In a book entitled Animated Photography, Cecil Hepworth drew a useful distinction between two types of films that were popular among producers in the late 1890s:[1]

  1. Films of “known” events—recorded, staged, and therefore controllable events.
  2. Films of “unknown” events—actual events that usually unfolded with a certain degree of unpredictability.

Much of the “action” of “unknown” events, of course, was not worth preserving because it wasn’t sufficiently interesting to watch. Hepworth suggested that cameramen “summarize” such events by controlling the nature and amount of the footage they shot: “However promising the beginning may be,” he advised,

long before the end all interesting incident may have given out. In which case, perhaps the best thing to do is to at once leave off turning [that is, filming], without moving the instrument [the camera], and resume turning when suitable incidents recur.

The Beginnings of Shot-to-Shot Continuity

Simply stopping and restarting the camera would produce the effect that we now call a jump cut—an abrupt in-scene transition that often violates continuity. The device was in fact used (as evidenced by the existence of slightly overexposed frames between the two shots), and although it was disorienting, it did underscore the value of dividing real space into shots that would hold an audience’s interest. “It is apparent,” contends British film historian Stephen Bottomore, “that by this time [ca. 1899] the makers of actualities were learning some key lessons about filmmaking. One of the most basic was that you don’t need to show everything.” Bottomore goes on to suggest that the effort of actuality cameramen to come to grips with the unpredictability of their subject matter and the unequal utility of their footage led to the practice of composing films of separate pieces of film—and thus to “the beginnings of a similar kind of shot-to-shot continuity” in both actuality films on the one hand and chase films on the other.

See the moving picture


                  Arrest in Chinatown,

                San Francisco: Shot 3

New-Found Flexibility: Intimations of the Reverse-Angle Shot  

Perhaps equally important was the realization that while the camera was turned off, there was no reason why it could not be moved, especially if some other position offered a better view of the evolving action. Among the results of this newfound flexibility were various prototypes of the reverse-angle shot. Bottomore cites an 1897 Edison Company actuality entitled Arrest in Chinatown, San Francisco (Figure R5.5), which consists of two shots:

  1. Surrounded by a throng on onlookers, a Chinese man is led up the street by a police officer; once the crowd has drawn to a halt, a detective directs the cameraman’s attention to a waiting police cart.
  2. The police cart departs with the arrested man, some policemen, and the detective, who waves to the crowd.

In Shot 1, the movement of the crowd is toward the camera. For Shot 2, the cameraman has effected a “reverse angle” by swiveling the camera in order to take in the departure of the police cart, which moves away from the camera. Clearly, the cameraman wants the effect of movement in opposite directions, but the reversal of movement is inherent in the action being photographed, not created by a change in camera position.

See the moving picture


       Taking President McKinley's Body

       from Train at Canton, Ohio: Shot 5

A more formal attempt at a reverse angle appears in a later Edison actuality, Taking President McKinley’s Body from Train at Canton, Ohio (1901—Figure R5.6). This film is composed of five shots:

  1. A long shot shows the pallbearers receiving the casket from the train and then proceeding left to right at 180-degree angle to the camera.
  2. In a medium long shot, a man who's been standing under some bunting moves out of camera range as the procession passes in the background.
  3. The procession moves on a right-left diagonal in medium long shot.
  4. The procession moves on a right-left diagonal in a closer shot.
  5. The procession continues, shot from behind in a medium long shot.

Shot 5 is supposed to be a reverse angle of Shot 4—the action that crosses the two shots is supposed to be a continuous action viewed from two different angles, one the reverse of the other. Such a cut, however, works only if the two shots are adequately matched—if, that is, the two shots are consistent in their representation of time, space, and action. In matched shots, the visual linkage, despite the shift in camera positions, is immediately clear, and that isn't quite the case in this instance.

Perhaps more interesting is the handling of Shots 1-3. The action in Shot 3 is clearly a continuation of the action in Shot 1, and although the camera changes position, the result is simply a closer and visually more interesting view. In any case, the interpolation of Shot 2 eliminates any other incidental impression that might have been given by a direct or straight cut from Shot 1 to Shot 3. Its purpose is to avoid the effect of a jump cut between the two shots of the procession. In other words, the filmmaker regarded the two shots of the procession as a single scene and preferred not to make a direct cut within the scene, fearing that such a cut, though eliminating dead time, would inadvertently suggest that the procession had changed directions.[2]

The Idea of Sequences Using Independent Shots

Unfortunately, concludes Bottomore, historians tend to ignore the role played by actuality films in the development of continuity techniques. Why? At least in part, he suspects, because the earliest multishot actualities were usually sold as separate films rather than as single films composed of “edited sequences.” Contemporary evidence suggests, however, that by 1901 actuality films “were employing a wide range of new practices . . . pointing toward the idea of sequences using interdependent shots and away from the original view of shots as individual items.”

It isn’t necessary, Bottomore admits, to argue from precedence (the actuality film experimented first) or influence (the actuality film showed the way). In fact, it’s likely that the development of the techniques in question issued from a process of problem solving common to both factual and fictional filmmaking. At the same time, however, the study of the actuality film throws fresh light on the role played by the practice of moving the camera in order to get a more satisfactory view of a subject: the presence of sequences in which the camera has been moved reaffirms the importance of the realization that films could be reconstructed as sequences from a series of “interdependent shots.”

[1] This reading is based on Stephen Bottomore, “Shots in the Dark: The Real Origins of Film Editing,” Sight and Sound 57 (Summer 1988), pp. 200-04; rpt. in Early Cinema: Space, Frame, Narrative, ed. Thomas Elsaesser (London: British Film Institute, 1990), pp. 104-13. Bottomore cites Hepworth’s Animated Photography: The ABC of the Cinematograph (1900; rpt. New York: Arno Press, 1970). Hepworth is the subject of Biographical Sketch 5.1.

[2] On Edison’s McKinley films, see Charles Musser, Before the Nickelodeon: Edwin S. Porter and the Edison Manufacturing Company (Berkeley, Los Angeles, Oxford: Univ. of California Press, 1991), pp. 184-90, at (accessed June 22, 2016).

Back to top

Back to CHAPTER 5/Part 2