While millions are watching movies at this moment, a few thousand are studying the films of the past. One person is trying to ascertain whether a certain film was made in 1904 or 1905. Another is tracing the fortunes of a short-lived Scandinavian production company. Another is poring over a 1927 Japanese film, shot by shot, to find out how it tells its story. Some researchers are comparing prints of an obscure film to determine which one can be considered the original. Other scholars are studying a group of films signed by the same director or
Set designer or producer. Some are scrutinizing patent records and technical diagrams, legal testimony, and production files. And still others are interviewing retired employees to discover how the Bijou Theater in their hometown was run during the 1950s.
Why?
One reason is evident. Most film historians—teachers, archivists, journalists, and freelancers—are cinephiles, lovers of cinema. Like bird-watchers, fans of 1960s television, art historians, and other devotees, they enjoy acquiring knowledge about the object of their affection.
Movie fans may stop there, regarding the accumulating of facts about their passion as an end in itself. But whatever the pleasure of knowing the names of all the Three Stooges’ wives, most film historians are not trivia buffs.
Film historians mount research programs, systematic inquiries into the past. A historian’s research program is organized around questions that require answers. A research program also consists of assumptions and background knowledge. For a film historian, a fact takes on significance only in the context of a research program.
Consider the image at the top of the page, from a film of the silent era. A film archivist—that is, someone who works in a library devoted to collecting and preserving motion pictures—often finds a film she cannot identify. Perhaps the title credit is missing or the print carries a title that differs from that of the original film. The archivist’s research program is, broadly, identification. The film presents a series of questions: What is the date of production or of release? In what country was it made? Who made the film? Who are the actors?
Our mysterious film carries only the title Wanda l’es-pione (“Wanda the Spy” )—most likely a title given to it by a distributor. It was probably imported rather than made in the small country in which the print was discovered. Fortunately there are some clues in the print itself that a knowledgeable historian can spot. Its lead actress, seated in the foreground, is a famous star, Francesca Bertini. Identifying her makes it almost certain that the film is Italian, made during the height of her career in the 1910s. The film’s style allows the researcher to narrow the range of dates even more. The camera frames straight toward the back wall of the set, and the actors seldom move closer to the camera than they are seen here. The editing pace is slow, and the action is staged so that performers enter and exit through a rear doorway. All these stylistic features are typical of European filmmaking of the mid-1910s. Such clues can be followed up by referring to a filmography (a list of films) of Bertini’s career. A plot description of a 1915 film in which she starred, Diana I’affascinatrice (“Diana the Seductress”), matches the action of the unidentified print.
Note that the identification depended on certain assumptions. For example, the researcher would have assumed that it is extremely unlikely, if not impossible, for, say, a 1977 filmmaker to scheme to bedevil archivists and make a fake 1915 Italian film. (Film historians seldom need worry about forgeries, as art historians must.) Note, too, that background knowledge was indispensable. The researcher had reason to believe that films staged and cut a certain way are characteristic of the mid-1910s, and the researcher recognized a star from other films of the period.
Consider another possibility. An archive holds many films made by the same production company, and it also has numerous filing cabinets bulging with documents concerning that company’s production process. Its collection also includes scripts in various drafts; memos passed among writers, directors, producers, and other staff; and sketches for sets and costumes. This is a rich lode of data—too rich, in fact, for one researcher to tackle. The historian’s problem is now selecting relevant data and salient facts.
What makes a datum relevant or a fact salient is the historian’s research program and its questions. One scholar might be interested in tracing common features of the company’s production process; he might ask something like, “In general, how did this firm typically go about making movies?” Another historian’s research program might concentrate on the films of a certain director who worked for the company. She might ask, “What aspects of visual style distinguish the director’s films?”
Some facts would be central to one program but peripheral to another. The historian interested in the company’s production routines might not particularly care about a daring stylistic innovation introduced by the director who is the focus of the other historian’s inquiry. Conversely, the latter historian might be uninterested in how the company’s producers promoted certain stars.
Again, assumptions exert pressure on the researcher’s framing of questions and pursuit of information. The company historian assumes that he can trace general tendencies of production organization, largely because film companies tend to make films by following fairly set routines. The director-centered researcher assumes—perhaps initially only as a hunch—that her director’s films do have a distinct style. And both historians would mobilize background knowledge, about how companies work and how directors direct, to guide their research.
Historians in any discipline do more than accumulate facts. No facts speak for themselves. Facts are interesting and important only as part of research programs. But facts help us ask and answer questions.
Film History as Description and Explanation
Inevitably, a historian needs at least a little information to prod him to ask questions. But the historian does not necessarily sift through mountains of facts and then judiciously ask a question. A historian may begin with a question, and sometimes that question might be better described as a hunch or an intuition or even just an itch.
For example, one young historian saw a few of the “anarchic” American comedies of the 1930s and noticed that their vulgar gags and absurd situations were very different from the more sophisticated comedy of the period. Suspecting that stage comedy might have been a source, he framed a question: “Might vaudeville and its performance style have shaped these particular comedies of the early 1930s?” He began to gather information, examining films, reading coverage of the comedians in the Hollywood trade press, and studying shifts in American taste in humor. The process of research led him to refine his question and to mount a detailed account of how comedians introduced a vaudeville aesthetic into sound films but then muted it in accord with Hollywood’s standards of taste.2
Nonhistorians often visualize the historical researcher as a cousin to Indiana Jones, braving library stacks and crawling through attics in quest of the treasure-lode of documents that overturn popular opinion. Certainly new documentation has a key role to play in historical research. One scholar gained entry to the long-inaccessible files of Hollywood’s self-censorship agency, the Hays Office, and she was able to put forth a new account of the office’s procedures and functions.3
Similarly, the increasing availability of films from cinema’s earliest era has created an entire subfield of cinema history.4
Still, many research programs rely more on asking new questions than on unearthing new data. In some cases, the research question seems to have been answered by previous historians, but another researcher comes along and suggests a more complete or complex answer. For example, no historian disputes the fact that Warner Bros. was quick to invest in talking pictures in the mid-1920s. For a long time most historians believed that the firm took this risky step because it was on the verge of bankruptcy and was desperate to save itself. But another historian with more knowledge of economics and how to read companies’ balance sheets concluded that evidence—which had long been publicly available to researchers—strongly indicated a quite different situation. He argued that, far from facing bankruptcy, Warners was quickly expanding and that investing in sound films was part of a carefully planned strategy for breaking into the ranks of the major studios.5
Our examples all indicate that the historian’s research program aims to do at least two things. First, the historian tries to describe a process or state of affairs. She asks What and who and where and when. What is this film, and who made it, and where and when? In what ways does this director’s work differ from that of others? What was the vaudeville comedic style? What evidence is there that a studio was nearly bankrupt? Who is the actor in this shot? Who was responsible for scripts at this company? Where was this film shown, and who might have seen it? Here the historian’s problem is largely one of finding information that will answer such questions.
Accurate description is indispensable for all historical research. Every scholar is indebted to descriptive work for identifying films, collating versions, compiling filmographies, establishing timelines, and creating reference works that supply names, dates, and the like. The more sophisticated and long-lived a historical discipline is, the richer and more complete its battery of descriptive reference material is.
Second, the historian tries to explain a process or state of affairs. He asks, How does this work? and Why did this happen? How did this company assign tasks, lay out responsibilities, carry a project to completion? How did this director’s work influence other films from the company? Why did Warners pursue talkies when larger companies were reluctant to do so? Why did some sound comedians adopt the vaudeville comedic style while others did not?
The film historian, like a historian of art or politics, proposes an explanatory argument. Having asked how or why, she puts forward an answer, based on an examination of evidence in light of assumptions and background knowledge. In reading historical writings, we need to recognize that the essay or book is not just a mass of facts but an argument. The historian’s argument consists of evidence marshaled to create a plausible explanation for an event or state of affairs. That is, the argument aims to answer some historical question.
Most arguments about empirical matters—and the history of film is principally an empirical matter—rely on evidence. Evidence consists of information that gives grounds for believing that the argument is sound. Evidence supports the expectation that the historian has presented a plausible answer to the original question.
Film historians work with evidence of many sorts. For many, copies of the films they study are central pieces of evidence. Historians also rely on print sources, both published (books, magazines, trade journals, newspapers) and unpublished (memoirs, letters, notes, production files, scripts, court testimony). Historians of film technology study cameras, sound recorders, and other equipment. A film studio or an important location might also serve as a source of evidence.
Usually historians must verify their sources of evidence. Often this depends on the sort of descriptive research we have already mentioned. The problem is particularly acute with film prints. Films have always circulated in differing versions. In the 1920s, Hollywood films were shot in two versions, one for the United States and one for export. These could differ considerably in length, content, and even visual style. To this day, many Hollywood films are released in Europe in more erotic or violent versions than are screened in the United States. In addition, many old films have deteriorated and been subject to cutting and revision. Even modern “restorations” do not necessarily result in a film identical to the original release version. (See “Notes and Queries,” Chapter 4.) Many current video versions of old films have been trimmed, expanded, or otherwise altered from their theatrical release format.
Often, then, the historian does not know whether the print she is seeing represents anything like an original, if indeed there can be said to be a single “original” version. Historians try to be aware of the differences among the versions of the films they are studying and try to account for them; indeed, the fact that there are different versions can itself be a source of questions.
Historians generally distinguish between primary and secondary sources. As applied to film, primary usually refers to the people directly involved in whatever objects or events are being studied. For example, if you were studying Japanese cinema of the 1920s, films, interviews with filmmakers or audience members, and contemporary trade journals would count as primary material. Later discussions concerning the period, usually by an earlier historian, would be considered secondary.
Often, though, one scholar’s secondary source is another’s primary source, because the researchers are asking different questions. A critic’s 1960s’ essay about a 1925 film would be a secondary source if your question centered on the 1925 film. If, however, you were writing a history of film criticism during the 1960s, the critic’s essay would be a primary source.
Explaining the Past: Basic Approaches
There are distinct types of explanation in film history. A standard list would include
Biographical history: focusing on an individual’s life history
Industrial or economic history: focusing on business practices
Aesthetic history: focusing on film art (form, style, genre)
Technological history: focusing on the materials and machines of film
Social/cultural/political history: focusing on the role of cinema in the larger society
This sort of inventory helps us understand that there is not one history of film but many possible histories, each adopting a different perspective. Typically, the researcher begins with an interest in one of these areas, which helps him to formulate his initial question.
Nevertheless, such typologies can be restricting if they are taken too rigidly. Not all questions the historian may ask will fall neatly into only one of these pigeonholes. If you want to know why a film looks the way it does, the question may not be purely aesthetic; it might be linked to the biography of the filmmaker or to the technological resources available when the film was made. A study of film genres might involve both aesthetic and cultural factors, and a person’s life cannot easily be separated from his or her working conditions within a film industry or from the contemporary political context.
We propose that the student of film history think chiefly in terms of questions, keeping in mind that these might well cut across typological boundaries. Indeed, one could argue that the most interesting questions will.
Explaining the Past: Organizing the Evidence
Finding an answer to a historical question may involve both description and explanation, in different mixtures. The techniques of descriptive research are specialized and require a wide range of background knowledge. For example, some experts on early silent cinema can determine when a film copy was made by examining the stock on which it is printed. The number and shape of the sprocket holes, along with the manner in which a manufacturer’s name is printed along the edge of the film strip, can help date the print. Knowing the age of the stock can in turn help narrow down the film’s date of production and country of origin.
Historical explanation also involves concepts to organize the evidence produced by specialized knowledge. Here are some of them.
Chronology Chronology is essential to historical explanation, and descriptive research is an indispensable aid to establishing the sequence of events. The historian needs to know that this film was made before that one or that event B took place after event A. But history is not mere chronology. A chronology stops short of explanation, just as a record of high and low tides gives no hint as to why tides change. History, as we have already seen, centrally involves explanation.
Causality Much historical explanation involves cause and effect. Historians work with conceptions of various kinds of causes.
Individual Causes People have beliefs and desires that affect how they act. In acting, they make things happen. It is often reasonable to explain a historical change or a past state of affairs in light of the attitudes or behavior of individuals. This is not to say that individuals make everything happen or that things always happen as people originally intended or that people always understand just why they did what they did. It is simply to say that historians may justifiably appeal to what people think and feel and do as part of an explanation.
Some historians believe that all historical explanation must appeal to person-based causes sooner or later.
This position is usually called methodological individualism. A different, and even more sweeping, assumption is that only individuals, and exceptional individuals at that, have the power to create historical change. This view is sometimes called the Great Man theory of history, even though it is applied to women as well.
Group Causes People often act in groups, and at times we speak of the group as having a kind of existence over and above the individuals who compose it. Groups have rules and roles, structures and routines, and often these factors make things happen. We speak of a government’s declaring war, yet this act may not be reducible to more detailed statements about what all the individuals involved believed and did.
When we say that Warner Bros. decided to adopt sound, we are making a meaningful claim, even if we have no information about the beliefs and desires of the individual decision makers at the company; we may not even fully know who they were. Some historians assert that any historical explanation must, sooner or later, ground itself in group-based causes. This position is usually called holism, or methodological collectivism, as opposed to methodological individualism.
Several sorts of groups are important to the history of cinema. Throughout this book we will be talking about institutions—government agencies, film studios, distribution firms, and other fairly formal, organized groups. We will also be talking about more informal affiliations of filmmakers. These are usually called movements or schools, small assemblies of filmmakers and critics who share the same interests, beliefs about cinema, conceptions of film form and style, and the like. (Movements are discussed in more detail in the introduction to Part 2.)
Influence Most historians use some notion of influence to explain change. Influence describes the inspiration that an individual, a group, or a film can provide for others. Members of a movement can deliberately influence a director to make a film a certain way, but a chance viewing of a movie can also influence a director.
Influence does not mean simple copying. You may have been influenced by a parent or a teacher, but you have not necessarily mimicked his or her behavior. In the arts, influence is often a matter of one artist’s getting ideas from other artists’ work but then pursuing those ideas in a personal way. The result may be quite different from the initial work that stimulated it. The contemporary director Jean-Luc Godard was influenced by Jean Renoir, although their films are markedly different. Sometimes we can detect the influence by examining the films; sometimes we rely on the testimony of the filmmaker.
A body of work by a group of directors may also influence later films. Soviet cinema of the 1920s influenced the documentary director John Grierson. The Hollywood cinema, as a set of films, has been enormously influential throughout film history, although all the directors influenced by it certainly did not see exactly the same films. Influences are particular kinds of causes, so it is not surprising that influences may involve both individual activity and group activity.
Trends and Generalizations Any historical question opens up a body of data for investigation. Once the historian starts to look closely at the data—to go through a studio’s records, examine the films, page through the trade press—she discovers that there is much more to explore than the initial question touches on. It is like looking into a microscope and discovering that a drop of water teems with organisms of confounding variety, all going about very different business.
Every historian omits certain material. For one thing, the historical record is already incomplete. Many events go unrecorded, and many documents are lost forever. Further, historians inevitably select. They reduce the messy complications of history to a more coherent, cogent story. A historian simplifies and streamlines according to the question he is pursuing.
One principal way historians go about such simplification is by postulating trends. Lots ofthings are going on, they admit, but “by and large” or “on the whole” or “for the most part,” we can identify a general tendency. Most Hollywood films of the 1940s were made in black and white, but most Hollywood films today are in color. On the whole, there has been a change, and we can see a trend toward the increasing use of color film stock between the 1940s and the 1960s. Our task is to explain how and why this trend occurred.
By positing trends, historians generalize. They necessarily set aside interesting exceptions and aberrations. But this is no sin, because the answer to a question is necessarily pitched at a certain level of generality. All historical explanations pull back from the throbbing messiness of reality. By recognizing that tendencies are “for-the-most-part” generalizations, the scholar can acknowledge that there is more going on than she is going to explain.
Periods Historical chronology and causation are without beginning or end. The child who incessantly asks what came before that or what made that happen soon discovers that we can trace out a sequence of events indefinitely. Historians necessarily limit the stretch of time they will explore, and they go on to divide that stretch into meaningful phases or segments.
For example, the historian studying American silent cinema already assumes that this period within film history ran from about 1894 to around 1929. The historian will probably further segment this stretch of time. She might break it down by decade (the 1900s, the 1910s, the 1920s), by changes external to film (say, pre-World War I, World War I, post-World War I), or by phases in the development of storytelling style (say, 1894-1907, 1908-1917, 1918-1929).
Every historian periodizes according to the research program he adopts and the question he asks. Historians recognize that periodization can’t be rigid: trends do not follow in neat order. It is illuminating to think of the American “structural” film of the early 1970s as a kind of response to the “underground” film of the 1960s, but underground films were still being made well into the 1970s. Histories of genres often mark periods by innovative films, but this is not to deny that there may be a great deal of continuity in less innovative works across periods.
Similarly, we ought not to expect that the history of technology or styles or genres will necessarily march in step with political or social history. The period after World War II was indeed distinctive, because this global conflict had major effects on film industries and filmmakers in most countries; but not all political events demarcate distinct periods in relation to changes in film form or the film market. The assassination of President Kennedy was a wrenching event, but it had little if any effect on the film world. Here, as ever, the historian’s research program and central question will shape her sense of the relevant periods and parallel events. (This is one reason that scholars often speak of film histories rather than a single film history.)
Significance In mounting explanations, historians of all arts make assumptions about the significance of the artworks they discuss. We might treat a work as a “monument,” studying it because it is a highly valued accomplishment. Alternatively, we might study a work as a “document” because it records some noteworthy historical activity, such as the state of a society at a given moment or a trend within the art form itself.
In this book, we assume that the films we discuss have significance on any or all of the following three criteria:
Intrinsic excellence: Some films are, simply, outstanding by artistic criteria. They are rich, moving, complex, thought-provoking, intricate, meaningful, or the like.
At least partly because of their quality, such films have played a key role in the history of cinema.
Influence: A film may be historically significant by virtue of its influence on other films. It may create or change a genre, inspire filmmakers to try something new, or gain such a wide popularity that it spawns imitations and tributes. Since influence is an important part of historical explanations, this sort of film plays a prominent role in this book.
Typicality: Some films are significant because they vividly represent instances or trends. They stand in for many other films of the same type.
A particular film might be significant on two or even all three of these counts. A highly accomplished genre film, such as Singin’ in the Rain or Rio Bravo, is often considered both excellent and highly typical. Many acclaimed masterworks, such as The Birth of a Nation or Citizen Kane, were also highly influential, and some also typify broader tendencies.
World History
