The Web Site for Critical Realism The WSCR Archive

[The following paper was originally posted to the Film-Philosophy: Electronic Salon. It is archived here with the permission of the author, who retains the copyright.]

Rothman and the Challenge of Critical Realism


Review of William Rothman, Documentary Film Classics (Cambridge: Cambridge University Press, 1997).


These are exciting times for those of us who are working in documentary film study. Where once the field had seemed almost dead there has been a veritable tsunami of interest. It is in this spirit that I welcomed the opportunity to review William Rothman's study of a number of classical documentaries.

Before engaging the text proper I would like to clarify what this book is not. It is not a history of documentary films, nor is it really an exercise in documentary theory though there is a good deal of theory informing the text and we will deal with that presently. What this book gives us, and which makes it fairly unique and extremely valuable, is an extended critique of a number of texts.

Rothman argues--and I would tend to agree with him--that the practise of critique is almost a forgotten art. Indeed, in these days when the academy generally is rushing to embrace the imperatives of late capitalism, a book which sets out to examine a range of texts rather than a set of policy prescriptions is something to celebrate. Three cheers for Rothman, I say. Long live critique!

There are some other preliminaries that I would like to deal with. We need to consider what we should expect of a critical work. First of all there are issues of selection here. Which texts will be chosen for study? Traditionally implied in such a choice is some concept of a canon. Rothman, however, does not entertain an explicit notion that he is setting up a tradition of great films. Nor is his selection defended--though he does mention a number of other films that he intended to deal with. Whatever the case in the end he settled for Nanook of the North, Land without Bread, Night and Fog, Chronicle of a Summer, A Happy Mother's Day, Don't Look Back and, in a postscript, Monterey Pop.

These are all important and even great films, but the choice is one which lends itself to a particular kind of critique, i.e., a particular way of interpreting the world. I myself would like to have seen Rothman deal with films such as Barbara Koople's Harlan County, Michael Moore's Roger and Me or Dennis O'Rourke's provocative classic The Good Woman of Bangkok. But, as I hope to show, if he had done so then the central thesis of this book would have to have been altered.

Aside from questions of choice what else should we expect of a critique? For me the answer is that we should demand that the critic illuminate the text for us, to enable us to read it at a greater number of levels and to respond in a richer number of ways. Ideally, I would argue, the critic should also "criticise" the text and place it in some sense with regard to other films.

The balance is difficult to strike here and Rothman is basically correct when he says that the practise of the critical study of documentary films is not fashionable. An exception here is Chris Berry et al's fine study of O'Rourke's The Good Woman of Bangkok (1997). However, it seems to me that we need to go back to examples such as Stuart Cunningham's brilliant response to John Heyer's Australian classic The Back Of Beyond (1956) to see what we have lost. Cunningham manages to convey to us a sense of enthusiasm for the film, to reveal it as a rich multi-layered text and at the same time to maintain a proper critical distance.

But it should be said from the outset that Rothman is an equally fine critic. His book is full of high seriousness and though I will disagree with aspects of his argument let me say that I feel that Documentary Film Classics is a very good book and it deserves the widest possible audience.

One final preliminary: unfortunately, like most academics, I process rather than read books, but this is a book which deserves to be read in a linear fashion. It builds up its arguments carefully and when it gets to Monterey Pop all that has gone before has been part of a carefully plotted aesthetic journey.

The Texts

Rothman begins with Flaherty's Nanook of the North. In many ways this is a brave choice. How can one find something new to say about such a venerable classic? But the film is a work of art and as such, like history, is capable of being rewritten. A particular problem with classics such as Nanook of the North is that they are excessively pre-read and so comment on them often substitutes for an actual engagement with the film. For instance one has only to contrast Macdonald and Cousin's glib remarks on Nanook of the North (19) to see what a thorough and careful critic Rothman is.

Rothman is drawn--as always throughout this book--to particular moments within the film. In this instance it is the look that Nanook gives Flaherty's camera when it intrudes on his eating. Here the hunter demands respect and is given it by the film maker.

But there is another aspect of this film. It is the structured absence of any attempt to explain the political economy of Nanook's life. Rothman touches upon this when he discusses the "high wall of hanging pelts" (10). For him this represents the appalling onslaught on nature sanctioned by the fur company. The same company also sponsored Flaherty's film. As with Louisiana Story (1948) the sponsors got a film which avoided the totality within which the lives of the people it filmed were embedded. As in his other films Flaherty substituted a fantastic totality for the real totality within which Nanook lived and operated. In so doing he avoided having to document that Nanook and the Inuit hunters and their families were deeply exploited in that they produced surplus value of great proportions for the fur company. A little part of this surplus was re-routed to Flaherty to help him ignore the destruction of a hunter/gatherer culture and instead "document" the epic battle that Nanook's wages with the wilderness.

To be fair, Rothman feels the exploitation that is being practised on Nanook. Thus he correctly condemns the attempt to make fun of the hunter in the gramophone sequence. However, just as Flaherty refuses to totalise his subject, so Rothman's criticism is inexorably drawn to the moment, and as such reduplicates in part Flaherty's own refusal. To be clear here, against Rothman I would maintain that Flaherty's "crime" lay not in his practising the "violence" of filming the Inuit. Nor did it even lie in his attempt to feel superior to Nanook. Rather it was his absolute refusal to acknowledge and document that totality that he and Nanook and the fur company occupied.

The chapter on Nanook of the North is followed by a discussion of Bunuel's Land without Bread (1932) which deals with the Hurdanos, an impoverished people cut off from the rest of Spain by a range of high mountains. Here we encounter one of the two key philosophical influences on Rothman. The first is Wittgenstein mediated by Stanley Cavell, and the second is Nietzsche. I will deal with the Cavell/Wittgenstein influence later, but I feel the principle vehicle of Nietzsche's influence is the motif of the "terrible wisdom of Silenus," namely that it would be better for humanity not to have been, to be nothing and the next best thing would be to die (22). This Nietzscheanism can be seen most clearly in the discussion of Bunuel's film. There Rothman claims that the key to understanding Land without Bread is a recognition--"That the existence of the Hurdanos is also our existence, that the horror that is their existence is our horror too, our horror of their existence, our horror of our own existence, our horror of everything that exists, our horror of existence itself" (24).

Not the least of the problems with this "absurdity and horror of existence" position is that it is at once dystopic and politically disabling. Nietzsche makes this clear when he is discussing Hamlet's reasons for not acting. The Prince does nothing, according to Nietzsche, because he sees that action cannot change the essence of things (39-40).

Fatalism inevitably follows such logic. Yet there are moments within Bunuel's text which could provide a challenge to the essential-horror-of-existence thesis. Thus we are shown the wealthier neighbours and of course are reminded forcibly of the affluence and power of the Catholic Church. As with Nanook and his people we are once more in the presence of exploitation. It could of course be argued that the existence that the clergy lead is no less absurd than that of the Hurdanos. Whatever the merits of such a case we need to recognise that the priests are a damn sight more comfortable.

As throughout this book one longs ever so much for the return of the repressed of our time--a class based analysis. It is this which would provide the key to the "fate" of the Hurdanos. Such an analysis would re-insert them into a totality and reveal them as the victims of combined and uneven development. But class is of course the taboo concept of our times. Yet even the most elementary grasp of basic Marxism would have prevented Rothman from perpetrating such absurdities as calling Capitalism the "modern feudalism," or saying of the Catholic Church that it was "exhausted as a spiritual force even in the midst of such poverty" (21, emphasis added).

More importantly a class based approach would have avoided the pitfall of universalising the particular experience of the Hurdanos and then reducing it to some metaphysical essence as in:

A class based approach may also have mediated Rothman's tendency to a romanticism which tends to see the fundamental problem of our times not as the existence of a social system based on exploitative class relations but rather on our supposed collective "horror of nature" (23, 38).

The discussion of Land without Bread is followed by an exposition of Resnais' Night and Fog. In these days of resurgent fascism it is good to be reminded of the "accomplishments" of National Socialism. Rothman is at his best here I feel. There is much sensitive and suggestive meditation on what Bernstein has termed "the fate of art," when it became separated from truth and morality and was reduced to the aesthetic.

Thus on page 57 we get a reproduction of a painting of a beautiful woman painted on human skin. One does not know what to say in the presence of such a monstrosity. Is it possible to paint an artistic masterpiece on the skin of a victim of the Holocaust? A parallel question is whether the Nazi scientists advanced knowledge in their hellish experiments on the inmates of their camps. One longs to answer "no" to both these questions, but in many ways to do so is to attempt to deny the full horror of the Final Solution. Better perhaps to follow Rothman and meditate on the "hand scraping shot,"

If the section on Night and Fog is for me the high point of Rothman's book, the next chapter, which consists of a long discussion of Rouch and Morin's Chronicle of a Summer, is much more laboured and far less interesting. Though there is the intriguing moment when Rothman's a priori commitment to overwhelming the particular with the universal almost traps him into to denying the specificity of the Holocaust as when he says in an aside: "(We all have numbers tattooed on our arms, metaphorically, and we are all responsible for tattooing numbers on the arms of others.)" (84).

From Rouch's "shared anthropology" Rothman returns home to cinema verite in America. While the chapter on Night and Fog moved me the most, I have to say that this is the section of the book where Rothman is most at home as a theorist. It is also where the Cavell/Wittgenstein influence is most at work.

There is much in the chapters on A Happy Mother's Day, Don't Look Back, and Monterey Pop that is valuable and insightful. Rothman also writes well about the affinity between cinema verite documentary and the Hollywood cinema. However I wish to comment in particular on the accumulation of the aesthetic throughout these last chapters. This begins at the moment in Leacock's A Happy Mother's Day when the mother of the quintuplets turns in the midst of one of the awful celebrations of her fate to smile, as Rothman puts it, at Leacock's camera. Rothman becomes positively lyrical here. Not only does he anthropomorphise the camera, he hails this grin as the "moment that cinema-verite is born, or reborn, as a movement of independent film" (142).

Leacock's film (or his camera, as Rothman would have it) has revealed a world where Mrs Fischer's babies have been commodified. Nevertheless, for Rothman, the smile has the power to affirm that the world is "capable of being viewed from a transfiguring perspective" (142).

From this epiphany it is a very short journey to the next film, Pennebaker's Don't Look Back--a chronicle of the troubadour Bob Dylan's 1965 British concert tour. Here the analysis is extraordinarily thorough and dense. It builds though to another revelatory moment--the final shots of Dylan at the Albert Hall. The writing is remarkable in this passage and matches perfectly what is an extraordinary moment in the film. It deserves to be quoted at length:

It seems almost sacrilegious, here in the presence of what is really "the god that dare not speak its name," to point out that the "we" of this passage glosses over a world where a large section of humanity have no power and are certainly not responsible for the "hell-on-earth" that Rothman refers to. But we must resist the aesthetic impulse here. We must insist, as always, on the need for the cognitive and the ethical, as well as the aesthetic.

In his preface to the book Rothman tells us of how he had intended to cover a wider range of films. Yet, in the end, he professes himself surprised but satisfied with the collection. Though his final choice of Monterey Pop should surprise no one. It is indeed the conclusion of a journey from the bleakness of the Arctic and the horrors of the concentration camp to the Dionysian joys of the festival.

Rothman's journey had been foretold long before when Nietzsche turned from contemplating the "horror and absurdity of existence" to the "redeeming, healing enchantress--art." For Nietzsche (and Rothman) art alone "can turn these thoughts of repulsion at the horror and absurdity of existence into ideas compatible with life" (40).

Philosophy: between conundrum and aporia

What kind of realism?

Rothman belongs to the school of philosophers who accept the existence of a reality, but there is a distinct preference for this reality to reveal itself rather than to be investigated or explained. Here we have the two contradictory aspects of Wittgenstein's heritage. On the one hand there is a positivistic acceptance of facts, especially in the Tractatus, and, on the other, a tendency to mysticism and Romanticism which is intrinsically a resistance to science (Pears 173-174). In Rothman's case this leads, I feel, to a preference for those documentary films which evince an almost passive attitude towards reality. The alternative documentary tradition which is built around explaining a particular phenomenon is stigmatised as the "voice of god." Rothman would have us believe that the key distinction here is between the assertion and the revelation of truth (110). Moreover, in his criticism, Rothman is drawn towards the key moment in the film when that revelation takes place. I find here an echo of Wittgenstein's injunction:

It is my argument here that Wittgensteinian/Cavellian philosophy suggests very powerfully a particular way of theorising (and indeed making) documentary films, and that Rothman's book is an excellent instance of this approach. This is made very explicit in Rothman's account of Pennebaker's Don't Look Back when he says of Pennebaker's approach that,

Central to this thinking is surely the Romantic fear of science's quest for knowledge of nature. Such a fear is in part a reaction to statements such as Bacon's "knowledge of nature is the same thing as power over nature" (in Cornforth 153). One has only to think of the history of nuclear science to see that such fears are by no means unjustifiable. But cognitive triumphalism and a Promethean scorn for natural limits are not an essential aspect of the human and natural sciences nor those of those philosophies which seek to underlabour for them.

I would offer Bhaskar's Critical Realism as an alternative to the both the evasions of Wittgenstein and the Prometheanism of Bacon (and Cornforth). Of particular importance is the Bhaskarian schema of the logic of scientific discovery where the goal is the discovery of what Bhaskar terms "alethic truth"--the reason for things (Dialectic 214-220).

"The Problems of Philosophy solved"?

Rothman compares the main characters in Flaherty's Nanook of the North and Griffith's True Heart Susie (played by Lillian Gish). He argues that in one film a real character, Nanook, plays "Nanook," a fictional character who is supposed to be real, while in the other a real person, Lillian Gish, plays "Susie," a fictional character who is supposed to be fictional. For Rothman the films are not different because in both cases Nanook and Gish reveal something of their real selves to the camera.

This is truly the Gordanian knot. If we think of Peter Watkins' film Culloden where people act the parts of historically real characters we can see how complex this question can get. How can we acknowledge the fictional aspects of films like Nanook of the North while retaining some notion of the distinction between the representation of a real, and of a fictional character? Mixed up in all this are questions such as the difference between historical and poetic truth, what we mean by "acting," what we mean by the self, and one of the most vexatious issues in philosophy, namely that of reference.

Fiction films can of course reveal aspects of the truth about a particular historical period or conjuncture, however they are ill-equipped to perform the analytical task of giving us the reason for things. But of course it is the type of documentary which attempts what Corner terms the expositional mode of documentary speech that Rothman avoids (Corner 30).

As for acting, Rothman wishes to use this, it would seem, in the ordinary language sense of "doing," of "being an agent." In this sense Nanook acts, Lillian Gish acts, we all act. Thus Rothman argues,

But it is precisely this putting and taking off of masks that good actors do. One wonders if Rothman has ever met an actor and had the experience of discovering that what one thought one knew about her or him is totally wrong in that it was based on the actor's acting. What I am suggesting here is that we cannot reduce acting to its ordinary language sense of "doing." Screen and stage acting are special arts, or language games if you like, which when practised at the highest level reveal little about the actor.

It might help us through the confusion here to discuss the notion of the self, the person, and the contrast between change and difference. Lillian Gish was a real person who developed and changed over a very long period. However she was always Lillian Gish. As Bhaskar puts it we are dealing with a "continuing thing in a tensed process" (Dialectic 45). Now if we take the case of Susie, then we are not dealing with change. It is possible though for an another actor to play her differently but Susie has not changed. Similarly there have been varying interpretations of the character Hamlet, but we cannot talk of Hamlet being a "continuing thing in a tensed process."

Great art (writing and acting) does of course attempt to convey something of the process of change. Thus in Coleridge's "The Ancient Mariner" the wedding guest is a "sadder and wiser man" after the old sailor has finished his tale. But here the change is not within a tensed process nor is it open as in the non-fictional world.

Reference is as I have said a complex and vexatious matter. At the heart of the difficulty is the tradition from Plato to Frege which has it that reference presupposes positive factual existence (Bhaskar, Dialectic 40). But we can refer to "Susie" and also to "Lillian Gish" and in both instances we are referring to different entities. One of these, Susie, is fictional, that is, she does not have a positive factual existence, nor can she ever cause anything to happen. Therefore she is not real. "Lillian Gish" on the other hand refers to someone who did have a positive actual existence and could make things happen. She was real. (Here we should note that "science employs two criteria for the ascription of reality to a posited object: a perceptual one and a causal criterion, neither of which are, of course satisfied by fictional objects" [Bhaskar, Philosophy and the Idea of Freedom 122].) However when we refer to Nanook, it is extremely unlikely that, despite what Rothman says, we are referring to a fictional "Nanook."

It is time perhaps to deal with what Rothman means by revelation. He quotes Cavell's "maxim" that "in the medium of film the only thing that really matters is that the subject be allowed to reveal itself" (4). However if we return this quote to its original context then a much more interesting, and arguably more controversial position emerges:

Again there is much involved here. One detects the hand of Bazin with his notion of allowing things "to exist for their own sakes." Moreover Cavell's is in essence a radical anti-documentary position. Crucial in all this is the key absence of a depth ontology.

For Cavell and Rothman and Wittgenstein there is of course a reality. But theirs is the ontology of empiricism. It is the common sense world of every day objects. It is the reality which Dr Johnson felt he had affirmed when he kicked the stone.

If on the other hand we conceive of reality as consisting of underlying structures and generative mechanisms then patently it is nonsense to talk of waiting for the moment when reality reveals itself. Allied to this is the denial of science as work. By contrast with Cavell and his followers, Bhaskar repeatedly emphasises the work that scientists have to do to uncover aspects of reality. Think for example of the labour that went into the decisive refutation of Newton's theory of gravitation (Cornforth 150-151).


There is intermittent sniping throughout Rothman's book at "Theory," especially film theory motivated by the French intelligentsia. As an old Marxist I have no great love for "theory" and so enjoyed the ambuscades. But seriousness keeps breaking out and one has to face up to the grounds for one's opposition to Theory. I feel we have to ask whether one's opposition is motivated by neorealist/empiricist concerns, or does it spring from a depth critical realist perspective which, as with Bhaskarian theory, contains a moment of theory/explanation as necessary for emancipation.

Rothman is no depth realist. However, his empiricism is softened by a complex aesthetic based, I believe ultimately, on Romanticism. Thus, in his discussion of Monterey Pop when he has finally arrived at the space where he can celebrate the illusion of art become reality, he says: "For the dream these films affirm to become real, all it would take is for the world to stop denying reality" (209).

This sounds significant and poetic, in a word "nice." But it is simply nonsense. Quite clearly this is the sort of political attitude and program which would hold little terror for the IMF or the World Bank or the rest of the powerful and rich who dominate this world of ours. Does Rothman really believe, one wonders, that the dominance of Capital is based on a denial of reality?

There is however a much less attractive side to his opposition to theory. In his preliminary to his discussion of Don't Look Back he strikes the note of American nationalism. Here theory is wrong because it "denies the American way of thinking championed by Dylan [and by Pennebaker], as if it [American thinking] could have no fruitful role to play in the serious study of film" (145).

I find these remarks unworthy of someone who is obviously such a fine, sensitive critic. The proper choice is not between American and French theory. Rather we in film studies and the broader social sciences have, as I see it, basically three choices. We can chose theories, such as Roy Bhaskar's Critical Realism, which seek to underlabour for human emancipation; or those theories, primarily post-modern, which attempt to ground resistance to prevailing power and truth regimes in notions of indeterminate negation; or we can turn to the varieties of new realism which either totally embrace the politics of capitulation, or attempt desperately to postpone this through a turn to aesthetics.

Queensland University of Technology
October 1997

Works Cited

Bernstein, J. M. The Fate Of Art: Aesthetic Alienation from Kant to Derrida and Adorno. Pennsylvania University Press, 1992.

Berry, Chris, Annette Hamilton, and Laleen Jayamanne, eds. The Filmmaker and the Prostitute: Dennis O'Rourke's The Good Woman of Bangkok. Sydney: Power Publications, 1996.

Bhaskar, Roy. Dialectic: The Pulse Of Freedom. London: Verso, 1993.

-----. Philosophy And The Idea Of Freedom. Oxford: Blackwell, 1991.

Cavell, Stanley. The World Viewed: Reflections of the Ontology of Film. New York: The Viking Press, 1971.

Corner, John. The Art of Record: A Critical Introduction to Documentary. Manchester: Manchester University Press, 1996.

Cornforth, Maurice Campbell. Science versus Idealism: In Defence of Philosophy against Positivism and Pragmatism. London: Lawrence and Wishart, 1955.

Cunningham, S. "To Go Back And Beyond." Continuum 2.1 (1988/9): 159-164.

Macdonald, Kevin, and Mark Cousins. Imagining Reality: The Faber Book of the Documentary. London: Faber and Faber, 1996.

Nietzsche, Friedrich. The Birth of Tragedy. Harmondsworth: Penguin, 1993.

Pears, David. Wittgenstein. London: Fontana, 1971.

Wittgenstein, Ludwig. Philosophical Investigations. Oxford: Blackwell, 1976.

Copyright © 1997 Gary MacLennan

The Web Site for Critical Realism The WSCR Archive