Dragon Gate Inn (a.k.a. Dragon Inn) is one of the best known and most influential wuxia films by the Chinese director known as King Hu. Both this film and the equally celebrated A Touch of Zen (1971) are currently streaming on MUBI. Dragon Gate Inn is also available on separate Blu-ray editions (Masters of Cinema in the UK and Criterion in the US) of the 2013 restoration of the film. I remember these films from the 1970s in the UK in cinemas and later on prints broadcast by Channel 4 and I think I must have seen at least extracts from then but it wasn’t until I bought a DVD of the 1992 remake that I really began to explore wuxia – more of that later.
King Hu was born in Peking in 1932 and attended art school before joining the flow of exiles to Hong Kong in 1949. He joined the Shaw Brothers Studio in 1958 a nd worked in various capacities on the studio’s Mandarin films and came under the influence of the Taiwanese director Li Han-hsiang. Hu directed his own films starting in 1965 and it was his second film, Come Drink With Me (1966) which introduced his update to the wuxia. Soon after this he left Shaw Brothers, moving to Taiwan where Dragon Inn was produced for the Union Film Company – a local producer giving Hu the chance to try out his own ideas.
Wuxia is an ancient Chinese form of fantasy literature that in the 20th century developed in new forms such as newspapers and magazines and, soon after, cinema. The term refers loosely to the idea of ‘martial chivalry’ and the idea that highly trained warriors would roam the country righting wrongs. After 1949 the nationalised Chinese film industry reduced its reliance on these kinds of action genre films and the development of the genre fell to the exilic directors in Hong Kong and Taiwan. There is a suggestion that wuxia began to be influenced by Japanese chanbara (swordplay) and Japanese cinema was accessible in HK and Taiwan. King Hu developed a particularly elegant and ‘open’ style of traditional wuxia in which he maintained roles for female martial artists and staged swordplay in spectacular natural environments, presented in CinemaScope formats and with traditional music scores.
The plot of the film is relatively straightforward. In 1457 AD, during the Ming Dynasty period, the Empire is being controlled by a group of eunuchs in the Imperial Palace who have already framed the Defence Minister Yu as a traitor and executed him. His two surviving children are to be sent to a remote destination in the desert but the Dong Chang ‘Eastern Agency’, in effect a military group controlled by the eunuch Ciao (Ying Bai), has been ordered to ambush and kill the children and their guards. Their first attempt to do so is thwarted by two sibling warriors (Polly Ling-Feng Shang-Kuan and her brother played by Han Hsieh) and the eunuch’s men decide to descend on the Dragon Gate Inn which is the children’s ultimate destination. The Eastern Agency announce themselves as working for the Ministry of Justice, expecting to find some Imperial Guards at the Inn, but they are not expecting the presence of other independent warriors including master swordsman Xiao (Chun Shih) and the innkeeper Wu (Chien Tsao) who turns out to be one of Yu’s warriors. At the inn a series of encounters between the Eastern Agency troops and the four warriors loyal to Yu’s memory escalates and Ciao himself arrives to try to finish the job. The ending may be predictable in that ‘good’ triumphs over evil, but each contest is different and the narrative is gripping throughout.
There are several reasons why Dragon Gate Inn became a massive hit in Taiwan and later Hong Kong (where Shaw Brothers delayed the film’s release to help one of their own films. Whatever you might think of action films and ‘martial arts’ films in particular, watching this film for ten or twenty minutes will convince you that Hu was a master filmmaker. Hu was originally an actor and an art designer before becoming a writer-director. In wuxia the characters are individuated through action and the visual qualities of their performances, including facial gestures and body postures as well as their martial arts skills and athleticism. Hu handles his actors well and he makes excellent use of his resources. There is just one studio set in the film as far as I can see – the inn itself which offers many opportunities for imaginatively-staged action set pieces – overturned tables, staircases, roofs, balconies, balustrades, windows/doors etc. This is then contrasted with the vast open spaces of deserts, mountains and river-beds. The Taiwanese interior offered so much more than the limited spaces of Hong Kong and Hu presents them on screen using CinemaScope ratio (2.35:1) and often extreme long shot framing by Hua Hui-Ying. But what makes him a special director is the ease with which he shifts from extreme long shot to close-up and a whole range of framings. His staging in depth is equally impressive.
Hu’s knowledge of Peking Opera informs the staging of his swordfight sequences and these are presented without the later wire work. Much is achieved by the editing which suggests movement rather than presenting it directly. The final ingredient is the music score by Chow Lan-Ping which is evocative of Japanese as well as Chinese cinema. Put all these ingredients and Hu’s skills together and stand back. You can recognise now that Hu’s masterpieces comes at a particular time in the late 1960s and early 1970s when American, Japanese and European cinema are converging around the Japanese chanbara and Sergio Leone and Sam Peckinpah’s Westerns (these latter two directors equally drawing on Kurosawa’s work). We might look even further back to John Ford Westerns (which influenced Kurosawa). The isolated inn in the desert is in many ways similar to the Edwards homestead in The Searchers or the inn in Stagecoach, both set in the landscapes of Monument Valley.
Dragon Gate Inn was more than a major hit film, it raised expectations of what a wuxia film could be and without it the later global blockbusters by Ang Lee with Crouching Tiger, Hidden Dragon (2000) and Zhang Yimou with Hero (2002) and House of Flying Daggers (2004) would not have had quite the impact they eventually achieved. But before that, Hu’s success encouraged the filmmakers of the Hong Kong New Waves with Wong Kar-wai’s Ashes of Time (1994) and Ann Hui’s Romance of Book and Sword and Princess Fragrance (both 1987). I haven’t seen these last three titles but my own introduction to appreciation of wuxia came through films produced by Tsui Hark in Hong Kong and starring action heroes such as Brigitte Lin and Michelle Yeoh. In particular, I enjoyed the remake of Dragon Gate Inn (New Dragon Gate Inn, 1992) with Brigitte Lin, Tony Leung Ka-fai and Maggie Cheung (as the inn-keeper) and Donnie Yen as the evil eunuch. The remake uses standard modern widescreen (1.85:1) and the exteriors are limited by comparison, but the performances and dialogue as well as action choreography make for an entertaining film.
In Taiwan, King Hu is still an important figure in Taiwanese film history and his status was confirmed by Goodbye Dragon Inn (Taiwan 2003) in which the auteurist direct Tsai Ming-liang constructs a narrative around the final screening at a major classic single screen cinema. Only a few fans are present and various encounters take place between them during a screening of Dragon Inn. I do struggle with Tsai’s films but I saw this one in Bradford with a small audience in Pictureville cinema if my memory serves. I wish I’d seen the whole of Dragon Inn at that point. I would have perhaps made even more sense of Tsai’s tribute.
I recommend the Blu-ray by Masters of Cinema (which also includes a DVD version of the film). The extras include a David Cairns visual essay and a newsreel of the films opening in Taipei plus a booklet with pieces by Tsui Hark, Tony Rayns and Edmond Wong. The Criterion release has similar material by different scholars/industry personnel/actors.
(My apologies about names for actors and crew in this posting. The various sources I’ve consulted don’t use either the Wade-Giles romanisation of Mandarin names (as in Taiwan at the time) or the Hanyu Pinyin which replaced it in the PRC in the 1950s consistently and I may have unintentionally mixed them up.
Trailer for the Criterion Blu-ray release:
Assassin is the kind of film that you don’t expect to understand after a single screening. As I left the cinema an audience member spoke to an usher who asked him what he thought of the film. “Well, it was very beautiful”, he said, “I didn’t understand it all, but that’s OK because I enjoyed the experience”. I feel much the same, except I thought I understood quite a bit of it until I spoke to my viewing companion and then started to read the reviewers who did understand it and who had actually discussed it with director Hou Hsiao-hsien (such as Tony Rayns in Sight and Sound February 2016). As I read more about it, the film made more sense but also revealed some of the aspects that I’d either missed altogether or seen but failed to make sense of. I do hope to watch the film again, although I’m not sure where. Assassin is not playing in many cinemas and I do worry about how StudioCanal are organising its distribution. In the meantime there are aspects of the film I’d like to discuss and I’m conscious that there is almost a ‘meta-text’ being constructed in the various discourses about the film both in print and on the internet.
The story of Assassin involves a young girl Nie Yinniang (Shu Qi) betrothed at 10 years-old to her cousin Tian Ji’an (Chang Chen) in Weibo, a province on the Northern edge of the empire. When a change in family policy prevents the marriage, the girl is taken to the imperial capital by her aunt who trains her as an assassin to serve the empire. Thirteen years later the young woman ‘fails’ to complete an assassination task and her aunt sends her back to Weibo with orders to kill her cousin, now the governor of the region and becoming a threat to the centre. The main part of the narrative deals with what happens when Yinniang clashes with her cousin.
Taiwanese director Hou Hsiao-hsien hadn’t made a feature since Le voyage du ballon rouge, a co-production with France in 2007, when he completed Assassin in 2015 and then won the Director’s prize at Cannes. Before 2007 he made two other films which got distribution in the UK – Three Times (2005) and Café Lumière, (2003) both also co-productions with France and Japan respectively. Before 2003, Hou’s work was quite difficult to see outside East Asia despite his status as one of the most important auteurs in global cinema. (His earlier films in the late 1980s were shown in the UK but have not remained in print.) As a consequence, I suspect some of the reviewers faced with Assassin had little context in which to try to ‘place’ his Cannes prizewinner. To confound critics further, Hou had not previously made a film set in the far distant past, so when he announced his interest in adapting a 9th century tale from the Tang period and exploring the wuxia or martial chivalry genre, a lot of blind alleys seemed to open up.
In many ways, approaching the film as a wuxia seems to me if not a ‘mistake’, at least a ‘problematic’ enterprise. For most viewers in the West, wuxia is only familiar through the work of a handful of filmmakers, most of whom are auteurs like Hou. Crouching Tiger, Hidden Dragon by Ang Lee and Hero by Zhang Yimou are the two most widely-seen examples of films with strong elements of wuxia and, even so, neither film is fully satisfying to Chinese fans of the genre. Wuxia implies a ‘period setting’ and a conflict which at its centre concerns the opposition of distinct groups or individuals who practice a school of martial arts – i.e follow a specific teacher and a code of honour. The action sequences will take place in what is known as jianghu. This term seems to have several meanings, but all of them suggest a different, alternate fictional world in which there are different ‘rules’ and identities and in which martial actions are directly linked to philosophical and spiritual questions. (A detailed discussion of jianghu and the elements of wuxia is included at the end of my notes on Hero.) Rayns (2016) suggests that the whole world of the Tang dynasty might be seen as jianghu in Hou’s envisioning of the period. What is certainly true is that there is a profound contradiction between Hou’s approach to the staging of the historical period and his use of certain familiar wuxia elements.
Wuxia narratives (popular in novel form as well as films – Hou seems to have remembered the novels of his youth rather than the films of the great Taiwanese master King Hu) feature the jianghu which can include super-powers for the warriors. This is famously represented by wire-work choreography that allows actors to fly or to leap up into a tree or on to a roof where swordfights can be staged in spectacular fashion. These warriors have sword skills that enable them to deflect arrows and athleticism to dodge flying blades. They can shoot arrows that split hairs etc. The jianghu also includes the possibility of the supernatural with ghosts and witchcraft. All of these elements are present in Assassin, but they sit alongside an intensely realist presentation of the ‘real world’. Hou’s inspiration for the some of the military scenes and also of the remote villages in Weibo is in the work of Japanese filmmakers and especially Kurosawa Akira’s approach to the production of Seven Samurai (Japan 1954). This approach relies on getting the historical details correct as far as possible:
I wanted to try my hand at the genre [i.e. wuxia] one day – but in the realist vein which suits my temperament. It’s not really my style to have fighters flying through the air or doing pirouettes on the ceiling; that’s not my way, and I couldn’t do it. I prefer to keep my feet on the ground. The fight scenes in The Assassin refer to those generic traditions, but they are certainly not the core of the drama. All else aside, I have to think about my actors. Even with protective padding and other safety precautions, even using wooden swords, such scenes are necessarily violent . . . Actually, the biggest influences on me were Japanese samurai films by Kurosawa and others, where what really matters are the philosophies that go with the strange business of being a samurai and not the action scenes themselves, which are merely a means to an end and basically anecdotal. (Hou quoted in the Assassin Press Notes)
It’s possible to see the problems for some critics (and even more so for some distributors) in this apparent contradiction. Hou seeks out the realist presentation and eschews too much reliance on action – which for many fans is the major attraction of wuxia. Comparisons with Zhang Yimou’s wuxia films are interesting because Zhang too is interested in those ‘philosophies’, but where Zhang stages the narratives in often spectacular settings – large palaces, hundreds of extras etc. – Hou chooses much more intimate settings – small palace chambers, clashes between groups of a dozen or so warriors etc. Hou also selects to use ‘narrow’ screen shapes – Academy 1:1.37 for the prologue (in monochrome) and something slightly wider for the main film (I thought 1:1.66 but IMDB says 1:1.41, which I’ve never come across before) with at least one insert of 1:1.85. Hou also favours long takes featuring a static or a slowly tracking camera. He doesn’t create the sense of movement with the camera or edits – only with the moments of swift movement by the actors within the frame. For much of the time, the principal character Yinniang waits quietly in the shadows, observing the scene before she acts. As a consequence, some audiences find the film ‘boring’ or ‘uninvolving’. Against this, many scenes are breathtakingly beautiful. Hou travelled to remote areas in Inner Mongolia and Hebei to find the silver birch woods, mountains and streams that become the ‘authentic’ settings for his story. Even with my limited knowledge of Chinese visual arts, I recognised the emotional power of the settings. The beauty of the settings is enhanced by Lee Ping-bing’s cinematography. A long-time collaborator with Hou, Lee uses monochrome and colour in startling ways creating a palpable texture for images featuring rain and mists. I was sat quite close to the screen and sometimes there was a high level of grain in the image and at other times the image seemed processed. There were also some very subtle shifts of focus in some of the long shots of figures moving through landscapes. As far as I can tell, Lee shot most of the film in 35mm (except perhaps for the monochrome prologue – on 16mm?). It’s frustrating that I haven’t as yet found any further details online. The interior mise en scène is just as meticulously constructed with costumes and sets designed by Hwarng Wern-Ying. Again the historical detail is more important than any melodrama excess but Yinniang often observes from behind curtains, gauzes etc. which match the mists in the exterior scenes.
Thinking about Assassin in relation to the films of Zhang Yimou, I remembered that Hou had been one of the producers of Zhang’s Raise the Red Lantern (1991), a film that features intrigues between the wives/concubines inside a war-lord’s house – itself a carefully constructed setting. Zhang also sought out new and spectacular settings for his second wuxia, The House of Flying Daggers (2004). Like Crouching Tiger, Hidden Dragon, Flying Daggers is a wuxia romance with female warriors in central roles and this is a description that might fit Assassin. However, it is another Zhang Yimou film that seems most relevant to me. The Curse of the Golden Flower (2006) has a similar structure of brief moments of action set between what is effectively a power struggle within a royal household. It’s much more spectacular than Assassin but the importance of the intrigue and the conflict of family ties and real politik is similar.
Hou’s final trick in making life more difficult for the audience – and adding layers to the intrigue – is to use another story told by a character as a kind of key. This is the story about the bluebird given as a gift. The bird fails to thrive until someone suggests that a mirror is put in the cage and then the bluebird revives, singing and dancing to its own reflection. Here is the clue to both the script and casting decisions. Many of the characters are ‘doubled’ and the casting and costumes/make-up seem to deliberately attempt to confuse the viewer – they certainly did for me. Thus it isn’t easy to distinguish between the wife and the concubine of Tian Ji’an and similarly Tian himself is sometimes easily confused with his officers. My first task when I re-watch the film will be to make sure I know who is doing what to whom.
Whatever my problems following the narrative, I have no doubts that this will be one of the most interesting films I will see this year. And I haven’t even mentioned the music by Lim Giong which also needs more of my attention. I’m sure I saw a reference in the credits to music from ‘Dakar’ (in Senegal?). I must find out more. Trailers can never possibly convey the pacing or complexity of a film like Assassin but you can get to see some of the beauty and some of the features outlined above in this trailer:
Hou Hsaio-hsien is a case study director in Chapter 11 of The Global Film Book.
The cinema audience in China is expanding very quickly. The final 2011 audience figures are likely to confirm the extraordinary growth rate shown in 2010 when four new cinema screens were opening each day. Since most screens are now in new multiplexes, often with 3D/IMAX possibilities, it isn’t surprising that locally-produced blockbusters compete on very favourable terms with Hollywood’s tentpole releases (which have restricted access to the market with only so many titles allowed and only at specific times).
Detective Dee: The Mystery of the Phantom Flame (HK/China 2010) was one of the big hits locally and regionally in East Asia in Autumn 2010 and it makes an interesting case study of the new Chinese commercial cinema. Like most such films it is a co-production between Hong Kong and China. Although mostly filmed in mainland China from a mainland story and script, the producer-director (Tsui Hark), action director (Sammo Hung) and three of the five acting leads (Andy Lau, Carina Lau and Tony Leung Ka-fai) are from the Hong Kong industry. The central character, Detective Dee, is based on a historical figure from the Tang Dynasty (7th century AD), Di Renjie, who was then fictionalised in a series of books by a Dutch sinologist, Robert van Gulik. Several American commentators have likened the film to Guy Ritchie’s take on Sherlock Holmes, i.e. as a mystery/thriller set in a historical period but played like a modern action film.
Outline (no spoilers)
In the year 679 AD (in European terms) the dowager Empress Wu Zetian has been the de facto power in the land for several years, often dealing harshly with any resistance to her plans. Now she intends to be crowned as Empress in her own right. The coronation is intended to take place after the completion of a giant 200 feet high Buddha. But when a foreign emissary is being shown around the interior of the soon to be completed Buddha things start to go wrong. Two senior officials internally combust without warning. Is the Empress in danger. She decides to send for ‘Detective Dee’ currently incarcerated in a maximum security prison. His release from prison attracts a gang of assassins who take their own lives rather than accept capture. With two reluctant assistants, the female bodyguard of the Empress and a senior officer of the Supreme Court, Detective Dee gets to work to solve the mystery.
I was interested in the genre mix here and how it has been interpreted outside China. The ‘meta-genre’ or ‘broad category’ is the historical action picture. I don’t know if there is an equivalent Chinese term for the Japanese concept of the jedai-geki or ‘period film’ but many Chinese and Korean films come into this category. Detective Dee seems to fall somewhere between the classical wuxia or martial chivalry film and the ‘kung fu’ film from Hong Kong. I take the former to be more ritualised and the latter to be more flexible and applicable across categories. I’m conscious that this may be a false distinction – any help gratefully received! Some of the Hong Kong films that I have seen add supernatural elements to the mix (A Chinese Ghost Story, 1987 – produced by Tsui Hark– and The Bride With White Hair, 1993). The latter film stars Brigitte Lin and Leslie Cheung and is one of several romance action dramas with strong female figures. Sammo Hung was also responsible for a sub-genre of kung fu zombie/vampire films starting with Encounters of the Spooky Kind, 1980. Detective Dee features supernatural elements which the detective must investigate in terms of rational explanations. The Chinese actor Li Bingbing plays the female bodyguard and thus for Western audiences is a reminder of the characters played by Michelle Yeoh, Zhang Ziyi and Maggie Cheung in the wuxia romance films Crouching Tiger, Hero and House of Flying Daggers. However, in Detective Dee, there is little or no romance element and only a frisson of erotic excitement in the first encounter between Li and Lau. Instead, the film plays on the mystery element. Perhaps surprisingly given the Sherlock Holmes comparison there isn’t a lot of comedy amongst the action scenes. Some commentators have suggested that Andy Lau, arguably the biggest male star in East Asia, is rather wasted in the film as he doesn’t have to be witty or prove that he is a heartthrob.What he has to do is think, fight and remain a man of principle. This makes Detective Dee rather different from the typical Hollywood action hero, creating possible barriers for Western audiences. The title too might be a problem. In China it was Di Renjie but the English title suggests something like Pirates of the Caribbean or Harry Potter.
If my genre description makes the film sound dull, fear not. It isn’t dull by any means. Hark doesn’t let the pace slacken over two hours. Any scene that requires exposition to set up/move on the mystery is quickly followed by a highly choreographed and spectacular action sequence courtesy of Sammy Hung and wire-work. There are also moments when you want the action to stop to allow time to take in the incredible sets and the CGI work, especially around the imperial palace and the Buddha statue. I’m no expert on CGI, but it looked OK to me, certainly no worse than Hollywood, Bollywood or South Korea. (And it’s certainly an achievement on a film budgeted at around US$20 million.) It does lend an air of fantasy to the film however. It’s interesting to compare the settings to those depicting Alexandria in Alejandro Amenábar’s Agora (Spain 2009). Amenábar’s understated realism still manages to represent the spectacular whereas Hark’s film has some very dodgy touches such as a harbour full of galleons, seemingly a Western European design of the 16th century. Added to this are the foreign emissaries who visit the court from Umayyad Caliphate speaking in Spanish rather than Arabic. But none of this matters. This isn’t a realist or aesthetically exact art film. It’s an entertainment and as such it works very well – though it could sometimes do with a touch of lightness. It’s sad that Leslie Cheung isn’t around as he was in the 1980s and 1990s. It’s also sad that because previous Chinese blockbusters like John Woo’s Red Cliff haven’t drawn UK audiences to cinemas, more recent films like Detective Dee are only available on DVD. They cry out for a big screen showing and, who knows, this might be the start of a franchise.
One last thought. There is an obvious parallel with Zhang Yimou’s Hero at the level of the ideological. Hero was widely criticised, especially in the US, as a film that legimitised the Chinese government through its implicit celebration of the formation of the Chinese state through conquest. Detective Dee similarly asks the question – should the Empress be saved even though she is cruel and ruthless because she represents stability and the promise of prosperity for the people?
UK English subtitled trailer: