1The ambiguity of the adaptive process in which an original work of art is reshaped into another, sometimes vividly different, form has been the focus of a lot of insightful research. This research, however, has tended to centre on certain issues at the expense of others. Narration, first and foremost, but also perspective, quantitative and qualitative changes have generally taken centre stage (Carcaud-Macaire and Clerc 2004; Cattrysse 1992; Jost 1987; Peters 1980). Other approaches have been underemphasized and chief among them the problematic issue of how to understand the phenomenon of adaptation through the analysis of its objects when the semiotic systems in which they are articulated differ.
2The ‘matters of expression’ (Hjelmslev’s term) which compose the different objects under scrutiny within adaptation studies present a double challenge: firstly they are ‘composite’ (we shall examine other, more suitable, terms below) and secondly they display variation between the initial original adapted product and the eventual adaptation.
3The scope of this variation can be very large indeed. Linda Hutcheon, for instance, proposes to consider adaptation(s) both as formal entities or products but also as processes of creation and reception (Hutcheon, 2006). Additionally she highlights three different modes of engagement (telling, showing, interacting) and examines the bi-directional movements between them. Within the restricted scope of this paper, our understanding of ‘adaptation’ will necessarily be narrower and designate only the adaptation of literary texts to film, probably the most prolific of all adaptive domains, at least from a critical perspective. My theoretical and exploratory approach, however, could easily be extended to other more open perceptions of adaptation.
4It cannot be denied that variation in semiotic systems is indeed a cumbersome obstacle. As it hinders comparative analyses, the issue of interface is often put at the core of the study of (text to film) adaptation. Boundaries are drawn opposing the verbal to the visual, written to spoken language, sometimes even signifiers to signifieds. These boundaries have ultimately often proved dissatisfying and at times reductive to the point of fallacy.
5In order to fathom the mechanisms of the procedure of adaptation and thus comprehend and possibly re-define the balance of power between the elements in presence a methodological framework is needed. It should be flexible enough to allow correlations and therefore contrast between the examined objects. It is only in this fashion that conclusive findings on the nature of adaptation as a process can be gained from observation of adaptations as products.
6Even as we acknowledge the essential differences at the core of the two systems, novel and film, we must find a way to transcend them in order to effectively compare them. To use Hutcheon’s terms, telling does not function like showing does; but to relate what has been told to what has been shown is feasible if we find a way of articulating divergences and convergences.
7To my knowledge, theories of discourse have seldom been used in the study of adaptation, literary or otherwise. It is my contention that these theories are an adequate – maybe the most adequate – model for this type of exploration. Discursive theories (and more precisely systemic-functional multimodal discourse, as developed below) indeed provide us with a frame and terminology which allow us to comment relevantly on the parallels to be drawn between text and film using a grounded, less impressionistic model than before. Michael Halliday’s systemic-functional grammar of the English Language and the extensions that scholars have recently made of his grammar to other systems of meaning that are not necessarily, or only partially, linguistic in scope will serve as a procedural basis to support this claim. The wide spectrum of the (systemic-functional) analytical apparatus customarily used by multimodal discourse(s) allows a flexibility that is more manageable than the traditional but sometimes binary and exclusive contrasts mentioned above.
8Before illustrating this model, using Choderlos de Laclos’ Dangerous Liaisons and Stephen Frears’ 1988 film adaptation, I will proceed to a – necessarily selective – overview of the processes adaptation studies has developed so far, and indeed some of the conclusions it has reached on their basis. This will at times call for detours through different academic disciplines; from traditional theories of discourse analysis to more recent ones, integrating paths taken by film theoreticians (especially semioticians) along the way. This overview will provide a methodological backdrop against which the framework that I propose as most appropriate may be appraised while simultaneously specifying some of its terms.
9The field of adaptation is a complex one, at once fairly specific but also sitting athwart different disciplines: film, literature, linguistics, visual and cultural studies to name but the main ones. With a finger in every pie, as always, the primary risk is unsavoury combinations leading to scientific indigestion. Yet one cannot afford too narrow a scope either, as has tended to be the case in the past. One common pitfall was the tendency to give almost exclusive consideration to content analysis over form. The unwillingness to battle the intricacies of moving from one semiotic system to another probably accounts for this perspective. It then became very difficult to circumvent the “fidelity” benchmark, whether it dared speak its name or not. This was in effect a doomed enterprise virtually always resulting in film being considered the poorer parent. A predictable outcome:
Fidelity to its source text whether it is conceived as success in re-creating specific textual details or the effect of the whole – is a hopelessly fallacious measure of a given adaptation’s value because it is unattainable, undesirable, and theoretically possible only in a trivial sense … [The] source texts will always be better at being themselves. (Leitch 2003, 161)
10More recently, adaptation criticism seems consensually to consider the fidelity issue as outmoded (Cartmell and Whelehan 2007, Stam 2000 and 2005, Hutcheon 2006, Corrigan 1999 etc…). Instead, it suggests viewing adaptation(s) as a form of intertextuality, although what that entails exactly varies from one author to another. The common denominator is their will to go beyond the value judgments, or even denigrations, which adaptations have formerly been subjected to. An intertextual point of view also entails that film is perceived as a type of text, a concept we shall return to below.
11In practice however, when adaptations are used as illustrations or being analysed, there still seems to be a marked preference for adopting a narrative perspective and considering the story as the core of the transposition. Some authors have acknowledged this tendency and justified it by highlighting the distinction between what is directly transferable (therefore observable) and what must necessarily be adapted, between the enunciated and the enunciation (Mc Farlane 1996, he also uses the terms narrated versus narration). Yet no system seems to have emerged as to how to identify these distinctions, account for them or indeed label them. Thus emphasis has remained on the enunciated rather than the enunciation providing a sometimes relevant but often incomplete picture.
12Although commonly circumvented, the issue of materiality remains, to my mind, at the very core of the study of adaptation and can be broached by resorting to theories of discourse. This perspective is not a new one; in 1984 Andrew already stated that:
Adaptation is a peculiar form of discourse, but not an unthinkable one. […] We need to study the films themselves as acts of discourse. We need to be sensitive to that discourse and to the forces that motivate it. (Andrew 2000, 37)
13In this tradition of viewing films as semiotic objects, Andrew has some illustrious forerunners; he mentions Keith Cohen, who in turn acknowledges his debt to Christian Metz. But all three consider film at the metatheoretical level and the practical analytical implications that it should inspire remain unfathomable in their abstraction. To continue the linguistic parallel, their enquiries into film as language are developed at the level of Saussure’s ‘langue’ rather than exemplified through instances of ‘parole’.
14To consider films as ‘acts of discourse’ as these authors suggest, one must embrace a wider semiotic perspective and view them as textual objects, since texts are generally regarded as the object of Discourse Analysis. In the work which may be considered as the cornerstone of his theory, Langage et Cinéma (1977), we can observe how defining films as texts is paramount to Metz’s approach.
Le seul principe de pertinence susceptible de définir actuellement la sémiologie du film est […] la volonté de traiter les films comme des textes, comme des unités de discours, en s’obligeant par là à rechercher les différents systèmes (qu’ils soient ou non des codes) qui viennent informer ces textes et s’impliciter en eux. Si on déclare que la sémiologie étudie la forme des films, ce doit être sans oublier que la forme n’est pas ce qui s’oppose au contenu, et qu’il existe une forme du contenu, tout aussi importante que la forme du signifiant. (Metz 1977, 14)
15Metz uses the label of ‘pluri-code’ to describe the composite nature of film. Even as he steered clear from exemplification, Metz conceded the layering and divisibility of filmic ‘sensorial supports’, all the while minding that it is the film itself which constitutes a text and that any separate analysis must eventually be re-projected onto a more global canvas.
16Other terminologies have been suggested to deal with the heterogeneous nature of film; Stam for instance distinguishes between the single track, uniquely verbal medium of the novel and a multitrack medium such as film.
17But to my mind, a more adequate labelling of film – and especially more conclusive to textual analysis—would be as a multimodal text, which I shall proceed to define.
18The conventional view within traditional discourse analysis (Jaworski and Coupland 1999; Schiffrin 1994; Brown and Yule 1983) is to consider it as the analysis of language in use; some definitions have extended discourse to language beyond use. Its object is commonly regarded as consisting of texts although precise definition of what constitutes texture remains elusive. The approach to these texts has, until recently, been overwhelmingly centred on language, disregarding other semiotic systems, meaning-making systems that function beyond or parallel to language or in combination with it. Although the existence of such systems has been posited through the acknowledgment of multimodal texts (Jaworski and Coupland 1999), i.e. texts which make use of more than one semiotic system, they have rarely been subjected to analysis due to the metalinguistic obstacles that such an analysis would entail. Thus discourse analysis was perceived as monomodal, the focus being primarily on the already rich perspectives offered by language-centred approaches. This model, with its overbearing emphasis on language, left little leeway to consider film as a textual object.
19Very recently however, some linguists have broadened that perspective and devised newfound ways of analysing multimodal texts. Taking Halliday’s (1973, 1994 ) systemic-functional grammar as a starting point, they have sought to apply his instruments to other systems of meaning that are not necessarily, or only partially, linguistic in scope. Thus, subscribing to Halliday’s organisation of meaning into three metafunctions (ideational, interpersonal and textual) which can be adapted to suit different modes (see for instance O’Toole 1994), they have sought to develop ‘grammars’ which are no longer lexico-grammatical but still reveal a system-structure cycle of paradigmatic choices built into a congruent whole which can in turn be decomposed along a rank scale.
20These theories started by extending the conception of language as an isolated phenomenon to include other semiotic, meaning-making, resources. This new outlook initially focused mainly on developing ‘grammars’ of visual design. (O’Toole 1994; Kress and Van Leeuwen 1996) but was soon widened to comprise other modes.
21Kress and Van Leeuwen’s definition of what constitutes multimodality illustrates the openness of this approach, especially in contrast with traditional monomodal discourse analysis:
The traditional linguistic account is one in which meaning is made once, so to speak. By contrast, we see the multimodal resources which are available in a culture used to make meanings in any and every sign, at any level, and in any mode. Where traditional linguistics had defined language as a system that worked through double articulation, where a message was an articulation as a form and as a meaning, we see multimodal texts as making meaning in multiple articulations. (Kress and Van Leeuwen 2001, 4)
22This emphatic shift towards meaning, regardless of the semiotic resource through which it is expressed ties in with traditional adaptation theory. As early as 1948, Bazin had pointed out that ‘faithfulness to a form, literary or otherwise, is illusory: what matters is the equivalence in meaning of the forms’ (Bazin 1997, 74).
23The use of a systemic-functional framework opens the path to a comparative examination, along the same metafunctional guidelines, of the meanings of texts, both in their traditional linguistic form and as film-texts.
24To test this system and the conclusions that may be drawn on its basis, I shall proceed, on the basis of existing systemic functional models, to a comparative analysis of quasi-identical passages in a novel and film.
25Amongst the scholars examining film within a multimodal perspective, Kay O’Halloran (2004) proposes a very complete systemic-functional model which I have chosen to use as a template for my own examination. Her framework is based on the one devised by O’Toole to examine two-dimensional paintings but given the dynamicity of the film medium it also considers resources which result in change, similarity and contrast along with the soundtrack. The categorizing terminology used (in rank constituent units and metafunctional manifestations) stems from Bordwell and Thompson (2001).
26For the sake of brevity I will focus only on the semiotic analysis and not dwell on adjacent issues such as type, form and genre.
- 1 To enable a more seamless comparison – and readable paper – I will be working on the basis of Hele (...)
27The passage which I have selected for analysis is from Choderlos de Laclos’s epistolary novel Dangerous Liaisons1, first published in 1787 and the identically titled film adaptation directed by Stephen Frears in 1988. Of the many reasons which prompted this choice, one stands out: to truly test the capacity of an adaptation to express similar meanings without resorting to similar resources, what better than a text which at face value presents very few adaptogenic qualities? Epistolary novels are generally a challenge to adapters as Bardet and Caron have pointed out:
Comment faire d’une relation écrite personnelle, qui passe par l’abstraction du signe verbal, une représentation, qui relève de l’imitation (mimesis) et qui sollicite avant tout les sens (la vue et l’ouïe)? (Bardet and Caron 2008, 139)
28But, as the scriptwriter (and dramatic adapter) Christopher Hampton mentioned, epistolary novels can also prove liberating since the scarcity of dialogues promotes re-creation and invention. (L’Avant-Scène Cinéma n°498, p.86)
29A further test was presented by the temporal and cultural gaps between novel and film: converting the concerns of XVIIIth Century France to a contemporary international audience entailed certain adjustments. To what extent and especially how this was achieved, using the cinematic means where language had formerly prevailed will then be the focus of this analysis.
- 2 Chapter 27 on the DVD. For Bardet and Caron: scenes 41 and 42 (la surenchère et la rupture). For F (...)
30Let us briefly set the background for our analysis. The marquise de Merteuil and the Vicomte de Valmont are two conniving former lovers who keep themselves entertained through seduction schemes. Their latest endeavour is to deflower young Cécile de Volanges before handing her over to her husband-to-be, a former ungrateful lover of the marquise whom she wants to avenge herself from. Simultaneously the Vicomte sets himself the challenge of seducing the Présidente de Tourvel, a devout married woman renowned for her virtue. To make it worth his while the marquise promises to resume their affair if he succeeds. The scene(s) which I propose to examine relate(s) to the filmic transposition of letter 141 in the novel2. Tourvel has surrendered but Merteuil, feeling betrayed by the Vicomte’s attachment to her, makes it implicit that she will not meet her end of the bargain unless the Vicomte gives up his mistress.
31Our lexicogrammatical analysis will focus on the passage in letter 141 where the Marquise suggests a subtly cruel dismissal of the Présidente to Valmont.
32If we examine the experiential function in this passage we notice that the overwhelming majority of the processes are relational: mostly attributive (especially with current attributes and the verb to be) or possessive (with to have). In itself, this abundance is not unusual; Downing and Locke have indicated that relational processes are ‘extremely common in all uses of English’ (Downing and Locke 2002, 135). Alongside it, however, we may observe a marked scarcity of material processes; when not relational the processes are mental (to find, to preoccupy, to realize…) or verbal (to demand, to say, to accuse…) instead. The verb to take which appears twice, once in vocative form, could be interpreted as material but also as possessive where the attribute – a lover in both cases – is possessed. Among the rare material processes that we may notice, only the verb to send truly stands out, with the Marquise’s fictional female as actor and the inspirational break-up letter as goal.
33The variation (or rather lack of variation) in process types in the passage seems to hint towards a fairly non-dynamic presentation of things in which depictions, and therefore appearances, take centre stage. The façade outweighs direct courses of action. Actions are described, perceived or judged more than actually being undertaken.
34This impression of superficiality, and relative detachment, ties in with the meanings that can be gleaned at the level of the interpersonal function. If we examine person selection in the mood block, we may observe that as the Marquise recounts the story, she distances herself – and also keeps Valmont at a distance - from the account by resorting exclusively to third person pronouns he and she. She never once resumes the first and second person perspectives which are, after all, the norm in letter-writing. In the break-up letter itself, person selection is also marked as we have successive oscillations between the first person I, generally at the onset of each paragraph and a more exterior indeterminate it, which is certainly detached and almost existential, especially in the refrain ‘It is not my fault’.
35This refrain is also salient at the textual level. If we look at the meaning patterns of thematic progression across the text, beyond the clausal level, the utterance systematically recurs at the end of each paragraph. This generates rhematic repetition which, along with the aforementioned features, gives both an almost poetic rhythm and cohesion to the whole passage.
36By combining experiential and textual meanings, we may also pinpoint the marked use of circumstances of condition or contingency. All four if-clauses appear in thematic position, and three of these have the ‘not my fault’ refrain as a dependent clause. This scene-setting function of circumstantial adjuncts as themes (Downing and Locke 2002, 230) contributes to the general impression of detachment.
- 4 At least double: one could indeed argue for a triple or even quadruple process, if we consider the (...)
37Thus at the interpretive level, we could summarize our lexicogrammatical conclusions as an overriding sense of disengagement on the part of the Marquise but also on that of her fictional protagonists in the letter-within–the-letter. This comparative impassiveness, which is to some extent feigned, cleverly mirrors the technical indifference of the double4 mise-en-abyme: Laclos carefully choosing words for the Marquise who in turn skilfully writes to Valmont using terms that she is sure will prompt him into action, all the while navigating between aloofness and implicit threats.
38What now remains to be seen is how Stephen Frears deals with the same passage; which meanings, if any, have operated an unspoiled shift to the silver screen, which ones have undergone subtle or drastic transformation and for what reasons? Are the changes rather ontological, dictated by the nature of the semiotic material itself, or preferential, to convey new meanings which the director wishes to project onto the passage?
39In her Chinatown analysis, O’Halloran chooses the mise-en-scène as basic unit for analysis ‘because the major systems for each metafunction across the semiotic resources are operational at this rank’ (O’Halloran 2004, 117). She discards the sequence as being too complex to be comprehensive and the frozen frame, which excludes sound.
40In this comparative analysis, we will need to extend our considerations beyond the rank of mise-en-scène to that of mise-en-scène complex or scene, where the various shots unfold and also one step further to that of sequence, in which the two scenes under scrutiny are unproblematically linked by cause-effect relations. Our aim is not to be exhaustive but rather to highlight marked choices, those likely to carry meaning and therefore open to interpretation.
41Immediately as we proceed to a close observation of the sequence, the main difference between the novel and film versions strikes us. The epistolary format of the novel is considerably un-cinematic and thus the role of letters has been kept relatively in the background. To resort to multimodal metalanguage, Frears does use letters, but only as props which contribute to cause-effect relations at the representational level. He also uses them compositionally, when they are read out loud in voiceover mode on the soundtrack, for instance. They then serve to organize the unfolding of the narrative by complementing visual imagery, even when the soundtrack is at odds with it. Frears has therefore chosen to symbolically retain some of the pragmatic force of language as embodied by letters. This wasn’t necessarily the only option; Milos Forman in his 1998 version entitled Valmont had come to the opposite conclusion. As Lefère points out ‘Forman considère qu’à transposer un roman épistolaire en film, les lettres en tant qu’artifice narratif éminemment littéraire perdent leur raison d’être’. (Lefère 1994, 141).
42The sequence which is our focus, however, is one of the many sequences in the film which have done away completely with letters and have been re-formatted as in vivo exchanges between characters. In the first scene composing this sequence, the transition is smooth enough; Valmont and Merteuil are re-negotiating the terms of their understanding, she recounts the story and its outcome:
He went round to see his mistress and bluntly announced he was leaving her. Well, as you’d expect she protested vociferously. But to everything she said, to every objection she made, he simply replied: ‘It’s beyond my control’
43The importance of the anecdote and its eventual implications is made modally salient to the viewer by two distinct means. Firstly, as the Marquise begins, the frame which was very wide in the previous mise-en-scène, encompassing the whole stairway, shifts to the two characters framing them from the waist up. The shot then continues throughout the remainder of the scene as it uninterruptedly tracks the characters descending the stairs, first from the side then gently spinning round to face them. Like the Marquise’s narrative, the shot is smooth, the perspective exterior and continuous, not resorting, for instance, to a shot / reverse-shot construction. Secondly, a musical theme (one of George Fenton’s original compositions) begins as Merteuil starts talking; it emerges gently but ends abruptly as she finishes her story and dismisses Valmont. Non-diegetic sound thus helps frame the narrative and direct the viewer’s engagement with the story as it unfolds.
44Additionally, as Bardet and Caron have pointed out (2008, 157), the spatial positioning of the characters in the staircase (as in other staircase scenes which are a visual leitmotiv in the film) is symbolic of the power they exert over one another. At the rank of temporal episode within this scene, although the marquise and the Vicomte move down together, when they reach the bottom of the stairs – and the end of the story – she re-ascends leaving Valmont by himself, thus asserting her superiority over him.
- 5 For the sake of illustration : ‘A man of my acquaintance had, like you, become embroiled with a wo (...)
45As to the dialogue itself, it is a simplified, summarized and modernized5, but otherwise relatively faithful emulation of the novel’s text. There are two notable differences. The first is that the Marquise doesn’t dictate the full break-up text but only the recurring catchphrase. The second is the catchphrase itself which has become ‘it’s beyond my control’. Bardet and Caron mistakenly, as the novel’s translation demonstrates, ascribed this change to English language constraints. The conversion probably has another more deliberate origin which can be inferred if we look at the first occurrences of the clauses in novel and film. Both actually originate with Valmont and the marquise is thus using his own words against him to provoke his downfall. The phrase stands out notably in the film a few scenes prior: an exchange takes place in which the libertines fondly remember their previous affair and fantasize about its renewal but just after gently kissing the marquise’s cheek, Valmont straightens and says ‘As for this present infatuation, it won’t last. But for the moment, it’s beyond my control.’ Her shock and disappointment are instantaneous, readily visible in close-up, and underlined by the brutal eruption of menacing string chords in the musical soundtrack before she turns away. The impact of the words, at the locutionary but also perlocutionary levels (through Merteuil’s reaction) is vividly made prominent.
46That is not the case in the novel where the sentence first appears at the beginning of letter 138 in which Valmont defends himself: ‘I insist, my love, I am not in love with her. And it is not my fault if circumstances force me to play the part.’ (Laclos, 2007, 339) The sentence recedes into the background of the letter and would probably not be remembered by an inattentive reader when the Marquise uses it again.
47The differences in phrasing between the two versions are not really major but do have some understated shades. In the film, a more exterior, quasi divine, force is driving the Vicomte: he is no longer in ‘control’ of himself or his decisions, it goes beyond notions of moral responsibility and ‘fault’.
48The shift in perspective, where Valmont is forced into action against his own volition, is particularly striking in the next scene, that of the break-up itself. Again, this scene confronts characters who were simply correspondents in the novel and here, contrary to the previous scene where the adjustment was more ornamental, the change is significant. Whereas the poetic, somewhat ethereal style of the letter made it, to some extent, unreal, the violence and cruelty of the break-up scene are very real as the words on paper become raw spoken words whose immediate effect we are direct witnesses to. The verbal violence is in turn reinforced by physical violence as Valmont pushes away the Présidente and later pulls her up by her hair before abandoning her. At the representational level, we can infer that Valmont’s saying something is also ‘doing’ something. Verbal and material processes intermingle in aggression, which, although it targets Tourvel, is ultimately aimed at Valmont himself, who is acting ‘beyond his control’. To return to Hutcheon’s comments on the different modes of engagement between adaptations, in this particular case viewing (being shown) is more painful than being told.
49That is all the more remarkable as the text in this scene is closer to the novel and its structure than at any other moment in the film. Additionally to the transfer into dialogues, simplifications and updatings of the language appear once again but the parallels between the two texts nevertheless remain striking. Tourvel says very little, towards the end she is even reduced to sobbing silence. Valmont’s speech, by contrast, is a spontaneous and continuous outpouring of cruelty which matches the letter in both structure and content. There are as many of Valmont’s lines as there are paragraphs to the letter and their substance is closely matched. In both cases, the refrain comes at the end of each one. Such a close adherence to the particulars of a novel over a fairly long scene (over two minutes) is unusual in adaptations, where reduction is usually the norm.
50However, the perspective we are offered in each medium is fairly different. The cold detachment of the letter and the deferring of its effect has made way for the easily recognizable image of suffering visible on Tourvel’s face but also on Valmont’s. Throughout the dialogue, we have an alternation of shots and reverse shots of each character in close-up: nine to-and-fro movements, framing each of Valmont’s lines. Thus as each arrow is shot, we can witness its impact. It is a fairly classical structure which, at the level of the modal metafunction, keeps us engaged, as viewers, with the characters’ predicament.
51Another important feature which, like the stairways, creates a symbolic leitmotiv in the film is the use of mirrors. The film opens and ends with Merteuil sitting at her commode and there are many other mirror scenes in the film. This is a particularly apt symbol as many authors have commented on the importance of appearances and the distortion of reality in the novel, often even conjuring the image of the mirror to ‘reflect’ the multiplicity of perspectives offered by the different correspondents. Michel Delon, for instance, said that: ‘Ces jeux de miroir et de contrepoint tendent à faire disparaître la notion même de réalité dans le tournoiement des points de vue, dans le tourniquet des apparences’. (Delon 1990, 50). In our scene, as he prepares to strike Tourvel with his first – verbal – blow, Valmont looks into the mirror as if to collect himself but also to reassure himself that what he is doing is just playacting, an illusion which cannot have real implications. It also ‘mirrors’ an earlier scene in which he had similarly stared at his reflection to gain inspiration and composure to actually seduce the Présidente. The pervasiveness of mirrors and reflections in the film helps us construe the role of appearances and deception as pivotal. The mirror also creates a distance between the characters but also between character and viewer, generating a breakdown in empathy.
52It is one of the respects in which the film echoes the novelistic features described above. This is also the case at the textual / compositional level where narrative progression is organized quite similarly.
53At other moments, film and novel come apart, most markedly at the level of the modal metafunction, where the film involves the viewer more overtly, and more emotionally, with the characters’ quandaries. The novel seems to keep the reader somewhat at bay. Many Laclos scholars have commented on his deliberate effort to retain an aura of mystery, to keep the characters and their motivations unfathomable, for instance by not including the Présidente’s love letter to Valmont or his last letter to her. A similar process can be observed at the stylistic level in letter 141 in which the marquise’s feigned detachment ends up spreading not only to her fictional creations (the break-up characters) but also, if marginally, to the reader.
54Frears is not as mysterious or as restrained as the novelist, but then neither is film: a director may have the option of showing something or not but once it is shown, it cannot help but be seen. As a modern British filmmaker, engaged in an international production, Frears’ outlook on the themes and centres of focus of the story was clearly different from Laclos’, it would be naïve to think otherwise. But it would be equally naïve to believe that the variation in semiotic systems is not accountable for some of these differences in perception, especially from the point of view of reception.
55Although not exhaustive, this analysis has hopefully put forward some of the marked choices at play in these two texts, and the way(s) in which a systemic-functional framework can help to interpret them into meanings, irrespective of the semiotic material used to shape them. The variation in semiotic systems is not per se an obstacle to understanding the processes underlying film adaptation if we accept a system of analysis which is target-oriented. Acknowledging the composite nature of this(ese) discourse(s) allows us to view and analyse its elements separately; as long as we heed the fact that it is the film itself which constitutes a text and that any specific observations (at any rank within a systemic-functional framework) must eventually be contemplated against the ‘gestalt’ of the text. Films can then be deconstructed into their constitutive elements, sensorial supports, semiotic systems and these can offer ground for meaningful comparison with other, more traditional, types of text such as novels.