a European model. It did not. Here is the Social Security Administration describing its own choice:

"This decision was not based on any philosophical principle or European precedent. It was, in fact, primarily pragmatic... One was a general observation about prevailing retirement ages in the few private pension systems in existence at the time and, more importantly, the 30 state old-age pension systems then in operation. Roughly half of the state pension systems used age 65 as the retirement age and half used age 70. The new federal Railroad Retirement System passed by Congress earlier in 1934, also used age 65... The CES planners made a rough judgment that age 65 was probably more reasonable than age 70. This judgment was then confirmed by the actuarial studies."

And then the closing line, which is the best sentence in the whole topic:

"With all due respect to Chancellor Bismarck, he had nothing to do with it."

Read the order of operations again, because that is the teaching point:

  1. Look around at what everyone else is already doing
  2. Split a rough judgment between 65 and 70
  3. Then run the math to check it works

They did not calculate their way to 65. They took a survey of existing habit, guessed, and back-filled the arithmetic.

The man in the room said it out loud

Robert J. Myers was on the committee's actuarial staff in 1934 and later ran Social Security's actuaries for twenty-three years. In his 1992 memoir, Within the System, he wrote:

"Now here's the arbitrary part. Age 65 was picked because 60 was too young and age 70 was too old. So we split the difference."

He called the decision "admittedly arbitrary and empirical." The number that has governed the shape of a billion working lives was, in its origin, a coin flip between two round numbers that the actuaries then blessed.

Say it as "Myers wrote in his 1992 memoir." The quote is traced to the memoir through a 2026 History.com article; the memoir itself has not been read. It is not on the SSA's own page and not in his oral history.

The second myth, while you are here

Everyone says: life expectancy in the 1930s was about 61, so Social Security was designed so that hardly anyone would live to collect.

The SSA's own answer: life expectancy at birth in 1930 was indeed low — 58 for men, 62 for women — but that number is dragged down by infant mortality, and a baby who dies never paid a payroll tax in the first place. The right measure is life expectancy after reaching adulthood.

  • Of men who reached 21, almost 54% could expect to reach 65
  • Men who reached 65 could expect to collect for nearly 13 more years
  • There were already 7.8 million Americans aged 65 or over in 1935
  • The CES's own projections expected 8.3 million by 1940, when monthly benefits began

The SSA's conclusion: "Social Security was not designed in such a way that few people would collect the benefits."

Two myths, two clean kills, both from the primary source. That is a segment on its own.


Chapter 5: The Rule

We talk about 65 as a norm. For a huge share of American men it was not a norm. It was a rule, and it was enforced.

Roughly 40 to 50% of US male employees in the 1960s and 70s worked under rules that mandated retirement at 65. Not encouraged. Mandated.

Then Congress took it apart:

Year What happened
1967 Age Discrimination in Employment Act protects workers aged 40–65
1978 Extended to 70 — killing forced retirement at the most common age
1986 Upper limit removed, with narrow exemptions

Get this right on air: say "with narrow exemptions," not "abolished entirely." High-policymaking executives, some public safety officers, and tenured faculty (until the end of 1993) kept theirs.

What happened when the rule died

Two results, and the second is the interesting one.

First: the labour force of workers 65 and older rose by an estimated 10 to 20%, and the sharp spike of retirements happening exactly at 65 flattened out significantly. So a meaningful chunk of "people choose to retire at 65" was never a choice. It was compliance.

Second, and this is the one to build a segment on: removing mandatory retirement had no effect on job tenure and no effect on the wage profiles of older workers.

That null result matters. The standard defence of forced retirement was an implicit-contract theory: firms pay younger workers less than they are worth and older workers more, and the whole arrangement only balances if there is a guaranteed exit at the end. Take away the exit, the theory says, and firms must stop over-paying older workers.

They did not. Nothing moved. The economic justification for forcing people out did not survive contact with the data.


Chapter 6: Selling the Dream

Making retirement legal and funded is not the same as making people want it. For a long stretch, stopping work read as failure — you were used up, you were dependent, you had been Oslerized. Two things changed that: a product, and a tax clause.

The product: Sun City

On 1 January 1960, Del Webb opened Sun City, Arizona — five model homes, a shopping centre, a recreation centre, and a golf course, on the site of a ghost town called Marinette. The pitch was the invention: not "you are finished," but "you have arrived." Not a poorhouse — a golf course. Retirement stopped being an ending and became a destination you buy.

The company says 100,000 people came that first weekend, ten times what it planned for, and that it sold 1,300 homes by the end of the year. Time put Webb on its cover in August 1962.

Say whose number it is. The 100,000 is Del Webb Corporation's own figure, repeated by the company's archive and by every history since. The university public-history project that covers it hedges to "tens of thousands... some estimates more than 100,000." No newspaper count from that weekend has been found. "The developer says 100,000 showed up" is true and still lands. "100,000 showed up" is not yet earned.

The tax clause: 401(k)

The popular version is that a benefits consultant named Ted Benna invented the 401(k) in 1980. The record says something less heroic and more interesting.

Some employers had long let workers take part of their pay as deferred compensation. In 1974, Congress froze new plans of that type pending a study. After the study, it wrote the arrangement into the tax code: Public Law 95-600, the Revenue Act of 1978, section 135, enacted 6 November 1978, added subsection (k) to section 401, effective for plan years beginning in 1980.

The first company on record moving to use it was Hughes Aircraft. In December 1978 — weeks after enactment — its outside lawyer, Ethan Lipsig, sent the company a letter recommending it begin converting its savings plan. Johnson & Johnson started in 1979. The IRS issued regulations in November 1981, and the first wave of plans — J&J, FMC, PepsiCo, JC Penney, Honeywell, Hughes — went live in January 1982. Within two years, half of large firms had one or were considering one.

The trade body that keeps the official history of the 401(k) does not mention Benna at all. He built one of the early plans, at his own firm, to cut taxes on a bonus plan for bank executives. "Inventor" is a headline writer's word.

On air: "Congress invented the 401(k) on November 6, 1978. Hughes Aircraft's lawyer was the first to move on it, that December. Ted Benna built one of the early ones. He did not invent it."

The quiet swap

What the 401(k) did over the next forty years was replace one kind of retirement with another, and almost nobody voted on it.

Active participants in private pension plans, from Department of Labor filings:

Year Traditional pension (defined benefit) Account plan (defined contribution)
1975 27 million 11 million
1984 30 million (the peak) 31 million
2022 11 million 93 million

By March 2025, 70 percent of private-industry workers had access to an account plan and 14 percent to a traditional pension.

The difference is not the money. It is who carries the risk. A pension is a promise the employer has to keep. An account is a balance the worker has to grow, and not outlive. The retirement your grandfather was sold and the retirement you are being sold share a name and nothing else.


Chapter 7: What It Actually Does To You

This is where the audience leans in, and where most content on this topic lies to them. The evidence is genuinely mixed — and the reason it is mixed is the most useful thing you can teach.

"Work longer, live longer" is mostly a statistical illusion

A meta-analysis of 25 longitudinal studies looked at retirement and death.

Run the analysis carelessly, and retiring on time looks dangerous: a hazard ratio of 1.56. That is the number that generates headlines.

Now control for how healthy people were before they retired:

Hazard ratio Verdict
On-time retirement, poorly adjusted 1.56 looks deadly
On-time retirement, adjusted for prior health 1.12 (CI 0.98–1.28) crosses 1 — gone
Early retirement 1.05 (CI 0.87–1.28) nothing

The effect nearly vanishes. What is left is the healthy worker effect: healthy people keep working, sick people stop. Work was never extending their lives. Their health was extending their work.

The line: "Does retiring kill you? No. The studies that said it did mostly forgot to check who was already sick."

"Use it or lose it" is much weaker than you have heard

A systematic review of retirement and cognitive decline found only seven longitudinal studies good enough to include. Its findings:

  • Weak evidence that retirement speeds decline in crystallised ability — and only for people leaving jobs high in complexity with people
  • For fluid abilities — processing speed, working memory — the evidence is conflicting

A 2025 review with twenty studies still reports negative, null, and positive results. Eight more years of work, still unsettled.

So: if you leave a job whose whole substance was dealing with people, there is a weak signal worth respecting. Beyond that, the confident version of "retirement rots your brain" is not supported.

The real answer: it depends, and we now know on what

An overview of 15 systematic reviews concluded that retirement's health impact remains inconclusive — positive, negative and null effects all reported.

But it is not random. Two things separate good outcomes from bad, and both are now sourced.

The first is socioeconomic status. Higher-SES retirees see mental health improve and activity rise. Lower-SES retirees see physical and mental health decline, sedentary time rise, and cardiovascular outcomes worsen.

The second is whether you chose it. A meta-analysis of 25 longitudinal studies measured depression after retirement by type of exit:

Type of retirement Effect on depressive symptoms (d)
All retirement, pooled 0.04 — a rounding error
Voluntary 0.09 — and the interval crosses zero
At a mandatory age 0.01 — nothing
Involuntary 0.18 — four times the average

A second meta-analysis, eight longitudinal studies, puts the risk of depression after forced retirement at 1.31 times that of people still working — and finds the arrow runs the other way too: people who are already depressed are 1.7 times as likely to be pushed out. A Dutch panel of 1,388 workers found that people who chose to leave were more satisfied than people who kept working, while people forced out by health or their employer were the least satisfied of anyone.

Which gives you the frame for the whole episode:

"Is retirement good for you?" is the wrong question. The right one is: good for whom, leaving what kind of job, and did they choose it or was it done to them?

That question also explains the mess. Every study mixing a forced-out 62-year-old labourer with a voluntarily-exiting 68-year-old professor is averaging two opposite experiences into a meaningless middle. The contradictions in the literature are not noise. They are the finding.


Chapter 8: The Psychology

This was the thinnest chapter and it is now the richest. Two warnings before you use it. First, the original books — Jung 1933, Erikson 1950, Cumming and Henry 1961, Atchley 1976 — are locked behind copyright and could not be read; what follows quotes them through peer-reviewed papers that quote them, and says so. Second, most of what gets called "the psychology of retirement" was never about retirement at all. That is the chapter.

Freud never said "love and work"

Ask any therapist what a healthy adult should be able to do and you will get Freud's answer: lieben und arbeiten, to love and to work. Some add the flourish: "Love and work are the cornerstones of our humanness."

Freud never wrote either sentence.

The line enters the record in 1950, in Erik Erikson's Childhood and Society. Erikson says Freud, asked what a normal person should be able to do well, "was reported to" have answered lieben und arbeiten. Reported. Erikson is passing on hearsay and says so. In 2001 Alan Elms went looking for it in Freud's collected writings for a paper called "Apocryphal Freud." It is not there. Two independent psychology papers since have repeated the finding: quoted by Erikson in 1950, "cannot be found in Freud's writings."

What Freud did write, in 1930, in Civilization and Its Discontents — and this one was read in the original German — is that human society "was doubly founded: on the compulsion to work, which external necessity created, and on the power of love." Same two nouns. Different sentence, doing a different job: it is about why people ever lived together, not about what a healthy mind can do.

On air: "Freud never said 'love and work.' Erikson said in 1950 that Freud was reported to have said it. Nobody has ever found it in Freud. The closest real sentence is from 1930, and it's about the origin of society, not mental health." That is a myth correction with a primary source under it.

Jung's afternoon starts at forty, not sixty-five

Every retirement book quotes Jung on "the afternoon of life." Two of his sentences survive in open scholarship, both from the 1933 English edition of his essay "The Stages of Life":

"The afternoon of human life must also have a significance of its own and cannot merely be a pitiful appendage to life's morning."

"The afternoon of life is just as full of meaning as the morning; only, its meaning and purpose are different."

Beautiful, and worth reading aloud. But note what Jung is describing. His afternoon begins around thirty-five to forty. It is an essay about midlife — about the moment when the goals that built a life stop working as a guide. Retirement writers borrowed it because the imagery fits. Jung was not writing about retirement, and the longer passage everyone quotes ("we cannot live the afternoon of life according to the programme of life's morning") could not be located in a checkable source this session — do not read it off this file.

Erikson, with one honest hole

Erikson's last two stages are the ones people reach for: generativity versus stagnation — making something that outlasts you, usually mapped onto the middle years — and integrity versus despair — looking back on a life and finding it was yours, usually mapped onto sixty-five and beyond. He returned to old age directly in The Life Cycle Completed (1982) and Vital Involvement in Old Age (1986), where integrity and despair are "two diametrically opposed tensions" the old must hold in balance, and wisdom is "the strength gained in the struggle" between them. His wife and collaborator Joan added a ninth stage for the eighties and nineties in 1997.

Two things you cannot say. The ages — 45 to 64, 65 and up — are textbook conventions; whether Erikson fixed them himself in 1950 could not be checked. And whether Erikson ever treated retirement as an event, rather than old age as a stage, is unverified. "Erikson's stage that lands at retirement age" is safe. "Erikson on retirement" is not.

The midlife crisis was invented in 1965, about 35-year-olds

Elliott Jaques, a Canadian psychoanalyst, coined "mid-life crisis" in a 1965 paper in the International Journal of Psycho-Analysis. Three things about it that nobody repeats.

He first read the paper to the British Psychoanalytical Society in 1957; the journal sat on it for eight years. It rested on a study of about 300 creative artists — when they produced, when they changed, when they died — plus his own patients. And the crisis he described struck around thirty-five: a depressive reckoning with mortality, "energetically masked by a manic determination to thwart advancing years" — the hypochondria, the affairs, the sudden religion. Jaques had started thinking about it in 1952, at thirty-five himself, after finishing his own analysis and reading Dante.

So the midlife crisis is a 1965 coinage, about the mid-thirties, from a sample of dead composers. It is not about fifty, it is not about retirement, and the historian who traced it argues it is a product of post-war demography and the American dream rather than anything in the body.

The first theory of ageing said withdrawal was natural. It was wrong.

Disengagement theory — Elaine Cumming and William Henry, 1961, Growing Old — was the first theory of ageing social scientists produced, built on the Kansas City Study of Adult Life (211 adults aged 50 to 90). It claimed the mutual withdrawal of the ageing person and society is natural and appropriate.

It fell apart because of how absolutely it was stated: disengagement was proposed as innate, universal, unidirectional, and irreversible once begun. Cumming was walking it back within two years, and again in 1968. Teach it as the cautionary tale: the first serious theory of ageing said old people are supposed to withdraw, and it was wrong.

Atchley: the theory is alive, the stages are a zombie

Robert Atchley gave the field two things. Continuity theory (1989) holds that ageing adults adapt by "preserving and maintaining existing internal and external structures," using "strategies tied to their past experiences of themselves and their social world." Continuity, he wrote, "is a grand adaptive strategy." That still stands.

The other thing is the stages of retirement — honeymoon, disenchantment, reorientation, stability — which every retirement-planning article recites as if it were a law. It is not. When researchers finally followed real retirees over time, they did not find a sequence. They found three different roads running at once: people whose wellbeing stays flat through retirement, people whose wellbeing rises the moment they leave work, and a smaller group who dip and then recover. That result came out of the US Health and Retirement Study in 2007 and replicated in 1,456 German retirees the same year: "retirement is not a uniform transition." A 2023 Australian study added a twist — mental health improves in the two years before retirement, in anticipation, and drifts back to baseline about two years after. "Retirement improves mental health before it happens, but not after."

So: continuity theory, yes. The honeymoon-then-crash story, only as one path among several. Never "the stages of retirement."

The retired husband, real and small and not Japanese

"Retired husband syndrome" — shujin zaitaku sutoresu shōkōgun, husband-at-home stress syndrome — is usually told as a Japanese discovery: Nobuo Kurokawa, 1991, wives falling ill when their salaryman husbands come home for good.

The name is American. It appears in the Western Journal of Medicine in October 1984, seven years before Kurokawa, in a paper by C. C. Johnson describing his own patients: "I am going nuts." "He is under my feet all the time." Kurokawa's contribution was the Japanese name and the Japanese framing.

Is it real? Yes, and here is the size. Two economists used a 2006 Japanese employment reform as a natural experiment and found that a husband's earlier retirement significantly increases his wife's reported symptoms — in the working-paper version, each extra year of his retirement raised her odds of symptoms by roughly six to fourteen percentage points. A separate 12-year Japanese panel of 3,794 wives measured the actual distress: on a 24-point scale where the average wife scored about 3.4, the score rose by about 0.2 points in each of the first three years after the husband retired, then went back to baseline. Wives with their own social lives, or who were already close to their husbands, showed no rise at all. That study's own conclusion: the syndrome has "limited relevance."

The BBC's line that sixty percent of older Japanese wives suffer from it has no study behind it. Do not repeat it.

On air: "It's real. A husband's retirement measurably raises his wife's stress. It's also small, it's gone in about three years, and it was named in California in 1984, not Tokyo in 1991."

What the chapter adds up to

Almost none of the famous psychology of retirement is about retirement. Freud's line is Erikson's hearsay. Jung's afternoon starts at forty. Jaques's crisis hits at thirty-five. Erikson wrote about old age, not the event. The one body of work that is actually about the transition — the longitudinal panels — says there is no single story, and that the variable doing the most work is the one in Chapter 7: did you choose it.


Chapter 9: Retirement Now

Two myths from Okinawa

"Okinawans have no word for retirement." The line is Dan Buettner's, from his 2009 TED talk, word for word: "In the Okinawan language there is not even a word for retirement." Japanese — the language every Okinawan speaks — has at least three: teinen, the fixed age at which a job ends; taishoku, leaving employment; intai, stepping down from a post. The claim is technically about the old Ryukyuan dialect, which nobody has checked and which would not matter if true, because a fishing village in 1900 had no pension to retire on. The line has since been repeated, uncited, in a 2025 neuroscience journal. That is how a TED talk becomes a footnote.

The Blue Zones themselves. In 2024 Saul Newman won the Ig Nobel Prize in Demography "for detective work to discover that many of the people famous for having the longest lives lived in places that had lousy birth-and-death recordkeeping." His preprint's findings, as stated: in the United States, supercentenarian status is predicted by the absence of birth registration; introducing birth certificates cut supercentenarian records by 69 to 82 percent; only 18 percent of "validated" supercentenarians worldwide have a birth certificate, and none in the US; their birthdates cluster on days divisible by five; and Sardinia, Okinawa, and Ikaria are, relative to their own countries, poorer, less literate, higher-crime, and shorter-lived. Hedge: it is a preprint. Say "a preprint that won an Ig Nobel," and let the audience enjoy it.

The ikigai diagram is from Guernsey

The four-circle ikigai diagram — what you love, what you're good at, what the world needs, what you can be paid for — is not Japanese. It is a blog post: Marc Winn, "What is your Ikigai?", 14 May 2014. And it is not original to him. A Spanish designer, Andrés Zuzunaga, drew the same four circles in 2012 and labelled the centre propósito — purpose. He said so in a comment on Winn's own post in 2017: "This is copy of a spanish graphic made Andrés Zuzunaga two years before." One word was swapped, and a Spanish purpose diagram became Japanese wisdom.

What the actual Japanese research asked is worth the contrast. The Ohsaki study followed 43,391 adults for seven years and asked one question: "Do you have ikigai in your life?" People who said no were 1.5 times as likely to die over the follow-up, mostly from heart disease and external causes, not cancer. One yes-or-no question. No circles.

Where Americans actually stand

From the Federal Reserve's 2022 Survey of Consumer Finances:

  • 54 percent of families have any retirement account. 46 percent have none.
  • Among families that do, the median balance is $86,900.
  • Among working-age families in the bottom half of incomes who have an account, the average balance is $54,700 — and it fell from 2019. In the top ten percent it is $913,300.

From the 2026 Social Security Trustees Report:

  • The retirement trust fund runs dry in the fourth quarter of 2032 — a quarter earlier than last year's estimate. After that, incoming taxes cover 78 percent of scheduled benefits.
  • If Congress merged the retirement and disability funds, the date moves to 2034 and the figure to 83 percent. That is the number you usually hear; it assumes a law change.

Put those together. Half the country has no account; the median account would not fund two years of a modest retirement; and the backstop is scheduled to shrink by a fifth in six years. The system built in 1935 for a country of 7.8 million people over 65 is carrying a very different country.

France tried to move the number and it took Article 49.3

In early 2023 the French government proposed raising the legal retirement age from 62 to 64. It could not find a majority. On 16 March 2023, Prime Minister Élisabeth Borne stood in the National Assembly and, over prolonged booing and deputies pounding their desks, invoked Article 49, paragraph 3 of the Constitution — which lets the government pass a budget-type bill without a vote unless the Assembly topples it. The parliamentary record notes the signs held up in the chamber: "64 ans c'est non." Four days later a censure motion failed by nine votes. On 14 April the Constitutional Council upheld the age change — rising three months per birth-year until people born in 1968 retire at 64 — and struck several riders. The law was promulgated that night.

The lesson is not about France. It is that the number is now so load-bearing that moving it by two years required a democracy to bypass its own legislature.

The proposal to abolish the three-stage life

Stanford's Center on Longevity — founded by the psychologist Laura Carstensen — has spent since 2018 on a project called The New Map of Life. Its premise: "The 100-year life is here. We're not ready." Its institutions "evolved when lives were only half as long."

The retirement proposal, in the report's own words: "Rather than fixed chapters that span 70 years and focus consecutively on education, work, and retirement, we envision several shorter, flexible intervals dedicated to learning, working, caregiving, and leisure that are woven as needed into the course of 100-year lives, with working intervals likely to include more than one primary career." Expect to work sixty years or more — but not forty-hour weeks the whole way. And stop marking the second half of life with "menopause, retirement, and death"; mark it instead with "volunteering and mentoring, returning to work, launching an encore career."

Notice what this is. It is not "retire later." It is "stop having a retirement." Which is, if you have followed the first eight chapters, a proposal to go back to the ramp — the on-the-job easing-out of 1880 — with better healthcare. The 140-year-old invention may turn out to have been a detour.


Chapter 10: What We Still Don't Know

Say this part out loud too. It is more credible than pretending the file is complete.

  • Erikson and retirement. Nobody this session read Vital Involvement in Old Age. Whether Erikson treated retirement as an event, and whether he fixed the stage ages himself in 1950, are open. A library copy would close both.
  • The Freud item is one step from airtight. The finding rests on two papers relaying Alan Elms's 2001 study. Reading Elms directly would move it from "secondary" to "primary."
  • Retirement savings by age. The medians by age band live in the SCF's supplementary tables and were not pulled. Do not quote by-age numbers.
  • Sun City's opening weekend. The 100,000 is the developer's figure. A January 1960 Arizona Republic count would settle it.
  • The pre-pension ramp. Whether Ransom and Sutch's gradual "on-the-job retirement" was real easing-out or disguised unemployment (Moen) is still unsettled. It decides whether Chapter 2 is taught as people eased out or people were pushed out and we renamed it.
  • Not covered at all: the post-war pension boom (Inland Steel 1949, the 1950 Treaty of Detroit), the origin of "golden years," unretirement rates, FIRE and the 4 percent rule, activity theory, and terror management. None of it is on the file. None of it goes on air.
Evidence file

Every claim, tagged

The prose above is only as good as the tag on each sentence here. Read the tag before you say the sentence out loud.

Narrative companion: Retirement - Curriculum — the read-through written to be taught. Web: https://concepts.rycolston.com/retirement/

Retirement — Evidence File

Every claim below carries a confidence tag. Read the tag before you say the sentence out loud.

Tag Means Safe to teach?
V Verified. Survived a 3-vote adversarial check (2 of 3 refutations would have killed it) Yes
P Primary. I pulled the original document and read the sentence myself Yes — strongest tier
S Secondary. From a reference work or a scholarly paper quoting the original. Not checked against the original Only with hedge
? Unsourced. Believed true, never checked No. Do not say on air.

Coverage (session 2, 2026-09-09): threads 4 (psychology) and 6 (today) are now filled. The 20th-century psychology canon — Jung 1933, Erikson 1950, Cumming & Henry 1961, Atchley 1976 — stays copyright-locked (every archive.org edition returns 401/403), so those items rest on open-access scholarship that quotes the originals, tagged S honestly. Where a primary was reachable (Freud's 1930 German text, PubMed abstracts, US Code, DOL/BLS/Fed tables, the French parliamentary record, the 2026 Trustees summary) it is tagged P.


1. Origin — who invented it, and why

The founding myth is false V

Germany's 1889 old-age insurance law set the qualifying age at 70, not 65. Bismarck was 74 when it passed. The age was not lowered to 65 until 1916 — 27 years later, and 18 years after Bismarck died (1898).

Every version of "Bismarck picked 65 because that was his age" or "because it conveniently excluded his rivals" is wrong.

  • Source: SSA, "Age 65 Retirement" — states it verbatim
  • Corroborated: The Gerontologist 54(1):5 (2014), titled "Otto Von Bismarck Is Not the Origin of Old Age at 65" (PubMed 24085527), quoting the 1889 act itself: it "determined the time of retirement at the age of 70 years"
  • Vote: 3-0 unanimous, three separate merged claims
  • Teaching nuance: 65 in Germany (1916) still beat 65 in the US (1935) by 19 years. The German precedent is real — it is just the 1916 amendment, not Bismarck's law.

Legislative dates V (2-1)

Reichstag voted 24 May 1889. Signed (Ausfertigung) 22 June 1889. Proclaimed (Bekanntmachung) 26 June 1889. Two different legal moments under German convention — not a contradiction. If a script needs one date, use 22 June 1889.

Bismarck's motive was dual, not purely cynical V (2-1, medium confidence)

SSA: "Bismarck was motivated to introduce social insurance in Germany both in order to promote the well-being of workers in order to keep the German economy operating at maximum efficiency, and to stave-off calls for more radical socialist alternatives."

Context: his 1878 Anti-Socialist Laws. Early German socialists themselves read it as an anti-socialist maneuver and opposed it.

Hedge on air: the anti-socialist motive is documented. Whether the welfare concern was sincere is live historical debate — a minority of historians read it as pure authoritarian calculation.

How the US actually chose 65 P — pulled from SSA directly

This is the answer to the biggest question in the whole topic, in the institution's own words:

"By the time America moved to social insurance in 1935 the German system was using age 65 as its retirement age. But this was not the major influence on the Committee on Economic Security (CES) when it proposed age 65... This decision was not based on any philosophical principle or European precedent. It was, in fact, primarily pragmatic, and stemmed from two sources. One was a general observation about prevailing retirement ages in the few private pension systems in existence at the time and, more importantly, the 30 state old-age pension systems then in operation. Roughly half of the state pension systems used age 65 as the retirement age and half used age 70. The new federal Railroad Retirement System passed by Congress earlier in 1934, also used age 65... The CES planners made a rough judgment that age 65 was probably more reasonable than age 70. This judgment was then confirmed by the actuarial studies.... With all due respect to Chancellor Bismarck, he had nothing to do with it."

  • Source: SSA, "Age 65 Retirement", fetched 2026-09-09 via Wayback (ssa.gov blocks direct fetch at its CDN). Re-fetched raw 2026-09-09 (session 2): the whole page is ~3,000 characters and this passage is its ending.
  • The teaching line: America did not reason its way to 65. It took a vote of existing practice and then checked the math afterward.

Robert J. Myers' "split the difference" S — traced (session 2)

Myers, a CES actuarial staffer and later SSA Chief Actuary, wrote in his 1992 memoir Within the System (title confirmed via Open Library; year per History.com):

"Now here's the arbitrary part. Age 65 was picked because 60 was too young and age 70 was too old. So we split the difference."

He also called the decision "admittedly arbitrary and empirical."

  • Source: History.com, "How 65 Became America's Default Retirement Age" (Jordan Friedman, published 21 Apr 2026), quoting the memoir. Fetched 2026-09-09.
  • Not on the SSA "Age 65" page (checked — the page never names Myers) and not in his 1996 SSA oral history (checked — no "too young / too old" passage).
  • S because the memoir itself was not read. Safe to say with "Myers wrote in his 1992 memoir." Best single line in the topic, and it now has a source.

2. Before retirement — what the end of work looked like

People eased out, they did not stop V

Ransom & Sutch re-estimated labour force participation for American men 60+ from 1870–1937 and named the mechanism "on-the-job retirement" — older industrial workers moving into less remunerative and less demanding occupations late in working life, "which may have made the transition from employment to full retirement less sudden than today."

  • Source: Ransom & Sutch, Journal of Economic History 46(1), 1986, pp. 1–30
  • Vote: 3-0
  • Mandatory caveat — say this: Jon R. Moen published a JEH rebuttal arguing what they measured as voluntary retirement was actually unemployment. What is verified is "this is what Ransom and Sutch argued and showed," not "this is consensus." The fight itself is good material: it is genuinely hard to see "retirement" in data collected before pensions existed.

The nursing home was created by subtraction V

American poorhouses were general-purpose institutions for all the destitute. Policy makers removed one group at a time until only the old were left:

Removed When How
Children (2–16) 1875 NY Children's Act
The mentally ill 1890 NY State Care Act — moved the insane poor to state hospitals
The able-bodied after 1880 police-lodging bans, 1896 NYC Raines Law, municipal wayfarers' lodges

Katz, p. 132: "By stripping away one group after another, policy makers came closer and closer to their goal: the transformation of poorhouses into old age homes. When everyone else had been siphoned off elsewhere, the only group remaining was the elderly."

  • Source: Michael B. Katz, "Poorhouses and the Origins of the Public Old Age Home", Milbank Memorial Fund Quarterly 62(1), 1984, pp. 110–140. Verified against the full PDF, not an abstract. Grounded in Erie County almshouse registers and Hoyt's 1874–75 survey of 12,614 inmates.
  • Vote: 3-0
  • REFUTED 0-3, do not say: that this was "substantially complete by the end of WWI." Katz's own wording is the looser "early decades of the twentieth century."

Old age had to be devalued first — Osler P — pulled from the original book

William Osler's farewell address at Johns Hopkins, "The Fixed Period" (1905), chapter XIX of Aequanimitas. He invoked Anthony Trollope's 1882 novel The Fixed Period, in which men retire at 67 and are then "peacefully extinguished by chloroform." The press ran it as "Osler recommends chloroform at sixty" and coined the verb "to Oslerize."

Osler's own reply, in his preface to the second edition — this is the passage worth reading aloud:

"I jokingly suggested for the relief of a senile professoriate an extension of Anthony Trollope's plan mentioned in his novel, The Fixed Period. To one who had all his life been devoted to old men, it was not a little distressing to be placarded in a world-wide way as their sworn enemy, and to every man over sixty whose spirit I may have thus unwittingly bruised, I tender my heartfelt regrets. Let me add, however, that the discussion which followed my remarks has not changed, but has rather strengthened my belief that the real work of life is done before the fortieth year and that after the sixtieth year it would be best for the world and best for themselves if men rested from their labours." — Oxford, July 1906

  • Source: Osler, Aequanimitas, archive.org aequanimitas00osle, full text fetched and read 2026-09-09
  • Why this matters: it was a joke, he apologised for the hurt, and he restated the belief anyway. The most respected physician in the English-speaking world held, on the record, that useful life ended at 40 and men should stop at 60 — 30 years before Social Security. Retirement did not arrive into a neutral world. The ground had been prepared.
  • Date resolved P (session 2): the book's own chapter heading reads "Johns Hopkins University, Feb. 22, 1905." Use 22 February 1905. The "25 February" variant is wrong.

Refuted — keep out V

  • Elderly coresidence with younger kin peaking around 1900 — refuted 0-3

3. How age 65 became compulsory, then stopped being

It was not culture, it was a rule V

Roughly 40–50% of US male employees in the 1960s–70s were covered by rules mandating retirement at 65.

  • Source: Till von Wachter, "The End of Mandatory Retirement in the US", UC Berkeley CLE WP #49, March 2002. Abstract: "In the 1970s, about 40% of male employees in the US were covered by rules mandating retirement at age 65."
  • Vote: 3-0 on the figure
  • Scope: male employees, not all workers. Post-1978 coverage fell to roughly 20%.
  • Caveat: unrefereed working paper. Its legislative history is corroborated by EEOC and CRS.

The legislative kill V (2-1)

Year What
1967 ADEA protects workers 40–65
1978 Amendments extend to 70 — abolishing mandatory retirement at the most common age
1986 Upper age limit removed with narrow exemptions

Script correction — say "with narrow exemptions," not "entirely": bona fide high-policymaking executives (29 CFR 1625.12), some public safety officers, and tenured faculty (exempt until 31 Dec 1993) survived.

Killing it moved behaviour V

"The labor force of workers 65 and older rose by 10% to 20% with the end of mandatory retirement. Neither job tenure nor wage-profiles of older workers were affected by the change."

  • von Wachter (2002), CPS 1968–2000. Vote 3-0.
  • The sharper point is the null result. If firms had really needed a forced exit age to sustain rising age-wage profiles, removing it should have changed tenure and wages. It did not. That argues against the implicit-contract defence of mandatory retirement.

Retirement sold as a product — Sun City S — the 100,000 traced (session 2)

  • June 1959: James G. Boswell II sells 20,000 acres to Del E. Webb Development Co for over $20 million S (Wikipedia)
  • Opened 1 January 1960 — five home models, a shopping centre, a recreation centre, a golf course S
  • Built on the former ghost town of Marinette; named by nationwide contest (prize: a house) S
  • The 100,000 figure is the developer's own number. Del Webb Corporation's corporate history document says the company opened Sun City "on Jan. 1, 1960, hosting 100,000 visitors on its first weekend and selling 1,300 homes by the end of that year." An ASU Archives photo caption (Del E. Webb Corporation collection, MSS 279, Box 70) repeats "over 100,000 people over opening day." Salt River Stories (Arizona State University public-history project) hedges: "tens of thousands of people visited, some estimates put the number at more than 100,000." No independent contemporary count (a January 1960 newspaper tally) was found.
  • Time put Del Webb on its cover on 3 August 1962 S (Wikipedia; Salt River Stories).
  • Sources: Del Webb History (corporate PDF, via Wayback); Salt River Stories, "Sun City: A Revolution". Both fetched 2026-09-09.
  • How to say it on air: "The developer says 100,000 people showed up that first weekend — ten times what they planned for. Take the number as Del Webb's; nobody outside the company counted." That is honest and still lands.

Not covered — real gaps in this thread ?

Inland Steel (1949), the UAW/GM 1950 "Treaty of Detroit", pension coverage growth numbers, and the origin of the phrase "golden years". Nothing survived verification on any of these.


4. The psychology — FILLED (session 2)

How this section was sourced. Jung (1933), Erikson (1950/1982/1986), Cumming & Henry (1961) and Atchley (1976) are copyright-locked: every archive.org edition returns 401 or 403. So the canon is read through open-access scholarship (PMC full text, PubMed abstracts) that quotes it, and tagged S. The empirical literature — the part that actually answers "what does retirement do to a mind" — is mostly P from the papers' own abstracts.

4.1 Freud never said "love and work" — Erikson reported it S strong, plus one P

The myth: Freud, asked what a healthy person should be able to do, answered "lieben und arbeiten." A popular variant adds: "Love and work are the cornerstones of our humanness."

The record: - The sentence appears in Erik Erikson, Childhood and Society (1950; 2nd ed. 1963, p. 265), where Erikson says Freud "was reported to" have answered this way. Erikson is reporting hearsay, and says so. - It cannot be found anywhere in Freud's writings. That is the finding of Alan C. Elms, "Apocryphal Freud: Sigmund Freud's Most Famous 'Quotations' and Their Actual Sources," Annual of Psychoanalysis 29 (2001): 83–104. - Two independent open-access papers state exactly this, citing Elms: Frontiers in Psychology 2020 (PMC7033473) — "The phrase was quoted in 1950 by Erik Homburger Erikson (1963, p. 265) but cannot be found in Freud's writings (Elms, 2001)"; and Frontiers in Psychology 2019 (PMC6834683), which attributes the "cornerstones of our humanness" variant to the same apocryphal tradition. - What Freud actually wrote P — read in the 1930 German first edition of Das Unbehagen in der Kultur (archive.org DasUnbehagenInDerKultur, fetched 2026-09-09):

"Das Zusammenleben der Menschen war also zweifach begründet durch den Zwang zur Arbeit, den die äußere Not schuf, und durch die Macht der Liebe..." ("The communal life of human beings was thus doubly founded: on the compulsion to work, which external necessity created, and on the power of love...") That is the closest genuine Freud gets to the slogan — and it is about the origin of society, not a definition of mental health. - Verdict for air: "Freud never wrote 'love and work.' Erikson said in 1950 that Freud was reported to have said it. Nobody has found it in Freud. What Freud did write, in 1930, is that human society rests on two things — the compulsion to work and the power of love. Close, but a different sentence with a different job." S on the attribution (Elms not read directly; two scholarly papers relaying him), P on the Freud 1930 text.

4.2 Voluntary vs involuntary — the moderator that explains the mess P (abstracts)

This is the missing piece of §5. The health literature averages chosen exits with forced ones. Separate them and the picture clears.

Study What it measured Result
Li, Ye, Zhu & He 2021, Am J Epidemiol 190(10):2220–2230 — meta-analysis of 25 longitudinal studies, 1980–2020 Depressive symptoms after retirement, by type Retirement overall: d = 0.044 (95% CI 0.008–0.080) — tiny. Involuntary: d = 0.180 (0.061–0.299). Voluntary: d = 0.086 (−0.018–0.190, crosses zero). Regulatory (mandatory-age): d = 0.009 (−0.079–0.097, nothing). Eastern developed countries d = 0.126 vs Western d = 0.016.
Zhai, Wang, Liu & Zhang 2022, Front Psychiatry 13:747334 (PMC8854640) — 8 longitudinal articles, 14,604 participants Involuntary retirement vs continued work → depression Pooled RR 1.31 (1.13–1.51). And the reverse: depression → later involuntary retirement, RR 1.70 (1.28–2.25). "There might be mutual causal relationship."
Hershey & Henkens 2014, The Gerontologist 54(2):232–244 — 1,388 Dutch workers, two waves six years apart Life satisfaction by exit type Voluntary leavers reported higher satisfaction than those still employed; involuntary leavers (health, organisational) the lowest of all.
van Solinge & Henkens 2008, Psychology and Aging 23(2):422–434 — 778 Dutch employees, multi-actor panel Adjustment vs satisfaction "Adjustment problems arise from preretirement anxiety about the social consequences of retirement and from a lack of control over the decision." Satisfaction tracks resources: finances, health, marriage. Two separate tasks: losing the work role, and building a life after it.
Dingemans & Henkens 2014, Tijdschr Gerontol Geriatr 45(6):312–320 — Dutch panel Involuntary exit + bridge jobs Involuntary exit → decline in wellbeing vs voluntary; a bridge job compensates; wanting one and failing to find it → decline; working on for money → worse, for intrinsic reasons → better.
van der Heide et al. 2013, BMC Public Health 13:1180 (PMC4029767) — 22 longitudinal studies Retirement → health "Strong evidence for retirement having a beneficial effect on mental health"; contradictory on physical health. "Few studies examined the differences between... voluntary, involuntary and regulatory retirement."
Luhmann, Hofmann, Eid & Lucas 2012, J Pers Soc Psychol 102(3):592–615 (PMC3289759) — 188 publications, N = 65,911 Adaptation to eight life events Retirement (13 prospective samples, 35 effect sizes) is "a typical example of a 'neutral' event that comes with costs and benefits." The reaction is more positive for people who retire earlier or later than usual (quadratic age effect, b = 0.009, CI 0.002–0.015).

The teaching line: retirement itself barely moves depression (d ≈ 0.04 — a rounding error). Forced retirement moves it four times as much, and the arrow runs both ways: depressed people get pushed out, and pushed-out people get depressed. Whenever a study says "retirement causes X," ask who chose it.

All rows P from the papers' own PubMed abstracts, fetched 2026-09-09. Full texts not read.

4.3 Erikson's stages 7 and 8 S — with one honest hole

  • Stage 7, generativity vs stagnation (virtue: care) and stage 8, ego integrity vs despair (virtue: wisdom) come from Childhood and Society (1950), ch. 7, "Eight Ages of Man." The ages usually attached — roughly 40/45–64 and 65+ — are textbook conventions. S Wikipedia, "Erikson's stages of psychosocial development."
  • ? Whether Erikson himself fixed those ages in 1950 could not be checked — the book is locked. Do not say "Erikson said stage 8 starts at 65." Say "the stage usually mapped onto retirement age."
  • Erikson returned to old age directly in The Life Cycle Completed (1982) and Vital Involvement in Old Age (1986, with Joan Erikson and Helen Kivnick). A 2025 open-access paper summarises: in Vital Involvement "he states more directly that integration and despair are two diametrically opposed tensions that the individual will face in old age," and in The Life Cycle Completed defines wisdom as "the strength gained in the struggle between integration and despair." S Frontiers in Public Health (PMC12138197).
  • Did Erikson apply the stages to retirement specifically? ? Not verified. Nothing fetched this session shows him treating retirement as an event. Treat "Erikson on retirement" as an extrapolation by later writers until someone reads Vital Involvement.
  • Joan Erikson added a ninth stage (the eighties and nineties) in the 1997 extended edition of The Life Cycle Completed. S Wikipedia.

4.4 Jung, "The Stages of Life" — the real afternoon-of-life lines S

Two sentences from the essay survive in open-access scholarship, quoted from the 1933 English edition (Modern Man in Search of a Soul):

"The afternoon of human life must also have a significance of its own and cannot merely be a pitiful appendage to life's morning." — Jung 1933, p. 109, as quoted in Margie Lachman, "Mind the Gap in the Middle," Research in Human Development 12(3–4), 2015 (PMC4734389)

"The afternoon of life is just as full of meaning as the morning; only, its meaning and purpose are different." — as quoted in Hastings Center Report, 2026 (PMC12872603)

  • Bibliography S: the essay is "The Stages of Life" (Die Lebenswende), a 1930 lecture published in German in 1931 and in English in 1933; later Collected Works vol. 8. The 1930/1931 dates were not verified against a primary this session.
  • Teaching correction: Jung's "afternoon" begins around 35–40, not 65. It is a midlife essay. Retirement writers borrowed it. Say "Jung wrote about the second half of life, starting at about forty" — never "Jung wrote about retirement."
  • The longer passage everyone quotes ("we cannot live the afternoon of life according to the programme of life's morning...") was not located in an open-access source this session. ? Do not read it on air off this file.

4.5 Atchley — continuity theory is alive; the phase model is a zombie as a sequence P/S

Continuity theory P — Robert C. Atchley, "A Continuity Theory of Normal Aging," The Gerontologist 29(2):183–190, 1989 (PubMed 2519525). Abstract, verbatim: "in making adaptive choices, middle-aged and older adults attempt to preserve and maintain existing internal and external structures; and they prefer to accomplish this objective by using strategies tied to their past experiences of themselves and their social world... Continuity is thus a grand adaptive strategy."

The phase model S — from Atchley's The Sociology of Retirement (1976; locked). As summarised by Butterworth, Gill, Rodgers & Anstey 2023 (Global Health Research and Policy 8:35, PMC10464218): "retirement comprises six stages: preretirement, retirement, disenchantment, reorientation, retirement routine, and termination of retirement. The first stage is often referred to as the 'honeymoon' phase."

The longitudinal test P (abstracts): - Mo Wang 2007, J Applied Psychology 92(2):455–474 (PubMed 17371091) — two Health and Retirement Study samples (n = 994; 1,066), growth-mixture modelling. Three trajectories coexist: a maintaining pattern (wellbeing flat through retirement), a recovering pattern (rises after leaving work), and a U-shaped pattern (dips, then recovers). "Retirees do not follow a uniform adjustment pattern... which reconciles inconsistent previous findings." - Pinquart & Schindler 2007, Psychology and Aging 22(3):442–455 — 1,456 German retirees, latent-class growth. Three groups again, differing by retirement age, gender, SES, marriage, health, prior unemployment. "Retirement is not a uniform transition, and resource-rich individuals are less likely to experience retirement-related change in satisfaction." - Butterworth et al. 2023 (above): "Retirement improves mental health before it happens, but not after" — an Ashenfelter's dip, with depression falling in the two years before retirement and returning to baseline about two years after.

Verdict: Atchley's honeymoon → disenchantment → reorientation → stability is dead as a universal sequence — most people never show the disenchantment dip, and the three-trajectory result has replicated in US and German panels. It survives as a menu: some retirees do walk that path. Say "one of several paths," never "the stages of retirement."

4.6 Elliott Jaques, "Death and the Mid-Life Crisis" (1965) P citation, S content

  • Citation P: E. Jaques, "Death and the mid-life crisis," International Journal of Psycho-Analysis 46(4):502–514, 1965. PubMed 5866085. Confirmed from the PubMed record.
  • Content S — from Mark Jackson, "Life begins at 40: the demographic and cultural roots of the midlife crisis," Notes and Records of the Royal Society 74(3):345, 2020 (PMC7434712, open access):
  • Jaques first presented the paper to the British Psychoanalytical Society in 1957; the IJP did not accept it for eight years.
  • It rested on "a study of 300 creative artists as well as case histories from his clinical practice."
  • The crisis he described was a depressive crisis around the age of 35, "energetically masked by a manic determination to thwart advancing years" — the hypochondria, promiscuity, and religiosity that later became the cliché.
  • Jaques began thinking about it in 1952, at 35 himself, after finishing his own analysis with Melanie Klein and reading Dante's Inferno.
  • Teaching correction: the midlife crisis was coined in 1965, about 35-year-olds, from a sample of dead artists. It is not a finding about retirement and not a finding about 50. Jackson's thesis: it is a product of post-war demography and the American dream, not a biological fact.

4.7 "Retired husband syndrome" — real, small, temporary, and American in origin P

  • The origin is not Japanese. The term appears in C. C. Johnson, "The retired husband syndrome," Western Journal of Medicine 141(4):542–545, October 1984 (PubMed 6506693, PMC1021891 — scanned, no OCR). That is seven years before Nobuo Kurokawa's 1991 presentation to the Japanese Society of Psychosomatic Medicine S (Wikipedia), where the Japanese name shujin zaitaku sutoresu shōkōgun (主人在宅ストレス症候群, "husband-at-home stress syndrome") comes from S (IZA paper below). Bertoni & Brunello's literature review opens with Johnson's 1984 American patients: "I am going nuts," "He is under my feet all the time."
  • Formal evidence P (abstracts):
  • Bertoni & Brunello 2017, "Pappa Ante Portas: The effect of the husband's retirement on the wife's mental health in Japan," Social Science & Medicine 175:135–142 (PubMed 28088619). "The 'Retired Husband Syndrome'... has been anecdotally documented but never formally investigated." Using a 2006 Japanese reform as an instrument: "the husband's earlier retirement significantly increases the probability that the wife reports symptoms." Retirement also hurt the husband's own mental health and the household's finances. Working-paper version (IZA DP 8350, 2014, read in full): one more year of the husband's retirement raises the wife's probability of symptoms by 5.8 to 13.7 percentage points (a 12–29% increase). S on those numbers — the published version may differ.
  • Oshio 2021, Journal of Epidemiology 31(5):308–314 (PMC8021876) — 3,794 Japanese wives, 12-wave panel. Kessler-6 distress (0–24 scale, mean 3.41) rose by 0.18, 0.18 and 0.19 points in the first three years after the husband retired, then declined toward baseline. Wives with active social lives, close interaction with the husband, or no paid job beforehand showed no rise. The author's own conclusion: "the limited relevance of the 'retired husband syndrome'."
  • The BBC's widely repeated claim that "over 60 percent of older Japanese women are affected" is reported by Bertoni & Brunello as a media figure with no study behind it. S Do not repeat the 60%.
  • Verdict for air: "It's real — a husband's retirement measurably raises his wife's stress. It's also small, it fades within about three years, and the name was coined in California in 1984, not Tokyo in 1991."

4.8 Disengagement theory S — unchanged, one citation firmed

Formulated by Elaine Cumming and William Henry, 1961, Growing Old — the first theory of aging social scientists developed — on the Kansas City Study of Adult Life (211 adults aged 50–90). Proposed mutual withdrawal as innate, universal, unidirectional and irreversible. Cumming walked it back in "Further Thoughts on the Theory of Disengagement" (1963) S and again in "New thoughts on the theory of disengagement," International Journal of Psychiatry 6(1):53–67, 1968 P citation (PubMed 5663544). Growing Old is locked (archive.org 403).

4.9 Mourning as a mechanism V (2-1) — read the scope flag

Rachel Gibbons, BJPsych Advances 30(2), 2023: psychiatric illness "can be understood to result from 'pathological mourning' due to arrests, or retreats, in the passage through the mourning process." Nothing in this source is about retirement. Applying it to role loss is an analytical bridge, not Gibbons' claim.

4.10 The map of the field P

Mo Wang & Junqi Shi, "Psychological Research on Retirement," Annual Review of Psychology 65:209–233, 2014 (PubMed 23751036). Reviews three models — the temporal process model, the multilevel model, and the resource-based dynamic model of adjustment — and the antecedents (individual, job, family, socioeconomic) of financial, physical and psychological wellbeing in retirement. If a listener wants one reading, this is it.

Still unsourced in this thread ?

Activity theory (Havighurst), role theory, terror-management readings of retirement, "retirement syndrome" as a clinical label, and the exact 1950 wording of Erikson's stage ages. Do not air.


5. What retirement actually does to people — STRONGEST SECTION

"Work longer, live longer" is mostly a statistical illusion V

The single best result in this file.

Comparison Hazard ratio Meaning
On-time retirement vs working on, poorly adjusted 1.56 (95% CI 1.41–1.73) looks strongly deadly
Same, adjusted for prior health 1.12 (95% CI 0.98–1.28) crosses 1 — effect gone
Early retirement vs working to retirement 1.05 (95% CI 0.87–1.28) no association at all

The residue is the healthy worker effect: healthier people select into staying employed. Work is not extending their life; their health is extending their work.

  • Source: Sewdas et al., "Association between retirement and mortality: working longer, living longer? A systematic review and meta-analysis," J Epidemiol Community Health 74(5):473–480, 2020 (PubMed 32079605, PMC7307664). Meta-analysis of 25 longitudinal studies (12 in the pooled estimates). Citation re-confirmed from PubMed 2026-09-09.
  • Vote: 3-0 on all three merged claims
  • Corroborated by a 2023 Journal of Economic Surveys meta-analysis (Filomena, 2000–2021) using near-identical healthy-worker language
  • Two honest hedges: the authors write "might reflect," not "is." And the wider literature is more mixed — a Swedish construction-worker cohort and some age-62 analyses do find higher mortality. Those do not refute this adjusted result, but the heterogeneity is real.
  • The teaching line: "Does retiring kill you? No. The studies that said it did mostly forgot to check who was already sick."

"Use it or lose it" is far weaker than you have been told V

A systematic review found only 7 longitudinal studies met inclusion criteria.

"We found weak evidence that retirement accelerates the rate of cognitive decline in crystallised abilities, but only for individuals retiring from jobs high in complexity with people." "The evidence of the impact of retirement on the rate of decline in fluid cognitive abilities is conflicting."

  • Source P citation resolved (session 2): Annette Meng, Mette Andersen Nexø & Vilhelm Borg, "The impact of retirement on age related cognitive decline – a systematic review," BMC Geriatrics 17(1):160, 2017 (PubMed 28732489, PMC5520232). PRISMA-guided. The earlier "17:259" was wrong; "Nagy et al." was wrong. It is Meng, Nexø & Borg, 17:160.
  • Vote: 3-0
  • A 2025 systematic review with 20 studies still reports negative, null and positive associations. Eight more years, still unsettled.
  • REFUTED 1-2, do not say: that the authors "explicitly state the evidence base is too sparse for firm conclusions." Do not put that sentence in their mouths.
  • ? The Rohwedder & Willis "mental retirement" hypothesis was not verified. Do not cite it off this file.

The honest overall answer: it depends, and on what is documented V

An overview of 15 systematic reviews (4 meta-analyses) concludes retirement's health impact remains inconclusive — positive, negative and null effects all reported.

What separates winners from losers is socioeconomic status: - Higher-SES retirees: improved mental health, increased physical activity - Lower-SES retirees: declines in physical and mental health, increased sedentary behaviour, adverse cardiovascular outcomes

  • Source: Vigezzi et al., "Impact of retirement transition on health, well-being and health behaviours: critical insights from an overview of reviews," Social Science & Medicine 375:118049, 2025 (PubMed 40250262). Citation re-confirmed 2026-09-09.
  • Vote: 3-0 on method and "inconclusive"; 2-1 on SES
  • Precision fix: say "the best-documented moderator," not "the key moderator" — the source lists SES alongside job characteristics and lifestyle.
  • Now add §4.2: the other moderator, chosen vs forced, has its own meta-analyses. Together they are the frame: "Good for whom, from what job, chosen or forced?"

6. Retirement today — FILLED (session 2)

6.1 "Okinawans have no word for retirement" — origin found, claim killed as stated P

  • Origin P: Dan Buettner, TED 2009, "How to live to be 100+," transcript: "In the Okinawan language there is not even a word for retirement. Instead there is one word that imbues your entire life, and that word is 'ikigai.'" Fetched via Wayback 2026-09-09.
  • Japanese has at least three P (ja.wikipedia, fetched 2026-09-09): 定年 teinen — the fixed age at which employment ends by rule ("定年退職"); 退職 taishoku — leaving a job, ending the labour contract; 引退 intai — withdrawing from a position or a sport. Every Okinawan alive speaks Japanese.
  • What the claim is actually about is Uchinaaguchi, the Ryukyuan language, and whether it had a native word before modern pensions existed. ? Unverified in either direction, and beside the point: a language spoken by subsistence farmers and fishermen before 1945 had no pension system to name.
  • The myth has reached the journals: a 2025 Frontiers in Aging Neuroscience paper repeats "there is no word for retirement in Okinawan language" as an uncited aside (PMC12350274). That is how a TED line becomes a citation.
  • Verdict for air: "The 'no word for retirement' line is Dan Buettner's, from a 2009 TED talk. Japanese has three words for it. The claim is about a dialect nobody checked — and it wouldn't matter if it were true, because there was nothing to retire on."

6.2 Saul Newman — the Blue Zones and bad paperwork P

  • Ig Nobel Prize in Demography, 2024 P (improbable.com winners list, fetched via Wayback): "Saul Justin Newman, for detective work to discover that many of the people famous for having the longest lives lived in places that had lousy birth-and-death recordkeeping."
  • The paper P (abstract, bioRxiv API, v4 dated 2025-11-03): Newman, "Supercentenarian and remarkable age records exhibit patterns indicative of clerical errors and pension fraud," bioRxiv 10.1101/704080. Findings as stated: in the US, "supercentenarian status is predicted by the absence of vital registration"; the state-by-state introduction of birth certificates "is associated with a 69–82% fall in the number of supercentenarian records"; "only 18% of exhaustively validated supercentenarians have a birth certificate, falling to zero percent in the USA"; supercentenarian birthdates "are concentrated on days divisible by five"; and "the designated blue zones of Sardinia, Okinawa, and Ikaria corresponded to regions with low incomes, low literacy, high crime rate and short life expectancy relative to their national average."
  • Hedge: a preprint, not peer-reviewed as of the fetch. Say "a preprint that won an Ig Nobel," not "a study proved."
  • Pairs with the koseki (family register) destruction in the 1945 Battle of Okinawa already in this file S.

6.3 The ikigai Venn diagram is a 2014 blog post, adapted from a 2012 Spanish graphic P

  • Marc Winn, "What is your Ikigai?", theviewinside.me, 14 May 2014 P (Wayback, fetched 2026-09-09). The post introduces the four-circle diagram (what you love / are good at / the world needs / can be paid for), credits nothing, and points readers to Buettner's TED talk.
  • Andrés Zuzunaga, in his own comment on that post, 29 April 2017 P: "This is copy of a spanish graphic made Andrés Zuzunaga two years before (2012): and it was published and well quoted in 2013 in this book. You can find many copies of them but no one with a creation date before 2012." His diagram was labelled propósito (purpose), not ikigai.
  • Wikipedia ("Ikigai") dates Zuzunaga's diagram to 2011 and says Winn's was "likely inspired" by it. S — the artist himself says 2012. Use 2012, his date.
  • ? Winn's own later admission that he "merged" the purpose diagram with the word ikigai was not found in the archived versions of his page. Do not quote it.
  • What the Japanese research actually asked P: Sone et al., "Sense of life worth living (ikigai) and mortality in Japan: Ohsaki Study," Psychosomatic Medicine 70(6):709–715, 2008 (PubMed 18596247) — 43,391 adults, 7-year follow-up, 3,048 deaths. The instrument was one question: "Do you have ikigai in your life?" People who answered no had all-cause mortality HR 1.5 (1.3–1.7), cardiovascular 1.6 (1.3–2.0), external-cause 1.9 (1.1–3.3), cancer 1.3 (1.0–1.6, not significant).
  • Ikigai was first popularised in Japan by psychiatrist Mieko Kamiya's 1966 Ikigai ni tsuite. S Wikipedia.
  • Verdict for air: "The diagram is a Guernsey blogger's 2014 post, built on a Spanish designer's 2012 'purpose' graphic, with one word swapped. The Japanese study behind the longevity claim asked one yes-or-no question. No circles."

6.4 401(k) — correction confirmed against the statute and EBRI P

  • Congress enacted it P: Pub. L. 95–600 (Revenue Act of 1978), title I, §135(a), Nov. 6, 1978, 92 Stat. 2785 "added subsec. (k)" to IRC §401, "applicable to plan years beginning after December 31, 1979." Source: 26 U.S.C. §401, historical and revision notes, uscode.house.gov, fetched 2026-09-09.
  • The 1974 freeze P: ERISA "barred the issuance of Treasury regulations prior to 1977 that would impact plans in place on June 27, 1974," freezing an IRS proposal that would have restricted cash-or-deferred plans; Congress extended the moratorium twice, the IRS withdrew the rule in 1978, and ERISA's mandated study "influenced 1978 legislation creating 401(k) plans."
  • Lipsig P: "1978 — In December, Ethan Lipsig, of the outside law firm for Hughes Aircraft Company, sent a letter to the company recommending that it begin the process of turning its savings plan into a 401(k) plan." Then: "1979 — Johnson & Johnson began the process"; "1979–1982 — Several companies... developed 401(k) plan proposals, many of which officially began operation in January 1982"; regulations issued November 1981. Source: EBRI, "History of 401(k) Plans: An Update", 5 Nov 2018, read in full.
  • Corrections to the draft in this file: 1. Wikipedia's "first implementation... about three weeks after Section 401(k) was enacted" overstates its own source. EBRI says a letter recommending a process, not a live plan. Say "the first documented move to build one." 2. EBRI's history never mentions Ted Benna. His "inventor" title is a press label (MarketWatch, 2016: "The inventor of the 401(k) says he created a 'monster'"). S The Johnson Companies plan and the 70% top marginal rate detail rest on Wikipedia/NYT 2024. S
  • Verdict for air: "Congress wrote 401(k) into the tax code on November 6, 1978. The first company on record moving to use it was Hughes Aircraft, on its lawyer's advice, that December. Plans went live in 1982. Ted Benna built one of the early ones — he did not invent it."

6.5 The pension-to-401(k) shift, in numbers P

Active participants in private pension plans (thousands) — DOL Form 5500 filings, Table E7:

Year Defined benefit Defined contribution
1975 27,214 11,217
1984 (DB peak) 30,073 30,545
1995 23,395 42,203
2005 20,310 62,355
2015 14,405 78,130
2022 11,333 92,602
  • Source: US Department of Labor, EBSA, Private Pension Plan Bulletin Historical Tables and Graphs 1975–2022, Table E7 (PDF via Wayback), fetched 2026-09-09.
  • Caveats printed in the table's own notes: "active participants" in DC plans includes workers eligible to defer who do not; the participant definition changed in 2004 (the 2004r row); one-participant plans are excluded. Use the direction and rough magnitude, not decimal precision.
  • Access today P: BLS, Employee Benefits in the United States — March 2025: "Retirement benefits were available to 72 percent of private industry workers... Seventy percent of private industry workers had access to defined contribution plans and 14 percent had access to defined benefit plans." (bls.gov news release, via Wayback), fetched 2026-09-09.

6.6 Where Americans actually stand P

Survey of Consumer Finances 2022 (Federal Reserve Bulletin, "Changes in U.S. Family Finances from 2019 to 2022," Oct 2023, PDF, read): - 54.3% of families hold any retirement account (2019: 50.5%). So 45.7% hold none. - Among holders: conditional median $86,900, mean $334,000 (2022 dollars). - Working-age participants (reference person 35–64) with an IRA or DC plan, mean balance by usual income: bottom half $54,700 (down from $66,600 in 2019); 50th–90th percentile $226,700; top decile $913,300. Overall mean $331,400. - ? Median retirement balance by age band was not pulled — it lives in the SCF's supplementary tables, not the bulletin. Do not quote by-age medians off this file.

Social Security — 2026 Trustees Report (summary at ssa.gov/OACT/TRSUM, fetched via Wayback snapshot of 2 Sep 2026): - OASI (retirement) trust fund depleted in Q4 2032 — one quarter earlier than the 2025 report — after which continuing income pays 78 percent of scheduled benefits, declining to 62% by 2100. - Combined OASDI: Q3 2034, then 83 percent payable. - Medicare HI: Q2 2033, then 89 percent. - DI: solvent through the 75-year window. - Say "78 percent in late 2032" for retirement benefits alone; "83 percent in 2034" is the combined figure that assumes a law change to merge the funds. Both are in the summary; the two get mixed up constantly.

6.7 Carstensen and "The New Map of Life" — what it actually proposes P

  • What it is: an initiative of the Stanford Center on Longevity, launched 2018; full report November 2021; short report April 2022. Laura Carstensen is the Center's founding director S (Wikipedia). It is a Center report with many authors — not a Carstensen book. The user's "(2022)" is the short report.
  • The frame P: "The 100-year life is here. We're not ready." The institutions "evolved when lives were only half as long."
  • The retirement proposal, verbatim P (short report, April 2022): "Rather than fixed chapters that span 70 years and focus consecutively on education, work, and retirement, we envision several shorter, flexible intervals dedicated to learning, working, caregiving, and leisure that are woven as needed into the course of 100-year lives, with working intervals likely to include more than one primary career." And: "Rather than associating the second half of life principally with menopause, retirement, and death, we can create social observances around meaningful milestones such as volunteering and mentoring, returning to work, launching an encore career or a new business."
  • On work P: "Over the course of 100-year lives, we can expect to work 60 years or more. But we won't work as we do now, cramming 40-hour weeks into lives impossibly packed... Workers seek flexibility... paid and unpaid intervals for caregiving, health needs, lifelong learning."
  • The named principles: Make the Most of the 100-Year Opportunity · Invest in Future Centenarians · Align Health Spans to Life Spans · Prepare to Be Amazed by the Future of Aging · Life Transitions Are a Feature, Not a Bug · Learn Throughout Life · Work More Years with More Flexibility · Build Financial Security From The Start · Build Longevity-Ready Communities · Age Diversity Is a Net Positive · Ensure Advances Are Distributed Deeply. Source: longevity.stanford.edu/the-new-map-of-life-report and the April 2022 short report PDF, both fetched 2026-09-09.
  • Teaching line: the proposal is not "retire later." It is "stop having one retirement." Replace education → work → retirement with many short intervals, and work sixty years because you work fewer hours in each of them.

6.8 France 2023 — 62 to 64, by Article 49.3 P

  • 16 March 2023 P: Compte rendu, deuxième séance du jeudi 16 mars 2023, Assemblée nationale. Prime Minister Élisabeth Borne: "Sur le fondement de l'article 49, alinéa 3, de la Constitution... j'engage la responsabilité de mon Gouvernement" — over prolonged booing, RN deputies pounding their desks, and LFI deputies holding up "64 ans c'est non" signs (the record notes all of it). The president of the Assembly: "L'Assemblée nationale prend acte de l'engagement de la responsabilité du Gouvernement conformément aux dispositions de l'article 49, alinéa 3." The text is deemed adopted unless a censure motion filed within 24 hours passes.
  • 20 March S (fr.wikipedia): the cross-party LIOT censure motion failed by nine votes — 278 for, 287 needed.
  • 14 April 2023 P: Conseil constitutionnel, décision n° 2023-849 DC. Article 10 of the law shifts "l'âge légal de départ à la retraite de soixante-deux à soixante-quatre ans," rising "à raison de trois mois par génération" until the 1968 cohort, and accelerates the contribution-period increase (172 quarters for those born from 1965). The Council found the 49.3 procedure constitutional, upheld the age change, and struck several riders as "cavaliers sociaux" (articles 2, 3, 6, part of 10, part of 17, 27 — including the "index seniors" S). Published JORF n° 0089, 15 April 2023.
  • Promulgated S the same night (Loi n° 2023-270 du 14 avril 2023). Légifrance blocked the fetch; the law number rests on fr.wikipedia.
  • Verdict for air: "France raised its retirement age from 62 to 64 in 2023 without a vote in the Assembly — the government used Article 49.3, survived a censure motion by nine votes, and the constitutional court signed off a month later. The age moves three months a year until people born in 1968 retire at 64."

Not covered ?

Unretirement rates, FIRE and the 4% rule, and SCF medians by age. Not fetched. Do not air.


Method, and what it cost

  • Session 1 (2026-09-09, run 1: broad harness, 111 agents, 8.1M tokens): 28 sources → 70 claims → 25 adversarially verified → 21 confirmed, 4 refuted → 12 findings. Runs 2–4 died when three concurrent workflows exhausted the 200-search session budget. A recovery pass of direct curl produced the SSA and Osler primaries.
  • Session 2 (2026-09-09, evening): no Workflow, no harness, two WebSearch calls in total (Myers; Sun City). Everything else was curl: NCBI E-utilities (PubMed abstracts + PMC full text — zero search budget, the single best tool in this file), archive.org full text and metadata, Wayback for ssa.gov / bls.gov / dol.gov / ted.com / ebri.org, Wikipedia and ja.wikipedia extracts as leads, uscode.house.gov, conseil-constitutionnel.fr, assemblee-nationale.fr, federalreserve.gov, longevity.stanford.edu, the Google Books API (dead — returned nothing for every query), Open Library. Roughly sixty fetches. Filled §4 and §6, resolved all four flagged items, and corrected one citation in §5.
  • The lesson, confirmed twice: the fan-out harness is for contested claims with no reachable primary. For everything else, a URL and grep beat it on cost, speed, and quote quality.

Open questions worth chasing

  1. Read Erikson's Vital Involvement in Old Age (1986) — does he treat retirement as an event at all, and does the 1950 text fix stage ages? (A library copy; the web will not give it up.)
  2. Read Elms 2001 directly (PEP-Web) to close the Freud item from S to P.
  3. SCF 2022 supplementary tables for median retirement balances by age band.
  4. A January 1960 Arizona Republic count for Sun City's opening weekend — the only thing that would move the 100,000 off the developer's say-so.
  5. Was Ransom & Sutch's "on-the-job retirement" real gradual withdrawal, or disguised unemployment (Moen)? Still open from session 1.