Three Real Men Bent by Jin Yong — and the Renaming That Fixed Nothing
**Câu trả lời cốt lõi:** Kim Dung đã bẻ cong ba nhân vật lịch sử trong tiểu thuyết võ hiệp: Lữ Văn Đức (đảo ngược tính cách), Doãn Chí Bình (gán tội cưỡng bức không có trong sử liệu) và Thi Lang (gắn nhãn Hán gian). Chỉ trường hợp Doãn Chí Bình được sửa, và chỉ bằng cách đổi tên thành Chân Chí Bính ở lần tu chính thứ ba. **Dữ kiện chính:** - Doãn Chí Bình (1169–1251), hiệu Thanh Hòa tử, là chân nhân đời thứ sáu phái Toàn Chân, kế vị Khâu Xử Cơ và nhường chức cho Lý Chí Thường. - Năm 2003, cộng đồng đạo sĩ chặn đoàn xe của Kim Dung tại Hoa Sơn; sau đó tên nhân vật được đổi trong vòng tu chính “Tân thế kỷ”. - Thi Lang (1621–1696) thắng Lưu Quốc Hiên ở Bành Hồ năm 1683 và dâng sớ xin giữ, lập phòng thủ Đài Loan. - Lữ Văn Hoán, em trai Lữ Văn Đức, mở cửa Tương Dương hàng quân Mông năm 1273 — chi tiết bị bài báo gốc bỏ qua. - Chín trong ba mươi tám điểm dữ kiện của bài báo gốc ghi “nguồn: không”, gồm toàn bộ tiểu sử Thi Lang. **Nguồn:** history.ifeng.com, bài tổng hợp không ghi tác giả (bản dịch và phân tích tiếp nhận); đối chiếu Tống sử quyển 45, Thanh Hòa chân nhân Bắc du ngữ lục, Hội Đạo giáo tỉnh Sơn Đông, Thông tấn xã Trung Quốc (2014), Bảo tàng Cố cung, Đại học Hạ Môn. **Hỏi đáp liên quan:** - Hỏi: Kim Dung có sửa nội dung vụ việc của Doãn Chí Bình không? Đáp: Không, chỉ đổi tên nhân vật thành Chân Chí Bính, cốt truyện giữ nguyên. - Hỏi: Vì sao chỉ một trong ba trường hợp được sửa? Đáp: Vì chỉ cộng đồng đạo sĩ có tổ chức mới tạo được sức ép, còn hậu duệ của Lữ Văn Đức và Thi Lang thì không. - Hỏi: Có thể coi Thi Lang là “Hán gian” theo nghĩa sử liệu không? Đáp: Đó là một phán xét chính trị gây tranh cãi, không phải dữ kiện kiểm chứng được; Bảo tàng Cố cung chỉ chứng minh năng lực quân sự của ông." } ```
In 2026, at the foot of Mount Hua, a group of Taoist priests blocked Jin Yong's convoy. They carried no long banners. They carried one name: Yin Zhiping, sixth-generation patriarch of the Quanzhen sect, dead since 1251. In The Return of the Condor Heroes, that name belongs to the man who violates Xiaolongnü and then falls on his own sword. In the record, the same name belongs to the successor of Qiu Chuji, who followed his master to meet Genghis Khan, raised the funding for the Da Yuan Xuandu Baozang canon, and later ceded the patriarchate to Li Zhichang.
The convoy stopped. And in the third revision cycle, the name in the book became Zhen Zhibing.

I came to this through an unsigned aggregation piece on history.ifeng.com. My trade is finding the piece of the puzzle the whole stadium walked past, so I did not read it to nod along. I ruled three columns: what the novel says, what the record says, and the third column — the one that matters — where each line came from.
Jin Yong wrote fifteen works, stopped serialising in 2026, and then spent decades revising himself. Three cycles. The last, the New Century edition, changed a run of character names. Yin Zhiping became Zhen Zhibing.
I make martial-arts documentaries; I am not a literary scholar, so let me state my ground plainly. How a martial culture tells its own past, and how it responds when caught out, is my professional territory. British boxing still holds fight tape from the 1950s; Japanese wrestling still holds its belt registers; wuxia holds its memory in novels. When the novel is the only source most readers ever touch, the novel quietly takes the historian's chair.
The piece picks three figures: Lü Wende in The Legend of the Condor Heroes and its sequel, Yin Zhiping in The Return of the Condor Heroes, Shi Lang in The Deer and the Cauldron. All three were real men with birth and death dates, offices and paper trails. All three were placed in situations the record never contains.
That is where I took out paper. I count by hand. I once counted every phase of Iceland's game and found a whole world inside it, and here I counted the same way: thirty-eight factual points in the piece, nine of them marked “source: none”.
Lü Wende first. In the novels he holds Xiangyang, trembles at the Mongol approach, and leaves Guo Jing and Huang Rong to carry the entire defence. The record draws someone else: a commoner who rose to general, fought on the Jianghuai, Jinghu and Sichuan fronts, relieved Anfeng by feint, scattered an enemy supply column at Shouchun, and in 1259 marched to the relief of Ezhou. The distortion here is a character inversion plus a transfer of agency: one real man's courage peeled off and pasted onto two fictional protagonists.
But my third column is not clean. The Anfeng and Shouchun operations are sourced to the Song Shi, volume 45 — solid. Most of the rest of Lü Wende's life is marked “source: none”. And the piece concedes something telling: Chinese historians had already criticised Lü Wende for military errors and for factionalism with Jia Sidao. The novel amplified an existing critical tradition rather than inventing one from nothing.
There is a gap the piece never mentions, and I think it is the heaviest one: Lü Wenhuan, Lü Wende's younger brother, held Xiangyang and opened the gates to the Mongols in 1273. The Lü family record is genuinely mixed. Silence on that point is a storytelling choice, not an accident.
Yin Zhiping, 1169–1251, sobriquet Qinghezi. This is where the evidence is thickest, and where I felt coldest. Four independent strands: the Taoist internal chronicle Qinghe zhenren Beiyou yulu, the Shandong Taoist Association, and Professor Zuo Hongtao speaking to China's national news agency in 2026. A religious order with a documented succession, with a named successor, had a sexual crime attached to its patriarch that no source outside the novel records. This is the only case that meets a strict definition of defamation.
Shi Lang, 1621–1696. Naval commander, victor over Liu Guoxuan at Penghu in 1683, author of the memorial arguing that Taiwan be retained and garrisoned rather than abandoned, humane in his handling of Zheng Keshuang and Liu Guoxuan after their surrender. In The Deer and the Cauldron he is humiliated by Wei Xiaobao again and again and branded a Han traitor. What is bent here is not an event but a verdict: calling a man a traitor is a political conclusion, not a falsifiable datum. The Palace Museum and Professor Wang Rigen of Xiamen University establish his professional competence, not the absence of a moral controversy. His entire core biography in the piece is marked “source: none”.
And here is what made me rewrite my outline: the piece carries its own evidentiary asymmetry. The case with the strongest sourcing — Yin Zhiping — is precisely the case the novel cannot defend. The cases with the thinnest sourcing — Lü Wende, Shi Lang — are precisely where the “slander” thesis is easiest to contest.
The renaming fixed nothing, and I think that is the larger lesson.
Changing Yin Zhiping to Zhen Zhibing is an identification fix. The plot beat survives intact. A fictional sect still keeps a rapist in its ranks; he simply no longer carries a real patriarch's name. What was protected was the name, not the character. The precedent is narrow: a fictional villain may exist, provided he does not borrow a real man's identity.
And that precedent exists only because an organised constituency wanted it. The Taoist communities of Shandong and Shaanxi are small but structurally coherent — they have spokespeople, and they can stand in front of a convoy. Lü Wende's and Shi Lang's descendants have no such structure. Neither do the historians who criticise Jin Yong. So they got nothing. Accountability in historical fiction runs on constituency, not on principle.
One more thing. The piece says Zheng Chenggong treated Shi Lang unfairly and executed his family over a minor fault. The sequence is reversed. The standard account runs: a dispute over a subordinate, Zheng imprisoned Shi Lang's family, Shi Lang defected to the Qing, and only then were his father and brother executed. Putting the killings before the defection flips the entire moral arithmetic. And Chen Jinnan, who stands opposite Shi Lang in the novel, is himself a fictional composite loosely built on Chen Yonghua — one invented man confronting one real one.
The piece also counts only the places where Jin Yong tore people down. It does not count the places where he built them up past the record. A count that runs in one direction is not a count.
Based on my experience covering fights and cutting martial-arts documentaries, I keep the old habit: when a detail cannot be verified, I write it as an estimate with its context attached, or I cut it. A 0.2-second discrepancy once taught me to trust my own eyes, and it also taught me that anything without a source must not be dressed in the clothes of data.
Wuxia is a shared language for hundreds of millions of people. Precisely because it is that strong, a name borrowed from a real man deserves a fuller return than a change of characters.
