Page 177
some film critics move to the other end of the spectrum and grant the moviemaker carte blanche to twist the text to their hearts’ content. This stance, taken for obvious reasons by many directors themselves, has caused a number of celebrated battles between Hollywood and novelists.
Two recent discussions neatly illustrate the troubling tension between the filmmaker’s theoretical freedom and his accountability. Writing in The New York Times Book Review in 1990, Molly Haskell betrays the film critic’s characteristic impatience with the literary world’s perceived half-learning.4 Haskell believes “the fact that a fine adaptation might be one that is freely reimagined rather than faithful to its source is a concept little understood by custodians of literature.” Yet later in the same essay, she denies ‘‘that an adaptation should be judged entirely on its merits as a film and without reference to the work on which it is based” (36). Just how the debts to sources acknowledged in Haskell’s analysis are to be paid by filmmakers who faithlessly “reimagine” remains less than clear. The same contradiction occurs more subtly in the extended television interview given by Orson Welles just before his death in 1985.5 The director rejects the idea that his adaptations of literary works owed any special allegiance to their authors. Yet later in the interview, Welles expresses continuing “bitterness” over the mutilation of The Magnificent Ambersons by others intent on giving it an upbeat reworking. Particularly galling to him was the substitution of a happy ending that utterly violated the integrity of “the story.” Of course, one is forced to conclude that Welles was embittered more by the “reimaginative” rape of his screenplay than by the rapists’ simultaneous assault on the Booth Tarkington novel that inspired the film. But this case should serve to remind us that any viable definition of authenticity in adaptation must measure both the original author’s vital interests and the filmmaker’s degree of freedom.
One of the more provocative propositions concerning adaptation is formulated in Keith Cohen’s essay (1977) about Eisenstein’s cinematic plans for An American Tragedy.6 Cohen argues that transferring any novel to film becomes “a truly artistic feat only when the new version carries with it a hidden criticism of its model or at least renders implicit (through a process we should call ‘deconstruction’) certain key contradictions implanted or glossed over in the original” (245). Few would argue that a work of literature could be adapted for film with deconstruction of the text as the primary goal, but surely such a theory is too narrow to cover more than a few actual cases. After all, it stands to reason that the vast majority of directors who adapt for serious artistic purposes
Page 178
are drawn to works they admire, not to those with which they wish to quarrel. This is not to say, however, that there is no legitimate room for interpretation of the literary text. The very process of turning written language into visual images, even though those images are accompanied by spoken language, necessitates interpretation. The more complex the literary source, the more varied the possible interpretations. In the case of Eisenstein’s scenario for An American Tragedy, as I will maintain presently, the great director did not subvert his source, as Cohen appreciatively argues, but merely subjected it to a legitimate interpretation. In fact, Dreiser approved of Eisenstein’s scenario and denounced as subversive von Sternberg’s film treatment of An American Tragedy.7 The key to legitimacy in film adaptation, however, must involve limiting interpretation to ideas, beliefs, and attitudes for which there is some evidence within the source text or in other of the source author’s closely related works.
Because most complex novels achieve structural tension through the clash of ideas about which their authors are undecided or even tortured, the resulting ambiguity offers the filmmaker considerable intellectual discretion in adaptations. Since making movies of Dreiser novels is the subject of this essay, a word or two about the latitude offered potential filmmakers by other “naturalistic” texts might prove useful. Naturalists are thought to be among the most doctrinaire of literary artists. Consider, however, the varying possibilities offered for filmic interpretation even by Stephen Crane’s intellectually rigorous short story “The Open Boat.” Two hypothetical directors might produce equally “artistic” and equally “authentic’’ film versions of the story even if one centered on the grim deterministic engine of cosmic indifference which the men in the boat are powerless to counter, while the other produced a tribute to the freely willed heroism of the bonded group that finally saves all but one of its members. Most naturalistic novels are considerably more complicated than “The Open Boat.” They offer, therefore, a fertile field for “contradictory” yet legitimate cinematic treatment. And most often it is the filmmaker’s ideological, political, or otherwise vested perspective that dictates the treatment of source materials in any adaptation. This should not surprise, since the source author’s own ideology almost always dictates his or her approach to the subject, as contemporary literary critics such as Walter Benn Michaels8 and Amy Kaplan9 have demonstrated.
As much as we might agree about the extent of the filmmaker’s freedom to interpret in adaptation, however, that agreement does not help us judge the relative appropriateness or success of the
Page 179
end product. And film criticism supplies us with few helpful guidelines. Cinema scholar Louis Giannetti gives a typical treatment to adaptations by dividing them into three categories: loose, faithful, and literal.10 His breakdown avoids value judgments, as no qualitative differences are imputed to the three categories. With some amplification, however, his scheme might further our evaluation of Dreiser on film. By a loose adaptation, Giannetti means one that is based on “only an idea, a situation, or a character … taken from a literary source, then developed independently” (329). The example he cites is Kurosawa’s Throne of Blood, which both retains and alters plot elements from Shakespeare’s Macbeth while transferring the action to medieval Japan. We might argue that such a treatment is legitimate, whatever its cinematic quality, because it does not purport to be more than it is. Kurosawa did not call it MacBeth, after all, or even MacShogun. But we should add a note of caution about the “loose” designation. Film adaptations are suspect when passed off as facsimiles of literary texts though they alter important elements so capriciously that the source author’s ideas are seriously diluted, distorted, or destroyed. The most infamous example of a fraudulent adaptation is the 1930 version of Moby Dick, in which Ahab kills the whale and takes a bride.11 We might consider as an example of a film marred by intermittent illegitimacy the NBC television treatment of The Old Man and the Sea.12 Hemingway’s relatively insignificant young tourists, whose only role in the novel is to comment on Santiago’s catch from a restaurant window, become major characters in this treatment, and the old man is given a daughter who nags him to retire. Not even the bravura performance of a marvelous mechanical marlin could make up for these two ill-advised departures from the book. Judgment of such loose adaptations will depend on where the films locate over a continuum embracing works roughly ranging from Kurosawa’s to the 1930 Moby Dick.
The movies in Giannetti’s second category, faithful adaptations, “attempt to recreate the literary source in filmic terms, keeping as close to the spirit of the original as possible” (330). For our purposes, this “faithful’’ category needs to be subdivided into at least two subcategories — successful and unsuccessful. Here judgment must be based both on how justifiable the interpretation of the source text turns out to be and how effectively the filmmaker has used the conventions of his medium to transform that interpretation into a work of art in its own right. Moreover, I would also suggest a third subcategory to be called “superior.” This designation would be reserved for those few films that manage the source material so that the complexity of the novelist’s vision is
Page 180
preserved rather than “interpreted out” by emphasizing only one or two strands of it. In our hypothetical film of the “The Open Boat,” a superior treatment would focus on the debate between determinism and free will implicit in the Crane text. An actual film adaptation which achieves the kind of superiority I am describing is Ken Russell’s Women in Love, an intricate masterpiece that actually clarifies some of D. H. Lawrence’s ideas.13
Giannetti applies his third category, the literal adaptation, almost exclusively to filmed versions of plays (331). But even these are rarely “literal,” because cameras are seldom if ever left at long shot to merely record a play. Certainly when film adaptations of novels are called literal, as von Stroheim’s forty-two reel Greed14 based on Frank Norris’s McTeague15 sometimes has been, they are simply mislabeled.
All five of the Dreiser screen projects discussed in this essay (including the unfilmed Eisenstein scenario) represent, to a greater or lesser degree, faithful adaptations. With the critical system outlined above in mind, let us now judge the strength of their interpretation and the quality of their art.