There is something quixotic about the pursuit of the ‘first screenplay’, which as Steven Maras warns ‘has proven unhelpful’ because ‘the search for firsts and origins can have the tendency to “fix” the landscape in particular ways, leading to a reductive view of the development and institutionalisation of screenwriting’.¹ Yet posing the question can nevertheless function as a worthwhile heuristic device, a means of opening up, if not necessarily providing definitive answers to, other, more substantial questions: What is a screenplay? In what ways, if any, does it function as a ‘planning’ document? Does it presuppose the existence of a written text? If so, must that text possess definable formal properties? In what ways is it distinct from other forms of writing? Is it a product and corollary of the division of labour in early studio systems? Is the terminology—scenario, continuity, screen play, screenplay and so on—significant?

No script from this period has survived, partly because at this date there was no need for anything possessing even the functions, let alone the form, of the modern screenplay. It is a logical inference that it was the introduction of longer and more complex narrative films in the early years of the new century that heralded the beginning of screenwriting. Narrative films have been estimated to comprise just 12 per cent of films made in 1900, but 96 per cent of films in 1908,³ with actualités outnumbering narrative films until as late as 1906.⁴ The length of films increased from the initial limit of 50 feet to 250–400 feet by 1900, and 300–600 feet by 1903. The greater duration entailed the construction of sequences linked into a coherent narrative. The ‘seemingly obvious yet strangely elusive’ conclusion drawn by Edward Azlant is ‘that this vital transition involves the birth of screenwriting, a craft that could be fairly described as the prearrangement of scenes, right up to the present’.⁵

Stempel’s description of a 1902 film directed by Edwin S. Porter advances a familiar argument: ‘It is obvious from the lavish production of Jack and the Beanstalk that considerable pre-production preparation was done, which suggests the use of a scenario of some kind.’¹¹ Yet while a complex sequence of events may require careful preparation, it does not follow that the plans will be translated into textual form. Accordingly, most commentators prefer to speak of the design, rather than the scenarios, of early films. For example, the first of the Lumière brothers’ pictures, Sortie d’Usine (1895), is not simply a documentary record of workers exiting the Lumière factory. It is clearly a staged event: the doors open on cue, and the workers exit the frame in an orderly fashion, indicating rehearsal. The film, then, ‘reflects a number of carefully chosen decisions about sequential narrative’, and ‘its organization reflects an order and direction akin to the movement one associates with traditional plot structure’.¹² This is even more apparent in L’Arroseur Arrosé (1895), the Lumière film widely accepted as ‘both the first film to tell a story and the first film comedy’.¹³ A boy stands on a gardener’s hosepipe: when the gardener inspects the nozzle the boy removes his foot, and the gardener is sprayed with water. Louis Lumière recalled in 1948 that ‘the idea of the scenario was suggested to me by a farce by my younger brother Edouard’.¹⁴ It is fascinating, if perhaps misleading, that many years after the event Louis would use the term ‘scenario’ in connection with a film of 1895; ‘scenario’ here probably implies a mental concept, or the filmic mise-en-scène, L’Arroseur Arrosé being ‘the prototype of what Méliès called “vues composées” or “artificially arranged scenes”’.¹⁵ There is no need to posit the existence of a written script for such a simple narrative, even though ‘the event depicted is not discovered but created, not recorded but acted, the whole a unified design’.¹⁶

In studies of early screenwriting, there is a widespread confusion between a source and a scenario. Several films of the late 1890s are demonstrably based on surviving written texts, which arguably performed the function of screenplays. The most commonly cited examples, however, are stage plays and newspaper articles that were not written with films in mind. They therefore provide no evidence of either screenplays or screenwriting practice, although they do reveal early film-makers working methodically from textual adaptation through shooting and perhaps to editing. The temptation to describe such texts as scenarios or screenplays, however tentatively, is a logical error that, repeated sufficiently often, has led to a distorted view of screenwriting activity in the period.