Light My Fire:
The Geology and Geography of Film Canons

  film canons

When I was invited to the Buenos Aires International Festival of Independent Cinema at the start of 2001, friends asked me: “What kind of festival is it?” I was truthfully able to reply: “They’re showing Béla Tarr’s seven hour Sátántango, and also the British comedy Kevin and Perry Go Large” – the latter helpfully translated for Spanish-speaking audiences as Sex in Ibiza.


Now, I have to tell you that, in the Australia of 2001, Sátántango (1993) has yet to be screened, even at a Film Festival; in fact, no Tarr film appeared there until The Werckmeister Harmonies (2000). Kevin and Perry Go Large (2000), on the other hand, had a commercial release in many cinemas, but I think virtually every film reviewer in the mass media voted it among the worst films of the year, because it was so crude in every way. In fact, I saw a damning Australian review of this film calling it “one of the worst movies ever”, with “an audience of zero”. (1)


I found this violent comment rather poignant. Suddenly, Sátántango and Kevin and Perry were linked in my mind – since both films were deemed to deserve an “audience of zero” in Australia. Needless to say, I was also then struck by a passionate conviction: I had to see both films.


Putting together two such different films in this kind of way is something I associate with a certain kind of film critic, a certain kind of film criticism, and a certain kind of film festival. Critics who are truly cinephiles, I believe, often champion extremes. They go for the highest and the lowest. They champion the most difficult, severe, rigorous, minimalist, experimental films; and, equally, they also champion the often despised, maligned and overlooked products of popular culture – like vulgar teenage comedies, gross horror, trashy exploitation, ultra-violent action, even pornography.


At both extremes, cinephile critics look for excess and intensity. A piece of their aesthetic credo is summed up in the words of critic Paul Willemen, who once proposed “frenzy, madness, neurosis, extravaganza, monstrosity, etc” as “positive values” in a work of art. (2) What such critics usually do not like, on principle, is a certain middle-of-the-road, middlebrow cinema – or, more exactly, a middle-of-the-road taste in cinema, safe and predictable, between those two extremes of the highest and lowest.


The topic today is canons, and what it would mean to propose a new canon. I am going to talk about canons, critics and lists. So, first, I will make a distinction between a canon and a list. At first glance, the difference is pretty clear-cut. A list is something you or I could compose at the drop of a hat: a list of our favourite films, or the ten best films of all time, or the best films of a country or a decade or a genre. I like it when people write on the top or bottom of such lists: “If you ask me tomorrow, my list will be different”. Why would it be different? Because the person would be different – their opinion, memory, reverie about cinema would change.


Lists tend to be very personal – individual, idiosyncratic, eccentric. But once anybody’s private list goes public, it can have a social or cultural effect. Circa 1995, when I wrote a fan letter to Jonathan Rosenbaum, I told him that a list he published in 1976 in Film Comment magazine of his favourite films and critical texts became my guide, at the age of 16, as to what films I felt I needed to see and what I needed to read about them. (3) In fact, I have not yet seen all the films on Jonathan’s old list – so I still have something to strive for. And now I also have other lists to follow, like Nicole Brenez’s lists of the best films of 2000, the best films of the ‘90s and the best films of all time, which nominate films such as Jonas MekasAs I Was Moving Ahead Occasionally I Saw Brief Glimpses of Happiness (2000) and Baise-moi (2000) alongside popular successes like Crouching Tiger, Hidden Dragon (2000).


A canon is very different from a list. A canon, as I imagine it, has the weight of impersonal, collective, institutional authority, like a law engraved on a tablet that Moses brought down from some mountain. A canon is always the result of a broad survey of serious opinions. It can often strike us as a bit stuffy, a bit dutiful, a bit dead. Canons sanctify and legitimate the greats, the classics, the perfect films. But here’s my big problem with such canons: something of the heart and soul of cinema, of the passionate experience, the passionate encounters each of us have with cinema, get erased from them. The filmmaker Bertrand Tavernier put this well when he said: “I don’t much like these [canonical] lists: too many beautiful and important films are missing, and they leave out the texture, the richness and life of cinema by not including all those ‘imperfect’ films which are more meaningful and alive than frozen, dated ‘classics’”. (4)


However, we have to note something strange and curious about film canons. There is not one, single canon that holds absolute sway in any country at any time. Rather, every now and again, a canon is proposed, put up by a magazine, a museum, a Festival, a Cinémathèque, a national Film Institute. And how is this canon put together, from where does it come? Of course, it usually comes from many individual lists. Canon-formation (as it is called in cultural studies) is a process of polling, of consensus. Many different, individual responses to cinema are sifted, and what comes to the surface is a canon. The problem is that this consensus list often seems like a bloodless compromise – or as Jonathan says in his book Movie Wars, a “corny hit parade”. (5)


I’ve been looking at about ten different canon-polls from various parts of the world. The same rough list of films keeps coming up: Citizen Kane  (1941), La Règle du jeu (1939), Vertigo  (1958), Tokyo Story (1953), Raging Bull  (1980), Lawrence of Arabia (1962), and so on. And of course, I love at least a few of these titles myself. But why do we need to affirm just this particular hit parade, over and over?


Actually, I think there are three major kinds of canons that circulate in global film culture. The first kind I won’t talk about for long because, frankly, it disgusts me. It is the supposedly populist canon, the people’s canon, and it’s based purely on box office and commercial success; in other words, on money and business. Naturally, it is massively biased towards blockbuster Hollywood product. This is the canon with Star Wars (1977), Jurassic Park (1993) and The Sound of Music (1965) on it. In fact, let’s call it, for ease of reference, the Star Wars Canon.


Then there is, for want of a better term, the old canon – the old guard, old-fashioned canon. This has on it De Sica’s Bicycle Thieves (1948), Kuosawa’s Rashomon (1950), Eisenstein’s Battleship Potemkin (1925), Fellini’s La Strada (1954) and Satyajit Ray’s Pather Panchali (1955), among others. This is a very respectable canon, reflecting the first great era of discoveries on the Film Festival circuit of the 1950s, and then the grand period of arthouse cinema, the New Wave years of the 1960s.


The aesthetic and political values of this canon tend to be humanist, naturalist and social-realist. It champions films it perceives as timeless, universal and noble. It disparages what it perceives as mere formalism or style without substance – so, to take a historical example, thhis canon always revered Italian neo-realism but had a big problem with Michelangelo Antonioni. In short, it is hard to avoid the conclusion that there is something terribly middle-of-the-road about the old canon – which I will call the Citizen Kane Canon.


I would expect that a new canon would try to correct some of the problems, absences, biases and exclusions of existing canons. Unfortunate problems like these:


* Canons massively favour the feature length format, and exclude short films.


* Canons massively favour narrative films, and exclude documentaries.


* Most canons have a heavy bias towards the classics of American cinema, because they reflect a long-ago period in (mainly Western) film culture before Asian cinema, Indian cinema, Iranian cinema, and so forth, finally broke into some people’s consciousnesses.


* Canons favour drama over comedy, just like at the Academy Awards. Buster Keaton always hovers just outside the established canons (while Chaplin sometimes scrapes in as timeless, universal and noble), while Jerry Lewis is a complete outcast – and that’s a crime.


* Canons have little regard for the achievements and traditions of many popular genres. At best, you might find one musical (usually Singin’ in the Rain [1951]), a token horror movie by a great auteur (like Alfred Hitchcock’s Psycho [1960]) and a single, exceptional, science fiction classic (Stanley Kubrick’s 2001: A Space Odyssey  [1968]). But there is so much more in the great sea of pop culture, from Louis Feuillade serials to Amy Heckerling’s Clueless  (1995).


* Virtually all forms of avant-garde or experimental cinema are banished from canons – which means, for example, that the best women filmmakers in cinema history, like Maya Deren and Chantal Akerman, are rarely honoured in such surveys.


* Canons like to take refuge in the past, and flee from the challenges of the present. Some canons are happy to shut up shop with Raging Bull in 1980.


* Canons favour an organic aesthetics – they valorise whole, entire films as perfect objects. This leaves no room for imperfect films, or brilliant bits or fragments of films. And we all know there are many films that are great for just ten minutes, maybe just for one scene.


* Canons valorise singular masterpieces over bodies or corpuses of work. But there is no single great masterpiece to be plucked from the careers of many important and influential directors, including Fassbinder, Pasolini and Preston Sturges.


… In short, canons simply ignore too many good, important, significant and pleasurable movies. But what, realistically, is the alternative? I recall a story by Jorge Luis Borges in which fanatical mapmakers create a map of their territory so precise, true and big that it covers the whole territory. And I sometimes think that this is what the best canon of cinema would be: the cinema itself, whole and entire.


But, of course, nobody’s grasp of cinema could ever actually happen in that kind of impossibly holistic, total way. We can’t just go out and see everything, alas! What any of us understands of cinema depends, at every moment, on material, social and political factors: what films are available, or lost; what films are in circulation and in which format (celluloid, video, DVD, etc); what decisions are being made by those in power concerning which available films audiences will actually get to see, where and for how long.


Jean-Luc Godard said that his TV/video series Histoire(s) du cinéma  (1989-99) is an investigation of what he called “the geology and geography” of cinema’s history and forms. (6) He meant that cinema not only spreads out around the globe, but also compacts itself into complex layers of visibility and invisibility. One could also trace or write a history of film canons according to these principles of geology and geography.


Film theorist Laura Mulvey puts this well: “No list can hope to do more than capture the level of research and aesthetic spirit of its time” (7) – and I would add, its place. Any canon is tied to the taste and the cultural agenda of its time and place. We constantly see, throughout the history of film canons, certain works and artists going wildly in and out of fashion. Some filmmakers have to wait a long time for their rendezvous with history, like Douglas Sirk or Michael Powell. Hitchcock’s Vertigo was not considered such a deathless masterpiece by so many people even twenty years ago. Other folk lose their footing in film history and are cast into a cruel oblivion, like (at least in some countries) Ingmar Bergman – at least until he ups and dies.


In terms of the dynamic and constant historic revision of canons, one must not underestimate the importance of public agitation – the kind of sudden awareness that can be prompted in an audience or community by an inspiring critic, teacher, film festival or Cinémathèque event. The language of the promotions or marketing industry is not so bad for describing this phenomenon: it’s a matter of a buzz or vibe, a small cult around a film or body of films. Look at what happened with the international, touring retrospective of Robert Bresson at the end of the 1990s. If such an event gets the level of cultural support it deserves from the media, suddenly young cinephiles and filmmakers everywhere discover Bresson as if he’s the newest, coolest, most exciting thing in the world – which he indeed is, and will always be.


So I have come back to the role of critics, and also to the individual list as opposed to the consensus canon. I’ve already mentioned the empirical reality that no one can really hope to see or cover all cinema. I find charming those enormous books (such as Jacques LourcellesDictionnaire du cinéma: les films – 1,725 pages of small print!), (8) or Internet sites by obsessed film maniacs who obviously believe they can establish a master-canon, and manage to see every important, great achievement in cinema history. But that is really a lost illusion from an old school of cinephilia.


Beyond all the obvious, material difficulties in getting to see and experience films, there is also a personal – I would even say existential – dimension to this dilemma. Life is short! None of us will ever watch all the films, read all the books, see all the art or hear all the music we think we should. We are all drowning in culture, stressed and overworked. So what do we want, what do we need?


Personally, I want a list, a crazy new canon, something that communicates the intensity and passion of someone else’s experience, that inspires me and lights my fire. Every one of us creates, in our own story of moviegoing experiences – whether we write, talk or just think about it – our unique history of the cinema. That is what I believe film criticism is about: multiplying these histories, bearing witness to them, deepening them. I’m reminded of a story told by André Weil, brother of the great, mystic philosopher Simone Weil:


She was once describing to me (…) some historical theory, or a historical fancy (…) I said to her: “This is a historical question. It must be discussed in terms of the evidence. What is your evidence for what you are saying?” She said: “I don’t need any evidence. It is beautiful, therefore it must be true”. (9)


It’s that sort of beauty and truth that I personally admire and seek out in other people’s wild histories, canons and maps of the cinema.


So, what would a new canon, an alternative canon (as it is sometimes called today) look and feel like, what would it try to do? Well, for starters, I think a new canon would embrace those extremes of cinema I mentioned at the outset: avant-garde cinema, cinema povera, underground cinema, exploitation cinema. A new canon would reclaim the overlooked, the unfairly rejected, the suppressed. Above all, the new canon would be a sustained argument, a long and bloody battle with the old canon. And since I named the mainstream canon after Star Wars and the old canon after Citizen Kane, what could I name the new canon after? What would definitely be on this list, as a flagship? A film by Kiarostami or Hou Hsiao-hsien? Abel Ferrara? John Cassavetes, the archetypal maverick/independent? Sergei Paradjanov or Ritwik Ghatak? Philippe Garrel or Raúl Ruiz? Stan Brakhage, Maya Deren, Jon Jost? Rosetta  (1999), Beau travail  (2000), Eyes Wide Shut   (1999)? Maybe a different Welles, not Kane but Touch of Evil (1958) or The Magnificent Ambersons (1942)? Perhaps we should put on an old master who is in danger of being forgotten in certain parts of the world, like F.W. Murnau; or an old master still to be discovered by most of us, like Boris Barnet?


Any of these choices might fit the bill for a new canon – alright, let’s call it the Kiarostami Canon – admirably. But in this realm, I think it’s healthy and important that there is precisely no consensus, and as little sifting or filtering as possible. In the original Movie Mutations dialogue that I did with Jonathan Rosenbaum, Nicole Brenez and other comrades, (10) I expressed my reservation about the latest kind of art-purism that sometimes emerges today, the sense that it’s only the new greats – the visionaries like Kiarostami or Tarr or Hou or Malick – who really matter, who must be elevated above all else at all cost. But cinema matters as much for what happens in its dirty valleys as in its lofty peaks. (Nicole asked later: Why say dirty? These valleys are dark, obscured, not dirty.) That is why we must also canonise films like I Spit on Your Grave (1978), Street Trash (1987) or Baise-moi.


A new canon, because it is an argument with the old canon, is obviously polemical. It announces an agenda, a program for action – it can even be a manifesto. It cries: this is what you have overlooked, this is what our culture must at last confront, these are pleasures and terrors beyond what is comfortable and familiar! Polemic is a more heated, more violent form of pedagogy. And I do think that canons worth anything, new or old, are pedagogical – which is nothing to be ashamed of.


In a canon, critics and programmers, film festival or Cinémathèque directors take on the authority of being public teachers. They stand up and shout: this is what you, the audience, must see, what you must know, what you must experience! The commercial Star Wars canon, on the contrary, isn’t about pedagogy; it’s only about confirming what the mass of people already see and know and do.


So, both the old Kane canon and the new Kiarostami canon are pedagogical, but here is a crucial difference between them: the middle-of-the-road canon tends to mummify or ossify film classics. It is too reverential. Everyone knows that when young students sit down to watch a film thinking, “Oh, this is some old classic that I am supposed to appreciate as a great work of art”, they are already dead to what that film can offer them; they’re shut down, closed off and bored before the projector even beams.


I’ve often thought that film culture (in teaching, for instance) should be all about revitalising these dead classics, releasing the sensational affect, the erotic or violent charge, the radical force that is waiting there in Vigo’s L’Atalante  (1934) or Battleship Potemkin or, indeed, Citizen Kane. That would constitute an act of trying to get these old films out from the middle of the road – alongside trying to move ahead on many roads simultaneously to embrace and understand what’s new, right now, and what’s coming in cinema’s future.


The organiser of this forum asked me for my own list, my own new canon. Composed randomly and completely spontaneously, it is not the list of my ten personal favourites, or even the ten I would consider the best or most significant or most representative in cinema history. My list is simply my agenda, my proposal, my polemic, weighted towards the recent past. Of course, it comes with that all-important proviso: tomorrow, the canon will be different. So here is my new canon, just for today:


1. The Killing of a Chinese Bookie (John Cassavetes, 1978)

2. Alone. Life Wastes Andy Hardy (Martin Arnold, 1998)

3. L’Atalante (Jean Vigo, 1934)

4.  The River (Tsai Ming-liang, 1997)

5. Haut bas fragile (Jacques Rivette, 1995)

6. The Night of the Hunter  (Charles Laughton, 1955)

7. Phenomena (aka Creepers, Dario Argento, 1984)

8. The Emperor’s Naked Army Marches On (Kazuo Hara, 1987)

9. The Spirit of the Beehive (Victor Erice, 1973)

10. F for Fake (Welles, 1973)



This paper was first presented at the 3rd Buenos Aires International Festival of Independent Cinema, 20 April 2001. More extensive attempts by me at formulating a personal canon can be found here.



   1. Kerrie Murphy, The Australian, 19 April 2001, “Media” supplement. back


   2. Paul Willemen, in Claire Johnston and Willemen (eds), Frank Tashlin (Edinburgh Film Festival, 1973), p. 17. back


   3. Jonathan Rosenbaum, “My Favourite Films/Texts/Things”, Film Comment, November-December 1976, pp. 51-7. back


   4. Bertrand Tavernier, 360 Film Classics, special supplement to Sight and Sound, 1998, p. 36. back


   5. Jonathan Rosenbaum, Movie Wars: How Hollywood and the Media Conspire to Limit What Films We Can See (Chicago: a cappella, 2000), p. 96. back


   6. Quoted in Michael Witt, “Montage, My Beautiful Care, or Histories of the Cinematograph”, in Michael Temple and James S. Williams (eds), The Cinema Alone: Essays on the Work of Jean-Luc Godard 1985-2000 (Amsterdam: Amsterdam University Press, 2000), p. 34. back


   7. 360 Film Classics, p. 36. back


   8. Jacques Lourcelles, Dictionnaire du cinéma: les films (Paris: Robert Laffont, 1995), p. 36. back


   9. Simone Weil, Gateway to God (Glasgow: Fontana, 1975), pp. 157-8. back


10. Cf. Martin and Rosenbaum (eds.), Movie Mutations: The Changing Face of World Cinephilia (London: British Film Institute, 2003). back



© Adrian Martin April 2001/February 2008

Film Critic: Adrian Martin
home    reviews    essays    search