Reviewing the Greatest Films of All Time, part one: continuity and change

Eight international correspondents post-mortem the biggest-ever iteration of our celebrated film polls.

Sátántangó (1994)

Editor’s note: Following the unveiling in our September 2012 issue of the initial results of our new Greatest Films of All Time poll, Kevin B. Lee initiated the following email correspondence with several fellow film critics analysing the outcome of the poll, following up a prior discussion on Indiewire that anticipated the poll. We made further data from the nether regions of the poll (now completely online) available to the circle as we finished processing it, so the discussion was able to dig deeper as it developed. We’ve broken the conversation into four parts for publication over consecutive days.

From: Kevin B. Lee
To: Nicole Brenez, Dan Callahan, Bill Georgaris, David Jenkins, Jonathan Rosenbaum, Dan Sallitt
Subject: Opening Stat Attack

Welcome to our conversation in response to the results of the Sight & Sound 2012 Film Poll. It’s an opportunity to follow through on a three-part discussion some of us had prior to the poll on Indiewire Press Play where we raised questions and speculations over the poll’s significance that we can now revisit.

In the Press Play discussion I let it be known that my own relationship to the poll doesn’t lie in its ostensible aim to list and rank the very best films of all time (anyone who loves cinema and has spent a significant portion of their lives exploring it in depth knows the inherent absurdity of such an enterprise). Rather, I see the list’s true value as a road map for encountering key examples of what cinema can be, across eras, genres and geographies. The very notion of ‘best’ deserves to be challenged at its base by questioning the historical, cultural and societal factors that shape how we value the art of film.

This is partly why I organised my list to allow for no more than one film from a given decade or country, as a way to shake things up and spark a sense of reorientation – for myself as much as for the total results, which I had no hope of really influencing anyway given the number of participants. I’m interested in seeing as many different kinds of cinema as possible reflected in the outcome, as I think that would be of most help to the budding cinephiles who will devour this list the same way I did 15 years ago.

This takes me to one thing I’d like to ask each of you, to share how you came upon your final top ten. How long did it take for you to compose your list, what considerations came into play?

Of course a list is a reflection of those who make it, so when I heard that there would be nearly six times more participants in this year’s voting than in 2002, I hoped the results would indicate a broader sense of backgrounds in different kinds of cinema (for the record, I sent the S&S editors contact information for people I considered the best film critics in China, as well as a prominent Latin American film programmer). Looking at the top ten, I’d say the status quo held intact, despite the brouhaha of Vertigo unseating Citizen Kane at the top, a non-event for those who could see it coming from last decade’s close call.

Man with a Movie Camera was the only film in the top ten that had never placed there before. As someone who voted for it, I was pleasantly stunned to see this 83-year-old film rise so high (up from number 27 in 2002), and I think it reflects the prominence of nonfiction filmmaking in contemporary cinema. It’s the first time a documentary has placed in the top ten since Louisiana Story back in 1952; the total number of documentaries in the top 50 doubled from two to four. I also think Man with a Movie Camera’s ranking reflects the rise to prominence of non-linear, non-narrative cinema, thanks to digital filmmaking and editing. Perhaps the film appeals more to the tastes and practices of today’s filmmakers and audiences than it ever has.

I’ve compiled some other statistics from looking at the top 100 and comparing the results from those of ten years ago, listed below. One encouraging development is the increase in countries of origin among the top 50, from nine in 2002 to 14 this year, with first time placements for China/Hong Kong, Belgium, Hungary, Iran and Algeria.

In descending order: the US, France, Japan, USSR, Italy, Sweden, Denmark, China/Hong Kong, Germany, Belgium, Hungary, India, Iran, Algeria, UK
Top films by country in the Sight & Sound Greatest Films of All Time poll 2012

I’ve complained in the past about the poll’s increasing inability to account for contemporary cinema post-1970, but here I will say that the 2012 edition made some advances. Within the top 50, the number of post-1970 titles rose from eight in 2002 to 13 in 2012. Two films from the past decade – In the Mood for Love and Mulholland Dr. – made the top 50, compared to zero films from the 90s placing that high in 2002.

Six from the 1920s, 3 from the 1930s, 2 from the 1940s, 13 from the 1950s, 15 from the 1960s, 7 from the 1970s, 1 from the 1980s, 3 from the 1990s, 2 from the 2000s
Top 50 films by decade in the Sight & Sound Greatest Films of All Time poll 2012

Regarding other modes of representation and inclusivity, Chantal Akerman’s Jeanne Dielman is the first film directed by a woman to crack the top 50 (Claire Denis’s Beau Travail also broke into the top 100). Battle of Algiers’s top 50 placement is the first for a film from Africa (Mambéty’s Touki Bouki also broke into the top 100), and along with Abbas Kiarostami’s Close-Up, the first from the Middle East. In a perverse post-colonial twist (all the more so given the source of the publication conducting the poll), for the first time no film from Great Britain made the Sight & Sound top 50.

I look forward to hearing your own major observations on the poll results.

In the meantime, here are the biggest gainers and losers of this year’s poll compared to 2002’s results. One should acknowledge the effect of a much larger voting pool leading to the perverse outcome of films like Fanny and Alexander and Intolerance dropping significantly despite getting three times more votes than they did ten years ago. Given the higher degree of consensus required to place well in the poll, it’s rather a relief that the results had as many intriguing titles as it did.

In relative order to their vote share in 2000: Histoire(s) du Cinèma, In the Mood for Love, Mulholland Dr., Sátántangó, Stalker, Playtime, La Jetée, Late Spring, Persona, Gertrud
Biggest gainers in the Sight & Sound Greatest Films of All Time poll 2012
In relative order to their vote share in 2000: The Story of the Last Chrysanthemums, Two or Three Things I Know About Her, Letter from an Unknown Woman, Ivan the Terrible, The Man Who Shot Liberty Vallance, The Seventh Seal, The Travelling Players, Jules et Jim, Intolerance, Fanny and Alexander
Biggest losers in the Sight & Sound Greatest Films of All Time poll 2012

From: David Jenkins
Subject: The Era of Big Films

Hey all,

Well, per Kevin’s list of new inductees, it’s certainly interesting that this poll has a wider political and cultural reach, even if via single films.

One reading you could take from the ascent of films like Shoah, Sátántangó, La Maman et la Putain and Jeanne Dielman is that, in 2012, size matters. If anyone were going to fully reject the taboo of the epic-length movie, it is film critics and academics, and those titles appear to have charged up the ranks in the new poll.

Perhaps one of the most surprising results is that Bela Tarr’s Sátántangó – which didn’t place within the top 100 in the 2002 poll – is now tied at number 34 with the similarly epic Jeanne Dielman. It topped the list of greatest films of the 90s, which itself contained two other four-hour-plus titles – Histoire(s) du Cinema (at number three) and A Brighter Summer Day (at number five). With placings for both Werckmeister Harmonies and The Turin Horse in the top films of the post-2000 period, Tarr is – assuming my maths is right – now the most popular director since 1990. Which is a very pleasant surprise.

Elsewhere, Wang Bing’s 551-minute documentary, West of the Tracks, was tied eighth as one of the films of the post-2000 period, while Shoah is the top film of the 80s. Looking through the sample of personal lists that have been printed in the September issue of Sight & Sound, there’s also a smattering of participants who have voted for Christian Marclay’s monolithic, 24-hour film-gallery piece hybrid, The Clock.

However, one key exponent of the Long Movie – Jacques Rivette – is notable by his absence: Celine and Julie Go Boating surprisingly failed to dent the top 100 despite an ecstatically received New York rerelease, while his celebrated film maudit – Out 1 – will remain just that for at least another decade.

Prior to all the personal top tens being published, it’s difficult to glean much firm insight from this observation beyond the superficial fact that epic movies by their very nature contain an innate cinephile allure, the equivalent of running a marathon or… painting a seed barn. When the full list of results is published, it will be interesting to see whether a film like Out 1 has closed the gap on Celine and Julie, perhaps addressing the idea that when critics are forced to select one film from a certain cherished director, the epic personal folly can become a ballot badge of honour.

From: Jonathan Rosenbaum
Subject: Listmaking 2.0

Hello everyone,

I think it’s important to stress that prior to 2002, whatever discussion took place about the Sight & Sound poll among cinephiles would have been, by necessity, a relatively private and low-profile affair, unless there was some public debate staged that I didn’t hear about or don’t recall now. The point, in other words, is that without the internet, one can’t even document what sort of discussions took place, apart from the odd mention in a few paper publications, just as without digital viewing, the prominence or lack or prominence of certain film titles obviously couldn’t and wouldn’t be the same.

Most significant of all, I think, is that lists of this kind are vastly more important nowadays than they used to be – which is precisely why the number of respondents went from 145 in 2002 to 846 a decade later. And they’re more important because, thanks to digital viewing, some people now have many more choices over what they watch – although for those who opt to stay out of these issues and debates, who wouldn’t dream of watching anything that hasn’t already been ratified by multi-million-dollar ad campaigns, there are correspondingly even fewer choices than there used to be and is more what we might call (in Stalinist terms) a ‘planned culture’.

We have to keep in mind that most Americans probably don’t even know that DVDs and Blu-rays exist in separate regions or zones, so the whole notion of having a multi-regional or multi-zone player and ordering films from abroad, even Hollywood films, would never occur to them. So we obviously have to differentiate between the sort of people who might read Sight & Sound and those who don’t and wouldn’t.

One thing has remained the same – then and now, the number of people interested in playing this game seriously as opposed to frivolously is limited. It isn’t quite a mainstream activity yet, even if it’s closer to being one than it used to be. And one thing that’s changed the overall cultural configuration, I suspect, are DVD extras – educational tools that tend to exist outside the institutional auspices of both film magazines and academia and are therefore somewhat harder to track institutionally.

To answer your question Kevin, my list was predicated on the exclusion of all titles that had appeared on any of my previous Sight & Sound lists, to avoid the monotony of repetitions – hence no Playtime, no Gertrud or Ordet, no Welles.

From: Nicole Brenez
Subject: Rules of the Game

Regarding the making of my list, it was easy, as the Australian website Senses of Cinema asked the same thing in 2000 for the millennium: I looked at that list time again, saw that, basically, I could sill agree with it, and actualised it by adding four principles:

  • To put the films in a chronological order (the first version was alphabetical), and choose only one film by decade. It was important for me to indicate that there are masterpieces all over the history of cinema, especially during its very beginning, where everything had to be invented. The poetry of early scientific cinema remains a major source of inspiration and of joy.
  • To indicate the many ways and tools involved into the word ‘cinema’: from the first series of photographic plates by Étienne-Jules Marey, to the last informatics programme created by Jacques Perconte. Same thing for the many ways through which a film could be produced: by a studio, but also by a single creator, by non-professionals, by collectives… Raoul Peck’s film, in this regard, is important, as Profit & Nothing But! Or Impolite Thoughts on the Class Struggle was produced and broadcast by Arte, a French-German TV channel. The film is a sparkle of this utopian TV that Roberto Rossellini wrote about, a creative, honest, educative and truly popular TV – something that could have been so useful, a dream many times re-actualised by many attempts of free channels, free websites, notably Indymedia and Democracy Now! Rossellini’s thought about TV is symbolically also crucial as he was one of the first to predict the supposed ‘death of cinema’, in 1953 if I remember correctly, because of TV. A prophecy that his disciple Jean-Luc Godard heard very well, so deeply that he worked all his life to create aesthetic bridges to transfer the ideals of cinema from the silver-reel to any kind of material, video, electronic, digital… as we can see in his amazing video series of the 80s and, as a synthesis, in the Histoire(s) du cinéma, so that the definition and conception of cinema won’t depend on the technology.
  • To propose category that has nothing to do with the usual categorisations, as ‘fiction/documentary/essay’, ‘industrial/author/experimental’ and so on, that are so poor, so useless and so false as soon as you analyse the films themselves.
  • To mention as many available titles as possible on the internet, so that if someone, on my list, would discover for example the existence of London 66 by Peter Whitehead, Afrique 50 by René Vautier, Gradiva by Raymonde Carasco or San Francisco by Anthony Stern, it can be found easily and for free. But it was not always possible.

Then, I thought to alternative options to constitute a list and wrote some drafts: a list made only with ten titles of the Étienne-Jules Marey experiments, aggregating to them the movies for which each title could retrospectively be considered as a seed; then, same thing with ten Lumière titles. But it was much too ‘determinising’ and systematic. So I choose the first option, because it’s very heterogeneous, which is for me one of the intrinsic qualities of the history of cinema.

From: Dan Sallitt
Subject: Continuities vs. Irregularities

Responding to Kevin, of course the concept of the Sight & Sound poll depends on it not getting too democratic… And I think its importance as an indicator of canon change also depends on a fair amount of continuity. Voters who switch their entire roster of films every ten years are voluntarily taking themselves out of the gene pool – and if that weren’t the case, I don’t think that cinephiles would pay much attention to the collective results.

I confess to some concern when I heard that the number of votes had increased seven-fold this year, but it looks to me as if the statistical centre held. Though I am a bit troubled at the bizarre dive that Jules and Jim and Story of the Last Chrysanthemums took: when I see discontinuities this great, I wonder whether they’re really due to changing tastes. Presumably there are comparable spikes and troughs in previous surveys that aren’t coming to my mind?

I’m actually embarrassed that my ten-best list leans so hard on the classical American cinema, and on films that I’ve loved for a long time. I did my best to challenge those old affections, and I did dislodge some timeworn favourites. But I wanted to vote helplessly, with as much sophistication as I am capable of, but as if no one else would see the choices.

From: Kevin B. Lee
Subject: Irregularities are Continuities

To answer your question, Dan, the sudden drops of certain titles – as well as the sudden appearances of others – occur with pretty much every edition of the poll. One only needs look at the history of Bicycle Thieves, the number one film from the first poll in 1952. It has since never returned to the top ten; it came as close as number 13 in 1992, only to fall to number 45 in 2002, and is now back up to number 33.

You mentioned Mizoguchi’s Late Chrysanthemums, a film that went from zero votes in 1992 to eight votes and a number 22 placement in 2002. Now it’s at number 183 with nine votes – even though it received more votes than it did in 2002, it was outpaced by more than a hundred other films within the exponential spike in total votes.

Jules and Jim has also had a fairly volatile history, going from number 12 in 1982, to 41, 19 and now down to 127. Here are some other volatile movers over the last three polls:

Volatile titles in the Sight & Sound Greatest Films of All Time poll 2012

The new issue of Sight and Sound

In this 21st-century cinema special: 25 critics choose an era-defining film from each year of the century, and J. Hoberman asks: what is a 21st-century film? Plus: ten talking points from Cannes – George Miller on Furiosa: A Mad Max Saga – remembering Roger Corman with a never-before-seen interview.

Get your copy