Tuesday, 12 September 1978. Alma-Ata, Soviet Kazakhstan.
The auditorium of the Lenin Convention Center holds 3,000 people and has a system for simultaneous interpretation. Delegations from 134 states sit in it, with 67 international organisations and dozens of voluntary groups, at the close of a conference on primary health care. A young African woman, a physician, in African dress, reads out the declaration they have agreed.
It asks that basic care be brought “as close as possible to where people live and work”, with the people’s own participation, and at a cost each community and country can afford to keep paying. It wants that care to rest on community workers as well as on doctors, nurses and midwives. Its list of what such care must include runs from nutrition and safe water to immunisation, treatment of common illnesses and essential drugs. And it sets a date. By the year 2000, all the world’s peoples are to reach a level of health that lets them lead socially and economically productive lives.
Halfdan Mahler, who runs the World Health Organization, is in the hall. He expects this to be the most decisive conference the organisation has held since its founding in 1948.
The conference is Soviet-sponsored, held with $2 million offered by the Soviet deputy health minister. China’s barefoot doctors are among the movement’s main inspirations. China is absent.
Thirty years later Mahler remembered people in the hall with tears in their eyes. “That was a sacred moment,” he said. Three months before it, an American delegation had watched the Chinese model at work. Between 7 June and 1 July 1978, a group of doctors, health planners and sociologists toured rural China as guests of the Chinese Medical Association. The National Institutes of Health published their report in 1980.
In Taoyuan county, Hunan, they visited a brigade of 872 people in 195 households. Its health station had examining rooms, two rooms where sicker patients could lie for a few hours, a room for births and minor surgery, a room for injections, one for preparing herbal medicines, and a dispensary. Two barefoot doctors ran it. Their day began early, and each saw about fifteen patients. The two also made 150 home visits a month, usually to very young babies and the old. Five health aides, one for each production team, inspected the brigade’s drinking water, latrines, pigsties and kitchens.
Above the brigade stood the commune hospital. One in Shaanxi had 64 staff and 50 beds, saw 100 outpatients a day, backed up 18 brigade health stations and trained their 43 barefoot doctors. There, barefoot doctors gave regular well-child checkups every three months, and paediatricians came out from the commune hospital twice a year. One barefoot doctor in each brigade kept the vaccinations, and because that doctor knew every family, every birth and where every child was, the visitors judged the coverage close to complete. Above that stood the county hospital, which for a villager could be a day’s journey by foot, bicycle or bus, and a day away meant lost work points.
A barefoot doctor was a peasant with a short training: anything from three months to two years, and about six months on the delegation’s average, at a cost it put at some 500 yuan a head. A typical brigade had three or four. Each staffed the clinic every third or fourth day and spent the rest of the time growing medicinal herbs or working in the fields, roughly a third of the time on each. They earned what an average farm worker in their brigade earned. Near Guilin that was 45 yuan a month, when a qualified doctor drew 50 to 300, and they took no fees from patients.
The money came from the collective. In one Guangdong commune in 1977, the brigades’ welfare funds, drawn from farm income, met 55 percent of the cooperative medical fund; the commune paid 13 percent and members’ premiums the rest. Premiums ran from 40 fen a year in one brigade near Beijing to 2.4 yuan in that commune, and most brigades spent 70 to 80 percent of their health budgets on drugs.
Each brigade also typically had one woman barefoot doctor trained in birth planning. She inserted and removed IUDs, could often perform abortions by vacuum aspiration, handed out pills and condoms, and kept a record of every married woman of childbearing age: the number, ages and sexes of her children and the contraceptive she used. Barefoot doctors were expected to lead the push for later marriage, wider spacing and fewer children.
The visitors were not starry-eyed. In Miyun county, outside Beijing, a barefoot doctor emptied her kit and could not say what chlorpromazine, one of her fifteen medicines, was for. They tried barefoot doctors on two imaginary patients. Most rightly called a fourteen-year-old boy with a day of pain low on his right side and a fever a case of appendicitis, but they disagreed on what to give him and when to send him on. An eight-year-old with a fever and pus on swollen tonsils was prescribed penicillin, aspirin or streptomycin, depending on whom they asked. One of their own escorts, a former barefoot doctor from Heilongjiang, told them she had several times taken pneumonia for a cold and so withheld the antibiotic the patient needed. Some figures they could not accept. A county hospital in Guangxi reported treating only three cases of mental illness in three years, where the World Health Organization’s studies put schizophrenia alone at 2 to 10 cases in every thousand people.
The design relied on referral. Barefoot doctors were taught to send a patient up to the commune hospital whenever in doubt, and they were said to handle 75 to 90 percent of cases themselves. Near Guilin, one had taken a sixty-year-old woman’s illness for arthritis. At the commune hospital it proved to be tuberculosis of the bone, and afterwards the barefoot doctor called at her home every day. The delegation doubted some of the record-keeping, but saw enough to give China’s claims general credence. Its report credited the system with immunising almost every child and with ending venereal disease and most flies. It cited one of the delegation’s own members, Pi-chao Chen: China’s realistic choice had not been between high-quality and low-quality services, but between competent care for some, at great distance and prohibitive cost, and some care for all, at a cost the peasants could bear. China had chosen the second. Nor had it spent unusually. By the delegation’s reading, China’s health budget was not above the developing-country average (Chen put it at 1.76 percent of GNP in 1973), and the state’s money was topped up with local labour and contributions from the communes.
The World Bank’s series show where this had left the two countries. In 1978 China produced $157 a head at official exchange rates and India $209: on that measure China was the poorer country. Yet a baby born in China that year could expect to live to 63, and one born in India to 52. Of every thousand children born, 70 died before the age of five in China, and 181 in India. A Chinese woman could expect 2.7 children in her lifetime and an Indian woman 4.9. At the censuses of the early 1980s, about two Chinese adults in three could read, against two Indians in five. Among women it was one in two against one in four.
By the time the reforms began, China's people lived eleven years longer
Life expectancy at birth. China's line starts in the depths of the famine of 1959–61, overtakes India's by 1962, and is eleven years ahead by 1978: the foundation was poured before the reform, at a monstrous price.
Source: World Bank, World Development Indicators (SP.DYN.LE00.IN), fetched Oct 2026
See the numbers
| Year | China | India | China's lead |
|---|---|---|---|
| 1960 | 33.4 | 45.6 | −12.2 |
| 1965 | 53.0 | 45.6 | +7.4 |
| 1970 | 56.8 | 48.6 | +8.2 |
| 1975 | 61.0 | 50.8 | +10.2 |
| 1980 | 64.2 | 53.6 | +10.6 |
| 1985 | 66.5 | 56.3 | +10.2 |
| 1990 | 68.2 | 58.6 | +9.5 |
| 1995 | 70.4 | 60.6 | +9.7 |
| 2000 | 72.3 | 62.8 | +9.5 |
| 2005 | 74.1 | 64.9 | +9.1 |
| 2010 | 75.7 | 67.2 | +8.5 |
| 2015 | 77.0 | 69.3 | +7.6 |
| 2020 | 78.0 | 70.2 | +7.9 |
| 2024 | 78.0 | 72.2 | +5.8 |
The basement was poured at a terrible price. In 1960, in the depth of the famine, a baby born in China could expect 33 years of life, against 46 in India. Jean Drèze and Amartya Sen, writing in 1989, found estimates of the famine’s extra deaths in 1958–61 ranging from 16.5 to 29.5 million, five to ten times the toll of the largest famine India suffered in the twentieth century. Later counts run higher: the Chinese journalist Yang Jisheng put the toll at about 36 million. China’s life expectancy passed India’s in 1962, and by 1978 it was eleven years ahead.
India paid in a different currency. It had suffered no large famine since independence, but its ordinary death rate stayed higher. Drèze and Sen also took the death rates of the mid-1980s, 12 per thousand in India against 7 in China, and calculated that the gap cost India about 3.9 million more deaths a year, so that every eight years or so, on their reckoning, more Indians died of it than had died in China’s famine.
Seventeen children in a hundred, against six
Of every hundred children born, how many died before their fifth birthday, in 1980.
Source: World Bank, World Development Indicators (SH.DYN.MORT), fetched Oct 2026
See the numbers
| Country | Deaths per 1,000 live births | Per 100 children (rounded) |
|---|---|---|
| India | 169.8 | 17 |
| China | 62.6 | 6 |
The same two economists drew the conclusion that should change how an Indian reads a growth chart. China’s growth, they wrote, had not been radically faster than India’s before the reforms of 1979, and by then the great gains in health and longevity had already been made. On their dating, the healthier worker did not come out of China’s boom. She was already there when it began.
Most accounts of China’s rise begin in the factories. This one begins underneath them.
India’s first Institute of Technology began in 1950 in an office at 5 Esplanade East, Calcutta, and moved that September to Hijli, near Kharagpur. On 18 August 1951 it was formally inaugurated, with 224 students and 42 teachers. Its classrooms, laboratories and office were in the buildings of the Hijli Detention Camp, which had opened in 1930 to hold Indian revolutionaries. By one account, Nehru chose the site on the advice of West Bengal’s chief minister, Bidhan Chandra Roy. In 1956 Parliament passed an act for the institute, and at its first convocation that year Nehru called the campus a symbol of “India’s future in the making”. At the census of 1951, 18 percent of Indians aged five and over were counted as literate. India built the penthouse first.
Chain I231951 → 2016Planted: the lift's graduates and the public rails of A Second Thumbprint19472026195107:00KalpaTen weeks later India began to vote. Kinnaur, high in the Himalaya, voted first, because early snow would soon close it, and the ballot material went up to its polling stations on mules. At seven in the morning of 25 October 1951, Shyam Saran Negi, a schoolteacher on election duty, thirty-four by his official records, reached the booth at the primary school in Kalpa, then called Chini, about 10,000 feet up. He is recorded as having cast the first vote of independent India’s first general election.
The man behind the election was Sukumar Sen: a civil servant since 1921, chief secretary of Bengal from 1947 to 1950, and from March 1950 India’s first Chief Election Commissioner. In September 1951 he and his staff rehearsed with mock elections in Delhi. He had also made a decision about names. Many women in northern India had asked to be entered on the rolls only as somebody’s mother or somebody’s wife. Sen called the custom a “curious senseless relic of the past” and refused to enrol them until they gave their own names. By one account nearly 2.8 million of about 80 million women voters never did.
Polling began on 25 October 1951 and ran into 1952. By one count there were more than 224,000 booths, and about a million officials ran them. For 489 seats in the Lok Sabha, 1,874 candidates stood from 53 parties. About 173 million people could vote, and each party’s symbol was painted on its ballot box, so that a voter who could not read could still choose. India gave the vote to every adult at once, before most of them could read the ballot.
173 million voters, more than four in five of whom could not read
Every adult got the vote at once, and parties were marked by pictures so that everyone could choose. Literacy in 1951 was about 18 percent. (Pictograms are illustrative, not party symbols.)
Source: Election Commission of India; Census of India 1951
See the numbers
| Voters | Of every 100 |
|---|---|
| Could read the ballot (literacy, Census 1951, about) | 18 |
| Voted by its pictures | 82 |
| Electorate, 1951–52 | 173 million |
China’s lower floors were ordered from the top. On 26 June 1965 Mao told the Ministry of Public Health that it worked for only about 15 percent of the population, mostly what he called gentlemen, while the peasants had neither doctors nor medicine. He asked why it should not be renamed the Ministry of Urban Gentlemen’s Health. Three years of training after higher primary school, he said, would be enough: even without much talent, such doctors would be better than quacks and witch doctors, and villages could better afford to keep them. A few of the less able recent graduates should stay in the cities, and the rest go to the countryside. Research, he added, spent too much on rare and difficult diseases and too little on preventing and treating common ones. Of medical schooling he said: “The more books one reads the more stupid one gets.” That year rural cooperative medical insurance was extended to ten provinces. During the Cultural Revolution decade, by one compilation, about 1.4 million barefoot doctors were at work and more than nine farmers in ten belonged to the scheme.
The same decade that filled the bottom floors closed the old way up to the top one. In June 1966 the year’s university enrolment was postponed by half a year, and in July the entrance examination was ordered abolished in favour of recommendation and selection; classes had stopped, and neither order was carried out at the time. By 1971, of China’s 434 institutions of higher education, 328 were still running. The way up ran through the villages now. In 1975 Zhou Enlai told the National People’s Congress that nearly ten million school graduates had gone to the countryside. When the Americans visited Zhongshan Medical College in Guangzhou in 1978, about one in ten of its 2,000 or so students had been a barefoot doctor, and an applicant needed a letter from fellow members of the brigade saying he or she had shown a willingness to serve the people. China built from the bottom and, for a decade, closed the way up.
1970JamkhedIndia’s working model came from below, and not from the state. In 1970 two doctors, Rajanikant and Mabelle Arole, set up a rural health project at Jamkhed, in Maharashtra. Each village chose its own health worker. The first were mostly widows or women from Dalit castes, and many could not read. Once a week they came to the project’s centre and stayed the night for training. By the founders’ own figures, infant deaths in the project area fell from 176 per thousand births to 52 within five years, and by 1985 the work reached 250 villages. A later independent evaluation found deaths among children aged one month to five years about 30 percent lower in Jamkhed’s villages than in comparison villages between 1992 and 2007. Perry and Rohde argue that Jamkhed helped shape the ideas written into the declaration read out at Alma-Ata.
1977The state’s version came later and took a different shape. A government committee set up in 1974, when rural death rates were nearly double urban ones, proposed a village worker drawn from the community, neither paid nor supervised by the state, and free to work on whatever trust she or he could earn. On 2 October 1977 the government launched a community health worker scheme, whose workers were later renamed village health guides, in 741 primary health centres across 28 districts. About 66,000 were working eighteen months later, and by 1980 about 150,000 had been trained, reaching a third of rural India. Each was paid Rs 200 during training and then Rs 50 a month, with Rs 50 worth of medicines and supplies. Between 75 and 94 percent of them were men, which, in Strodel and Perry’s words, made their selection “in effect a part of the political patronage system”. Almost half had no drug kit in 1979. That year the states were asked to pay half the cost. Several dropped the scheme at once, and most dropped it when the next government made the states share the cost equally. It was abandoned for a year and revived in 1981 with central funding. The government’s planners, Perry and Rohde write, had tried to copy only Jamkhed’s training.
The purse did not last. Ten weeks after Alma-Ata, by the date the village later worked out, came the meeting in Xiaogang that opens this book, where household heads agreed to divide the team’s land. The American report had already noted the first signs: newspaper accounts, based on later visits by health officials, that the barefoot doctor concept “may be in the process of being downgraded”. According to those accounts, Chinese medical officials no longer praised the barefoot doctors, and many peasants insisted on going straight to a county hospital. In 1992 two Chinese health-policy researchers, Liu Xingzhu and Cao Huaijie, wrote that the rural reform of the early 1980s had destroyed the collective economy on which the scheme rested. By their estimate its coverage fell from about nine rural Chinese in ten at the end of the 1970s to about one in twenty.
1985In 1985 the title of barefoot doctor was formally abolished. A survey of three counties in 1987 found infant mortality, immunisation and infectious disease worse where the scheme had declined, even as incomes rose. Doctors and patients had generally been satisfied with it, the scheme had been financially sound, and without it the barefoot doctors lost their continuing training.
- Villages with a cooperative medical scheme (Chen, 2001)
- Villages with some form of cooperative insurance (Young, World Bank, 1988)
Cooperative medicine drained from about 9 villages in 10 to about 1 in 20
Each mark is a share that a source states for that year; no annual series exists, so none is drawn, and the dashed lines only join the marks. The two sources count differently and disagree about 1986: 4.8% of villages had a cooperative medical scheme, while the World Bank put some form of cooperative insurance at about 50%. By 1997 the first measure stood at 7%.
Source: Chen, 2001, citing Feng et al., 1995; the World Bank (Young, 1988). Page numbers are in the data table.
See the numbers
| Year | Share of villages | What was counted | Source |
|---|---|---|---|
| 1976 | over 90% | Villages with a cooperative medical scheme | Chen, 2001, p. 467, citing Feng et al., 1995 |
| 1982 | almost 90% | Villages with some form of cooperative insurance | Young, World Bank, 1988, p. 14 |
| 1983–84 | under 25% | Villages with some form of cooperative insurance | Young, World Bank, 1988, p. 14 |
| 1986 | 4.8% | Villages with a cooperative medical scheme | Chen, 2001, p. 467, citing Feng et al., 1995 |
| 1986 | about 50% | Villages with some form of cooperative insurance | Young, World Bank, 1988, p. 15 |
| 1997 (not drawn) | 7% | Villages with a cooperative medical scheme | Chen, 2001, p. 467, citing Tang, 1997 |
China’s state-funded rebuilding began in 2003, when a New Rural Cooperative Medical Scheme started as a pilot. Before it, about one rural Chinese in five had any health insurance. The aim was to cover every rural resident by 2010.
Chain C041978 → 2003Planted: the rural scheme is rebuilt in A Job Card194720262005India’s village worker came back by another road. In Chhattisgarh the state government and civil groups began the Mitanin programme in 2002; the health ministry dates it to 2003. In the local dialect, Mitanin means “female friend”. When the National Rural Health Mission was launched in April 2005, it created a national cadre of Accredited Social Health Activists, or ASHAs, with the Mitanin as its prototype. By the accounts of Perry and his co-authors, the design also drew on Jamkhed and on the lessons of 1977.
The ASHA’s job, as the 2006 guidelines set it out, was to counsel women on safe delivery, breastfeeding, immunisation and contraception, to take pregnant women and sick children to the nearest health centre, to treat diarrhoea, fevers and minor injuries, and to keep a stock of oral rehydration salts, iron tablets, delivery kits, pills and condoms. By 2019 there were almost a million ASHAs, all of them women. What set the programme apart from earlier efforts, the health ministry wrote in 2012, was the support around the worker: full-time supervisors and mentors, drug kits, and performance-based incentives.
| Design | Who the worker is | How she is chosen | How she is paid | Who supervises her | How she is supplied |
|---|---|---|---|---|---|
| Jamkhed 1970 A rural health project in Maharashtra | A village woman. At the start many could not read, and many were widows or from the Dalit castes once called untouchable. | Chosen by her community. The source does not say who in the village decided. | A volunteer, taught skills such as running a small enterprise so that she can earn her own living. | She comes to the project's training centre each week. A mobile team of a nurse, a paramedic and a social worker visits each village and helps solve problems. | — The source describes no supply system. It says only that the workers learned herbal medicines, home remedies and homemade oral rehydration. |
| India's village health guides 1977 A national scheme, first called the Community Health Worker scheme | A community member trained for three months at the local primary health centre. The aim was one for every 1,000 people, and a woman where possible, but 75 to 94 percent were men. | On paper, her community chose her. In practice a handful of people did: the centre's medical officers and village leaders, and later district officials. | A small honorarium: Rs 200 during training, then Rs 50 a month. The committee that designed the scheme had advised no pay. | On paper, her community, with help from the primary health centre. In practice the centre, which was poorly placed to do it. From 1981, village health committees. | On paper, Rs 50 worth of medicines and supplies a month. In 1979 almost half of the guides had no drug kit. |
| Mitanin, Chhattisgarh 2002 A state programme, run with civil-society groups | A woman from the same hamlet and the same economic background as her neighbours. By 2010 there were nearly 60,000, covering almost every rural hamlet. | By her community, in a series of meetings with the village council. Women mobilisers, picked by block coordinators and nurse-midwives, ran the meetings. | A volunteer with no regular pay. Small incentives began with the ASHA programme. | A trainer from her own area supports each cluster of 20 to 25 Mitanins. The design gave the health department a supporting role, not a supervising one. The 20 to 25 figure is from a 2023 description of the programme. The supporting role is from its 2003 plan, as Nandi and Schneider report it. | — No source read for this figure says what the 2002 design supplied. By 2011-12 every Mitanin (60,092) had a drug kit, refilled each year. |
| ASHA, India 2005 A national cadre under the National Rural Health Mission | A woman living in the village, married, widowed or divorced, preferably aged 25 to 45 and literate to Class 8, though that can be relaxed. One for every 1,000 people. | Facilitators hold village discussions and shortlist at least three names. The village assembly (Gram Sabha) then picks one. | An honorary volunteer: no salary or honorarium. She can be compensated for training, for meetings and for set tasks, and group awards are encouraged. | The auxiliary nurse-midwife meets her every week or fortnight and guides her. The Anganwadi worker guides her too, and block facilitators can give field support. | A drug kit for every ASHA. The Anganwadi worker holds the kits and can replace what she uses. |
The same village health worker, designed four ways
Each cell says how the design set up the worker's job, as far as a source says; 'in practice' marks what a source reports happened. India's health ministry wrote in 2012 that what set the ASHA programme apart from earlier efforts was the support around the worker: supervisors and mentors, drug kits and incentives linked to performance. A dash means no source read for this figure answers the question.
Source: Perry and Rohde, 2019; Strodel and Perry, 2019; Nandi and Schneider, 2014; Garg et al., 2023; Health ministry, ASHA guidelines, 2005; NHSRC, 2012
In outline, then: a directive in 1965 sent health workers into the villages, and the collective paid them. The healthier generation they helped raise was already in place when the factories opened after 1978, and the reform that opened the factories emptied the purse. India built an elite technical institute first, and reached a design for its village worker, with supervisors, drug kits and incentive pay, only in 2005.
China poured its basement first, by command and through a famine. India put up its top floor first, in a former prison camp, and laid one more thing in its foundations: a vote for every adult. The next chapter puts that vote to its first hard test.