Genres are often perceived to be rigid, predictable, ideologically pernicious, and associated with commercial mainstream. "Put simply, genre movies are those commercial feature films which, through repetition and variation, tell familiar stories with familiar characters in familiar situations." is the opening phrase of Barry Keith Grant's book Film Genre: From Iconography to Ideology, and it is a good example of the conventional wisdom regarding genres.
Yet genres are in reality wonderfully fluid, malleable and flexible, and can contain everything from derivative mainstream sequels to great films at the height of what the art form is capable of. This gap between a perceived idea and the reality is apparent both among scholars and the ordinary viewer.
The problems with much of genre theory, or discussions of genre in general, are the same as with much else within film studies/criticism: it is rarely based on that complicated reality but on ideal types, and often founded on a lack of awareness of the historical nuances, developments and complexities of the subject being studied. To borrow a quote from Jason Mittell, many "scholars seem content to take genres at face value, using the labels that are culturally commonplace without giving much consideration to the meanings or usefulness of those labels." In this article I want to point out some of the difficulties with genres, and why the ways they are frequently seen and discussed are unsatisfying.
The basic problem is that the common belief that we can easily define and specify genres is wrong. We can not, because, being made up by us, they are subjective. Therefore, there is no consensus about what constitutes a given genre, or which films should be counted as belonging to a given genre. To give a seemingly trivial example: once in a video store I overheard two teenage girls talking in front of the horror section, and one of them said, outraged, "Why is that film in the horror section?!?! It is not a horror film!" Her friend though it was. That is the reality of the situation. As soon as you try to define a genre it slips away, and the arbitrariness of it becomes apparent. Another example: is Fort Apache (John Ford 1948) a western or a war movie? Or does it not belong to a particular genre? Or is it a cavalry film, as a distinct genre that is neither a western nor a war movie. There is no correct answer to these questions, because it entirely depends upon how the person answering it has chosen to define either of those categories. Or compare She Wore a Yellow Ribbon (John Ford 1949) and The Gunfighter (Henry King 1950). They are both routinely referred to as westerns, and that is understandable, but do they have anything in common, whether in terms of narrative, aesthetics, themes, or characters? Is not the only thing they have in common, other than being non-animated, feature-length, fiction films, that they are set some time towards the end of the 19th century? You could make a film on the same script as She Wore a Yellow Ribbon but change the setting to, say British India in the 1920s, or far in the future at some distant planet. It would tell the same story yet not be called a western then.
Thomas Schatz writes about the "deep structure, those rules and conventions which render this film a Western and that film a musical." but I do not believe that such rules and general conventions exist. The reason a musical is called a musical is because the characters sing in it in a context in which people would not sing in the real world. But that is neither a rule nor a convention (any less than being green is a rule or convention for a green apple) but a banal necessity needed for those films to be called musicals. While there are conventions in musicals, as there are in most things in life, there are no uniform conventions for all musicals. Different kinds of musicals have different kinds of conventions, just as musicals from different eras or decades have different conventions, and musicals from different studios have different conventions, musicals from different countries have different conventions, and musicals by different directors or choreographers or actors have different conventions. And there are no rules. Your musical can look, feel or move any which way. It can be upbeat or sad; it can be progressive or reactionary; it can be fast-paced or slow; it can be serious or playful; songs might be integrated or free-standing. Sometimes a film is made as a musical, but then released without the songs, as a non-musical; sometimes a film is written as a musical but the director ignores that, and refuses to do the song and dance numbers, so it ends up as a non-musical. But the only difference is that they have no songs.
Another difficulty is that there are so many films, most of which are forgotten, and for there to be a theory that explains or defines a genre, it must cover the genre in all its historic appearances, all of these forgotten films from past decades. Another complication is that a lot of older films do not belong to genres that we talk of today, or to genres that have been forgotten. If you look at the top-grossing Hollywood films of the 1930s, many of them cannot be categorised by using our contemporary genres. It instead happens that older films are today placed into genres that did not exist back then, when they were considered as part of some other genre, or not of any genre, but instead considered as part of a cycle for example, like the journalist films of the 1930s, or the same decade's prison films. What, by the way, is the difference between a genre and a cycle. Does a cycle become a genre when it has been going on long enough?
A famous conundrum is film noir. As is well-known, it was not something that existed in the 1940s when the films that today are considered film noirs were made, and they were then instead referred to as all kinds of different genres: detective stories, thrillers, melodramas, spy films, action films, and so on. While some French critics began writing about what they called "film noir" in the late 1940s, they were using it in a slightly different way then how it is being done now, and it was not until the early 1970s that all of these films suddenly became film noirs in the modern sense. A pioneer was Paul Schrader and his article "Notes on Film Noir" from 1972, in which he tried to explain what it was, and how it had evolved from 1941 to 1958. But whether there was an evolution of film noir was dependent upon which films he choose to refer to as such. Laura (Otto Preminger 1944) is clearly different from Kiss Me Deadly (Robert Aldrich 1955), but that might as well be because only one of them is a film noir, or neither is. One of the few things that are consistent regarding film noir is the fact that there is no consensus at all about which films qualify as noir. This is not strange though, but, as Steve Neale has pointed out, it is inevitable: "as a concept film noir seeks to homogenize a set of distinct and heterogeneous phenomena; this inevitably generates contradictions, exceptions and anomalies and is doomed, in the end, to incoherence."
Given all of this, it is no surprise that almost all attempts to present a coherent theory of a genre or genres are flawed from the beginning. Whoever has developed a theory has chosen to include some films and exclude some others, either out of ignorance or deliberately in order for the theory to work. Many thousands of films have been made that are considered westerns, and you are inevitably going to have to narrow it down a lot, but the 10 or 20 films you have chosen to build your grand theory on are not enough. At best, your theory will be valid for those films, but not for westerns in general. (I say at best, because something like Will Wright's famous and influential Sixguns and Society (1975), does not even work for the small sample he has chosen to focus on.)
An obvious risk is that if the claim about a certain genre is that it is rigid and rule-bound, you can only include films that are rigid and rule-bound, and where does that take us? You may well define westerns as films that always exhibit traits A, B, C, and D, and then criticise westerns for being an inferior genre as it only relies on the traits A, B, C, and D, and make ideological hay out of that, but to what extent we have learned something by this is unclear. The flaw is not in the genre westerns, in all their manifold representations, but in your chosen definitions of it.
This aspect of definitions is central to genre theory, yet it too often is the case that the person writing about a certain genre is not explaining how they define it but seemingly assumes that we all define things the same way, even though we do not. Some are loose and open in their definitions, and others are very strict and narrow. If your theory requires a narrow understanding of the subject for which you have developed the theory, considering it in some pure state in which it never existed, you are setting yourself up for tautology.
So far I have only spoken of some genres in relation to Hollywood, which is the way it is usually done. But why is that? Brazilian, Chinese, French, German, Japanese, Swedish, Italian cinema, and others, can also be discussed from generic perspectives if one wants to do so, in the same way Hollywood is discussed. That would also add some needed perspectives to the field, because it would give rise to questions like whether for example neorealism is a genre. It is not treated as such, yet it is easier to define and specify than most of the American genres, and the films have recurring tropes and traits. Considering the fact that many critics and historians have referred to various films from other countries than Italy as being neorealist, it would seem that they might be open to calling it a genre. Or is it merely a style? Another Italian kind of film is giallo. Is that a genre? Is it derivative of Hollywood genres or its own homegrown thing? (I need not take that discussion though because Alexia Kannas already has.)
Genre is often said to be the opposite of art, and to exist only in mainstream cinema, but that is to have a very narrow view of art, and also to believe that a lot of literature and music, including Shakespeare and Mozart, is not art. Some of the world's best filmmakers make/made genre films and many of the best films are genre films. It would be a strange art form of its greatest examples were not to be considered art. The claim that genre equals mainstream (as Grant does in the quote above) is also a dead end. To take two obvious example, Andrei Tarkovsky and Stanley Kubrick worked within genres. You might disagree that they made genre films, but then you need to re-define for example war films and science fiction, so that when they make them they are no longer genre films, only when others make them. (Nancy Meyers's The Intern (2015) is a mainstream film yet less of a genre film than Tarkovsky's Solaris (1972).) Another example is Yasujiro Ozu. He is not considered mainstream yet he was working within established genres. Or is Ozu to be considered as an example of mainstream cinema, albeit within the Japanese context?
Considering the inevitability of such questions, I often wonder if we do not needlessly waste a considerable amount of ink and time because of our abstruse need to think of cinema, and most things in life, in crude binary terms, and mistake our subjective idealisations for some larger truth.
Genres exist, and they can be interesting to talk about, and production companies, filmmakers, scholars, and the regular viewer, refer to genres on a daily basis. But it is rare for those references to acknowledge the complexity and vagueness of the reality. Genres are hardly what we say they are. This does not matter much when choosing a film at the video store, but scholars and historians should thread more carefully, or else their work risk becoming pointless.
Schrader's influential article about film noir was published in 1972, which, as it happens, was also the year Thomas Elsaesser's even more influential article "Tales of Sound and Fury: Observations on the Family Melodrama" was first published. Just as the contemporary understanding of film noir to a large extent comes from Schrader's article, so a contemporary understanding of melodrama in films comes from Elsaesser, even though he cautious the reader that "what I want to say should at this stage be taken to be provocative rather than proven."
Elsaesser, Thomas, (1972) "Tales of Sound and Fury: Observations on the Family Melodrama" published in Monogram, No. 4, 1972.
Grant, Barry Keith, (2007) Film Genre: From Iconography to Ideology. (He must like the sentence I quoted because he has opened other books with it too.)
Kannas, Alexia, (2017) “All the colours of the dark: Film genre and the Italian giallo” published in Journal of Italian Cinema & Media Studies, Vol. 5 Issue 2, March 2017.
Mittell, Jason, (2001) "A Cultural Approach to Television Genre Theory" published in Cinema Journal, Vol. 40, No. 3, Spring 2001.
Neale, Steve, (2000) Genre and Hollywood.
Schatz, Thomas, (1981) Hollywood Genres: Formulas, Filmmaking, and The Studio System.
Schrader, Paul, (1972) "Notes on Film Noir" published in Film Comment, Vol. 8, No. 1, Spring 1972.
Wright, Will, (1975) Sixguns and Society - A Structural Study of the Western.