Who evaluates whom and what in Jane Austen's novels?
Many commentators of Jane Austen's fiction have expressed the view that she is particularly difficult to catch at her narrative game. Speaking of E, (1) almost universally held to be the most complex and the most elusive of her novels, Virginia Woolf famously wrote that if you read it twelve times over, "at every fresh reading you feel anew that you never understood anything like the widening sum of its delights" (Southam 266). Lionel Trilling added that "the difficulty of Emma is never overcome. We never know where to have it. If we finish it at night and think we know what it is up to, we wake the next morning to believe it is up to something else; it has become a different book" (122). More recently and more generally, Irvin Ehrenpreis has expressed the bafflement of all those who try to establish what Austen (or, her narrator) is "up to" in her novels:
So the explicitness of the novelist is sometimes only apparent, and at other times is a game played with the audience. By sounding blunt and outspoken in many of her judgments, Austen entices unwary readers into assuming that she is straightforward.... But it remains true that when Austen does plainly set forth her judgment, it is as I have said--quite reliable. (118)
However, while certain commentators have put their fingers on Austen's invisibility (for a recent example, see Miller's 2003 Jane Austen, or The Secret of Style), others have seen her novels as mirroring a definite world view--a world view which has been interpreted in diametrically opposing ways. A traditional reading of Austen as an upholder of the patriarchal values of her society has been challenged by "revolutionary" and/or feminist readings of the novels as subtle critiques of those same values. S&S has been interpreted as a satire on (excessive) sensibility, but some readers have observed that Marianne/sensibility is shown to be much more fascinating than Elinor/sense (Nardin 10). MP--the litmus test of Austen studies in this department--has been read as an evangelical plea for old gentrified England and as a covert manifesto against the moral and social strictness of Austen's time. (2) How can these positions be reconciled with Reginald Farter's (1917) image of the author as a joycean divinity, indifferently paring her fingernails elsewhere?
... impersonality comes as the first ingredient in the specific for immortality. The self-revelation of the writer must be as severely implicit as it is universally pervasive; it must never be conscious or obtruded.... She is there all the time, indeed, but never in propria persona, except when she gaily smiles through the opener texture of "Northanger Abbey," or, with her consummate sense of art, mitigates for us the transition out of her paradises back into the grey light of ordinary life, by letting the word 'T' demurely peer forth at last, as the fantasmagoria in "Mansfield Park," "Emma" or "Northanger Abbey" begins to thin out to its final pages. (Southam 248)
It comes as no surprise, of course, that Austen's novels generate opposing interpretations: all great literature is supposed to do so, and conversely, the ability to instigate different readings has long been identified as a stigma of literary greatness. What is at once interesting and baffling is that these opposing readings appear to be equally justified, that there is ample textual material in Austen's novels to support them both. At the same time, there seems to be uncertainty as to whether Jane Austen watches over her novels as a Victorian commentator or as a modernist detached observer. Is there a "point" to her depiction of English gentry between the eighteenth and the nineteenth centuries, or is there not? And if that "point" is there, what is it exactly?
From our postmodern position in history, we might dismiss all the business of finding a "point" in Austen's fiction--in all fiction--as self-evidently irrelevant. Have we not been taught that novels are not pamphlets? And is not the point of any narrative the telling of a story? On the most basic level, it most certainly is: novels are not pamphlets--and narrators, in written fiction, are not authors. Yet any story, besides telling itself, expresses, is "symptomatic of," a "world-view," in Roger Fowler's terms. (3) And in Austen's case, this world-view seems to be, at one and the same time, contradictory and elusive. And while we might, once again, conclude that the "point" of (great) literature always eludes us, there may be a lesson to be learnt about how Austen's particular elusiveness is constructed--how readers are enticed into looking for a point which consistently evades their grasp.
Before embarking on a linguistic investigation of how Austen's elusiveness is created, it may be useful to remind ourselves of two cultural facts. The first one is contemporary: in our time, we have come to accept that the text does not contain its author--it contains a narrator, and can at most presuppose an "implied author" with whom all readers ideally wish to be acquainted. The second one is contemporary with Austen: in her time, it was customary to think of novels as ethical/ideological mirrors, and she would have expected at least certain categories of readers to deduce the author's opinions from her writings--to conflate narrator with author, implied author with "the real Jane Austen."
In his seminal study on oral narratives told by young black Americans, Labov wrote that a story, in its simplest form, consists of two temporally ordered clauses. Besides this minimal definition, however, he also provided a more detailed pattern, in order to account for the higher degree of complexity to be found in some of the stories he analyzed. The six parts or stages of this pattern can be and have been used to examine and dissect written as well as oral narratives (cf. Pratt, Black):
3. Complicating action
5. Result or resolution
Some of these parts or stages can be present or not, in written as well as in oral narratives. In written fiction, the "abstract" is usually provided by the title; the "orientation," if it is to be found at all, is most often found at the start (it is the "who, what, where, when," of the story); the "complicating action" unsettles the initial balance and prepares the "resolution;" the "coda," usually placed at the end of the narrative, is where things are rounded off--where the "(implied) author," or the "narrator," parts company with the "reader." "Evaluation," though it comes up fourth, is the most difficult "part" or "stage" to place, because though it tends to cluster in certain areas of a text (traditionally, at the beginning and end; but there is a lot of variation along the genre and period axes), it can be found anywhere, and evaluative elements are hard to identify with any certainty. "Evaluation," as Labov himself defined it, is the "point" of a story: it can be a moral, a religious, or a didactic point; more generally, it is what demonstrates that the story is worth telling?
In a novel, as well as in any other kind of story, evaluation is endemic, and no two readers will exactly agree as to which stretches of text are evaluative and which are not--though certain stretches are quite unequivocally evaluative. Evaluative elements can be found, to begin with, in dialogue as well as in narrative, in the characters' as well as in the narrator's discourse. Some of Jane Austen's novels (P&P, E, the unfinished TW) are mostly made up of dialogue, and when this is the case much of the evaluative work is as it were "embedded" in direct (or indirect) speech. In P&P, we learn something about Mr. and Mrs. Bennet before the narrator tells us "who and what they are:"
"My dear Mr. Bennet," said his lady to him one day, "have you heard that Netherfield is let at last?"
Mr. Bennet replied that he had not.
"But it is," returned she; "for Mrs. Long has just been here, and she told me all about it."
Mr. Bennet made no answer.
"Do not you want to know who has taken it?" cried his wife impatiently.
"You want to tell me, and I have no objection to hearing it." (1)
The narrator, however, is always instinctively held to be a more reliable evaluator than any single character. As Michael Toolan puts it, "narrators are typically trusted by their addressees ... To narrate is to bid for a kind of power" (3). Even first-person "homodiegetic" narrators, who take part in the story and can therefore be suspected of having a personal interest in directing audience reactions, are considered more "authoritative" than any other character, and can be confused with the "author" to a lesser or greater degree. Modernist writers such as Conrad and James, and latter-day followers like Ishiguro, have deliberately played with readers' expectations by exploiting this "authoritativeness." Third-person "heterodiegetic" narrators acquire an extra degree of authoritativeness by being impersonal (if that is the case) and situating themselves out of the action: within the space of their fictional world, they are like gods, and readers will tend to treat them as such--i.e., they will tend to believe all they say.
All of Austen's narrators are third-person heterodiegetic narrators, and as such command the reader's blind faith (or his/her gullibility, if we believe in the author mocking her audience). These narrators never take part in the action, and only rarely come out of hiding to speak in the first person. Therefore, their evaluative comments tend to have a ponderous weight in our interpretation of the novels--of what is going on, who are the good guys and the villains, what is likely to happen, etc. It is my point, though, that these narratorial figures variously undermine their own authoritativeness and leave readers more or less stranded between the waves of conflicting interpretations.
However, before looking at how Austen's narrators evaluate their fictional worlds (and at how they undermine their own evaluative work), some preliminary definitions of "evaluation" are needed in order to define the range of textual data we are looking for. Evaluation is very difficult to individuate, because it is not necessarily linked to any particular linguistic items, and it is not consistently signalled by any linguistic or metalinguistic means (at least in literary texts: other textual types, e.g. manuals or academic articles, can display specifically signalled evaluative techniques). Linguistic studies on evaluation have worked with different definitions of a very elusive quality, and have attributed that quality to words, sentences/utterances, text/discourse, speakers/writers, etc.: some terms of art are "connotation," "affective meaning," "attitude." Scholars belonging to the field of stylistics have preferred to speak of the evaluative, attitudinal force of language as "modality" (Fowler 131-32; Simpson 46-55), but they too have had to admit that modal elements are only the tip of the iceberg of attitude. More recently, a very promising new field of research on "evaluation" as such has opened: the linguists working in this field are trying to unify the terminology and to build a general evaluative theory--and though the concept of "evaluation" still seems ultimately irreducible to any satisfying unity, some interesting results have been obtained in the analysis of literary texts (Cortazzi and Jin), argumentative prose (Hoey), written and oral academic discourse (Anderson and Bamford).
Textual (or discoursive) evaluation is not simply a question of assessing what is good and what is bad, what is important and what is not (though both axes are relevant). Geoff Thompson and Susan Hunston, editors of a volume on Evaluation in Text, have identified three main functions of evaluation, which are respectively expressive, interpersonal, and textual: (5)
(1) to express the speaker' s or writer's opinion, and in doing so to reflect the value system of that person and their community;
(2) to construct and maintain relations between the speaker or writer and hearer or reader;
(3) to organize the discourse. (6)
If one thinks of evaluation in a novel, and conveniently substitutes "narrator" for "writer," one immediately sees the cogency of this tripartite definition. The narrator, overtly or covertly, offers his/her point of view on the fictional world he presents, and in so doing reflects (directly or indirectly) the value system of the community which has spawned him/her, or of a part of that community. (6) By telling his/her novelistic story, the narrator maintains certain kinds of relations with his/ her readers, and/or, within the space of the text, with the shadowy figure of the narratee (if given). Finally, even dispositio is a form of evaluation, and a certain kind of ideological slant (using "ideological" in the widest possible sense) brings about certain forms of textual organization. To give one very straightforward example, the modernists' sense that the world could no longer be described from an external point of view, within the four walls of unified personality, sequential chronology, narrative reliability and providential finality led to the freer structures employed by Virginia Woolf, James Joyce, Ford Madox Ford.
Though they insist that evaluation is essentially a unified phenomenon, Thompson and Hunston identify four main parameters: "good-bad, certainty, expectedness, and importance" (25). While it is important to insist on the interpenetration of these axes (judging the importance of an event can also imply gauging its expectedness and goodness), and though "expectedness" and "certainty" can certainly be conflated, there is no doubt that the three remaining parameters sum up the evaluative work performed by the authority (or authorities) in charge of a text. (7) In a novel, events are arranged, facts and characters are judged along those three axes, as shown in the incipit of Dickens' David Copperfield (expectedness/certainty) and at the start of the second chapter of William McIlvanney's Docherty (good-bad, importance):
Whether I shall turn out to be the hero of my life, or whether that station will be held by anybody else, these pages must show. (13) High Street was the capital of Conn's childhood and boyhood. The rest of Graithnock wets just the provinces. High Street, both as a terrain and as a population, was special. Everyone whom circumstances had herded into its hundred-and-so yards had failed in the same way. It was a penal colony for those who had committed poverty, a vice which was usually hereditary. (24)
As shown by these textual stretches, however, though it is sometimes very evident that one is faced with an evaluative passage, it is by no means easy to determine the linguistic means by which evaluation is effected. In a novel, the most evident case, s of evaluation are those in which the narrator commits him/herself to a categorical assertion along the good-bad, important-unimportant axes, or reflects on the probability of an event taking place (all of Austen's narrators do all of these things). On other, subtler, occasions, an event or a character may be compared to another, a single modal expression used to determine probability or desirability ("must," "may," "certainly," "luckily"). Even more trickily, evaluation may be hidden in lexical choice, collocation, or contextual elements. The beginning of NA can be used as an illustration; it is only when one identifies the intertextual link with the gothic genre that the evaluative contours of the narrator's discourse become visible:
No one who had ever seen Catherine Morland in her infancy, would have supposed her born to be an heroine. Her situation in life, the character of her father and mother, her own person and disposition, were all equally against her. Her father was a clergyman, without being neglected, or poor and a very respectable man, though his name was Richard--and he had never been handsome.... Her mother was a woman of useful plain sense, with a good temper, and, what is more remarkable, with a good constitution. She had three sons before Catherine was born; and instead of dying in bringing the latter to the world, as any body might expect, she still lived on--(NA 5)
Attempts at identifying an exhaustive list of linguistic evaluative tools are of course doomed to fail or to produce Lewis Carroll's 1:1 map of the world. (8) It is true, however, that evaluation tends to be effected by certain linguistic means, and that certain linguistic items almost invariably carry evaluative force. Stylisticians have classified narrators and narrative techniques according to the dominating presence of one or more "modality" systems: "deontic" ("You may/must/should leave," "it is necessary that you leave"), "boulomaic" ("I hope that you will leave," "I wish you'd leave," "hopefully you'll leave"), "epistemic" ("You may/must be right," "You're certainly/possibly right"), and "perceptive" ("it was evident that he was tired") (Simpson 46-55). Even so, they have freely admitted that there is no fixed connection between modality and language, and that certain forms of evaluation seem to be too pervasive to be identified with any certainty.
Evaluation scholars, in this as in other matters, adopt an all-embracing approach. Geoff Thompson and Susan Hunston start out from the assumption that evaluation can appear at the level of lexis, grammar, and text; but after trying to isolate specific lexical, grammatical and textual evaluative elements, they have to admit that the task is ultimately impossible or useless. Consequently, they decide to work with more general "conceptual" entities. Evaluation, they say, can be comparative, subjective, value-laden. "Of these three groups, the third seems inherently more lexical in nature; but the first and the second are primarily grammatical" (22):
(1) Evaluation involves comparison of the object of evaluation against a yardstick of some kind: the comparators. These include: comparative adjectives and adverbs; adverbs of degree; comparator adverbs such as just, only, at least; expressions of negativity (morphological, such as un- and other affixes; grammatical, such as not, never, hardly; and lexical, such as fail, lack).
(2) Evaluation is subjective: the markers of subjectivity. This is a very large group including: modals and other markers of (un)certainty; non-identifying adjectives; certain adverbs, nouns, and verbs; sentence adverbs and conjunctions; report and attribution structures; marked clause structures, including patterns beginning with it and there, and "Special Operations Clauses" ... such as pseudo-clefts.
(3) Evaluation is value-laden: the markers of value. These may be divided into two groups: lexical items whose typical use is in an evaluative environment (the circularity of this definition seems unavoidable); and indications of the existence of goals and their (non-)achievement ("what is good" may be glossed as "what achieves our goals" and "what is bad" may be glossed as "what impedes the achievement of our goals"). (21)
In the end, the catalogue is so vast that the critic is left to his own resources, and must brave the pitfalls of "epistemic circularity" on his own (Leech and Short 13). However, some of these tools, analyses and definitions may be useful in defining how evaluation works in Austen's novels.
We are now in a position to redefine the difficulties many readers have experienced who have tried to put their fingers on "what Jane Austen is saying," or "what narrative game Jane Austen is playing." It is my contention that these difficulties can be summed up in one general problem--the problem of tracing evaluative patterns in Austen's novels. This problem is not merely a consequence of the pervasive and elusive qualities of evaluation in general, Or of the fact that novels are not pamphlets and have, therefore, no clear "point" to make: such novels as Dickens' Hard Times or D.H. Lawrence's Lady Chatterley's Lover do display a clear evaluative pattern--whatever different directions their fictional structures may take in spite of it. As seen above, Austen's novels also seem to invite evaluative scrutiny, while many novels of similar complexity do not--no serious critic would dream of finding the "point" of Woolf's Mrs Dalloway, for instance, unless the point is a universal one about the nature of mankind. (9) But the reader's search for Austen's "point" is frustrated as consistently as it is instigated, and a web of "evaluative opacity" is created which does not coincide with evaluative absence.
Austen creates this web of opacity not by withdrawing evaluation, but by dismantling the authority of the evaluative sources she sets up. In fiction, the evaluative sourcepar excellence is of course the narrator: even when the physically present spinner of yarns becomes a disembodied voice hovering over a neutrally-told story, a narrative function still remains to give substance to facts and words. (10) In Austen's novels, the narrator as an evaluative centre can still be identified, sometimes even personally, but his/her evaluations cannot be relied upon to provide a centripetal interpretation of events.
Some literary critics, though stopping short of a linguistic analysis of Jane Austen's evaluative opacity, have grappled with the problem of narrative unreliability in her fiction. Many of these critics have brought post-structuralist exegetic concepts to bear against some of her novels, particularly E (see Holly) Richard F. Patteson has written about "the multiplicity of narrative voice" which "makes the reader's search for determinacy even more difficult than the characters'" (465). Tara Goshal Wallace has observed the moves by which Austen's narrators renounce omniscience, or partially disappear from their narratives, from LS to P. In a totally different vein, Bernard J. Paris has suggested that while narrative structures may present "an abstract moral perspective," "Realistic characterization fights against theme as well as against form" (20): in MP, for instance, Fanny's "real" character "is not compatible with her aesthetic and thematic roles," i.e., with Austen's project of making her the heroine of a conservative evangelical novel (22).
Post-structuralist and psycho-analytic views highlight two different but related sources of evaluative unreliability in Austen's fiction: on the one hand, the narrator tends to undermine his/her own authority by contradicting him/herself, claiming or showing ignorance of facts or thoughts, etc.; on the other hand, the novel is an inherently "heteroglossal" genre, where the narrator's is only one of the voices at play, if a highly privileged one (Bakhtin, Problems). In Austen's novels, the narrators tend to undermine their own authority and/or leave evaluative room to other characters--whose views, in their turn, are often refuted by the unfolding of events. In these conditions, it becomes hard, and ultimately impossible, to establish "who evaluates what, in and through narrative, and why" (Cortazzi and Jin 104).
The problem with Austen's narrators is that they seem to change during the course of each novel, or, to look at it from another angle, that they react differently to the various characters and events it falls to their lot to introduce and describe. They are, all of them, third-person heterodiegetic narrators, yet their level of detachment varies greatly, and their vision disturbingly hovers between omniscience and ignorance. Furthermore, they sometimes employ (and mingle with) one or more reflectors through which the action is shown. In the terms of Paul Simpson's exhaustive classification, Austen's novels always display "category B" narrators, but these oscillate between "narratorial" and "reflector" modes. These narrators employ both "negative" and "positive" strategies: at times they speak from outside the consciousness of their characters, whereas on other occasions they claim knowledge of thoughts, feelings, and past actions (Simpson 55-75). Given these premises, it is not surprising that many Austen readers feel the interpretive ground slipping from under their feet: they look to the evaluative centre of the novel to know where they stand, but that centre is continually shifting or disappearing from view.
Sometimes the centre does hold. There are moments when Austen's "category B" narrators work in the "narratorial positive" mode (i.e., they profess omniscience, or rather, they move freely between past and future events, characters' thoughts and feelings): these moments are usually situated at crucial spots in the narrative, so that readers are encouraged to think that they will set the norm for the rest. When working in the "narratorial positive" mode, Austen's narrators offer "strong evaluations" of people and actions, mostly on the "good-bad" and "importance" axis. To fall back on Labov's 1972 pattern, these "strong evaluations" mostly occur during the initial "Orientation" and the final "Result" and "Coda," as well as when new characters have to be introduced. In the "Orientation," Austen's narrators usually provide social and financial information on the main dramatispersonae and judge their moral and social character. The first Chapter of S&S provides most of the practical and psychological information we need on the Dashwood family:
The family of Dashwood had been long settled in Sussex. Their estate was large, and their residence was at Norland Park, in the centre of their property ... The late owner of this estate was a single man ... he invited and received into his house the family of his nephew Mr. Henry Dashwood, the legal inheritor of the Norland estate, and the person to whom he intended to bequeath it.... By a former marriage, Mr. Henry Dashwood had one son; by his present lady, three daughters. The son, a steady respectable young man, was amply provided for by the fortune of his mother ... To him therefore the succession to the Norland estate was not so really important as to his sisters; for their fortune ... could be but small.... The old gentleman died; ... to [Henry Dashwood's] son, and his son's son, [Norland Park] was secured, in such a way, as to leave him no power of providing for those who were most dear to him ... He [Henry Dashwood's uncle] meant not to be unkind, however, and, as a mark of his affection for the three girls, he left them a thousand pounds a-piece.... [Mr. John Dashwood, Henry Dashwood's son] was not an ill-disposed young man, unless to be rather cold hearted, and rather selfish, is to be ill-disposed; ... But Mrs. John Dashwood was a strong caricature of himself;--more narrow-minded and selfish.... Elinor ... possessed a strength of understanding, and coolness of Judgment ... Marianne's abilities were, in many respects, quite equal to Elinor's. She was sensible and clever; but eager in every thing ... she was every thing but prudent. The resemblance between her and her mother was strikingly great.... Margaret, the other sister, was a good-humoured well-disposed girl; but as she had already imbibed a good deal of Marianne's romance, without having much of her sense, she did not, at thirteen, bid fair to equal his sisters at a more advanced period of life. (1-5)
The information Austen's narrators provide, almost invariably, when a new actor appears on the stage, is complementary to this initial orientation. Even in E, where the narrator is mostly noted for his/her absence and reticence, no new character is launched without a few introductory words:
Lady Lucas was a very good kind of woman, not too clever to be a valuable neighbour to Mrs. Bennet. (P&P 12) Mr. Perry was an intelligent, gentlemanlike man ... (E 16) Mr. and Mrs. Musgrove were a very good sort of people; friendly and hospitable, not much educated, and not at all elegant. (P 38)
However, this "positive" evaluative handling soon disappears, and readers are left to their own resources. For the main part of each novel, after the initial orientation, the narrator employs many strategies of invisibility and reticence: he/she does not vanish completely, or fall into absolute silence, yet the moments when he/she demonstrably evaluates the narrative are few and far between (though of course, even organization, dispositio, is a form of "textual" evaluation).
The narrative voice comes back to the surface only at the end, in the "result" and the "coda," when it takes charge to condense certain parts of the story, or, particularly in the coda, to judge past events and anticipate future developments. (11) Generally speaking, the "result" of all of Austen's novels is marriage (between the heroine and the most desirable man, between another woman and the second-best man, etc.). When a marriage proposal takes place, the narrator prefers to make a summary of the facts rather than merely repeat the characters' words. In Leech and Short's classification of speech and thought presentation modes, what readers are offered is a prolonged and reticent Narrative Report of Speech Act(s) (318-36), sometimes supplemented by half-moral comments, often uttered with half a tongue in the narrator's cheek: (12)
How soon he had walked himself into the proper resolution, however, how soon an opportunity of exercising it occurred, in what manner he expressed himself, and how he was received, need not be particularly told. (S&S 317) What did she say?--Just what she ought, of course. A lady always does. (E 391)
In the summing-up coda, Austen's narrators oscillate between the light irony and benevolence of E (where Mrs. Elton's ventriloquized criticism of Emma and Mr Knightley's wedding is counterbalanced by their friends' concluding faith in "the perfect happiness of the union;") (E 440), and the harsh retributive morality of MP, where villains are both punished and reproached. In this case, the narrator employs a variety of value-laden expressions ("the indignities of stupidity," "the disappointments of selfish passion," "punishment," "conduct," "guilt," "mortified," "reproach") which leave the reader in no doubt as to the deontic character of that final "must":
Mr. Rushworth had no difficulty in procuring a divorce; and so ended a marriage contracted under such circumstances as to make any better end, the effect of good luck, not to be reckoned on. She had despised him, and loved another--and he had been very much aware that it was so. The indignities of stupidity, and the disappointments of selfish passion, can excite little pity. His punishment followed his conduct, as did a deeper punishment, the deeper guilt of his wife. He was released from the engagement to be mortified and unhappy, till some other pretty girl could attract him into matrimony again, and he might set forward on a second, and it is to be hoped, more prosperous trial of the state--if duped, to be duped at least with good humour and good luck; while she must withdraw with infinitely stronger feelings to a retirement and reproach, which could allow no second spring of hope or character. (MP 364-65)
These narrative apparitions can have a twofold effect on readers: on the one hand, they create in their minds a conflation of the narrator with the "author" (because if an opinion is forcibly expressed by a third-person heterodiegetic narrator, it must be the author's); on the other, they convince them that the author or the narrator will be or has been in charge throughout--which is very far from being the case. Firstly, the greater part of Austen's novels is made up of dialogue (free and bound direct and indirect speech, narrative reports of speech acts), often unmediated by the narrator (above all, but by no means only, in P&P and E). Secondly, even when the narrator is present, his/her opinions cannot be readily identified--because they are not expressed explicitly enough, because they are contradictory, or because it is not clear who is speaking/thinking.
Let us first examine the narrator's evaluative reticence, which is apparently at odds with his/her openness in the openings and closings. Sometimes, this reticence is a function of mystery, or of what Leech calls the "interest principle:" (13) the narrator does not want to uncover his/her plans, as he/she would if he/she offered explicit evaluation of a character or an event. When Willoughby first appears in S&S, for instance, the narrator does not judge him on the ethical plane (on the "good-bad" axis), nor does he/she ply into his real feelings; so, even when readers start to suspect him of double dealing, they can still hope he will act honourably by Marianne. In the Chawton novels, narrative reticence becomes the rule--if orientations, results and codas be excepted; in E, it takes the peculiar form of listing both good and bad qualities of(almost) all characters, from the eponymous heroine downwards:
Emma Woodhouse, handsome, clever, and rich, with a comfortable home and happy disposition, seemed to unite some of the best blessings of existence; ... The real evils indeed of Emma's situation were the power of having rather too much her own way, and a disposition to think a little too well of herself; (3-4)
This quotation also illustrates the second source of narrative unreliability, or of evaluative confusion in the narrator's discourse. In the space of a few paragraphs, the narrator shifts from a "negative" to a "positive" mode: in the first sentence, he/ she adopts an external point of view which forces him/her to make conjectures about the real state of affairs (Emma "seemed to unite some of the best blessings of existence"); whereas in the second, he/she falls back on a positive perspective which allows him/her to establish the "real evils indeed of Emma's situation."
This kind of oscillation produces epistemological uncertainty, because readers cannot be sure whether the narrator knows or does not know about people' s morals and feelings, about past and future events. Owing to narratorial ignorance or reticence, Jane Austen's novels are permeated with evaluative opacity, on the good-bad, certainty, and importance axes. We have already seen how on certain occasions, her narrators refuse to provide open evaluations of certain characters (Willoughby, but also, initially, the Crawfords in MP). The same is true of events: sometimes an omniscient narrator informs his/her readers about facts which have taken place on a different temporal plane (see for instance Mr Weston's story, told in Chapter 2 of E); on different occasions, though, an ignorant or reticent narrator omits crucial information--or does not signal the importance of certain details. In E, we know that Frank Churchill and Jane Fairfax are engaged only towards the end of the novel, even though various hints are dropped before that stage; (14) in P&P, we are never told who betrays Elizabeth Bennet and Darcy's secret to Lady Catherine (see Sutherland). Not infrequently, the narrator openly pleads his/her ignorance, from the early LS to the mature P:
Whether Lady Susan was, or was not happy in her second Choice--I do not see how it can ever be ascertained--for who would take her assurance of it, on either side of the question? The World must judge from probability. (LS 249) Mrs. Clay' s affections had overpowered her interest, and she had sacrificed, for the young man' s sake, the possibility of scheming longer for Sir Walter. She has abilities, however, as well as affections; and it is now a doubtful point whether his cunning, or hers, may finally carry the day; (P 201)
In the first case, the narrator/editor (Lady Susan is a short epistolary novel with a very short narrative coda) professes him/herself unable to guess a character's thoughts and feelings; in the second, he/she pleads ignorance of the future. In both instances, the narrator's powers are limited much as a character's would, by being situated in time and in an individual psyche.
The very individual quality of Austen's narrators clashes with the omniscience they also claim at certain stages: if at times they seem to be looking at the action from above, on other occasions they descend upon the earth and betray their position--they say "I," they become characters. In Jane Austen, or The Secret of Style, D.A. Miller rightly observes that Austen's novels lack a strong evaluative centre ("Austen's divinity is free of all accents that might identify it with a socially accredited broker of power/knowledge in the world under narration;") (32), but simplistically attributes this centrifugal quality to the absence of a perceptible narrator ("Austen's work most fundamentally consists in dematerializing the voice that speaks it;") (6-7). While Miller's identification of a void, a "cut" at the heart of the novels is instructive, his conflation of this "cut" with the narrator's disappearance clashes with all those instances in which the narrator makes a nameless, but not impersonal, appearance (34). (15)
If, on the other hand, the narrator is seen as a character among many (though one with a special functional status), his/her personal interventions need no longer be seen as intrusions, and his/her evaluative and epistemological uncertainties become a sign of human, no longer godlike, authority. In his seminal study of E in The Rhetoric of Fiction, Wayne C. Booth identified the narrator with "Jane Austen," in inverted commas--both a character and a dramatized projection of the "implied author." Booth observed that this dramatized "Jane Austen" is sometimes unreliable ("Is the mystery purchased at the price of shaking the reader's faith in Jane Austen's integrity?") (254), and that her presence is a key structural element of the novel ("The dramatic illusion of her presence as a character is as important as any other element in the story;") (266); but he failed to link unreliability and presence, personification and fallibility.
Whereas it is probably excessive to say that "an omniscient narrator destroys his authority the moment he says I" (Black 14), it is true that absence and omniscience often go together (a prejudice having to do with our received ideas on God)--and a humorous narrator speaking in the first person, as well as alternatively knowing and guessing, certainly does lose a great part of his/her reliability. (16)
I come now to the relation of a misfortune, which about this time befell Mrs. John Dashwood. (S&S 216) I wish I could say, for the sake of [Mrs Bennet's] family, that the accomplishment of her earnest desire in the establishment of so many of her children, produced so happy an effect as to make her a sensible, amiable, well-informed woman for the rest of her life; (P&P 295) ... although there doubtless are such unconquerable young ladies of eighteen (or one should not read about them) as are never to be persuaded into love against their judgment by all that talent, manner, attention, and flattery can do, I have no inclination to think Fanny one of them ... (MP 181) Mrs Goddard was the mistress of a school--not of a seminary, or an establishment, or any thing which professed, in long sentences of refined nonsense, to combine liberal acquirements with elegant morality upon new principles and new systems--and where young ladies for enormous pay might be screwed out of health and into vanity--but a real, honest, old-fashioned Boarding-school ... (E 18)
In all these cases, the narrator-as-a-character (a figure conflated by many with Austen herself) comes out of impersonal hiding, though in different ways and with different degrees of omniscience (or, renunciation of omniscience). In the quotations from S&S and P&P, the narrator, though speaking in the first person, is in full charge of the narrative: he/she knows what has happened and will happen, and is free to move in the consciousness of his/her characters. In MP, the narrator is guessing at possibilities and measuring a (limited) knowledge of Fanny's inclinations against her background and situation (though one implicature of the text between parentheses is that the narrator does know what would happen if Henry Crawford persisted, because Fanny is not one of those romantic characters peopling the novels the narrator is having a swipe at). Finally, in E, the narrator comes out not by saying "I," but by expressing in a very direct manner his/her personal opinions on contemporary affairs (in this case, the confusion with the historical Jane Austen is almost inevitable).
Another fluctuating movement exhibited by Jane Austen's narrators--one that further complicates the network of evaluation--is the continual shift of perspective from the narrator to one or more characters--with the heroine usually taking up the role of reflector. From the mock-gothic part of NA onwards, all of Austen's novels pivot on one central character whose consciousness the narrator can penetrate, but who also mixes with the narrator him/herself, at times in an inextricable manner. The mixture is stylistic as well as narratological: as literary critics have noted and computational linguists have statistically demonstrated, there is a greater linguistic similarity between heroines and narrators than between any two characters in each novel--what Bakhtin calls "stylisation" (Bakhtin, The Dialogic 315). (17)
A distinction is needed, however, between the cases in which a central character functions as a reflector, and the many instances in which Austen's narrators and reflectors become virtually indistinguishable. When the narrator consistently lets the narrative focus remain within a character's consciousness (it happens throughout E), readers may tend to see things from that character's visual, psychological, and ideological perspective. But when this central consciousness and the narrator are blended, and readers no longer know who is speaking, evaluative confusion reaches a peak. In order to discriminate between these two techniques (both of them alternately used in all novels, though with different proportions), I propose a distinction between "mimetic reflector" and "non-mimetic reflector."
In all of Austen's novels, a central reflecting consciousness is employed. After the initial orientation conducted by the narrator, when all characters have been introduced and events set in motion, readers are plunged into the consciousness of the heroine. This plunge takes place later in the early works, while in the Chawton novels the reflector is employed more consistently, and in E we are allowed to abandon Emma's gaze and feelings only once (Volume I, Chapter V). The narrator penetrates the reflector's consciousness by the use of "mental process clauses" (Halliday, An Introduction to Functional Grammar 107). (18) Readers are told what the reflector is feeling, thinking, or perceiving:
Every object in the next day's journey was new and interesting to Elizabeth; and her spirits were in a state of enjoyment; (P&P 119) Fanny could not wonder that Edmund was at the parsonage every morning; (MP 52) On the morning appointed for Admiral and Mrs. Croft's seeing Kellynch-hall, Anne found it most natural to take her almost daily walk to Lady Russel's, and keep out of the way till all was over; (p 31), (19)
From this first plunge onwards, and with the exception of all those cases in which the narrator comes into the open, the heroine becomes the temporal, spatial, and psychological pivot of the narrative. When a scene is described, this pivot functions as a deictic centre; and throughout the reflector narrative, occasional "sensing" reminders are inserted to signal that the angle of vision has not been shifted:
Fanny supposed she must have been mistaken, and meant to think differently in future; but with all that submission to Edmund could do, and all the help of the coinciding looks and hints which she occasionally noticed in some of the others, and which seemed to say that Julia was Mr. Crawford's choice, she knew not always what to think. She was privy, one evening, to the hopes of her aunt Norris on this subject, as well as to her feelings, and the feelings of Mrs. Rushworth, on a point of some similarity, and could not help wondering as she listened; and glad would she have been not to be obliged to listen ... 'I think, ma'am,' said Mrs. Norris--her eyes directed towards Mr. Rushworth and Maria, who were partners for the second time--'we shall see some happy faces again now.' ... Miss Bertram did indeed look happy, her eyes were sparkling with pleasure, and she was speaking with great animation, for Julia and her partner, Mr. Crawford, were close to her; they were all in a cluster together. How she had looked before, Fanny could not recollect, for she had been dancing with Edmund herself, and had not thought about her. (MP 92-3; emphasis added)
The narrator first selects Fanny as "senser," then describes a scene and reports a dialogue as seen through her eyes and heard by her ears. At the end of the dialogue, readers are used to Fanny functioning as a deictic/psychological centre, and will tend to interpret that "indeed" as belonging to her (as if she had turned her gaze towards Julia Bertram and Henry Crawford to verify the probability of her aunt's propositions). (20) If any doubts should arise, however, Fanny is again selected as senser before the end of the paragraph ("How she had looked before, Fanny could not recollect"): readers are reminded that they are looking at the fictional world through her eyes, and that they are not allowed to see what she does not.
In an article comparing modernist narrative techniques with Austen's, I have already observed how narrators employ the reflector technique in order to sift knowledge before presenting it to their readers (Morini 77, 83-93). In E, in particular, many mysteries are unveiled for the reader when they are unveiled for Emma. Even in the above passage, we may suspect the reflecting narrator of a (psychological) omission: we are told that Fanny had not thought about Julia before because she was dancing with Edmund--but the reason is given in passing, and it is left to the reader to guess (or learn later) that since she is in love with Edmund, very little else is likely to engage her attention when she is with him.
Things become even more complicated, and evaluation becomes even more elusive, when a mimetic reflector is substituted for its non-mimetic counterpart. When the narrator and the reflector are completely conflated, it becomes impossible to attribute evaluative comments with any certainty to one or the other. Even in the above passage from MP, which is clearly non-mimetic, certain words and clauses ("indeed," "she was speaking with great animation") could be grammatically assigned to the narrator as well as the reflector (though logic leads elsewhere). In other instances, neither grammar nor logic offer any guidance. The following passage from P&P describes a social occasion which the narrator shows through Elizabeth Bennet's eyes (The chapter begins: "Convinced as Elizabeth now was ... she could not help feeling ..."):
By Mrs. Hurst and Miss Bingley, they were noticed only by a curtsey; and on their being seated, a pause, awkward as such pauses must always be, succeeded for a few moments. It was first broken by Mrs. Annesley, a genteel, agreeable-looking woman, whose endeavour to introduce some kind of discourse, proved her to be more truly well bred than either of the others; (21) ... In Darcy's presence [Elizabeth] dared not mention Wickham's name; ... but, exerting herself vigorously to repel the ill-natured attack, she presently answered the question in a tolerably disengaged tone. (205; emphasis added)
As many have noted, Jane Austen is an eclectic master of thought, as well as speech, presentation. She employs direct and indirect Halliday, An Introduction to Functional Grammar hought, free direct and indirect thought, narrative reports of thought acts (Leech and Short 337-51). However, the nature of speech and thought presentation is such that the boundaries between different techniques are not always clear-cut. Free indirect thought is particularly tricky: in the absence of clear signals (and in a novel whose narrator seems to be omniscient and not omniscient, and appears to be working both in the "positive" and in the "negative mode"), how are readers to tell if the thought is attributable to the narrator or the reflector? In the above passages, are the italicized evaluative comments to be allotted to Elizabeth, the narrator, or both? (22)
In other words, to go back to Simpson's taxonomy of narrators, in these cases we no longer know if the narrator is working in the narratorial or in the reflector mode. One could say that it is not so crucial to distinguish between the narrator's and the reflector's voices, because each heroine is a spokesman for her narrator (or, for Jane Austen), and therefore heroine and narrator are one from the ideological point of view. Though that may be the case, however, it is as simplistic to equate narrator with character as it is unfeasible to identify narrator with author: it has been often noted that the "point" of E and NA, for instance, is to prove the heroines wrong; in E, MP and P, the narrators occasionally use not only their heroines, but even such unsympathetic characters as Mrs Norris as reflectors.
More generally still, the narrator is apt to "ventriloquize" the speech of all characters, as well as the kind of "tittle-tattle" one could hear in Bath or in the village of Highbury (see Finch and Bowen). The famous incipit of P&P is a typical example: "It is a truth universally acknowledged, that a single man in possession of a good fortune, must be in want of a wife" (1). In this case, the narrator overtly appears to endorse the proposition expressed by the relative clause, but a certain degree of exaggeration suggests that the endorsement may not be totally heart-felt ("It is a truth universally acknowledged"). This kind of "stylisation," or ventriloquizing appropriation, illustrates very well the concept of "irony" in relevance theory: according to Wilson and Sperber, irony is "echoic language": "The speaker echoes a thought [he/she] attributes to someone else, while dissociating [him/herself] from it with anything from mild ridicule to savage scorn" (265). While the definition of irony as echoic language does not hold for all ironical statements (unless the definition is circular), it applies very well to the kind of appropriating game Austen's narrator plays: and since it is often difficult to determine whether the narrator's voice aligns itself with or detaches itself from (and if so, in what measure) from the other voices it swallows, the reader is left without a firm evaluative ground to stand on.
In the end, we find that we cannot "catch" Jane Austen in her novels, because she is simply not there to be caught; only Booth's "Jane Austen" walks through the rooms of Barton Cottage or in the Mansfield Park grounds, silently watching, loudly commenting on, openly or covertly conniving with the (other) characters. The presence of "Jane Austen" awakens the reader's desire to know Jane Austen's mind, and at the same time it posits access to "the real Jane Austen" as impossible. With a further complicating move, however, even "Jane Austen" goes into hiding behind her reflectors, or in the meanderings of description. The reader can rely on no stable evaluative center, and opacity becomes the rule of the most crystal-clear of narrative creations.
Traditional "ironical" readings of Jane Austen (see Mudrick's Jane Austen: Irony as Defense and Discovery) set naive "first impressions" against a more sophisticated reading of the novels, promoted by the narrative structure itself. But in Jane Austen's novels, whenever readers' expectations are frustrated, one reading is not simply substituted for another: interpretations are heaped upon interpretations, and if certain evaluative comments are presented as more authoritative than others, in other cases readers do not know whether they are allowed an insight into the heart of the matter, or whether they are only following this or that character (or the narrator-as-a-character) in their misreadings. Logic, linguistic knowledge, and literary expectations cannot unravel one discourse from another, one interpretation from another. On the one hand, Austen tricks us into believing that certain evaluative comments are more reliable than others; while on the other, she allows us no stable source of authoritativeness, by proving that a chance word, or a silence, can contain a bigger grain of truth than a long "authorized" speech.
Anderson, Laurie, and Julia Bamford, eds. Evaluation in Oral and Written Academic Discourse. Roma: Officina Edizioni, 2004.
Austen, Jane. Emma. 1815. Oxford: Oxford UP, 1998.
--. Mansfield Park. 1814. Oxford: Oxford UP, 2003.
--. Northanger Abbey, Lady Susan, The Watsons, Sanditon. Oxford: Oxford UP, 2003.
--. Persuasion. 1818. Oxford: Oxford UP, 2004.
--. Pride and Prejudice. 1813. Oxford: Oxford UP, 2004.
--. Sense and Sensibility. 1811. Oxford: Oxford UP, 1992.
Bakhtin, Mikhail. The Dialogic Imagination. Trans. Caryl Emerson and Michael Holquist. Austin: U of Texas P, 1981.
--. Problems of Dostoevsky's Poetics. Ed. Caryl Emerson. Manchester: U of Minnesota and Manchester UP, 1984.
Black, Elizabeth. Pragmatic Stylistics. Edinburgh: Edinburgh UP, 2006.
Booth, Wayne C. The Rhetoric of Fiction. London: U of Chicago P, 1961.
Carroll, Lewis. The Complete Works of Lewis Carroll. Harmondsworth: Penguin, 1988.
Cortazzi, Martin, and Lixian Jin. "Evaluating Evaluation in Narrative." Hunston and Thompson 102-20.
DeForest, Mary, and Eric Johnson. "The Density of Latinate Words in the Speeches of Jane Austen's Characters." Literary and Linguistic Computing 16.4 (2001): 389-401.
Dickens, Charles. David Copperfield. Harmondsworth: Penguin, 1994.
Ehrenpreis, Irvin. Acts of Implication: Suggestion and Covert Meaning in the Works of Dryden, Swift, Pope, and Austen. Berkeley: U of California P, 1980.
Farrer, Reginald. On the Eaves of the World. London: Edward Arnold. 1917.
Finch, Casey, and Peter Bowen. "'The Tittle-Tattle of Highbury': Gossip and the Free Indirect Style of Emma." Representations 31 (1990): 1-18.
Fludernik, Monika. The Fictions of Language and the Languages of Fiction. London: Routledge, 1993.
Fowler, Roger, Linguistic Criticism. Oxford: Oxford UP, 1986. 1990.
Gay, Penny. Jane Austen and the Theatre. Cambridge: Cambridge UP, 2002.
Halliday, M.A.K. An Introduction to Functional Grammar. London, Edward Arnold, 1985.
--. "Types of Process." Halliday: System and Function in Language. Selected Papers Edited by Gunther Kress. Oxford: Oxford UP. 1976. 159-73.
Hoey, Michael. "Persuasive Rhetoric in Linguistics: A Stylistic Study of Some Features of the Language of Noam Chomsky." Evaluation in Text, Authorial Stance and the Construction of Discourse. Eds. Susan Hunston and Geoff Thompson. Oxford, Oxford UP. 28-37.
Holly, Grant I. "Emmagrammatology." Studies in Eighteenth-Century Culture 19 (1989): 39-51.
Hunston, Susan, and Geoff Thompson, eds. Evaluation in Text. Oxford: Oxford UP, 2000.
Labov, William. Language in the Inner City. Philadelphia: U of Pennsylvania P, 1972.
Leech, Geoffrey N. Principles of Pragmatics. London: Longman, 1983.
Leech, Geoffrey N., and Michael H. Short. Style in Fiction: A Linguistic Introduction to English Fictional Prose. London: Longman, 1981. 1983.
McIlvanney, William. Docherty. London: Hodder & Stoughton, 1996.
Miller, D.A. Jane Austen, or The Secret of Style. Princeton: Princeton UP, 2003.
Morini, Massimiliano. "Kazuo Ishiguro's An Artist of the Floating World as a Pretext for Jane Austen's Emma." Jane Austen: Oggi e ieri. Ed. Beatrice Battaglia. Ravenna: Longo, 2002.77-93.
Mudrick, Marvin. Jane Austen." Irony as Defense and Discovery. Princeton, NJ: Princeton UP, 1952.
Nardin, Jane. Those Elegant Decorums: The Concept of Propriety in Jane Austen's Novels. Albany: State U of New York P, 1973.
Paris, Bernard J. Character and Conflict in Jane Austen's Novels: A Psychological Approach. Brighton: Harvester, 1978.
Patteson, Richard F. "Truth, Certitude, and Stability in Jane Austen's Fiction." Philological Quarterly 60 (1981): 455-69.
Pratt, Mary Louise. Toward a Speech-Act Theory of Literary Discourse. Bloomington: Indiana UP, 1977.
Simpson, Paul. Language, Ideology and Point of View. London: Routledge, 1993.
Southam, B. C., ed. Jane Austen: The Critical Heritage, 1870-1940. London: Routledge, 1987.
Sutherland, John. Who Betrays Elizabeth Bennet? Oxford: Oxford UP, 1999.
Thompson, Geoff, and Susan Hunston. "Evaluation: An Introduction." Evaluation in Text, Authorial Stance and the Construction of Discourse. Eds. Susan Hunston and Geoff Thompson. Oxford, Oxford UP. 1-27.
Thompson, James. Between Self and World: The Novels of Jane Austen. University Park: The Pennsylvania State UP, 1988.
Toolan, Michael. Narrative: A Critical Linguistic Introduction. London: Routledge, 2001.
Trilling, Lionel. "Emma, and the Legend of Jane Austen." 1957. Jane Austen: Emma. A Casebook. Ed. David Lodge. Houndmills: Macmillan, 1991. 118-37.
Wallace, Tara Ghoshal. Jane Austen and Narrative Authority. Basingstoke: Macmillan, 1995.
Wilson, Deirdre, and Dan Sperber. "On Verbal Irony." The Stylistics Reader. Ed. Jean Jacques Weber. London: Arnold, 1996. 260-79.
University of Udine, Italy
(1) In what follows, Austen's novels are identified by a system of abbreviations: S&S (Sense and Sensibility, 1811), P&P (Pride and Prejudice, 1813), MP (Mansfield Park, 1814), E (Emma, 1815), P (Persuasion, 1818), NA (Northanger Abbey), LS (Lady Susan).
(2) Lovers' Vows, the unacted play at the centre of Mansfield Park, has itself elicited similar divided comments as to its ideological function: as Penny Gay has recently written, "As the critical literature demonstrates, Kotzebue's play, as adapted by Inchbald, can be used to support both a conservative and a radical reading of the novel" (107).
(3) Indeed, for Fowler it is (fictional) language in general that expresses a world view: "modal devices ... make explicit (though sometimes ironic) announcements of beliefs; other parts of language, indirectly but nevertheless convincingly, may be symptomatic of world-view: it has traditionally been assumed in stylistics that the different ways people express their thoughts indicate, consciously or unconsciously, their personalities and attitudes" (132).
(4) Certain systemic linguists working on storytelling have argued that Labov's is only one of several narrative genres, not all of them comprehending six parts. An "anecdote", for instance, has no resolution and displays only a minimum of evaluation. This open-endedness may be one of the reasons why Miss Bates' speeches in E are regarded as boring and inconclusive (while they often hide crucial truths).
(5) These three functions seem closely allied to Halliday's three grammatical "metafunctions" (Halliday, An Introduction to Functional Grammar xiii, passim).
(6) This is true even for "subversive" novels, because subversion is a form of reverse reflection.
(7) The notion of "implied author" looms behind the idea of "responsible authority," because even if we cannot re-construct an author, we can think of an authorial figure creating a narrator within a fictional world. For the sake of simplicity, however, and because we could by the same token identify countless intermediate stages, we can skip the "implied author" and think of the narrator as the person in charge of a narrative, even when that narrator is also a character. At the same time, we will see how the narrator him/herself can undermine and disperse his/her authority by conferring it on others, or by showing the uncertain grounds on which that authority is founded.
(8) The reference is to Sylvie and Bruno Concluded (Carroll 556).
(9) Toolan interestingly defines narrative "point" in a didactic manner: "A narrative is a perceived sequence of non-randomly connected events, typically involving, as the experiencing agonist, humans or quasi-humans, or other sentient beings, from whose experience we humans can 'learn'" (8).
(10) See Fludernik (443): "I am here modifying and refining Chatman's views on the narrator ... I reject Chatman's 'narrator at all times' (including a 'cinematic narrator'), but decisively maintain the existence of narration, and a gradual scale between an overt (personalized) narrator persona and a more covert narrative voice all the way to an objective backgrounded narrative function in reflector mode narrative."
(11) Penny Gay offers a theatrical interpretation of these final interventions. In the late eighteenth century, such famous actresses as Dorothy Jordan and Frances Abington were often assigned the task of reciting the final prologue, in Shakespeare's As You Like It as well as in more recent productions. "For a gleeful extra minute or so of theatre time", Shakespeare's Rosalind "plays with the audience's expectations of conventional gender behaviour, and invites applause for her--and her company's--performance. In doing so she both ironises the apparent closure of the story that the audience has just enjoyed, and leads the audience to appreciation of an even more sophisticated pleasure: recognition of the creative energy of the author and the actors". Gay suggests that "Jane Austen ... takes a similar position on the stage of her own creations, her novels, putting on with a flourish the mask of 'author' and speaking with affectionate irony of the story that we have all--author, actors, and audience--been involved in ... Austen, like the principal actress, is both inside and outside the novel as it ends: both authoritatively knowledgeable about her fictional world, and ironically dismissive of its reality" (166-67).
(12) James Thompson has noted the conventional character of these final summaries, which were also employed, among others, by Scott, Edgeworth, and Inchbald: such strategies represented a reaction against "a generation of overblown language of sentimentality ..., implying that [previous] novelists had used up the language of emotion. If this most important emotion cannot be expressed well, Austen and contemporary novelists imply, it ought not to be expressed at all" (72-3).
(13) Leech formulates this principle to account for certain "uncommunicative" features of conversation, but the definition can be easily adapted to the written word: "I shall tentatively propose ... an Interest Principle, by which conversation which is interesting, in the sense of having unpredictability or news value, is preferred to conversation which is boring and predictable" (146).
(14) Once again, narrative information is withheld for the sake of mystery--Leech's "interest principle." Readers, however, may feel they are being unfairly treated, because if the narrator knows, then he/she should tell, and if he/she does not, there is no apparent reason why he/she should occasionally judge the "real" state of affairs.
(15) Predictably, Miller confines the novels which most evidently disprove his theory of impersonality to the periphery of the Austen canon: NA is "The least revised of Austen's early novels, while P becomes the great false step of Austen Style" (33, 68).
(16) The "author" also addresses his/her "audience," i.e., the "narrator" addresses his/her "narratees," thus foregrounding him/herself as a character. When the narrator of P says that Lady Russel's aversion to the idea of a second marriage "needs no apology to the public" (P 11), we are encouraged to think of him/her as a person, or at least a persona.
(17) See DeForest and Johnson, who also comment on how the conflation of narrator and heroine/reflector tricks the reader into giving the latter more credibility than she deserves: "It is no coincidence that Austen's four great heroines are all within two percentage points of their narrators. Their voices blend with the narrators', and we unconsciously give them the authority of the actual storyteller. ... One of the great pleasures in reading Austen is being tricked by the heroine's mistake" (398).
(18) For a definition and a four-way taxonomy of mental process clauses applicable to Jane Austen's novels, cf. Halliday: "Mental process clauses are of four main types: perception (e.g. verbs see, look), reaction (e.g. please, like, smile), cognition (e.g. convince, believe, wonder), and verbalization (e.g. say, speak). The last is in fact rather different from the other three" (Halliday, "Types of Process" 165).
(19) The process here is one of doing rather than sensing, yet we are also allowed an insight into Anne's feelings (her unwillingness to be in the way when the Crofts visit her family house as prospective tenants).
(20) In Thompson and Hunston's three-way definition, this would count as "comparative evaluation" (the narrator' s, or Fanny's, vision--physical and moral--is compared with Mrs Norris').
(21) This passage contains a different example of "comparative" ("more truly well bred") mixed with "value-laden evaluation" ("a genteel, agreeable-looking woman").
(22) Michael Toolan identifies "free indirect discourse" as a privileged means to "align" the narrator with a character: "I favour the word 'alignment' because it doesn't prescribe whether that closeness of narrator to character is going to be used for purposes of irony, empathy, as a vehicle for stream-of-consciousness or the clashing of two voices, or whatever: the alignment is perceived, then the function (or 'naturalization') is worked out by the reader. The term 'alignment' also helps us keep in mind that, in terms of lexicogrammatical markers and aesthetic or narrative effect, there is a continuum from pure narrative words to pure character words, with any number of points on that continuum" (135).