Friday, 23 February 2018

Muddled modernism

[t]he earliest example of modernist critical reflexivity is Bergman's first entirely independent work as scriptwriter and director, Prison (1948, released in March 1949), which introduced modern reflexivity into European cinema at least ten years before modernism proper and seven years before the critical conception of auteurship. 
"Modernism" is a word that it seems it is inevitable that most film scholars and critics will use at one point or another. And some devote their whole careers to writing about it. For some reason though many do so in a very confused and contradictory way and when they try to define it by using examples of films and filmmakers they feel are modernist it often becomes impossible to understand why they believe this one to qualify but not that one. There is also a need to date it, to say that it began this year and ended that year. Take the quote above from page 228 of Screening Modernism: European Art Cinema, 1950–1980, by András Bálint Kovács, one of the most comprehensive attempts to investigate and specify modernism within cinema. By what possible definition could a film made as late as 1949 be regarded as the "earliest example of modernist critical reflexivity"? Bálint Kovács mentions two earlier reflexive films which he feels do not count as "critical self-reflexivity", Sherlock Jr. (Buster Keaton 1924) and The Man with the Movie Camera (Dziga Vertov 1929), because the filmmakers are not critical of cinema itself and do not try to put themselves forward as auteurs. ("There is no room for auteurs in this system." p. 229) While I can agree with the first part, that the two films do not criticise the medium of film itself, I do not understand what the second part means. And what about all other films, from Mauritz Stiller's Thomas Graals bästa film (1917) and Die Puppe (Ernst Lubitsch 1919) to Marcel Blistène's Etoile Sans Lumiere (1946) that are reflexive? But even if it would actually be true that Prison was first, how would you know if you had not seen all films made before 1949? I am not denying that Prison can be seen as a good example of "modernist critical reflexivity" but if it is to be regarded as the first then you must prove how and why.

The next question the quote raise is why 1949 was ten years before "modernism proper" and "seven years before the critical conception of auteurship." What happened in 1956 and 1959? There was for one thing no specific point in time about which you can say "Here was the birth of auteurship." Already in the early 1920s critics wrote about auteurs and the importance of the director. Alexandre Astruc wrote his frequently mentioned article "La Camera stylo" in 1948. The article about "a certain tendency" by François Truffaut which is often mentioned as seminal (whether it was or nor) in the development of his generation's "politique des auteurs" was published in 1954.

Prison was produced by Lorens Marmstedt and his company Terrafilm, who also produced a few earlier films by Bergman, so it is questionable to call it "Bergman's first entirely independent work". It was the first film directed by Bergman that was based on an original screenplay by him but that is not the definition of independence. There is also an important dimension of Prison which is not addressed at all, and that is that the director in the film is played by Hasse Ekman, who in real life was Bergman's competitor as to who was Sweden's greatest filmmaker. Since this adds another layer of critical self-reflexivity it is strange that this is not discussed at all.


That modernist cinema flourished in the 1960s and early 1970s is a common argument, and not necessarily wrong in the sense that there was a very palpable outburst of creativity across Europe at the time. Another scholar focused on modernity, John Orr, argues in Cinema and Modernity and elsewhere that the years 1958 - 1978 were the true years of modernist cinema. Some, like Miriam Bratu Hansen, argues that cinema in itself is an example of modernist art, or, as she put it in her essay "The Mass Production of the Senses: Classical Cinema as Vernacular Modernism" from 1999, "classical Hollywood cinema could be imagined as a cultural practice on a par with the experience of modernity, as an industrially-produced, mass-based, vernacular modernism." But such a position is not as prevalent as the Bálint Kovács/Orr position. Another more popular take is that Italian neorealism of the late 1940s is where modernist cinema got going. This is, in a way, the position of Gilles Deleuze for example. Robert Kolker too takes that view, in A Cinema of Loneliness from 2011:
Post-World War II cinema modernism, which began with the Italian neorealists, flowered in the work of the French New Wave in the early sixties, and moved through Europe and America, defining a view of the world within the structures of cinema and revitalizing those structures in the process. Cinema modernism foregrounded form, celebrated the history of film by incorporating it into the work of the film itself, while at the same time resisting the conventions that bound filmmaking for decades. That movement ran its course by the eighties, even though Martin Scorsese, Robert Altman, and Stanley Kubrick maintained strong ties to it and continued to experiment with the expressive potentials of the medium. There were few others. (p. xi)
(If modernism foregrounded form and incorporated film history, then neorealism is a peculiar place to begin since it did neither of those things.)

Even those, like the ones I have mentioned, who consider modernist cinema to have appeared after World War 2, still admit that the 1920s was also a strong decade for modernist cinema. Orr says that "one can really speak of two 'modern' cinemas, a silent cinema of Murnau, Dreyer, Lang, Buñuel, and Eisenstein and a sound cinema which crystallizes in the 1960s and early 1970s." (p. 2). Bálint Kovács says that there was a first round of modernism in the 1920s, exemplifying it with German Expressionism, "pure cinema" and French Impressionism (p. 17-19) and "Dreyer was obviously a great modernist auteur throughout his career, while Ozu and Mizoguchi are the only names in this list that do not fit this category." (p. 58)

Trends come and go, and different kinds of films are popular at different times. Westerns are not as popular now as they were from the 1940s to the 1970s. But Westerns have not disappeared. The way discussions about modernist cinema is framed it is like there were no modernist films in the 1930s and 1940s, and, depending on whom you ask, none in the 1950s either. And then they would disappear altogether in the late 1970s. But is that plausible? Could it be a question of which films are famous and which are forgotten? Or that the definitions of modernism are based on the films from a certain period rather than from a set of principles which are then applied to all films to see which correspond with these principles? It is not like scholars do a statistical investigation of each decade and list all films in either the modernist column or the non-modernist column. But if you did that then you could see the ratio of modernist films to non-modernist ones, and give some foundation to your argument. Who knows, maybe the 1940s was really when modernist cinema peaked.

Hold Back the Dawn (Mitchell Leisen 1941)

Bálint Kovács and others are arguing that modernism was primarily a European thing. Kolker on the other hand believes there was some modernism in American cinema and that "it is Hitchcock's Psycho (1960) that marks the passage of American film from its classical to its modernist stage." (p. 18) John Orr has a similar idea although he puts it a bit earlier, in 1958 with Hitchcock's Vertigo and Welles's Touch of Evil. (p. 17). He then blames the alleged disappearance of modernist cinema in the late 1970s on European integration (p. 18) because apparently modernism is dependent on the individual nation.

But what is modernism then? Here is a suggestion: if you want to use the terms "classical" and "modernist" and put them against each other, then a classical film would be one that tells a story which is comprehensible, told in a style that is primarily interested in forwarding that story and not draw too much attention to itself, stays close to common ideas about realism and is not self-conscious. A modernist film would be one that does not follow these classical ideals but instead engage with them or questions them or abandons them. Prison is a good example. It has a complicated, layered narrative, it wrestles with itself and its own artform, it plays with concepts of narrative and narration and is stylistically explicit. This is how most people would define it, including Bálint Kovács, if asked to give the minimum requirements. Orr calls it "the reflexive nature of the modern film, its capacity for irony, for pastiche, for constant self-reflection" (p. 2) Kolker says that "[m]odernist works - Seven (1995) or JFK (1991), any film by Stanley Kubrick - create pleasure with care, with a sense of the fragility of narratives, either political or personal, that presume to represent the world as it is or was." (p. 266) Although, if all films by Kubrick are modernist then how can Psycho be the first American one, as Kolker also said? (Kolker's book is full of questionable and/or contradictory statements.)

But few films are so pure that they have nothing of the classical or nothing of the modernist in them. Therefore, whenever one film or filmmaker is called modernist and another not, it is tempting to ask "Why?" Bálint Kovács dismisses Kurosawa's Rashomon (1950), Ozu and Mizoguchi and almost all of American cinema as not being modernist at all. "The most Hollywood could tolerate of modernism in this period was the slightly neorealistic style of Paul Mazursky, John Schlesinger, John Cassavetes, or Bob Rafelson." (p. 60) he says but calling their films "slightly neorealistic" is not accurate and what about Lilith (Robert Rossen 1964), or The Pawnbroker (Sidney Lumet 1964) or Mickey One (Arthur Penn 1965) or Seconds (John Frankenheimer 1966) or Point Blank (John Boorman 1967)? Bergman of the 1950s or Roberto Rossellini do not count either. This is partly because Bálint Kovács makes a distinction between art cinema and modernist cinema. Explaining why Bergman was not a modernist filmmaker in the 1940s and 1950s he says:
He almost never quit this type of art-cinema form even during his modernist phase. What Bergman did in the beginning of the 1960s was that he modernized this form by adding stylistic and narrative features of modernism to it. He locates his stories in abstract time and space, as in Silence (1963), he made them open-ended, as in Winter Light (1962), he made them self-reflexive and ambiguous, as in Persona. When modernism became obsolete at the end of the 1970s, he just returned to his classical narrative form and to a classical style adapted to the trend of the 1970s and 1980s. (p. 63)
Ambiguity and open endings are usually two key features used to separate art cinema from classical cinema, so it is unusual to use them as something that modernist cinema has but not art cinema. I do not see how Winter Light and Silence are that different from all that was made in the 1950s, by Bergman or many others, to qualify them as being of a different kind of cinema. (Persona though is different.) It is the same with The 400 Blows (François Truffaut 1959), The Loneliness of the Long Distance Runner (Tony Richardson 1962) and Raven's End (Bo Widerberg 1963), three films Bálint Kovács also considers modernist. This is not bringing any clarity to the issue, because what is it that makes them so radically different? It is not that he is too narrow or too wide in his definition but too bewildering. Imagine a book about musicals in which the writer said that 42nd Street (Lloyd Bacon 1933) was the first musical and The Tender Trap (Charles Waters 1955) was the last and that while Singin' in the Rain (Stanley Donen, Gene Kelly 1952) was a musical An American in Paris (Vincente Minnelli 1951) was not, and then not elaborated as to why. That is not different from these books. Two other things Bálint Kovács thinks are typical of modernism are genre parody and narrative ambiguity, arguing that "Genre became a focus for parodies only from the late 1950s on." (p. 115) But has there ever been a time when people were not making genre parodies? As for when ambiguity first appeared he has two opposing propositions: "Another main trend was informed by the problem of narrative ambiguity appearing for the first time in Kurosawa's Rashomon." (p. 271) but on page 60 narrative ambiguity "was introduced into modern cinema by Alain Resnais and Alain Robbe-Grillet". But obviously, narrative ambiguity is also almost as old as narrative cinema itself. (Incidentally, the exact same concept in Rashomon was used in Anthony Asquith's The Woman in Question, also from 1950 but neither could have influenced the other.)

Another important aspect of many definitions of modernist cinema (and art cinema in general) is the idea of the auteur. To again quote Bálint Kovács: "It is with the idea that the film has an individual auteur who has his own personal relationship to reality and to the medium that critical reflection appears in the cinema." (p. 224) but it is peculiar that an idea from critics should matter here and not what the filmmakers themselves believed, and filmmakers from early on most definitely had this relationship and awareness. Buster Keaton, Ernst Lubitsch, Fritz Lang, Mauritz Stiller, Billy Wilder, Yasujiro Ozu and Howard Hawks are examples of filmmakers who signalled such self-awareness long before the late 1950s.

But what about the definition of modernist films I gave above? Does it make any sense? It depends on which specific films I would call modernist. The common argument that neorealist films such as Bicycle Thieves (Vittorio De Sica 1948) or La terra trema (Luchino Visconti 1948) are modernist and that films such as Last Year in Marienbad (Alain Resnais 1961), Trans-Europ-Express (Alain Robbe-Grillet 1966) or India Song (Marguerite Duras 1975) are modernist too makes it a very expansive term because these films are very different from one another. The story and structure of Bicycle Thieves is conventional, straightforward and melodramatic and Last Year in Marienbad is nothing like it. Does one specific term such as modernism really capture all these films? I would be in favour of treating someone like Frank Tashlin as a modernist filmmaker rather than De Sica. A lot of films made in Hollywood, Japan and India during these decades are more modernist, by most definitions, than neorealism and various French New Wave films. But this does not mean others are wrong to claim Bicycle Thieves as a work of modernist cinema and disregard Tashlin. I do not have a monopoly on the proper definition of modernism. But what I mean is that it is reasonable to demand some semblance of coherence and logic.

This is not something unique for modernism or these books but a general problem. Or rather two problems. One is not having the film historic knowledge necessary for making the claims in question and the other is not being consistent in the argumentation. If you define X as an item having the properties A, B and C, then you cannot also later claim that an item which lacks A, B and C is still an X or say that an item with properties A, B and C is not an X, without carefully explaining way. If you do not it does suggest you are making it up as you go along.

Inevitably, politics is involved too. The common, generic assumption is that the classical is conservative and "in support of the status quo," that most vacuous of terms ("It is called entertainment, but it is in fact ideology reproducing itself." (p. 348) as Kolker puts it) whereas the modernist is radical and against the status quo. Again, the proper response is "Why?" and "How?" Bergman was a modernist artistically but politically close to the Social Democratic consensus of the time, and was criticised by the far left in the 1960s and 1970s. Resnais, Rivette and Rohmer (perhaps) were modernists but what were their politics and how did their films challenge the rule of de Gaulle? The themes and the messages of the works must also be considered when discussing politics and ideology, not just their form or their style, and there is no reason to assume that a film's style or form is indicative of a certain idea it might have. A classical film can be politically radical and a modernist film politically conservative. (Of course, sometimes being conservative can be quite radical depending upon the circumstances.)


For all that have been said about modernism the situation is still very muddled. Which is probably inevitable considering the complexity of the issue. In order to define modernism above I simplified by providing a binary situation, classical vs. modernist film. But these are of course not the only two kinds there are. I have elsewhere suggested that another distinction can be made; between classical and romantic, with the second a cinema of emotional and visual excess exemplified with, for example, Borzage, Minnelli, Powell/Pressburger, Ophüls. Modernism would be a third kind, distinct from both classical and romantic. (Ophüls's Lola Montès (1955) could be seen as bordering on both romantic and modernist cinema, a link between Minnelli and Rivette.) Bálint Kovács as we have seen wants to separate modernist films from art films, seeing modernist cinema as a reaction against art cinema. (p. 62) Some want to make a distinction between modernist cinema and avant garde cinema. ("Modernism, according to [Peter] Wollen, is characterised by reflexivity, semiotic reduction, foregrounding of the signifier and suppression or suspension of the signified, whereas the avant-garde rejects purism and ontological speculation in favour of semiotic expansion and a heterogeneity of signifiers and signified." Well then.) In Narration in the Fiction Film (1985) David Bordwell provides at least four different kinds of modes of narration (which makes for four different kinds of films): classical narration, art-cinema narration, Soviet historical-materialist narration, and parametric narration. He then asks whether any of them could be aligned with the term modernism, and if so, which one? His answer is that it depends, but that it is not that important. (p. 310)

In his most recent book, Reinventing Hollywood, Bordwell talks about "moderate modernism" by which he means various stylistic techniques used in Hollywood in the 1940s that go beyond the classical conventions. That appeals to me, considering what I said above how there are few films that are pure one thing or the other. Films exist along a continuum, making many of these definitions and distinctions difficult judgement calls. But if you have to make them, at least take care that what you argue is historically accurate and theoretically coherent.

Equinox Flower (Yasujiro Ozu 1958)


Books cited:
András Bálint Kovács Screening Modernism: European Art Cinema, 1950–1980 (2007)
David Bordwell Narration in Fiction Film (1985)
David Bordwell Reinventing Hollywood (2017)
John Orr Cinema and Modernity (1993)
Robert Kolker A Cinema of Loneliness (2011)

Miriam Bratu Hansen's article was first published in Modernism/Modernity Volume 6 #2 (1999).

The quote about Peter Wollen is by Alison Butler in a chapter in The Cinema Book (2nd Edition 1999) p. 117

A link to Alexandre Astruc's "Le camera stylo"

An earlier piece by me about ideology.

This quote from Screening Modernism is not related to the topic of this piece, but it is so strange that I wanted to share it anyway: "During at least the first sixty years of film history, one could not reasonable speak about a cinematic tradition whatsoever. Cinema as a cultural tradition was first invented by the auteurs of the French new wave." (p. 16) Is there any way of interpreting that in a way that makes it even remotely true?

Friday, 9 February 2018

Save the Cat!

When the screenwriting manual Save the Cat! The Last Book on Screenwriting You'll Ever Need was released in 2005 it became an immediate success, and has been reprinted multiple times. It has two sequels and there is a Twitter account and a website carrying on the ideas of the original book. There have been conferences, works shops and lectures around it. Blake Snyder, who wrote it, clearly managed to fill a need out there for a quick read on how to write a screenplay.

The book has of course also been criticised many times, primarily from people who see it as having ruined Hollywood, and from people who believe that there is more to film than just Hollywood mainstream cinema even though Save the Cat! might pretend otherwise. Snyder is only interested in profit, and his book is about writing the most commercially successful script possible. If it does not satisfy an impatient teenager, it is of no use to Snyder. But he is upfront with this and there is no need to disparage a book just for wanting to teach how to write a successful teen movie. Such films also have a place in the world.

At long last I have now finally read Save the Cat! myself. The book has eight chapters which go through the steps Snyder thinks a writer should take. Some of these are useful, such as having a good logline (a sentence or two which captures what the film is about) and having a board on which you can see your whole script and the structure of it, so you can easily arrange scenes and plot points, and move them around. But other than these basic suggestions, which are pretty standard advice from such books and not unique to Save the Cat!, nothing else in it makes sense, even for the kinds of films Snyder wants his readers to write. Snyder has put so little thought into it, and has such peculiar ideas about films, audiences and, well, screenwriting, that it is hard to understand why it was even published in the first place.

Let's consider the logline. According to Snyder a great logline should be ironic, "create a compelling mental picture", tell the production company what the potential "audience and cost" will be, and suggest "a killer title". Fair enough. Then he provides this as an example of a great logline: "A just-hired employee goes on a company weekend and soon discovers someone's trying to kill him - The Retreat". (p. 5) I do not see how this satisfies any of the key ingredients Snyder just said a logline must have. There is no irony. It gives no idea what kind of a film it is, as it could be a comedy, a thriller, an action film or a drama. Hence it tells us nothing about either audience or costs. By his own favourite example he has negated his own argument. And, surely, The Retreat is not a killer title. It is a rather lame and generic title. Snyder is confusing about titles elsewhere too since he mentions these four Hitchcock films as examples of what he thinks are great titles: Rear Window (1954), Vertigo (1958), North by Northwest (1959) and Psycho (1960). These he claims are titles that sell, and clearly tell what genre and style the films are of. "All of them, across the board, certainly say what it is and they do so in a way that's not on the nose or stupid" (p. 15). But, with the possible exception of Psycho, they certainly do not. Nobody would be able to guess what kind of film North by Northwest or Rear Window is if all they had to go by was the title. So what does Snyder mean?

One chapter is called "The Immutable Laws of Screenplay Physics" and Snyder gives eight examples of such laws which he calls:
Save the Cat
Pope in the Pool
Double Mumbo Jumbo
Laying Pipe
Black Vet
Watch Out for That Glacier!
The Covenant of the Arc
Keep the Press Out

These rules Snyder says are always true for all good films. But they are not. Not at all. He knows this too because he mentions very successful films that break them so even by his own examples they are not in fact "immutable laws" and it is unclear why he claims that they are.

It is not just that the "laws" are not immutable, they are also weird to begin with. I will just briefly discuss three of them, "Pope in the Pool", "Double Mumbo Jumbo" and "Laying Pipe". The first one is about what to do in scenes with boring exposition so as not to lose the interest of the audience. His example, from which the name comes, is to present the exposition while the pope is going for a swim in a pool in the Vatican. But what is left unclear is why, if the exposition is boring for the audience, included it at all? Why have a scene in which things are said which the audience will not bother about? If the scene is constructed to keep the audience distracted from the dialogue it is likely that they will not pay any attention at all to said dialogue, which is even more reason to delete the whole scene to begin with.

"Double Mumbo Jumbo" refers to Snyder's belief that the audience will accept one piece of weird, unrealistic thing in a film but not more than one. As an example he mentions that in Spider-Man (Sam Raimi 2002), first Peter Parker develops superpowers and then Norman Osborn also develops superpowers and becomes the villain The Green Goblin. This, according to Snyder, is a bad thing and an example of double mumbo jumbo which is supposedly off-putting to the audience. But is not the opposite true here? A key aspect of the whole superhero concept is the idea of one hero and one villain who both have developed superpowers and often are in same way mirror images of each other, and audiences since at least the 1930s have loved this concept. Why Snyder would consider this a major flaw, and something to avoid, is a mystery.

"Laying Pipe" is what Snyder calls it when there is too much exposition and too much information in the beginning of a film, before we get to the important stuff. Here he mentions Minority Report (Steven Spielberg 2002) as an example as it is 40 minutes into the film that John Anderton becomes hunted by his own team. Since Snyder sees the film's hook to be "a detective discovers he is the criminal" (p. 129) it is according to him way too late to have this moment occur after 40 minutes. The audience will be bored by then. This is another example of a Snyder argument that makes no sense. There have been plenty of action and drama before this moment, and if an audience member is bored it will not be because of this moment happening after 40 minutes. And is that really the hook of the film? Is it not more of a twist? (Not to nitpick too much but Anderton does not discover that he is a criminal, he discovers that the system says he is one, which is not the same thing.) Snyder's complaint is similar to arguing that The Sixth Sense (M. Night Shyamalan 1999) is a failure because it is not until the very end that we are told that Malcolm Crowe is dead.

So much for the immutable laws. Another argument Snyder makes in the book is that all buddy films, like Butch Cassidy and the Sundance Kid (George Roy Hill 1969) or Thelma & Louise (Ridley Scott 1991), begin with the buddies disliking each other. This is not true at all. It is not even true for Butch Cassidy and the Sundance Kid or Thelma & Louise. In a film like 48 Hrs. (Walter Hill 1981) the two leads dislike being together at first and then their relationship deepens during the course of the story but that is just one kind of buddy film, not a rule for all of them.

Snyder also says that nobody wants to see a film about a man who is "a little world-weary and yet bravely wise" (p. 53) even though that description applies to a large number of exceptionally successful male movie stars, then and now. He also talks about the importance of the first image of a film, how it must immediately grab the audience and tell them what the film is about and what it will be like. The audience he says, is supposed to think "This is gonna be good!" (p. 73) As an example he uses Lawrence of Arabia (David Lean 1962). This is the first image of Lawrence of Arabia:

Snyder says a couple of times that by following his advice success is assured. About his immutable laws he says "they always work." (p. 38) and later he says about his rules that "The point I'm trying to get across here is - it works. And it works for a reason." (p. 42) But works in what way? Success comes not just from a script; acting, directing and marketing also matters for example. Most films are not particularly successful even though many of them since 2005 have been written following Snyder's guidance, such as it is. I suppose this is one reason why his book is so successful. He makes the claim that if you buy the book and follow the rules in it you will not fail. For an aspiring writer that might be just what you want to hear, I can understand that, and Save the Cat! is a very quick read too. But this book is not the answer to their needs. There are many other books written by people who actually know what they are doing, understand their own arguments and have good advice to give.

I have singled out a few things about the book but there is a lot more that I could have brought up. But for now, just one more thing. Snyder measure things according to box office returns. He scoffs at people who think Memento (Christopher Nolan 2000) is worth talking about. "But be ready for one hell of an argument from me!! I know how much Memento made." (p. 96) I am sure he does. But if box office is what matters why is he criticising Spider-Man and Minority Report? According to Snyder's logic they cannot have any flaws considering how successful they were. In the part where he waves away Memento he uses Miss Congeniality (Donald Petrie 2000) as an example of a film with a perfect script, yet it made a lot less money than the other two. Memento, while not making as much money as Miss Congeniality, did pretty good for a cheap indie production and in terms of return on investment it was the more successful of the two.


Before he wrote the book Blake Snyder was a screenwriter himself, and he seems to be proud of his career and his accomplishments, which he uses as examples to emulate. I do not begrudge him that. (He co-wrote Stop! Or My Mom Will Shoot (Roger Spottiswoode 1992) and Blank Check aka Blank Cheque (Rupert Wainwright 1994).) But he is also referred to as "one of Hollywood’s most successful spec screenwriters" and sometimes just "Hollywood's most successful spec writer." Yet I have failed to figure out who it was that made that claim. The closest I got was that some trade journal allegedly said it, but not which one or in what context. (If some trade journal did say it, it cannot have been an important one because if it was its name would obviously be mentioned by Snyder.) It is one of those cases where a quote gets a life of its own, and there is never a source but only circular references. It is quite possible that nobody ever made that claim.

"I had this great premise, which was: 'Dirty Harry gets a new partner - his mother." (p. 81)