Origins and History of Baguazhang

This page sets out what is documented, examines the two competing accounts of the art's source, and describes how it divided into branches within a single generation.

Dong Haichuan

Dong Haichuan came from Wen'an county in Hebei, in the plain south of Beijing. His birth year is uncertain — the two figures usually given differ by about fifteen years — and little is known of his early life beyond that he had some martial training, as many men in that region did.

At some point he left home and travelled, and by the 1860s he was in Beijing. The traditional account of his arrival is that he took a servant's position in the household of a Manchu prince, and that his skill was discovered by accident when he crossed a crowded room carrying dishes with a speed and agility nobody could account for. The prince, so the story goes, then appointed him a martial instructor.

Whether or not the banquet episode happened, the substance is corroborated: Dong was attached to a princely household and taught martial arts in Beijing from around that period until his death in 1882.

One persistent question concerns whether Dong was a eunuch, which would have been a normal condition for service in an inner household. The matter has been debated for over a century and was raised again when his grave was relocated in the early 1980s. We do not think the available reports settle it.

The stele

The year after his death, Dong's students erected a stone stele at his grave. Further stelae were added in later decades, and the texts have been recovered and re-erected. This is the single most valuable documentary source on the art's early history.

The inscriptions record where Dong came from, give an account of how he acquired his method, and list the names of his students. The list of names is the important part: it lets us establish who the first generation actually were, rather than relying on lineage claims made much later by people with an interest in the answer.

The stele also matters because of its date. It was put up within a year of Dong's death by people who had known him. Whatever the inscription says is at least a record of what his own students believed in 1883, which is a far better evidential position than taijiquan's earliest documents occupy.

The mountain Taoist account

What the inscription says about the art's source is that Dong received it in the mountains from a Taoist.

This is the account Dong gave, and his students recorded it on stone. It is unverifiable in the ordinary way: no such teacher has been identified, no lineage before Dong can be traced, and the story has the structure of an authorising legend of exactly the kind we treat with caution elsewhere on this site.

Two points, however, distinguish it from the Zhang Sanfeng case. Dong made the claim during his own lifetime about his own training, not centuries after the fact about somebody else. And circle walking, the art's central practice, has a real precedent in Taoist ritual circumambulation, so the story is at least pointing at something that exists.

The sect hypothesis

There is a substantial alternative explanation, and it deserves to be taken seriously because it explains something the standard account does not: why Dong might have wanted his origins to be vague.

Hebei and Shandong in the late eighteenth and early nineteenth centuries were the heartland of the Eight Trigram teaching, a millenarian religious movement that used trigram symbolism to organise its structure, practised forms of circular walking as devotion, and mounted a serious armed uprising in 1813 which was crushed with great severity. Association with it remained dangerous for generations.

The hypothesis is that baguazhang's trigram framework and its circle walking derive from this milieu rather than from mountain Taoists, and that Dong, teaching in the imperial capital in a prince's household, had compelling reasons not to say so.

The evidence is circumstantial: geographical overlap, the shared eight-fold organisation, and the shared practice of walking a circle. It has not been proved and probably cannot be. But it is the most substantive alternative available, and it has the merit of explaining an otherwise puzzling vagueness.

Modern research

From the 1980s, Chinese researchers investigated Dong's home region directly, examining local records, genealogies and surviving traditions, and looking for candidate teachers among identifiable local martial artists rather than among immortals.

This work produced plausible local candidates and established that the region had martial traditions Dong could have drawn on. It did not produce a documented transmission, and the question remains open.

What this research did settle is that Dong did not invent the art from nothing in Beijing. He arrived with something, and the something had local antecedents. The methodological point is the same one Tang Hao demonstrated for taijiquan: go to the place, read the records, and see what is actually there.

Why the branches diverged immediately

Dong taught in Beijing for roughly twenty years and had a substantial number of students, many of them already accomplished martial artists when they came to him.

The consistent tradition is that he taught them differently, building his method onto whatever each already knew. This is unusual and it had a permanent effect: the major branches were already distinct in the first generation, and each carries visible traces of its founder's earlier training.

The consequence for anyone studying the art today is that there is no original version to recover. The branches are not corruptions of a lost standard; they are what the transmission produced by design.

Yin Fu and Cheng Tinghua

Two students dominate the subsequent history.

Yin Fu came from a background in a hard external tradition, and his baguazhang reflects it. The hand shape associated with his line is flat with extended fingers, the movements are more linear and piercing, the frame is smaller, and the emphasis falls on penetrating rather than seizing.

Cheng Tinghua worked as a spectacle maker in Beijing and had trained in wrestling. His line uses a curved, claw-like hand, larger and rounder movements, and makes far more use of grabbing, twisting and throwing — recognisably the contribution of a wrestler.

Cheng Tinghua's death is itself a documented historical event: he was killed in 1900 by foreign troops during the suppression of the Boxer uprising. It is a reminder that these were not figures from a distant legendary past but men living through a violent and well-recorded period.

Other first-generation students founded lines of their own, including one noted for kicking methods and one whose lineage became particularly widespread through his own long teaching career.

The alliance with xingyiquan

In the 1890s senior practitioners of baguazhang and xingyiquan in Beijing agreed to treat their arts as related, to exchange students, and to teach each other's methods.

This was a practical arrangement between colleagues, not a discovery of common ancestry. But it had large consequences. It produced a generation of practitioners fluent in both arts, it created the technical exchange visible in later forms of each, and it laid the groundwork for the grouping of taiji, bagua and xingyi as a single internal family — a category that, as our Foundations pages explain, was formalised shortly afterwards.

The twentieth century

Baguazhang entered the twentieth century as an established Beijing art with a documented lineage and a strong practical reputation. It was taught at the national martial arts institute in the Republican period, and Sun Lutang, who had learned it from Cheng Tinghua, wrote about it and folded elements of it into his own taijiquan.

Compared with taijiquan it never became a mass health practice, which has had mixed effects. It remained smaller and less commercially pressured, and rather more of its martial content survived. It also received far less standardisation, so the branches remain genuinely distinct.

What remains open

Where Dong Haichuan learned the method. Whether the sect connection is real. How much of the trigram theory was his and how much was added by students — the balance of evidence favours the students, as our page on the Yijing discusses. And how much each first-generation branch changed the art it received.

These are answerable questions in principle, and further work in local archives could still move them. That is more than can be said for most origin questions in Chinese martial arts.

Key points

• Dong Haichuan taught in Beijing from around the 1860s until his death in 1882, attached to a princely household, and his students erected a dated stele naming themselves.
• The stele records that he received the art from a Taoist in the mountains; the claim is contemporaneous with him but cannot be verified.
• The main alternative is that the trigram framework and circle walking derive from the Eight Trigram sect milieu in Hebei, which would explain why Dong was vague about his origins.
• Modern fieldwork in Dong's home region found plausible local antecedents but no documented transmission, leaving the question open.
• Dong taught each student differently according to their prior art, so the branches were distinct in the first generation and no original version exists to recover.
• The 1890s alliance with xingyiquan in Beijing produced real technical exchange and prepared the ground for the internal family category.