Go to site main page,
China Resources main page.

Content created: 2012-06-08
File last modified:

Origin and Migrations of the Hakkas


The Hakka. The Hakka people (Mandarin: Kèjiā rén 客家人, literally “people from the guest families”) are an important minority “ethnicity” in southern China, speaking a distinctive language, and living in distinctive but discontinuous communities in Guǎngdōng 广东 Province and, to a lesser extent, Fújiàn 福建 and Táiwān 台湾 Provinces and occasionally elsewhere. (At the time this paper was published, Táiwān was part of Japan, and the author refers to it as “Formosa.”)

The populations around them are referred to in English as “Punti” people (Mandarin: Běndì rén 本地人, literally “original place people” or, less literally, “natives”) implying that they were there first and that the Hakka are intrusive. (This is the usage even in Hakka communities into which the “Punti” are known to have moved later, and it is accurate to note that the “guest” business is usually out of focus except for the etymologically finicky.)

In Guǎngdōng most Hakka are surrounded by speakers of the Cantonese family of Chinese dialects, and are usually bilingual in Cantonese. In Fújiàn and Táiwān, most Hakka are surrounded by speakers of the Hokkien family of Chinese dialects (variously named Hokkien, Hoklo, Holo, Mǐn, Mǐnnán, Amoy &c.) and are usually bilingual in Hokkien. Today, virtually all people in China (excluding Hong Kong but including Taiwan) also speak Mandarin.

The Hakka-Punti linguistic boundary is a critical part of Hakka identity, of course, and helps to sustain a strong sense of cultural difference, which is also represented by minor differences in dress, preferred divinities, the layout of home altars, and so on. And, like almost any other ethnic boundary in the world, Hakka vs Punti identity has a rich history as a fault line for local-level feuding, fussing, and fighting.

The “Hakka Origins Problem.” Linked to the term kè , or “guest,” Hakka have traditions of having migrated from somewhere else, probably somewhere to the north. (The Hakka language is sometimes described as “somewhere between” northern Mandarin and southern Cantonese.) Not surprisingly, this has given birth to silly ideas about the Hakka being the only “pure” Chinese or the Hakka being “half-breed mixes of Chinese and savage southern tribes,” and such notions have long been discredited. But for the historian the term “guest,” the language difference, and the traditions of migration in the remote past combine to suggest that the original differentiation of this people and the history of its probable peregrinations can be a matter of some interest. Unfortunately, whether one views the data as rich or as scant, it has proven difficult to achieve perfect clarity about Hakka origins.

The following paper, derived from archival research by a Hakka Ph.D. student nearly a century ago, still stands as a useful introduction to the Hakka story, the kinds of factors that have influenced movements of Hakka populations, and the kinds of data that must be taken account of in developing a comprehensive Hakka history. The author is what today might be called a “Greater Hakka Chauvinist,” and is quite happy to make sweeping generalizations about Hakka virtues. More disturbingly, he has not yet divorced the concept of race from that of culture, and he is untroubled by seeing all early non-Hàn south Chinese peoples as savages and barbarians. However, the paper remains useful, and it has passed into the public domain for free use by anyone.

Web Presentation. The original paper included Chinese characters for many terms, and this has made it possible to create an updated version for modern readers. Throughout this text, I have substituted Pinyin spellings with tone marks for the outdated original spellings. Obvious proof errors have been corrected (although I may accidentally have introduced some new ones), and occasional words have been inserted in brackets where clarification seemed desirable. A few words have been replaced to conform to modern usage (For example, the name Canton has been changed to Guǎngzhōu throughout.)

To facilitate on-line reading, brief footnotes —usually bibliographic references— have been inserted into the text. Longer ones have been set off near the points where they are cited. In both cases, they are printed in purple. I have also broken the long paragraphs of the original text into shorter ones and have inserted numbered subtitles.

Because of the large number of place names, I have placed a very simple map of provinces at several points in the text. There is room on it only for abbreviated province names, but clicking on it will open a second page where they are spelt out. More detailed maps provided in the original article could not be successfully adapted for web viewing.

Chinese characters in the main text and footnotes conform to the current “simplified” standard (here printed in red). Those in the bibliography remain in the “traditional” character set in use when the paper was published (here printed in blue). The bibliography and associated citations do not meet modern citation standards, but I have not made modifications, since doing so would require adding information that is not easily available.


HSIEH T’ing-yu
1929 Origin and migrations of the Hakkas. The Chinese Social & Political Science Review (Běijīng), vol 13, pp. 202-227

Origin and Migrations of the Hakkas

By Xiè Tíngyù (“Hsieh T’ing-yü”) ( 谢廷玉), B. A.

Page Outline

  1. Introduction: Who Are the Hakka?
  2. Hakka Locations: Guǎngdōng
  3. Hakka Locations: Outside Guǎngdōng
  4. Hakka Glory
  5. Hakka Daily Life
  6. Hakka Language
  7. Misunderstanding Hakka History
  8. Hakka Origins: List of Theories
  9. Text Sources for Studying Hakka Origins
  10. Historical Records of Hakkas in the North
  11. Punti and Hakka in the Great Migrations
  12. The Hakkas of Jiāyìngzhōu
  13. Leaving Fújiàn and Jiāngxī: The First Wave
  14. Leaving Fújiàn and Jiāngxī: The Second Wave
  15. Why Leave Jiāngxī and Fújiàn? What’s Special About Guǎngdōng?
  16. Theory 1: Mongol Garrison Soldiers
  17. Theory 2: Fújiàn Aborigines
  18. Theory 3: Qín Shǐhuáng’s Soldiers
  19. Theory 4: People Leaving Hénán at the Breakup of the Táng Dynasty
  20. A Complication: Mayflower-Like Claims
  21. The Term “Hakka”
  22. Six Generalizations From the Evidence
  23. The People the Hakkas Displaced: The Kingdom of Yuè
  24. Hakka Spread Within Guǎngdōng Province & Beyond
  25. Hakkas at War
  26. Hakkas Defeated
  27. Hakkas Eternally Glorious
  28. Appendix: Table of Events
  29. Bibliography

Author’s Note: This is a dissertation to fulfill the requirements for the degree of bachelor of arts from Yenching University (Yānjīng Dàxué 燕京大学). The work was begun a year ago [1928] when the writer, descended from a Hakka family that had migrated from Wàizhōu 外州 [?] [in Húnán 湖南 ] to Xiāngshān 香山 or Zhōngshān 中山 district in Guǎngdōng 广东 three generations ago, was advised to use it as his thesis. Since the results cannot claim to be conclusive or final, he will be very glad to accept corrections, additional information, or suggestions for improvement. The writer desires to express his indebtedness to all his friends, particularly Professor William Hung, who had first suggested the subject and given much valuable assistance in the preparation of this paper.

Yenching University, May 1st, 1928.

1. Introduction: Who Are the Hakka?

The term Hakka or Kèjiā 客家 has aroused much interest as to the origin and history of this people. Their name alone implies that they are “strangers” or “guests,” and not the original inhabitants of the regions which they now occupy.

It is difficult to determine whether only those shall be counted as Hakkas who call themselves such, and who in every respect bear the characteristics of Hakkas, or whether included within them should be added those groups whom it has been customary so to classify; as for example, some people of the North River 北江, who renounce the name as one degrading to their dignity but whose speech is closely related to Hakka, in spite of a few variations.


Outside of Guǎngdōng 广东 it is more difficult to demarcate the regions occupied by them. For example, in Southern Jiāngxī 江西 the colloquial [speech] or tǔtán 土谈, which we readily identify with the Hakka, becomes less intelligible than the Mandarin or Guānhuà 官话 as we go northward.

For practical purposes it may be said that the Hakkas are a distinct group of people found in parts of Guǎngdōng 广东, Guǎngxī 广西, Fújiàn 福建, Jiāngxī 江西, Húnán 湖南, Sìchuān 四川, Formosa (Táiwān 台湾), and overseas, who speak a [group of] closely related dialect[s], and whose characteristics and customs are very similar. Their main point of concentration is in the province of Guǎngdōng.

2. Hakka Locations: Guǎngdōng

In certain districts there they monopolize the whole countryside, as in the former prefecture of Jiāyìngzhōu 嘉应州.* In other parts of the province they form a half, a third or less of the population, being interspersed among the Puntis (Běndì 本地)** and Hoklos (Mandarin: Fúlǎo 福佬)***. In some places partly populated by them, they have settled on the higher lands, and so have been called “Chinese Highlanders;” but this is a misnomer only capable of local application, since in other places, they are spread over the plains as well as the hilly districts. (Ball, Hakka Made Easy. Introduction.)

*-Also known as Méizhōu 梅州. It consists of the present districts of Méi Xiàn 梅县, Xīngníng 兴宁, Wǔhuá 五华, Píngyuǎn 平远, and Jiāolíng 蕉岭 or Zhènpíng 镇平.

**-Older inhabitants of Guǎngdōng, who consider themselves the rightful owners of the soil. They form over half of the total population.

***-Also known as Haklos (Mandarin: Xuélǎo 学佬), a third group of people who settled along the seacoast from the border of Fújiàn down to Hongkong 香港. They are said to be descendants of emigrants from Fújiàn.

3. Hakka Locations: Outside Guǎngdōng

The Hakkas are not only confined to the province of Guǎngdōng where they form less than one third of the population.* There are large numbers of them living in Fújiàn, especially in Zhāngzhōu 漳州, and also in Tīngzhōu 汀州, the old Fújiàn 福建 home of the Hakkas. They occupy Southern Jiāngxī 江西and parts of Guǎngxī 广西. There are scattered communities of them in Sìchuān 四川, Húnán 湖南, and Formosa (Táiwān 台湾).**

*-The estimates vary greatly. More than thirty xiàn [counties] were represented in the mass meeting of Hakkas at Guǎngzhōu 广州 in 1921; the total number of Hakka constituencies could at least be twice as much.

**-There are probably 10,000,000 Hakkas in Guǎngdōng, 1,000,000 in Guǎngxī, 1,000,000 in Jiāngxī, 2,000,000 in Fújiàn, 200,000 in Húnán, 50,000 in Sìchuān , and 250,000 in Formosa, making a total of 15,000,000 in China.

Their emigrants abroad are second to the Puntis from Guǎngdōng, [for] they have settled in Japan, the Philippines, Siam, Annam, the Malay Peninsula, British North Borneo, Batavia, Ceylon, Sabang, Natal, the Transvaal, the West Indies, Cuba, California, Mexico, South America, Hawaii, and Australia. Thus a line of Hakkas stretches around the world.

Return to top, outline

4. Hakka Glory

The character of the Hakkas is shown quite clearly in their name and history. They are a strong, hardy, energetic, fearless race with simple habits but a very contentious and litigious disposition. Self-reliant and active, their rapid expansion, and fondness of property have often brought them into conflict with their neighbors.

Jiāyìngzhōu 嘉应州, the stronghold of the Hakkas, is the famous abode of scholars. It furnished a large number of licentiates and yamen officials for the past dynasties. Five out of the eleven successful candidates for the imperial jìnshì 进士 degree in 1752 came from Jiāyìngzhōu. See Jiāyìngzhōu Annals 20/6. Even the poorest want to give their children a chance for education.

Today Méi Xiàn 梅县 [Jiāyìngzhōu and its environs] is dotted with over six hundred schools and claims the highest percentage of literacy in the province. Huntington estimates the percentage of male literacy in Méi Xiàn as high as 80%. (Huntington 1924, p. 167.) The Hakkas are proud of the literary accomplishments of their ancestors; they claim many well known literati.

The Hakkas are a “people of the future,” unhampered by the prejudices or the easy-going slackness of the old landowners who are proud of their riches and of their fancied superiority. (Stauffer 1922, p. 352.)

Their village is a natural expression of the character of the inhabitants. The houses are clean and well-constructed. Foreigners visiting Méi Xiàn are surprised at the cleanliness and general appearance of the place. “It would not be easy to find an inland district where the people are as well housed as they are in Méizhōu.” (Huntington, op. cit., p, 167.)

5. Hakka Daily Life

The people live in scattered hamlets or houses located, preferably in the valleys, where the farmer can live surrounded by his fields. In the villages the elders of the clan make all important decisions.

In other places they are dispersed in small villages or hamlets between the hills and paying ground rent to the Puntis, or congregated in larger villages and then continually fighting with the Puntis for the ownership of the hills and fields occupied by them. (Eitel, Ethno. Sketches of Hakkas, p. 265.) Being independent and thrifty, little by little they move out of their hilly tracts and displace the coastal people to the south and east of them. This is why so many Puntis have feared and reviled them.

Fundamentally the Hakka is a farmer, forced by poverty to struggle with the unproductive soil and wresting a bare livelihood therefrom. “Zhènpíng 镇平 (one of the four districts in Jiāyìngzhōu) has not less than 300,000 families and its annual produce can suffice for only three months...” (Jiāyìngzhōu Annals, 32/16.) They usually occupy the hilly and less fertile districts, while the Puntis remain in possession of the fertile deltas and plains and the Hoklos inhabit the coastal regions.

The peasant class is very poor indeed; only a small percentage of them own their fields and they are often involved in debt. Hence many vocations despised by the Puntis, are taken up by the Hakkas, who become barbers, itinerant blacksmiths and stone-masons. Some of them are traders, particularly those from Xīngníng 兴宁, the richest industrial district of the Hakkas, which at one time produced enough paper fans to supply half of China. Japan is now handling this trade and the weaving of cotton stuff has taken the place of fan making. (Stauffer, op. cit., p. 352.)

There are few industries producing articles of luxury in these poor Guǎngdōng districts. Such are the conditions which have made the inhabitants depend more upon their learning than upon their fields to get a living since the Sòng dynasty [960-1279] (Jiāyìngzhōu Annals, 8/2.) , and migrate to other more fertile districts or seek their fortunes overseas during the past [Qīng ] dynasty [1644-1911].

There is not much to choose between the dress and customs of the Hakkas and those of the Puntis. If there are differences at all, they are due to geographical rather than racial [i.e., cultural] reasons.

“The food of the people is, on the average, inferior in quality to that of the Punti, but better than in most parts of North China. Rice forms the basis, but it is often ‘stretched’ to yield quantity by the addition of sweet potato. Other additional dishes are very frugal.” (Stauffer, op. cit., p. 352.) However, this description applies to the poorer and more mountainous districts only.

The sexes are not so strictly separated in domestic life as in the case with some of the other Chinese. The women folk are strong and energetic, and have never adopted the foot-binding custom. Alongside with their domestic duties they go out to till the fields, gather grass for fuel, and feed the pigs. “There are no women who are-so industrious as those of the Hakkas.” (Jiāyìngzhōu Annals, 8/53.) Their girls are rarely sold as slaves or concubines ; polygamy is not common mainly because of the poverty of the people. As it is written in the “Jiāyìngzhōu Annals”:

The soil is not fertile and the people are poor. There are lots of hills but few fields for cultivation. The men hence want to emigrate and leave the management of the home to their women. As soon as the ban on emigration overseas was lifted, many of the inhabitants flocked to Nanyang 南洋.* They began as menial workers but gradually amassed a fortune. Some returned in three or five years ‘while others delayed till ten or twenty years later. Some even left their homes in their youth and returned with hoary hairs. When they departed from their homes, they left their old folks and young children, the fields, ancestral graves, and houses in charge of their wives. … This is why the people can migrate abroad. The women in other places bind their feet and have to depend upon servants for help; hence their men have lots of domestic cares and cannot think of leaving for distant places ….
(Jiāyìngzhōu Annals, 8/34.)

*-A general term for southeastern Asia and the neighbouring islands where numerous emigrants from Guǎngdōng and Fújiàn have settled.

The Hakkas place great emphasis on ancestral worship ; this is characteristic of the Chinese people as a whole. The ancestral halls and tombs are carefully preserved. In the village, no matter how small, each clan has its family temple 祠堂. The Hakkas are great believers in “fēng shuǐ” 风水.

In their worship at the temple, Guānyīn 观音, is the most popular. Special temples are set aside for the God of War and for the Patron of Literature. Confucius and his disciples are revered In the schools and private homes. Their religious worship, on the whole, is closely similar to that of the Puntis. However, in some districts where the Puntis predominate, the Hakkas are considered as intruders and have no share in the worship of local idols and ancestral halls. The German missionaries have been very active in the Hakka districts.

Return to top, outline

6. Hakka Language

Many authorities agree that “the Hakka dialect is not a mere local growth, nor a patois of some one of the other dialects,. but an independent growth of the common Chinese language, and, so to speak, the crystallized relic of one of the different phases through which the language passed in developing from Punti, which is the oldest relic of the original form of the Chinese language, to Mandarin, which represents the latest phase in the process of development through which the Chinese language passed.” (Eitel, Ethno. Sketches of Hakkas. p. 265.)

It still “retains some of the old finals for which feature Cantonese is also noted, while in the more modern Mandarin and some of the languages of China they are nearly gone, the detritions of ages of use having denuded many of the words of these relics of antiquity.” (Ball, op. cit.)

It is less polite than the Mandarin and has clearer sounds than the Punti. Prof E. H. Parker characterizes it as having “to a large extent the vowels of Běijīng the diphthongs of Fúzhōu 福州, the finals of Guǎngzhōu, and the tones of Hànkǒu 汉口.” (Ibid)

In common with other dialects of China, it contains within its limits not a few variations. At first sight the variations do not seem so numerous as in some of the Chinese speeches; in fact, however, there are enough colloquial differences to cause at least a perfect understanding between the people concerned rather difficult.

7. Misunderstanding Hakka History

Concerning the origin and history of the Hakka people, there has been much controversy and misunderstanding. The” Encyclopedia of Missions” is bold enough to state that “the Hakkas are a peculiar race or tribe, inhabiting the mountains near Guǎngzhōu or Shàntóu 汕头, who are of a lower social rank than the native Chinese. Their language is written with the Chinese characters.” (Quoted in Campbell, Origin and Mig. of Hakkas, p. 474) The “International Encyclopedia” adds that for two thousand years the Hakkas have been “the object of the most persistent and inveterate hostility and persecution at the hands of the native or punti part of the population of China.” (International Encyclopedia, VIII 943.)

The traditional antipathy of a section of the Cantonese is probably the cause of this widespread impression that the Hakkas are a mongrel race more civilized than the aborigines but hardly entitled to rank with the Chinese.

Popular interest was aroused by the controversy over the statement made in Roger D. Wolcott’s “Geography of the World,” printed by. the Commercial Press, Limited, Shanghai, that the Hakkas are a backward people: “In the mountains are many wild tribes and backward people, such as Hakkas and Ikias …” (p. 132 of April edition, 1920. Quoted from Jiāyìng Magazine, 2/103)

Indignant protests came pouring in from Hakka organizations in China and abroad. On April 3rd, 1921, a mass meeting of the Hakkas was called at Guǎngzhōu. Over a thousand delegates representing thirty odd xiàn from the five provinces —Fújiàn 福建, Guǎngdōng 广东, Jiāngxī 江西, Húnán 湖南, Guǎngxī 广西— elected Ráo Fúcháng (“Yao Fu-shang”) 饶芙裳, former member of the House of Representatives, as presiding officer. Resolutions were passed to take necessary measures to make the Commercial Press retract this statement. The minutes and resolutions of this meeting have been printed in a pamphlet called 中华客语攷原总会章程.

As a result, the fourth revised edition in May of the same year came out with the following changes: “… a part of the population of Guǎngdōng is made up of Hakkas ( 客家), whose ancestors migrated southward on account of disorder and lawlessness in other provinces during the ‘Five Dynasties’ ( 五代). The language and customs of the Hakkas partly coincide with those of the ancient people of northern China.”*

*Wolcott, Geography of the world, p. 144. The writer wrote a letter on March 18, 1928 to Mr. Fong, F. Sec of the Commercial Press, English Editorial Department, to ask for detailed information concerning the matter. The following reply was written on April 3rd:

“… some years ago some of the Hakka people in Shanghai called on us to protest against some passages that Mr Wolcott made concerning their people. We referred the matter to Mr. Wolcott. who afterwards modified his references to them. The change was satisfactory to the Hakka people and the matter was closed … .

Return to top, outline

8. Hakka Origins: List of Theories

There are at least five theories which attempt to explain the origin and history of the Hakka people:

(1) The Hakkas are descendants of the Mongol garrison soldiers. (Eitel, Ethno. Sketches of Hakkas, p. 265.)

(2) The Hakkas are aborigines from Fújiàn who had been assimilated by the Chinese. (Stauffer, op. cit., p. 351.)

(3) The Hakkas are descendants of the half million soldiers sent by Qín Shǐhuáng 秦始皇帝, many of whom intermarried with Yíjiā 夷家 women. (Mesney, Chinese Miscellany, II, 475.)

(4) The Hakkas are descended from the remnants of the Kingdom of Yuè which was destroyed by [the state of ] Chǔ in 333 B.C. (Bái , Mínguó Dìzhì 民国地志. 31/4/43.)

(5) The Hakkas are descendants of Chinese coming from the Northern section of China following the Jìn and Táng dynasties.” (Campbell, op. cit., p. 476.)

9. Text Sources for Studying Hakka Origins

Thus we are confronted with a difficult problem that requires careful handling and weighing of historical evidences to get the facts. Much information for this paper is obtained from the provincial [gazetteers or annals] 通志, prefectural [annals] 府志, and district annals 县志 of which the “Jiāyìngzhōu Annals” 嘉应州志 is especially valuable. These are voluminous works and require much patience and time to sift the mass of materials.

The geographical works of the various periods, such as the “Yuánhé Jùnxiàn Túzhì” 元和郡县图志 of [the] Táng [dynasty], and the “Tàipíng Huányǔ Jì” 太平寰宇记 and the “Yǔdì Jìshèng” 与地纪胜 of [the] Sòng [dynasty], prove to be very useful sources. The dynastic histories of Jìn , Táng, Wǔdài 五代, and Sòng [periods] are also very, very valuable.

The family genealogies are useful but rather inaccurate and unreliable at times (The writer has been able to gather only three family registers. He will appreciate it very much if he can have access to more of these materials.

The works by foreign sinologists are helpful in giving the lead to the Chinese sources; they are better classified but often seem to mix traditions with facts and fail to mention the exact sources. Aside from the examination into the history of the Hakkas, a study of their customs and language reveals a great deal of important information concerning their relations with the past.

Return to top, outline

10. Historical Records of Hakkas in the North


How far back into the past can we trace the history of the Hakkas? Their traditions and family records place them mostly in western Shāndōng 山东 Province at the end of the Zhōu dynasty. There were a few living on the southeastern borders of Shānxī 山西, while some occupied the northwestern frontiers of Ānhuī 安徽 Province.

The claim that they were living in the border regions south and southwest of Shāndōng is confirmed by popular ballads among the Hakkas with allusions to localities situated in these regions. (Eitel, History of Hakkas, p, 160. The writer is unable to find evidence to check this statement.)

Qín Shǐhuáng 秦始皇 unified China and proclaimed himself Emperor in 246 B.C. Under his despotic rule many persecutions and wholesale removal of populations took place. (Shǐjì 史记, Vol. 6.)

The First Migration. The Hakka clans were said to be among the unfortunate people persecuted. So merciless was the blow that some clans became extinct while some altered their names to avoid recognition.* The remnants fled to the mountains of Hénán 河南 and Ānhuī 安徽, while a few went as far south as the borders of Jiāngxī. Here they settled and more prosperous times followed. In the Hàn and Jìn dynasties, many gained high offices in the government.

*-The alteration of names to escape persecution was not an uncommon occurrence in the past history. For example, the remaining members of the family of Hán Xìn 韩信, who was punished by Hàn [dynasty emperor] Gāozǔ 汉高祖 (206-194 B.C.), changed their surname to Wéi and fled to Sìchuān for safety.

The Second Migration. The second migration occurred in the early part of the fourth century. It was a period of successive invasions by the barbarians in the north. Jìn [dynasty emperor] Huái Dì 晋怀帝 was captured by the invader Liú Zǒng 刘总 in 313 and was compelled to wait on him in a menial capacity. He was finally put to death and his successor, Mǐn Dì 愍帝, also met the same fate.* These insults and humiliations broke the spirit of the people. When the founder of the Eastern Jìn 东晋 dynasty, Yuán Dì 元帝 moved his capital to Nánjīng 南京, many left their homes and took their families across the Dàjiāng 大江 (another name for the Yángzǐ Jiāng 扬子江 or “Yangtze River”). This was indeed a radical and desperate venture for these people to cross the “Great River” in to a new and sparsely settled region.**

*-Several writers have mistaken Liú Yuān 刘渊 for Liú Zǒng 刘总. See Jìn Shū 晋书 Vol. 5.

**-Campbell compares the importance of this occasion with the crossing of the Atlantic by the Pilgrims in 1620. op. cit., p. 476.

Seeking refuge from the ravages of the incoming barbarians, they migrated southward in search of a permanent home where they could live in peace. It was a general stampede — even those few clans remaining in Shāndōng before now fled to south Hénán 河南. Most of the Hakka family records mention further shifting and renewed migration during this period. (Eitel, History of Hakkas, p. 161.) The emigrants settled mostly in Jiāngxī. Some went to Zhèjiāng 浙江 and on to Fújiàn.* Here they lived in comparative peace and prosperity.

*-Were these Fújiàn newcomers the forefathers of the present Fukienese? A large number of the refugees were scholars and officials. Their coming influenced the culture and refinement of the raw regions.

“In the second year of Yǒngjiā 永嘉 (308 A.D.), the intelligentsia from Zhōngyuán 中元 [“Central Plain”] ( a geographical term applied to the ancient seat of Chinese civilization, in the region now occupied by Hénán 河南, Shǎanxī 陕西, Shānxī 山西, Zhílì 直隶 [modern Héběi 河北], Shāndōng 山东) coming to Fújiàn numbered eight clans. Because of the many troubles in their home regions, they had no desire of returning there.”

See Fújiàn Annals 55/1.

The Third Migration. At the end of the Táng dynasty when the barbarians again descended and ravaged the northern provinces, many refugees fled southward. Even those who had been living in Jiāngxī were compelled to move again. A separation then took place; the majority sought refuge in the mountains of Fújiàn while a few reached the high mountain chains which separate Jiāngxī and Guǎngdōng. (Eitel, History of the Hakkas, p. 162.)

The Fourth Migration. The fourth and last move of migration from the north occurred in the Sòng dynasty. The incoming Tartars pushed the Chinese southward until the [Sòng] Emperor Gāozōng 高宗 crossed the Yangtze and established the Southern Sòng 南宋 dynasty in 1127. Large numbers of loyal adherents followed him and settled in Jiāngxī and Fújiàn. Little by little some of them spread to Gànzhōu 赣州 [in southern Jiāngxī] and Tīngzhōu 汀州 [in Fújiàn], occupying the mountain lands and isolating themselves. Till the end of the Sòng dynasty, very little was heard of these settlements which were far from the highways of travel.

Return to top, outline

11. Punti and Hakka in the Great Migrations

So far we have had only a bird’s-eye view of the migrations to the south of the Chinese in general, not only of the Hakkas. We notice that such migrations occurred almost at the end of every dynasty. The Hakkas were probably a group of related people leaving their ancestral homes of disturbances to seek the Promised Land. Hakka traditions all point to North China as the old home and their family registers support this claim. Punti records, however, also trace back their migrations in a similar manner. Who then, are Hakkas?

We can state that about the time of Sòng, about 960 A.D., there were inhabitants, probably of northern origin, living in southern Jiāngxī and southwestern Fújiàn, especially in Tīngzhōu and Gànzhōu, who later crossed the mountain boundaries into Guǎngdōng and became known as Hakkas.

12. The Hakkas of Jiāyìngzhōu

The history of the occupation of Jiāyìngzhōu by the Hakkas is especially interesting. As early as 900 A.D. there were wandering farm laborers from Fújiàn and Jiāngxī. A census of Méi Xiàn in 976 showed that there were 367 such “alien” and 1210 “native” families. (Tàipíng Huányǔ Jì 太平寰宇记, 160,7.)

A contemporary of the period wrote: “The country is extensive but the people are lazy, depending on wandering farmers from Tīngzhōu and Gànzhōu to till their soil.” (Yúdì Jìshèng 舆地纪胜, 102, 3.)

One hundred years later, in 1078, the census for the place revealed 6,548 “alien” and 5,824 “native” families. It was not until 1049 that a mud wall was built around the city and thirty-five years later a brick wall was constructed. (Ibid. 32/1.)

The influx of newcomers from Fújiàn and Jiāngxī became rapid and gradually displaced the old inhabitants. The few inhabitants at the time of early Sòng disappeared, while only one-tenth or two-tenths of those that came at the end of the dynasty remained. It was in the Yuán , and early Míng dynasties that this region was substantially settled. (Jiāyìngzhōu Annals, 9, 1.)

13. Leaving Fújiàn and Jiāngxī: The First Wave

The first wave of their migration into Guǎngdōng began at the close of the southern Sòng dynasty with Jiāyìngzhōu as the local point. What made the people leave their homes in Fújiàn and Jiāngxī?

First, it was because the mountainous regions could not support the increasing population, while the region across the Guǎngdōng border, although hilly, was sparsely settled and offered more opportunities.


Moreover, numerous bands of robbers and bandits were a constant menace to the people. An official sent in 1171 a memorandum to the Court saying: “Tīngzhōu has the most bandit troubles in Fújiàn. In a period of ten years, we have been forced to take arms against the robbers thrice. … Many people are out of employment and become bandits.” (Fújiàn Annals, 266,15.)

The coming to [Jiāngxī and Fújiàn] of Mongol invaders [ of the Yuán dynasty] caused many to flee for safety. Some served as volunteers in the imperial army and many died with the last prince of the Sòng house in 1279 at Yái Shān 崖山, [in Guǎngdōng] west of Macao. When the Mongols reached Méizhōu in 1279, Cài Méngjí 蔡蒙吉, the local official, was captured and put to death after he had reviled them. (Cài was pupil of Hóu Ānguó 侯安国 who was one of the first scholars to come from Tīngzhōu to Méizhōu.)

A year later Wén Tiānxiáng 文天祥 retook the place and honored Cài. Then with his army enlarged by recruits from the region, he left for Jiāngxī. (There is a tradition that 800 men of the Zhuó clan joined him and only one, [a certain] Zhuó Mǎn 卓满, ever saw his home again. See Jiāyìngzhōu Annals, 32/12.)

When the Mongols re-entered Méizhōu, they devastated the region. A native writer wondered if all the people had turned into foxes and birds. (Campbell, op. cit., p. 479.) The census then revealed only 2,478 families, — no mention of “aliens” and “natives” was made — less than one-fifth of the Sòng statistics two hundred years ago. (Yuánshǐ dìlǐ zhì 元史地理志.) It was reported that only three clans (Yáng , Gǔ , Bǔ ) were left, and many Fújiàn emigrants, especially from the district of Nínghuà 宁化, came streaming in to take up the waste lands.

Return to top, outline

14. Leaving Fújiàn and Jiāngxī: The Second Wave

The Mongols having been driven away by the new Míng dynasty, the second wave of immigrants crossed the mountain borders into Guǎngdōng and settled in Jiāyìngzhōu 嘉应州. The Fújiàn Hakkas came in such overwhelming numbers that they drove everything before them in Jiāyìngzhōu, which has remained their exclusive possession since. About the same time, others came from southern Jiāngxī and settled southwest of Jiāyìngzhōu. The following census reports of the Nínghuà district give a graphic illustration of the Fújiàn migrations: (Nínghuà Annals, 3/48.)

Dynasty Census Date Migrant Families
Northern Sòng 1078 A.D. 13,700 families
Southern Sòng 1253 35,000 families
Beginning of Míng 1391 12,588 families
Middle of Míng 1492 6,565 families
Close of Míng 1613 5,279 families

15. Why Leave Jiāngxī and Fújiàn? What’s Special About Guǎngdōng?

Thus far we have seen that the Hakkas migrated into Guǎngdōng from Jiāngxī and Fújiàn in two large movements — one in the southern Sòng period and the second at the close of the Mongol (Yuán ) dynasty, a [total] period of about three hundred years. But how did they come to Fújiàn and Jiāngxī? Were they the original inhabitants there? If not, when and [from] where did they move to these regions?

We shall now examine the five hypotheses which attempt to answer these questions. Note that there is a difference of over a thousand years and that the theories seem so varied and contradictory. We must resort to a process of logical elimination in their respective order of probability.

16. Theory 1: Mongol Garrison Soldiers

One theory places the Hakkas as descendants of the Mongol garrison soldiers who were stranded in the south following the overthrow of their (Yuán) dynasty. This would place the Hakkas in the same category with those ostracized Mongol clans in Zhèjiāng who surrendered to the Chinese and are now treated as outcasts and slaves.

However, there is very little probability to this answer, because it is too late a period to place the origin of the Hakkas who had lived in Fújiàn and Jiāngxī before the Táng dynasty and migrated to Guǎngdōng before the Mongols came. Moreover, their speech and customs bear few similarities to those of the Mongols.

17. Theory 2: Fújiàn Aborigines

Some claim that the Hakkas are not Chinese but descendants of Fújiàn aborigines.

The coming of this people from a mountainous region causes this doubt and suspicion as to their Chinese origin. But the Hakkas did not come from Fújiàn only; some of them came from parts of Jiāngxī.

And also this argument in counteracted by their similarities of language and customs with those of the other Chinese. There are still some aborigines living in Fújiàn and other parts of the south, but they are generally weak, oppressed, and backward, while the Hakkas have strong characteristics of their virile Chinese strain in their pride of race and history, independence, and fearlessness to migrate and spread out.

Return to top, outline

18. Theory 3: Qín Shǐhuáng’s Soldiers

There are others who believe that the Hakkas are descendants of the five hundred thousand soldiers drafted by Qín Shǐhuáng in 214 B.C. for the conquest of the south. (Shǐjì 史记.) Twelve thousand widows were sent along later. (Liang. See below.) A great number of the military colonists, however, intermarried with the semi-civilized tribes of the region. When [the] Hàn [dynasty emperor] Wǔ Dì 汉武帝 made, a reconquest of Guǎngdōng, he followed his policy of shifting the population to the Jiānghuái 江淮 region [the region between the Yangtze river and the generally parallel Huái river lying to the north of it].* This absorption of peoples and intermingling of cultures produced a “Chinafied” civilization.

*-In 138 B.C. the Mǐn Yuè 闽越 of Fújiàn and the Ōu Yuè 瓯越 of Zhèjiāng came to blows. The latter people were granted permission by the Emperor to move to the Jiānghuái 江淮 region, in the present province of Ānhuī 安徽. In 110 B.C. Hàn Wǔ Dì also moved the Mǐn Yuè to the same region in the north. The regions left empty by this wholesale removal of populations were then occupied by Chinese colonizers from the north. This was a favorite method pursued by successive Chinese conquerors, which has contributed greatly to homogeneity of the present Chinese people. See Liang’s article on the Research of the Chinese Races, Geographical Magazine, Vol. XIII, No. 6-7, p. 17, 18.

The people were more probably the ancestors of the bulk of the present Punti population instead of the Hakkas.* There is no such thing as a pure Chinese stock; the Hakkas cannot claim to be the purest and the “cream of the Chinese people.” (Huntington, op. cit., p. 167.) The first Chinese, after settling in the Huánghé 黄河 [Yellow River] basin, intermarried with the subjugated natives, called the Miáo , or the nomads, thus making the present day Chinese a heterogeneous stock.

*-Ibid. Liáng Qǐchāo 梁启超 thinks that the claim that the Cantonese or Puntis are descendants of immigrants coming into Guǎngdōng through Nánxióng 南雄 at the time of the Sòng dynasty is not convincing. He believes that their blood is a mixture of the Chinese and the Bǎi Yí 摆夷 tribes or Ikias. [Yí is an ancient cover-term for non-Hàn peoples of eastern China, all of whom were eventually assimilated into the Hàn population. No peoples are identifiable as Yí today. —DKJ]

In the Zhōu dynasty, intermarriages between the ruling houses of the Chinese feudal states and the aborigines were not uncommon. In the period of the Northern and Southern dynasties (Nánběi Cháo 南北朝) (420-590 A.D.) various tribes of barbarians came in overwhelming numbers and were assimilated in blood and culture by the Chinese. (Liu, in Chinese Soc. and Pol. Sc. Rev., XI. 492.)

We can stretch our imagination at Piton’s far-fetched suggestion that the Hakkas may be descendants of the remnants of the Róng and Yí tribes in North China at the close of the Zhōu dynasty, which would then place the Hakkas in the same category as the Miáo. (On Origin and History of Hakkas, China Rev. II, 224.)

19. Theory 4: People Leaving Hénán at the Breakup of the Táng Dynasty


Hakka family records and traditions, however, persistently trace to Hénán as their original home and their migrations to the South at the end of the Jìn and Táng dynasties when they fled from the incoming barbarian invaders.

The Hakkas point to their similarity in marriage and burial customs; after centuries of separation from their old home, they are able to preserve their Hénán speech and customs. In each family there is one or more t’ang [hall] names handed down from the past, and in most cases these are from Hénán. For example, the Yeh and Zhāng clans claim the Hénán district of Nányáng 南阳 as their ancestral home.

Like the European settlers in the new American colonies, the Hakka emigrants founded a “New Hénán” wherever they settled and reproduced the conditions and customs of their homeland. From generation to generation they were able to transmit the heritage of their forefathers. Their pride of race is so strong that they are able to preserve their own speech and customs even in a strange country; even those who migrated overseas have not forgotten their ancestral heritage.

Below is a table showing the increase of population (in families) caused by the influx of immigrants to Jiāngxī (Jiāngxī Annals, 47i 1-4.) and Fújiàn. (Fújiàn Annals. 48/12.)

Period A.D. Jiāngxī Fújiàn
Jìn (265-420) 63,000 4,300
Táng (618-907) - 91,186
Sòng (950-1280) 493,500 211,552
Yuán (1280-1368) 320,000 700,817
Míng (1368-1644) 1,553,923 815 527

Conclusion: Thus we see that the three great migrations were after the Jìn, Táng, and Northern Sòng dynasties.

Return to top, outline

20. A Complication: Mayflower-Like Claims

It is a mere coincidence that the Hakkas claim Guāngshān district 光山县 in Guāngzhōu 光州 as their ancestral home while the Fukienese trace back their origin to Gùshǐ district 固始县 in the same prefecture? It is said that if the claims of many people in America that their ancestors on the Mayflower in 1620 were allowed, a [whole] fleet [of] vessels would have been needed to carry the [throngs of] Pilgrims.

Similarly, how many of the Hakkas have legitimate claims to be the descendants of Guāngshān emigrants and how many of the Fukienese actually come from Gùshǐ stock?*

*-Zhèng Qiáo 郑樵,the great Sòng historian, and Liáng Qǐchāo 梁启超 both give little weight to such claims. Moreover, most of the Hakkas of eastern Guǎngdōng claim that their Fújiàn ancestors came from the township of Shíbì 石壁 in Nínghuà district 宁化 of Tīngzhōu 汀州. Why is it that so many clans claim Nínghuà as their former residence when the usual practice has been for one or two clans to stay in one district?

It seems that both of them base their claim on the Wáng brothers (Wáng Cháo 王潮 and Wáng Shěnzhī 王审知) , who fled from Guǎngzhōu 广州 with five thousand soldiers after incurring the displeasure of the Court in 885 A.D. and finally settled as an independent colony in the present region of Tīngzhōu 汀州, Zhāngzhōu 漳州 [in Fújiàn], and Gànzhōu [in southern Jiāngxī] 赣州. (3 The Five Dynasties History 五代史68/1.)

Hénán was then in the utmost chaos and disorder as a result of the ravages of the barbarians and the numerous uprisings of rebels. Therefore many left their homes and crossed the Dàjiāng 大江 [Yangtze River] into Jiāngxī and then to Fúkiàn. Coming to the territory controlled by the Wáng family, they tactfully claimed Guāngshān as their home district [in order to have a claim on Wáng hospitality based on common place of origin].

Many of them perhaps had legitimate claims, but a great number simply “adopted” Guāngshān. Thus this region, being isolated and mountainous, became a “New Hénán.” They lived here for four centuries (about 900-1300) and during this period probably absorbed the old inhabitants, adopted some of the latter’s customs, and modified their speech which we may characterize as “corrupted Mandarin.”

Return to top, outline

21. The Term “Hakka”

The term Hakka does not imply a racial distinction. It had its origin in the Táng census circa 780.* The Sòng census reports, from 968 to 1223, all use the terms “zhù” [resident] and “kè” [guest] for the general estimate of the country’s population.

*-The terms “zhù” and “kè” were said to occur later in the Sòng period. Yet volume XX of the “Yùhǎi” 玉海 puts the national census for the early Táng period, circa 780. as 1,800,000 “zhù” families and 1,300,000 “kè”. families. It is strange, however, that the “Yuánhéjùn Xiànzhì” 元和郡县志 does not include such terms for the census reports of 713 and 806 of the same dynasty. See Jiāyìngzhōu Annals, 7/86.

After the Sòng dynasty, no such terms are found in the dynastic census surveys. In the first report of A.D. 968, during the early reign of Sòng [emperor] Tàizōng 宋太宗, are the following notes: “The Hàn families, including ‘zhù’ and ‘keh’ …” (“Hànhù Zhù-Kè” 汉户住客 in “Tàipíng Huányǔ Jì” 太平寰宇记, 36/10.) And “the population of Léizhōu 雷州 includes 101 ‘zhù,’ 5 ‘kè,’ and 2 Dànjiā 蛋家 [Boat People] families.” (Ibid., 169/6.)

Moreover, in the third year of Sòng [emperor] Rénzōng 宋仁宗 (A.D. 1051), an edict was issued to decrease the tax burden of Zhāngzhōu 漳州, Quánzhōu 泉州, and Xīnghuà [garrison] 兴化军 because of the poverty of the region and the suffering caused by military disturbances; the reduction for the ‘kè’ families was larger than that of the “zhù” families. (Fújiàn Annals, 52/2.)

In the census report of A.D. 1085 there were 10,109,542 “zhù” and 4,743,144 “kè” families. Included among the former were soldiers, hermits, mountaineers, monks, nuns, boat people, and the semi-civilized Lí tribes; while wanderers and rovers along the borders, those with no definite homes, and outside tribes were included in the latter. (“Wénxiàn Tōngkǎo” 文献通考, Vol. XI.)

According to the “Nínghuà Annals,” native dwellers were called “zhù” while “kè” were strangers from outside districts coming to make a living. (34/8.)

Return to top, outline

22. Six Generalizations From the Evidence

Hence, by a study of the above evidences, we can make the following generalizations:

  1. “Zhù” means “old-timers” or “native inhabitants” while “kè” means “newcomers,” “aliens,” or “guests.”
  2. The terms did not apply to only a certain locality. They were used for the whole country in the Sòng census reports.
  3. The Hakkas got their name because they were not the original inhabitants of the regions which they now occupy.
  4. They were called such by the native or Punti population when the immigrants came to Guǎngdōng. Nothing was implied as to their racial difference or inferiority.
  5. The Hakkas are not “strangers” in some of the regions they occupy. The older inhabitants were absorbed or driven away so that the “strangers” became the sole possessors of the new land.
  6. These terns are still used to differentiate “new-comers” from the older inhabitants.* The Hakkas took their name everywhere they went because their language and customs were little changed. They were inassimilable elements in many ways in the new regions.

*-For example, the first Chinese colonizers to come to Hainan 海南 [island] were the Cantonese. They [and not the Hakkas] were called Kè People 客人. The Hakkas, who came later, were called Ái People 捱人, because their pronunciation for the pronoun “I” [Mandarin: wǒ , Cantonese: ngóh ] in their dialect is “Ngai” [Mandarin: ái ].

We have examined the suppositions that the Hakkas are descended from the Mongol garrison soldiers, or from the mixture of the Qín [dynasty] military colonists and Ikias, or from the aborigines of Fújiàn, but they all seem improbable after careful investigations.

Most probably the ancestors of the Hakkas emigrated from North China to the south following the disorders at the end of Jìn and Táng dynasties [i.e., the early 400s and early 900s]. Many were refugees from the disturbances of the north. Some came as exiles while others came as officials and made their permanent homes there. It is not unusual for a whole village to migrate, as in the case of many Guǎngdōng villages leaving for the overseas. When a clan heard that a neighbor had migrated to a promising region in the south, it also packed up its belongings and worked its long way to the new destination.

Their conservatism and instinctive tendency to cling to their ancestral plots could not check their departure. This is especially true of the Hakka forefathers who migrated as a last resort to a distant land in spite of extreme travelling difficulties. It was the hardy and sturdy ones that survived. (Huntington, op. cit., p. 167.)

It was this spirit of fight and “never-say-die” that made the Hakkas so full of independence, initiative, and pride of race. Their traditions, their characteristics, their customs, and their speech all point to a past historical background in north China. However, the Hakkas are not the “cream of the people” or the “purest of the Chinese.” Their ancestors probably intermarried with the inhabitants with whom they had come into contact.

Return to top, outline

23. The People the Hakkas Displaced: The Kingdom of Yuè

Who were the natives they met? Were they Chinese people or were they semi-civilized tribes?


We must remember that the Kingdom of Yuè [extending over southern China and the northern part of Vietnam — Yuè is the Mandarin pronunciation of “Viet”] under the rule of Gōu Jiàn 勾践 had become very powerful and destroyed its neighbor, the state of Wú in 473 B.C. After this, Yuè declined in power and was finally broken up by [the state of] Chǔ in 333 B.C. into the “Hundred Yuè” 百越 states. The remnants of Yuè [population] then scattered to Fújiàn, Jiāngxī, and Guǎngdōng.

They were not allowed to remain unmolested, however. The Hàn [dynasty emperor] Wǔ Dì 汉武帝 several times exchanged the population of this region with the people in the Jiānghuái 江淮 area [the region between the Yangtze river and the generally parallel Huái river lying to the north of it]. Thus by the time the first wave of refugee Chinese came in the Jìn dynasty, the character of the existing population there had become very complex indeed.

The theory that the Hakkas are descended from these Yuè refugees, therefore, cannot be refuted as being without foundation. But do any of the Hakka traditions point back to the Yuè region or do their popular allusions speak of Yuè? If the Hakkas were descendants of Yuè, why had they lost their identity and become “Hénán-ized”?

It seems probable that the northern immigrants, being stronger in number and culture, drove away or assimilated those in whose regions they had penetrated. They probably had to modify their customs and speech by the contact. Living in a region which was isolated, mountainous, and far from the highways of travel, they were able to consolidate into a homogenous group, united in customs and language and differing from the Puntis or Hoklos [Fukienese].

24. Hakka Spread Within Guǎngdōng Province & Beyond

Thus we have attempted to trace the origin and the migrations of the Hakkas until they finally crossed the Fújiàn-Jiāngxī borders into Guǎngdōng. The second part of this paper deals with the later history of the Hakkas — how they spread over Guǎngdōng and the neighboring provinces, and how they migrated to the overseas.

Jiāyìngzhōu at the end of the Míng dynasty was centre of a raw, rough, new population. There were some quarrelsome elements among the immigrants, and fights were not infrequent. (Citation in Jiāyìngzhōu Annals, 8/1.) There were bandits, ex-soldiers, tramps, exiles, adventurers, as well as farmers, traders, and scholars.

The closing years of the [Míng] dynasty saw this region suffering from frequent political disturbances and local disorders.* Many moved to Guǎngxī, Formosa, and other more promising regions to make their permanent homes.

*-“Heretofore, Bóluó 博罗. (“Poklo”) consisted of the native inhabitants only. The fields were extensive and the people few. … During the reign of [the] Jiālóng 嘉隆 [i.e., Jiājìng 嘉靖 and Lóngqìng 隆庆 emperors] (1522-1573), there were increasing mountain bandits in the northeastern part of the province, but this district was the least disturbed by the disorders. ... In the year 1548-1549, the people from Xīngníng 兴宁 and Chánglè 长乐 (the districts in Jiāyìngzhōu) carried their belongings and came... . The local inhabitants objected, but the officials said that these districts were crowded and poor and that these Cài people (meaning the newcomers as being from Càizhōu 蔡州 [?] in Hénán) were of similar origin; it was better to let them stay together with the natives. After this the newcomers from the two districts settled here. Others from Tīngzhōu and Zhāngzhōu in Fújiàn also came. ... The natives were weaker and the newcomers stronger. Hence disputes and quarrels arose... .” See “Tiānxià Jùnguó Lìbìng Shū” 天下郡国利病书, 100/6.

When Zhāng Xiànzhōng 张献忠, invaded Sìchuān 四川 in 1646, he captured and burned [the capital city of Ch’engtu 成都. (Míngshǐ 明史, Vol. 309.) His extreme cruelty and massacre of the people everywhere he went made the province [of Sìchuān] a waste desert. When conditions became more peaceful, immigrants from the neighbering provinces — Húnán, Húběi, and Jiāngxī — flocked to occupy the waste lands.

One party of Hakkas from Guǎngxī also went along and their descendants now occupy two districts fifty miles west of Chóngqìng 重庆 [in Sìchuān]. (Little, The Far East, pp. 73-74.) They managed to maintain their customs and preserve their speech, although some of the present generation have gradually broken from their clannish ties and adopted Sìchuān as their native home.

The change of dynasty which resulted in the late Qīng [dynastic] house being established by the invading Manchus was also an occasion for many Hakkas to spread west and southwest of Guǎngzhōu. (In the present districts of Huà Xiàn 化县 Hèshān 鹤山, Xīnníng 新宁.)

Since the time of [the early Qīng dynasty] Kāngxī 康熙 [emperor], large numbers of Hakkas were employed to cultivate the military fields, replacing the Puntis and Halos. (Eitel, Outline History of the Hakkas, Rev., II, 142.) Many enlisted in the reign of [the] Qiánlóng 乾隆 [emperor] under the eight banners, a favorite Tartar division. (Ibid.)

Gradually the Hakkas gained favor and influence in the Court. They were admitted to the lower ranks of the government; they took part in the competitive examinations for literary and military degrees; individual Hakkas were appointed as district magistrates and prefects. This caused much opposition from the Puntis. The Hakkas sent a deputation to Běijīng and succeeded in their mission to maintain their rights. (Also from Eitel. The writer has not been able to locate the source concerning the deputation of the Hakkas to Běijīng.)

“It is an impressive picture, that of the old Punti villages, enclosed by ivy-covered brick walls and towers half in ruins and surrounded by ditches, and all about the rows of white houses of the Hakkas, into whose hands the greater part of the lands have fallen. Soon after the appearance of a Hakka house inside the walls of a Punti village, the Punti disappeared completely. It is, however, in the main, a peaceable acquisition by diligence and thrift; though in some instances trickery and force are said not to have been entirely absent from the methods of occupation ...” (Stauffer, op. cit., p. 351.)

This occupation of Punti land soon made the Hakkas feared and disliked; clan fights with the Puntis became more frequent.

Return to top, outline

25. Hakkas at War

The nucleus of the Tàipíng 太平 Rebellion was formed of Hakkas from Guǎngdōng and Guǎngxī. The Tàipíng “Emperor,” Hóng Xiùquán 洪秀全 was a Hakka from Huà Xiàn, northwest of Guǎngzhōu. Most of the ministers or kings (as styled), generals, and administrators also came from Hakka homes, “Through all the vicissitude of their march from Guǎngdōng to Nánjīng they succeeded in retaining the supremacy among the ill-assorted cohorts of rebels from all the eighteen provinces.” (Also from Eitel.) A Hakka Bamboo Rifle corps was formed in the 1860 war to fight for the allied Anglo-French troops against Běijīng.*

*-This is one of the peculiar characteristics of the Chinese people under the rule of the Manchus. There was no national consciousness, patriotism, or collective co-operation, When one part of the country was with the foreigners, the other part carried on friendly relations and even helped them in their military operations.

The clan fights between the Hakkas and Puntis reached a climax in 1854. The Tàipíng Rebellion had spread to Zhàoqìng 肇庆, where numerous Punti people joined the rebels.

The Hakka clans located there, however, for the most part remained loyal to the government. Much bad feeling was aroused and, when the Tàipíng rebels went north, the internecine war began first in Hèshān 鹤山 and spread all over the southwest districts.

The Puntis, being stronger in men and means, defeated the Hakkas in these districts and expelled those not killed. In 1862 the contest was at its height when imperial authority was entirely suspended in several districts. By the end of the year, the Hakkas were outnumbered and driven towards the coast. They then stormed and occupied the fortified town of Guǎnghǎi 广海west of Macao (Mandarin: Àomén 澳门. They were soon driven out by the Puntis with the help of the imperial forces, this being the first overt act of official interference in the contest.

26. Hakkas Defeated

In 1864 there were at least two hundred thousand wandering Hakkas who turned half-bandits and half-refugees. They collected in the mountains of the west districts and defied the Puntis. There was no means for reconciliation, as the Hakkas only wanted liberty to live while the Puntis refused to be “sheltered under the same heaven with them,”*

*-Most of the materials concerning this strife come from Eitel — Outline History of the Hakkas. Rev., II, 163-164.

In September, 1866, a new governor came and sent eight thousand troops under the Grain Intendant of Guǎngzhōu to the west districts to compel the Hakkas to give up their arms and disperse. Two hundred thousand taels were set aside for distribution in proportion of eight taels to each adult and four to each child, with passes and protection to enable them to migrate to Guǎngxī, Hǎinán 海南, and other regions where waste lands exist in abundance. The small district of Chìxī 赤溪 was marked out from Xīnníng and set apart for the Hakkas to reside.

Thus ended the long bloody strife in which more than one hundred fifty thousand Hakkas perished. Some became bandits; some were sold as slaves to coolie ships at Macao; some escaped to other provinces; some migrated to Formosa, Saigon, and Singapore; others stayed back and intermingled with the Puntis. It was this internecine war through which many people came to have a wrong impression that the Hakkas have always been the object of merciless persecution since the time of Qín Shǐhuáng and indulged in far-fetched-theories concerning their origin.

27. Hakkas Eternally Glorious

This misfortune does not mean the end of Hakka influence. The exclusive Hakka regions around Jiāyìngzhōu were untouched by the strife and still remain under Hakka domination. The fight for the republic against the Manchus revealed many Hakka patriots and martyrs. The Hakkas abroad contributed great sums of money to support the Revolution of 1912. Today in all lines of work, Hakka influence is felt. The military genius of this people has not diminished — Chén Jiǒngmíng 陈炯明, Zhāng Fǎkuí 张法奎, Huáng Qíxiáng 黄祺翔 are of Hakka ancestry. Their political aptitude is unquestioned — many of them hold high positions in the government.

As leaders in commerce and industry, the Hakkas are well known in China and abroad; the Cháozhōu-Shàntóu Railway (Cháo-Shàn Tiělù 潮汕铁路) was built by Hakka contractors and is now largely under the control of Hakka merchants. With the modernization of China where facilities for communication and transportation have advanced and opportunities for contact are more frequent, the Hakkas have gradually mingled with the Cantonese and others.

Many of them as students, merchants, or workers in Guǎngzhōu or Hongkong eventually give up their clannish ways and use Cantonese as a medium of speech. This action is commendable, because we want to see all China speaking one language and the people calling themselves Chinese, not Hakkas or Puntis, not Cantonese or Hunanese, not Northerner or Southerner, not Manchu or Mongol. We all should cast off our clannish ways and work together for the welfare of China.

Return to top, outline

Appendix: Table of Events

Dates B.C.
333 Kingdom of Yuè destroyed by Chǔ .
ca 300 Hakkas located in Shāndōng 山东, Shānxī 山西, Ānhuī 安徽.
249-209 Qín Shǐhuáng 秦始皇. Persecutions. First migration of Hakkas to Hénán 河南, Jiāngxī 江西.
214 500,000 military colonists sent to Guǎngdōng 广东 and Guǎngxī 广西.
206 Beginning of Hàn dynasty.
140-86 Hàn [Emperor] Wǔ Dì 汉武帝 exchanges populations of the Huái Hé 淮河 [river basin] and Zhèjiāng 浙江, Fújiàn 福建, Guǎngdōng 广东.

Dates A.D.
221 End of Hàn dynasty. Chaos.
265 West Jìn 西晋 dynasty begins.
313 Jìn [Emperor] Huái Dì 晋怀地 captured and killed by incoming barbarians.
317 Nánjīng 南京 now capital. East Jìn 东晋 dynasty begins. Jìn Min-Ti 晋明帝also captured and killed. Second migration of Hakkas reaches Jiāngxī 江西 and Fújiàn 福建.
420-618 Period of Barbarian Dynasties.
618-907 Táng dynasty. Substantial colonization of the south.
780 National census first mentions such terms as “native” (zhù ) and “newcomer” (kè ).
885 5000 soldiers led by the Wáng brothers leave Hénán 河南 for Fújiàn 福建 for safety.
907-960 Period of the Five Dynasties. Barbarian invasions and ravaging. Third migration of Hakkas reaches Southern Jiāngxī-Fújiàn 江西/福建 border.
960 Beginning of Northern Sòng 北宋 dynasty.
972 First Sòng census to separate “natives” and “newcomers”.
1049 Méizhōu 梅州 builds a mud wall.
1084 Méizhōu 梅州 Changes to brick will.
1127-1280 Southern Sòng dynasty. Barbarian invasions. Fourth and last migration down south.
1278 Mongols capture Méizhōu 梅州. Devastation.
1280 Beginning of Yuán dynasty. First wave of Hakka migration into Guǎngdōng.
1368 End of Yuán dynasty. Beginning of Míng Period. Start of second wave of migration from Jiāngxī 江西 and Fújiàn 福建 to Jiāyìngzhōu 嘉应州.
1644 Qīng dynasty begins. Chaos and disturbances.
1646 Zhāng Xiànzhōng 张献忠 ravages Sìchuān 四川. Guǎngxī 广西 Hakkas migrate to Chungking region.
1662-1723 Kāngxī 清康熙 Emperor. Hakkas replace Punti and Hoklo military cultivators.
1723-1735 Hakkas spread west and southwest of Guǎngzhōu 广州.
1736-1796 Qiánlóng 清乾隆 Emperor. Hakkas enlist under the eight banners. Hakkas send deputation to Běijīng 北京.
1450-1864 Tàipíng 太平 Rebellion.
1851-1862 Tóngzhì 清同治 Emperor.
1854 Internecine war begins at Hèshān 鹤山.
1860 Anglo-French war against China. Hakkas serve in Bamboo Rifle Corps for Allies.
1866 Internecine strife ends. Hakkas migrate to Guǎngxī 广西, Hǎinán 海南, Formosa, and overseas.
1912 Establishment of Chinese Republic.

Return to top, outline



BALL, James Dyer
Hakka Made Easy (London, 1884). Introduction.
BROOMHALL, Marshall (ed)
The Chinese Empire (London, Morgan & Scott, 1907). Pp. 43-45 on the Hakkas.
“Origin and Migrations of the Hakkas,” in the Chinese Recorder, XLIII, (Shanghai).
COOLING, Samuel (ed.)
Article on Hakkas. p, 222, in Encyclopedia Sinica (Shanghai, Kelly & Walsh, 1915).
A series of articles on the Hakka Chinese in The China Review (London), such as
Vol. II. pp. 160-164, on the Outline History of the Hakkas;
Vol. VIII. pp 316-318 on the Hakka Dialect;
pp, 316-321 on Marriage Customs ; and
Vol. XX, pp. 263-267 on Ethnological Sketches of the Hakkas.
Encyclopedia Britannica,
Article on Hakkas, p. 828, in Vol. XII, 11th Edition (Cambridge, 1910).
Character of Races (N. Y., Scribner, 1924), pp, 149-171 on migrations of Chinese to the South.
LECHLER, Rev. Rudolf
“The Hakkas Chinese” in the Chinese Recorder (1878) IX. 352-359.
LITTLE, Archibald
The Far East (Oxford, 1905), p. 149 on history of the Hakkas.
LIU, Ch’iang (Liú Qiáng 劉強)
“Isolation and Contact as Factors in the Cultural Evolution of China, Korea, and Japan prior to 1842,” in the Chinese Soc. and Pol. Science Review (1927) XI, pp. 492-493, on the intermingling of the early Chinese settlers.
MEADOWS, Thomas Taylor
The Chinese and Their Rebellions (London, Smith, 1856). pp. 36-48.
The Chinese Miscellany (Shanghai, 1896) II, p. 475.
New International Encyclopedia (N. Y., 1905), IX, p. 461.
PITON, Rev. Charles
“The Hia-K’ah in the Chekiang Province and the Hakka in the Canton Province,” in the Chinese Recorder (1870) II, pp. 218-220.
“The Origin and History of the Hakkas ; in China Review (London, 1873-1874) II, pp. 222-224.
Comprehensive Geography of China (Shanghai, 1908), pp. 199, 204-207, 343-344.
STAUFFER, Milton T. (ed.)
Christian Occupation of China (Shanghai, 1922). See article on Hakkas by Rev. D. Oehler, pp. 351-353.
WILLIAMS, Samuel Wells
The Middle Kingdom (N. Y.. Scribner’s, 1901), I, p. 138, and II., p. 586-591, on the Hakkas.
Geography of the world (Shanghai, Commercial Press. 1926, 4th revised edition) p. 144.
Sur cinq cranes d’Hakkas et les origines chinoises. Article in Bulletin de la Société d’Anthropologie de Paris, Série 3, Tome 2, pp. 557-578 (Librairie de l’Académie de Médicine, 1879).

Foreign (Not Cited)

Lettre on the Hakkas Article in Mission Catholique, 1879. Vol. Xl, pp. 505 -606.
de RIALLE, Girard
The Hakkas. Article in Review d'Anthropologie (Jan. and April, 1885).
GIRARD, Dr. Henry
Notes sur les Chinois du Quangsi, Article in L'anthropologie, 1898, Vol. IX, pp. 144-170.
Die Hakka Chinesen. Article in Verb. Berliner Ges. f. Antli., 1879.
Ann English-Chinese Dictionary in the Vernacular of the Hakka People. (Am. Presbyterian Mission Press, Shanghai, 1905).
North China Herald, June 29, 1867 (Shanghai).
Article on Hakka reprinted from the China Mail, Hongkong.
PITON. Rev. Charles
Une visite au pays des Hakkas dans la province de Canton. See Bul. Soc. Neuchâteloise Geog., VII. pp. 31-51 (1892-1893).
Contribution a l'étude anthropologie des Chinois-Hakka de la province de Moncey (Tonkin). See L'Anthropologie, XXX. No 1-2, pp. 83-109 (June, 1920).
VOMEL, Johann Heinrich Von
Der Hakka Dialect. See T'oung Pao 通報 (Dec. 1913) pp. 597-696.


BÁI Yuèhéng 白月恆 著
民國地誌.( 化京高師圖書館總發行民十初版.)
CÉNG Guófān 曾國藩 等重修
CHÉNG Mèngjiǎn 程夢簡 修
DÈNG 鄧氏族譜
GÙ Yánwǔ 顧炎武 著
HÉ Kūn 和坤 等纂修
HUÁNG 黃氏族譜
KǑNG Tíngzhāng 孔廷璋 等譯編
中華地理全誌.( 上侮中華書局民四第三版.)
LǏ Jífǔ 李吉甫 選
LǏ Xián 李賢 等撰·
LǏ Zhòngyuán 李仲元 撰修
LÙ Yìngyáng 陸應陽 撰·
MǍ Duānlín 馬端臨 著·
RUǍN Yuán 阮元 等重修
SHĚN Zuògān 沈作乾 著
畬民調責記. ( 見東方雜誌廿一卷 第七號 十四年十月.)
SĪMǍ Guāng 司馬光 著·
SŪN Ěrzhǔn 孫爾準 等市纂
WÁNG Cún 王存 等纂修
WÁNG Xiàngzhī 王象之 褊
WÁNG Yìnglín 王應麟 撰
WÁNG 王錫祺 輯
WĒN Shēnhé 溫伸和 等修
XIÈQǏ Kūn 謝啟昆 等修
YÁNG 楊氏續修族譜
YUÈ Shǐ 樂史 選
ZHĀNG Tàiyán 章太炎 著
章氏叢書. ( 新方言篇有嶺外三州一章.)
ZHĀNG Tiānpéi 張天培 纂修
ZHĀNG Xījīng 張希京 等修
Zhōng dúfó 鍾獨佛 著
粵省民族攷原.( 見嘉應雜誌 第二期 民十年五月.)
Zhōnghuá Dìxué Huì 中華地學會
地學雜誌 (Geographic Magazine)
Zhōnghuá Kèyǔ 中華客語攷原總會章程
( 客系大同盟 民十年四月在廣州開會.)

Chinese (Not Cited)

HUÁNG Gōngdù 黃公度
HUÁNG Xiāngtiě 黃香鐵
Méizhōu Tújīng 梅州圖經
Shàntóu Dàtóng Bào 汕頭大同報
( 民十一年?)
Shàntóu Gōngyán Bào 汕頭公言報
( 民十年一月二十九日.)

Return to top, outline