Touch of Evil between restitutio textus and creative restoration

Federica Rovelli

Università degli Studi di Cagliari

This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Download full text (.pdf)

Many of the films of Orson Welles as we know them are not exactly as the director himself intended, on account of disagreements that arose during their making or of complex negotiations concerning the rights. Touch of Evil is no exception: Welles was appointed to direct the film almost by chance, following a misunderstanding between Universal and Charlton Heston, and was excluded from the project during the last phases of post-production. In fact the version distributed by the major in 1958 did not have his approval. Nonetheless public and critics alike never seem to have had any doubts as to whose film it was. The initial reaction of André Bazin was surely significant, and may indeed have given rise to the ensuing ambivalence. Hailing the film as one of Welles’s greatest achievements, he dubbed it the first example of a new cinema of ambiguity (Bazin [1958] 2005: 172-6). Over the following de­cades Touch of Evil was restored on two occasions with the aim of providing a version which was closer to Welles’s own wishes. The two subsequent rele­ases, dating from 1976 and 1998, were each hailed as the ‘new, definitive version’, influencing new generations of critics, film directors and spectators and becoming the focus for analytical studies. In view of this history, and before going on to consider more strictly audiovisual matters, it may be hel­pful to go over the stages that led to the release of the three versions currently in existence, to see just what sort of text (or ‘texts’) we are dealing with.

The first version of Touch of Evil (1958) appeared once Welles had been excluded from the project, at the instigation of the head of production of Universal Edward Muhl. Acting in conjunction with Jonathan Rosenbaum, a protégé of Welles, Muhl gave instructions for Ernst Nims to re-edit the film and Harry Keller to film four new scenes in order to clarify the narrative, whereupon he terminated the post-production phase (Leeper 2001: 227). Welles was only able to view the result of this operation once all the way through. He gave vent to his bitterness about many of the editing decisions in his famous dossier, running to 58 pages, which contained precise instruc­tions for 50 modifications (Welles [1957] 2008). For the sake of completeness (for this detail is all too often passed over), it should be pointed out that in some instances Welles specifically approved the end re­sult. It must also be said that Universal did not entirely refuse to meet the requests of the director: more than one scene was revised according to his indications. The second version of the film (distributed in 1976) was made with the declared intent of providing a closer approximation to the film Welles had had in mind. In fact, however, it was simply the version that had appeared in 1958 with the scenes that had been cut by Muhl reinstated, but still with Nims’s re-editing. The third version was produced by a team work­ing under the supervision of Rick Schmidlin and with Rosenbaum, the pupil of Welles mentioned above, in a consulting role. The team included Walter Murch as sound editor, Bob O’Neil in charge of picture restoration and Bill Varney of re-recording. This new version, which its makers referred to as a ‘restoration’, took the 1976 version as an authentic copytext, comprising a negative of the video track and a magnetic master of the audio track with di­alogues, music and effects (DME). The original negative for the opening sequence, without the credits superimposed, was found in one of the cans containing the 1976 version and substituted for the corresponding excerpt in the copytext (Ondaatje 2002: 186). On the basis of this material the team attempted to carry out the 50 modifications specified by Welles. In many cases they relied not only on the dossier but also on other writings left by Welles, in particular nine pages of sound notes he produced for Joe Gershen­son (Tully 1999) – head of music in 1958 – and some annotations conserved by Nims. The sound notes, jotted down while filming was in progress, con­tain indications subsequently used by Henry Mancini when he composed the sound track. In them Welles goes into great detail concerning the genres of music to be used in order to create the feeling of a border outpost that cha­racterises the film. He spelt out the distinction between ‘background music’ – “[…] ‘realistic’, in the sense that it is literally playing during the action” – and ‘underscoring music’ “[….] which accommodates dramatic action and which does not come from radios, night clubs, orchestras or juke boxes”, making it clear that the former should unmistakably predominate over the latter (ibidem). The annotations conserved by Nims are only referred to in an interview with Murch (Ondaatje 2002: 196). According to the latter, they contained some thoughts of Welles concerning the strategies to be adopted for the film’s au­dio. Unfortunately we do not know which year they date from, but there is no doubt that they were not written for Touch of Evil. For the moment we shall suspend judgement concerning the validity of the restoration criteria de­scribed thus far. The prime aim of this essay is to verify how the audio dimension was involved in the various attempts to reconstruct the text, and to show how the modifications that were effected influenced the audiovisual end product as a whole. Then, after examining some examples, we shall end by considering the problem of restoration/reconstruction.