p.5
PART I
Genette’s concept of the paratext and its development across disciplines
p.7
1
GENETTE’S PARATEXT
Seuils in context
A major figure in the French academic establishment since the 1960s, Gérard Genette has published almost twenty monographs or collections of essays over six decades and made key contributions to literary criticism and aesthetics. While it is difficult to summarise the achievements of such a long and productive career, it is perhaps helpful – as an introduction to this book, at least – to think of them as dividing into three broad domains, with three corresponding points of focus. The first point of focus is the literary text; the second (and the one with which we will be concerned) is the relation of the literary text to other texts around it; and the third is the relation between literature and the arts. The first corresponds roughly to Genette’s first four major publications (Figures I (1966), Figures II (1969), Figures III (1972), Mimologiques (1976)),1 in which Genette makes seminal contributions to poetics and narratology. The second corresponds to the three works that followed (Introduction à l’architexte (1979), Palimpsestes (1982), Seuils (1987)), in which Genette shifts the focus to transtextuality, or in other words to ‘everything that brings [the text] into relation (manifest or hidden) with other texts’ (Genette 1992, 81). The turn towards the third domain was anticipated to some extent in Fiction et diction (1991) but established more definitively in the two-volume L’Oeuvre de l’art (1994, 1997a); in these and subsequent works, notably Figures IV (1999) and Figures V (2000), Genette broadens out from literature to address questions on the nature of art and aesthetic response, drawing on a vast range of material that includes music, television, art and architecture.2
The work which is the focus of our concern, Seuils (1987), thus dates from the second phase of Genette’s long career and is the third in a trilogy of works exploring a range of types of textual ‘transcendance’ (Macksey 1997, xviii). In Palimpsestes, Genette (1982) summarises the types of transcendence as intertextuality, paratextuality, metatextuality, hypertextuality and architextuality, stressing that these should not be viewed as ‘separate and absolute categories without any reciprocal contact or overlapping’ (Genette 1997b, 7). He defines paratextuality as the
p.8
Genette 1997b, 33
In Seuils, Genette (1997c) carries out an extensive study of the paratext, thus foregrounding an aspect of literary texts which, as he argues, had hitherto been ‘disregarded or misperceived’ (14).4 Some work on individual paratextual features did exist, as Genette acknowledges: Claude Duchet, Leo Hoek, Charles Moncelet and others were working in the domain of ‘titrologie’ (55n1), studying titles of literary works; Genette also acknowledges Jacques Derrida’s discussion of prefaces (196n1) and justifies the brevity of the section on epitexts on the basis that ‘critics and literary historians have long made extensive use of the epitext in commenting on works’ (346).5 However, it is true to say that, particularly since the emergence of New Criticism as the dominant paradigm in the early twentieth century,6 the focus of literary criticism was on close reading of the text rather than consideration of external factors.7
In Seuils, through the interrogation of myriad examples of texts and their paratexts, Genette shows that reading of a text never occurs in isolation from the paratext around it, since a reader never comes to a text, but always to a book; and the book, furthermore, circulates in a context which also affects its reception. Genette describes Seuils as a ‘synchronic and not a diachronic study’ (13), in other words ‘an attempt at the general picture, not a history of the paratext’ (13), and adopts a general, universalising terminology, speaking of ‘the paratext’, rather than specifying the focus more precisely. However, his examples, which are drawn for the most part from the French literary canon, together with his sketches of developments in uses of particular paratextual elements, do edge his study towards an ‘essay on the customs and institutions of the Republic of Letters’ (14) at several points, a tendency which he himself acknowledges. Aware of these limitations, Genette (14–15) himself cautions that Seuils represents neither a universal theory of the paratext, nor even, as a survey of French literary paratextual practices, an exhaustive study: ‘what follows is only a wholly inceptive exploration, at the very provisional service of what – thanks to others – will perhaps come after’. As Chapter 3 in particular will make clear, Genette’s hope that his research might prompt further enquiry has been more than fulfilled, giving rise to studies of paratextual elements in other national literary traditions as well as in relation to other domains of cultural expression.
Genette’s concept of the paratext
What is the paratext?
When getting to grips with any new theoretical framework, it is as well to start with the simplest of questions. At first glance, the question ‘what is the paratext?’ would appear to have a straightforward answer. In the opening paragraph of Seuils, Genette declares: ‘the paratext is what enables a text to become a book and to be offered as such to its readers and, more generally, to the public’ (1), something which ‘ensure[s] the text’s presence in the world’ (1). Even in antiquity, when texts ‘often circulated . . . in the form of manuscripts devoid of any formula of presentation’ (3), Genette argues that the notion of paratext is still relevant, since ‘the sole fact of transcription . . . brings to the ideality of the text some degree of materialization’ (3). The paratext, then, is what turns a text – defined by Genette as ‘a more or less long sequence of verbal statements that are more or less endowed with significance’ (1) – into a physical, material thing, capable of being sold, distributed, read: it is the format of a book and its binding, as well as the various elements that are placed around the text proper in the process of turning the text into a book (title page, cover, blurbs etc.). This emphasis on the physical aspect of the paratext finds expression in the series of metaphors on which Genette draws to further explain the concept: the paratext is a ‘threshold’ (2), a ‘vestibule’ (2), an undefined ‘zone’ (2) between the inside and the outside, an ‘edge’ (2), a ‘fringe’ (2), a ‘privileged place of a pragmatics and a strategy’ (2). This physical definition is the one most commonly used by scholars in translation studies, as we will see in the next chapter.
p.9
However, at several points in the discussion, Genette evokes the possibility of the immateriality of the paratext. For example, when setting out his approach to the study of paratexts, Genette (4) states: ‘A paratextual element, at least if it consists of a message that has taken on material form, necessarily has a location’ (bold added; italics in original). In this scenario, Genette suggests that we can identify something as a paratextual element even if it is invisible, and indicates that the paratext is not a physical thing, but a ‘message’. That we are to conceive of the paratext as a ‘message’ rather than a material element is given further support when Genette describes the proliferation of peripheral elements (‘the jacket, the band, and the slipcase’ (32)) not as an expansion of the paratext, but as ‘an expansion – some will say an inflation – of at least the opportunities (that is, of the possible supports) for a paratext’ (32). The difference between what Genette might have said and what he does say here is significant, for with this wording he asserts that the paratext is not the peripheral element itself; such material elements simply provide opportunities for a paratext – a paratext being, by implication, something else. With this statement, then, Genette appears to complicate his earlier assertion that the paratext is ‘what enables a text to become a book’ (1), for in a hypothetical situation in which the material elements surrounding a text carried no paratextual message (a situation which Genette’s description of peripheral elements as mere opportunities for paratexts would appear to allow), there would be no paratext. It is not clear what the logical conclusion of this position would be. (Would the text fail to become a book? What should we call the material thing that we hold between our hands as we read?) In actual fact, Genette’s careful analysis of the various messages conveyed through peripheral or ephemeral elements – however minor or innocuous those elements first appear – means that a hypothetical situation of this kind could never become a reality. As Genette confidently asserts, ‘a text without a paratext does not exist and never has existed’ (3). Still, the question of what exactly a paratext is remains.
p.10
To explore this further, let us consider Genette’s reflections on the substantial status of the paratext. Genette notes that almost all of the paratexts that he considers are ‘of a textual . . . kind’ (7), but stresses that ‘paratextual value . . . may be vested in other types of manifestation’ (7), including the ‘purely factual’ (7) such as the age or sex of the author, the era in which the text was written, or the genre to which it belongs. Genette (7) explains: ‘By factual I mean the paratext that consists not of an explicit message . . . but of a fact whose existence alone, if known to the public, provides some commentary on the text and influences how the text is received.’ In relation to what he terms ‘contextual affiliation’ (8), Genette suggests that ‘in principle, every context serves as a paratext’ (8), whether or not it is ‘brought to the public’s attention by a mention that, itself, belongs to the textual paratext’ (8). These remarks on the factual paratext indicate that the definition of a paratext depends not on materiality but on function: anything that ‘provides some commentary on the text and influences how the text is received’ (7) is part of the paratext.
The importance of this function-based criterion emerges at several other points in Genette’s discussion, notably when he is discussing the dividing line between paratext and text on the one hand, and paratext and external context on the other. With regard to the first of these divisions, Genette discusses the case of notes added to the text by the author and clarifies that if the note is connected to a text ‘that is itself discursive and with which it has a relation of continuity and formal homogeneity’ (328), then the note ‘belongs more to the text, which the note extends, ramifies, modulates rather than comments on’ (328, my emphasis). The criterion used here for deciding whether notes of this kind belong to the text or the paratext has nothing to do with their material realisation or physical location; rather, the criterion is functional, or in other words based on what the note does. In simple terms, if the note comments on the text, then it is part of the paratext.
Genette uses the same criterion for the second type of dividing line, i.e. that between paratext and external context. In his preliminary observations on the epitext, defined as ‘the distanced elements . . . located outside the book’ (5) and contrasting with the ‘peritext’, which is physically attached to the text, Genette states that ‘the epitext – in contrast to the peritext – consists of a group of discourses whose function is not always basically paratextual (that is, to present and comment on the text)’ (345, my emphasis). When considering such discourses (which include, for example, interviews or correspondence with the author), Genette speaks of them as potentially containing paratextual information, as the following citations make clear, but not as paratexts in and of themselves: ‘we must look on these various exercises [authors’ conversations, correspondence, journals] as occasions capable of furnishing us with paratextual scraps’ (346); ‘the . . . mass of collected conversations constitutes a mine of paratextual evidence (364); ‘recordings . . . are a mine of paratextual information’ (370); ‘let us not conclude . . . that the journal in general is paratextually destitute’ (392). The common point that emerges is that the paratext is not the element itself (the interview, correspondence, recording, journal, etc.), but only that small part of the element which serves to present or comment on the text in question.
p.11
If the answer to the question of what a paratext is, then, is functional rather than material, why does Genette prioritise spatial metaphors and open his book with a description of the paratext that encourages readers to conceptualise it in terms of its physical qualities? Furthermore, why does he structure his book along the same lines, constructing a typology t...