Conclusions

Having considered these two examples, we can now go back to the issues we raised at the outset. The involvement of the sound dimension in operations of ‘restoration’ is manifest and comparable to, if not indeed more conspicuous than, that of the video dimension. In the case of Touch of Evil, once Welles had been sidelined, someone had to be brought in to take charge of the sound di­mension before the film could be released. However, the main issues concern the definition of the term ‘restoration’ and the principle, usually considered axiomatic, of respect for ‘the author’s wishes’. The occurrence of both prob­lems in the same context actually denotes an adherence to philological principles which in the case in question were continually suborned in the in­terest of artistic considerations. Leaving to one side the problem of how the term ‘restoration’ should be defined and whether it can strictly be used in the case of genuine operations of ‘reconstruction’ (Canosa 2001: 1072-1083), it is clear that the 1998 text is the result of an operation that borders on philo­logical malpractice. One only has to consider the differing status of the texts by Welles that the authors of this version drew on: while the dossier he compiled in 1958 can indeed stand as a list of corrections to serve for an emendatio of the text carried out at a later date, the annotations contained in the sound notes contain elements which are of considerable interest for a study of Welles’s creative process but cannot properly be used for a restitutio textus. And if one wanted to pursue the matter further, there is one consideration that puts a stop to any serious philological discussion. Quite simply the film is an unfinished product: restoration of the parts that do exist may indeed be car­ried out with respect for ‘the author’s wishes’, but there can be nothing philological about completing it. Here, we can note in passing, another highly contentious issue rears its head: how should one go about establishing, on the basis of the different versions available, which portion of text was in fact com­plete when Welles was excluded? There can be no doubt that ‘the author’s wishes’ is an elusive concept, only to be approached asymptotically; in any case it cannot act as the sole orientation in an operation like the one in ques­tion. Even if it were possible to reconstruct the wishes of the author without any margin of doubt, this would not by any means give us Touch of Evil as Welles himself conceived of the film. A film text is fundamentally the outcome of a negotiation between several authors: hence even if Welles had had charge of the final phases of the film’s post-production, he still would not have been directly involved in the audio. He would merely have reviewed the proposals of his collaborators, rejecting them, suggesting alternatives or adopting them en bloc. And as a matter of fact, this last reflection obliges us to reconsider the idea that Touch of Evil can even be viewed as an unfinished product.

Clearly the logical short circuit that characterises discussion of this film has its roots in the difficulty encountered in handling a text ‘with a complex structure’ such as a film, whose status has been a focus for debate for some time now (Canosa 2001, Mazzanti – Farinelli 2001, Micciché 2002). The problems increase when the text in question is the outcome of the work of a director whose approach came to revolutionise the very concept of ‘author’ as it is viewed in the film world. What has to be emphasised here is that reflec­tions concerning the audio component within this type of debate are still very limited. The issue has recently gained due recognition (Calabretto 2010: 277-286) and certainly deserves to receive a systematic treatment.