Navigation – Plan du site

AccueilNuméros5VariaCognitive and Semiotic Aspects of...


Cognitive and Semiotic Aspects of Endings in Self‑Narrations: The Example of Germaine de Staël

Anna Cabak Rédei
p. 315-332


Le but de cet essai est de discuter et d’illustrer les relations entre la vie et les narrations sur la base de recherches faites en sémiotique et en psychologie, inspirées d’abord, de M.M. Bakhtine, Jérôme Bruner et Katherine Nelson. L’envergure de cet essai est multidisciplinaire du point de vue théorique, tout en aspirant à illuminer les aspects cognitifs et sémiotiques de l’écriture littéraire. Pour ce faire, quelques exemples de la littérature du xixe siècle ont été choisis, à savoir, Dix années d’exil (1820–1821) et Corinne (1807) de Germaine de Staël (1766–1817), auteur français et philosophe de la littérature et de la politique. En plus, et peut-être contrairement à ce que l’on pourrait anticiper, la discussion sur la vie et sur les narrations se concentrera sur des fins, littérairement et littéralement. Germaine de Staël a terminé ses deux œuvres autobiographiques d’une façon assez particulière, comme si sa vie était entrelacée avec ses textes. Pourquoi se concentrer sur les fins ? Premièrement, parce que je trouve que la fin d’une narration est importante pour une étude qui aspire à analyser les différences entre la vie et les narrations. Les genres autobiographiques comme l’autobiographie (non-fictive) et le roman autobiographique (fictif) sont définis à partir de conventions spécifiques à la littérature, déterminantes pour les choix faits par l’écrivain et pour la façon dont le lecteur peut les lire. Deuxièmement, en étudiant les fins de ces deux œuvres, la relation compliquée entre la vie et la narration fait surface ; comment peut-on captiver le courant de la vie dans une narration, nécessairement construite, limitée et, dans un sens, exceptionnelle ? Si la fin contient une clôture ou une conclusion de quelque sorte — par exemple, concernant la vie d’un personnage (autobiographique ou fictif) — cela implique inévitablement une limite entre la vie et le texte, un moment où l’auteur a arrêté à écrire et le lecteur a arrêté de lire.

Haut de page

Entrées d’index

Haut de page

Texte intégral

[t]here is nothing absurd in speaking about the narrative unity of life, under the sign of narratives that teach us how to articulate narratively retrospection and prospection. / The conclusion of this discussion, then, is that literary narratives and life histories, far from being mutually exclusive, are complementary, despite, or even because of, their contrast (Ricœur 1990, p. 160 eng. tr.)

1. Introduction

1How do narratives relate to life, and how may these help to create meaningful interpretations of ourselves in relation to the world? These are some questions that occupy the developmental psychologist J. Bruner in his seminal work Acts of meaning from 1990. And as the semiotician C. Paolucci (2012, p. 304) affirms, if one wants to study these structures one has to do it through empirical analyses (as narratives since they are prototypical) in order to find the structures of meaning. But narratives may also be regarded, more generally, as heuristic models (Paolucci, 2012, p. 304). Thus, the scope of this essay is to illustrate and discuss the relations between life and narratives on the basis of semiotics, foremost M.M. Bakhtin, philosophy, notably P. Ricœur and the work in psychology by foremost J. Bruner and K. Nelson. Moreover, other psychological and literary research will be used in my attempt to illuminate the connection between narratives and life. In order to do so, some examples from the 19th century French author, literary and political philosopher Germaine de Staël’s (1766-1817) self-narratives will be used. The focus will be on her autobiographical travel account Dix années d’exil (1820-1821) [Ten years’ exile] and her autobiographical novel Corinne (1807), and more specifically on the endings of these self-narratives. Why studying the endings?

2Firstly, because I think that the ending of a story is important for an inquiry aiming at scrutinizing differences between life and narratives. Autobiographical genres such as autobiography (non-fictive) and autobiographical novels (fictive) are defined according to certain literary conventions, deciding the author’s specific narrative choices and the reader’s specific way of reading the text. Secondly, by studying the endings of these works, the complicated relation between life and narrative comes to the fore; how is it possible to capture the continual flow of life in a narrative, necessarily constructed, limited and, in some way, exceptional? Whether or not the end involves a closure or a conclusion of some sort — for instance concerning the life of a character (autobiographical or fictive) — it inevitably means a limit between life and text, a moment when the author stopped writing and when the reader stops reading. By studying the endings the analysis may also go in the other direction, focusing on literary text as a continuation of the self and life, within the frames of P. Ricœur’s (1990) philosophy of the self (and the other) we may call this continuation for narrative identity.

3The disposition of the essay is as follows: The discussion will start with a brief overview of the autobiographical genre per se from the viewpoints of 1) semiotics and 2) philosophy and 3) psychology. Concerning semiotics a special interest will be on M.M. Bakhtin’s philosophy of the dialogic word, and classifications of genres, notably his account of similarities and differences between self-narratives and travel accounts. Those are of a particular interest here, since Germaine de Staël’s autobiographical travel account Dix années d’exil, is a mixed text, containing elements of both genres. The philosophical aspects of the problem of self-narratives and identity are thoroughly discussed by P. Ricœur, and are illuminating indeed in the analyse of the two autobiographical texts by Germaine de Staël. Thirdly, I’ll account for a study in cognitive psychology where it has been shown that endings of self-narratives have special implications for how the reader evaluates the life of the protagonist. I’ll also recount for insights made in folk psychology, as they have been elaborated by D.D. Hutto and S. Gallagher. These accounts are then followed by analyses of Germaine de Staël’s autobiographical works.

2. Self-narratives from the point of view of literary semiotic studies

4The Russian philosopher of language, semiotician and literary scholar M.M. Bakhtin devoted his writings to explore the dialogical word and to classify literary genres. How does a text relate to the socio-cultural context in which it emerged? How do words enter the text? In what way are they dialogical? These questions were thoroughly explored by M.M. Bakhtin in his seminal work on Dostoevsky, the book Problems of Dostoevsky’s poetics (1963). But he was also interested in the way the hero was organised within a literary work, an interest that extended to (auto)biographical genres:

In a biographical novel (especially autobiographical and confessional), the only essential change in the hero himself is his crisis and rebirth (the biographical hagiographies of the crisis type, Augustine’s Confessions, and so on). The conception of life (idea of life) that underlines a biographical novel is determined either by life’s results (works, services, deeds, feats) or by the category of happiness/unhappiness (with all of its variations) (Bakhtin 1979, p. 17 eng. tr.).

5When it comes to Germaine de Staël in Corinne and Ten years’ exile, I’m tempted to say that M.M. Bakhtin’s conclusive analysis of categories determining the autobiography (as well as the biography, M.M. Bakhtin does not make an analytical difference between the two) do fit. However, Ten years’ exile is a hybrid when it comes to genre: in parts it is both an example of a travel account and an autobiography. This will be discussed more in detail below.

6Her most conspicuous autobiographical works, thus, seem to be dominated by the category “happiness/unhappiness”, the latter being the dominating: death in Corinne, exile in Ten years’ exile (exile perhaps, being even worse than death, according to Germaine de Staël). In fact, unhappiness might have been Germaine de Staël’s main topic. The opening lines in Réflexions sur le suicide [Reflections on suicide] — written in Stockholm 1812, where she stopped for some months on her grand tour as an exile, before reaching England in the spring 1813 — reflects this: “It is for the woeful one has to write […]” (de Staël 1814, 1. Translation ACR). Now, the organizing dichotomy of “happiness/unhappiness” needs further explanation.

7When M.M. Bakhtin (1979) refers to the categories hero, time and plot, he discusses five types of texts. Two of those are of the utmost interest for the present study: the travel account and the autobiography. In the former it is the environment (cities, cultures, nationalities and such) that is in focus, not the hero or other human beings. If pictures of human being exist, they are static and room is given to their personal evolution. The world is described in differences and contrasts, and life oscillates between different opposing conditions: success/failure, happiness/unhappiness, victory/loss etc. The travel account is based on a static worldview and organized in binary polarities, and the world emerges as fragmentary and disintegrated. The time perspective is not very developed and historical dimensions are missing. The biographical time is not considered and therefore descriptions of the hero’s ageing, the way from childhood to old age, are lacking. The only time dimension to be found is the “adventure time”, which is characterized by its immediacy: time markers referred to, in an arbitrary way, are moments, hours and days (Bakhtin, 1979, pp. 10-11 eng. tr.).

8According to M.M. Bakhtin, autobiographies never existed in an unmixed form. It is only a principle for creating the hero. Moreover, all self-narrations are created in relation to something — or someone(s) — outside the Ego in time and space, they are, otherwise put, dialogic (Cabak Rédei, 2007). There are similarities between the heroes in the autobiography and the travel account in the sense that they are both rather static and are not developed on a personal level, apart from when they focus on crises or happiness/unhappiness. Thus, the binary opposition of happiness/unhappiness also arranges the faith of the autobiographical hero, and thereby resembles the one found in the travel account. However, the plot differs in the two different genres. The autobiography is built on characteristic elements that are common to all life-telling narratives: birth, school years, marriage and so on (Bakhtin, 1979, pp. 10-11 eng. tr.). Also, the use of descriptions of minor characters and the environment in the autobiography, diverges from the travel account. As discussed earlier, these elements may be pictured rather superficially in the travel account, whereas in the autobiography they may on the contrary be given an important role in the relation to the hero. According to M.M. Bakhtin:

This makes it possible, in depicting the world, to surmount […] the naturalistic fragmentation of the travel novel […]. Because of the link with historical time and with the epoch, it becomes possible to reflect reality in a more profoundly realistic way. (Position, occupation, and kinship were masks in the travel novel, […]; here they acquire a life-determining essence. The links with secondary characters, institutions, countries, and so on are no longer superficially adventurist by nature) (1979, pp. 10-11 eng. tr.).

9It is the hero who moves in the travel account, deprived of personal traits, but this is not the case generally with the autobiography. The hero’s character is given from the outset in the travel account and does not change in the course of the narrative. The events, M.M. Bakhtin writes, “shape not the man, but his destiny” (1979, p. 19 en. tr.). Thus, the aims of the travel account and the autobiography are not primarily dedicated to the continuous development of the heroes, their focus lie elsewhere. But this is not always the case, as the well-known example of Rousseau’s Confessions shows; here the development of the self constitutes the central theme of the book. On the whole, M.M. Bakhtin’s dichotomy seems too sharp, over time a vast variety in the genre of autobiography has emerged. However, it is useful in the analyse of Germaine de Staël’s Ten years’ exile because when it comes to her autobiographical travel account the discussion of genre becomes complex in fact. The book is composed of two parts: the first (covers the years 1800-1804) is more autobiographical in a traditional sense and the other (covers the years 1810-1812) is rather a travel account. That is, several genres come into play in the book, which may explain the peculiar ending of it.

10At the core of M.M. Bakhtin’s reasoning lies, as noted above, his conception of the dialogic word. The text is always connected to the outside world, to the sociocultural context in which it emerged. The text is a result of the author’s appropriation of the other’s word. The dialogic word is thus constituted of two voices, the author’s and the Other’s.

11This assumption about the dialogic word has bearings also on his conception of what an autobiography is, alongside with the principle, tied to the genre, of orchestrating the hero. This is highly relevant when approaching Ten years’ exile, since it might on the one hand be considered to be a project to save the memories of Mme de Staël’s father and herself, in the light of Napoleon’s persecutions, on the other hand it might be seen as a travel account (Germaine de Staël’s telling about her exile). Bakhtin stresses not only the categories happiness and unhappiness as determinants for the autobiography, but also “life’s results” (1979, p. 17 eng. tr.). These ways of reasoning, especially the latter aspect, are very much in line with recent results from experimental psychology and have bearings on the aims of this essay focusing on endings of self-narrations and posthumous effects (Rozin & Stellar, 2009).

12To sum up: Ten years’ exile is fragmented as M.M. Bakhtin defined the travel account to be, and the protagonist is not continuously developed, not in the first part nor in the second part. Germaine de Staël’s self-narrative is a good illustration of what one, as a reader senses, namely that the book’s “focus lies elsewhere”. The absence of a closure, that is, the absence of final marker that ends the narrative, hinders the reader from evaluating the narrative identity, and the narrative as such.

13Before moving to cognitive psychological aspects of the problem of closures of biographies, something will be said about P. Ricœur’s discussion on the personal identity vs. the narrative identity, which sheds further light on the problem of closure and self-narratives.

3. The problem of identity and the philosophy of the self

14There is a link between M.M. Bakhtin and P. Ricœur (1990) in their view of the role of the other for the formation on the ego, in the sense that the other is existentially necessary to the ego. In P. Ricœur’s terminology this is put in the dialectics between ego-idem and ego-ipse, between sameness (for instance, the genetic code) of character and “selfhood, self-constancy” (Ricœur 1990, p. 148 eng. tr.) which is the pole through which the other enters — the other through whom one passes. The former answers to the question “What am I?” and the latter to the question: “Who am I?” (in the sense of keeping one’s word). This dialectic constitutes what P. Ricœur calls “l’herméneutique du soi”, and it is now, within narrative theory, that the problem of it becomes visible. This is so, because constancy in time creates a confrontation between the sameness and selfhood, what is subject to change in the identity. They are both models of constancy in time, but of a contradictory kind. Only through narration might these two poles in the character (personal identity) be modulated, more precisely in the study of the role, le personnage (narrative identity). Moreover, narratives allow actions to be viewed from two perspectives: descriptive and prescriptive. How may the discussion of the narrative identity contribute to the discussion of the personal identity? For now, P. Ricœur’s important question will be kept in mind, in order to be dealt with later on.

15P. Ricœur gives a known example from literary history to illustrate an example of “fictions of the loss of identity” (1990, p. 149 eng. tr.): Robert Musil’s The man without qualities. In the novel the dissolution of the identity of the role is parallel to the dissolution of the narrative, which leads to a crisis of closure of the narrative. What type of dissolution is The man without qualities an example of P. Ricœur answers: the selfhood, the ipse deprived of idem, “the support of sameness” (ibid.). Now, H. White’s take on the crisis of closure is in the spirit of P. Ricœur when he writes, in connection with discussing the importance of closure for any narrative: “What else could narrative closure consist of than the passage from one moral order to another?” (White, 1987, pp. 1-25). The same idea is brought forth also by P. Ricœur when he says that literary imitations of actions (mimesis) are, like the ones in real life, subjected to embodied and worldly conditions under which we live. “Worldly” is not the same as “earthly”, but instead connected to the phenomenological idea of the lifeworld (the world we take for granted), once denoted by Husserl. Thus, the importance of context to our ability to create meaning is made clear: without praxis (habits) actions are deprived of their meaning. P. Ricœur (1990, p. 158 eng. tr.) writes:

[T]he idea of gathering together one’s life in the form of a narrative is destined to serve as a basis for the aim of a “good life”, the cornerstone of his [MacIntyre] ethics, as it will be of ours. How, indeed, could a subject of action give an ethical character to his or her own life taken as a whole, if this life were not gathered together in some way, and how could this occur if not, precisely, in the form of a narrative?

16A problem connected to narratives as a form of giving meaning to life (if we may agree to that closures are prerequisites for a hermeneutics putting wholes and parts into play) is the relation between author, narrator and character on the level of fiction (where they are distinct) in comparison to the autobiography: “When I interpret myself in terms of a life story, am I all three at once […]?” Another problem in this connection is of course the ideas of beginning and closure, which differs in life and in narrative form. There are no beginning, nor end in the course of life that are equivalent to the beginning and end respectively in narratives. Childhoods are clouded because of lack of memories, and deaths may only be recounted for in posthumous narratives and stories (Ricœur, 1990, p. 160 eng. tr.). Well, may “one still speak of the narrative unity of life?” (Ibid.). Ricœur answers:

As for the notion of the narrative unity of life, it must be seen as an unstable mixture of fabulation and actual experience. It is precisely because of the elusive character of real life that we need the help of fiction to organize life retrospectively, after the fact, prepared to take as provisional and open to revision any figure of emplotment borrowed from fiction or from history (p. 162 eng. tr.).

17Thus, a philosophy of identity has taught us that the unity of life implicates some sort of closure that can only be provided by narratives, which are to their nature meaningful wholes that come into being in relation to their parts. Now, what does this mean and how may we deal with this empirically? Some examples from cognitive psychology might elucidate this question and be a good complement to the initial philosophical inquiries done so far.

4. Life-narratives from the point of view of psychology and cognitive science

  • 1 Interestingly enough the results showed that there was almost no difference in answers between the (...)

18PRozin and J. Stellar (2009) wanted to study the importance of closures for interpreting self-narratives from the perspectives of goodness of life and happiness of life. The inquiry was designed to shed some light on these problems by using experimental set ups in order to study how posthumous events change the way readers first conceived of a biographical account. Participants were given a questionnaire with 18 short life stories: “‘For you personally, how good is this life?’, ‘How good is this life in general?’, ‘How good a life would this be for your child to live?’, ‘How much total happiness is in this life?’, and ‘How much total unhappiness is in this life?’” For each story the participant was asked to answer five questions by rating on scale from 1 to 100 the amount of 1) goodness and badness of life respectively and 2) happiness and unhappiness respectively (Rozin & Stella, 2009, p. 279).1 Depending on the answers, P. Rozin and J. Stellar (2009) decided to keep the first, fourth and fifth question in their analysis. Only the four experimental stories (which could vary from 2 sentences to four sentences in length) designed to test the posthumous effect were included in the actual study, together with a “neutral” baseline story, being “flat” to its nature and including no particular events. However, the results showed that this story did not differ significantly from the positive stories in terms of ratings of goodness of life, happiness and unhappiness of life respectively, but did so significantly in comparison with the negative ones.

19Two of these stories described a life course to the point of death, and each of these two was followed by an identical story including a posthumous reversal, which continued after death and gave information of an opposite valence of the foregoing presented.

20The results of the study showed that when given a posthumous ending, the “meaning” of the previous ending which might have been favourable is reversed to the opposite in concordance with the posthumous version. Thus, Rozin and Stellar show the importance of the (posthumous) ending for the reader’s conception of the protagonist’s life, especially for the evaluation of “goodness of life”. Rozin and Stellar writes:

That is, posthumous events affected judgments of the goodness of lives, and even the judgment of the happiness of lives. The effect was always in the direction of moving the total life evaluation in the valence direction of the posthumous event. The mean change for posthumous effect size for happiness/unhappiness is about one half of the change for goodness of life (2009, p. 275).

21These results of Rozin and Stellar are interesting in our connection, since they seem to indicate something central in human understanding of narratives and their connections to personal life experiences. In this case one might bring in some insights made by Frederic Bartlett (1932) in his famous experiment using a story called “The war of the ghosts”. Results of that study points to the fact that as humans we are depending on schemas, acquired early on and used (consciously or unconsciously) in order to understand human interactions in the sociocultural context belonging to us. Schemas help us to fill in gaps in stories, when coherence is missing.

In The War of the Ghosts events follow one another, but their connexion is not, as a rule, actually stated. The situation is like that which would confront the spectator of one of the earlier cinematograph films with the usual explanatory connecting tags omitted. The subjects, in the experiments, supplied the tags, but without realising what they were doing: […] (Bartlett, 1932, p. 86).

22The posthumous effect discussed by Rozin and Stellar (2009) might also perhaps partly be explained by the participants’ readiness to fill in the gaps needed in order to make the story coherent and congruent with its new ending, despite the fact that its valence is incongruent with the first emotional understanding of the story. Bartlett’s experiment was designed to study memory, how memory is influenced by time and background (cultural) knowledge (schemas). His results illustrated that with time we tend to fill in memory gaps with knowledge from our own culture, rather than from what we have perceived in the original story. The “gaps” between what is perceived and what is reported, the observer fills in with some help from previous experiences. He or she picks from them something that seems “fit” to the situation (Bartlett 1932, p. 14). Memory and its connection with narratives is something that also occupies J. Bruner in his research on autobiographical accounts (1990). Bruner, as does Bartlett, stresses the feature of coherence inherent in narratives:

[I]ts inherent sequentiality: a narrative is composed of a unique sequence of events, mental states, happenings involving human beings as characters or actors. These are its constituents. […] Their meaning is given by their place in the overall configuration of the sequence as a whole — its plot or fabula (Bruner, 1990, p. 43).

  • 2 Since pictorial narratives are somewhat differently composed, see Sonesson (1995).

23To this we might perhaps add that the ending has a special role to play in narratives for the understanding of the plot. According to J. Bruner (1990) this particularity of narrative, linguistic narrative to be precise2 has similarities with how memory is constructed, although such a claim has been challenged (Nelson, 2003, p. 126). In spite of that, the developmental psychologist K. Nelson (2003) proposes that “memory for temporally sequenced events”, something that is a property of episodic memory (autobiographical memory), may have developed “as functional for certain kinds of knowledge: how to build a nest, for example, or how to locate and dig out termites” (Nelson, 2003, p. 126). Perhaps more importantly in our connection, memory can, according to Nelson be “ascribed with an “epistemic value” as it keeps an account of both common and novel events, it gives a platform for actions in a now as well as anticipating the future, and its needs” (ibid.). The common feature to the chain of events in practice and in narratives seems to be sequentiality, starting with a beginning and ending with a closure.

24The philosophers S. Gallagher and D.D. Hutto (2008, p. 20) make similar evolutionary connections from the point of view of cognitive science and the term “embodied practices” (including “emotional, sensory-motor, perceptual and nonconceptual” practices) and narrative competence as it develops in the child. In doing so they argue against theory-of-mind methods (“theory theory” and “simulation theory”) since they do not consider primary perceptual intersubjective practices that the child is engaged in from early on, implicitly already when new-born (and persistent through out development). Instead they propose a new take on the matter, comprised in the Narrative Practice Hypotheses (NPH) (Gallagher & Hutto, 2008, p. 17).

25Secondary intersubjectivity, a stage in the infant’s development that occurs at the age of around 1 year, implies the ability of shared attention in contexts. That entails to observe another person’s movements and expression as having a meaning and purpose in a given situation, or “pragmatic context” (Gallagher & Hutto, 2008, p. 23). Contexts (norms) thus influence the way we perceive others, and therefore our way of relating to and interacting with others depends on that. With time, as we practice cultural norms, they become embodied. These habits and practices form the background against which we understand undertakings of others in our ordinary world, at the same time as it forms the background against which “we make sense of reasons for action via narratives of the folk psychological variety” (Gallagher & Hutto, 2008, p. 28, italics in the original). With reference to empirical work by J. Bruner, among others, S. Gallagher and D.D. Hutto propose that there are important connections between our abilities to understand narratives and other written texts. Previously, the idea that narratives are inherently sequential was suggested, and S. Gallagher and D.D. Hutto (2008, p. 30), when defining the nature of folk psychological narratives (Little Red Riding Hood being prototypical), add yet another definition: they “are distinguished by being about agents who act for reasons”. Moreover, these narratives may function, according to NPH, “as objects of joint attention in early learning” (Gallagher and Hutto, 2008, p. 30). They may also function as heuristic devices and help us out when we encounter actions that deviate from norms and that we do not fully come to grips with.

26Narratives more generally speaking provide us with knowledge about the lifeworld, the ordinary world we take for granted (Schutz, 1964), that is: social roles that come with norms governing different contexts such as restaurants, banks, shops, cinemas etc. Our actions—performed in one or another context—provide the other with information about the reasons for our actions. To sum up the core of the implication of NPH: There is, thus, no need for a theory of mind because actions are displayed right there where we can perceive them, and guide us in our efforts to understand others and interact with them.

27NPH provides yet another aspect of the logic of the posthumous effect, the importance of context as a holistic framework (which needs a closure of some sort to be holistic) for understanding action and the reasons behind it — the importance of viewing narratives as constituting of wholes and parts have been discussed also earlier in connection with M.M. Bakhtin (the dialogic word) and P. Ricœur (narrative identity; narratives as “unity of life”). Now, narratives are “‘instruments of culture’” as D.D. Hutto (2008, p. 37), at the same time as they are models of culture. This notion of narratives has similarities with the social psychologist and semiotician J.V. Wertsch’s view of narratives being “cultural tools”, involved in “mediated actions”, that is, in processes of semioisis where several narratives are put into play in the creative act of the “writer” (Wertsch, 1998). Thus, every narrative is a result of a dialogic interplay between other narratives and the writer. To uncover the meaning, is to bring forth the relationship between sociocultural context and the narrative (or any other artefact). The process is triadic involving the interpreter, the agent (writer) and the narrative (the mediated action) (Wertsch, 1998).

28Germaine de Staël, at the time for her writing of the autobiography, had important matters still to be settled in her life, to mention a few: the “Necker millions (2 million francs once lent to the state by her father, and partly paid back in 1815), the political rehabilitation of her father and herself. The ending of the book, which we will have a closer look at next, might be a reflection of that state of fact (if we agree with P. Ricœur and his philosophy of the unity of life, identity and the entanglement of life courses and narratives), since it fails to provide us with a closure — and therefore also fails to give the reader a holistic framework for understanding the actions undertaken by the protagonist, that is, Germaine de Staël herself.

5. Ten years’ exile : Germaine de Staël’s autobiography

29Germaine de Staël ends Ten years’ exile as follows:

  • 3 “J’ai toujours été fort sujette à l’ennui et, loin de savoir m’occuper dans ces moments tout à fait (...)

I have always been very subject to ennui, and far from knowing how to occupy myself at those moments of entire leisure, which seem destined for study… (de Staël, 1821, p. 433).3

30Why this ending? It ends with “…” as her life would go on endlessly. Some have explained this peculiar ending by saying that Germaine de Staël abandoned the work in order to complete and start others (more urgent perhaps to her?). It might be as simple as that, but in the light of the insights made in philosophy and psychology, might there not be other reasons more directly involving Germaine de Staël herself and her life? What de Staël wanted to seize and describe in her autobiography was still going on outside the text: her life, her struggle with Napoleon and the urge for glory for herself and her father. Germaine de Staël could not foresee (to her own distress) the end of the story of Napoleon’s persecutions — which probably was one of the main points of the book, that is, to establish the “winner”, who ultimately, one might on good grounds assume, would be her father and herself. Thus, she was unable to give her self-narrative a form of “unity of life”, that only narratives may provided, following P. Ricœur here, and the general view that narratives only may provide us with a holistic perspective on life courses.

31Moreover, Germaine de Staël might have been well aware of the historical moment, her importance as a salonnière, writer and political thinker — and of the importance of printed texts. Through them she could secure a place for her life-story in history. In writing her autobiography she might also have been inspired by Jean-Jacques Rousseau, and his writing of his autobiography Confessions (1782), a book she knew well and admired (Cabak Rédei, 2007)? As the semiotician J. Lotman points out in his chapter on the “‘I’ and ‘I’” and the advent of the (auto)biographical in art (Lotman, 1992, pp. 148-149 eng. tr.), in connection with Rousseau’s Confessions and Diderot’s Salons (1759–1781): “The artificial behaves like the living” (p. 149). Lotman underlines Rousseau’s and Diderot’s new ways of regarding the “I” and “their elevation of portraiture” as a non-allegoric way of painting man, in literature and in painting respectively. In folk psychological terms one might suggest that Confessions functioned as a cultural instrument to Germaine de Staël, and as such gives us an additional historical framework (although a particular historical perspective is beyond the scope of this essay), a context for understanding why and how she wrote her autobiography. And perhaps more importantly, artworks as Confessions give the reader and interpreter an understanding of which context Germaine de Staël’s self-narrative emerged, and with which sociocultural framework the text is in dialog, to put it within the frames of M.M. Bakhtin’s philosophy of language.

32In cultural semiotic terms, on the other hand, one may speak of this new way of viewing the autobiographical genre, established by Rousseau, as a new set of canonical works entering culture — a term used by the Prague school in the 1930s and 1940s to define a set of artistic rules governing the act of creating (the artist) as well as the one of receiving a piece of art, literary or other (see Cabak Rédei, 2007). Germaine de Staël was probably highly affected by this new take on the “I” in her own autobiographical writings, something that might be recognised not only in the way she ends the book, but also in the opening of Ten years’ exile where she makes excuses for writing about herself, if it was not for the higher purpose of telling about contemporary times. Thus, to put the “I” in focus was not unproblematic for her, in the sense that it was a break with the literary tradition of her times, at the same time as it was the start for a new literary genre. Thus, cultural reasons might yet be other explanations, alongside “pure” cognitive and philosophical ones, to the way Germaine de Staël used the pronoun “I” in her self-narratives.

33In ending her autobiographical account in the middle of a sentence, Mme de Staël leaves to the reader to end the story by other means than her own autobiography. The gap between the ending of the self-narrative in Ten years’ exile and the end of Germaine de Staël’s life has to be filled in by the reader posthumously. There is no final word. Thus, Germaine de Staël seemed to have refused even to make a summary of her life, at the time she wrote the autobiography. J. Bruner acknowledges this dilemma about autobiography:

There is something curious about autobiography. It is an account given by a narrator in the here and now about a protagonist bearing his name who existed in the there and then, the story terminating in the present when the protagonist fuses with the narrator (Bruner 1990, p. 121).

34Germaine de Staël’s way of ending her autobiography may be a good illustration of what J. Bruner pinpoints as a curiosity of autobiography—the fact that the protagonist is the same as the author. But, Germaine de Staël refuses to “fuse” with the “protagonist” captured in the text. The “autobiographical contract” as Lejeune (1981) describes this connection between protagonist and author — from the perspective of the reader — is based on two conventions: population data and publishing agreements. Thus an answer to how we may tell a fictive story from a non-fictive one seems to be found on the level of socio-cultural norms (in which the text emerged) — and not in the text itself, when one feature of narratives is that they can be “real” or “imaginary” without loss of their power as a story, as Bruner puts it (1990, p. 43).

35To sum up: Germaine de Staël’s autobiography illustrates the anticipatory aspect of remembering (to use a term from the psychologist F. Bartlett), stressed by K. Nelson (2003). And as J. Bruner interprets F. Bartlett’s “memory schemata”: “It is as if unity of affect (in contrast to ‘conflict’) is a condition for economical schematization of memory” (Bruner 1990, p. 58). The topic of congruence, and incongruence, is also discussed by Wertsch (1998). Wertsch has a theory about how narratives tend to become less coherent when hesitation occurs about which cultural tool (i.e. narratives in a general sense) to use in a specific situation. In this case one might illustrate this problem by suggesting that Germaine de Staël experienced an inner conflict when ending her self-narratives — a conflict based on other narratives known to her, and her own cognitive schemas about endings, fictive and non-fictive — which she did not resolve fully, at least not in terms of literary convention. Moreover, the reader is left to fill in the gaps, to provide the autobiography with an ending in order to be able to interpret it in terms of goodness of life, happiness or unhappiness in life (or in any other respect for that matter). And this task to provide this self-narrative with an ending, a final word, could not in this case be anything else than posthumous, since it was never finished during her life time.

36However, are autobiographies more “mimetic” in relation to “reality” (P. Ricœur, 1990) than, for instance, autobiographical novels? Or may both be said to be “re-constructions” at the same level of the self, set in different perspectives (by creating different “contracts” in Lejeune’s sense, with the reader)? Or might autobiographies be considered to be more embodied than fictive ones, for the simple reason that the protagonist overlaps with its author in terms of the memories accounted for.

37Let us have a look at Madame de Staëls novel Corinne.

6. Corinne: Germaine de Staël’s protagonist and alter ego

  • 4 “Mme de Staël avoue elle-même, que son but, en écrivant Delphine, fut de se peindre en profil, et p (...)

38Germaine de Staël chose the name Corinne for her protagonist and thereby she alluded to the tradition from Greek antiquity. Korinna (c. 518–442 BC) was the name of the famous poet who describes in a poem how she won over her male colleague and rival Pindar at a contest (Korinna, 1988, pp. 150-151). In that way Germaine de Staël placed her protagonist, and perhaps also herself, in the history of literary fiction. In fact, von Brinkman — a Swedish friend of Germaine de Staël’s — wrote in a letter: “Mme de Staël confesses that her goal in writing Delphine was to paint herself in profile, and with Corinne she wanted to mediate herself in total” (Brinkman, 1813/1987, p. 48. Translation, ACR, italics in the original).4 In that sense one might say that Germaine de Staël’s wish to seek redress in posterity, as it is manifested in Ten years’ exile, finds another form in Corinne, since the latter is written not as a pure autobiographic account but well as a hybrid between the former and a fictive one. Thus, Corinne might be said to be an autobiographical novel. The basis for that assumption will be further developed below.

39P.H. Vincent (2004, p. xviii) proposes that Corinne was a modern rewriting of Sappho’s life story. And that would, in fact, reinforce the claim made here that one of Germaine de Staël’s ambitions with Corinne was to acquire a place in history and hence in the public sphere, a sphere she always had belonged to since birth. So, what features in Corinne give good grounds for classifying it as autobiographical?

40M-C. Vallois points out that Germaine de Staël’s own voice replaces that of the narrator at the very end of the book, a few pages after the protagonist has faded away (Vallois, 1987, p. 179; de Staël, 1807/1985, p. 587; de Staël, 1998, p. 404). Also, the novel is provided with footnotes, which point to Germaine de Staël’s own experiences in her life outside the novel. Thus, one might say that Corinne has two endings, the latter reminiscent of an autobiography as it ends with “I”. And this latter ending might be the best illustration of to what extent Corinne was autobiographical, since the narrator switches to an “I” reflecting on the future. Germaine de Staël ends the fate of Corinne, the protagonist, as follows:

  • 5 “Le grand mystère de la mort, quel qu’il soit, doit donner du calme. Vous m’en répondez tombeaux si (...)

May be the mighty mystery of death, that mystery at least must give repose. Ye do not answer me, ye silent tombs! Merciful God, thou dost not answer me! I made my choice on earth, and now my heart has no asylum. Ye decide for me, and such a destiny is best. / Thus ended the last song of Corinne (de Staël, 1870, p. 389).5

And a few pages further on Germaine de Staël ends the novel:

  • 6 “Lord Nelvil donna l’exemple de la vie domestique la plus régulière et la plus pure. Mais se pardon (...)

Lord Nevil’s domestic life became most exemplary: but did he ever pardon his past conduct? Could the approving world console him? After the fate he had enjoyed, could he content himself with common life? I know not: nor will I; on that head, either absolve or condemn him (ibid.).6

41These two endings may illustrate the difficulty Germaine de Staël had when ending Corinne. She lets her alter ego, the protagonist, fade away, but the “I” ending the book, on the contrary, takes a grip on the future by leaving some evaluations of the male protagonist open. Be as it may with that, at least Prosper de Barante recognised himself in Lord Nelvil and felt that the author had “locked him” (“l’avait enfermé”) in the male protagonist (Diesbach, 1984, p. 391). Thus, the ending of the book might have been meant to illustrate a continuation in the life outside the book. Germaine de Staël (by switching to “I”) evaluates Corinne’s life via diffuse, but yet transparent, regrets of the male protagonist. In sum, Germaine de Staël gives her protagonist a posthumous ending by addressing the lover in a regretful way, and thereby initiating the reader too, i.e., to regret the lover’s deceitfulness. By doing this she differentiates herself from Corinne, her alter-ego. In that sense, the ending of the novel converges with findings made by Rozin and Stellar:

Judgments of the quality of complete lives must be determined by the self, near or at the end of life, or by third parties based on experiences of the life of another or narrative life descriptions. […] The ending of a life is typically considered to be at death, however some aspects of a life, particularly those having to do with the life’s meaning, can be affected by events that occur after a life is over (Rozin and Stellar, 2009, p. 273).

  • 7 “[…] Cette clôture décisive, qui fixe à jamais le désir en l’empêchant d’arriver à sa fin normale” (...)

42Now, what “is left” at these endings is a desire that prevents the narratives from coming to an end, to give them a “unity of life”. As A. Kotin Mortimer (1985) describes in connection with the ending of Corinne that, according to him, has a closure “[t]hat seizes the desire by hampering it to arrive at its normal end” (Mortimer, 1985, p. 113. Translation, ACR).7 Thus, the “posthumous effect” is avoided. However Mortimer’s description might also have bearings on Ten years’ exile, and thereby have further implications for a more general problem in respect of the relation between life and narrative. Since the ending of the book, which is outspokenly autobiographical, constitutes the ending of the narrative on one hand, but not of course of Mme de Staël’s life story on the other, she might have wished to avoid the “posthumous effect”.

43To sum up: Since Ten years’ exile was written for posterity, the importance of the ending comes into the fore. Put otherwise, one might say that Germaine de Staël lacked a schema, in Bartlett’s (1932) sense, for filling in the gap between what she could possibly know of and what she could not know anything of, namely posterity and future circumstances for her own death. Therefore one cannot perhaps exclude the possibility that she deliberately refrained from giving Ten years’ exile a proper ending, but on the contrary gave it a “non-ending” to avoid posthumous judgments. In fact, the ending of Corinne might also be said to illustrate Germaine de Staël’s dilemma, only in a different way. Ending the fictive self-narrative with an “I”, who is not the proper protagonist of the story, the posthumous effect is cancelled. It only applies to the protagonist, since it is the protagonist life story that ends. But again, the posthumous effect is partly cancelled here as well, since Germaine de Staël provides the reader with it when switching to the pronoun “I” and thus referring to herself at the same time as this “I” is merged with the protagonist Corinne, and thus the reader is left out from that process — the protagonist’s life continues with Germaine de Staël, the author outside the narrative. Thus, the self-narrative refuses to be subordinated the narrative rules, one might perhaps say, of rendering a “unity of life”. Otherwise put with the aid of P. Ricœur (1990), in the end Germaine de Staël’s personal identity refuses the narrative identity of Corinne, with which it is partly merged (in the sense that Corinne was the author’s alter ego, and the role became later on part of Germaine de Staël’s character in real life as ipse) until the closure of the autobiographical novel.

7. Conclusions

44The aim of the essay has been to shed some light on the theoretical affinities between semiotics, philosophy and psychology on the one hand, and on the other to apply these empirically when analysing Germaine de Staël’s self-narratives Ten years’ of exile and Corinne. This analysis has foremost served the purpose to illustrate how life and narratives may differ on the one hand, and on the other to show how they are entangled. Endings of self-narratives, as the focus point of such an inquiry, proved to be fruitful in this purpose. To study opening lines would, in contrast, not be as enlightening when trying to disentangle life and narratives. Germaine de Staël’s Ten years’ of exile and Corinne, I suggest, are manifest examples of resistance to closure, due to the difficulty of capturing real life in its continuous course within the constraints of a textual construction. I suggest, moreover, that Germaine de Staël’s literary endings also illustrate the difficulty to enhance life’s ending and its possible continuation in a life after the one on earth. The posthumous effect, the judgement of “goodness of life”, “happiness of life” or “unhappiness of life” is in the end left to others, to other people or/and to God, if one is a believer. Germaine de Staël seemed to have been very well aware of that, and perhaps she wanted to control the effect, after all, by refusing to give her self-narrative accounts proper endings? Or perhaps she was incapable of doing so, lacking the proper schema to fill in the gap between experience and an imagined posterity? Or, the crisis of closure in Corinne might, one is tempted to say, illustrate another way of dissolution than the one P. Ricœur gives an example of when referring to R. Musil’s The man without qualities, namely she gives an example of what happens when the narrative role merges with the personal identity.

45Insights from cognitive science and philosophy have been used to illuminate similarities between a cognitive approach to semiotics and the Narrative Practice Hypothesis (NPH) (Gallagher & Hutto, 2008).

46Now, both semiotics and psychology deal with the mental, but in different ways. In the first case, the mental is referred to indirectly in terms of content in the processes of meaning making, inherent in all interpretations of signs and sign systems (such as for instance narratives). This entails analyses not only of the mental (content) but also of the material (expression) plane of the sign, as done by for instance Bakhtin in his classifications of genres and dialogic words. Genres and the dialogic word (that is, how the word of the author enters into the text) are of outmost importance in order to understand the function and meaning of a literary text. In the second case, the mental is studied in terms of elucidating the relations between perception and cognitive structures in mental functions as for instance, in this case, memory (Bruner, 1990; Nelson, 2003). Also, P. Rozin and J. Stellar (2009) have shown that semiotic and philosophical inquiries may be tested experimentally in order to find statistical and important correlations between text and readers. Results that in their turn inspire to new problems for semiotic inquiries: in this case, for instance, how may religion influence our way of approaching “a unity of life” provided by narratives? More particularly: how did Germaine de Staël’s Calvinism influence her way of looking at narratives as a mediating function between personal identity and narrative identity from the perspective of “a unity of life?

47By this multidisciplinary approach to the analysis of the endings of Germaine de Staël’s self-narratives Ten years’ of exile and Corinne I hope to have shown how fruitful such take may be when analysing (literary) narratives as cognitive devices for interpreting life and actions.

Haut de page


Bakhtin, Mikhail M. (1963), Problemy poètiki Dostoevskogo, Moskva, Sovestkii pisatel’; eng. tr. Problems of Dostoevsky’s Poetics, Ann Arbor (Mich.), Ardis, 1973.

Bakhtin, Mikhail M. (1979), Estetika slovesnogo tvortchestva, Moskva, Iskusstvo, posthumous selection; engl. tr. Speech Genres and Other Late Essays, Austin, University of Texas Press, 1986.

Bartlett, Frederic (1932), Remembering. A study in experimental and social psychology, Cambridge, Cambridge University Press; last ed. 1995.

Brinkman, Carl Gustaf von (1813/1987), “Lettre sur l’auteur de Corinne”, Cahiers Staëliens, 39, pp. 139-181.

Bruner, Jerome (1990), Acts of Meaning, Cambridge (Mass.) & London, Harvard University Press.

Diesbach, Ghislian de (1984), Madame de Staël, Paris, Perrin.

Gallagher, Shaun & Hutto, Daniel D. (2008), “Primary interaction and narrative practice”, in Jordan Zlatev (ed.), The Shared Mind. Perspectives on Intersubjectivity, Amsterdam, John Benjamins.

Hutto, Daniel D. (2008), Folk Psychological Narratives. The Sociocultural Basis of Understanding Reasons, Cambridge (Mass.), MIT Press.

Korinna (1988), “On Her Self”, in Willis Barnstone (ed.), Sappho and the Greek Lyric Poets, New York, Schocken Books.

Lejeune, Philippe (1981), On Autobiography, Minneapolis, University of Minnesota Press, selection.

Lotman, Juri (1992), Kul’tura i vztyv, Moskva, Gnosis; eng. tr. Culture and Explosion, Berlin & New York, Mouton de Gruyter, 2004.

Mortimer, Armine Kotim (1985), La clôture narrative, Paris, José Corti.

Nelson, Katherine (2003), “Self and Social Functions: Individual autobiographical memory and collective narrative”, Memory, 11(2), pp. 125–136.

Paolucci, Claudio (2012), “Sens et cognition : La narrativité entre sémiotique et sciences cognitives”, Signata, 3, pp. 299–316.

Rédei, Anna C. (2007), An Inquiry into Cultural Semiotics. Germaine de Staël’s Autobiographical Travel Accounts, Lund, Lund University.

Ricœur, Paul (1990), Soi-même comme un autre, Paris, Seuil ; eng. tr. Oneself as Another, London & Chicago: University of Chicago Press, 1992.

Rozin, Paul & Stellar, Jennifer (2009), “Posthumous Events Affect Rated Quality and Happiness of Lives”, Judgment and Decision Making, 4 (4), pp. 273-279.

Schutz, Alfred (1964), Collected Papers II. Studies in Social Theory, posthumous (ed. Arvid Brodersen), The Hague, Martinus Nijhoff.

Sonesson, Göran (1995), Mute Narratives: New Issues in the Study of Pictorial Texts”, in Ulla-Britta Lagerroth et al. (ed.), Interart Poetics. Essays on the Interrelations of the Arts and Media (Acts of the congres “Interart Studies: New Perspectives”, Lund, May 1995), Amsterdam & Atlanta, Rodopi, pp. 243-252.

Staël, Germaine de (1807/1985), Corinne ou l’Italie, Paris, Gallimard.

Staël, Germaine de (1814), Réflexions sur le suicide, Paris, H. Nicolle.

Staël, Germaine de (1818/1996), Dix années d’exil, Paris, Fayard.

Staël, Germaine de (1821), Ten years’ exile, London, Treuttel and Wûrt.

Staël, Germaine de (1870), Corinne or Italy, New York, W.I. Pooley.

Staël, Germaine de (1998), Corinne, or Italy, Oxford & New York, Oxford University Press.

Vallois, Marie-Claire (1987), Fictions féminines. Mme de Staël et les voix de la Sibylle, Saratoga (Calif.), Anma Libri.

Vincent, Patrick H. (2004), The Romantic Poetess. European Culture, Politics, and Gender, 1820–1840, Durham (N.H.), University of New Hampshire Press.

Wertsch, James V. (1998), Mind as Action, New York & Oxford, Oxford University Press.

Wertsch, James V. (2002), Voices of Collective Remembering, Cambridge, Cambridge University Press.

White, Hayden (1987), The content and the form. Narrative discourse and the historical representation, London, John Hopkins University Press.

Haut de page


1 Interestingly enough the results showed that there was almost no difference in answers between the first person question and the second more general one (Rozin & Stellar, 2009, p. 279).

2 Since pictorial narratives are somewhat differently composed, see Sonesson (1995).

3 “J’ai toujours été fort sujette à l’ennui et, loin de savoir m’occuper dans ces moments tout à fait vides, qui semblent destinés à l’étude…” (de Staël, 1818/1996, p. 313).

4 “Mme de Staël avoue elle-même, que son but, en écrivant Delphine, fut de se peindre en profil, et par Corinne elle a voulu se communiquer toute entière” (Brinkman, 1813/1987, p. 48).

5 “Le grand mystère de la mort, quel qu’il soit, doit donner du calme. Vous m’en répondez tombeaux silencieux ; vous m’en répondez, divinité bienfaisante ! J’avais choisi sur la terre, et mon cœur n’a plus d’asile. Vous décidez pour moi : mon sort en vaudra mieux. // Ainsi finit le dernier chant de Corinne” (de Staël, 1807/1985, p. 584).

6 “Lord Nelvil donna l’exemple de la vie domestique la plus régulière et la plus pure. Mais se pardonna-t-il sa conduite passée ? Je l’ignore, et ne veux pas, à cet égard, ni le blâmer, ni l’absoudre” (de Staël, 1807/1985, p. 587). “Lord Nevil’s domestic life became most exemplary: but did he ever pardon his past conduct? Could the approving world console him? After the fate he had enjoyed, could he content himself with common life? I know not: nor will I; on that head, either absolve or condemn him” (de Staël, 1870, p. 391).

7 “[…] Cette clôture décisive, qui fixe à jamais le désir en l’empêchant d’arriver à sa fin normale” (Mortimer, 1985, p. 113).

Haut de page

Pour citer cet article

Référence papier

Anna Cabak Rédei, « Cognitive and Semiotic Aspects of Endings in Self‑Narrations: The Example of Germaine de Staël »Signata, 5 | 2014, 315-332.

Référence électronique

Anna Cabak Rédei, « Cognitive and Semiotic Aspects of Endings in Self‑Narrations: The Example of Germaine de Staël »Signata [En ligne], 5 | 2014, mis en ligne le 30 octobre 2016, consulté le 19 mai 2024. URL : ; DOI :

Haut de page


Anna Cabak Rédei

Anna Cabak Rédei is Research Fellow at the section of Cognitive Semiotics, at the Centre for Languages and Literature, at Lund University (Sweden). Her research field includes cognitive semiotics, (visual) perception psychology, social psychology, pictorial semiotics (especially film), cultural semiotics, narrativity and translation theory. Recent publications include: Film, music and experimental psychology: reflections and further directions. Music and the Moving Image, 2014. University of Illinois Press; Dialogue, embodiment and semioethics.Writing, Voice, Undertaking. New York, Ottowa, Toronto: Legas Publishing, 2013; Cognition and stereotypes in Guess Who is Coming to Dinner: a semiotic and social psychological perspective. Southern Semiotic Review, 2013. An Inquiry into Cultural Semiotics: Germaine de Staël’s Autobiographical Travel Accounts. Diss., Lund University, 2007.

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC BY 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search