Chinese Reign-Era Converter

Reading a date written in Chinese reign years

For two thousand years Chinese documents did not count years from a fixed point. They counted from the start of an era — a name the emperor chose, such as 康熙 or 光緒 — so 光緒二十三年 means the twenty-third year of the Guangxu era. The month and day that follow are lunar: 五月初五 is the fifth day of the fifth lunar month, not 5 May. Turning such a date into a Western one therefore needs two things, the year the era began and the exact lengths of every lunar month since, because Chinese months ran 29 or 30 days by observation and calculation rather than by a fixed rule.

Where the months come from

This page answers from the Chinese calendar of the DDBC Time Authority Database, compiled by the library of Dharma Drum Buddhist College in Taiwan between 2008 and 2010 and released in 2012. It records every lunar month of every era from the Qin to the end of the Qing: its number, whether it was a leap month, how many days it had, and the Western day it began on. The converter does no astronomy of its own; it looks the month up and counts days.

One day, several eras

China was often divided, and rival courts kept their own calendars. On 1 January 1000 the Northern Song and the Liao were each in a different era, and the converter lists both. In 1644 the Ming and the Qing overlap: the day the Chongzhen emperor died is 崇禎十七年三月十九日 in the Ming record and 順治元年三月十九日 in the Qing one. Two dynasties could also share an era name — 太安 appears under five regimes in the record, one of them the Former Liang keeping the Western Jin's era — and a few names were proclaimed twice by the same dynasty: the Tang used 上元 in 674 and again in 760. An era name alone is ambiguous until you choose the era, and the page shows each one.

Dates kept after the reign

People did not always stop using an era when it ended. After the Chongzhen emperor died in 1644, the Southern Ming court at Nanjing went on dating officially by 崇禎十七年 until its own era began the next New Year; Ming loyalists wrote 崇禎 years long after; and the Former Liang counted years of a Western Jin era, 建興, for almost half a century after that dynasty fell. The database keeps such records apart and calls them proleptic, and for some eras it extends them a long way — its 崇禎 count runs to the 1890s — so that any date a source might write can be placed. The converter answers from the reign itself first and lists proleptic readings separately, labelled "outside the reign": official for Nanjing in 1644, real for a loyalist's letter, hypothetical for most others.

When an era began part-way through a year

The database starts a new era on the day it was decreed: 大曆 begins on 18 December 766, and the earlier months of that lunar year are filed under 永泰二年. Academia Sinica's converter works the same way. That is a way of coding the record, not a claim that the first year lasted two months. The histories often renamed the whole year — the Old Book of Tang records the decree as 「改永泰二年為大歷元年」, changing the second year of 永泰 into the first of 大曆 — and the Book of the Later Han dates the Yellow Turban rising to 中平元年二月, although 中平 was decreed in the twelfth month. If a month of an era's first year is not found, the converter says so and names the era the database files it under.

Singapore, Malaysia and the 民國 years

The same system dates much of the Chinese heritage of Southeast Asia. Older Chinese graves in Singapore and Malaysia — Bukit Brown is the best-known example — commonly carry a Qing reign year such as 光緒 or 宣統, and many from after 1912 carry a Republic year instead. Republic years follow a simple rule, year N being 1911 + N, on the Gregorian calendar, so the converter computes them rather than looking them up. A 民國 date written with a lunar month is a different matter: it needs that year's lunar calendar, which the lunar–solar date converter has. A 民國 date written with lunar labels — 正月, 臘月, a 閏 month, a day such as 初五 — is refused here rather than read as a Gregorian date.

What to trust, and what to check

Western dates before 15 October 1582 are given in the Julian calendar, as historians write them; in 1582 the Gregorian date was ten days ahead, and the gap shrinks the further back you go. The database's own notes warn that its research into exactly when each reign began and ended is incomplete, so the first and last days of a reign can be off; the month lengths inside a reign are the reliable part. Some smaller regimes are not included, and the data stops with the Qing. We corrected one record: the release labels the last nine days of the tenth month of 光緒三十四年 (15–23 November 1908) as the twelfth month, and Academia Sinica's converter gives 光緒三十四年十月廿二日 as 15 November 1908, as this page now does.