Skip to main content
For families who want their stories to last. learn about the founding circle.

Culture · Essay

The jiapu: the longest family records anyone has ever kept

Chinese lineages have kept written genealogies for two thousand years, some running to eighty generations and thousands of pages. On what goes in a jiapu, what was deliberately left out, and how many were burned in a single decade.

Written to last.

By Confinity Editorial · 2026-08-21 · 6-minute read
Quiet tools, kept out of the way.
The deepest continuously documented family line on earth belongs to a Chinese teacher who died in 479 BC. The descendants of Confucius have been recorded, generation by generation, for roughly two and a half millennia. The main line is traced past the eightieth generation, and the most recent full revision of the genealogy, completed in this century, added around two million names and went looking for branches in Korea and among the diaspora. Whatever you think about the accuracy of the earliest centuries, and scholars argue about it, the fact of the document is remarkable on its own. Somebody in every generation for eighty generations decided that the record was worth the work. A jiapu is a lineage's own book, compiled by the lineage, held by the lineage, and revised on a schedule. It is not a state record. Chinese officialdom kept household registers for taxation and conscription, and those are a different thing with a different purpose. The jiapu belongs to the family, was traditionally stored in the ancestral hall alongside the spirit tablets, and answered to the family's needs.
A page from the genealogy of the Ma family, a Chinese lineage book
A full one is a substantial book, sometimes many volumes, and contains far more than a tree: The descent chart itself, generation by generation. A preface explaining where the lineage came from and why the current edition was made. The generational poem, of which more below. Biographies of members who did something worth recording. The rules of the lineage, on conduct, on marriage, on how the common property is managed. Maps and records of the graves, so they can be found and swept at Qingming. Records of the ancestral hall and its endowment. Sometimes the whole thing opens with a claim of descent from somebody famous, which is where the scepticism belongs. Revision happened roughly every thirty years, by a committee, funded by the lineage, and printing a new edition was itself a major undertaking that marked the lineage's standing. This is the detail that gives the system away as a designed thing rather than an accumulated one. Many lineages fixed a poem, of perhaps twenty or forty characters, in advance. Every member of the first generation takes the first character of the poem as part of their given name; the next generation takes the second character; and so on down. So a person's name states their generation. Two strangers with the same surname can establish, in one exchange, which of them is senior to the other, without either knowing a single shared ancestor by name. Across a diaspora, across centuries, with no central register, the naming system itself carries the position. The Confucius lineage's poem was granted by emperors and extended when it ran short. That is a family planning its naming scheme several hundred years ahead. A jiapu is not a neutral record and reading one requires knowing what it excludes. Women appear, in the main, as surnames. A wife is recorded as "of the Wang family", often without a personal name; daughters frequently do not appear at all, because they will marry into another lineage and be recorded there, if anywhere. The book follows the male line and is explicit that this is its subject. Adoptions in and out are recorded, but so is the reasoning, and lines that failed are marked as failed. Members who disgraced the lineage could be struck out. The book is an argument about the lineage's standing as much as a record of who was born, and the two purposes pull against each other in ways a modern genealogist has to allow for. There is something to admire even in the omissions. The document knows what it is doing. Compared with a modern family's photograph library, which excludes just as much and has no stated principle at all, the jiapu is at least honest about its own shape. For a system built to last centuries, the twentieth century was very nearly fatal. Lineage organisations were dismantled as feudal remnants after 1949, ancestral halls were repurposed or destroyed, and the Cultural Revolution treated genealogies as one of the Four Olds. Jiapu were burned in large numbers, in some regions systematically. Families hid them in walls, under floors and in coffins, and a great many did not survive. What has happened since is one of the more striking reversals in modern cultural history. From the 1980s lineages began reconstituting their books, often working from a single surviving copy in a distant branch, from an elderly member's memory, or from copies that had left the country with emigrants and were sitting in a library in Salt Lake City or Taipei. Shanghai Library holds one of the largest collections in the world and it is consulted constantly. The rebuilding is patchy and some of it is invention. It is also enormous. If you take one thing from this, take the schedule. The jiapu was not a document somebody wrote once. It was a document a family committed to maintaining, on a known cycle, by a named group of people, funded out of common property, with a physical place to keep it. Every element of that is a design decision, and every element is missing from how most families handle their records now. We have better tools than woodblock printing and no equivalent of the committee. The record gets made when somebody dies and there is a scramble, which is the worst possible moment and produces the worst possible record. The griots solved the same problem with a person instead of a book. The Zoroastrian priest holding the Muktad list solved it with an institution. The common element is never the medium. It is that somebody was responsible, on a schedule, before anyone died. A thirty-year revision cycle is optional. Having somewhere for the book to live, and a reason to open it more than once, is not, and that is the whole of what Confinity does.
What we preserve here

An essay is a smaller archive, a way of saying something is worth keeping in words. Confinity is the larger version of the same instinct.

More essays

All essays
Heritage library
Start your own archive