Film canon

Film canon is the limited group of movies that serve as the measuring stick for the highest quality in the genre of film.

Criticism of canons

The idea of a film canon has been attacked as elitist. Thus some movie fans and critics prefer to simply compile “lists”. But canon simply means “the best” and proponents of film canon argue that it is a useful exercise to identify and experience a limited number of the “best” films, if only as a starting point for film students. All writing about film, including reviews, arguably serves to create a film canon. Certainly, some film canons can be elitist, but others can be “populist”. For instance, Paul Schrader's essay on film canon includes his personal canon of sixty movies. [Paul Schrader, “Canon Fodder: As the sun finally sets on the century of cinema, by what criteria do we determine its masterworks?” "Film Comment" (September/October 2006):33-49.] He openly describes his canon as “elitist” and argues that this is a virtue. A more populist film canon might be a poll of readers in a film magazine, or an internet poll on a website.

In literature

In literature, the “canon” (presumably the academic canon) has been attacked as dominated by “dead white European males”, and one might attack typical film canons on the same basis. (For instance, a feminist might disagree with many personal film canons because few women directors are represented in them.) However, literary critic Harold Bloom’s arguments for the traditional literary canon (in his "The Western Canon; The Books and School of the Age" (1994)) can be applied to film also. He argues that works must be evaluated on their aesthetic merits, not on the basis of politics. Those disagreeing with Bloom have argued that ethics is a central component of great art, and that politics is often ethics on a broad, social level.

Another problem that faces all aesthetic canons is: what person, persons or groups of persons selects that shortlist of ultimate movies or works of art? A critic? A group of critics? Librarians? Academics? How many academics (10 or 50 or 500)? Fellow artists (who know the craft better than anyone else; but who can be narrowly critical of work competing with their own)? Pure popularity? If the latter, chronology must be factored in, for many works of art are very popular for a brief period, then are totally forgotten. (And some movies make little impact upon release, but then build in reputation over time).

In fact, there is no absolute answer to any of those questions. There is no absolute film canon. There are many canons, that change over time, from country to country. The fact that they all directly contradict each other makes them even more useful. Personally, we may each trust some canons above others; we may hold to the canon of André Bazin or Pauline Kael over Jonathan Rosenbaum and Manny Farber, or vice versa. We may trust collective polls of critics or directors, such as the "Sight & Sound" polls. We may respond more to canon based on great directors (in film we might call that an "auteur" canon) than on canon based on specific films. Some individuals will construct their canons based on films he or she regards as great or influential; others will formulate their canons based on pure pleasure, selecting films that they enjoy watching repeatedly. Some may create a canon as a conscious rejection of the movies that have been widely regarded as great or influential.

Two main methods of assessing film canon are canon from individuals (critics, directors or film archivists) and canon from collective polls (in which a number of individuals are surveyed).

Film canon from individuals critics or directors

A critic or director may write a book about film in which he or she discusses many of his or her favorite, most-admired movies. A critic or director may contribute a top ten list to a magazine or poll. A director may give interviews in which he outlines the movies that have influenced him most. In all these cases, an individual is arguably, if informally, creating a personal film canon. So the following articles and books are examples of more or less formal personal canons: James Agee, "Agee on Film: Criticism and Comment on the Movies" (1958); Manny Farber, “White Elephant Art vs Termite Art” (1966); André Bazin, "What Is Cinema?" (two volumes, 1967 and 1971); Pauline Kael, "5001 Nights at the Movies: A Guide from A to Z" (1982); Orson Welles and Peter Bogdanovich, "This is Orson Welles" (1992); Tom Lasica, “Tarkovsky's Choice” (1993); Cameron Crowe, "Conversations with Wilder" (1999). Popular critic Roger Ebert has written "The Great Movies" (2002) and "The Great Movies II" (2005).

However, some critics, books and essays have special connections with canon. One influential book has been film critic Andrew Sarris’ "The American Cinema; Directors and Directions, 1929-1968" (1968), which has a section on fourteen “Pantheon Directors”, [These are Charles Chaplin, Robert Flaherty, John Ford, D. W. Griffith, Howard Hawks, Alfred Hitchcock, Buster Keaton, Fritz Lang, Ernst Lubitsch, F. W. Murnau, Max Ophuls, Jean Renoir, Josef von Sternberg, and Orson Welles.] and then sections on admirable but less important directors, and at least two sections on directors who are regarded as subpar to a certain extent (“Less Than Meets the Eye,” including David Lean, Carol Reed and Billy Wilder and “Strained Seriousness”, including Stanley Kubrick). This book thus has an "auteur" view of canon, with the style of the director and his body of work (not just one or two outstanding movies, but consistent achievement over a career) as the main criteria for inclusion. "The American Cinema" is consciously limited by its national focus and thus is not a holistic canon of cinema. (Sarris has written elsewhere that his nomination for the greatest movie ever made is Ophuls' "Madame de...", and this film is obviously outside of the focus of a book on American cinema, though Sarris discusses Ophuls’ American films. He also ranks Mizoguchi’s "Ugetsu" and Renoir's "The Rules of the Game" as the second and third greatest films ever made.) [Andrew Sarris, “The Greatest Film of All Time: Ophüls’ "Madame de . . ." Is Coming Back to Town,” "NY Observor" (March 12, 2007), p. 10. (Accessible online through cache.)]

Director/critic Peter Bogdanovich has written an essay entitled “Favorite directors” (1973) that proceeds from ten favorite directors to their best works. Again, this is substantially an "auteur" canon.

Critic Jonathan Rosenbaum's "Essential Movies: On the Necessity of Film Canons" (2004) argues that meditating on and creating film canons is necessary to offset the increasing commercialization of movies, where the public accepts or rejects a movie based on marketing campaigns, rather than because of the movie’s intrinsic value. He accepts Bloom’s argument for the necessity for canon, but disagrees with him in some ways. First, he rejects the limitation of “western” canon. Second, he accepts “informational” criteria in the formation of canon, not just aesthetic considerations, and argues that good art can make us “better citizens”. Third, he regards canon formation as “an active process of selection rather than a passive one of reportage.” Fourth, unlike Bloom who argues that canon formation is necessarily elitist and should not include cultural criticism, Rosenbaum holds that valid criticism can include cultural criticism, and that canon formation can be elitist or populist.

In the book’s appendix Rosenbaum provides a “Personal Canon” of a thousand movies, arranged chronologically, and a hundred of these are asterisked as the “crème de la crème.” He gives his criteria for inclusion in this canon as “pleasure and edification”.

Screenwriter/director/critic Paul Schrader has written an influential essay on film canon in "Film Comment" (September/October 2006) entitled “Canon Fodder”, in which he chooses twenty movies that he regards as canonical (with forty runner-ups). [Schrader's top twenty films are: 1. "The Rules of the Game" (1939, Jean Renoir); 2. "Tokyo Story" (1953, Yasujiro Ozu); 3. "City Lights" (1931, Charles Chaplin); 4. "Pickpocket" (1959, Robert Bresson); 5. "Metropolis" (1927, Fritz Lang); 6. "Citizen Kane" (1941, Orson Welles); 7. "Orphée" (1950, Jean Cocteau); 8. "Masculin, féminin" (1966, Jean-Luc Godard); 9. "Persona" (1966, Ingmar Bergman); 10. "Vertigo" (1958, Alfred Hitchcock); 11. "Sunrise" (1927, F.W. Murnau); 12. "The Searchers" (1956, John Ford); 13. "The Lady Eve" (1941, Preston Sturges); 14. "The Conformist" (1970, Bernardo Bertolucci); 15. "8½" (1963, Federico Fellini); 16. "The Godfather" (1972, Francis Ford Coppola); 17. "In the Mood for Love" (2000, Wong Kar-wai); 18. "The Third Man" (1949, Carol Reed); 19. "Performance" (1970, Donald Cammell/Nicholas Roeg); 20. "La Notte" (1961, Michelangelo Antonioni).] (The contrast of Rosenbaum's thousand and Schrader's twenty shows that even the number of films in a film canon is entirely arbitrary.) “As the sun finally sets on the century of cinema”, the article’s subtitle asks, “by what criteria do we determine its masterworks?” Schrader proposes seven criteria for formulating a film canon: beauty, strangeness (an “unpredictable burst of originality”), unity of form and subject matter, tradition (“The greatness of a film or filmmaker must be judged not only on its own terms but by its place in the evolution of film”), repeatability, viewer engagement (a great film “frees the viewer from” the passivity of the typical film-watching experience “and engages him or her in a creative process of viewing”), and morality (“no work that fails to strike moral chords can be canonical”). [Paul Schrader, “Canon Fodder: As the sun finally sets on the century of cinema, by what criteria do we determine its masterworks?” "Film Comment" (September/October 2006):43-46.] He rejects the idea of auteur canon, stating that “The film canon . . . consists of films, not people”, arguing that the moving picture medium is too collaborative to focus merely on the director. Schrader's personal canon excludes documentaries and non-narrative films, and he also eschews multiple films by the same director.

Film canon from collective polls

In 1952, the The Festival Mondial du Film et des Beaux Arts in Brussels polled 55 “cinema personalities, mainly directors,” and published the resulting poll in a pamphlet that was widely reported in film journals. Later in the year, the magazine of the British Film Institute, "Sight & Sound", decided to poll critics to see if they agreed or disagreed with the directors, and published the poll in its July-September issue. Collective film polls—of critics, directors, film archivists, readers, and film industry professionals—have continued through the century and into the new millennium. "Sight & Sound" has continued to poll critics (and in recent decades, both critics and directors) every ten years, creating a valuable record of change in critical taste over the years. The "Sight & Sound" polls have been admirable for their attempts at including a varied, international electorate (though the bias of any poll toward films in its own country and language is undeniable) and for the fact that they publish all of the ballots—the top ten lists of individual critics and directors, which provide a priceless record of many personal film canons.

While one might argue that the canons of individual critics and directors are most important for the formation of a holistic “film canon”, the interest and impact of these film polls have been substantial. Jonathan Rosenbaum has stated, in the introduction to "Essential Movies: On the Necessity of Film Canons", that his chance purchase of the 1962 "Sight & Sound" issue that contained the second critics poll was the most important event that contributed to his own personal film canon. Critic Roger Ebert has said that when students ask him which great movies to see first, he refers them to the "Sight & Sound" polls.

Film polls have all the variety of individual film canons. They can be consciously non-holistic (i.e., a poll may cover only French or Russian or American film); they can focus on the movie or the director or both, depending on how their data is presented; they can be elitist (with an electorate of a comparatively few critics) or populist (with an electorate of hundreds of critics or readers). Internet film polls tend to be populist: two examples of internet polls are the Internet Movie Database's [ [http://us.imdb.com/chart/top “Top 250 movies as voted by our users”] ] and the movie poll of the website “They Shoot Pictures, Don't They?”, [ [http://www.theyshootpictures.com/gf1000.htm “The 1,000 Greatest Films: As voted by 1,320 critics, reviewers, scholars, filmmakers and other likely film types.”] ]

A poll can consciously judge films by the criteria of historical importance (as is understandably the case of the 1995 poll of film archivists, FIAF, La Fédération Internationale des Archives du Film (The International Federation of Film Archives)) or by the criteria of pure “film pleasure.” A poll may be consciously biased; for example a group of poll-proprietors with a certain bias may choose an electorate of like-minded critics. Polls with particular biases can be useful, as long as the reader understands the bias.

The American Film Institute's 1998 “100 Years... 100 Movies” poll of American movies was controversial on a number of grounds. First, it sent ballots with a pre-selected number of movies to its electorate. Jonathan Rosenbaum, an outspoken critic of this poll, noted that his favorite American comedy, Chaplin’s "Monsieur Verdoux", was left off the ballot. Another classic, Whale’s "Bride of Frankenstein", also was inexplicably not included. Second, the AFI poll was tied to a marketing scheme after it was released. Third, in the publicity surrounding the poll's release, it implied that it was definitive.

A history of film polls, Todd Compton's "In Search of a Canon: Movie Polls Through the Years" (2007) (which tabulates ballots to expand many polls beyond the top ten or twenty and into a top 100) shows, if nothing else, that a general film canon cannot be created by film polls, because each film poll inevitably conflicts with every other film poll. At best, movie polls show how collective critical taste changes over time and by nationality. Thus, polls are useful, fascinating compendiums that will help in the formulation of personal canons (especially if they include individual ballots); but they do not create “the” film canon.

A major example of variation in the film polls is Vittorio De Sica's "The Bicycle Thief". In the first "Sight & Sound" critics poll, in 1952, this movie is ranked as the greatest movie ever made. It seems a solidly placed star in the cinematic firmament. Forty years later, the 1995 FIAF archivists poll ranks it at 8. However, the 1995 "Time Out" poll leaves the movie entirely out its top 100, and the National Society of Film Critics' 2002 “100 Essential Films” does the same. The 2002 "Sight & Sound" critics take it out of their pantheon, putting it at 45, but the 2002 "Sight & Sound" directors put it firmly in their pantheon, at 6. (One finds in the polls that there are “critics’ films” and “directors’ films.”)

Another example of poll variation is Federico Fellini. In most polls, he is high in the pantheon. In the 1992 "Sight & Sound" directors’ poll, there are two Fellini films in the top four ("8½" at 2 and "La strada" at 4) and three in the top 14 (with "La dolce vita" at 14). The 1992 "Positif" poll has three Fellini movies in its top 23. "8½" has appeared near the top of most movie polls since its release in 1963. However, "La strada" does not appear in the top 100 of the "Sight & Sound" critics poll in 1992, a stark split between critics and directors at the same moment in time. The 2000 "Village Voice" critics poll does not place "8½" in its top 100; curiously, no Fellini movie places in the top 100 of this poll.

Focus on older films in recent polls

A striking pattern in the film polls is that the older polls include in their top tens and twenties films that were recent at the time of the poll; but polls from 1980 on generally do not include films from recent decades in their top tens. For example, in the 1952 "Sight & Sound" critics poll, four movies in the top ten were from the previous decade, including the highest-ranked movie, "The Bicycle Thief", which had appeared just four years before the poll. Antonioni's "L'Avventura" was released in 1960; in the 1962 "Sight & Sound" critics poll, it ranked at number 2. Also in that poll’s top six were two other movies from the previous decade, "Ugetsu monogatari" (at 4) and "Ivan The Terrible, Part II" (at 6). However, in the 1982 and 1992 "Sight & Sound" critics’ polls, there was no movie from the previous decade in the top ten. In the top ten of the 2002 "Sight & Sound" critics poll there were two films from the 1920s, one film from the 1930s, one from the 1940s, three from the 1950s, two from the 1960s and one from the 1970s (the first two "Godfather" movies, 1972 and 1974, regarded as one movie). The most recent movie in this poll's film canon of ten is thus some thirty years old. In the second twenty, there are no films from the 1970s, 1980s or 1990s.

This phenomenon, which has been widely noticed and discussed, is perplexing to many critics and film historians. For example, Dan Sallitt wrote an article entitled “Sight Unchanged: How did the film canon get so stodgy?” in response to the 2002 "Sight & Sound" polls. This focus on the past, or fear of the present, in film polls has been explained in various ways. Some commentators have suggested that there was a period of cinematic florescence, a golden age, and that films in recent decades are simply not as good as they were in the 1950s and 1960s. This theory has been strongly rejected by critics who feel that contemporary cinema ranks with “classic” cinema. Another theory is that a canon of film (headed by usual suspects such as Welles' "Citizen Kane" (1941), Renoir's "Rules of the Game" (1939), and Eisenstein's "The Battleship Potemkin" (1925)) was put in place by a rough academic consensus in the 1960s or 1970s, and this “academic canon” has been frozen into place. Another theory is that critics are impacted most powerfully by films they saw in their youth, and that film criticism is a greying profession.

Australian film critic Adrian Martin regards “the Citizen Kane canon” (“the old canon—the old guard, old-fashioned canon”)—as uninspiring. [ [http://www.sensesofcinema.com/contents/01/14/light_my_fire.html “Light My Fire (The Geology and Geography of Film Canons)"] , "Senses of Cinema", Issue 14 (June 2001). This was taken from a paper originally presented on the "Canon 2000" panel at the 3rd Buenos Aires International Festival of Independent Cinema, 20 April 2001.] He suggests as a replacement what could be called a “Kiarostami canon”, referring to the great contemporary Iranian director Abbas Kiarostami.

Notes

ources

*Bloom, Harold. "The Western Canon: the Books and School of the Ages." New York: Harcourt Brace, 1994.
*Bogdanovich, Peter. “Favorite directors.” In Philip Nobile, ed., "Favorite Movies: Critics’ Choice". New York, Macmillan, 1973.
*Compton, Todd M. " [http://www.geocities.com/athens/oracle/7207/pollsTOC.htm In Search of a Canon: Movie Polls Through the Years] ". Mountain View, California: Magos Press, 2007.
*Lasica, Tom. “Tarkovsky's Choice.” "Sight and Sound" 3.3 (March 1993).
*Rosenbaum, Jonathan. "Essential Movies: On the Necessity of Film Canons." Baltimore: The Johns Hopkins University Press, 2004.
*Sallitt, Dan. [http://www.slate.com/?id=2069759 “Sight Unchanged: How did the film canon get so stodgy?”] "Slate" (Aug. 20, 2002).
*Sarris, Andrew. "The American Cinema; Directors and Directions, 1929-1968." New York, Dutton, 1968.
*Schrader, Paul. “Canon Fodder: As the sun finally sets on the century of cinema, by what criteria do we determine its masterworks?” "Film Comment" (September/October 2006): 33-49.